Business Models for Family Physicians – by, David Filhart

Download PDF

Business Models for Family Physicians

By, David Filhart

Capstone Project

Davenport University

October 2014

Published in the Student Journal of Medicine: November, 2014

*Special thanks to Professor Dr. Timothy Delicath

for mentoring, editing, and suggestions.


Table of Contents



Table of Contents. 2

Abstract 3

Introduction.. 4

Secondary Research (Review of Literature). 9

Research methodology and design………………………………………………………………..……………….44

Data collection process…………………………………………………………………………………………………48

Data Analysis……………….……………………………………………………………..………………………………..52




Appendix ………………………………………………………………………………………..……………………………68



This project will closely analyze the different business models that are in play in the specialization of family medicine in the United States.  In this work, one will find that clear definitions and differentiations of the business models used in family medicine today are given.  With that, a careful look as to the various advantages and disadvantages are given herein.  Furthermore, an in-depth look into why there was a need to evolve and innovate from the traditional family medicine business model is looked into.  When examining this point there is analysis on the economic factors and other variables that came into play and which business models take care of each problem.  Also we examined which business models were the most viable and beneficial to the physician and the patient.  Specifically we looked at the happiness of the physician, the patient satisfaction, the annual income of the physician, and also their stress level.


Business Models for Family Physicians – by, David Filhart

As we examine the various business models for family physicians, the reader will come to recognize that they may come in many different varieties.  This paper examines why there are so many different business models in family medicine.  The reason that we are looking into this is because it seems that all physicians are playing with the same rules, yet they have so many different ways that they are playing within those rules.  Could there be a set of practices or an ultimate business model that would be more than satisfactory for all physicians?

We will also examine the different types of business models in detail so that we can try to understand this relatively new phenomenon as even today most of us are familiar and have the idea in our head of what and how a solo family physician would run their practice.  As we do this, we will look at the pros and cons of the different business models.  As we look into the various advantages and disadvantages of each business model, it is the hope of the author that this will also help to identify, define, and differentiate between the various business models for the reader.

We will also analyze the data that we received from first and second hand research.  The firsthand research was conducted through a variety of ways.  For instance, there were interviews conducted to the physicians via email, and also over the telephone, and finally face to face with the author.  The secondary research was gleaned from various scholarly articles that had been published and that the author of this paper thought was appropriate for this topic.  Finally we will synthesize the data that was gathered through the various methods to determine which business model would make the best business model for a family physician depending on a variety of factors such as their ultimate values and goals in practicing medicine and delivering excellent patient care.

Going through all of these stages of research and writing, our hope is that in this paper we will help students, physicians, and anyone else that is interested in the business models of family medicine to understand the different business models that are being used today.   As we do so, it is also the authors hope that this may possibly help to open the discussion as to what would be an even better business model in the future, that may not even be in existence today.


As there are many different ways to run a family physician to run a practice, it would be surprising to some to see the different and innovative ways that have evolved over the years.  This is due to the many different variables that the family physician is forced to deal with.  Having been exposed to many different ways that family physicians have as a business model, the author thought that it would be an interesting task to further research the various business models and to try to gather data that may help one to conclude the best set of practices that a family physician can use to be successful studies.

 Conducting the research by both first and secondhand sources has been really eye opening to the author as to the different business models those family physicians and other doctors are practicing.  It is the hope of the author is that an attempt to have synthesized the best parts here and that this will be beneficial and interesting to the reader, in such a way that there will be a better understanding not only to the difficulties that the family physician faces today, but also to some possible solutions to help these physicians.


The main objectives of this study are first to define and differentiate the different business models that family physicians are practicing in the United States today.  Another objective will be to understand why family physicians are looking at new and innovative ways to practice medicine as it relates to the business aspect.  Identifying the pros and cons of the various types of business models will also be an objective. The final objective is to help the reader to come to a conclusion of the viability of each business model.

Problem statement

The problem this study will address in particular are various and complex.  Because of different economic factors, various government laws and changing mandates Family physicians have been struggling to survive in the free market.  Other factors such as quality of patient care and time management have also been affected due to these factors. The family physicians have had to struggle to adapt their business model.  In order to stay in business, deliver quality healthcare, and to have a lifestyle by implementing new business practices that are more cutting edge than traditional the family physician in many cases is still struggling to survive in the healthcare industry today in the United States.  Finding the best business model for the family physician has been an ongoing process and is still up for debate (Goldberg, 2008).


Purpose of the Study

The purpose of this study is help the struggling family physician, and aspiring family physicians.  We will do this as we explore different business models.  This will better help us in order to identify a set of best practices.  A concomitant benefit will that this will also to help the reader to become better acquainted and familiar with the different business models that family physicians are using in the United States today.

 As this is done, hopefully the pros and cons of each respective model will be made clear, and the reader will be able to decide for themselves which business model, or combination thereof would be viable to them best suit their individual needs, wants, and style for their own preference.  Also, the reader will be able to see the why as to there being different business models for the family physician, any why they are continuing to evolve from traditional business models conventionally used in family practice.

            Finally, it is the author’s with that hopefully this paper may in some way help the reader to become inspired and to think of an even better business model for the family physician than what is presented herein, or what is currently being used today.

Research Questions

 When conducting the research, it was important to ask what the business model was exactly.  Also important was the why this business model would be necessary, and its viability in the real world.  What are the pros and cons of the various business models?  To go along with what was just said, what was it that the physicians in family medicine wanted to make them want to change the way that they practiced medicine?  Also, which practice was more beneficial, or what benefits did the patient see from the various business models?  Finally, what impact will each business model have on the lifestyle of the physicians?  When we attempt to answer these questions we will be able to solve the problem that family physicians are having today in their business practice as answering these questions will help us to identify the best and ideal set of practices that physicians can use, or are using today in their innovative business models.

Significance of the Study

For various reasons, family physicians have evolved and morphed the traditional and conventional business model into various forms (Goldberg, 2008).  This study will be significant insofar as it will help the reader to understand the different business models that the family physicians in the United States are practicing, and to further understand the pros and cons of each business model.


Literature review

In this section we will examine the various types of business models used by family physicians in the United States today.  We will also look at the advantages and disadvantages that each business model may present.  The way the research was organized was to give a brief overview of the business model.  After this we will examine why, and how this business model came about.  Next we will analyze the pros and the cons of the business model, as well as any other pertinent information that is useful to the reader.

Within each topic will be cited firsthand and secondhand research from personal surveys and researched literature respectively.  The different business models that will be examined are concierge medicine, the micro practice, the solo practice, the solo-solo practice, HMO’s, PPO’s, group practices, a traditional practice, health care homes,  and cash only practices, amongst any other variations.

Debra Goldberg put it well in her dissertation when she said that basically the external environment that many family medical practices face is so extremely complex.  Some of the variables that the family physician or practice has to deal with on a regular basis are the pressures and constant changes that come from regulatory sources.  Also an added pressure is the fact that reimbursement levels are decreasing instead of rising.  Another factor that is cumbersome for the family physician is having to deal with the changing of technology which seems to happen at a very rapid rate.  Finally, there is also a huge increase in patient and community expectations from their physician (Goldberg, 2008).

Concierge Medicine

To begin with, concierge medicine is gaining popularity in the field of family practice. Concierge medicine isn’t an entirely new idea in medicine however.  Although, the actual business model today is a bit different than it was fifty years ago (Hover, 2008).  It was interesting when talking to different physicians that are not in a concierge practice, that they would like to be in one and have thought about making the transition into a concierge practice. Needless to say, today’s definition of concierge medicine is different than the definition from yesteryear (Signature, 2013).

It is important to note, that in concierge medicine that there is a focus on preventative care.  This means that there is less hospitalization of the patients.  What is also great is that instead of a patient panel of say 3,000, most concierge physicians cap their practice at around 600 patients.  This definitely helps for better patient care.  This also helps to ease the stress of the physician.  This will not however help with the shortage of family practice physicians in the United States (Signature, 2013).

Concierge medicine is kind of like paying a physician a retainer fee in the legal world.  The fee that you pay can be a yearly amount that is spread over twelve months or in a lump sum, or whatever time table of payment is decided upon (Marshall, 2011). Originally concierge medicine was geared towards those with stratospheric incomes, now however, retainer fees at some practices have gone down and this has led to others with lower incomes to take advantage of and experience concierge service in medicine.

Concierge medicine is great for the patient if they really like the doctor, and they do not want to be inconvenienced by waiting more than a day for an appointment to see their physician.  The patient also is usually guaranteed more time in this type of business model with the physician.  This is especially appealing to patient who would like thirty minutes to an hour with their physician (Markiewicz, 2014).

Some of the reasons why concierge medicine became a popular niche in medicine are because a lot of physicians felt like they were seeing a huge number of patients, and not spending a lot of time with them.  For example, a physician could be booked with over twenty patients a day.  Oftentimes seeing 24 patients in a work day is the goal.  Sometimes this can average out to seeing a patient for 7-9 minutes maximum.  As one can empathize, it would be very advantageous to have such a schedule where you can see about a quarter or a third of the patients, and also be able to spend thirty minutes to an hour with them.  Not only would this be great for the patient, but it would be great for the physician, and this would equate to a mutually beneficial relationship and ultimately equate to phenomenal patient health care.

On the patient’s perspective, they often felt like they were paying too much for their healthcare and not getting enough time with the physician, and or the physician seemed rushed.  Another reason is because some people elected to pay for concierge medicine instead of having to deal with the hassle and cost of insurance, and concierge medicine allowed the patient a way to save money.  The patient would be able to save money with concierge medicine if they were to be able to pay less to the physician for their subscriber fee than to their insurance company premium.

There are many different variations of the concierge business model.  One thing that they all have in common is the fact that the patient has to pay a subscriber fee to the physician.  Usually the fee will ensure that the patient has 24/7 access to a physician. It usually also means that when a patient and a physician are in a concierge agreement, that the physician will have allotted at least an hour per patient visit.  In many instances the physician may actually then charge more to the patient if the allotted one hour visit with the physician is over for the month and then the physician may charge a fee by the minute for every minute that goes over the hour.

As stated earlier, the way a concierge practice is set up allows for the physician to accept an annual fee from the patient, for example.  This fee will cover whatever the physician and patient decide.  It could just be a fee for the patient to be accepted into the practice.  The fee might also cover the actual visit.  Sometimes the fee may even cover medications.  This is entirely up to the physician, and the patient.  The great thing about the retainer fee, for the physician and the patient, is that this allows the physician the luxury of seeing the patient for a longer visit instead of the rushed ten minute visit.

Most patients that can afford to pay for concierge medicine, also do keep their insurance for other reasons such as surgeries, or if they have to see a specialist.  Depending on the insurance the patient has, the medications may also be covered by the insurance, and even reimburse the physician on top of the subscriber fee that the patient pays to be a part of the concierge practice.  The concierge fee or subscriber fee is never covered by the insurance company however.

This translates into the physician only needing to see so many patients a day to make ends meet, which would be a fraction of what they were seeing, because they have the retainer fee coming in regularly.  For example, usually a healthy practice has around 2500 patients.  In concierge medicine, often the practice caps off the patient load at only 600 patients.

On a daily basis, a physician practicing concierge medicine may see 6 to 10 patients a day.  They also spend as much as thirty minutes to a half of an hour with each patient.  Elderly patients especially appreciate this as they are often managing multiple issues or problems with their health.  Oftentimes the patient, especially the elderly patient, has a caretaker that comes along with them to appointments.

The bringing of an additional caretaker can extend the meeting time with the patient as the caretaker is another person with information and who often expresses what the patient needs to, or at least prods the patient into saying what they should say if the patient does not remember or perhaps does not necessarily want to discuss a particular problem due to embarrassment or preference personally.  Concierge medicine is great for these types of patients as the doctor is not worried about the time constraints that may come if he or she had to see thirty patients a day.  Furthermore the caretaker and the patient do not feel they are imposing on the physician when they go in for a doctor visit.

The pros of a concierge practice are that the physician can see fewer patients in a day, and also see the patient for a longer duration of appointment.  This means that the work load will be a lot less, which equals more personal time for the physician to do other activities such as recreation or family time if they choose it.  At the very least, this is an option that is very inviting for a lot of physicians (Growth, 2014).

As far as the patient is concerned, the patient gets to get in to see the physician quicker, usually the same day that they call, and for a longer period of time.  Also, for the physician, there is a steady stream of income no matter how many patients they see, or do not see.  This type of business model also translates to less workload for the office staff.   Because there are less patients to manage for the practice, the workload is also lessened for the staff.

Some of the cons to concierge medicine are that the practice needs to have subscribers, and it could take time to get subscribers to this system.  Another con is that it may be more difficult for a physician to take time off as the patient will most likely expect service from their physician that they are retaining as opposed to a mid-level health provider such as a physician’s assistant or a nurse practitioner.

Sometimes it can be very difficult for a physician to get subscribers to their plan.  This is especially true because most patients are not accustomed to paying a subscriber fee for their healthcare, and perhaps do not want to, especially if they are already paying for big premiums in health insurance (Growth, 2014).

Most patients that are paying for concierge service want to make sure that they see their physician that they are paying for, and no one else.  Another con is that the physician may not feel comfortable asking their patients for a subscriber fee.  However, it may be surprising to the physician how many patients are willing to pay to be a subscriber to a concierge practice if the quality of care is increased.

Another difficulty a physician may have when transitioning to concierge medicine is which consulting company they would like to work with.  There are a variety of concierge consulting companies that will help the physician to make the transition to concierge.  Depending on a variety of factors the physician may want to use one concierge consulting company over the other as some have policies that allow for the physician to be more autonomous and more personalization in designing their concierge practice (Growth, 2014).

Solo-solo Practice

 A solo-solo practice is one where the physician is the only actual person working at the practice.  In other words, there is no insurance specialist, and there is not receptionist.  That means that the physician is dealing with the insurance company and they are answering their own phone.  This also means that there is no medical assistant to take the blood pressure or to handle the urine sample of the patients.  The physician answers their own phone, runs the EKG’s, takes care of the billing, etc.  This can be a very difficult way to practice for these very reasons.  A physician working for a big corporation has an administrative team that takes care of the hiring and firing.  They also probably have a department for billing.  In corporate medicine the physician does not clean the toilets or take out the garbage, but in a solo-solo practice the person that is cleaning the toilets and taking out the trash is the physician.  It is easy to see how one could lower the overhead by a solo-solo practice, but then one has to consider the opportunity cost that comes from having to do things like cleaning the toilets and taking out the trash, while the could be seeing patients and actually making money.  In other words, the main way to make money as a physician in family practice is to see patients.  This is the physicians core competency.  However, if they are busy doing other things like taking out the trash, then they are in essence losing money.

Sometimes certain tasks are outsourced to other entities.  For example the billing for a medical practice could be outsourced to another company for a nominal fee.  The billing can be very difficult to handle as this can be very time consuming and a tedious task having to deal with insurance companies and the like.  In other words the physician is the only person running the practice and performing all of the duties, save it be running the labs which are also usually outsourced, although the physician is still having to draw the blood from the patients if need be.

A physician may choose to have a solo practice because they want control of their whole practice.  Another reason why a physician would choose a solo practice is because the overhead will be cut down by a lot if they do not have to pay for other employees, such as the receptionist, billing department, medical assistant, nurse, phlebotomist, etc.  The physician may not want to deal with any other staff.

A patient may decide to go to a physician who has a solo practice because they feel that this helps them to have more confidentiality in their health care between their physician and themselves.  Also they may choose to see a solo practice physician because it may seem to be more personable to them as a physician running a solo practice has less of a patient load as they can only see so many patients in a day.  A patient may also find that it is easier to set up an appointment to see a solo family physician because of the lower patient load.

Usually, the way a solo practice is set up is self-explanatory.  They are usually run from the physician’s home.  However, some are also run from a small office.  With the advent of various technologies it is quite possible for a physician to take on all of the tasks that it takes to run an office.  For example, the physician may not have to spend time answering phones and setting appointments as appointments can be set up online and confirmations can be automated by email or telephone.

Some physicians don’t mind answering their own phone, and even go so far as to make their cellular  phone number available to the patients  Also, tasks like taking blood pressure can also be automated as there are electronic blood pressure cuffs that are very accurate and can be done as the physician is taking the history of the patient.  These are just a few of the things that physicians in family medicine are doing today that may seem atypical to the traditional family medical practice.

The pros of a solo practice are that you do not have to deal with any office politics.  This is especially nice if you are more introverted and does not care to make small talk with the staff.  Another plus about not having staff is that you do not have to spend the time or the resources to have to train them.  The physician also does not have to solve problems that can arise from personality conflicts that may arise between co-workers.  The physician also does not have to supervise or be a task manager for any other person except for themselves.  Also, the overhead is a lot lower due to the lack of staff or employees.

Because you are responsible for doing everything, this may be nice as there will not be any monotony and you will always be busy doing something, from billing, cleaning, replying to emails, and marketing.   Another plus is that if the practice fails, there is no one to blame it on except for you.

Some of the cons are that there is no one to substitute for you, for example, if you need to take time off.  You are the only employee.  Also, you have to do all of the work.  This means everything.  A lot of the things some physicians would feel are beneath them after all of their hard work in school.  Also, all of these other tasks can take away from the big money maker which is seeing patients.  Having to do a bunch of other tasks may cut into the time available for actually seeing patients. Another con is that if the practice fails, you only have yourself to blame (also see pros).

Many solo practice models are what is used in small towns because there is less competition from corporations.  Also the small town doctor may choose this practice model in a rural setting because the physician is actually the right fit for this type of lifestyle and environment as it takes a certain type of physician personality to make it in a solo practice in a rural area as evidenced by the high turnover of family physicians in rural areas in the United States (Stucke, 2009).  In fact, in states like Florida, the family physician solo practice has fallen from sixty-nine percent to just over forty percent (Bryant, 1998).

An entity that is really putting a strain also on the solo practice is that of the managed care practice.  That is what health maintenance organizations are.  These organizations are forcing solo practice physicians out of business because insurance companies give them contracts with a lot of patients, and the patients are charged less as there is a volume discount to be in the HMO or the PPO. This is in turn making it very difficult for the solo family physician as they are losing patients to the HMO’s and the PPO’s and they do not have the negating power of a large group to deal with the negotiation of the contract that would come if physicians banded together into a group.  So this means that the intimate care of patients that has been practiced for centuries in the traditional solo physician family practice is becoming extinct in a lot of ways.  This then means that patient care is becoming less intimate and more managed by a team of physicians or a health care team.  Some argue that for this reason managed care is better than solo practice care as far as the quality of healthcare that the patient gets is concerned (Bryant, 1998).

On the other hand, one study suggests that the productivity of the physician is much more for a physician in solo practice as opposed to a physician that is working for a managed health care team.  This is kind of like saying that competition between the two models is give and take as one has strengths where the other has weaknesses (Bryant, 1998).

Micro Practice

A micro practice is very similar to a solo-solo  practice, with the main difference that of being a physician who is in a micro practice will often have at least one other staff member, usually a receptionist/medical assistant.  This is so the physician will not have to answer their own phones, draw labs, or take vital signs. These are just a few of the unique tasks and practices that you will see a physician in a micro practice perform (Guglielmo, 2006).

A physician in Portland named Dr. Ott said that she did a lot of research before she actually switched to the micro practice from a more traditional practice in the year 2008.  Before Dr. Ott did this however, she, ultimately realized that even though her overhead would be low, and the stress level would be low, that she would indeed have more responsibilities that she did not worry about before, like the cleaning of the building, collections, insurance negotiations, and other responsibilities that are usually delegated to other people in a larger practice or group setting (Hands, 2011).

The reason that a physician may want to have a micro-practice is because they want to cut the costs of having a lot of employees, while at the same time keeping someone on that can do a lot of other tasks, thus allowing the physician to focus more on the medical side of the office.  Virtually every other aspect of a micro practice is similar to that of a solo-solo practice.  Although there is a crisis in family medicine due to the shortage, it has been said that the micro practice will not help to alleviate this in any way because they usually see less patients per practice (Davis, 2008).

To go back to Dr. Ott, it is interesting to note that since the six years that she started her micro practice, and taking on doing everything, that she let herself have a little bit of help. Although she is still considered a micro practice, she allowed for help in the cleaning department, as well as some billing, and an answering service (Hands, 2011).

Physicians that practice medicine in the micro practice model seem to be very happy (Happy, 2011).  They really seem to enjoy the interaction with the patients and the flexibility to make their own hours.  This is probably due to the less stress they feel with the lower overhead costs.  Also contributing to their happiness is the sense of owning their very own practice (Happy, 2011).

Even though on average they get paid less, they are also under less stress, and usually feel that this trade-off is well worth it. In fact, in a recent article, it was found that because of the less overhead, and the stresses that come with a larger staff, and running a larger organization, that this actually equates to more fulfilling practice for the physician as they do not have to worry about the things that are not related to patient care (Jespersen, 2006).

In addition to this the patients also enjoy going to a micro practice.  This could very well be because they feel that they are getting quality care for their money.  This is because the patient gets to spend more time with the physician.  This could also be because the patient is in an environment that seems more warm and friendly, due to the atmosphere (Kerr, 2007).

In a recent article published by the Morning Sentinel, it follows a physician who worked in corporate medicine for a long time.  The physician stated that she was secure in her corporate life, but then saw that there was a different way to practice in an article that she had read.  This physician decided to start a micro practice.  One of the patients that was seeing this new physician stated how she was especially surprised to find that she could get into the practice without a long wait.  One of the previous family physicians that she called said that she would have to wait for three to four months for an appointment, whereas the physician that was practicing in the new micro practice accepted the patient that day.  This really made the patient satisfied.

What was even more satisfying for the patient however, was that the physician from the micro practice was able to direct the ill patient to the necessary specialists that were needed to get the patient back in good health.  The patient felt that, after she got healthy again, that she did not want to bother the micro practice doctor, but to her surprise, the doctor called her to follow up and see how she was doing.  This really made the patient happy, and the patient came to the realization that the doctor did want to hear from her (Jespersen, 2006).

Many physicians describe the frustration that they have to spend so much time doing paperwork that they feel is unnecessary.  So, this makes sense also because the less paperwork there is to do, the less stress there will be, and furthermore this could mean more time with patients, and it usually does (Lactis, 2008).  For all of these reasons it seems that the combination makes it much more of a pleasurable experience for the patient to go to a physician that practices in the micro practice business model in the United States (Painter, 2006).

Traditional Solo-Practice

  A more traditional solo practice is where a physician has a full staff, consisting of a medical assistant, who would take the vital signs and bring the patient back.  There may also be a receptionist that works the phones and greets the patients.  There may also be an insurance and or billing specialist. Also, there may be a person or a department for marketing.   Finally, there may be an office manager who helps to make sure all of the pieces are running smoothly and that can supervise all of the office workers.  There can be of course many variations of the way an office is set up, however this is the most conventional way (Practice, 2011).

The office manager in a traditional solo practice helps they physician to be able to take care the patient care and also to let the physician know of any problems or concerns that may arise in the day to day business of the practice.  An office manager can also serve as a liaison between the staff and the physician.   All of these tasks may also be combined, shared, or overlapped with other responsibilities and duties that are necessary to run a smooth office.

In addition to the office staff, a physician may elect to have a number of mid-level providers such as nurse practitioners and or physician’s assistants. The mid-level health providers can be an excellent source of revenue as they can see patients on behalf of the physician.  This can increase the amount of patients that a practice sees in a day.  This can also add quite a bit to the revenue stream of the practice as well as the profitability of the said practice (Journal, 2004).

A physician may elect to be in a solo practice because they like things done their way (Beaulieu-Volk, 2014).  They also enjoy having the appendages of an office staff and or mid-level health providers to keep the practice running smoothly.  The traditional solo physician has their independence, as well as a team to help the practice.

Some of the pros of having a traditional solo practice are that of enjoying the camaraderie of the office staff.  This can be like a team with the purpose of the team being to provide excellent patient care (Bryant, 1998).  This can in turn have a synergistic effect and increase the positive energy of the office.  Another pro is that the physician can delegate certain responsibilities and duties to the staff as they may deem necessary.

Additionally, depending on the state laws, the physician with mid-level health providers may be able to be out of the office while the nurse practitioner or the physician’s assistant is seeing patients.  This would allow for the physician to be more flexible in taking time off for whatever reason they may have.  However in some states the physician does need to be in close proximity to the practice, even in the building, when managing mid-level health providers.

Another important pro in being a traditional solo practice is that the physician is independent.  They are independent from the policies and rules that are outlined by administration.  They also do not have to deal with being treated with disrespect from the any administration and do not have to worry about getting written up or being evaluated every few months.  The physician also does not have to worry about getting their pay docked if they do not see a certain number of patients.  This type of practice can be a huge reduction of stress if the physician is not up to dealing with administration looking over their shoulder (W______, personal communication, October 5, 2014).

In a traditional solo practice the physician is the one who gets to call the shots on how they want the practice to be run.  They have no other supervisor.  They have no other partner or partners that they have to deal with, or to consult.  When there is a decision to be made, or a change needs to be made, the physician can decide what they feel is best and executes their vision.  The physician is ultimately the head boss, unless the physician has a wife, then she is the head boss.

Even though many of these advantages are great, the disadvantages of a solo practice can also be great.  For example, there is a family medicine doctor that is said to not even be able to give away his practice (Gardiner, 2011). This is a huge disadvantage, especially if the physician was banking on the sale of the practice for money to place towards retirement.

Due to the ever competitive medical field, with corporate medicine being the stiffest competition, it is very difficult for a solo practice to survive (Girion, 2008).  With that being said however, with the advent of new technology and the costs coming down from such, many physicians in solo practice are taking advantage of this.  The technology that is available today is actually giving a lot of family physicians a chance whereas if they did not have this they could be in even more trouble economically than they are now (Naik, 2007).

In a recent article that was referred to earlier, we learn of a family practice physician that could not even give his practice away (Gardliner, 2011).  We learn that this family physician is in a predicament that many solo practice family physicians are in.  The article argues that quite possibly this type of medical business model will soon become extinct due to the mandates that are forcing this way of practicing medicine very difficult and some may say obsolete.

The article follows Dr. Sroka, who himself employs the equivalent of five full time employees.  In order for Dr. Sroka to make a profit, he needs to see four patients an hour.  If Dr. Sroka only sees three patients an hour he will just barely break even.  The Doctor works long hours.  Many of his patients he has had for over thirty years.  One year he made over three hundred thousand dollars, and this year he will be lucky if he makes over one hundred thousand dollars.  The reason why he can’t give away his practice which boasts about four thousand patients is because he is working so hard for so little money (Gardiner, 2011).

Most new physicians that may be looking to buy a practice are burdened with a lot of debt and also want a lifestyle.  In the article it states that over half of the family physicians coming out of residency are women.  These female physicians as it is stated in the article that they preferred or wanted the weekends off, the ability to go to their kids soccer games, and want to go shopping and enjoy life while paying off their student debts.  This dream is virtually impossible if they were to buy Dr. Sroka’s practice (Gardiner, 2011).

Dr. Sroka’s son says that he feels that his dad is just going to lose it one of these days as there are new mandates that come out all of the time that he just can’t keep up with.  One example of this is the electronic medical record which Dr. Sroka may not even be able to afford, or let alone have the desire to learn how to use so late in his career (Gardiner, 2011).

Group Practice

 Although the official definition from Medicare and Medicaid for a group practice is 25 or more physicians together, for the purposes of this paper, and what is known and what actually makes sense and is used in the real business world, we will not be using that definition here (Lee, 2013).

A group practice is a medical practice where there is more than one physician in the practice.  This can range from two physicians to very large groups of over eighty physicians.  They are also fully staffed in the office and may utilize mid-level health care providers to help bring in more revenue and profitability to the practice.

A physician may choose to be in a group practice because of the various advantages. One major advantage is being able to rotate being on-call.  This is very hard if the physician is solo, and a solo physician may feel tied to the practice.  Physicians in a group practice can also cover for each other when they are going on vacation or emergencies.

The phrase power in numbers definitely rings true here in the group practice.  This is because the group of physicians will have increased bargaining power when working with and negotiating contracts with the insurance companies.  There is also usually more structure as far as vacations and people that are willing and able to cover for you when you are on vacation.

Being in a group practice can come in various forms.  Usually, there are a set of standards or guidelines that must be adhered to in order to be a part of the group.  For example, some of the mandated polices could reflect the hours of operation, or standard procedures.  Group practices come in all different shapes and sizes.  The autonomy of the physician can also vary by degrees depending on the group.  Some group practices will recruit physicians and they will staff the office and even the building to the physician.

This is because often times, resources are pooled to help the physicians in their practice.  For example, a certain percentage of the budget may go to advertising and marketing for the physicians in the group. This would obviously benefit all of the physicians that practice under the group’s banner.  Also, when physicians are part of a group, they often refer patients to other physicians in the group.  This is also mutually beneficial between the physicians in the group as this helps to keep patients coming through the doors.

Some disadvantages of being in a group are the loss of some control over the way that you would want the practice set up.  For example, the physician may not be in charge of hiring staff, and may be stuck with whoever is assigned to the office.  This may prove to be difficult as certain personalities can clash.  Also, the person that is hired for the job may not be performing as well as a physician would want, but not bad enough to get fired, and there really would not be much the physician could do about it.  This is only in some cases, like there are many different variations of the group practice.

Depending on the deal made with the group, there may be certain criteria and quotas that need to be met in order for the physician to be a part of the group.  For example the physician may have to see so many patients a day in order to receive a quarterly bonus.  This is because oftentimes resources are pooled to help the physicians in their practice, so everyone has to pull their weight.  This could add a lot of extra stress to the physician.  For example, a certain percentage of the budget may go to advertising and marketing for the physicians in the group. This would obviously benefit all of the physicians that practice under the group’s banner.  Also, when physicians are part of a group, they often refer patients to other physicians in the group.  This is also mutually beneficial between the physicians in the group as this helps to keep patients coming through the doors.

Another disadvantage would be is having to deal with any negative repercussions that could come if there was a physician or controversial scandal that occurred that would tarnish the brand equity of the group’s name.  This is because things that happen under the group name could negatively impact you as a physician as your name would be associated with it.  This may be a rare incidence, but it does happen unfortunately. This would be very frustrating if you were an ethical physician, to have your name associated with something that you may have had nothing to do with.

Another situation that may be difficult to deal with in a group would be the hassle of making changes.  For example, if there was a certain electronic medical record that the group was using, but that the physician did not like for some reason or the other, the physician may end up being stuck with that software and be irritated by it on a daily basis.  This could be frustrating in the form of the Electro cardiogram machine that is used, or to the company that is contracted to run the labs, the insurance companies that the group decides to accept or reject, and to any other policy and procedure that is agreed upon in the group.  This may be the case if the group thinks that they could save money by using one type or brand of equipment, and another physician may not be satisfied with its performance.  Furthermore, if the equipment were not as good as they thought, and they spent a lot of money on the equipment, then the group would have wasted a lot of money and the frustration and the morale of the physicians and employees would go down, but the physician would almost be stuck as they would be under contract.

Another big disadvantage would be that when a person or a physician signs a contract or becomes a part of a group, there is usually a non-compete clause in the agreement.  This means that if as a physician you become fed up with the group, and decide to leave, you may not be able to practice in the same location, within a certain radius of any of the physicians in the group, and you would not be able to have your patients follow you.  This could be very difficult for the physician that has roots in the area in which they live, or possibly have children that are integrated into the community, church, or school.  Another difficulty would be the consideration of the spouse and what they would want to do.

This would mean that you would have to wait out your no compete clause and or move to another location and start all over.  This would be indeed very difficult as it can take a lot of time and effort to build a practice.  Because of this the physician may decide to stay with the group even though it is frustrating and difficult, thus making life somewhat miserable when it comes to practicing medicine and working (Editors, 2009).  The larger the medical group, the more difficult it may be to get things done as far as change is concerned as this usually means more channels and red tape to deal with.


Medical Home Practice

This is a unique niche in the medical world that has actually been around since the 1960’s (Business, 2011).  In this business model, the physician basically runs the medical home, and the medical home is a place where people can go for all of their health care needs from cradle to grave.  So a patient could come in to see their family physician, and if they needed to see a specialist, the specialist would be available in the same location.  To be clear, there is a team of medical health care providers that are able to communicate with each other, and communicate with the patient.  This is a great business model for the patient especially if they want a one stop type of shop for their healthcare.  This is really great for the physician as communication between colleagues is much easier as they are in close proximity to one another.

The advantage of this type of practice is that the patient will, in theory, be better taken care of because of the effective channels of communication that are in close proximity.  This is good for the patients of a family physician as the health care team will be working close together for the benefit of the patient.

One major disadvantage of the medical home business model is that it has been found to be more expensive to run than most other medical practice models.  This would most likely equate to less money being taken home by the family physician.  This would also equate to less money for everyone working in the medical home.  For this reason it seems that the medical home has been stagnant as far as growth is concerned, especially when comparing it to new and trendier business models for the family physician such as the micro-practice.

Preferred Provider Organization

Otherwise known as a PPO, this type of health organization is good for the patient that does not necessarily want all of their care coordinated through the primary care physician.  In other words, if a patient wanted to see a specialist like a dermatologist, the patient could do that if the dermatologist that they went to see was in the preferred provider organization.  The patient would not a referral from another physician first in order to get a referral.

For the physician, being a part of this organization may be good as the subscribers would have you as an option should they choose to use your services.  For example, if a patient does want to go to a family physician, and you are a member of the PPO, or under contract by them, then the patient would see your name on the list and possibly choose you as a physician (Bayley, 1998).

Another advantage for the family physician that is a part of the Preferred Provider Organization would be that, similar to a group practice, the team infrastructure would be there and because there is usually a large group of physicians that are contracted with the Preferred provider organization, there is bargaining power in this.

Some of the disadvantages would be that what would come with being a part of any large organization, which would be for example, having to deal with administration, personalities, policies, and other nonsense that one may not find in a smaller situation.  For example, a physician may have to see so many patients a month to fulfill a quota.  Furthermore, the physician may be told how to practice medicine.

To amplify this, let’s take for example, if the physician feels that a certain test is necessary, the preferred provider organization may ultimately say no and request another procedure be in effect first.  This would either be to save the preferred provider organization money, or to make the organization money. This ultimately helps the bottom line of the PPO.

If a physician feels that a patient needs a particular treatment, this could be very frustrating for a physician that feels that for the best interest of the patient that something else needs to be accomplished for that patient’s benefit, and yet it is rejected by the PPO.   It is incredibly interesting that some articles have stated that PPO’s are actually really popular amongst patients (Hurley, 2004).  This could be because the patient enjoys the ability to be able to go to a specialist without having to get a referral from a family physician.  The way that PPO’s keep this type of health care costs down is by offering the physician a bonus for keeping the costs of healthcare low and by practicing conservative medicine (Hurley, 2004).

Health Maintenance Organization

Otherwise known as an HMO, it is similar to a preferred care provider in the sense that the family physician will be a part of a network of physicians in this organization (McBride, 1995).  One of the main differences between a health maintenance organization and a preferred provider organization is that in the health maintenance organization the patient cannot just see any physician in any specialty that they want without a referral, even if the specialist is in the organization (Bayley, 1998).

In a health maintenance organization, the patient must first go to their primary care provider, which could be the family physician that is listed as a provider in the health maintenance organization.  If the primary care physician sees that the patient needs to go to a specialist, the physician will then make a referral to the specialist that is in the health maintenance organization.  In other words, all referrals have to go through the primary care physician first, which in a lot of cases would be the family physician.

It is argued by health maintenance organizations that this type of business model actually saves a lot of money for healthcare.  Health maintenance organizations argue that a lot of unnecessary procedures and services are avoided by this as much can be handled by the family physician instead of the premiums that are paid for seeing a specialist (Bayley, 1998).  This can be especially frustrating to the patient however, especially if a patient really believes that they need to see a specialist and their primary care physician has a different opinion.

All of the rest of the advantages and disadvantages would probably be similar in a health maintenance organization then they are for the preferred provider organization for the physician as far as dealing with administration and policies and procedures.  Although there is a certain amount of security that comes with working in a big organization, such as the luxury of working with other physicians and health care providers on a health care team, it definitely comes with a price.  One of those prices is physician autonomy.

It seems that physicians that work for health maintenance organizations are the most dissatisfied with their work.  In fact, they often fee overworked, and underpaid.  It has been said that in one article that they feel powerless and undervalued.  This article stated that out of the 24 physicians that were working for the health maintenance organization that they had a huge concern and that was of coping with their frustrations.

An advocate for the physicians stated that it was imperative to make sure that the physicians felt valued and also that a sense of autonomy in practicing medicine was also restored to them in order to ensure their well-being (Healthcare, 2004).  This would be very important for the health maintenance organization to do if they wanted to keep their physicians happy and for them to have longevity in their career.  One could only empathize how frustrating this could be, especially after having struggled and gone through all of the work it takes to become a physician, to be not happy with your work.

Cash Only Practice

A cash only practice is a business model that is being increasingly popular although it accounts for a small percentage of medical practices today.  In a cash only practice, the patient pays the physician directly.  In other words, the physician does not receive any money from insurance companies or any other organization (Castens, 2009).  The patient either pays with cash, check, or credit card.

A lot of times, when a physician has what is referred to as the cash only practice, they will give the receipt to the patient.  This receipt that the patient is given will have the proper billing codes on it.  What the patient can do with these billing codes on the receipt is take that and present it to their insurance company if they have one and get reimbursed by them (Chen, 2010). This practice saves the physician and their practice the time and resources that it takes trying to fight with the insurance companies in order to get reimbursed by them.

The advantages of having a cash only practice is that for the physician, they get paid immediately. They also do not have to deal with insurance companies that are notorious for not paying physicians.   They may not do this if they feel that a procedure is not necessary.  Also, insurance companies have been known to change their policies, so that if a policy is proper one month, it may not be proper the next month, thus causing a headache for the physician and their billing department.

Needless to say, not having to deal with insurance companies in a cash-only business model is a great relief for family physicians.  In fact, it has been said that insurance companies are so hard to deal with, and physicians are so frustrated with dealing with insurance companies, that the cash-only business model was born out of that frustration.

It is interesting how out of adversity sometimes the need to overcome them gives us creative solutions, as was the case for cash-only practices.  Another positive is that physicians and patients alike are coming to the conclusion that there is actually a mutually beneficial relationship that can be had between a patient and the physician without health insurance companies meddling in their business (Wlazelek, 2008).

Some of the advantages to the patient is that sometimes paying the physician directly for services is cheaper than actually having to pay premiums to their insurance every month.  The case may be that paying the physician for services may actually be cheaper than even paying their copay that the patient has to pay on top of their monthly premium (J. Fortes, personal communication, September 15, 2014).  This would be a great choice for a patient and a physician as the hassle of dealing with the insurance company is out of the equation.  Furthermore, insurance companies often charge or bill way more than a cash-only practice does.

The disadvantages of a cash only practice are that some people do not have the cash to pay for services, such as Medicaid patients.  This could dramatically decrease the number of potential patients that a physician may have available to them. This can definitely be a reason why some cash-only practices fail (Frank, 2008).

One interesting aspect of the cash-only practice model is that this can actually help the uninsured.  Some practices make it very difficult to see patients with no insurance.  However, the practices that are not dependent on insurance, and can lower their prices, are actually places where those that are uninsured may afford to go.  For example, a group of Texas physicians designed a cash only practice to specifically go after the uninsured patient demographic (Shinkman, 2014).

This business model of the cash-only practice really helped the people as they could not afford health insurance, or because the people felt that the insurance company was ripping them off.  When the word got out that there were physicians that were seeing patients without insurance for a reasonable price the people without insurance were glad to go to the cash-only medical practice that were affordable with a great quality of care (Shinkman, 2014).

One interesting example of the cash-only model is by a practice called Medical Associates of Lehigh Valley.  This group of physicians served approximately 100,000 patients.  The group became increasingly frustrated with dealing with the different insurance agencies as they would each have a different set of guidelines for the physicians to follow.  To add to this frustration, the guidelines are constantly changing.  So, for example, if a medication is covered by the insurance company to treat a diabetic patient one month, well, three months later on a follow up visit the same medication may not be covered so the patient has to switch prescriptions.  The insurance companies will do this without notifying the physician (Wlazelek, 2008).

This practice by the insurance companies would be especially frustrating if the physician and the patient know that the medication that was approved actually worked well for the patient, and now do to insurance constraints has to switch.  Medical Associates of Lehigh Valley figures that they can reduce the hassle of dealing with the health insurance companies by charging the patient and letting the patient seek reimbursement from their insurance company for either all or part of the fee.  This will cut costs as the Medical Associates of Lehigh Valley will not have to worry about a billing department, and will also not have to spend hours on the phone with the insurance companies arguing for reimbursement for proper patient care (Wlazelek, 2008).


Research Methodology and Design

As the author contemplated and pondered what would be the best method for gathering primary research and information, it was concluded that a set of direct and open ended questions would be the best way to find out from the actual physicians themselves about their business model that they are using in their family medicine practice.  These questions were reviewed and validated by the author’s peers before they were used in the primary research.  As the information from the interviews were recorded, it was then analyzed mathematically to come to conclusions of the effectivity of the business model when compared to the variables of income, lifestyle, stress level, happiness, and patient satisfaction.


A huge bulk of the information that was gleaned for this project was found from the literature review.  Fortunately there has been a lot of information in the form of various publications that document the different business models that physicians, specifically those in the field of family medicine, are currently implementing in the United States today.  Some of the aspects that this literature review revealed as far as the various family medicine business models are concerned were the reasons why there were possibly so many different business models for the family physician, as well as the differentiation between the different business models.

There was also gleaned from this secondary research the advantages and disadvantages of the various business models as they apply both to the physician and the patient.  Also, there was insight as to the strengths and weaknesses of the various business models and which business models would be most viable in the future for the family physician.

Furthermore, there was also, when appropriate, an in depth look to various stories or anecdotes that the author felt better grasped the big picture and or illustrated various points as to the general feeling of the physician, practice, and the patients.  This would come in the form of stories that were related to the author of the various publications that the author felt was appropriate to share as they were related to the various business models that the family physicians are using and evolving today in the United States.

There was also primary research conducted by the author in the form of face to face interviews, as well as interviews that were conducted via email.  The interviews were conducted from physicians that are practicing various business models all over the United States in order to try to get a general flavor and generalizations or common threads that these various business models may have to make them better classified and defined when attempting to differentiate between them.  While conducting the email correspondence, the author made sure that a good number of the questions were open ended to allow for the respondents to open up and put their own expression into the interview.

Interview Format

            The author conducted a variety of interviews that were face to face.  These face to face interviews were with Dr. E___ , a group practice family physician.  Dr. W_______, a physician that is practicing in the corporate medical system, Dr. J_______, who runs a cash-only practice, and Dr. B______, who operates a micro practice.

As stated earlier, these questions were reviewed and validated by the author’s peers before they were used in the primary research.  As the information from the interviews was recorded, it was then analyzed mathematically to come to conclusions of the effectivity of the business model when compared to the variables of income, lifestyle, stress level, happiness, and patient satisfaction.

The author actually also interned with each of those physicians and was able to also see firsthand how the practices are operated on a daily basis.  During this time the author was also able to get a general feeling of the office staff, practice, and patients.  The author was also able to see firsthand the interaction that the physicians had with their patients.  When appropriate, the author was also able to see the physicians interact with staff and administration, insurance companies, and other aspects of running a medical practice.

There were also a couple of telephone interviews, as well as a number of interviews that were conducted via an email questionnaire.  The author found various physicians across the United States to send these interviews to by finding their practices on the internet and their email addresses in the contact section of their websites.

The interviewees were at the very minimum asked a variety of questions (see Appendix A) that were related to finding out more about their particular business model.

Limitations and Assumptions of the Interview Questions and also in the Recording of the Data

            Although the primary research yielded valuable data, it is important to realize that due to time constraints and the relatively small sample size that the data may be limited.  The author attempted around 60 interviews, with less than half responding, and about half of those that did respond, responded with an answer of not being interested in being interviewed.

It is important to note that another flaw in the data could be that the people being interviewed may have skewed their responses due to the Hawthorne Effect, meaning that because they were being interviewed for an academic project that they could have exaggerated some of their responses in order to make their business model or situations look better or worse than it really was.

However, regardless of the Hawthorne Effect or any other bias, it is assumed that the interviewees were honest and their data and answers that were provided are presented as such.  The Hawthorne Effect was a concern as some of the data and answers could have been exaggerated.  It would have been ideal to have some of the information come from other sources such as staff, or accountants, to get a more accurate and unbiased answer.

It is also assumed that the author would not skew the information due to an observer expectancy bias, as the author has hopes and dreams of one day running their own family medicine practice with a particular business model in mind.  The observer expectancy bias indicates that the person that is conducting the research can unknowingly or knowingly skew the data and results to the expected or desired conclusions that the person has desired in their head.

Data Collection Process – Findings from primary research

The primary data that was collected through face-to-face interviews, telephone interviews, and interviews via email, all provided different perspectives and points of view that were unique to the physician and to their practice business model.  The face-to-face interviews were with physicians that the author actually interned with and spent time working in their practice.  The phone interviews and the interviews via email were randomly conducted with any family physician that would respond to the request for interview submitted by the author.

It was interesting to find that although each physician considered themselves as a part of a certain business model, that the author found that within each business model there were variations and differences between each physician’s style of implementation of that business model.  In addition to the style of the practice, other factors that would contribute to a physician’s variations between certain business models of the same niche was the stage in which the physician is in their career.  This meaning that the physician could be at the end of their career, nearing retirement, or just beginning their practice, or anywhere in between those stages.  For example, a physician that considered himself in a micro-practice but was almost into retirement and considered himself at the end of his career.  The difference between this physician and a physician just transitioning into a micro practice would be that the physician nearing retirement was making more money and working more.  Also, it was interesting to see the older physician took on more duties such as maintenance.

Generally speaking, the happiest physicians the author interviewed were those that were working for themselves in some sort of solo capacity, and the physicians that worked for a group or corporate medicine were the least happy.  It also seemed that the physicians that worked for a group or for a corporation did however make more money annually on average.  What was also interesting is that the physicians that worked for a corporation or a group were very vocal when in saying that they often considered going into concierge medicine, or a cash-only practice, or a micro practice.

The happiest physician that I encountered was a physician that had a micro practice that was cash-only.  This physician also had multiple mid-level health care providers working for her.  She was happy because she did not have to deal with insurance companies.  She also was well paid and appreciated by her patients.  She had a huge inflow of patients, but a lot of them were seen by her mid-level health care providers.  She had a low overhead.  Before she started her cash-only micro practice she had twenty years of experience working for corporate medicine.  She said that as she was older that she wanted to work smarter, not harder.  She was able to take all of the time off that she wanted and still made a great living as her mid-level health providers saw a lot of the patients (J_____, personal communication, October 01, 2014).

On the other end of the spectrum, the most miserable appearing physician was working for a big corporation.  He was making a lot of money, and had some great incentives if he made certain goals.  However, he was always stressed out, and worked long hours in order to meet his goal.  He also had problems with staff that seemed to be out of his control as he was not the one to be able to hire or fire employees.  He also mentioned to me personally that he really did not appreciate having to be evaluated by the administration, and he did not like getting written up when a patient complained.  He usually put in around sixty hours a week at work, but then volunteered at a free clinic a couple of nights a week also.  The author has chosen to withhold this physician’s name (Name withheld, personal communication, October 01, 2014).


Comparing the Primary Data to the Literature Review

  When we compare the primary data collected to the literature review, we see that they do in fact closely correlate one with another.  For example, the physicians that were read about in the literature review, we can see that the happier ones are the ones that are doing their own practice, and the physicians that are the most stressed out seem to be the physicians that are working in corporate medicine.  This seems to be the truth in family practice.

Another correlation that was found was that the physicians that made the most money annually were not necessarily the most happiest.  This was found in the literature to be true as well as from primary research.  However, it is also important to note that some of the physicians that made a lot of money comparatively to their colleagues could also be very happy.

It also seemed that a downside to practicing medicine in the United States today, especially as it applied to the family medicine specialty, was that all of the different rules and regulations seemed to be “superfluous” to practicing medicine as the rules and regulations do not really progress the practice of medicine or the quality of patient care (John Mochata, personal communication, October 02, 2014).  This seems to be the downsides as all physicians seem to agree on this common belief, no matter what their business model of practicing medicine is.

One difference that the author did note in the literature review that the author did not come across in primary research was the failures in certain business models that could happen.  For example there was in the literature review a cash only practice that failed (Frank, 2008).  This would be understandable as it is difficult to interview physicians in a failed business practice as they are currently working somewhere else and do not necessarily care to advertise that they were in a practice that did in fact fail.


Data Analysis

            The way the data was analyzed was by taking the recorded answers from the physicians that responded, then mathematically figuring out the average of the individual responses were, within the business model that the physician claimed was implemented in their family medical practice. When we analyzed the data, we could see that out of the physicians that responded to being interviewed, and also partially in combination with the data found in the literature review, we could illustrate the following graphs from the results: As we can see the family physician on average makes between $150-220 thousand a year for full time hours.  We can also see the stress level for the physicians that are working in a corporate setting are amongst those with the highest stress level.  We can also see that physicians in corporate medicine are also amongst the highest paid.  Interestingly enough, as far as their personal happiness is concerned, the corporate family physician is the least happy.

We can also see that the happiest physicians seem to be those in solo-solo practices, micro practices, or cash only practices.  We can also see the stress levels are lower in these areas of particular business models.  It is interesting to note that these physicians on average do not necessarily make the most money compared to their colleagues. However there were some outliers that made an annual salary that could be comparable to the highest paid corporate physician.


According to the data and research, the author believes that this implies that physicians that do not work for corporate medicine, or that do not have to deal with insurance companies, are the happiest.  This is because the physician has a primary focus of delivering the best possible patient care that he or she has the ability to deliver.  This is a very difficult thing to do in and of itself, let alone having to deal with the distractors that a physician would have to deal with just in order to pay his staff, and any other overhead that a physicians practice may have, just to stay in business.  This seems to be correlated with the fact that they have on top of all of this, an increase in stress in their jobs is also due to administrative watch-dogs and the like.  It can be very frustrating for a physician that is working as hard as they can to deliver quality care to their patients, only to have administration say that they are not seeing enough patients per hour, or that they need to see more patients in a day in order to get a pay raise.  It is also very frustrating for a physician to get evaluated by administration, especially as the evaluation can influence their income.  It is also an added stress to a physician when they have to deal with getting written up by administration when a patient complains.  All of these things combined can dramatically raise a stress level of a physician.

It also seems that the more government makes mandates and rules that this adds to the stress of medicine.  Additionally these extra rules and mandates do not necessarily improve the quality of care.  The author recalls working with a physician that was well respected by patients, as well as colleagues.  This physician had been practicing medicine for over forty years, and was very successful and skilled as a physician.  Then the government mandate that physicians were supposed to start using electronic medical records in order to get an increase in reimbursements from Medicare and Medicaid.  This physician, even though he was very competent, and his hand taking notes on paper worked just fine over the last forty years, found that he was increasingly frustrated having to learn how to use the computerized electronic medical record.  This also was frustrating for the physician as this took a huge amount of time, more than ten times that of taking notes by hand.

We can also see that the amount of money that a physician makes does not seem to correlate necessarily with the level of happiness that they have in their lives.  Although, it is also important to note that physicians that made a lot of money compared to their colleagues were not necessarily depressed or saddened either.

The author also believes that this implies that the happier the physician, the happier the patient.  If the physician is happy then the patient can benefit from this.  The patient will benefit by the way the service that they get in their healthcare from their physician.  Therefore it is very important for the physician to work in an environment where they are happy in order to deliver the best possible patient care in the family medicine specialization.



            It is the recommendation of the author, that after much research and consideration, that the ultimate business model for the family physician now and in the future would be that of a cash-only micro practice.  The reason that the author recommends this is because these business models tend to be the ones with the least amount of stress, thereby increasing the happiness level of the physician and the patient. One of the reasons for a happier patient is because they are receiving better patient care, and one of the reasons that the patient feels this way is because in these particular business models the patient can see the physician for thirty minutes to an hour sometimes, and they do not feel rushed with the visit from the physician.

The combination of the cash-only micro practice business model is also very lucrative as a business model.  This is most likely because the overhead costs are cut dramatically in the business models of this type.  This in turn increases the percentage of net earnings.

The ultimate family medicine business model practice should also hire as many mid-level providers as the state law will allow.  This will allow the physician to increase their income as well as see as many patients as they desire, and not have to worry if the block their schedule out for other responsibilities.   In this business model however, it is important to note that in some cases the physician feels more like an office manager as they have to manage the mid-level health providers, such as the nurse practitioner and the physician’s assistant.

The author would also ultimately and strongly recommend that for the physician to try to keep costs down, and to keep the stress level as low as possible when practicing medicine.  Remember that patient care is the primary focus.  When this is done, the physician can appropriately align their business design and personal goals in the best way.  If the physician does this, their business model will most likely be a successful one.




            It seems that there are many different business models to choose from, and there are so many factors to consider.  Keeping it simple however is going to be the best way for a physician to maximize their happiness and well-being.  This will also translate to patients being satisfied with their physician as they receive excellent care.

There should not be unnecessary time restrictions when a patient visits their physician.  In corporate medicine a physician is lucky to spend seven actual minutes with the patient face to face.  This can be really frustrating for the patient as they may feel rushed.  This is also frustrating for the physician as they do not like to feel like they are rushed either.  The patient may feel like they are burdening the physician if they take any more of their time, and this can be very dangerous.  This can lead to the patient being unwilling to give a full and proper history to the patient, thereby possibly jeopardizing the patients’ health, and this is the last thing that should be done in healthcare, as patient care should always be the primary focus.

There should not be excess mandates and government to take away from the practice of medicine.  It is widely felt that these mandates do not enhance or improve the practice of medicine.  Therefore, they do not increase the quality of patient care.  This is just a problem that adds to the stress, time, and ultimately monetary resources of the family physician.

There should not be administration always looking over the doctor’s shoulder to make sure that profits are maximized.  This type of business model adds a lot of stress to the physician.  It can also be humiliating to them.  This in turn becomes a distraction to the primary goal of healthcare, which is patient care.

When a physician decides on what business model is for them, they need to take a close look at themselves, their values, why they are in medicine, and their goals.  When these factors are taken into consideration, the physician will better be able to excise the unnecessary variables that are taking away from the practice of good medicine.  Also they will be able to implement the best set of business practices, which will allow them to be able to maximize and deliver the best possible patient care, which should be the goal of health care.



Bayley, E. B. (1998). The effects of managed care growth on physicians’ practices. (Order No. 9916578, The University of Rochester). ProQuest Dissertations and Theses, , 129-129 p. Retrieved from (304450143).

Beaulieu-Volk, D. (2014). Employment or solo practice: Which is right for you?. Newton: Questex Media Group LLC. Retrieved from

Bryant-Friedland, B. (1998, May 04). Ailing family practices managed care is making it harder for doctors in solo practice to survive. Florida Times Union Retrieved from

Business news; the brooklyn hospital center family medicine receives national recognition as A patient centered medical home. (2011). Pediatrics Week, , 1355. Retrieved from

Castens, E. (2009, Oct 25). Cash-only clinic gaining ground. McClatchy – Tribune Business News Retrieved from

Chen, M. Y. (2010, Jan 31). Doc’s cash-only billing bucks the insurance morass: A burnsville family physician totes a black bag on his house calls. McClatchy – Tribune Business News Retrieved from

Davis, L. (2008). Micropractice model not the answer to primary care crisis. Jebsen Beach: Basil & Spice. Retrieved from

Editors, B. (1999, May 03). Towers perrin study finds average health system-employed family physician more productive than group practice counterpart. Business Wire Retrieved from

Frank, E. K., M.D. (2008). Why my cash-only practice failed. Medical Economics, 85(5), 54-54, 57. Retrieved from

Gardiner, H. (2011, Apr 23). Family physician can’t give away solo practice. New York Times Retrieved from

Girion, L. (2008, Dec 15). COLUMN ONE; why the doctor won’t see you now; these days, a solo practice is increasingly unsustainable, as one young beverly hills physician learns by slowly going broke. Los Angeles Times Retrieved from

Goldberg, D. G. (2008). The influence of environmental factors and organizational characteristics on innovations in family medicine practices in virginia. (Order No. 3326668, Virginia Commonwealth University). ProQuest Dissertations and Theses, , 257-n/a. Retrieved from (304825335).

Guglielmo, W. J. (2006). What’s a micropractice? Medical Economics, 83(23), 51-7. Retrieved from

Growth in concierge medicine leads to more choices, and some challenges for physicians, according to specialdocs. (2014, Jul 28). PR Newswire Retrieved from

HANDS-ON AND HAPPIER THAN EVER. (2011). Medical Economics, 88(10), 31. Retrieved from

Healthcare models; family physicians in group model HMO feel powerless, undervalued. (2004). Health & Medicine Week, , 429. Retrieved from

Hover, L. M. (2008). Concierge medicine: Something old, something new. (Order No. 3334871, Drew University). ProQuest Dissertations and Theses, , 212-n/a. Retrieved from (304634047).

Hurley, R. E., Strunk, B. C., & White, J. S. (2004). The puzzling popularity of the PPO. Health Affairs, 23(2), 56-68. Retrieved from

Jespersen, B. (2006, Sep 18). Less proves to be more for this family doctor. Morning Sentinel Retrieved from

Journal, R. s. (2004, Oct 07). Family physician sets up practice at midwest regional medical center. Journal Record Retrieved from

Kerr, K. (2007, May 30). A new approach to family medicine gives doctors more access to their patients, practice made ‘micro’. Newsday Retrieved from

Lee, D., Fiack, K. J., & Knapp, K. M. (2013). A PROFILE OF SOLO/TWO-PHYSICIAN PRACTICES. Journal of Health and Human Services Administration, 36(3), 297-322. Retrieved from

Lacitis, E. (2008, May 04). Doctors going solo with micropractices; leaving a group practice can mean less paperwork, more time with patients. The Washington Post Retrieved from

Markiewicz, D. (2014, Jul 27). HEALTH CARE: Extra medical service, for a fee: In concierge medicine, doctors can have more time with patients. The Atlanta Journal – Constitution Retrieved from

Marshall, S. (2011, 01). Concierge medicine. Town and Country, 165, 104. Retrieved from

McBride, W., & Rearick, D. (1995). Physician employment experiences in health maintenance organizations: Analysis of an urban HMO. Health Marketing Quarterly, 13(2), 3. Retrieved from

Naik, G. (2007, Feb 23). Micromanagement: Faltering family M.D.s get technology lifeline; doctors think small to revive solo role for primary care. Wall Street Journal Retrieved from

Painter, K. (2006, May 22). Micro practice makes perfect? ; smaller, efficient doctors’ offices appeal to patients. USA TODAY Retrieved from

Practice management Q&A. (2011). Medical Economics, 88(7), 75-76. Retrieved from

Shinkman, R. (2014). TX physicians create cash-only practices for uninsured. Newton: Questex Media Group LLC. Retrieved from

SignatureMD; it’s time to redefine concierge medicine. (2013). Managed Care Weekly Digest, , 72. Retrieved from

Stucke, J. (2009, Oct 11). RURAL HOSPITALS SEEK RX FOR SURVIVAL; making medicine work in small towns takes the right doctor – and the right business model. Spokesman Review Retrieved from

Sweeney, J. F. (2011). Cutting EDGE. Medical Economics, 88(10), 22-24,31-32,34,36. Retrieved from

Wlazelek, A. (2008, Aug 09). Physician group eyes cash-only system ** fed up with insurers: Medical associates’ move would affect 100,000 valley patients. Morning Call Retrieved from



Interview Questions – Business Models for Family Medicine

  1. How many years have you been in family medicine?
  2. What type of business model is your practice, or that you work in (you can choose more than one answer here)?
  3. How many hours per week is your average work week?
  4. What is the average number of patients that you see in a day?
  5. What is the average amount of time that you spend per patient visit?
  6.  What is your daily stress level on a scale of “1-10″ (with “10″ being the most stressful)?
  7. What is your average annual gross income (the average of the last two years)?
  8. What are the pros of your business model?
  9. What are the cons of your business model?
  10. On a scale of “1 – 10″, what would you say is you level of happiness (with “10″ being the happiest)?
  11.  What recommendations and or advice do you have?



Polyphasic characterization and genetic relatedness of low-virulence and virulent Listeria monocytogenes isolates

Download PDF



Currently, food regulatory authorities consider all Listeria monocytogenes isolates as equally virulent. However, an increasing number of studies demonstrate extensive variations in virulence and pathogenicity of L. monocytogenes strains. Up to now, there is no comprehensive overview of the population genetic structure of L. monocytogenes taking into account virulence level. We have previously demonstrated that different low-virulence strains exhibit the same mutations in virulence genes suggesting that they could have common evolutionary pathways. New low-virulence strains were identified and assigned to phenotypic and genotypic Groups using cluster analysis. Pulsed-field gel electrophoresis, virulence gene sequencing and multi-locus sequence typing analyses were performed to study the genetic relatedness and the population structure between the studied low-virulence isolates and virulent strains.


These methods showed that low-virulence strains are widely distributed in the two major lineages, but some are also clustered according to their genetic mutations. These analyses showed that low-virulence strains initially grouped according to their lineage, then to their serotypes and after which, they lost their virulence suggesting a relatively recent emergence.


Loss of virulence in lineage II strains was related to point mutation in a few virulence genes (prfA, inlA, inlB, plcA). These strains thus form a tightly clustered, monophyletic group with limited diversity. In contrast, low-virulence strains of lineage I were more dispersed among the virulence strains and the origin of their loss of virulence has not been identified yet, even if some strains exhibited different mutations in prfA or inlA.


Listeria monocytogenes, a facultative intracellular pathogen, is one of the major causes of food-borne infection in humans [1]. Although rare, invasive listeriosis is a public health concern due mainly to its high fatality rate evaluated at 20-30% [2]. The clinical outcome of listeriosis is influenced by the pathogenic potential of the infecting strain which is in part related to its serotype [3]. It is now known that isolates 1/2a, 1/2b and 4b are responsible for 96% of human infections and most outbreaks are caused by strains of serotype 4b whereas serotype 1/2a has been associated with sporadic cases [4]. Serotypes 4a and 4c are predominant in animal, food or environment [5].

Unfortunately, there is currently no standard definition of virulence levels and no comprehensive overview of the evolution of L. monocytogenes strains taking into account the presence of low-virulence strains [5]. Different studies have shown that L. monocytogenes isolates form a structured population, composed of divergent lineages [6]. The large majority of isolates clusters into two lineages, but two additional lineages have been identified. However, these lineages correspond more to different but overlapping niches than to virulence-related clusters. We previously described low-virulence L. monocytogenes strains using a method that combines a plaque-forming (PF) assay with the subcutaneous (s.c.) inoculation of mice [3]. Using the results of cell infection assays and phospholipase activities, the low-virulence strains were assigned to one of four groups by cluster analysis. Sequencing of virulence-related genes highlighted the molecular causes of low virulence. Group I included strains that exhibited two different types of mutation in the prfA gene: either a single amino acid substitution, PrfAK220T, or a truncated PrfA, PrfAΔ174-237 [7]. In Group III, strains exhibited the same mutations in the plcA, inlA and inlB genes that lead to a lack of InlA protein, an absence of PI-PLC activity and a mutated InlB [8]. The fact that numerous strains exhibit the same substitutions in virulence genes suggests that they could have common evolutionary pathways. In contrast, Ragon et al. reported that numerous L. monocytogenes strains exhibit different mutations in the inlA gene due to convergent evolution [9]. These data emphasize the interest of providing a framework for the population study based on the virulence of this bacterium.

The aim of this study was to assign the new low-virulence strains identified by different methods to phenotypic and genotypic Groups using cluster analysis, and to study their relatedness with virulent Listeria monocytogenes strains using pulsed-field gel electrophoresis and multi-locus sequence typing analyses


Phenotypic characterisation of the low-virulence strains

The combination of PF assays followed by s.c. injections of immunocompetent mice, allowed us through different studies, to collect 43 low-virulence strains mainly of serotypes 1/2a (51%) and 4b (28%), which are usually related to sporadic and epidemic human cases of listeriosis, respectively [4] (Table ​(Table1).1). In this study, a strain is considered a low-virulence strain when fewer than 4 mice out of 5 inoculated become infected with a mean number of bacteria in the spleen less than 3.45 ± 0.77 log [3].

Table 1

Characterization of the low-virulence L. monocytogenes strains

As previously performed, these low-virulence strains were classified using an ascendant clustering hierarchical technique [3]. Six groups according to the values of four factors (level of cell invasion, number of plaques formed, and enzymatic activities of the two phospholipases C) have been obtained (Table ​(Table1).1). Group-I included 15 strains that did not enter cells, formed no plaques and had no phospholipase activity. Group-II consisted of only one strain entering cells, forming no plaques and only expressing PI-PLC activity. Group-III comprised nine strains entering cells, forming no plaques and only expressing PC-PLC activity. In this new analysis, the previously described Group-IV [7] has now been divided into 3 sub-Groups. The new Group-IV included nine strains forming plaques but fewer than virulent strains (mean 3 log versus 5). Three out of 9 strains were also characterized by a very low level of PC- and PI-PLC. The new Group-V comprised six strains also forming plaques but fewer than virulent strains and characterized by their very high PI-PLC activity. Finally, Group-VI contained three strains forming plaques within 48 h. In contrast the other strains formed plaques within 24 h, classic time necessary to count the plaque number.

Genotypic characterisation of the low-virulence strains

Sequencing the prfA, plcA, plcB, inlA and inlB genes allowed us to observe that some phenotypes correlate with genotypic mutations which have been demonstrated to be the cause of the low virulence (Table ​(Table1)1) [7]. The sequences of the PrfA, InlA and ActA fragment were compared to those of the EGDe strain (serotype 1/2 – GenBank accession number AL591824) or F2365 strain (serotype 4 – GenBank accession number AE017262), according to the serotypes of the strains.

The phenotypic Group-I strains exhibited mutations in PrfA compared to the EGDe strain and were subdivised into 2 genotypic Groups: the PrfAK220T (genotypic Group-Ia) and the truncated PrfAΔ174-237 (genotypic Group-Ib) previously described [8,11]. One strain (NP26) exhibited a new putative causal mutation in prfA, K130Q, and is the only one of serotype 4b exhibiting a PrfA mutation (herein defined as genotypic Group-Ic).

Two genotypic Groups were also identified for the phenotypic Group-III strains. One harbored exactly the same mutations in the plcA, inlA and inlB genes, characteristic of the previously genotypic Group-IIIa [8]. Only one strain (AF105) belonged to Group-IIIb and harbored a mutation at least in the inlA gene.

No genotyping Group has been defined for the phenotypic Groups-II because this Group is formed by only one strain. The Group-IV, -V and –VI strains did not exhibit specific DNA sequence of the prfA, inlA and actA fragment genes, that allowed us to assign genotyping Groups. No causal mutations could have been displayed explaining the low virulence of these Groups.

PFGE profiles

To study the genetic relationships between the low-virulence strains, the 43 low-virulence strains were compared with 49 virulent strains (based on both the mouse s.c. inoculation and PF assays) selected on the basis of matching serotypes and origins (Additional file 1).

This analysis revealed three major branches (Figure ​(Figure1)1) probably corresponding to the lineages I, II and IV described by Ward et al. by a SNP analysis [12]. In their study lineages I and III isolates formed, indeed, a sister group to lineage II strains, while the lineage IV represented a divergent sister clade. However, the small number of lineage IV strains did not allow us to conclude in this distribution. Nonetheless, as observed by Ward et al., lineage I included strains of serotype 1/2b, 4b, 4d, 4e, 3b and 7, whereas lineage II included strains of serotype 1/2a, 1/2c and 3a. Lineage III and IV included strains of serotype 4a, 4b and 4c. PFGE typing of the 92 isolates resulted in 69 different patterns, most of them grouped into 16 clusters with a similarity percentage above 85%. All strains gave interpretable PFGE patterns after restriction by AscI enzyme, whereas three virulent strains of lineage III/IV (serotype 4a and 4c) gave no profiles after ApaI restriction, possibly due to the methylation of restriction sites [13,14].

Figure 1

Dendrogram constructed for PFGE analysis using the UPGMA method with BioNumerics v.4.6 software showing the genetic relationships between 92 L. monocytogenes strains. The low-virulence strains are in red. Green lines indicate the division into

No clear correlation could be made between the PFGE clusters and the virulence levels of the strains and even though seven clusters included only virulent strains, the low-virulence strains were distributed in 9 clusters out of 16 (indicated by green lines in Figure ​Figure1),1), often mixed with virulent strains. Within the same lineage, the low-virulence strains were clustered according to their serotype. This observation is supported by the fact that strain NP26 belongs to the phenotypic Group-I which was grouped in lineage I with serotype 4b strains, whereas all the other strains of the phenotypic Group-I were grouped in lineage II with serotype 1/2a strains.

In the lineage II, the low-virulence strains were grouped according to their genotyping Groups, but were sometimes clustered with virulent strains. Only strains of the genotypic Group-Ia formed one specific cluster. All strains of the genotypic Group-IIIa were grouped together, but on the same branch as strain A23 (similarity percentage >80%). This clustering can be explained by the demonstration that the A23 strain had the same genotypic mutations as the Group-IIIa strains, but exhibited some virulence in our in vivo and in vitro virulence tests [15]. In the same way, all strains of the genotypic Group-Ib belonged to the same cluster, but with two virulent strains.

In the lineage I, the phenotypic Groups-IV, -V and -VI did not form specific clusters but were mixed with virulent strains (Figure ​(Figure1).1). This is probably related to the absence of a genotypic Group and probably corresponds to multiple genomic backgrounds. No low-virulence strain was found in lineage III/IV, but the small number of strains in this lineage hampered us to conclude in the rate of low-virulence strains.

Sequencing of virulence and housekeeping genes

To investigate the population structure and diversity of the low-virulence strains compared to virulent strains, three virulence genes were sequenced (prfA, inlA and actA) as well as seven housekeeping genes (acbZ, bglA, cat, dapE, dat, ldh, and lhkA). The dendrograms of the concatenated nucleotide sequences of virulence and housekeeping genes performed with the NJ method were presented Figure ​Figure2A2A and ​and2B,2B, respectively. They showed different relationships among lineages and in part for some lineage I low-virulence strains. In the housekeeping-gene tree, lineage III/IV strains formed a sister group to lineage I isolates as previously described [16]. However, as also observed by Tsai et al.[16], this was not the case with the virulence-gene tree where the strains of serotype 4a and 4c formed different branches. In the same way, all strains of serotype 4b were on the same branch in the housekeeping-gene tree. That was not the case in the virulence-gene tree where few strains of serotype 4b were on the same branch as strains of serotype 1/2b and 3b. Similar variations were observed for strains of serotype 1/2a which were on the same branch in the housekeeping-gene tree, whereas with the virulence-gene tree, 7 strains were on different branches than the other 34 serotype 1/2a strains (bootstrap 100%). This observation comforted the hypothesis that numerous recombinations have occurred with the virulence genes.

Figure 2

A Dendrogram of the prfA, actA and inlA gene sequencing using the NJ method with BioNumerics v.4.6 software showing the genetic relationships between 92 L. monocytogenes strains. The tree was constructed on the basis of the mean matrix distances

Similar variations between the two trees were also observed for low-virulence strains of lineage I. For example, with the virulence-gene tree 2 low-virulence strains of serotype 4b and 2 of serotype 4d were on the same branch as virulent strains of serotype 1/2b, 3b, and 7. This is not the case for the housekeeping-gene tree. As observed with PFGE, for the lineage II, both trees suggested that i) all the low-virulence strains of the same genotyping Group are on the same branch, and ii) the genotypic Group-Ia was closer to the genotypic Group-IIIa than to the genotypic Group-Ib. In lineage I, the low-virulence strains of phenotypic Groups-IV, -V and -VI were, in contrast, mixed with virulent strains showing that evolution of their virulence genes had occurred independently. This is also related to the fact that no genotyping group has been detected for these lineage I strains.

Twenty-six out of the 43 low-virulence strains (60%) and 11 out of the 49 virulent strains (22%) had a truncated InlA protein (Table ​(Table2),2), grouped in only 7 ST. Remarkably, all low-virulence strains of lineage II had a truncated InlA protein, compared to only three out of 18 low-virulence strains of lineage I. In addition, a correlation exists between the genotyping Groups and inlA mutations. All strains of the genotypic Group-Ia harboring the PrfAK220T mutation exhibited the inlA mutation at codon 77. Similarly, all strains of the genotypic Group-Ib harboring the PrfAΔ174-237 mutation exhibited a stop-codon at codon 189, and all strains of genotypic Group-IIIa had an insertion after the codon 13, leading to a truncated InlA.

Table 2

Mutational events in the inlA gene

MSTree analysis

To analyze in greater detail the population structure of the low-virulence strains, the 92 strains were analyzed and compared with the 656 L. monocytogenes isolates included in a previous study [18]. As no low-virulence strain was found in lineage III/IV, we presented only the lineages I and II.

This analysis showed that low-virulence strains of genotypic Group-Ia, -Ib, and -IIIa were distributed among three specific closely related STs (13, 31, 193) (Figure ​(Figure3).3). The ST 13 was formed with 10 Group-Ia low-virulence strains and one strain (Lm74905) belonging to the comparative set (in white). The analysis of this strain revealed that it exhibited the PrfAK220T mutation and the same truncated InlA characterizing the genotypic Group-Ia. Likewise, the Lm85820 strain which grouped in the ST31 (in white) exhibited the same mutation in InlA than the low-virulence strains of this ST, but no mutation in PfrA. Remarkably, although all strains of the ST31 had InlA mutations, only half of these strains also had the PrfAΔ174-237 mutation. In this analysis, the A23 strain corresponds to a singleton (ST196) with only one mismatch with Group-IIIa and two with Group-Ia. It is related to Group-Ib through ST11.

Figure 3

Minimum spanning tree based on allelic profiles by using BioNumerics version 4.6. (Applied-Maths, Sint-Martens-Latem, Belgium). The comparative set included 656 L. monocytogenes strains from the French Reference Centre for Listeria and the WHO

Overall, half of the low-virulence strains (22 out of 43), belonging to the genotyping Groups-Ia, -Ib and -IIIa, are likely to have descended from a single virulent 1/2a ancestral bacterium. In contrast, the other strains were distributed into five clonal complexes and 10 STs and may be regarded as virulence variants of L. monocytogenes strains.

Contribution of the optical mapping

To investigate the genomic relationship between the A23 strain and the closely related low-virulence strains belonging to Group-IIIa strains, two strains (BO43 and 416) were compared with the A23 strain using optical mapping and the in silico reference EGDe map (Figure ​(Figure4).4). The EGDe optical map was approximately 20% different from the maps of the Group-IIIa and A23 strains, whereas the A23 strain showed 99% similarities with Group-IIIa. Two fragments (3 and 4) (63 and 47 kb, respectively) had been inserted in the chromosome of the A23 strain but not in the EGDe strain. Fragments 5, 6 and 7 (52, 50 and 41 kb, respectively) represent the fragments inserted in the chromosomes of the BO43 and 416 strains. A supplementary fragment 8 (125 kb) was inserted in the chromosome of the BO43 strain.

Figure 4

Aligned optical maps for Group-III (BO34, 416) and A23 strains and in silico reference EGDe map. In the pair-wise alignments, lines connecting two chromosomal maps indicate a discontinuity in the alignment of fragments. Chromosomal inversions are indicated

This analysis confirms that all the Group-IIIa strains are very similar to each other and to the A23 strain. Indeed the insertion of the fragment 4 is located at the same place as the fragment 7 and could be inserted in the region of the lmo2589 gene annotated as similar to a transcription regulator T and R / AcrR family. The fragment 3 present in the A23 strain is different from the fragment 5, present in the Group III strains and could explain the increase of virulence of the A23 strain. The fragment 3 could be inserted in the region of the lmo2073 gene annotated as similar to ABC transporter and the region of the lmo2074 gene (similar to unknown proteins). The fragment 5 could be inserted in the region of the lmo2105 gene, annotated as similar to ferrous iron transport protein B. The fragment 6 present in the Group III strains could explain the decrease of virulence of these strains compared to the A23 strain. Indeed the annotation of the EGDe strain indicates that this insertion was found in the lmo2467 gene, located upstream of the clpP gene and its promoter, involved in the rapid and adaptive response of intracellular pathogens during the infectious process [19].


For a long time, all L. monocytogenes isolates were regarded as strictly pathogenic at the species level, and were always related to disease. However, from the experimental data collected over recent years, it has become clear that L. monocytogenes demonstrates serotype/strain variations in virulence and pathogenicity rate [5]. The population structure of 43 low-virulence strains was investigated with that of 49 virulent strains to estimate their diversity from virulent strains. We also investigated whether low-virulence strains formed a homogeneous subpopulation of L. monocytogenes or whether they originated from a random loss of virulence genes and thus diversified in multiple distinct directions.

We based our analysis on PFGE and different DNA-sequence-based approaches. The PFGE gave the greatest discriminatory power. Indeed PFGE gave profiles for different strains that by another way were grouped together in MSTrees. For example, ST2 (Figure ​(Figure3)3) comprised low-virulence strains of the phenotypic Groups-I, -V, and -VI, which had different PFGE profiles. Similarly, the low-virulence strains AF105 and LSEA-99-23 exhibited the same MLST profile but had distinct profiles in PFGE. Interestingly, MSTree identified specific ST for half of the low-virulence strains belonging to lineage II.

Overall, we identified low-virulence L. monocytogenes strains in both lineages I and II. No hypothesis could be advanced for the lineage III/IV, as they were few strains studied here represented these lineages. Our population structure showed that low-virulence strains are linked firstly according to their lineage, then to their serotypes and after which, they lost their virulence suggesting a relatively recent emergence. MSTree analyses showed that low-virulence strains belonging to lineage II formed a tightly clustered, monophyletic group with limited diversity, in contrast to the low-virulence strains of lineage I. All our observations further supported the fact that some correlations existed between virulence level and point mutations, base substitutions inducing a stop-codon, or inactivation of different virulence proteins, rather than on horizontal transfer or gene loss [7,8,20]. A characteristic of lineage II low-virulence strains was that all strains had a point mutation in the virulence inlA gene. Interestingly, there was a strong correlation between the inlA mutation and the genotypic group which were based on the mutations responsible for the virulence lost. Moreover, all strains of ST31 had only two different inlA mutations, but only the strains with the mutation type 5, according to Van Stelten also have the PrfAK220T mutation [17]. This observation suggested that the inlA mutation appeared before the prfA mutation. Regardless of the nature of mutations in inlA in the different low-virulence strains, there was clearly a link between their prevalence in food environments and the inlA mutations. Indeed, the inlA mutations were identified mainly in serotypes 1/2a and 1/2c from lineage II isolated from food and food-processing environments [17,21]. As such, it is reasonable to hypothesize that variations within these groups have been shaped to a greater extent by selective constraints operating in food manufacturing-plants.

It is intriguing that InlA, and to a lesser extent PrfA, which are important bacterial factors for host colonization, were lost. This pattern could be explained either by relaxation of the selective constraint to maintain InlA and PrfA function or by a selective advantage provided by the loss of functional virulence proteins in the ecological niche occupied by these strains. Clonal families might be adapted to different niches, and their occurrence as mammalian pathogens may be of limited significance for their evolutionary success in the long term. Considering all altered factors, the low-virulence strains could represent over 50% of the L. monocytogenes strains [5]. The fact that the growth of some low-virulence L. monocytogenes strains was impaired on selective medium suggests that the prevalence of these strains may be higher than that currently reported [22]. Moreover, only a few L. monocytogenes strains isolated from the environment and/or food have been analyzed, in contrast to strains of human origin. Developing reliable and easy-to-perform virulence tests could be useful, particularly for risk analysis, where it is important to evaluate the risk associated with the consumption of food products contaminated with L. monocytogenes not only on the basis of levels of bacterial contamination but also on the virulence level of the strains.

In this complex diversity scheme, the case of the A23 strain is very intriguing. Indeed, it is still virulent in mice, despite non-functional major virulence genes, due to point mutations in inlA, inlB and plcA that characterize the genotypic Group-IIIa [15]. This strain was found to be in the same cluster as the Group-IIIa strains using PFGE and MLST analyses, but to be in a specific ST using MSTree (ST 196 and 193, respectively). The fact that this strain has an additional mutation in mpl compared to Group-IIIa strains [15] suggests that it evolved from this group and thus reacquired virulence genes after initial virulence-gene loss. However, optical mapping does not support this hypothesis, since compared to the EGDe genome, specific fragments have been inserted in the genome of the Group-IIIa strains but not in strain A23, suggesting that the Group-IIIa strains have evolved from the latter. The complete sequencing of the genome of these strains should clarify this question.

This analysis corroborated the classification obtained for the phenotypic Groups-I and –III. Moreover the new detected low-virulence strains exhibiting the same phenotypes and harbouring the same mutations in the virulence genes, as previously observed, reinforced our observations. The new results allowed us to subdivide the former Group-IV into 3 new Group-IV, -V and –VI and to suggest different hypothesis concerning the population structure and diversity of the low-virulence strains compared to virulent strains.


The data presented in the present study show indeed that the diversity and population structure according to the virulence level of L. monocytogenes strains is complex and based on different mechanisms which seem to differ according to the lineage of the strains and thus to their ecological niches. However, from a practical perspective, this strain population does not correspond to a new species within Listeria. The relatively clear differences between virulent and non-virulent strains or species make these bacteria an attractive model for examining the lost of pathogenicity in this genus and for applying these principles to logical predictions of how certain pathogens will behave in a population over evolutionary time.


Strains and culture conditions

The 92 L. monocytogenes strains used in this study are described in the Additional file 1. The non-virulent L. innocua BUG499 strain was used as negative reference. All isolates were collected from independent sources at different dates. L. monocytogenes strains were defined as virulent or low-virulence using a virulence test combining a PF assay performed with the human colon adenocarcinoma cell line HT-29 and subcutaneous inoculation of mice into the hind footpads of immunocompetent Swiss mice as previously described [3]. Animal experiments were carried out in strict accordance with French recommendations. The protocol was approved by the Val de Loire Ethics Committee for Animal Experiments (n° 2011-07-02). For analysis, strains were cultured for 8 h in brain-heart infusion broth (Becton Dickinson, Fisher, Illkirch, France) at 37°C.

The collection of 656 L. monocytogenes strains from the French Reference Centre for Listeria and the WHO Collaborative Centre for Foodborne Listeriosis were used for the minimum spanning tree (MSTree) (comparative set; Figure ​Figure3)3) as previously described [9,18].

Phenotypic characterization of the low-virulence strains

The PF assay performed on HT-29 cells and invasion assays performed on Caco-2 and Vero cells were previously described [8]. The detection of the PI-PLC activity assays were analyzed in the culture supernatant with tritium-labelled L-α- phosphatidyl-inositol [8] and the PC-PLC activity was assessed after incubating with lecithin suspension, at 510 nm [7]. Experiments were carried out in duplicate and repeated twice for each strain. The values obtained allowed us to perform an agglomerative hierarchical clustering, based on Ward’s method and the Euclidean distance, to identify groups (clusters).

Pulsed-Field Gel electrophoresis (PFGE)

The PFGE protocol used in this study was the PulseNet standardized molecular subtyping protocol in accordance with Graves and Swaminathan [23].

The gels were photographed under UV transillumination, and the images were digitized and analyzed using BioNumerics v4.6 software (Applied-Maths, Sint-Martens-Latem, Belgium). The matching of band patterns was based on the DICE coefficient. Dendrograms were created using the Unweighted Pair Group Method with arithmetic mean. Strains were considered to be indistinguishable and were assigned to the same PFGE profile when the dendrogram indicated an index of relatedness of 100% verified by visual examination of band patterns.

Gene sequencing and multi-locus sequence typing (MLST)

The nucleotide sequencing of prfA, inlA, inlB and plcA genes and sequence analyses were described previously [7,8]. The clpP gene and its flanking regions (lmo2467 and lmo2469) were amplified from total isolated DNA using PCR. Primers and temperature annealing are listed in the Additional file 2.

The prfA and inlA virulence genes were fully sequenced, whereas the actA gene was partially sequenced. Seven housekeeping genes (acbZ, bglA, cat, dapE, dat, ldh, and lhkA) were selected for the MLST analyses (Additional file 2: Table S2) [9]. Alleles and sequence types (ST) are freely available at For analyses, sequences were concatenated either for the virulence or the housekeeping genes in an MLST scheme. For each MLST locus, including the 748 L. monocytogenes strains, an allele number was given to each distinct sequence variant. MLST analysis links profiles so that the sum of the distances (number of distinct alleles between two profiles) is minimized [24]. Each circle represented in Figure ​Figure33 corresponds to a ST number, attributed to each distinct combination of alleles on the seven genes. The size of the circle corresponds to the number of strains with that particular profile.

The dendrograms of the concatenated nucleotide sequences of virulence and housekeeping genes with the Neighbor-Joining (NJ) method and MLST analysis were performed using BioNumerics v4.6.

Optical mapping

Optical maps were prepared on the Argus™ Optical Mapping System by OpGen (Gaithersburg, MD USA), as described previously [25]. This method scans and assesses the architecture of complete bacterial genomes. Briefly, following cell lysis, genomic DNA molecules were spread and immobilized onto derivatized glass slides and digested by NcoI. After restriction digestion, a small gap in the DNA at the precise location of the restriction endonuclease cleavage site is left. The DNA digests were stained with YOYO-1 fluorescent dye, and photographed with a fluorescence microscope interfaced with a digital camera. Automated image-analysis software located and sized fragments, based on YOYO-1 binding and assembled multiple scans, into whole-chromosome optical maps. The average size of each restriction fragment (measured in 30–100 different molecules in the assembly) was determined and used to create a linear “consensus map” on which each restriction site is represented by a vertical line.

Nucleotide sequences

The DNA sequences of the MLST loci have been deposited in GenBank under accession numbers EU294615-EU294706 (abcZ), EU294707-EU294797 (bglA), EU294798-EU294889 (cat), EU294890-EU294981 (dapE), EU294982-EU295073 (dat), (EU295074-EU295165 (ldh), EU295166-EU295257 (lhkA), EU294523-EU294614 (prfA), EU295258-EU295336 (actA), and EU295337-EU295423 (inlA).

Competing interests

The authors declare that they have no competing interests.

Authors’ contributions

OG and ST carried out the molecular genetic studies, participated in the sequence alignment. AK carried out the PFGE analysis. MR and AL carried out the MLST analysis. SMR carried out the phenotypic studies. BS performed the statistical analysis. GK carried out the optical mapping. LM and ALM participated in the design of the study. PhV and SMR conceived of the study, and participated in its design and coordination, helped to draft the manuscript. All authors read and approved the final manuscript.

Supplementary Material

Additional file 1:

Describes theListeria strains used in this article[[7],[8],[10],[15],[26-30]].

Additional file 2:

Describes the primers used for the amplification and sequencing of the housekeeping genesabcZ,bglA,dapE, dta, kat,ldh and lhkAand the virulence genes prfA, actAandinlA. The primers used for the verification of an inserted fragment in the “clpP” region have been also given.


This study was supported by grants from the Conseil Régional du Centre and the Ministère de l’Agriculture et de la Forêt, by Institut Pasteur (Paris, France), and the Institut de Veille Sanitaire (Saint-Maurice, France). It was also funded by an INRA food research programme. S. Témoin holds a Doctoral fellowship from the Région Centre and the Institut National de Recherche Agronomique.


  • Westrell T, Ciampa N, Boelaert F, Helwigh B, Korsgaard H, Chriel M, Ammon A, Makela P. Zoonotic infections in Europe in 2007: a summary of the EFSA-ECDC annual report. Euro Surveill. 2009;14(3):1–3. [PubMed]
  • Rocourt J, Hogue A, Toyofuku H, Jacquet C, Schlundt J. Listeria and listeriosis: Risk assessment as a new tool to unravel a multifaceted problem. Am J Infect Control. 2001;29(4):225–227. doi: 10.1067/mic.2001.115681. [PubMed] [Cross Ref]
  • Roche SM, Velge P, Bottreau E, Durier C, Marquet-van der Mee N, Pardon P. Assessment of the virulence of Listeria monocytogenes: agreement between a plaque-forming assay with HT-29 cells and infection of immunocompetent mice. Int J Food Microbiol. 2001;68(1–2):33–44. [PubMed]
  • Swaminathan B, Gerner-Smidt P. The epidemiology of human listeriosis. Microb Infect. 2007;9(10):1236–1243. doi: 10.1016/j.micinf.2007.05.011. [PubMed] [Cross Ref]
  • Velge P, Roche SM. Variability of Listeria monocytogenes virulence: a result of the evolution between saprophytism and virulence? Futur Microbiol. 2010;5(12):1799–1821. doi: 10.2217/fmb.10.134. [PubMed] [Cross Ref]
  • Orsi RH, den Bakker HC, Wiedmann M. Listeria monocytogenes lineages: Genomics, evolution, ecology, and phenotypic characteristics. Int J Med Microbiol. 2011;301(2):79–96. doi: 10.1016/j.ijmm.2010.05.002. [PubMed] [Cross Ref]
  • Roche SM, Gracieux P, Milohanic E, Albert I, Virlogeux-Payant I, Temoin S, Grepinet O, Kerouanton A, Jacquet C, Cossart P. et al. Investigation of specific substitutions in virulence genes characterizing phenotypic groups of low-virulence field strains of Listeria monocytogenes. Appl Environ Microbiol. 2005;71(10):6039–6048. doi: 10.1128/AEM.71.10.6039-6048.2005. [PMC free article] [PubMed] [Cross Ref]
  • Temoin S, Roche SM, Grepinet O, Fardini Y, Velge P. Multiple point mutations in virulence genes explain the low virulence of Listeria monocytogenes field strains. Microbiology. 2008;154(Pt 3):939–948. [PubMed]
  • Ragon M, Wirth T, Hollandt F, Lavenir R, Lecuit M, Le Monnier A, Brisse S. A new perspective on Listeria monocytogenes evolution. PLoS Pathog. 2008;4(9):e1000146. doi: 10.1371/journal.ppat.1000146. [PMC free article] [PubMed] [Cross Ref]
  • Kerouanton A, Roche SM, Marault M, Velge P, Pourcher AM, Brisabois A, Federighi M, Garrec N. Characterization of isolates of Listeria monocytogenes from sludge using pulsed-field gel electrophoresis and virulence assays. J Appl Microbiol. 2010;108(4):1380–1388. doi: 10.1111/j.1365-2672.2009.04531.x. [PubMed] [Cross Ref]
  • Velge P, Herler M, Johansson J, Roche SM, Temoin S, Fedorov AA, Gracieux P, Almo SC, Goebel W, Cossart P. A naturally occurring mutation K220T in the pleiotropic activator PrfA of Listeria monocytogenes results in a loss of virulence due to decreasing DNA-binding affinity. Microbiology. 2007;153(Pt 4):995–1005. [PubMed]
  • Ward TJ, Ducey TF, Usgaard T, Dunn KA, Bielawski JP. Multilocus genotyping assays for single nucleotide polymorphism-based subtyping of Listeria monocytogenes isolates. Appl Environ Microbiol. 2008;74(24):7629–7642. doi: 10.1128/AEM.01127-08. [PMC free article] [PubMed] [Cross Ref]
  • Zheng W, Kathariou S. Differentiation of epidemic-associated strains of Listeria monocytogenes by restriction fragment length polymorphism in a gene region essential for growth at low temperatures (4 degrees c) Appl Environ Microbiol. 1995;61(12):4310–4314. [PMC free article] [PubMed]
  • Yildirim S, Lin W, Hitchins AD, Jaykus LA, Altermann E, Klaenhammer TR, Kathariou S. Epidemic clone I-specific genetic markers in strains of Listeria monocytogenes serotype 4b from foods. Appl Environ Microbiol. 2004;70(7):4158–4164. doi: 10.1128/AEM.70.7.4158-4164.2004. [PMC free article] [PubMed] [Cross Ref]
  • Roche SM, Grepinet O, Corde Y, Teixeira AP, Kerouanton A, Temoin S, Mereghetti L, Brisabois A, Velge P. A Listeria monocytogenes strain is still virulent despite nonfunctional major virulence genes. J Infect Dis. 2009;200(12):1944–1948. doi: 10.1086/648402. [PubMed] [Cross Ref]
  • Tsai YH, Maron SB, McGann P, Nightingale KK, Wiedmann M, Orsi RH. Recombination and positive selection contributed to the evolution of Listeria monocytogenes lineages III and IV, two distinct and well supported uncommon L. monocytogenes lineages. Infect Genet Evol. 2011;11(8):1881–1890. doi: 10.1016/j.meegid.2011.08.001. [PMC free article] [PubMed] [Cross Ref]
  • Van Stelten A, Simpson JM, Ward TJ, Nightingale KK. Revelation by single-nucleotide polymorphism genotyping that mutations leading to a premature stop codon in InlA are common among Listeria monocytogenes isolates from ready-to-eat foods but not human listeriosis cases. Appl Environ Microbiol. 2010;76(9):2783–2790. doi: 10.1128/AEM.02651-09. [PMC free article] [PubMed] [Cross Ref]
  • Chenal-Francisque V, Lopez J, Cantinelli T, Caro V, Tran C, Leclercq A, Lecuit M, Brisse S. Worldwide distribution of major clones of Listeria monocytogenes. Emerg Infect Dis. 2011;17(6):1110–1112. doi: 10.3201/eid1706.101778. [PMC free article] [PubMed] [Cross Ref]
  • Gaillot O, Pellegrini E, Bregenholt S, Nair S, Berche P. The ClpP serine protease is essential for the intracellular parasitism and virulence of Listeria monocytogenes. Mol Microbiol. 2000;35(6):1286–1294. [PubMed]
  • Jacquet C, Gouin E, Jeannel D, Cossart P, Rocourt J. Expression of ActA, Ami, InlB, and Listeriolysin O in Listeria monocytogenes of human and food origin. Appl Environ Microbiol. 2002;68(2):616–622. doi: 10.1128/AEM.68.2.616-622.2002. [PMC free article] [PubMed] [Cross Ref]
  • Nightingale KK, Ivy RA, Ho AJ, Fortes ED, Njaa BL, Peters RM, Wiedmann M. inlA premature stop codons are common among Listeria monocytogenes isolates from foods and yield virulence-attenuated strains that confer protection against fully virulent strains. Appl Environ Microbiol. 2008;74(21):6570–6583. doi: 10.1128/AEM.00997-08. [PMC free article] [PubMed] [Cross Ref]
  • Roche SM, Kerouanton A, Minet J, Le Monnier A, Brisabois A, Velge P. Prevalence of low-virulence Listeria monocytogenes strains from different foods and environments. Int J Food Microbiol. 2009;130(2):151–155. doi: 10.1016/j.ijfoodmicro.2009.01.012. [PubMed] [Cross Ref]
  • Graves LM, Swaminathan B. PulseNet standardized protocol for subtyping Listeria monocytogenes by macrorestriction and pulsed-field gel electrophoresis. Int J Food Microbiol. 2001;65(1–2):55–62. [PubMed]
  • Schouls LM, van der Heide HG, Vauterin L, Vauterin P, Mooi FR. Multiple-locus variable-number tandem repeat analysis of Dutch Bordetella pertussis strains reveals rapid genetic changes with clonal expansion during the late 1990s. J Bacteriol. 2004;186(16):5496–5505. doi: 10.1128/JB.186.16.5496-5505.2004. [PMC free article] [PubMed] [Cross Ref]
  • Cai W, Jing J, Irvin B, Ohler L, Rose E, Shizuya H, Kim UJ, Simon M, Anantharaman T, Mishra B. et al. High-resolution restriction maps of bacterial artificial chromosomes constructed by optical mapping. Proc Natl Acad Sci U S A. 1998;95(7):3390–3395. doi: 10.1073/pnas.95.7.3390. [PMC free article] [PubMed] [Cross Ref]
  • Glaser P, Frangeul L, Buchrieser C, Rusniok C, Amend A, Baquero F, Berche P, Bloecker H, Brandt P, Chakraborty T. et al. Comparative genomics of Listeria species. Science. 2001;294(5543):849–852. [PubMed]
  • Vicente MF, Mengaud J, Chenevert J, Perez-Diaz JC, Geoffroy C, Baquero F, Cossart P, Berche P. Reacquisition of virulence of haemolysin-negative Listeria monocytogenes mutants by complementation with a plasmid carrying the hlyA gene. Acta Microbiol Hung. 1989;36(2–3):199–203. [PubMed]
  • Mereghetti L, Roche SM, Lanotte P, Watt S, van der Mee-Marquet N, Velge P, Quentin R. Virulence and cord blood mononuclear cells cytokine production induced by perinatal Listeria monocytogenes strains from different phylogenetic lineages. Biol Neonate. 2004;86(1):66–72. doi: 10.1159/000077714. [PubMed] [Cross Ref]
  • Seeliger HPR. Listeriosis. New York: Hafner Publishing Co; 1961.
  • Bille J. In: Foodborne Listeriosis. Miller AJ, Smith JL, Somkuti GA, editor. Elsevier, New York: Society for industrial Microbiology; 1990. Epidemiology of human listeriosis in Europe, with special reference to the Swiss oubreak; pp. 71–74.

Articles from BMC Microbiology are provided here courtesy of BioMed Central

Helminth Communities of Owls (Strigiformes) Indicate Strong Biological and Ecological Differences from Birds of Prey (Accipitriformes and Falconiformes) in Southern Italy

Download PDF
Brock Fenton, Editor


We compared the helminth communities of 5 owl species from Calabria (Italy) and evaluated the effect of phylogenetic and ecological factors on community structure. Two host taxonomic scales were considered, i.e., owl species, and owls vs. birds of prey. The latter scale was dealt with by comparing the data here obtained with that of birds of prey from the same locality and with those published previously on owls and birds of prey from Galicia (Spain). A total of 19 helminth taxa were found in owls from Calabria. Statistical comparison showed only marginal differences between scops owls (Otus scops) and little owls (Athene noctua) and tawny owls (Strix aluco). It would indicate that all owl species are exposed to a common pool of ‘owl generalist’ helminth taxa, with quantitative differences being determined by differences in diet within a range of prey relatively narrow. In contrast, birds of prey from the same region exhibited strong differences because they feed on different and wider spectra of prey. In Calabria, owls can be separated as a whole from birds of prey with regard to the structure of their helminth communities while in Galicia helminths of owls represent a subset of those of birds of prey. This difference is related to the occurrence in Calabria, but not Galicia, of a pool of ‘owl specialist’ species. The wide geographical occurrence of these taxa suggest that local conditions may determine fundamental differences in the composition of local communities. Finally, in both Calabria and Galicia, helminth communities from owls were species-poor compared to those from sympatric birds of prey. However, birds of prey appear to share a greater pool of specific helmith taxa derived from cospeciation processes, and a greater potential exchange of parasites between them than with owls because of phylogenetic closeness.


In the last 30 years a number of papers on helminths of European owls (Strigiformes) have been published [1], [2], [3], [4], [5], [6], [7] including an exhaustive review of endoparasites found worldwide in raptors [8]. Most of those papers listed the helminth species identified and reported on basic statistical parameters of infection. When most than one owl species was studied from the same area, few attempts were made to investigate statistical differences between hosts and/or the factors influencing their helminth communities. Sanmartín et al. [7] concluded that in Galicia (northwest Spain) the helminth community of owls represented basically a “subset” of that observed in the birds of prey (Accipitriformes and Falconiformes) from the same region. This observation would agree with the observation that owls and birds of prey, although phylogenetically not closely related, have similar ecological niches and food habits, dividing the habitat not spatially but temporally [9]. Accordingly, their helminth faunas would be expected to be quite similar [10]. However, this prediction is at odds with the observed differences in composition of parasite faunas in geographical regions other than Galicia, i,e., Florida (USA) and Catalonia (northeast Spain), where a sizeable part of the faunas of each raptor group is not shared [6], [10], [11], [12], [13], [14]. These observations would therefore suggest that host specificity may play a contrasting role in structuring parasite communities in each raptor group depending on the geographical region.

Sanmartín et al. [7] also noted that helminth species richness of owl species was lower than that from birds of prey, and this result was considered unexpected given the similarity in hosts’ dietary spectrum. Ferrer et al. [6], [14] also indicated that, in Catalonia, owls exhibited lower diversity of helminths than birds of prey, and a similar conclusion can be derived from data by Kinsella et al. [10], [11], [12], [13] from Florida when values of helminth species richness are corrected for host sample size (MJ Kinsella, unpub. data). In attempting to account for these differences between raptor groups, Sanmartín et al. [7] suggested that a different explanation than feeding habits should be investigated. In this context, Kinsella et al. [10] pointed out body size as a potential determinant of helminth diversity among owl species; in fact, body size often correlates with key variables that affect transmission, i.e. host’s population density, rate of food intake or dietary breath ([15], [16] and references therein). The question is therefore whether body size could also help explaining the apparent differences in helminth richness between owls and birds of prey.

In southern Italy, Strigiformes include at least 7 species, displaying a wide variety of ecological and life-history patterns, including biological, environmental and dietary requirements [17]. In a recent published study from Calabria we found significant differences in both diversity and composition among helminths communities of 5 species of birds of prey [18]. Because several intrinsic and extrinsic factors including host age, sex, size, diet, habitat, behavior, migration, distribution and geographical range have all been recognized as variables influencing richness and diversity of parasite communities [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], we used a large sample of owl carcasses from southern Italy to evaluate the relative importance of the above mentioned variables on host biology, ecology and phylogeny on the structure of host-parasite associations at two taxonomical scales, i.e., between owl species, and between owls and birds of prey. The analysis benefits from the putative similarity in the regional pool of parasite species and the overall environmental characteristics from where owls and birds of prey were obtained.

Here we studied the helminth community of 5 owl species in Italy investigating the factors which may influence their community structure and comparing patterns of diversity and composition with those obtained among birds of prey from the same geographical area [18]. In addition, we evaluated overall differences in richness and the composition of helminth communities of owls and birds of prey, and compared the results with those previously obtained in Galicia [7]. The analyses were driven by the following research questions: (i) do the helminth communities of owl species from Calabria exhibit the same variability in composition and structure as that observed between birds of prey from the same region [18]? (ii) Are the helminth species from owls in Calabria “a subset” of the species found in birds of prey? (iii) Do owls have a lower diversity of helminth richness than birds of prey? (iv) What factors might account for the similarities and differences at each host’s taxonomical scale? And finally, do host body size play a role as an explaining factor?

Materials and Methods

Data Collection

A total of 122 owls that died between January 2004 and December 2011 at the Wildlife Rescue Centre in Rende, Cosenza (Calabria region), in southern Italy, were examined for helminth parasites. The birds belonging to 5 species of strigiforms including 30 little owls Athene noctua, 31 tawny owls Strix aluco, 41 barn owls Tyto alba, 10 long-eared owls Asio otus, and 10 scops owls Otus scops were all from the Calabria region. All owls included in the present study had a clinical course less than 7 days to minimize parasite losses; and no anthelmintic treatments were used in these birds [18], [27], [29].

All owl individuals were weighed prior to parasitological analysis. During necropsy examination, the trachea, lungs, air sacs, kidneys, spleen, liver, gallbladder, and the whole digestive tract of birds including oesophagus, stomach and intestines (duodenum, jejuno-ileum, ceca, and cloaca) were examined and helminths were collected, counted and identified following the techniques by Krone [30]. Worms were washed in saline solution and fixed in 70% ethanol; trematodes and cestodes were stained with Mayer’s acid carmine and mounted in Canada balsam, and nematodes and acanthocephalans were cleared in lactophenol on a glass slide for identification and then returned to the preservative. Voucher specimens are deposited in the U.S. National Parasite Collection, Beltsville, Maryland (Accession numbers: 105610 to 105625).

The whole pectoral muscles (depending by owl species approximately from 2 to 15 grams) and an aliquot of leg muscles (approximately from 2 to 5 grams) from each bird were examined for Trichinella spp. by artificial pepsin digestion [31].

Comparison between Owl Species

Total abundance, species richness, Brillouin’s index of diversity, and the Berger-Parker dominance index were used as overall descriptors of infracommunities. Total abundance is the number of individuals of all helminth species, and species richness the number of helminth species, harbored by each individual owl. The 95% confidence interval (CI) for prevalence was calculated with Sterne’s exact method [32], and for mean values of intensity, total abundance, species richness, Brillouin’s index, and Berger-Parker index, with the bias-corrected and accelerated bootstrap method using 20,000 replications [33].

For each owl species, differences of total abundance, species richness, Brillouin’s index, and Berger-Parker index between genders were compared with Mann–Whitney U- tests, respectively. These parameters were also compared between owl species with Kruskal-Wallis tests using post hoc comparisons [34]. Inferential statistics on compositional differences between owl species were carried out with a nonparametric analysis of similarities (ANOSIM) [35]. The number of individuals of each helminth species from each infracommunity was square-root transformed, and the Bray-Curtis similarity coefficient was calculated between individual hosts that harbored at least 1 helminth species. The ANOSIM ranks the Bray-Curtis similarity matrix and tests whether the ranks of similarities between and within owl species are the same on average. This is evaluated with the statistic R [35]. The null hypothesis was constructed by calculating 20,000 R values with random permutation on host individuals regardless of species. The overall comparison was followed by pairwise comparisons between host species. When significant differences were found, the Similarity Percentage (SIMPER) procedure was used for assessing which helminth taxa were primarily responsible for the observed differences between groups [35].

Compositional Differences of Helminth Faunas between Owls and Birds of Prey

Helminth data from the owl species analyzed in this study were compared with those obtained from 6 species of birds of prey examined by us in the same recovery centre between January 2000 and December 2010, i.e., 35 Eurasian buzzards Buteo buteo; 20 European sparrow hawks Accipiter nisus; 21 western honey buzzards Pernis apivorus; 17 marsh harriers Circus aeruginosus; 25 common kestrels Falco tinnunculus, and 17 peregrine falcons Falco peregrinus [18].

In Galicia, Sanmartín et al. [7] published infection data from 10 birds of prey species (110 Eurasian buzzards; 35 European sparrow hawks; 21 northern goshawks Accipiter gentilis; 12 common kestrels; 5 Montagu’s harrier Circus pygargus; 3 western honey buzzards, 3 Eurasian hobbies Falco subbuteo; 2 peregrine falcons; 1 red kite Milvus milvus and 1 black kite Milvus migrans) and 4 owl species (49 barn owls, 26 tawny owls, 9 little owls and 8 long-eared owls) that were collected in four recovery centers from 1991 to 1996. This dataset provide a unique opportunity to replicate the above comparisons between owls and birds of prey in another geographical region.

To examine compositional differences between owls and birds of prey, prevalence values for each helminth species scaled to unity were used to obtain a matrix of similarities between raptor species using the Bray-Curtis coefficient. The resulting matrix was used to perform a group average hierarchical cluster of owl and birds of prey species [18]. To examine for statistical evidence of genuine clusters among species, 20,000 random permutations of prevalence values were employed in the matrix [35]. The finding of statistically significant clustering could assist in interpreting whether phylogenetic, and/or ecological, relatedness between raptor species could influence the similarity between their helminth faunas.

To interpret differences in overall composition of helminth faunas of owls compared to birds of prey we derived a measure of specificity for each helminth species found in the samples of owls from both Calabria and Galicia based on records on each species worldwide. For each helminth species we checked all references for synonymies and looked for taxonomic updates, assuming that original identifications were correct. Then, we established the following categories: a helminth species was considered ‘specialist’ if it had been reported in single host species; ‘owl specialist’ if it had been reported mainly, or only, in owls (Strigiformes); ‘birds of prey specialist’, if it had been reported mainly, or only, in birds of prey (Accipitriformes and Falconiformes); ‘raptor generalist’, if it had extensively been reported in both owls and birds of prey, and ‘bird generalists’, if it also occurred extensively in birds other than raptors. The use of host-parasite lists may suffer from well-known problems of representativity ([36], and references therein), namely, records may equate common and rare species, and suitable and unsuitable hosts (i.e., nonhosts). Therefore, estimations of the degree of specificity are conservative. For each helminth species, data were also gathered about its geographical distribution and the intermediate/paratenic hosts, which may assist in interpreting patterns of specifity and geographical differences in helminth faunas, respectively.

Diversity differences of Helminth Communities between Owls and Birds of Prey

At the component community level (i.e., helminth communities for each host species considering the sample of hosts as a whole), we compared differences of species richness between owls and birds of prey from Calabria with an ANCOVA, considering the number of helminth taxa in each host species as the dependent variable, ‘raptor group’ as a fixed factor and ‘sample size’ (log10-transformed) as a covariate that correct for differences in sampling effort [22]. We firstly included the interaction term ‘raptor group×sample size’, but when it was not statistically significant, it was removed from the final model to increase the sensitivity of the analysis and to correctly interpret main effects [37]. The same ANCOVA analysis was carried out at infracommunity level, using data of mean species richness per host. Also, we investigated whether overall parasite recruitment differed between owls and birds of prey. Mean total abundance was discarded as an index of recruitment per host species because some small parasites (e.g., digeneans) were more numerous in birds of prey [18] and would strongly influence overall values. Instead, we calculated, for each host species, the median value of mean intensity of all parasites in the community since medians are very resistant to extreme values. This parameter was included as the dependent variable of an ANCOVA with the same predictors above.

We performed the same analyses described above with the data set from Galicia. For the comparison at the infracommunity analysis, Sanmartín et al. [7] only provided data for species with n≥8. Also, these authors did not provide values of mean species richness per host, but this value can easily be calculated for each host species as the sum of prevalences expressed on a per unit basis [38].

Effects of host body size upon community were explored as follows. Weight data from all raptor species included in the study were obtained from Snow et al. [39], and mean values for males and females throughout all seasons was averaged to obtain a single value per species. Then, for bird samples of both Calabria and Galicia, we examined whether residuals of component community richness, infracommunity richness, and the median value of mean intensity of all parasites in the community, corrected for host sample size, increased with host weight. One-tailed Spearman correlation tests were used. The package Primer v.6 [35] was used for the ANOSIM and cluster analyses, the free software Quantitative Parasitology v. 3 [40] to set 95% confidence intervals, and the statistical package SPSS v. 17 for the remaining analyses. Statistical significance was set at P<0.05.


Comparison between Owl Species

A total of 19 helminth taxa (10 nematodes, 3 acanthocephalans,3 cestodes and 3 digeneans) and 758 helminth individuals were found in the total sample of owls (Table 1). All helminth taxa were found in the gastrointestinal tract except for a single specimen of Excisa excisiformis which was collected from the trachea of 1 long-eared owl. No Trichinella infection was found by artificial pepsin digestion of muscular tissues. Gravid individuals were found in all helminth taxa regardless of owl species. The total species richness in the sample ranged from 2 (in the long-eared owl) to 12 (in the tawny owl) (Table 1). No helminth species was shared between the 5 owl species, but Centrorhynchus aluconis and Synhimantus affinis were shared between 4 owl species. Centrorhynchus aluconis was also the most frequent species in little owls, tawny owls and barn owls, whereas S. affinis was the most prevalent species in scops owls (Table 1). Four helminth species were shared between 3 owl species and, as many as 12 helminth species were restricted to single owl species (Table 1). However, this restriction was not coupled with high specificity because, within this group, only Paruterina candelabraria and Choanotaenia littoriae is specific to a single owl species (Table 2), and prevalences were low or very low for all helminth species of this group (range: 3.2–20%, see Table 1).

Table 1

Infection parameters of helminths found in 5 species of owls from Calabria (southern Italy).
Table 2

Classification of helminth taxa collected from owls in Calabria (C), southern Italy, and Galicia (G), northwest Spain, according to host specificity.

The proportion of individual hosts that were infected in the sample differed significantly among host species (Fisher test, p<0.003), ranging from 2 out of 10 (10%) in long-eared owls to 7 out of 10 (70%) in scops owls (Table 3). Infracommunity parameters for each host species are shown in Table 3. There were no significant differences in mean species richness (Kruskal-Wallis test, χ2 = 8.64, 4 d.f., p = 0.071), mean total abundance (χ2 = 7.11, 4 d.f., p = 0.130) and Brillouin’s diversity index (χ2 = 2.71, 4 d.f., p = 0.607) of helminths between owl species. The Berger-Parker index did not differ also between owl species (χ2 = 6.68, 4 d.f., p = 0.154) and the most abundant helminth species in infracommunities accounted for a very high proportion of total helminth abundance (mean Berger-Parker index >0.80 in all host species, Table 3). Centrorhynchus aluconis was numerically dominant in little owls, tawny owls and barn owls (in the latter shared with S. laticeps), whereas S. affinis was dominant in scops owls and Synhimantus laticeps in long-eared owls (shared with E. excisiformis) (Table 3). Mean similarity values of helminth infracommunities between owl species are shown in Table 4. Similarities ranged from 33.6% to 51.5%. Note that only 2 long-eared owls were infected (Table 3) and, therefore, comparisons with the other species are uncertain. Overall, helminth infracommunities of scops owls had the lowest similarity with those from the remaining species (Table 4). Statistical comparison of compositional differences between owl species (excluding the long-eared owl) revealed modest, but significant differences of composition among owl species (ANOSIM, R = 0.173, p = 0.0005). Two pairwise comparisons were found to be significant, namely, those involving scops owls and little owls (R = 0.402, p = 0.004), and scops owls and tawny owls (R = 0.338, p = 0.002); the comparison between scops owls and barn owls was close to significance (R = 0.162, p = 0.059). In the two significant comparisons, C. aluconis and S. affinis ranked as the first and second species providing dissimilarity between owl species according to The SIMPER procedure. Together, these 2 species accounted for 48.7% (scops owls vs. little owls) and 38.9% (scops owls vs. tawny owls) of mean dissimilarity. We found no statistically significant effects of host weight (Spearman correlation test, minimum nominal p = 0.145) or sex (Mann-Withney test, minimum nominal p = 0.183) on any 4 infracommunity parameters of Table 3 for any owl species. In the case of long-eared owls the tests involving Brillouin’s diversity index and Berge-Parker index could not be performed because only 2 hosts were infected (Table 3).

Table 3

Mean values (95% C.I.) of 4 parameters of gastrointestinal helminth communities in 5 owl species in Calabria (southern Italy).
Table 4

Matrix of mean values (with standard devistion in parentheses) of Bray-Curtis index of similarity (expresed as percentage) of helminth infracommunity composition between 5 owl species from the Calabria region, southern Italy.

Compositional Comparison of Helminth Communities between Owls and Birds of Prey

The group-average hierarchical cluster of raptor species based on prevalence of their helminth fauna is shown in Figure 1. In Calabria, a major significant subdivision (p = 0.0005) separated owls and birds of prey (Fig. 1A). However, in Galicia the cluster did not have any significant nodes, and species were not arranged according to the subdivision between owls and birds of prey.

Figure 1

Group-average hierarchical cluster analysis of helminth fauna from samples of birds of prey and owls in two geographical regions based on a bray-Curtis resemblance matrix using prevalence data scaled to unity.

Data about specificity of each helminth species are shown in Table 2. In Calabria, specificity could be established in 18 out of 19 helminth taxa, and they were distributed as follows: 1 species was classified as ‘specialist’; 6 as ‘owl specialists’, 5 as ‘raptor generalists’ and 6 as ‘bird generalists’. Species typical from owls (the two former categories) summed up 410 helminth individuals, or 54.1% of all helminth individuals found in the total sample of owls (see Table 1). In Galicia, 8 helminth species were reported, of which 1 species can be classified as ‘owl specialist’, 3 as ‘birds of prey specialists’, and 4 as ‘raptor generalists’. The single species typical from owls, P. candelabria, was found only in a single species (Table 2).

Diversity differences of Helminth Communities between Owls and Birds of Prey

At the component community level, the ANCOVA for species richness indicated that the interaction between host sample size and raptor group was significant neither in Calabria nor in Galicia and, therefore, interactions were removed from the models. Host sample size had an overall significant positive effect on species richness (Calabria: F(1,8) = 6.124, p = 0.038, Galicia: F(1,11) = 26.532, p<0.001); differences between raptor groups were also significant in both regions (Calabria: F(1,8) = 8.568, p = 0.019, Galicia: F(1,11) = 8.602, p = 0.014), with birds of prey having higher values of species richness in their helminth communities (Fig. 2 A, B). At infracommunity level, the ANCOVA for mean species richness also revealed that the interaction between host sample size and raptor group was not significant in either region. Also, there were no significant effects of host sample size (Calabria: F(1,8) = 0.001, p = 0.995, Galicia: F(1,5) = 2.694, p = 0.162), although sample size in Galicia was very low (Fig. 2 C, D). Concerning raptor group, helminth infracommunities from birds of prey in Calabria had higher values that those from owls (Fig. 2C) and the difference was significant (F(1,8) = 5.518, p = 0.045). In Galicia, this pattern was less marked (Fig. 2D) and the difference was not significant (F(1,5) = 1.771, p = 0.241).

Figure 2

Comparison of community parameters between species of birds of prey (solid dots) and owls (empty dots) in two geographical regions, Calabria, Italy (on the left) and Galicia, Spain (on the right).

The ANCOVA for the median values of mean intensity (MMI) also offered contrasting pattern between geographical regions. In Calabria, the interaction between host sample size and raptor group was significant (F(1,7) = 11.861, p = 0.011). Apparently, host sample size influenced MMI only in birds of prey (Fig. 2E). Ignoring the effects of host sample size, the comparison of MMI between raptor groups was significant (F(1,8) = 5.445, p = 0.048), and birds of prey tended to exhibit higher values of MMI. In Galicia, none of the predictors was significant (results not shown), and MMI was similar between owls and birds of prey (Fig. 2F). Host weight did not significantly correlate with host sample-size-corrected residuals of component community richness, mean infracommunity richness, and the median value of mean intensity of all parasites in the community either in Calabria or Galicia (Spearman correlation, all one-tailed p>>0.05).


Comparison Among Owls Species

Because most of the helminths in birds are acquired through the ingestion of their prey, the overall environment with its included habitats influencing the survival and potential transmission of a parasite species have been considered as the most important extrinsic determinants of pattern in helminth communities of avian hosts [19], [23], [24], [28], [41]. Host vagility, a broad host diet, and selective feeding by a host on prey that serve as intermediate hosts for a wide variety of helminths represent the main intrinsic determinants influencing their helminth communities [19], [23], [24], [41].

Ecological determinants are important when considering the similarities and differences of helminth communities between owls in Calabria. Three types of helminth species can be recognized, namely, species typical from owls (including an apparently very specific species, C. littoriae), species shared also with birds of prey, and generalist parasites common to other birds. It is not possible to determine whether all owl species are equally suitable hosts for each of these parasites, but patterns of specificity, and the absence of significant subdivisions of owls in the cluster analysis, strongly suggest that there are not fundamental barriers for exchange of helminth taxa among owl species. Therefore, factors driving the contact between owls and parasites [16], especially diet [15] are predicted to mainly account for the similarities and differences in their helminth faunas.

The owl species here studied are crepuscular and nocturnal feeders. According to Snow et al. [39], in the western Palaearctic scops owls feed mostly on insects and other invertebrates, whereas the remaining species rely more on small mammals and other vertebrates. However, each owl species can adjust their diet according to local availability, including a variable portion of birds, reptiles, amphibians and invertebrates [39], [42], [43]. Unfortunately no studies on feeding ecology of owls were available from Calabria, but results from studies in other regions of Italy largely conform to the general pattern described above, with scops owls feeding mainly on insects (orthopterans and moths) [44], barn owls, tawny owls and long-eared owls feeding more on small mammals [45], [46], [47], and little owls having a mixed diet of insects and small vertebrates [48].

Statistical comparison of helminth communities in owls from Calabria showed only marginal differences between scops owls and little owls and tawny owls. These differences are largely accounted for variability of infection levels of 2 helminth species which account for over 80% of total helminth abundance, i.e., C. aluconis (higher in little owls and tawny owls) and S. affinis (higher in scops owls). Centrorhynchus aluconis is known to use a wide range of micro-mammals and reptiles as paratenic hosts [49], [50], [51] in which the parasite accumulates. The life cycle of S. affinis is not known, but data from allied species indicates that insects and terrestrial isopods act as intermediate hosts [52] and lizards could act as paratenic hosts [53]. We therefore interpret that the largely insectivorous diet of scops owls would led them to recruit more individuals of S. affinis, and less of C. aluconis, compared to the other owl species.

The otherwise strong similarities in community structure of helminth communities of owls from Calabria are in contrast to the strong differences observed in birds of prey from the same region. Santoro et al. [18] interpreted that these differences resulted from diverse feeding habits among hosts (e.g., insectivory in western honey buzzards, ornithophagy in peregrine falcons, or a more catholic diet in Eurasian buzzards). Conversely, we submit that the small differences found in owls would indicate that all the studied species feed on a narrower range of prey, consuming different proportions of invertebrates, micro-mammals, and small vertebrates depending on both species and local availability. For example, in Greece, the barn owl preyed mainly on mammals, while birds and amphibians were only of local importance, and, accordingly, diet showed low diversity; the long-eared owl preyed mainly on mammals, but also took other prey (particularly birds and reptiles), having a more diverse diet. In contrast, the diet of the little owl was more variable, in two of the study areas the main prey were mammals but other prey involved resulted in relatively high diversity. In the other three areas the species took mainly insects, thus showing a more restricted diet based on small-sized prey [43]. In a study from Chile, Spain and California was observed that in Spain barn owl feed on significant amount of insects, reptiles and amphibians respect to those from Chile and California, and also the mean size of small mammals in its diet was considerably smaller than that from other two areas [42]. This was attributed to the reduced abundance of larger-sized small mammals in Spain, which presumably forces the barn owl to prey more heavily on the smallest mammals available and also on low-reward non-mammalian prey [42]. This suggests that owls may adapt their trophic requirements to the reduced prey occurring in a particular geographical area.

Compositional Differences of Helminth Faunas between Owls and Birds of Prey

Cluster analysis indicated that, in Calabria, owls can be separated as a whole from birds of prey with regard to the structure of their helminth communities; no further subdivisions among owl species were significant. This pattern results largely from the occurrence of ‘owl specialist’ species, which account for over 50% of total helminth abundance. It is also important to note that owls and birds of prey share just 4 of the 50 helminth taxa found in total (19 in owls and 31 in birds of prey) showing different infection levels; shared parasites included C. globocaudatus, C. falconis, S. laticeps, and B. fuscatum (see [18], [26]). The first 3 parasite species are very common in birds of prey from southern Italy, while in owls had lower prevalence and intensity; and only immature B. fuscatum were found in birds of prey and mature specimens in owls, respectively [18], [26].

Interestingly, ‘owl specialists’ are species shared only among owls, not just in Calabria, but apparently throughout their entire geographical distribution. For instance, the cestode P. candelabraria has extensively been reported only in owl species from Europe and North America (Table 2). This raises the question of what factors could produce these patterns of specificity. The encounter/compatibility paradigm [54], [55] states that specificity is determined by two sequential filters. The encounter filter prevents infections of potential hosts that cannot contact the parasite, whereas the compatibility filter excludes contacted hosts in which the parasite is unable to find the appropriate resources and/or escape or deter the host’s defences. The compatibility filter is directly associated to the history of co-adaptation between parasites and their hosts, and predicts that hosts that are phylogenetically related will tend to share parasites, among other factors, because they have a similar physiology [16].

Because of the lack of information, it is difficult to assess the role of the encounter and compatibility filters in shaping specificity of the ‘owl specialist’ helminths that were found in Calabria. Does, for instance, P. candelabraria use intermediate and paratenic hosts that are consumed only by owls and/or is it specialized for the microhabitat conditions provided by owls as hosts? We noted above that no dietary data exists for owls in Calabria but, in general, owls foraging at dusk and during the night are predicted to encounter only certain prey compared to birds of prey which are diurnal predators feeding generally on a wider spectrum of prey. There is only a subset of prey whose active times overlap with that of the both raptor groups, which are the ones more likely to be caught by both of them [56], [57]. Therefore, owls and birds of prey might share a limited number of prey species, constraining exchange of parasites. However, it is likely that some ‘owl-specialist’ species that contact non-owl hosts are also unable to establish and reproduce in them. For example, in North America, P. candelabraria have been reported in shrews, deer mice, voles and squirrels [58], which are regularly consumed by birds of prey [59] but none of them has been reported as a host for P. candelabraria, suggesting that the parasite cannot established in them.

Four out of the 5 owls species (barn, long-eared, tawny and little owls) included in the present study were also examined for helminths in Galicia [7]. Interestingly, ‘owl specialists’ were missing in this sample except for P. candelabraria, and owls essentially harboured a subset of the helminths found in birds of prey [7] (Table 2). This striking difference in composition can hardly be related to biogeographical factors because ‘owl specialist’ species have generally very wide geographical distributions (Table 2). Alternatively, compositional variability might be related to differences in the local pool of parasites [16], [24]. In support of hypothesis, of 27 total helminths found in owls from Calabria (19) and Galicia (8) just 3 were common in both localities (Table 2). In fact, local variability seems to be a common theme in other geographical areas. In Netherlands, for instance, Borgsteede et al. [5] analyzed 84 owls of 5 species (including barn, long-eared, tawny and little owls) and identified 12 helminth species excluding cestodes, of which only Porrocaecum spirale can be considered as an ‘owl specialist’ (Table 2). The role of local conditions, especially the availability of intermediate and paratenic hosts cannot be overestimated in accounting for these local differences [16].

Of the helminths species found here, the nematodes including Dispharynx spp., Excisa spp., Synhimantus spp., Skrjabinura spp., and Subulura spp. use a wide range of insects as intermediate hosts, and Capillaria falconis and Heterakis gallinarum use earthworms; Porrocaecum spp. use insectivorous mammals [52]; cestodes within Choanotaenia spp. use coleopterans and dipterans, Passerilepis spp. use insects, and P. candelabraria uses micro-mammals [60]; among digeneans Neodiplostomum spp. use amphibians, B. fuscatum uses terrestrial snails [61] and Skrjabinus spp. use terrestrial mollusks and arthropods [62]; acanthocephalans within Centrorhynchus spp. use orthopteran insects as intermediate hosts and mammals, reptiles and anurans as paratenic hosts [49], [50], [51] (Table 2).

Diversity Differences of Helminth Communities between Owls and Birds of Prey

The statistical comparison of diversity of helminth communities between owls and birds of prey assume that observations are independent. This is not the case because species within each bird group are related through phylogenetic relationships [15]. However, given the small sample of raptor species included in the study, our exploratory comparison was considered as a useful starting for future analyses that will include more raptor species and will explicitly control for phylogenetic effects ([15], and references therein). Currently, results indicate that the helminth fauna of owls from both Calabria and Galicia was less diverse than that from birds of prey in the same regions, thus statistically confirming conclusions previously obtained by Sanmartín et al. [7] for Galicia. The pattern could indeed be more general. In Catalonia, Ferrer et al. [6], [14] also observed that compared to birds of prey, owls had lower numbers of genera of helminths (14 vs 22), and generally lower prevalence rates among shared genera. Data from Kinsella et al. [10], [11], [12], [13] in Florida would also point to a similar conclusion when the effects of sampling effort are accounted for (MJ Kinsella, unpub. data). Overall, evidence would suggest that there are significant differences in the diversity of helmith faunas between owls and birds of prey regardless of the actual pool of species that can potentially infect host species in each region (see above). We therefore interpret that there might be a common factor producing this pattern.

A number of host-related factors have been put forward to account for differences in species richness of parasites among vertebrates, of which factors related to host body size often play a prominent role [15], [16], [23], [36]. Although our analyses should be interpreted with care because of small host sample sizes, host body size did not significantly correlate with species richness, neither at infracommunity nor at component community levels. Also, median intensity of helminths did not increase in large-bodied species, suggesting that the rate of parasite recruitment was not related to the amount of food consumed. Also, there was not an evident relationship between helminth infracommunity parameters and body size in owls (Table 3).

We suspect that other factors probably blur the expected influence of host body size upon helminth communities. One potential candidate is trophic niche breath [16]. Kinsella et al. [10] speculated that species richness in the helminth fauna of owls from Florida was primarily related to the variety of prey items consumed, with specialized feeders like barn owls and screech owls (Otus asio) harbouring fewer species than euryphagic species like barred owls (Strix varia). However, at a larger taxonomic scale, Sanmartín et al. [7] argued against a direct influence of diet because both owls and diurnal raptors share the same basic pool of prey in Galicia.

A factor missing in latter explanation is the influence of parasite specificity, which is much more apparent in birds from Calabria. In the previous section we pointed out that, regardless of contacts between parasites and hosts, the compatibility filter prevents some parasites from being established in certain hosts, but the filter should be more relaxed insofar as hosts species are phylogenetically closer. Accipitriformes plus Falconiformes represent a more speciose group than Strigiformes (ca. 58 vs. 19 spp., respectively, in the western Palaearctic, see Snow et al. [39]). It is therefore possible that birds of prey, as a group, harbor more specific helminth taxa than owls [8]. Also, the diversity of birds of prey generally outnumbers that of owls in any locality in Europe [39]. Following both arguments, we could expect, in any locality, that birds of prey share a greater pool of specific helmith taxa derived from cospeciation processes, and a greater exchange of parasites between them than with owls. The observation that both in Calabria and Galicia there are a number of helminth species shared between diurnal raptors with diverse trophic habits, but that do not occur in sympatric owls would lend support to this hypothesis. We urge researchers to develop specific analysis to test this hypothesis when more quantitative data about helminth communities from raptors are gathered in the future.


We thank the Centro Italiano Protezione Rapaci (CIPR) in Rende (Cosenza) for permission to use the owl carcasses, and the Istituto Zooprofilattico Sperimentale del Mezzogiorno, Section of Cosenza, for their collaboration in this study.

Funding Statement

This work was supported by Istituto Zooprofilattico Sperimentale del Mezzogiorno, Portici, Italy (project IZSME 06/10 RC). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.


1. Kutzer E, Frey H, Nöbauer H (1982) Zur parasitenfauna österreichischer eulenvögel (Strigiformes). Angew Parasitol 23: 190–197. [PubMed]
2. Ewald JA, Crompton DWT (1993) Centrorhynchus aluconis (Acanthocephala) and other helminth species in tawny owls (Strix aluco) in Great Britain. J Parasitol 79: 954–958. [PubMed]
3. Illescas-Gomez MP, Rodriguez-Osorio M, Aranda-Maza F (1993) Parasitation of falconiform, strigiform and passeriform (Corvidae) birds by helminths in Spain. Res Rev Parasitol 53: 129–135.
4. Sitko J (2001) Trematodes of owls (Aves: Strigiformes) in the Czech Republic. Acta Soc Zool Bohem 65: 45–46.
5. Borgsteede FHM, Okulewicz A, Zoun PEF, Okulewicz J (2003) The helminth fauna of birds of prey (Accipitriformes, Falconiformes, and Strigiformes) in the Netherlands. Acta Parasitol 48: 200–207.
6. Ferrer D, Molina R, Castella J, Kinsella JM (2004) Parasitic helminths in the digestive tract of six species of owls (Strigiformes) in Spain. Vet J 167: 181–185. [PubMed]
7. Sanmartín ML, Alvarez A, Barreiro G, Leiro J (2004) Helminth fauna of falconiform and strigiform birds of prey in Galicia, northwest Spain. Parasitol Res 92: 255–263. [PubMed]
8. Lacina D, Bird DM (2000) Endoparasites of raptors. A review and an update. In: Lumeij JT, Remple JD, Redig PT, Lierz M, Cooper JE editors. Raptor Biomedicine III. Lake Worth: Zoological Education Network Inc. 65–99.
9. Bosakowski T, Smith DG (1992) Comparative diets of sympatric nesting raptors in the eastern deciduous forest biome. Canadian J Zool 70: 984–992.
10. Kinsella JM, Foster GW, Forrester DJ (2001) Parasitic helminths of five species of owls from Florida, USA. Comp Parasitol 68: 130–134.
11. Kinsella JM, Foster GW, Forrester DJ (1995) Parasitic helminths of six species of hawks and falcons in Florida. J Raptor Res 19: 117–122.
12. Kinsella JM, Cole RA, Forrester DJ, Roderick CL (1996) Helminth parasites of the osprey, Pandion haliaetus, in North America. J Helminth Soc Wash 63: 262–265.
13. Kinsella JM, Foster GW, Cole RA, Forrester DJ (1998) Helminth parasites of the bald eagle, Haliaeetus leucocephalus, in Florida. J Helminth Soc Wash 65: 65–68.
14. Ferrer D, Molina R, Adelantado C, Kinsella JM (2004) Helminths isolated from the digestive tract of diurnal raptors in Catalonia, Spain. Vet Rec 154: 17–20. [PubMed]
15. Morand S (2000) Wormy word: Comparative tests of theoretical hypotheses on parasite species richness. In: Poulin R, Morand S, Skorping A editors. Evolutionary biology of host-parasite relationships: Reality meets models. Amsterdam: Elsevier Science. 63–79.
16. Poulin R (2007) Evolutionary ecology of parasites. Princeton: Princeton University Press. 332 p.
17. Peterson RT, Mountfort G, Hollom PAD (1983) A field guide to the birds of Britain and Europe. London and Glasgow: W Collins & Sons & Co Ltd. 312 p.
18. Santoro M, Kinsella JM, Galiero G, degli Uberti B, Aznar FJ (2012) Helminth community structure in birds of prey (Accipitriformes and Falconiformes) in southern Italy. J Parasitol 98: 22–29. [PubMed]
19. Bush AO (1990) Helminth communities in avian hosts: Determinants of pattern. In: Esch GW, Bush AO, Aho JM editors. Parasite communities: Patterns and processes. London and New York: Chapman and Hall. 197–232.
20. Fedynich AM, Pence DB (1994) Helminth community structure and pattern in a migratory host (Anas platyrhynchos). Can J Zool 72: 496–505.
21. Fedynich AM, Pence DB, Gray PN, Bergan JF (1996) Helminth community structure and pattern in two allopatric populations of a non-migratory waterfowl species (Anas fulvigula). Can J Zool 74: 1253–1259.
22. Poulin R (1995) Phylogeny, ecology, and the richness of parasite communities in vertebrates. Ecol Monogr 65: 283–302.
23. Poulin R (1997) Species richness of parasite assemblages: Evolution and patterns. Annu Rev Ecol Syst 28: 341–358.
24. Poulin R, Leung TLF (2011) Latitudinal gradient in the taxonomic composition of parasite communities. J Helminthol 85: 228–233. [PubMed]
25. Santoro M, Mattiucci S, Kinsella JM, Aznar FJ, Giordano D, et al. (2011) Helminth community structure of the Mediterranean gull (Ichthyaetus melanocephalus) in southern Italy. J Parasitol 97: 364–366. [PubMed]
26. Santoro M, Tripepi M, Kinsella JM, Panebianco A, Mattiucci S (2010) Helminth infestation in birds of prey (Accipitriformes and Falconiformes) in southern Italy. Vet J 186: 119–122. [PubMed]
27. Krone O, Streich WJ (2000) Strigea falconispalumbi in Eurasian buzzards from Germany. J Wildl Dis 36: 559–561. [PubMed]
28. Mutafchiev Y, Santoro M, Georgiev BB (2010) Parachordatortilis n. g. (Nematoda: Spirurida: Acuariidae), with a redescription of P. mathevossianae (Petrov & Chertkova, 1950) n. comb., a parasite of Falco tinnunculus L. (Aves: Falconiformes). Syst Parasitol 76: 191–197. [PubMed]
29. Krone O (2000) Endoparasites in free-ranging birds of prey in Germany. In: Lumeij JT, Remple JD, Redig PT, Lierz M, Cooper JE editors. Raptor Biomedicine III. Lake Worth: Zoological Education Network, Inc., 101–116.
30. Krone O (2007) Endoparasites. In: Bird DM, Bildstein KL editors. Raptor, research and management techniques. Surrey: Hancock House Publishers. 318–328.
31. Nöckler K, Kapel CMO (2007) Detection and surveillance for Trichinella: Meat inspection and hygiene and legislation. In: Dupouy-Camet J, Murrell KD editors. FAO/WHO/OIE Guidelines for the surveillance, management, prevention and control of Trichinellosis. Paris: World Organisation for Animal Health Press. 69–97.
32. Reiczigel J (2003) Confidence intervals for the binomial parameter: Some new considerations. Stat Med 22: 611–621. [PubMed]
33. Rózsa L, Reiczigel J, Majoros G (2000) Quantifying parasites in samples of hosts. J Parasitol 86: 228–232. [PubMed]
34. Conover WJ (1999) Practical nonparametric statistics. New York: John Wiley & Sons. 584 p.
35. Clarke KR, Warwick RM (2001) Change in marine communities: An approach to statistical analysis and interpretation. Plymouth: PRIMER-E. 172 p.
36. Poulin R, Morand S (2004) The parasite biodiversity. Washington DC: Smithsonian Institution Press. 216 p.
37. Engqvist L (2005) The mistreatment of covariate interaction terms in linear model analyses of behavioural and evolutionary ecology studies. Anim Behav 70: 967–971.
38. Lotz JM, Font WF (1991) The role of positive and negative interspecific associations in the organization of communities of intestinal helminths of bats. Parasitology 103: 127–138. [PubMed]
39. Snow D, Perrins CM, Doherty P, Cramp S (1998) The complete birds of the western Palaearctic on CD-ROM. Oxford University Press.
40. Reiczigel J, Rózsa L (2005) Quantitative parasitology 3.0. Budapest, Hungary. Available: 2011 Jun 20.
41. Kennedy CR, Bush AO, Aho JM (1986) Patterns in helminths communities: Why are birds and fish different? Parasitology 93: 205–215. [PubMed]
42. Jaksić FM, Seib RL, Herrera CH (1982) Predation by the barn owl (Tyto alba) in Mediterranean habitats of Chile, Spain and California: A comparative approach. Am Midl Nat 107: 151–162.
43. Alivizatos H, Goutner V, Zogaris S (2005) Contribution to the study of the diet of four owl species (Aves, Strigiformes) from mainland and island areas of Greece. Belg J Zool 135: 109–118.
44. Marchesi L, Sergio F (2005) Distribution, density, diet and productivity of the scops owl Otus scops in the Italian Alps. Ibis 147: 176–187.
45. Bosè M, Guidali F (2000) Seasonal and geographic differences in the diet of the barn owl in an agro-ecosystem in northern Italy. J Raptor Res 35: 240–246.
46. Cecere F, Vicini G (2000) Micromammals in the diet of the long eared owl (Asio otus) at the WWF’s Oasi San Giuliano (Matera, South Italy). Hystrix 11: 47–53.
47. Capizzi D (2000) Diet shifts of the tawny owl Strix aluco in central and northern Italy. Ital J Zool 67: 73–79.
48. Arcidiacono G, Donati C, Mastrorilli M (2007) Diet of the little owl Athene noctua in rural and urban areas: A bibliographical review. Studi Trent Sci Nat Acta Biol 83: 243–247.
49. Ewald JA, Crompton DWT, Johnston I, Stoddard RC (1991) The occurrence of Centrorhynchus (Acanthocephala) in shrews (Sorex araneus and S. minutus) in the United Kingdom. J Parasitol 77: 485–487. [PubMed]
50. Krone O, Cooper JE (2002) Parasitic disease. In: Cooper JE editor. Birds of Prey. Health and diseases. Iowa: Iowa State Press. 105–120.
51. Santoro M, Aznar FJ, Mattiucci S, Kinsella JM, Pellegrino F, et al. (2013) Parasite assemblages in the Western whip snake Hierophis viridiflavus carbonarius(Colubridae) from southern Italy. J Helminthol In Press. doi:10.1017/S0022149X12000338. [PubMed]
52. Anderson RC (2000) Nematode parasites of vertebrates: Their development and transmission. Wallingford: CABI Publishing. 650 p.
53. Perera A, Maia JP Jorge F, Harris DJ (2012) Molecular screening of nematodes in lacertid lizards from the Iberian Peninsula and Balearic Islands using 18S rRNA sequences. J Helminthol 2: 1–6. [PubMed]
54. Combes C (2001) Parasitism: The ecology and evolution of intimate interactions. Chicago: University of Chicago Press. 728 p.
55. Kuris AM, Goddard JHR, Torchin ME, Murphy N, Gurney R, et al. (2007) An experimental evaluation of host specificity: The role of encounter and compatibility filters for a rhizocephalan parasite of crabs. Int J Parasitol 37: 539–545. [PubMed]
56. Jaksić FM (1982) Inadequacy of activity time as a niche difference: The case of diurnal and nocturnal raptors. Oecologia 52: 171–175.
57. Cui Q, Su J, Jiang Z (2008) Summer diet of two sympatric species of raptors upland buzzard (Buteo hemilasius) and Eurasian eagle owl (Bubo bubo) in Alpine meadow: Problem of coexistence. Pol J Ecol 56: 173–179.
58. Kinsella JM (2007) Helminths of the vagrant shrew, Sorex vagrans, from western Montana, USA. Acta Parasitol 52: 151–155.
59. Johnsgard PA (1990) Hawks, eagles and falcons of North America: Biology and natural history. Washington DC: Smithsonian Institution Press. 403 p.
60. Khalil LF, Jones A, Bray RA (1994) Keys to the cestode parasites of vertebrates. Wallingford: CAB International. 751 p.
61. Gibson DI, Jones A, Bray RA (2002) Keys to the Trematoda, Vol. 1. London: CAB International and Natural History Museum. 521 p.
62. Pojmańska T (2008) Family Dicrocoeliidae Looss, 1899. In: Bray RA, Gibson DI, Jones A, editors. Keys to the Trematoda, Vol. 3. London: CAB International and Natural History Museum. 233–260.
63. Furmaga S (1957) The helminth fauna of predatory birds (Accipitres et Striges) of the environment of Lublin. Acta Parasitol Pol 13: 215–297.
64. De Muro P (1934) Choanotaenia littoriae sp. n., nuovo cestode nell’intestino della civetta (Athene noctua). Croce Rossa Italiana, Rivista Mensile 9: 486–490.
65. Dubois G (1938) Monographie des Strigeida (Trematoda). Mem Soc Neuch Sci Nat 6: 1–535.
66. Dimitrova ZM, Gibson DI (2005) Some species of Centrorhynchus Lühe, 1911 (Acanthocephala: Centrorhynchidae) from the collection of the Natural History Museum, London. Syst Parasitol 62: 117–134. [PubMed]
67. San Martín ÓJ (2009) Diversidad de la superfamilia Trichinelloidea (Enoplea, Nematoda) en Chile y su implicancia en salud humana, animal y ecosistémica. Lundiana 10: 19–52.
68. Host parasite database of the Natural History Museum, London, UK (2012) Available: 2012 Nov 5.

Articles from PLoS ONE are provided here courtesy of Public Library of Science

Pulmonary tuberculosis diagnostic delays in Chad: a multicenter, hospital-based survey in Ndjamena and Moundou

Download PDF



Tuberculosis remains one of the leading causes of morbidity and mortality in low-resource countries. One contagious patient can infect 10 to 20 contacts in these settings. Delays in diagnosing TB therefore contribute to the spread of the disease and sustain the epidemic.


The aim of this study was to assess delays in diagnosing tuberculosis and the factors associated with these delays in the public hospitals in Moundou and Ndjamena, Chad.


A structured questionnaire was administered to 286 new tuberculosis patients to evaluate patient delay (time from the onset of symptoms to the first formal or informal care), health-care system delay (time from the first health care to tuberculosis treatment) and total delay (sum of the patient and system delays). Logistic regression was used to identify risk factors associated with long diagnostic delays (defined as greater than the median).

Results and discussion

The median [interquartile range] patient delay, system delay and total delay were 15 [7–30], 36 [19–65] and 57.5 [33–95] days, respectively. Low economic status (aOR [adjusted odds ratio] =2.38 [1.08-5.25]), not being referred to a health service (aOR = 1.75 [1.02- 3.02]) and a secondary level education (aOR = 0.33 [0.12-0.92]) were associated with a long patient delay. Risk factors for a long system delay were a low level of education (aOR = 4.71 [1.34-16.51]) and the belief that traditional medicine and informal care can cure TB (aOR = 5.46 [2.37-12.60]).


Targeted strengthening of the health-care system, including improving patient access, addressing deficiencies in health-related human resources, and improving laboratory networks and linkages as well as community mobilization will make for better outcomes in tuberculosis diagnosis.

Keywords: Tuberculosis, Delay, Diagnosis, Treatment


Tuberculosis (TB) is one of the leading causes of morbidity and mortality: 9.2 million new cases of TB and 1.7 million deaths due to this disease were reported worldwide in 2007. The majority of these cases occurred in developing countries, particularly in Asia and Africa [1]. In limited-resource countries, one contagious patient can infect 10 to 20 people during the natural history of the disease [2]. Lin X et al. found that 30 days of infectious disease is enough for the bacillus to pass from the index case to the household members [3]. Consequently, any delay in the diagnosis, care and treatment of TB patients not only exposes them to severe morbidity and a greater risk of mortality, but it also contributes to the spread of the epidemic [4-7]. Thus, one of the main goals of TB control programs should be the prompt diagnosis and treatment of TB patients.

TB is one of Chad’s major public health concerns [8]. In 2009, the prevalence of TB was 480/100,000 population, with an annual incidence estimated at 283/100,000 population and a specific mortality of 63/100,000 population [8]. The disease has been the target of a national control program since 1990, and the DOTS strategy was adopted in 1994. TB care and treatment are free in Chad. Patients with symptoms suggestive of TB are identified when they visit a first-level health service and are subsequently referred to a hospital, where a diagnosis of TB can be confirmed. The main diagnostic tools used are the sputum smear test and chest radiography. When the diagnosis is confirmed, standard treatment regimens are prescribed in accordance with World Health Organization (WHO) recommendations.

A study conducted at a hospital in Ndjamena in 2003 determined the TB diagnostic delay to be 75 days. However, the authors did not clearly distinguish between the patient delay and the health-care system delay [9]. The objectives of our study were to investigate pulmonary TB diagnostic delays and to identify factors associated with these delays in order to strengthen the TB prevention program. For the period from the onset of symptoms to the initiation of TB treatment, we sought to distinguish the “patient delay” (time to the first access to care, whether formal or informal) and the “health-care system delay” (time from the first access to care to the initiation of TB treatment).



A multicenter questionnaire-type survey was conducted from August to October 2009 in three hospitals, two of which are in the Chadian capital, which has the largest number of TB patients (the Hôpital Général de Référence de Ndjaména [HGRN] and the Hôpital de l’Union [HU]). Both serve mainly local and urban TB patients. The third hospital, Hôpital de Moundou (HM), is the regional hospital for the Western Logone region (440 km south of Ndjamena). Regular hospitals are designed to serve a population of 100,000 to 200,000, but referral hospitals have a population base larger than this. The population of Ndjamena is 833,531, and of the 650,000 inhabitants of Western Logone, 142,000 live in Moundou. Patients are supposed to visit a health center first. From there, under the referral system, the more severe cases are sent to district hospitals, then to regional hospitals and, lastly, to the HGRN.

Study population

Newly diagnosed cases of pulmonary TB aged 15 years or older were recruited consecutively and prospectively. The TB cases were classified according to the guidelines of the Chadian TB control program (WHO guidelines). Patients with other lung diseases or extrapulmonary TB, those who declined to give their consent and those who were too weak to answer the questionnaire were excluded from this study. Assuming a frequency of extended total delay of 60% among individuals exposed to a risk factor and of 40% among those not exposed, the study required a sample size of least 225 patients.


A semi-structured questionnaire was used to collect the data. It was translated into Arabic and Sara when necessary. The questionnaires were filled out by trained interviewers who conducted face-to-face interviews shortly after diagnosis. The patients’ medical records were cross-checked to confirm and complete the data.

The outcome variables were the patient delay (PD; defined as the time interval between the onset of a cough lasting more than 15 days and/or of major symptoms according to the national TB control program guidelines, i.e., night sweats, weight loss, fever and respiratory symptoms− all the cases were reviewed by a pneumologist to date the onset of TB symptoms − and the first formal or informal health care received); the health-care system delay (HSD; defined as the time interval between the previously mentioned care and the initiation of TB treatment); and the total delay (TD; defined as the sum of the patient and system delays). The delays were estimated in number of days. Delays were considered extended when they were longer than their respective median values.

The independent variables to be studied were chosen after an intensive literature review. They were the individual’s demographic and socioeconomic characteristics, such as gender, age (divided into five groups), rural residency, defined as living outside the city (yes/no), health insurance status (yes/no), and level of education (in five groups of increasing numbers of years of education). Economic status was assessed by calculating a wealth score based on housing status, the construction quality of the dwelling, the sources of drinking water and electricity, the type of sanitation, the ownership of certain items (such as a car, a motorbike, a bicycle, a refrigerator or a television) and the case’s occupational status. We also asked the patients how they would pay the additional expenses. The answers were grouped into five categories: the household’s savings, a loan, financial help from relatives or friends, selling his/her belongings, and earnings from continuing to work. We also asked the patients if one of their friends or relatives was a health-care worker (yes/no).

Three medical findings were considered as well: the presence of hemoptysis (yes/no), the result of the smear test (positive or negative) and the patient’s HIV serological status (positive, negative, unknown).

Knowledge and attitudes concerning TB were assessed with questions regarding the cause of TB, its mode of transmission, its treatment, the link between TB and AIDS, and the primary care received.

Distance between the patient’s residence and the closest health facility was divided into three categories (≤ 1 km, between 1 and 5 km, and  5 km). Lastly, whether or not the case had been referred to the hospital by a primary care facility was examined.

Statistical analysis

The distributions of the independent variables with the three different delays were compared using a chi-square test (or Fischer’s exact test where the numbers were small), and quantitative variables were compared using the (non-parametric) Wilcoxon test and the Kruskal and Wallis test. The associations between the ordinal variables (age and wealth score) and the outcomes of interest were assessed for trends. Next, since the delays differed according to the three hospitals, we performed bivariate analysis to make the same comparisons after adjusting for the study site and examined whether there were any interactions. Lastly, we included all the variables with a p-value  0.20 in bivariate analysis and selected them by backward analysis, fitting a logistic regression model for each delay separately. In multivariate analysis, the categories for knowledge of TB treatment were medical care, no response and other responses. The categories for the first health care received were formal (health center, hospital, pharmacist or private doctor) and informal (other responses), and the means for paying the additional expenses were classified according to the ability (savings, work) or inability (other responses) to pay. Epidata 3.1 software was used to build the database. Statistical analyses were performed with SAS 9.2.

Ethical issues

Since there is no ethics committee in Chad, research authorization was obtained from the Chadian Health Ministry. Each patient had been informed of the study’s objectives and his/her right to decline to participate. Verbal informed consent was obtained before every interview. No act that could harm the patients’ dignity or physical integrity was committed during this study.


Population characteristics

Two-hundred and eighty-six newly diagnosed patients were included in the analysis (Figure ​(Figure1).1). They were mainly men (67.1%). The median age was 32 years, with less than a fourth of this population being over the age of 41. The education level was low: one-fifth of the population had no education, and only one-tenth of the patients had reached a postgraduate level. Only a minority (17.5%) of the patients lived in a rural area. The average size of the patients’ households was 6.1 persons. Half of them were unemployed and had no income. More than 80% of the smear tests were positive. One-fifth of the patients were HIV-positive, and one-third of them had not been tested for HIV. Very few patients (13%) had health insurance, and more than half of them (60.4%) expected financial help from their relatives. One-third of the patients sought treatment by visiting a hospital, 22% by buying drugs on the informal market, 21% by visiting a health center, 13% by using traditional medicine, less than 8% by consulting a private doctor, and 3.5% by consulting a pharmacist. Only 2.1% of them did not seek health care.

Figure 1

Study recruitment.

Comparison of the hospital populations

The patients at the HM and the HGRN seemed to be older than those at the HU (p≤0.001), and the patients at the HU were likely to be more educated (p < 0.0001) (Table ​(Table1).1). The wealth scores were higher for the HM and the HGRN than for the HU (p = 0.02). Unemployment also seemed to be more frequent for the HU than for the other two facilities (p < 0.01). There was a higher rate of HIV-positive serology for the HGRN (29.4%) than for the HM (20.3%) and the HU (5.8%), and the HIV serological status of more than half of the patients was unknown at the HU and the HM compared to only one-fifth at the HGRN (p < 0.0001). Because the HSD (p < 0.0001) and TD (p = 0.0002) were much longer for the HGRN, bivariate analysis was adjusted for the hospital.

Table 1

Characteristics of the study hospitals

Risk factors associated with an extended patient delay

Once adjusted for the study hospital (Table ​(Table2),2), protective factors were a higher level of education, having health insurance, the belief that people hide their TB, having a health professional among one’s relatives, and the primary care having been obtained by consulting a pharmacist. On the other hand, an extended PD was associated with a remote community health facility, selling one’s belongings in order to pay the additional expenses, and not knowing how TB is transmitted. In multivariate analysis (Table ​(Table3),3), an extended PD was associated with a low wealth score, an intermediate education level, misconceptions about TB treatment, and having no referral to a hospital.

Table 2

Factors associated with delays exceeding their median value (univariate analysis )
Table 3

Factors associated with delays exceeding their median value (bivariate analysis, adjusted for the hospital)

Risk factors associated with an extended health-care system delay

In bivariate analysis, knowing that TB treatment is free and having received the primary care in a hospital were associated with a shorter HSD, while a low level of education, a low economic status, remote residence, living in a rural area, and the belief that traditional medicine can cure TB were associated with an extended HSD. In multivariate analysis, a low wealth score, having no knowledge about the correlation between AIDS and TB, a poor knowledge of TB treatment, and being treated at the HGRN were the three characteristics associated with an extended HSD.

Factors associated with an extended total delay

Univariate analysis (Table ​(Table4),4), showed that having health insurance, unknown HIV serological status, knowing that TB treatment is free, and not knowing about the link between AIDS and TB were associated with a shorter TD. Living in a rural area, believing that traditional healing can cure TB and having started to undertake health care by using a traditional treatment appeared to be significantly associated with an extended TD. In multivariate regression analysis, a low economic status, the absence of hemoptysis, the belief in the efficacy of traditional and informal treatments, and being treated at either of Ndjamena’s hospitals were four characteristics associated with a longer TD.

Table 4

Comparison of the PD, HSD and TD with the findings in the literature


This study reveals a long delay in TB diagnosis, with an HSD 2.4 times longer than the PD (Table ​(Table1).1). The results show that a low economic status, a low level of education and the belief in the efficacy of traditional treatments were associated with extended diagnostic delays.

Patient delay, health-care system delay and total delay

Lin X et al. found that TB infection spreads in the index case’s household after 30 days [3]. Three-fourths of the patients in this study began their TB treatment at least 33 days after the onset of symptoms (Table ​(Table1).1). Therefore, the delays in diagnosing TB observed in this study are likely to be important in the spread of this disease.

The median PD of 15 days is equal to the duration of a cough that should be considered suspicious for TB, according to the national program guidelines. The median HSD in this study is one of the longest observed, while the PD is one of the shortest compared to the findings in other settings (Table ​(Table5).5). This could be explained by the decision to include informal care in the definition of the primary care received by the patients in this study. Indeed, some authors consider the PD to be the time interval between the onset of symptoms and the first formal medical treatment received. Thus, the exclusion of informal and traditional health care from the definition of the primary care received seems to compound the patient’s role in the delay in TB diagnosis [6,7]. Therefore, the impact of informal care on the TD may be underestimated in resource-limited countries. For example, we observed that more than half of the patients visited a conventional care provider first and that those with formal care trajectories were likely to be diagnosed earlier. Therefore, traditional medicine and informal care should be considered part of the health-care system in studies conducted in developing countries.

Table 5

Factors associated with delays exceeding their median value (multivariate analysis)

Determinants of patient delay

Several studies have shown that the inability to pay for health care is a barrier to seeking it [10-12]. Surprisingly, this was also the finding in this study, even though TB treatment is free. Indeed, patients bear certain direct and indirect costs (drugs, consultations, investigations, transportation, lost days of work, etc.) from the onset of symptoms to when TB is suspected. Although tests for TB are performed free of charge, patients still pay the rest of the expenses: food, transportation, lost income and so on. This prediagnostic cost can represent 7.1% of the median annual household income in Kenya, and patients may spend up to 125% of their monthly income to get a proper diagnosis in Ethiopia [13,14]. Mesfin et al found that spending time seeking care instead of earning money worsens TB patients’ financial burden and impoverishes their households [14]. This economic pressure may lead patients to delay their first visit to a doctor if the symptoms appear to be mild.

The PD seems to decrease when the level of education increases [15]. A higher level of education may be associated with a better knowledge of TB and a better understanding of the health-care system. Thus, more educated patients promptly consult a health professional shortly after the onset of symptoms. However, a higher level of education might also be associated with self-medication and the postponement of the first visit to a doctor.

Typically, patients with suspected TB would be seen in lower-level facilities and referred to the next level for further management. Thus, the referral system needs to be simple and efficient in order to reduce delays. When patients are not familiar with the referral system, they are likely to seek treatment outside the conventional services or make multiple visits to the same lower-level facilities without progressing upward. In our study, referral was associated with a shorter PD, which is contrary to the findings of other studies, where referral was associated with a longer PD (more obvious symptoms of TB due to a delayed first visit to a doctor) [16]. Surprisingly, there were few referrals in our study, despite the fact that the entire study population consisted of TB cases. This may be a reflection of the poor case-detection skills of lower-level health-care providers.

Determinants of the health-care system delay

Similar to other studies which found that low income was associated with longer delays, we noted that low economic status lengthened the HSD [17,18]. Spending time seeking care and having to pay the necessary expenses to access it may impede the patient’s progression through the health-care system [19]. In Myanmar, Lönnroth et al showed that implementing measures to address the financial burden of TB can significantly shorten diagnostic delays [20]. Economic impediments to accessing health care are likely to contribute to the lengthening of the HSD in Chad, despite the fact that TB treatment is free there.

The organization of health care and its quality may affect the HSD [19]. Indeed, the centralization of TB diagnosis requires a visit to a hospital for the sputum smear test and a chest radiograph. In this study, the longest HSDs were associated with having been diagnosed in Ndjamena. This could be explained by the fact that this is a larger city with more-substantial health-care facilities, with the result that there are a larger number of potential steps in the pathway of care. Storla et al. do, in fact, call attention to the harmful role of repeated visits at the same level of care as one of the mechanisms that can contribute to diagnostic delay in TB [7]. The hierarchical level of care might also increase the risk of lengthening the HSD, given that the patients diagnosed at the HGRN seem to have had a longer HSD.

A poor knowledge of TB may lead to a longer HSD [19]. Believing in the efficacy of informal care and especially of traditional medicine in curing TB was significantly associated with longer HSDs in this study. The literature shows similar findings in different contexts, such as Vietnam, Nepal and South Africa [21-23]. These patients may use traditional healers as gatekeepers to enter the health-care system. The ability of these healers to identify TB symptoms and to then promptly refer the patient to a trained health professional could impact the HSD. Thus, training traditional healers on and involving them in the TB detection strategy might reduce the HSD.

Determinants of total delay

The centralization of the point of diagnosis of TB, the referral pattern, the cost of care and the misunderstanding of the requirements of TB treatment influenced the TD in the same manner as they influenced the PD and the HSD. As a result, a longer TD was associated with a lower economic status, with the belief in the efficacy of informal treatment and with having been diagnosed at a Ndjamena hospital.

A low sensitivity of the TB screening criteria may also be a key factor in delays. Indeed, the TD was longer in the absence of hemoptysis. The inability of a health-care provider to suspect TB when the pulmonary signs are mild might explain this association [19]. This is probably one of the reasons why the TB detection rate remains so low in Chad.

Some factors seem to affect the first and the second phase of the pathway of care in opposite directions [19], and their effects on the TD may be the result of this opposite influence on the PD and the HSD. For example, being a woman was associated with a shorter PD, but paradoxically, it may be associated with a longer HSD. The women’s behaviour was unlikely to be significantly different from that of the men at the beginning of the trajectory of care, but afterwards, they were likely to encounter some gender-specific barriers once they entered the health-care system. The gender-specific parameters that may have been associated with the slow progression of women through the health-care system include a lack of financial independence, a lower social status, family responsibilities and a lack of respect from health-care providers. Consequently, public health interventions should be tailored to different circumstances.


Since it excluded patients who died before reaching the hospital and those who were too ill to be interviewed, this study may underestimate TB diagnostic delays in Chad. This should be taken into account when interpreting the results of this study. These results concern patients who had access to public tertiary hospitals in Ndjamena and Moundou. Since WHO estimated the TB case-detection rate at 26% in Chad in 2009 [8], there is a need to understand the behavior of patients who are not detected. Another study should help identify the determinants of their health care trajectories.

The multicenter design of this study enabled us to investigate the factors associated with the delayed initiation of TB treatment at two different levels of the health-care system and in two different cities and regions.


The TD in Ndjamena and Moundou is too long. A fourth of the patients began their TB treatment at least 95 days after the onset of symptoms. The 286 patients in this study may have exposed 1740 members of their respective households to a risk of TB infection when they were infectious. The ability to pay for care, the level of education, knowledge of TB and knowledge of the organization of health care may determine the length of the delay in the diagnosis of TB. Significant differences in diagnostic delays might also depend on the quality of care, on the ability of health professionals to use the TB detection protocol, and on how they interact with the patients.

Implementing measures to inform the general public about TB and the availability of free TB treatment could help shorten diagnostic delays. Certain measures, such as microfinance, might improve the performance of the referral pattern by reducing the financial burden of TB for patients. Transporting sputum specimens from first-level facilities to the nearest hospitals could decentralize TB diagnosis without decreasing the quality of the sputum smear test. This decentralization would also reduce the cost incurred by patients to get diagnosed.

Training health workers on the management of TB via regular mentoring and supervision could improve the management of TB. The need to limit the transmission of the bacillus may encourage active screening of the households of contagious patients, despite the cost of this measure. Involving traditional healers and informal health professionals in the screening strategy might also facilitate patient access to TB diagnosis. Lastly, regular monitoring, a TB control program and the evaluation of this program are necessary to facilitate the use of public TB services.

Competing interests

The authors declare no conflict of interest.

Authors’ contribution

NNN designed the study protocol, collected and analysed the data and drafted the article. DN and NR revised the study protocol, collected the data and revised the article. MNN revised the study protocol and the article. MGS revised the article. VHF and PC revised the study protocol, supervised the data analysis and revised the article. All authors approve this submitted version of the article.

Pre-publication history

The pre-publication history for this paper can be accessed here:


This study was supported by the Chadian Health Ministry. We thank Dr. Abdelatti, Mr. Fina-Teysou and Mr. Guinloungoum of the Chadian TB control program for their advice and help.

The Chadian Health Ministry was not involved in the design, the analysis, the interpretation of the results or in the writing of this article. We also thank Dr. P. Izulla (Kenya) for his assistance in editing the English version of the article.


  • Glaziou P, Floyd K, Raviglione M. Global burden and epidemiology of tuberculosis. Clin Chest Med. 2009;30(4):621–636. [PubMed]
  • Lawn SD, Afful B, Acheampong JW. Pulmonary tuberculosis: diagnostic delay in Ghanaian adults. Int J Tuberc Lung Dis. 1998;2(8):635–640. [PubMed]
  • Lin X, Chongsuvivatwong V, Lin L, Geater A, Lijuan R. Dose–response relationship between treatment delay of smear-positive tuberculosis patients and intra-household transmission: a cross-sectional study. Trans R Soc Trop Med Hyg. 2008;102(8):797–804. [PubMed]
  • Golub JE, Bur S, Cronin WA, Gange S, Baruch N, Comstock GW, Chaisson RE. Delayed tuberculosis diagnosis and tuberculosis transmission. Int J Tuberc Lung Dis. 2006;10(1):24–30. [PubMed]
  • Asch S, Leake B, Anderson R, Gelberg L. Why do symptomatic patients delay obtaining care for tuberculosis? Am J Respir Crit Care Med. 1998;157(4):1244–1248. [PubMed]
  • Sreeramareddy CT, Panduru KV, Menten J, Van den Ende J. Time delays in diagnosis of pulmonary tuberculosis: a systematic review of literature. BMC Infect Dis. 2009;9:91. [PMC free article] [PubMed]
  • Storla DG, Yimer S, Bjune GA. A systematic review of delay in the diagnosis and treatment of tuberculosis. Bmc Public Health. 2008;8:15. [PMC free article] [PubMed]
  • Wordl Health Organization. Global Tuberculosis Control 2010. In: Global tuberculosis control: WHO report 2010. vol. WHO/HTM/TB/2010.7, Wordl Health Organization. World Health Organization, Geneva; 2010.
  • Martin A, Baptiste JP, Krieger G. Respiratory infections: SARS and tuberculosis. Clin Occup Environ Med. 2004;4(1):189–204. [PubMed]
  • Wang JM, Fei Y, Shen HB, Xu B. Gender difference in knowledge of tuberculosis and associated health-care seeking behaviors: a cross-sectional study in a rural area of China. Bmc Public Health. 2008;8:354. [PMC free article] [PubMed]
  • Wang Y, Long Q, Liu Q, Tolhurst R, Tang SL. Treatment seeking for symptoms suggestive of TB: comparison between migrants and permanent urban residents in Chongqing, China. Trop Med Int Health. 2008;13(7):927–933. [PubMed]
  • Xu B, Jiang QW, Xiu Y, Diwan VK. Diagnostic delays in access to tuberculosis care in counties with or without the National Tuberculosis Control Programme in rural China. Int J Tuberc Lung Dis. 2005;9(7):784–790. [PubMed]
  • Mauch V, Woods N, Kirubi B, Kipruto H, Sitienei J, Klinkenberg E. Assessing access barriers to tuberculosis care with the tool to Estimate Patients’ Costs: pilot results from two districts in Kenya. Bmc Public Health. 2011;11:43. [PMC free article] [PubMed]
  • Mesfin MM, Newell JN, Madeley RJ, Mirzoev TN, Tareke IG, Kifle YT, Gessessew A, Walley JD. Cost implications of delays to tuberculosis diagnosis among pulmonary tuberculosis patients in Ethiopia. Bmc Public Health. 2010;10:173. [PMC free article] [PubMed]
  • Xu B, Diwan VK, Bogg L. Access to tuberculosis care: What did chronic cough patients experience in the way of healthcare-seeking? Scand J Public Health. 2007;35(4):396–402. [PubMed]
  • Meintjes G, Schoeman H, Morroni C, Wilson D, Maartens G. Patient and provider delay in tuberculosis suspects from communities with a high HIV prevalence in South Africa: A cross-sectional study. BMC Infect Dis. 2008;8:72. [PMC free article] [PubMed]
  • Tobgay KJ, Sarma PS, Thankappan KR. Predictors of treatment delays for tuberculosis in Sikkim. Natl Med J India. 2006;19(2):60–63. [PubMed]
  • Lönnroth K, Thuong LM, Linh PD, Diwan VK. Delay and discontinuity–a survey of TB patients’ search of a diagnosis in a diversified health care system. Int J Tuberc Lung Dis. 1999;3(11):992–1000. [PubMed]
  • Ndeikoundam Ngangro N, Chauvin P, HalleydesFontaines V. Determinants of tuberculosis diagnosis delay in limited resources countries. Rev Epidemiol Sante Publique. 2012;60(1):47–57. [PubMed]
  • Lonnroth K, Aung T, Maung W, Kluge H, Uplekar M. Social franchising of TB care through private GPs in Myanmar: an assessment of treatment results, access, equity and financial protection. Health Policy Plan. 2007;22(3):156–166. [PubMed]
  • Barker RD, Millard FJC, Malatsi J, Mkoana L, Ngoatwana T, Agarawal S, de Valliere S. Traditional healers, treatment delay, performance status and death from TB in rural South Africa. Int J Tuberc Lung Dis. 2006;10(6):670–675. [PubMed]
  • Huong NT, Vree M, Duong BD, Khanh VT, Loan VT, Co NV, Borgdorff MW, Cobelens FG. Delays in the diagnosis and treatment of tuberculosis patients in Vietnam: a cross-sectional study. Bmc Public Health. 2007;7:110. [PMC free article] [PubMed]
  • Yamasaki-Nakagawa M, Ozasa K, Yamada N, Osuga K, Shimouchi A, Ishikawa N, Bam DS, Mori T. Gender difference in delays to diagnosis and health care seeking behaviour in a rural area of Nepal. Int J Tuberc Lung Dis. 2001;5(1):24–31. [PubMed]
  • Kiwuwa MS, Charles K, Harriet MK. Patient and health service delay in pulmonary tuberculosis patients attending a referral hospital: a cross-sectional study. Bmc Public Health. 2005;5:122. [PMC free article] [PubMed]
  • Lienhardt C, Rowley J, Manneh K, Lahai G, Needham D, Milligan P, McAdam KP. Factors affecting time delay to treatment in a tuberculosis control programme in a sub-Saharan African country: the experience of The Gambia. Int J Tuberc Lung Dis. 2001;5(3):233–239. [PubMed]
  • Ayuo PO, Diero LO, Owino-Ong’or WD, Mwangi AW. Causes of delay in diagnosis of pulmonary tuberculosis in patients attending a referral hospital in Western Kenya. East Afr Med J. 2008;85(6)):263–268. [PubMed]
  • Demissie M, Lindtjorn B, Berhane Y. Patient and health service delay in the diagnosis of pulmonary tuberculosis in Ethiopia. Bmc Public Health. 2002;2:23. [PMC free article] [PubMed]
  • Gele AA, Bjune G, Abebe F. Pastoralism and delay in diagnosis of TB in Ethiopia. Bmc Public Health. 2009;9:5. [PMC free article] [PubMed]
  • Lorent N, Mugwaneza P, Mugabekazi J, Gasana M, Van Bastelaere S, Clerinx J, Van den Ende J. Risk factors for delay in the diagnosis and treatment of tuberculosis at a referral hospital in Rwanda. Int J Tuberc Lung Dis. 2008;12(4):392–396. [PubMed]
  • Odusanya OO, Babafemi JO. Patterns of delays amongst pulmonary tuberculosis patients in Lagos. Nigeria. Bmc Public Health. 2004;4:18. [PMC free article] [PubMed]
  • Yimer S, Bjune G, Alene G. Diagnostic and treatment delay among pulmonary tuberculosis patients in Ethiopia: a cross sectional study. BMC Infect Dis. 2005;5:112. [PMC free article] [PubMed]
  • Harries AD, Salaniponi FM, Kwanjana JH. Directly observed treatment for tuberculosis. Lancet. 1999;353(9147):146–147. [PubMed]
  • Cambanis A, Ramsay A, Yassin MA, Cuevas LE. Duration and associated factors of patient delay during tuberculosis screening in rural Cameroon. Trop Med Int Health. 2007;12(11):1309–1314. [PubMed]
  • Kasse Y, Jasseh M, Corrah T, Donkor S, Antonnio M, Jallow A, Adegbola R, Hill P. Health seeking behaviour, health system experience and tuberculosis case finding in Gambians with cough. BMC Public Health. 2006;6:143. [PMC free article] [PubMed]
  • Mfinanga SG, Mutayoba BK, Kahwa A, Kimaro G, Mtandu R, Ngadaya E, Egwaga S, Kitua AY. The magnitude and factors associated with delays in management of smear positive tuberculosis in Dar es Salaam. Tanzania. BMC Health Serv Res. 2008;8:158. [PMC free article] [PubMed]
  • Steen TW, Mazonde GN. Pulmonary tuberculosis in Kweneng district, Botswana: delays in diagnosis in 212 smear-positive patients. Int J Tuberc Lung Dis. 1998;2(8):627–634. [PubMed]
  • Salaniponi FML, Harries AD, Banda HT, Kang’ombe C, Mphasa N, Mwale A, Upindi B, Nyirenda TE, Banerjee A, Boeree MJ. Care seeking behaviour and diagnostic processes in patients with smear-positive pulmonary tuberculosis in Malawi. Int J Tuberc Lung Dis. 2000;4(4):327. [PubMed]
  • Camara A, Diallo A, Camara LM, Fielding K, Sow OY, Chaperon J. Factors linked to delayed diagnosis of tuberculosis in Conakry (Guinea) Sante Publique. 2006;18(1)):63––70. [PubMed]
  • Cambanis A, Yassin MA, Ramsay A, Squire SB, Arbide I, Cuevas LE. Rural poverty and delayed presentation to tuberculosis services in Ethiopia. Trop Med Int Health. 2005;10(4):330–335. [PubMed]
  • Kilale A, Mushi A, Lema L, Kunda J, Makasi C, Mwaseba D, Range N, Mfinanga G. Perceptions of tuberculosis and treatment seeking behaviour in Ilala and Kinondoni Municipalities in Tanzania. Tanzan J Health Res. 2008;10(2):89–94. [PubMed]
  • Okeibunor JC, Onyeneho NG, Chukwu JN, Post E. Where do tuberculosis patients go for treatment before reporting to DOTS clinics in southern Nigeria? Tanzan Health Res Bul. 2007;9(2):94–101. [PubMed]
  • Ouédraogo M, Kouanda S, Boncoungou K, Dembélé M, Zoubga ZA, Ouédraogo SM, Coulibaly G. Treatment seeking behaviour of smear-positive tuberculosis patients diagnosed in Burkina Faso. Int J Tuberc Lung Dis. 2006;10(2):184–187. [PubMed]

Articles from BMC Public Health are provided here courtesy of BioMed Central

Characterization of the Modes of Binding between Human Sweet Taste Receptor and Low-Molecular-Weight Sweet Compounds

Download PDF
Wolfgang Meyerhof, Editor
This article has been cited by other articles in PMC.


One of the most distinctive features of human sweet taste perception is its broad tuning to chemically diverse compounds ranging from low-molecular-weight sweeteners to sweet-tasting proteins. Many reports suggest that the human sweet taste receptor (hT1R2–hT1R3), a heteromeric complex composed of T1R2 and T1R3 subunits belonging to the class C G protein–coupled receptor family, has multiple binding sites for these sweeteners. However, it remains unclear how the same receptor recognizes such diverse structures. Here we aim to characterize the modes of binding between hT1R2–hT1R3 and low-molecular-weight sweet compounds by functional analysis of a series of site-directed mutants and by molecular modeling–based docking simulation at the binding pocket formed on the large extracellular amino-terminal domain (ATD) of hT1R2. We successfully determined the amino acid residues responsible for binding to sweeteners in the cleft of hT1R2 ATD. Our results suggest that individual ligands have sets of specific residues for binding in correspondence with the chemical structures and other residues responsible for interacting with multiple ligands.


The human sweet taste receptor (hT1R2–hT1R3) is a heteromeric complex composed of two subunits, T1R2 and T1R3, which are class C G protein–coupled receptors (GPCRs) [1], [2], [3]. Each subunit has a large amino-terminal domain (ATD) linked by an extracellular cysteine-rich domain (CRD) to a seven-transmembrane helical domain (TMD) [4]. hT1R2–hT1R3 responds to a wide variety of chemical substances including naturally occurring sugars (glucose, sucrose, fructose and sugar alcohols), D-amino acids (D-tryptophan and D-phenylalanine) and glycosides (stevioside and glycyrrhizin), as well as artificial chemical compounds such as sucralose, aspartame, neotame, saccharin Na, acesulfame K (AceK), and cyclamate (Fig. 1) [5]. Moreover, naturally occurring sweet proteins, such as brazzein, thaumatin, and monellin, and naturally occurring taste-modifying proteins, such as neoculin and miraculin, also bind to hT1R2–hT1R3 [6], [7], [8], [9], [10], [11]. hT1R2–hT1R3 has multiple ligand-binding sites for these various sweeteners. For example, the ATD of hT1R2 is responsible for binding to aspartame and sugar derivatives [9]. Neoculin binds the ATD of hT1R3 [12]. In contrast, cyclamate and neohesperidin dihydrochalcone (NHDC) bind the TMD of hT1R3 as agonists [13], whereas this region also serves as the allosteric binding site for saccharin and lactisole as antagonists [14].

Figure 1

Chemical structures of the small molecular sweeteners used in this study.

The structural features of the ATD of the homodimeric metabotropic glutamate type 1 receptor (mGluR1) have been identified by X-ray crystal structure analysis, and this was the first example to reveal the structure of a class C GPCR [15]. The ATD of mGluR1 comprises two lobes (LB1 and LB2) that form the glutamate-binding domain lying between LB1 and LB2. The structure of ATD exists in an equilibrium of two different conformations, and the structural change strongly depends on glutamate binding. In the ligand-free state, both LB1 and LB2 tend to show open conformations (open-open), whereas an agonist induces a closed conformation for LB1 and LB2 of one ATD, while the other remains in an open conformation. This closed-open structure is thought to contribute to the active state of mGluR1 [15].

Because hT1R2 and hT1R3 share sequence homology (24% and 23%) with mGluR1 (Fig. S1), they also share some common structural features with mGluR1 [16]. hT1R2–hT1R3 can form a heterodimer, with the open-open form representing an inactive structure and the closed-open form representing an active structure. When low-molecular-weight sweeteners are applied, hT1R2 probably exhibits a closed conformation because the ATD of hT1R2 receives aspartame and sugar derivatives [17], [18]. Not only these small sweeteners but also cyclic sulfamate derivatives such as saccharin sodium and AceK probably bind at the cleft formed by LB1 and LB2 of hT1R2 ATD; they differ from each other in their hydrophobicity, electric charge, molecular size and other parameters (Fig. 1). Naturally occurring hydrophilic sugars are generally different in chemical structure from rather hydrophobic artificial amino acid derivatives and cyclic sulfamate derivatives. Moreover, amino acid derivatives and cyclic sulfamate derivatives have charged groups, whereas sugar derivatives are neutral.

Several ligand-binding sites were proposed by a molecular modeling–based docking simulation for the sweet taste receptor [6], [8], [11], [16], [19]. Thus, the wedge site of an open form of the ATD of the T1R3 was proposed for sweet proteins [6], [8], [16], whereas the involvement of the CRD of the T1R3 was proposed for brazzein, a sweet protein [11]. On the other hand, the cavity of the closed form formed by LB1 and LB2 of either T1R2 or T1R3 [11], [16] is suggested for small sweeteners as glutamate bound in the glutamate receptor [15]. In this study, we found that the various structures of low-molecular-weight sweeteners were recognized by the sweet taste receptor hT1R2–hT1R3 through the different residues at the ligand-binding site of the ATD of T1R2. Modes of binding between hT1R2–hT1R3 and low-molecular-weight sweet chemical substances were characterized both by response profiles of cells expressing the mutated hT1R2–hT1R3 to sweeteners and by a molecular modeling–based docking simulation at the binding cleft formed by LB1 and LB2 of hT1R2. The candidate amino acid residues at the binding cleft of hT1R2 were targeted to produce mutated hT1R2, which was then heterologously expressed in cultured cells together with hT1R3 and its coupling Gα protein. Using the functional analysis of cell-based assays, we successfully determined the residues responsible for binding to each sweetener in the ligand-binding cleft of hT1R2 ATD and found that individual molecules use characteristic residues for binding. A mechanism of receptor activation is also discussed according to a molecular model of the receptor–ligand complex.

Materials and Methods

Site-directed mutagenesis of hT1R2 cDNA

cDNA fragments with point mutations in hT1R2 were synthesized by the overlap PCR method using mutated primer pairs. The following 15 residues in hT1R2 were mutated individually to Ala: S40, K65, Y103, D142, S144, S165, Y215, P277, D278, Y282, E302, S303, D307, E382, and R383. In the cases of Y103, D142, Y215, P277, and R383, each residue was also replaced with residues other than Ala (Y103F, D142R, Y215F, P277G, P277Q, P277S, R383D, R383Q, R383L, and R383H).

Calcium imaging analysis of the heterologously transfected cultured cells

cDNA fragments were subcloned into the pEAK10 vector (Edge Biosystems, Gaithersburg, MD, USA). Each hT1R2 mutant was transiently cotransfected together with hT1R3 and G16-gust44 [20] into HEK293T cells (kindly provided by Dr. Hiroaki Matsunami, Duke University), and calcium imaging analysis was carried out as described previously [12]. Briefly, transfected cells were seeded into 96-well Lumox multiwell black-wall plates (SARSTEDT AG & Co., Nümbrecht, Germany). After 40–46 hours, the cells were loaded with 5 µM of fura-2/AM (Invitrogen, Carlsbad, CA, USA) in assay buffer for 30 min at 37°C, and then washed with assay buffer, prior to incubation in 100 µl of assay buffer for more than 10 min at room temperature. The cells were stimulated with sweet tastants by adding 100 µl of 2× ligands. The intensities of fura-2 fluorescence emissions resulting from excitation at 340 and 380 nm were measured at 510 nm using a CCD camera. The images were recorded at 4 sec intervals and analyzed using MetaFluor software (Molecular Devices, Sunnyvale, CA, USA).

Construction of stable cell lines expressing the mutated human sweet taste receptor

The entire coding regions of hT1R2, hT1R3, and G16-gust44 were subcloned into the pcDNA5/FRT vector (Invitrogen) according to the procedure described previously [21]. To generate the expression plasmid for the mutated receptor, the hT1R2 cDNA fragment with the point mutation was used instead of using the wild-type (WT) hT1R2 cDNA.

Stable cell lines expressing mutant hT1R2 together with hT1R3 and G16-gust44 were generated to prepare the following hT1R2 mutants: Y103A, Y103F, D142A, S144A, S165A, P277A, P277G, P277S, P277Q, D278A, E302A, D307A, E382A, and R383H. The stable cell lines were generated using Flp-In 293 cells (Invitrogen) and the plasmid we constructed according to the manufacturer’s protocol for the Flp-In pcDNA5/FRT Complete System (Invitrogen) as described in our previous publication [21]. Hygromycin-resistant cells were collected, cultured, and used to measure the cellular responses to sweet tastants. The cells for these measurements were cultured in low-glucose (1.0 g/l) Dulbecco’s modified Eagle’s medium with 10% fetal bovine serum.

Measurement of cellular responses by the cell-based assay

Trypsinized cells were seeded at a density of 80,000 cells per well into 96-well black-wall CellBIND surface plates (Corning, Corning, NY, USA) and 24 hours later were washed with assay buffer prior to loading with a calcium indicator dye from the FLIPR Calcium 4 Assay Kit (Molecular Devices) diluted with assay buffer. The cells were incubated for 60 min at 37°C, and measurements were made using FlexStation 3 (Molecular Devices) at 37°C. Fluorescence changes by excitation at 485 nm, emission at 525 nm, and cutoff at 515 nm were monitored at 2 s intervals, an aliquot of 100 µl of assay buffer supplemented with 2× ligands was added at 20 s, and scanning was continued for an additional 100 s. The response of each well was represented as ΔRFU (delta relative fluorescence unit) and defined as maximum fluorescence value minus minimum fluorescence value. To calculate EC50 values, plots of amplitude versus concentration were prepared in Clampfit Version 9.2 (Molecular Devices). Nonlinear regression of the plots produced the function:

equation image

where x is the ligand concentration and h is the Hill coefficient used to calculate the EC50 values for ligand–receptor interactions. When the EC50 value of the mutated receptor-expressing cells was changed more than 5 fold compared with wild type receptor, the corresponding mutation was judged to be largely affected.

Structure modeling of receptor and receptor–ligand complexes

The crystal structures of the ATD of mGluR1 solved in both inactive (glutamate-unbound) and active (glutamate-bound) forms (PDB: 1EWT and 1EWK, respectively) were used to construct the ATDs of hT1R2 and hT1R3. The structural model of the ATDs of the hT1R2 and hT1R3 heterodimer was constructed with homology modeling according to their sequence homology with mGluR1. For the active form of the heterodimer model, the closed form of mGluR1 was used for hT1R2 and the open form for hT1R3. Conversely, the open form of the crystal structure of mGluR1 was used to construct the inactive form of T1R2 and T1R3. Each heterodimeric structure was then energy-minimized with molecular mechanics using Discover 3 (Accelrys Inc., CA, USA), and the main chain was tethered at the conserved position.

Sweet small ligands were docked into the ligand-binding cleft of the hT1R2 model where glutamate is bound in the mGluR1; this was pursuant to the plausible interactions between the charged or hydrophilic groups of the ligands and the receptor that were deduced from the mutational experiments. Conformations of the ligands were then generated and energy-minimized with molecular mechanics using Discover 3. The minimized complex structures were then structure-optimized with molecular dynamics using Discover 3, and the residues were tethered beyond 12 Å from the ligands.


Mutagenesis studies for screening the residues responsible for sweetener recognition

To define the binding modes of sweeteners at the cleft formed by LB1 and LB2 of hT1R2 ATD, we carried out a series of mutagenesis studies on hT1R2 ATD. First, a molecular model of hT1R2 ATD based on the ligand-binding structure of the closed form of mGluR1 was constructed. Based on the residues resided in the glutamate-binding cleft in the structure of mGluR1 ATD, 15 residues of hT1R2 were arbitrarily selected to introduce the point mutation (Fig. S1, Table S1), and 25 single hT1R2 mutants for the 15 residues were then constructed. The selected residues were almost hydrophilic, and were expected to form ionic or hydrogen bonds with the ligands. The responses to sweeteners were examined by a calcium imaging assay using HEK293T cells transiently expressing the T1R2 mutant and T1R3. Ten out of the 15 residues (Y103, D142, S144, S165, P277, D278, E302, D307, E382, and R383) were selected from the results of the 25 mutants because receptors mutated at these 10 residues retained the responsiveness and exhibited largely changed activities toward the sweeteners tested (Table S1).

As for the 10 residues, stable cell lines expressing the hT1R2 mutant and hT1R3 were constructed, and the cell-based assay was performed to determine the dose–response relationship with the half-maximal effective concentration (EC50) value for each sweetener. To validate the activity of each mutated receptor, we used an artificial sweetener cyclamate, which was recognized by the TMD of hT1R3, as positive controls [22]. Because all the hT1R2 mutant cell lines clearly responded to cyclamate, showing similar EC50 values to those expressing the WT receptor (Table 1), the mutated receptors were determined to be functionally expressed. The response profiles of the mutated receptors to the sweeteners are summarized in Table 1.

Table 1

Summary of point mutations in hT1R2–hT1R3.

Residues responsible for aspartame and D-tryptophan reception in hT1R2 ATD

The response to aspartame was completely lost in the cell lines expressing E302A, S144A, D142A and Y103A (Fig. 2A), and EC50 values largely increased in those expressing D278A, with a decrease in potency (EC50 value 8.14-fold increase versus WT, Fig. 2B). These results suggest that the residues E302, S144, D142, Y103, and D278 are crucial for aspartame reception, among which E302 and S144 have also been previously reported as important residues for aspartame recognition [17].

Figure 2

Dose-dependent responses of hT1R2/hT1R3-expressing cells to amino acid derivatives.

In contrast, only the application of d-tryptophan (d-Trp) to E302A-expressing cells elicited no response (Fig. 2C), and large increases in EC50 values were observed for D307A, D142A, D278A, S165A, Y103A, and P277A mutants (>5-fold increase versus WT) (Fig. 2D and Table 1). Although aspartame elicited no response in D142A and Y103A mutants (Fig. 2A and Table 1), d-Trp considerably reduced the response potency to these mutants within an 8-fold EC50 increase (Fig. 2D and Table 1). In the cases of S165A and P277A mutants, EC50 of d-Trp increases 5.40- and 6.31-fold, respectively (Fig. 2D), while those of aspartame were only changed (Table 1). Although the carboxylate of aspartame and d-Trp is located near S165 and R383 in their complex models, the carboxylate of d-Trp would interact with S165, but that of aspartame would be located at slightly different position not to directly interact with S165. A similar case is also the interactions of P277 with d-Trp and aspartame, in which the indole moiety is locate closer to P277 than the phenylalanine moiety is. The roles of S165 and P277 in receptor activation are thus ligand depended.

Residues responsible for saccharin Na and acesulfame K reception in hT1R2 ATD

Saccharin Na and AceK activated WT hT1R2–hT1R3 in a dose-dependent manner at lower concentrations, but the response was suppressed at higher concentrations (>3 mM and >10 mM, respectively, Figs. 3A and 3B), which has been observed and investigated in detail by Galindo-Cuspinera et al. [14]. Therefore, EC50 values for saccharin Na and AceK were estimated at the lower concentrations.

Figure 3

Dose-dependent responses of hT1R2/hT1R3-expressing cells to sulfamates.

The cellular responses to saccharin Na and AceK were lost in R383H, D142A and E382A (Figs. 3A and 3B). These results indicate that R383, D142, and E382 are crucial residues for activation by saccharin Na and AceK. The mutations E302, S144 and D278 scarcely affected the EC50 values for saccharin Na and AceK, unlike aspartame and d-Trp (Fig. 2 and Table 1). Moreover, the other mutations tested in this study were not sensitive to saccharin Na and AceK (Table 1), suggesting that the binding region for saccharin Na and AceK is limited to a region around R383 (see Discussion).

Residues responsible for sucralose reception in hT1R2 ATD

The response to sucralose was almost completely lost in D278A and Y103A (Fig. 4A). E302A, D307A, D142A, and P277A largely increased the EC50 values of sucralose and decreased the potency (Fig. 4B). Most of the crucial residues for sucralose reception (E302, D142, Y103, D278, and D307) appeared to overlap with those for d-Trp and aspartame reception (Table 1). However, unlike aspartame, the EC50 value of sucralose for S144A did not change dramatically (0.27 mM), and P277A elicited a remarkable increase of the EC50 value. These results indicate that sucralose partially shares the binding region with aspartame, but also interacts with sucralose-specific residue such as P277.

Figure 4

Dose-dependent responses of hT1R2/hT1R3-expressing cells to sucralose.

Roles of Y103 and P277 at the entry of the lobes

Six out of the 10 critical residues (D142, D278, E302, D307, E382, and R383) are acidic or basic residues that probably bind to ligands via electrostatic interactions (Table 1). Furthermore, S144 and S165 were important for the reception of the amino acid derivatives aspartame and d-Trp, respectively (Figs. 2A and 2D). We next evaluated the role of the hydrophobic residues, Y103 and P277, located across the cleft of LB1 and LB2, respectively (See Discussion). To further examine the effect of Y103 on receptor activity, the responses of stable cell lines expressing additional mutants (in which Y103 was replaced with Phe in addition to Ala) were evaluated. When sucralose was applied to Y103 mutants, the response was almost completely lost in Y103A but was only slightly reduced in Y103F (Fig. 5A). These results indicate that the aromatic ring of Y103 is specifically essential to sucralose binding.

Figure 5

Roles of Y103 and P277 for the reception of the sweeteners.

To evaluate the role of P277, the additional mutants P277G, P277Q and P277S were constructed. The P277Q mutant showed severely reduced responses to aspartame (Fig. 5B) and d-Trp (Table 1), while P277G and P277S did not (Fig. 5B). In contrast, these three mutants responded almost equally to saccharin Na and AceK (Table 1). These results suggest that P277 plays an important role in allowing the sweet taste receptor to discriminate amino acid derivatives (aspartame and d-Trp) from the other sweeteners.


Critical residues for small molecular sweetener recognition in hT1R2 ATD

To clarify the roles of the 10 residues in small molecular sweetener recognition, we mapped them on the model of the open form of the hT1R2 ATD (Fig. 6). They were divided into four classes based on the results of a single point mutant analysis of hT1R2–hT1R3 corresponding to three chemically different types of ligands: amino acid derivatives (aspartame and d-Trp), sulfamates (saccharin Na and AceK), and a sugar analog (sucralose) (Table 1). Our data strongly suggest that the binding sites in hT1R2 ATD are quite different from each other, although all of them are recognized in the cleft of hT1R2 ATD. As shown in Figs. 7 and ​and8,8, aspartame, d-Trp, and sucralose share LB1 residues (Y103 and D142) and LB2 residues (D278, E302, and D307) for binding, but each compound also needs specific residues for individual interaction with the receptor (S144 for aspartame (Fig. 2A) and P277 for sucralose (Fig. 4B)). By contrast, these residues are not involved in binding saccharin Na and AceK, but the residues (D142, E382 and R383) located in another site of LB1 are indispensable for their binding (Figs. 6A).

Figure 6

Model of the open form of hT1R2 ATD.
Figure 7

Complex model of the sucralose-bound hT1R2 ATD.
Figure 8

Model of the aspartame-bound hT1R2 ATD.

The low-molecular-weight sweeteners bind in the cleft composed of LB1 and LB2 with a different binding mode at each characteristic residue. To examine further characteristic interactions between ligands and the 10 residues, we built ligand–hT1R2 ATD (closed form) complex models for sucralose, aspartame and saccharin Na (Figs. 7, ​,8,8, ​,9,9, Methods S1, S2, S3).

Figure 9

Complex model of the saccharin Na–bound hT1R2 ATD.

(i) Roles of Y103 at the entry of LB1 and D278 at the entry of LB2

The complex models of sucralose–hT1R2 and aspartame–hT1R2 suggested different roles of Y103 in receptor activation. The C2-H and C4-Cl of the hexose portion of sucralose bind to the aromatic ring of Y103 (Fig. 7C), and the hydroxyl groups in the hexose moiety of sucralose form hydrogen bonds with D278 (Fig. 7C). The binding of the hexose portion to Y103 in LB1 and D278 in LB2 may thus facilitate the formation of the closed form of hT1R2 ATD. The importance of these residues for binding of sucralose is consistent with the results reported by Zhang et al. [18].

Conversely, the phenol group of Y103 forms a hydrogen bond with D278 in the aspartame–hT1R2 model (Fig. 8C), stabilizing the closed form of hT1R2. The hydrogen bond appears to be important for the d-Trp-binding. However, the role of the phenol group in the aspartame-binding would be more significant, since the phenol group would interact with the carboxylate of aspartame. The phenol group of Y103 is thus important for the binding of aspartame, while the aromatic group is necessary for the binding of sucralose, as in the cases of the Y103A and Y103F mutants (Fig. 5A). On the other hand, Zhang et al. [18] suggested a contribution of a hydrogen bond between D278 and K65 to the stabilization of the closed form in the binding of sweet taste enhancers. However, a transiently expressed K65A mutant receptor did not show a significant difference from the native receptor in the binding of aspartame and sucralose (Table S1), being consistent with the results reported by Zhang et al. [18] and Liu et al. [23], in which K65 is not important for the binding of aspartame and sugar derivatives.

(ii) Roles of E302 at the center of LB2

The negatively charged E302 residue forms a salt bond with the positively charged amine group of aspartame (Fig. 8C), whereas a hydroxyl group of the pentose moiety of sucralose forms a hydrogen bond with E302 (Fig. 7C). E302 in the LB2 should thus be a crucial residue for the ligands, with hydrogen bond donors contributing to the formation of the closed form in receptor activation. In contrast, the E302 residue makes no electrostatic interaction with saccharin Na (Fig. 9C), so the contribution of this residue to receptor activation should be little, if any (Fig. 3B).

(iii) Roles of D142, E382, and R383 at the center of LB1

Because R383 forms a hydrogen bond network with D142 and E382 in the hT1R2 model, R383 plays a crucial role in the recognition of negatively charged groups of ligands (Fig. 9C). D142 or E382 may not directly interact with the negatively charged ligands but would play an important role in localizing the flexible R383 residue at a proper position for interacting with the ligands (Fig. 9C). For aspartame recognition, binding of both the carboxylate moiety to R383 in LB1 and the amino group to E302 in LB2 may facilitate the formation of the closed form of the ATD (Fig. 8C). The negatively charged group of saccharin and the cationic sodium ion attached to saccharin would play similar roles in the formation of the closed form (Fig. 9C). Liu et al. [23] showed that S40 and V66 contribute to the species specificity in the binding of aspartame. The S40 residue is located at the hydrogen bond distance to D142 and the V66 residue is close to R383 in the aspartame-bound model. The mutation of these residues would electronically and sterically affect the interaction of D142 and R383 which are important for the recognition of the carboxylate of aspartame. This is somewhat similar to the roles of S40 and V66 in the species specific recognition of aspartame.

The neutral ligand sucralose may directly interact with D142 through a hydrogen bond with the vicinal hydroxyl groups of the furanose moiety (Fig. 7C). This hydrogen bond probably leads to the formation of a hydrogen bond between R383 and E302 to facilitate receptor activation.

(iv) Role of P277 at the entry of LB2

Aspartame and saccharin do not bind P277 (Figs. 7C and ​and8C).8C). However, aspartame is located near the residue because the Gln mutant for P277 interrupts receptor activation by aspartame. In contrast, the mutation of smaller residues such as Gly and Ser does not affect activation (Fig. 7C). The smaller ligand, saccharin Na, may be located far from P277 and thus may not be influenced by the mutation (Fig. 9C). Still, P277 should be an important binding site for d-Trp, as observed in the P277A and P277Q mutants (Table 1). These results suggest that saccharin Na is located far from P277 whereas d-Trp is located close to P277. The distance between aspartame and P277 would be intermediate between those of saccharin Na and d-Trp.

The chloride at C1′ of the furanose moiety of sucralose showed favorable van der Waals contact with P277 (Fig. 7C), and the P277Q mutant caused unfavorable steric interactions with the chloride; however, the favorable hydrophobic interactions are lost in the P277G and P277S mutants (Table 1).

Characteristic features in receptor activation mechanisms of the human sweet taste receptor

As described above, the interaction at the core of LB1 and LB2 appears to be essential for reception of all the sweeteners, and the interaction at the entry of LB1 and LB2 would reinforce the formation of the closed structure of the receptor for activation. These results strongly suggest that the activation mechanism of the human sweet taste receptor is similar to that of mGluR1.

X-ray crystal structural analysis, molecular modeling, and many mutagenesis studies have revealed the existence of critical residues for ligand binding in other class C GPCRs, such as mGluRs [15], [24], [25], the GABA receptor [26], [27], the calcium sensing receptor [28], [29], and the human umami taste receptor (hT1R1–hT1R3) [30]. In comparison with previous data [31], our model of hT1R2–hT1R3 based on a mutagenesis analysis suggests that hT1R2–hT1R3 uses five acidic residues (D142, D278, E302, D307, or E382) for the recognition of its agonists; the other receptors use one or two acidic residues. These results suggest that hT1R2 ATD forms different sites of binding with specific sets of these residues to receive chemically diverse low-molecular-weight sweeteners, although their affinities for hT1R2 ATD are quite low.

It should be noted that we could not determine the binding mode of sugars such as sucrose. Sugars generally elicit the strong sweet taste, and they are the most common natural ligands for the receptor. Although it would be important to elucidate the key residues for the recognition of sugars, the cellular response to sucrose was quietly faint compared with the other sweeteners used in this study, and EC50 values of the mutated receptors to sucrose could not be accurately calculated. Further studies should be required to improve the sensitivity of the functional assay system for the human sweet taste receptor.

In this study, we defined how hT1R2–hT1R3 acquires the ability to recognize chemically diverse sweeteners. These results will not only provide insights into molecular recognition patterns of GPCRs but may also help develop novel sweeteners.

Supporting Information

Table S1

Summary of point mutations determined by a calcium imaging assay using HEK293T cells transiently expressing the T1R2 mutant and T1R3.


Figure S1

Sequence alignment of the ATDs of hT1R2 and rat mGluR1. The mutated residues in hT1R2 used for initial screening are shown in blue and magenta. Stable cell lines were also constructed for the residues shown in magenta. Critical ligand-binding residues in the rat mGluR1 ATD that interact with the carboxylate side chain and the α-amino acid moiety are shown in red and green, respectively.


Methods S1

Modeling for sucralose-T1R2ATD complex (Fig. 7A9C9C).


Methods S2

Modeling for aspartame-T1R2ATD complex (Fig. 8A–C).


Methods S3

Modeling for saccharin-T1R2ATD complex (Fig. 9A–C).



Competing Interests: The authors have declared that no competing interests exist.

Funding: This study was performed with a grant from the Research and Development Program for New Bio-industry Initiatives of the Bio-oriented Technology Research Advancement Institution. This work was also supported by the Japan Society for the Promotion of Science Research Fellowship for Young Scientists (to AK) and by Grants-in-aid for Scientific Research 21880015 (to KN), 20688015 and 21658046 (to TM) and 20380183 (to KA) from the Ministry of Education, Culture, Sports, Science and Technology of Japan. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.


1. Nelson G, Hoon MA, Chandrashekar J, Zhang Y, Ryba NJ, et al. Mammalian sweet taste receptors. Cell. 2001;106:381–390. [PubMed]
2. Li X, Staszewski L, Xu H, Durick K, Zoller M, et al. Human receptors for sweet and umami taste. Proc Natl Acad Sci U S A. 2002;99:4692–4696. [PMC free article] [PubMed]
3. Zhao GQ, Zhang Y, Hoon MA, Chandrashekar J, Erlenbach I, et al. The receptors for mammalian sweet and umami taste. Cell. 2003;115:255–266. [PubMed]
4. Pin JP, Galvez T, Prezeau L. Evolution, structure, and activation mechanism of family 3/C G-protein-coupled receptors. Pharmacol Ther. 2003;98:325–354. [PubMed]
5. Chandrashekar J, Hoon MA, Ryba NJ, Zuker CS. The receptors and cells for mammalian taste. Nature. 2006;444:288–294. [PubMed]
6. Temussi PA. Why are sweet proteins sweet? Interaction of brazzein, monellin and thaumatin with the T1R2–T1R3 receptor. FEBS Lett. 2002;526:1–4. [PubMed]
7. Jiang P, Ji Q, Liu Z, Snyder LA, Benard LM, et al. The cysteine-rich region of T1R3 determines responses to intensely sweet proteins. J Biol Chem. 2004;279:45068–45075. [PubMed]
8. Walters DE, Hellekant G. Interactions of the sweet protein brazzein with the sweet taste receptor. J Agric Food Chem. 2006;54:10129–10133. [PMC free article] [PubMed]
9. Cui M, Jiang P, Maillet E, Max M, Margolskee RF, et al. The heterodimeric sweet taste receptor has multiple potential ligand binding sites. Curr Pharm Des. 2006;12:4591–4600. [PubMed]
10. Nakajima K, Asakura T, Oike H, Morita Y, Shimizu-Ibuka A, et al. Neoculin, a taste-modifying protein, is recognized by human sweet taste receptor. Neuroreport. 2006;17:1241–1244. [PubMed]
11. Assadi-Porter FM, Maillet EL, Radek JT, Quijada J, Markley JL, et al. Key amino acid residues involved in multi-point binding interactions between brazzein, a sweet protein, and the T1R2–T1R3 human sweet receptor. J Mol Biol. 2010;398:584–599. [PMC free article] [PubMed]
12. Koizumi A, Nakajima K, Asakura T, Morita Y, Ito K, et al. Taste-modifying sweet protein, neoculin, is received at human T1R3 amino terminal domain. Biochem Biophys Res Commun. 2007;358:585–589. [PubMed]
13. Winnig M, Bufe B, Kratochwil NA, Slack JP, Meyerhof W. The binding site for neohesperidin dihydrochalcone at the human sweet taste receptor. BMC Struct Biol. 2007;7:66. [PMC free article] [PubMed]
14. Galindo-Cuspinera V, Winnig M, Bufe B, Meyerhof W, Breslin PA. A TAS1R receptor-based explanation of sweet ‘water-taste’. Nature. 2006;441:354–357. [PubMed]
15. Kunishima N, Shimada Y, Tsuji Y, Sato T, Yamamoto M, et al. Structural basis of glutamate recognition by a dimeric metabotropic glutamate receptor. Nature. 2000;407:971–977. [PubMed]
16. Morini G, Bassoli A, Temussi PA. From small sweeteners to sweet proteins: anatomy of the binding sites of the human T1R2_T1R3 receptor. J Med Chem. 2005;48:5520–5529. [PubMed]
17. Xu H, Staszewski L, Tang H, Adler E, Zoller M, et al. Different functional roles of T1R subunits in the heteromeric taste receptors. Proc Natl Acad Sci U S A. 2004;101:14258–14263. [PMC free article] [PubMed]
18. Zhang F, Klebansky B, Fine RM, Liu H, Xu H, et al. Molecular mechanism of the sweet taste enhancers. Proc Natl Acad Sci U S A. 2010;107:4752–4757. [PMC free article] [PubMed]
19. Walters DE. Homology-based model of the extracellular domain of the taste receptor T1R3. Pure Appl Chem. 2002;74:1117–1123.
20. Ueda T, Ugawa S, Yamamura H, Imaizumi Y, Shimada S. Functional interaction between T2R taste receptors and G-protein alpha subunits expressed in taste receptor cells. J Neurosci. 2003;23:7376–7380. [PubMed]
21. Imada T, Misaka T, Fujiwara S, Okada S, Fukuda Y, et al. Amiloride reduces the sweet taste intensity by inhibiting the human sweet taste receptor. Biochem Biophys Res Commun. 2010;397:220–225. [PubMed]
22. Jiang P, Cui M, Zhao B, Snyder LA, Benard LM, et al. Identification of the cyclamate interaction site within the transmembrane domain of the human sweet taste receptor subunit T1R3. J Biol Chem. 2005;280:34296–34305. [PubMed]
23. Liu B, Ha M, Meng XY, Kaur T, Khaleduzzaman M, et al. Molecular mechanism of species-dependent sweet taste toward artificial sweeteners. J Neurosci. 2011;31:11070–11076. [PMC free article] [PubMed]
24. Muto T, Tsuchiya D, Morikawa K, Jingami H. Structures of the extracellular regions of the group II/III metabotropic glutamate receptors. Proc Natl Acad Sci U S A. 2007;104:3759–3764. [PMC free article] [PubMed]
25. Tsuchiya D, Kunishima N, Kamiya N, Jingami H, Morikawa K. Structural views of the ligand-binding cores of a metabotropic glutamate receptor complexed with an antagonist and both glutamate and Gd3+. Proc Natl Acad Sci U S A. 2002;99:2660–2665. [PMC free article] [PubMed]
26. Galvez T, Parmentier ML, Joly C, Malitschek B, Kaupmann K, et al. Mutagenesis and modeling of the GABAB receptor extracellular domain support a venus flytrap mechanism for ligand binding. J Biol Chem. 1999;274:13362–13369. [PubMed]
27. Galvez T, Prezeau L, Milioti G, Franek M, Joly C, et al. Mapping the agonist-binding site of GABAB type 1 subunit sheds light on the activation process of GABAB receptors. J Biol Chem. 2000;275:41166–41174. [PubMed]
28. Brauner-Osborne H, Jensen AA, Sheppard PO, O’Hara P, Krogsgaard-Larsen P. The agonist-binding domain of the calcium-sensing receptor is located at the amino-terminal domain. J Biol Chem. 1999;274:18382–18386. [PubMed]
29. Hammerland LG, Krapcho KJ, Garrett JE, Alasti N, Hung BC, et al. Domains determining ligand specificity for Ca2+ receptors. Mol Pharmacol. 1999;55:642–648. [PubMed]
30. Zhang F, Klebansky B, Fine RM, Xu H, Pronin A, et al. Molecular mechanism for the umami taste synergism. Proc Natl Acad Sci U S A. 2008;105:20930–20934. [PMC free article] [PubMed]
31. Wellendorph P, Brauner-Osborne H. Molecular basis for amino acid sensing by family C G-protein-coupled receptors. Br J Pharmacol. 2009;156:869–884. [PMC free article] [PubMed]

Articles from PLoS ONE are provided here courtesy of Public Library of Science