How should hospital doctors be paid?

Doctors are a significant portion of the costs for the health care system. In 2009, almost a quarter of Ontario’s health care budget was spent on paying doctors.

Most doctors who work in hospitals are not paid by the hospital directly, but by the Ontario Health Insurance Plan, which pays doctors for each procedure and clinical service. This system is known as “fee-for-service”.

There is a growing debate about whether doctors who work in hospitals should instead be paid by salary, instead of fee-for-service.

The history of paying doctors in Ontario

Prior to the advent of Medicare in the late 1960s, many doctors charged patients directly. The Ontario Health Insurance Plan (OHIP), created after the passage of federal legislation in 1966, allowed patients to receive medically necessary services in hospitals and doctors’ offices without having to pay directly. Instead of sending their bills to patients or private insurers, doctors began to bill OHIP, which gets its funding from the Ontario Ministry of Health and Long-Term Care. Doctors went from being able to set their fees independently to having to negotiate with the government. Every few years, the Ontario Medical Association, which represents doctors, and the Ministry of Health and Long-Term Care enter into negotiations to determine how much doctors are paid for each visit, diagnostic test and procedure. The current Schedule of Benefits for doctors is 812 pages long, and contains thousands of different fees associated with different consultations and procedures.

Hospitals are not directly represented in these negotiations, even though for many doctors, hospitals are where they deliver health care to patients. Hospitals provide doctors with the tools and equipment to practice as well as clinical space, a team of health care providers to work with, and, in some cases, administrative support for booking appointments and operating room times. Hospitals also incur the costs of all the procedures, tests and surgeries that doctors perform. For example, for hip replacement surgery, hospitals pay for the artificial hip joints, operating room staff and the equipment needed to do the surgery, while the surgeon bills OHIP for performing the surgery and providing post-operative care.

Fee-for-service vs. salary: incentives for care

Most Ontario doctors receive a substantial proportion of their income from fee-for-service, meaning a payment for each service that is provided, according to the fees listed in the Schedule of Benefits. The incentive is clear – see more patients or do more procedures, and you will get paid more.

Many health care policy analysts argue that these incentives can lead to unintended outcomes. Fee-for-service may encourage doctors to do unnecessary tests and procedures and neglect the time-consuming, financially unrewarding work of care coordination and chronic disease management. Carolyn Baker, CEO of St. Joseph’s Health Centre says that fee-for-service “rewards volume, not complexity.”

Many doctors, however, do not agree. Dr. Kenneth Pace, a urologist at St. Michael’s Hospital, says that fee-for-service encourages doctors to be as productive as possible. Pace says “in our system we’re constrained by resource availability” and that a shortage of doctors, plus increasing demand for health care services from the population means that “over-utilizing, and doing more procedures would just add to an already overbooked waiting list.” Indeed, unlike the USA, there are long waiting lists for some tests and operations in Canada, raising concerns that we are not providing enough care. If this is true, fee-for-service might be what we need. And in fact, governments have reduced wait times recently by remunerating hospitals on a fee-for-service basis for priority procedures such as cataract operations.

Doctors also value their independence. Many believe that fee-for-service ensures they are practicing medicine with only the best interests of each patient in mind. Other doctors argue that the modern practice of hospital medicine requires doctors to consider the financial constraints faced by hospitals and their funders. After all, a dollar spent on one patient is a dollar not spent on another. This perspective suggests that if doctors were paid by hospitals, they might better balance their patients’ needs with the need to control costs.

Opponents of the fee-for-service system also argue that the logic behind fee-for-service is becoming less appropriate as hospitals today are increasingly faced with complex, older patients who not only require procedures, but also care coordination and continuous disease management and surveillance. Under the current system, a hospital-based physician is paid each time a patient is readmitted. Successfully coordinating care and keeping a patient out of hospital paradoxically results in less remuneration for the physician.

Other payment methods, such as salary and capitation—paying doctors a fixed amount for each patient each year—have been proposed as a way to focus attention on complex care and quality.

Although these models have made significant inroads into primary care in Ontario, hospital doctors have seen relatively little change. One exception is that most doctors who work in Ontario’s cancer centres are paid with a combination of salary and fee-for-service.

Like fee-for-service, capitation and salary models have their own problems. Because the salary model provides doctors with the same salary no matter how many hours they work or how many patients they see, there can be an incentive for doctors to work fewer hours and see less patients. A study using data from the Canadian National Physician Survey in 2004 found that salaried doctors see about half the number of patients as doctors paid fee-for-service, although patient complexity was not considered in this study. Another study found that salaried doctors have different models of practice, and in general order less diagnostic tests and procedures but spend more time with each patient and provide more preventive care.

Models to align incentives

Carolyn Baker says that there are lessons we can learn on how to balance incentives for doctors from health care systems outside Canada. Kaiser Permanente, in California, pays doctors a salary and then provides them with the opportunity to earn additional bonuses based on performance incentives. The highly lauded Veterans Affairs system in the United States does the same. Even many for-profit health care organizations are moving towards models where doctors are hospital employees, with a recent study showing that more than half of American doctors are employed by a hospital or integrated health system.

“There is no one size fits all model for paying doctors” says Tom Closson, President and CEO of the Ontario Hospital Association. Closson notes the implications of the fact that “different doctors do different kinds of work” and that effective payment systems are reflective of the type of work each type of doctor does, as well as “what the organization they work for is accountable to achieve in terms of clinical service volumes, access and quality as well as academics.” Closson also notes the implications of the diversity of settings in which Ontario’s doctors work. He says ” a doctor in a small community will have a different role to play in practicing medicine than doctors in large communities and this needs to be reflected in the design of their compensation scheme to ensure that they have the appropriate financial support and incentives to best serve the patients in their community.”

Carolyn Baker agrees, saying “I don’t think that there is any one right solution.” She suggests that “there needs to be more experimentation, which is challenging because the current ‘union’ mentality of the Ontario Medical Association is focused on classic collective bargaining, to benefit the many, and doesn’t encourage experimentation.” Erik Hellsten, who works in the Negotiations and Accountability Management Division of the Ontario Ministry of Health and Long-Term Care agrees, saying “there is no perfect set of provider incentives. Each funding mechanism has its own strengths and weaknesses, and ultimately we make decisions around the policy consequences we can live with.”

The comments section is closed.

  • Laith Bustani says:

    I am a hospital doctor. I am paid primarily by fee for service and receive a stipend by the hospital to cover the additional work performed that is not directly related to patient care.

    The renumeration of hospital based physicians, especially those who act as Most Responsible Physician for admitted inpatients is currently undergoing intense multilateral review.

    It is generally recognized that the current fee for service system does not adequately capture or foster quality improvement initiatives by physicians. There is currently a “e code” program intended to transition physicians out of the hospital payroll, as hospital budgets are not intended to remunerate physicians. This is a 30% premium code on patient billings. A physician receives $ 55.45 to manage a patient on the 1st, 2nd, and last day of their hospital stay. Every other day they receive $29.20 unless they have a scheduled meeting with family. The e codes add 30% to these amounts in exchange for a direct reduction of the “top-ups” that physicians receive.

    Things that are not remunerated/rewarded are: telephone calls to family, to other care providers previously involved (note there is a new code that will allow billing of conversations with other physicians “telephone consults” greater than 10 minutes, which are rare – it usually takes 10 minutes to reach the physician and 2 minutes of conversation to sort out the problem). Multidisciplinary rounds and communication is not directly compensated. It is up to the hospital groups to develop their own standards/expectations for how often this should happen.

    I am actively involved in these negotiations and will be representing my colleagues at the OMA level after June 27th as the Chair of the Section of Hospital Medicine. As part of this role I hope to learn from my colleagues and the broader public how they perceive the system succeeding and failing in driving what is most important, providing the best health for those they are privileged and entrusted to help heal.

    • Laith Bustani says:

      In my haste and reckless enthusiasm, I forgot to recount my experience as a hospital doctor in a salaried system. I worked for two years where I was paid a salary and would contribute my billings back to the hospital. My billings accounted for about 40-60% of my salary (i.e. the hospital did not recoup its investment).

      I was paid more than I am now, but I also saw more patients and worked longer hours.
      We contributed to quality improvement initiatives, and as in most cases where physicians work as a group, met regularly to troubleshoot issues and work to improve the program.

      There was a gradual push by the hospital to increase our patient loads (and thus recoup more money) to the point where the more experienced physicians had to stand up and say: “i refuse to take any more patients”. You see, the physician is responsible to know their limits, and if they perceive that they jeopardize their patients by taking on additional responsibilities, then they are ethically bound to refuse. This is harder to do when you consider the “hospital”, and not “the patient” as the person you are working for.

      The disadvantage was we were treated like employees. There was “responsibility creep” (a problem everywhere during periods of budgetary constraints). On several occasions there would be imposed changes to the program that were somewhat mandated without fully understanding/adapting to the impact.

      In short, I’ve worked in both environments (FFS/Salary). They each have advantages/disadvantages and each can and should be refined. In both scenarios most physicians did what they could to provide the best care they knew how. Many (in both systems) leave/burn sooner than they would if the system had been more efficient and responsive to their patients needs. Until we hear why they left and what would bring them back, the care for our sickest, most vulnerable patients will not be as good as it should.

  • Doctor M says:

    If health was so simple, there wouldn’t be thousands in the health bureaucracy.

    I guess I can see the list of reasons why productivity drops – bed closures, teaching, more complex patients, nursing shortages, etc, etc, etc, Computers don’t sort that out and don’t make judgements.

    I think what we really need is an incentive system that has 4 legs – high volumes, high quality, lowest cost, chronic disease management. Sounds like a blend of FFS, P4P, capitation, and profit sharing.

    • Jeremy Petch says:

      I think it’s a bit rich to on one hand claim that alternative payment plans themselves decrease physician performance, but then claim on the other that any system that would hold physicians accountable for such a drop would be flawed, since a drop in performance of a physician under an APP might be attributable to other factors.

      Also, it was never suggested that computers would be used to make judgements, but rather that it is a fairly simple matter to alert a group of humans accustomed to dealing with such issues when an anomaly arises. Moreover, if drops in productivity were due to bed closures or nursing shortages, associated drops would be seen in entire departments, heading off any worries that these would lead to unfair disciplinary actions for individuals. In cases where a physician’s patient volume dropped due to spending more time managing chronic conditions for complex patients, any accountability program would likely congratulate the physician in question, since this is precisely one of the outcomes the move away from FFS is meant to facilitate. Patient volume is, after all, not the same thing as productivity.

      The example of computer monitoring with HR or supervisory action was raised to illustrate that the specter of needing ‘thousands in a health bureaucracy’ is a red hearing. Monitoring physician productivity under a salary system needn’t and shouldn’t be conducted at the provincial level – it will be done by established departments inside the individual hospitals at which physicians are employed.

  • Doctor M says:

    We have a clear example in Ontario that a large group that shifted from FFS to an APP – and saw a 30 percent drop in service. That drop was seen as an increase in surrounding communities – communities that had no increase in funding to deal with the increased demand.

    It’s pretty obvious, if you remove one incentive, you change behaviour. If you are paid the same amount regardless of whether or not you work hard to get that last case in, well you might just choose not to. That’s human. Doctors respond in the same way that others do.

    At the same time, those who advocate moving away from FFS seem to suggest that there is a big volume of overservice currently. I don’t believe it. We have a huge UNMET need that we currently are ignoring. I think it is likely appropriate that we figure out what work physicians should be doing, and incentivize that. I suggest that too much non-complicated work is being done by doctors that should be done by others. To suggest that a move from FFS to other models will reduce unnecessary work intimates that doctors currently provide services that are not medically required. Strong words. I’d counter that we need to find out how to start to address the myriad of issues/treatments/patients that are virtually abandoned in our current environment

    • Irfan Dhalla says:

      I agree with you that there is substantial unmet need (e.g., it is still difficult for many patients, particularly those with complex problems, to find family physicians) and that other health professionals could do some of what doctors do today. But there is also good evidence of unneccessary care in some areas — imaging would be one example. Each payment model has its pros and cons. Your opening example suggests to me that tinkering with the payment model alone will not get us very far. We might also need more accountability.

    • Jeremy Petch says:

      Doctor M, you have twice referenced research in Ontario that shows a significant drop in volume (30-35%), with no demonstrable benefits. Please provide a link this research. I would happily reconsider my position on the issue if there is in fact definitive evidence that it is mistaken.

  • Nicholas Leyland says:

    This is a very complex issue that requires different models for different physicians and for on call and “unsociable” hours work. Capitation is excellent in my view for complex work that incentivizes the provider to keep patients well. Appropriately alligned incentives and accountability agreements for services is the way to go. For emergency surgical care and after hours obstetrical work there needs to be some incentive to provide this care as these providers are called upon to work many more hours in the week than the 40 or 50 hours for many in society. The issue of productivity in surgical cases has to be managed or we will see what happened in some Alternate funding plans where volume dropped and patients had to go elswhere for care (the leakage phenomenon). Simply putting doctors on salary without addressing these complex issues will be problematic

  • Doctor M says:

    For those who wonder if changing the model changes behavior, I’m glad to be able to say that doctors are human. Full salary models reduce productivity – well known information in Ontario demonstrated a 35 percent reduction in volume, no demonstrated increase in quality, and no demonstrated increase in other outputs.
    If you want to change to salary models then you all of a sudden need a new bureaucracy to monitor/enforce the output expectations.
    People assume that in a FFS model that unnecessary work is being done. It may be true in some areas but in the vast majority of areas, the provider sees no benefit from doing additional work other than the fee – and the last time I checked there seems still to be a gross underservicing of unmet need in this province.
    Salary models make great sense if you want to reduce productivity and health care expenses but I can’t see any way that it helps the population

    • Jeremy Petch says:

      I disagree that changing salary models for hospital based physicians suddenly requires the implementation of a new bureaucracy. Businesses the world over have become very adept at managing the performance of salaried employees. A simple computer program can easily monitor physician productivity through a hospital’s EMR and automatically notify the Human Resources department when a physician’s productivity drops off. This need not be cumbersome or expensive.

      • Laith Bustani says:

        Keep in mind, physicians are not employees. HR has no interaction/authority over physicians. Physicians apply for “privileges” to work within a hospital. CEOs are now becoming more involved with the hiring process, but once hired, it is up to the other physicians to monitor each other. I have seen examples of where non-physician managers try and get involved and it is not a pretty sight.

        EMR software logs will not provide a true reflection of a physicians productivity. I am keenly interested in metrics in health care as a means of advancing efficient practices. I have the luxury of doing most of my documentation on a laptop that connects to the hospitals EMR. I have been experimenting with automated time management software for several months now. I track and analyze my own progress. I am actively controlling the quality of the data metrics it records. It tells me how much time I spend on various activities (writing, email, etc.) and lets me create categories for non-computerized activities. When I come back to my computer after having spent time with a patient, it asks me “What have you been doing”, and i’ve programmed it to capture the most descriptive/common activities.

        I have ownership of this data. No one else has access to it.

        The hospital also collects and analyzes performance data on me attributed to length of stays, turn around time for dictations, patient satisfaction surveys, etc. I am not given full access to this data and it tends to reassure/embarrass me, but not help me become more efficient.

        Also, something that is not talked about but is very, very significant is the impact of professional reputation amongst your peers. This is the most important metric for most physicians when they are thinking about who to refer their patients/family to. Very difficult to measure from computer data, but go and ask five doctors who the “go-to guy” is for a given condition/procedure and you’ve got your answer.

      • Jeremy Petch says:

        I am aware that physicians are not currently employees of hospitals. The discussion Doctor M and I have been having is around the claim that some have made that any move to an alternative payment plan would necessarily result in a drop in physician productivity. This argument operates on the premise that alternative payment plans remove incentives for physicians to be productive. The problem with this line of argument is that it relies on the assumption that a move to an alternative payment plan would not be an accompanied by a different sort of incentive: accountability.

        In discussing human resource departments, I was outlining one way such an accountability system could work, a system that could ensure accountability for productivity without creating a burdensome layer of provincial bureaucracy. While your comments above make it clear that your own arrangement does not make you an employee of your hospital, and it is certainly interesting to hear how your hospital does things, other alternative payment plans do exist, some of which do make physicians employees.

      • Laith Bustani says:

        I agree that a well designed AFP for hospital medicine has potential and should be fleshed out and demonstrated with keen and well matched centres. I would love your help to put together a proposal for one for the upcoming 2012 negotiations. There are a lot of people who are working on this as we speak, and maybe we need to facilitate them (us) putting their (our) heads together. What background are you coming from? Transparency and full disclosure of conflicts of interest is an absolute prerequisite to constructive participation in this process.

        I agree that engineering in metrics could help keep people honest, and help front line staff care for their patients more efficiently. What metrics do you think should be collected? How should we use this data once we have collected it?

        Let’s get a handful of dedicated people with a broad skill set together, set aside their various vested interests, and work on building a better model.

        More importantly, I think we should expand this intellectual exercise and ask up front: “how are we going to measure/define efficiency/effectiveness?” How do we know if what we are creating is better than what we had? The users are the best judges. This includes both the patients moving through the system and the providers working within it.

        I can’t tell you how many times poorly managed implementations of what would otherwise have been a great idea have negatively impacted patient care by placing extra, often redundant and inefficient,steps to a complex and highly individualized workflow (both at the level of the hospital and the provider).

        But we’re not going to make that mistake. We are going to learn from others experiences (and our own). We are going to ask the front line workers who do this day-in-day-out what they think would work for them. We are going to ask not just in Toronto, but in Renfrew and Sioux Lookout. We are going to ask each and every doctor/nurse/patient in each and every hospital how we can help them do their job more effectively. And we are going to listen. And we are going to have to work our buts off to earn their trust and be upfront about our own vested interests/ambitions/biases, and recognize that we are going to produce an imperfect solution. (We only have six months to get the first draft ready).

        We are going to implement information technology ourselves to organize all this data and analyze it and try and digest the layers of complexity and extract truths from it.

        In the end, we need to let individual “units”, be it clinics, hospitals, decide what is right for them. We don’t ram it down their throats and tell them they are all a bunch of dinosaurs that should be forced into retirement. If they tell us “we don’t want this”, then we ask them why and go back to the drawing board and try again.

        It’s a long road ahead. Personally, I think we need some radical thinking on the part of our leaders. Everyone has a stake in healthcare. It should be a civic responsibility to contribute to the prosperity of the system (in its broadest sense – including environmental and public health). I’m not talking about taxes, but feedback and stories and ideas and energy. I’m bias, but I don’t think anything is more important to our country’s future.

  • Jeremy Petch says:

    No doubt the most common worry when discussing a move away from fee for service is that physician productivity will fall without the adequate financial incentives. I am not convinced that this is really much of an issue.

    First, high productivity is endemic to the culture of medicine. The profession has long recruited only those who display a nearly super-human work ethic. Medical school is grueling and it culminates in years of residency, where newly minted physicians put in 100+ hours a week for nearly no pay (and certainly no overtime!). Some economists might argue that these trainees are simply responding to some future incentive – the promise of a good income, but the truth is that most have internalized an incredible work ethic and they will keep producing regardless of how they are compensated.

    Second, the argument that productivity will drop under a salary system presupposes that physicians would retain their current independence, where they are essentially entrepreneurs who operate within a hospital. But a key component of the move to a salary system is to make physicians employees of the hospital, and thus accountable to hospitals for issues like productivity. We have enormous experience under fee for service to tell us how many patients particular physicians can treat effectively in a week. A physicians whose productivity drops after moving to salary can easily be identified by hospital administration, to whom he/she would now be accountable, and disciplined appropriately – just like any other employee of any other organization. A move to salary does not remove incentives completely – it simply removes one set of incentives (economic), and replaces it with another (accountability).


Karen Born


Karen is a PhD candidate at the University of Toronto and is currently on maternity leave from her role as a researcher/writer with

Terrence Sullivan


Terrence Sullivan is an editor of Healthy Debate, the former CEO of Cancer Care Ontario and the current Chair of the Board of Public Health Ontario.

Republish this article

Republish this article on your website under the creative commons licence.

Learn more