I am unsure if I would advertise it as a proven success but in the UK there is the general practice quality and outcomes framework - a pay for performance system that is linked to the provision of specific (largely evidence based) services to patients as well as performance on some clinical (proxy) outcome measures. So practices get points for measuring the blood pressure of people with diabetes (proportion checked in the last 12 months) and for the proportion with HbA1C below certain thresholds. Points translate into enhanced payments. There is some evidence (see Martin Roland's work including his NEJM paper) that quality has increased:

Campbell, S.M., Reeves, D., Kontopantelis, E., Sibbald, B., Roland, M., 2009. Effects of pay for performance on the quality of primary care in England. New England Journal of Medicine 361 (4), 368-378.

Also in England there are commissioning arrangements where in addition to volume based tariffs / reimbursement structures there is reimbursement for the quality of services. For example a component of payment might be linked to the quality of patient feedback. There are also punitive tariff regimens designed to enhance quality - so no payments for unplanned re-admissions to hospital (there is some evidence on this from the US I think).

All of these initiatives attempt to counter simple "quantity not quality" incentives. Of course all are heavily dependent on the availability of data!

Here is the link for the Summit we organize in Skopje, Macedonia http://healthgrouper.com/en/page/Summit-P4P-Macedonia-first-announcement

...I would add that the UK system has not been universally welcomed by doctors. There is certainly evidence of some gaming the system and concerns that issues incentivised on the tariff are prioritised (which is fine) t o the detriment of issues that are not (so unintended consequences). As far as the evidence stacks up it is broadly favourable overall but there remains doubt.

In Catalonia (Spain), GPs are payed regularly and once a year they receive an extra salary linked to the health outcomes of their patients. Health outcomes are measured by the doctor or analytical parameters. This objectives should be established months before the work had to be done, but unfortunately this is not the usual case. And now, with the budget cuts, this economic incentives are decreasing dramatically.

Yes, dear colleagues, I agree with all your comments. Thanks a lot. There is no ideal payment method for providers as health care services and outcomes are interconnected with various factors which are not easy measurable with real quantification and side effects prevention and control. That is why the market mechanisms are not easy implementable in the health care. I prepared a teaching module for the teachers and students of the postgraduate MPH studies in the South Eastern Europe Region on Payment Methods and Regulation of Providers, published in the Book entitled Health Systems and Their Evidence Based Development, 2004. Here is the link for those who are interested to read it: http://www.snz.unizg.hr/ph-see/publications.htm. Payment for performance is somehow new method introduced in many countries within the last about ten years without sufficient evidence for desirable positive effects so far. Unfortunately, the side effects are predominant in many countries as some of you pointed out. That is especially the case when the budget neutral model is implemented taking 20% from the salary of doctors with low performance and rewarding 20% to the salaries of doctors with high performance measured just by the number of services performed without introducing the indicators for the quality of services and health outcomes. This model was implemented in Macedonia within the last about five months, from July 2012, and caused a lot of turbulence and strike of doctors for the last five weeks.

I agree with the discussants that all payment methods have benefits and flaws.generally,payment per performance in terms of fees-for-service is a passive methods of payment that incetivises providers to gives as much services as possible without due attention to quality and efficiency.However,some methods like case-based payments gives some incetive to get the right input mix which may skew the provider towards cost effective quality services since the money follows the patient.the assumption here is that the patient has a choice over the provider they recieve care from.In that case,the better the quality ,the more patients the provider will attract and the more emoney they will receive.The Macedonian budget neutral approach of reducing salaries for some while increasing for others is surely bound to fail.The ideal should be a basic salary with top-ups for extra services.By the way,what happens when thosewho perform better exceed those who perform poorly?

In the US, P4P has used quality measures for some time. They are not perfect and work on them continues, particularly in the area of clinical outcomes. Here are three articles that might help.

http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&ved=0CDcQFjAA&url=http%3A%2F%2Fwww.rti.org%2Fpubs%2Frtipress%2Fmitchell%2FBK-0002-1103-Ch04.pdf&ei=TAPmUMjPDYyL0QGJ94GgDA&usg=AFQjCNFJ5nc9OgpulgWNCH_qCAHTpCL43A&sig2=qRtON5vawlhl7UBYHJtZSQ&bvm=bv.1355534169,d.dmQ&cad=rja

http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=2&ved=0CEMQFjAB&url=http%3A%2F%2Fwww.cms.gov%2FRegulations-and-Guidance%2FGuidance%2FFACA%2Fdownloads%2Ftab_H.pdf&ei=rwPmUNeoC9S50QHR3oGwBg&usg=AFQjCNHIYbV0iFNhNDrTWFwWKFm9PDNwDQ&sig2=cwZRiBJ7oH-yg03yC2TusQ&bvm=bv.1355534169,d.dmQ&cad=rja

http://www.healthaffairs.org/healthpolicybriefs/brief.php?brief_id=78

Pay for performance (P4P has two components: (1) up front incentive (usually additional "bonus" money) for doctor to provide better care (with appropriate outcomes measures that reflect "better care"), and (2) a safety net "back-end" system to detect poor clinical practices that can harm patients. Some recent studies question the value of P4P in the U.S. but there are few if any studies that examine the relative amount of "bonus" money threshold necessary to give doctors incentive to apply more effort and time to achieve improved care and clinical outcomes. If a doctor has to take extra time that 2 more office visits could be scheduled just to improve the care for a single patient and receive a "bonus" equivalent of 10 percent of an office visit, then the doctor has a net negative financial loss: 0.10 visit minus 2 visits. This is basic application of incentives, behavior, and rewards in the field of economics but it has not been used much in the health care cost/payment arena. The "safety-net" for patients usually involves systems that include unannounced "spot-check" site visits and surveys of sampling of individual case patient records, anonymous "hot-line" reporting of suspected events of clinical mismanagement to a regulatory authority/agency, and centralized periodic monitoring (automated and computerized) of sentinel events that can trigger an investigation. In other words, the doctors know that "the system" is designed to detect and take action on poor clinical care.

P4P must be linked to health outcome measures. In the U.S we are beginning to implement shared risk / shared reward structures, known as accountable care organizations. The ACOs will be responsible for the provision of care for their panels at or below the previous years reimbursement levels for their specific patients while meeting certain quality guidelines. If net savings are realized the ACO shares that savings with the payer (usually the federal government). It must be reinforced that this is an outcome based scheme, that is patients must remain healthy (relatively speaking). An ACO cannot simply withhold care in order to save money because in the long term they would fail to meet health outcome criteria.

This represents a shift from episodic, reactive intervention to actually paying for "health".

One of the biggest challenges to P4P is creating a system that risk adjusts for the population that the provider or facility is serving. Without an equitable risk adjustment system, there will not be substantial buy-in from providers. This issue is particularly true for P4P schemes that attempt to provided rewards/penalties for comprehensive case management, as opposed to simply incenting the referral of the patient to certain services (eg, glaucoma exams, Hb1ac tests, etc.)

I'm impressed, as a thirty year veteran of solo and group private practice, and now a faculty member, by the confidence expressed of health care managersin their ability to manage the doctor patient relationship. I'm perplexed by the blizzard of mandates. Unlike my patientsn the requirements seem to be one size fits all.

I'm a surgical sub specialist. The limited personnel resources in our clinic are increasingly consumed complying with requirements for ritual measurements unrelated to my patient's needs.

I'm also bemused by the evolution of the notion of "quality of care". It appears to me the definition of quality in these discussions is compliance. In my view," quality of care" is very much like "pornography"...hard to define, but we know it when we see it.

Taylor, who designed the assembly line for Henry Ford set up a system which ultimately led to General Motors which finally smothered under the weight of the mamagement required to keep it all functioning. Taylorism in medicine will not produce a better outcome...and certainly not a less expensive one.

"Quality of care" is not too difficult to define for large deviations from basic outcomes such as complications after surgery like wound infection rates, bleeding (transfusions), unplanned return to OR, bowel obstruction, postop pneumonia, deep venous thrombosis/pulmonary embolism, acute MI or stroke, etc. These events are easily documented and the denominator patient population can be risk-adjusted to account for acuity or severity of patient medical status and underlying surgical problem. The key is adequate risk adjustment and large denominators to assure statistical significance and not simply "chance" occurrence. On the other hand, process measures as indicators of quality care are more problematic because many "accepted" process measures are in fact nothing more than expert opionion and have not been validated to actually result in improved outcomes in large enough populations sufficient to demonstrate statistical significance (and assuming clinical significance also).

I wonder if the people who constructed "Pay for Perfomance" in the US have read the book 'Drive" by Daniel Pink. I agree that looking at oucomes is important but while in theory, carrots and sticks may motivate good behavior or discourage bad behavior, Pink's book turns some the idea of carrots and sticks on its head based on research in areas of work outside medicine. For example, one idea he puts forth is P4P doesn't work as well if people are aren't paid a minimally satsifying wage to begin with. As a primary care doc, I know most of us are compensated better than the average US worker but may have left the field because overhead, malpractice, and other costs have outstripped income and generally, primary care docs aren't paid as well for thinking as for doing procedures even if the former takes more effort. As well, Pink writes about autonomy (whether it's of time, who you work with, how you work, etc.) and how this is a motivating factor as well yet physician autonomy in the US continues to be restricted.

Not working for Amazon here but here's the book:

http://www.amazon.com/Drive-Surprising-Truth-About-Motivates/dp/1594484805

I take it as a bad sign when innovations require carrots and sticks. The copier, the word processor, the FAX machine and the computer didn't need carrots or sticks to transform how we saw patients. Their benefits were obvious.

Carrots and sticks are often signs of bad ideas.