In the healthcare industry, pay for performance (P4P), also known as “value–basedpurchasing”, is a payment model that offers financial incentives to physicians, hospitals, medical groups, and other healthcare providers for meeting certain performance measures.
In the United States, most professional medical societies have been nominally supportive of incentive programs to increase the quality of health care. However, these organizations also express concern over the choice and validity of measurements of improvement, which may include process measures that do not directly tie to outcomes. The American Medical Association (AMA) has published principles for pay-for performance programs, with emphasis on voluntary participation, data accuracy, positive incentives and fostering the doctor-patient relationship, and detailed guidelines for designing and implementing these programs. Positions by other physician organizations question the validity of performance measures, and whether it will preserve an individual physician’s clinical judgement, a patient’s preferences, autonomy and privacy. They question whether it will lower costs, although it will increase administrative costs.
- American Academy of Family Physicians: “there are a multitude of organizational, technical, legal and ethical challenges to designing and implementing pay for performance programs”
- American College of Physicians: “adoption of appropriate quality improvement strategies, if done right, will result in higher quality patient care leading to increased physician and patient satisfaction. But the College is also concerned that these changes could lead to more paperwork, more expense, and less revenue; detract from the time that internists spend with patients, and have unintended adverse consequences for sicker and non-compliant patients.” “… concerned about using a limited set of clinical practice parameters to assess quality, especially if payment for good performance is grafted onto the current payment system, which does not reward robust comprehensive care.”
- American Geriatrics Society: “quality measures (must) target not only care for specific diseases, but also care that addresses multiple, concurrent illnesses and (are) tested among vulnerable older adults. Using indicators that have been developed for a commercially insured population…may not be relevant”
- American Academy of Neurology (AAN): “An unintended consequence is that current relative payments are distorted and represent a misaligned incentive system, encouraging diagnostic tests over thoughtful and skilled patient care. The AAN recommends addressing these underlying inequities before a P4P program is adopted.
- The Endocrine Society: “it is difficult to develop standardized measure across medical specialties…variations must be allowed to meet the unique needs of the individual patient…P4P programs should not place financial or administrative burdens on practices that care for underserved patient populations”
In the United Kingdom, the National Health Service (NHS) began a major pay for performance initiative in 2004, known as the Quality and Outcomes Framework (QOF). General practitioners agreed to increases in existing income according to performance with respect to 146 quality indicators covering clinical care for 10 chronic diseases, organization of care, and patient experience. For example, family practitioners got points for clinically reviewing patients with asthma every 15 months. Unlike proposed quality incentive programs in the United States, funding for primary care was increased 20% over previous levels. This allowed practices to invest in extra staff and technology; 90% of general practitioners use the NHS Electronic Prescription Service, and up to 50% use electronic health records for the majority of clinical care.
A 2006 study found that most of the doctors actually did get most of the points, although some practices seemed to have gotten high scores by excluding patients with high-risk factors from their percentage targets. The 8,000 family practitioners included in the study had an increase in revenue by $40,000 by collecting nearly 97% of the points available.
A 2014 study examined 1.8 million hospital charts, and found that the mortality in control hospitals fell more than mortality in the pay-for-performance hospitals. Short-term improvements were not maintained. There was some evidence that mortality was declining in the three conditions covered by the program, at the expense of other conditions not covered in the program.
A 2015 population based study investigated the relationship between mortality and performance on the scheme across the whole of England. Although all-cause and cause-specific mortality rates declined over time, there was no significant relationship between practice performance on quality indicators and all-cause or cause-specific mortality rates in the practice locality. Higher mortality was associated with other well-known predictors: greater area deprivation, urban location, and higher proportion of a non-white population.
Responding to public backlash to managed care in the 1990s, California health care plans and physician groups developed a set of quality performance measures and public “report cards”, emerging in 2001 as the California Pay for Performance Program, now the largest pay-for-performance program in the country. Financial incentives based on utilization management were changed to those based on quality measures. Provider participation is voluntary, and physician organizations are accountable through public scorecards, and provided financial incentives by participating health plans based on their performance.
In the United States, Medicare has various pay-for-performance (“P4P”) initiatives in offices, clinics and hospitals, seeking to improve quality and avoid unnecessary health care costs. The Centers for Medicare and Medicaid Services (CMS) has several demonstration projects underway offering compensation for improvements:
- Payments for better care coordination between home, hospital and offices for patients with chronic illnesses. In April 2005, CMS launched its first value-based purchasing pilot or “demonstration” project- the three-year Medicare Physician Group Practice (PGP) Demonstration. The project involves ten large, multi-specialty physician practices caring for more than 200,000 Medicare fee-for-service beneficiaries. Participating practices will phase in quality standards for preventive care and the management of common chronic illnesses such as diabetes. Practices meeting these standards will be eligible for rewards from savings due to resulting improvements in patient management. The First Evaluation Report to Congress in 2006 showed that the model rewarded high quality, efficient provision of health care, but the lack of up-front payment for the investment in new systems of case management “have made for an uncertain future with respect for any payments under the demonstration.”
- A set of 10 hospital quality measures which, if reported to CMS, will increase the payments that hospitals receive for each discharge. By the third year of the demonstration, those hospitals that do not meet a threshold on quality will be subject to reductions in payment. Preliminary data from the second year of the study indicates that pay for performance was associated with a roughly 2.5% to 4.0% improvement in compliance with quality measures, compared with the control hospitals. Dr. Arnold Epstein of the Harvard School of Public Health commented in an accompanying editorial that pay-for-performance “is fundamentally a social experiment likely to have only modest incremental value.”
- Rewards to physicians for improving health outcomes by the use of health information technology in the care of chronically ill Medicare patients.
As a disincentive, CMS has proposed eliminating payments for negative consequences of care that results in injury, illness or death. This rule, effective October 2008, would reduce payments for medical complications such as “never events” as defined by the National Quality Forum, including hospital infections. Other private health payers are considering similar actions; the Leapfrog Group is exploring how to provide support to its members who are interested in ensuring that their employees do not get billed for such an event, and who do not wish to reimburse for these events themselves. Physician groups involved in the management of complications, such as the Infectious Diseases Society of America, have voiced objections to these proposals, observing that “some patients develop infections despite application of all evidence-based practices known to avoid infection”, and that a punitive response may discourage further study and slow the dramatic improvements that have already been made.
Multiple providers for complex disorders
Pay for performance programs often target patients with serious and complex illnesses; such patients commonly interact with multiple healthcare providers and facilities. However, pilot programs now underway focus on simple indicators such as improvement in lab values or use of emergency services, avoiding areas of complexity such as multiple complications or several treating specialists. A 2007 study analyzing Medicare beneficiaries’ healthcare visits showed that a median of two primary care physicians and five specialists provide care for a single patient. The authors doubt that pay-for-performance systems can accurately attribute responsibility for the outcome of care for such patients. The American College of Physicians Ethics has expressed concern:
Pay-for-performance initiatives that provide incentives for good performance on a few specific elements of a single disease or condition may lead to neglect of other, potentially more important elements of care for that condition or a comorbid condition. The elderly patient with multiple chronic conditions is especially vulnerable to this unwanted effect of powerful incentives.
Deselection, ethical issues
Present pay-for-performance systems measure performance based on specified clinical measurements, such as reductions in glycohemoglobin (HbA1c) for patients with diabetes. Healthcare providers who are monitored by such limited criteria have a powerful incentive to deselect (dismiss or refuse to accept) patients whose outcome measures fall below the quality standard and therefore worsen the provider’s assessment. Patients with low health literacy, inadequate financial resources to afford expensive medications or treatments, and ethnic groups traditionally subject to healthcare inequities may also be deselected by providers seeking improved performance measures.
In Minnesota, Minnesota Community Measurement ranks providers on multiple categories such as patient experience and total cost of care and provides public reporting online to inform consumers. The service is designed to help purchasers make better decisions when seeking care and to provide feedback to providers for areas that need improvement.