Studies with data and outcome information designated as influential and important to P4P across both United States and international health care systems are included in this section. Individual sections exist for data and outcomes specific to:
- United States - National
- United States - State
- United States - Local/Private Insurance
- Provider Viewpoint
A systematic review by Lin showed positive clinical effects of P4P for most diseases, but indicated that implementation may bring about negative unintended consequences, particularly related to health equity (420). In a large blow to P4P, the United Kingdom's Quality and Outcomes Framework (QOF) showed no significant association with improved population mortality for any assessed disease area, including those targeted and not targeted by the QOF when data from before and after its 2004 implementation was analyzed (425).
The basic question of whether P4P is successful in achieving better health outcomes is actually quite complex. The results vary based on the primary goal of the P4P program, most commonly quality improvement, which has proved to be challenging to measure. It is rare that programs actually decline in overall performance; so the rate of quality improvement is an important consideration as well. Additionally, a significant limitation for most studies is provider competition. If one hospital demonstrates improvement, market forces drive the other hospital to improve as well. Therefore, P4P projects are extremely difficult to provide control groups for. Perhaps the best example of this can be seen by Tu, et al.'s work (317) in the Public Reporting section of Additional Performance Based Programs.
Website Authors' Opinion
Data thus far has been mostly positive. Most trials are not ideal for rigorous scientific investigations, and that could be detrimental in the eyes of some clinicians, yet satisfactory in others. Meanwhile, most programs have only rewarded physicians for process measures, mainly because establishing a link to outcomes is difficult, and getting bonuses from outcomes of patients would be less convincing for many physicians' approval. Therefore, the data is skewed towards programs that are highly unlikely to fail. This is probably the proper first step for P4P programs because unintended consequences, and potentially severe repercussions in the form of lowered patient trust, must be avoided. Either way, there is much more work that must be done to research the effectiveness of P4P, and the United States offers a great chance to analyze many different P4P implementations.
General Data and Outcomes Literature
- Systematic review of 44 studies to assess the impact of P4P on primary care physician behavior and patient outcomes
- Overall positive effect was found for the management of disease although process outcomes often improved more than endpoint outcomes
- Baseline quality of medical care and the size of practice both limit performance improvement
- Unintended consequences associated with P4P were found to include:
- Rising medical costs for programs without financial metrics
- Inconsistent effects on health equity - some programs improved and some programs exacerbated inequities related to sex, age, ethnicity, socioeconomic status, comorbidity/severity, duration of illness, and size of practice.
- Inconsistency in patient satisfaction (some increased and some decreased patient satisfaction)
Significance to Literature:
Evidence for P4P has shown positive clinical effects for most diseases, but implementation may bring about negative unintended consequences, particularly related to health equity.
(425) Ryan et al. Long-term evidence for the effect of pay-for-performance in primary care on mortality in the UK: a population study. Lancet. 2016 Jul 16;388(10041):268-274. doi: 10.1016/S0140-6736(16)00276-2. Epub 2016 May 17.
- The United Kingdom’s (UK) Quality and Outcomes Framework (QOF), introduced in 2004, is the world’s largest primary care P4P program
- Population-level mortality statistics from 1994-2010 for the UK and other high-income countries were assessed
- Primary outcome was mortality per 100,000 population for “a composite outcome of disease areas that were targeted by the QOF from the beginning of the program”
- Secondary outcomes included mortality for ischemic heart disease, cancer, and a composite of all causes of death not included in the primary outcome
- No significant association with improved population mortality was found for any assessed disease area, both those targeted and not targeted by the QOF
Significance to Literature:
Results indicates that the viability of P4P to improve population mortality is questionable, and the comparison of the cost-effectiveness of P4P to other health system interventions is necessary
(439) Ogundeji YK, Bland JM, Sheldon TA. The effectiveness of payment for performance in health care: A meta-analysis and exploration of variation in outcomes. Health Policy. 2016 Oct;120(10):1141-1150. doi: 10.1016/j.healthpol.2016.09.002. Epub 2016 Sep 5.
- With P4P popularity continuing to increase despite considerable variation in the results of program evaluations, a systematic analysis of P4P schemes worldwide was conducted
- Each of the 37 studies included were categorized according to whether or not the study found a positive effect
- Analysis concluded:
- Randomized controlled trials had lower estimates of effect than studies with no controls
- Studies measuring outcomes (e.g. hospital mortality, smoking cessation) showed no effect from P4P
- Studies measuring intermediate outcomes (e.g. blood pressure and cholesterol control) showed modest effects from P4P
- Studies measuring process (e.g. cancer screening, smoking cessation counselling) showed a significant positive effect from P4P
- Likelihood of showing a positive effect was three times higher for schemes with larger incentives (defined as >5% of salary or usual budget)
- Reducing the perceived risk of not receiving the incentive payment and paying incentives to individuals instead of groups both showed non-statistically significant positive effects
Significance to Literature:
Effectiveness of P4P schemes are likely inflated due to prevalence of poorly designed studies which focus on process measures, but certain characteristics such as size of incentive can improve the effect of schemes on provider behavior