|
Does performance-based remuneration for individual health care practitioners affect patient care? A systematic review |
Houle SK, McAlister FA, Jackevicius CA, Chuck AW, Tsuyuki RT |
|
|
CRD summary The authors concluded that Pay-for-Performance modestly improved preventive activities, such as immunisation rates, but there was little evidence that it was effective for other outcomes. Implementation of these schemes should be experimental. This was a well-conducted review and the authors' conclusions reflect the findings and seem reliable. Authors' objectives To assess the effect of Pay-for-Performance remuneration, for individual health care practitioners, on the patient care outcomes. Searching PubMed, EMBASE, The Cochrane Library, OpenSIGLE, the Canadian Evaluation Society's Unpublished Literature Bank, and the Grey Literature Collection of the New York Academy of Medicine's Library were searched up to June 2012. Search terms were reported. Reference lists were manually searched. Study selection Eligible for inclusion were randomised controlled trials (RCTs), interrupted time series, uncontrolled and controlled before-and-after studies, and cohort comparisons. Eligible studies had to compare the effects of Pay-for-Performance versus at least one other payment model, or compare performance before versus after the initiation of Pay-for-Performance. Remuneration schemes had to target health care practitioners and payment had to be made on the basis of achievement of quality indicators in patients under the practitioner's direct care. There were no limitations on the patients, remuneration scheme variables, study duration, or patient-related outcomes. Included studies were conducted in the USA, Canada, UK, or Germany, between 1995 and 2012. Some studies were of patients with diabetes, most studies assessed the effects of Pay-for-Performance on care for chronic medical conditions, others assessed preventive care or screening, or both preventive and long-term care. Where reported, most of the comparator groups used Fee-for-Service or Quality and Outcomes Framework, other comparators were capitation or salary, as defined in the review. Outcome measures for preventive care or screening included change in rates of patients receiving a variety of vaccinations, blood tests, screening tests for different cancers, foot or eye examinations, and advice on taking daily aspirin. For chronic conditions they included quality of care scores for coronary heart disease, asthma, and diabetes; blood pressure; hypertension-related outcomes; change in drug treatments; and provision of smoking cessation advice. Two reviewers independently screened studies for inclusion. Disagreements were resolved by consensus. Assessment of study quality Two reviewers independently assessed study quality, according to the Cochrane risk of bias scale, which included criteria for allocation concealment, similar baseline characteristics, complete outcome reporting, and protection against contamination. Discrepancies were resolved by consensus and data were checked by a third reviewer. Data extraction Two reviewers independently extracted outcome data, including probability values, odds ratios or relative risks, 95% confidence intervals, and percentage changes. Study authors were contacted for further information, where necessary. Discrepancies were resolved through consensus and data were checked by a third reviewer. Methods of synthesis Due to substantial heterogeneity across studies, the data were presented as a narrative synthesis, grouped by study design and performance-related target (care for chronic medical conditions, or preventive care and screening). Results of the review Thirty studies were included in the review. Four were RCTs, with more than 200 physicians and more than 11,648 patients; five were interrupted time series, with at least 633,104 patients; three were controlled before-and-after studies, with at least 21 physicians and 21,419 patients; one was a non-randomised controlled study, with 110 physicians and two assistants; 15 were uncontrolled before-and-after studies, with at least 2,865 health care providers and 3,821,681 patients; and two were uncontrolled cohort studies, with 948 physicians and 6,194 patients. The quality of the studies was generally low to moderate; only RCTs had comparable baseline characteristics and only one study had adequate patient allocation concealment (full results were reported). Follow-up was between six months and nine years. Preventive care or screening: Ten studies were found. Two RCTs showed small improvements in immunisation rates with Pay-for Performance versus Fee-for-Service (p<0.05 and p=0.03). One RCT showed no significant differences in mammography referral between intervention groups. Two controlled before-and-after studies showed improved cervical screening (p=0.02) or influenza vaccination rates (OR 1.79, 95% CI 1.37 to 2.35) with Pay-for-Performance, but no other findings were statistically significant. Three uncontrolled before-and-after studies showed statistically significant improvement after the introduction of Pay-for-Performance (vaccination, cholesterol screening, colorectal cancer screening, and glycated haemoglobin testing; all p<0.01). One non-RCT and multivariate analysis of one cohort study showed no statistically significant differences in outcome measures between intervention groups. Chronic conditions: Twenty studies were found. One RCT reported no statistically significant differences in smoking cessation rates between intervention groups. Thirteen uncontrolled before-and-after studies reported mixed results, but most indicated improvements in quality of care after the implementation of Pay-for-Performance. Multivariate analysis of a non-RCT showed no statistically significant differences in outcomes between intervention groups. Five interrupted time series showed inconsistent results; two reported significant improvements in blood pressure and total cholesterol targets, one reported initial improvements in quality of care for diabetes and asthma, but this dissipated over time. Two studies reported no statistically significant differences in quality of care between intervention groups. Authors' conclusions Evidence suggested that Pay-for-Performance modestly improved preventive activities, such as immunisation rates, but there was little evidence that it improved other outcomes. Implementation of these schemes should be experimental and not considered to be evidence based. CRD commentary The review question was clearly stated; the accompanying inclusion criteria were broad. A comprehensive literature search was undertaken and included sources of unpublished data, reducing the potential for publication bias. Each stage of the review was performed in duplicate, minimising the potential for reviewer error and bias. Study quality was assessed and indicated that there was a risk of bias in the studies. A large amount of evidence was included in the review. Given the heterogeneity among studies, a narrative synthesis, based on study design, was appropriate. The authors acknowledged the limitations of the uncontrolled studies and the inability to draw firm conclusions from them. This was a well-conducted review and the authors' conclusions reflect the findings and seem reliable. Implications of the review for practice and research Practice: The authors stated that the evidence was insufficient to support the widespread implementation of the pay schemes into health policy. Research: The authors stated that further high quality research was needed to assess the potential of Pay-for-Performance to affect patient care, outcomes, and the cost of health services. This research should evaluate organisational factors that facilitate or impede the implementation and effectiveness of Pay-for-Performance, and it should focus on the best motivators for changing professional behaviour. They stated that research into the effects of Pay-for-Performance on other health care providers, such as nurses and pharmacists, was urgently needed. Bibliographic details Houle SK, McAlister FA, Jackevicius CA, Chuck AW, Tsuyuki RT. Does performance-based remuneration for individual health care practitioners affect patient care? A systematic review. Annals of Internal Medicine 2012; 157(12): 889-899 Indexing Status Subject indexing assigned by NLM MeSH Clinical Trials as Topic; Great Britain; Health Care Costs; Humans; Primary Health Care /economics /standards; Quality of Health Care /economics; Reimbursement, Incentive; Research Design; United States AccessionNumber 12012056910 Date bibliographic record published 18/12/2012 Date abstract record published 08/01/2013 Record Status This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn. |
|
|
|