Most primary care practices are not large enough for significant differences in performance to be assessed using national quality and cost benchmarks, according to a report in JAMA.
Nationally, fewer than 2% of all primary care practices were able to be reliably assessed because their caseloads were too small. Even when their case loads were pooled with those of other physicians in the practice, and even if 2-3 years' worth of cases were included, the numbers were too small to reliably assess quality, according to David J. Nyweide, Ph.D., of the Centers for Medicare and Medicaid Services and his associates.
The CMS has “been overseeing a series of value-based purchasing initiatives,” including pay-for-performance projects and the Physician Quality Reporting Initiative.
Dr. Nyweide and his colleagues questioned whether individual physicians see a sufficient number of patients with various disorders such that their performance can be judged against commonly used quality and cost measures.
Using national mean ambulatory Medicare spending data, the researchers calculated the caseloads that would be necessary to detect meaningful differences on each commonly used performance measure, including rates at which 66- to 69-year-old women received mammography, rates of hemoglobin A1c testing for diabetics aged 65-75 years, rates of preventable hospitalizations associated with 13 specific adult conditions, and rates of hospital readmission for heart failure patients.
In all, 71,980 primary care physicians who were affiliated with 30,794 practices were included in the study. Most of the practices (61%) were solo. Caseloads ranged from a median of 170 patients for solo practitioners to 13,400 for practices with more than 50 primary care physicians.
The investigators found that “only the largest primary care physician practices, which are also the most uncommon, can be expected to have sufficient caseloads to measure significant differences in performance.”
A year-long caseload of 328 women aged 66-69 years old would be needed to detect a 10% difference in the rate of mammography for that age group, and 19,069 patients would be needed to reliably detect a 10% difference in the rate of preventable hospitalizations.
Overall, fewer than 2% of the practices could be reliably compared on any of the performance measures.
Even grouping caseloads by 2-year and 3-year periods failed to amass sufficient sample sizes for reliable comparisons among practices.
“The results from this study call into question the wisdom of pay-for-performance programs and quality reporting initiatives that focus on differentiating the value of care delivered to the Medicare population by primary care physicians,” Dr. Nyweide and his colleagues wrote (JAMA 2009;302:2444-50).
Disclosures: None was reported.