Abstract
Background/Aims Understanding variation in resource use among primary care physicians (PCPs) and its impact on clinical quality and patient outcomes is critical for designing policies to encourage efficiency in delivery without sacrificing quality. This study examined the relationship between PCPs’ use of resources, and clinical quality and patient-assessed quality indicators.
Methods We used a dataset linking electronic health records, administrative claims, and patient satisfaction surveys from a large ambulatory group practice with mixed sources of payment. We studied PCPs practicing family or general internal medicine in 2010 (n = 208). For the measure of resource use, we report fee-weighted differences for ambulatory services relative to the overall average (“costs”). For the measures of clinical quality, we combined various indicators pertinent to primary care practice and created a composite average score (number of patients who met the target/number of eligible patients). For patient-rated care quality, we examined the percent of patients who indicated in the survey that they were “very likely” (5 on 1–5 scale) to recommend their provider to others. We tested whether costs differ between PCPs who have superior quality and satisfaction scores and those who have less exceptional scores. Based on tertiles of satisfaction and quality scores, PCPs were classified into three subsets: (1) high-satisfaction/high-quality (HS-HQ), (2) high-satisfaction/ medium-quality (HS-MQ) or medium-satisfaction/high-quality (MS-HQ), and (3) the remainder (non-superior). PCPs with low-satisfaction and low-quality scores were not studied further.
Results Costs were highest in the HS-HQ group (9.0% [± 3.6%] above the non-superior group, and 8.6% [± 4.4%] above the HS-MQ or MS-HQ group). The difference in costs was statistically significant between PCPs with HS-HQ scores and those with non-superior scores (P = 0.01). Differences were not significant between PCPs with HS-MQ or MS-HQ scores and those with non-superior scores.
Conclusions These findings are preliminary, as we plan to apply a more detailed risk-adjustment approach. However, the findings suggest that PCPs with both high clinical quality and patient satisfaction scores used approximately 9% more services for their patients (age and sex adjusted) than PCPs with less exceptional scores.
