Background:Physicians often receive feedback on their quality of diabetes care. However, it is unclear what performance measures (e.g., blood pressure targets) are used in this feedback, whether these measures correspond with diabetes care guideline recommendations, and whether feedback affects their clinical practice.

Methods: In 2016, we mailed a national survey to 1,200 U.S. primary care physicians and endocrinologists. We excluded physicians who did not provide longitudinal care to patients with type 2 diabetes (adjusted response rate, 41% (N=356)). Physicians were asked to report whether they received feedback on their quality of diabetes care, the hemoglobin A1c (A1C), cholesterol, and blood pressure (BP) goals they were encouraged to reach, and the usual A1C, cholesterol, and BP goals they recommend to their patients.

Results: In total, 56% of physicians reported receiving feedback. Among those who received feedback, the most frequent goals were an A1C<7.0% (51%), LDL<100 mg/dL (58%), and BP<130/80 mmHg (37%). In clinical practice, physicians frequently used goals that did not correspond with performance measure goals (A1C, 53%; cholesterol, 51%; systolic BP, 45%; diastolic BP, 47%). Only 22% of physicians reported that receiving feedback on their quality of diabetes care affected their treatment decisions “most of the time”/“almost always,” compared to 40% and 38% reporting “sometimes” and “never”/“rarely,” respectively.

Conclusions:Many U.S. physicians are being measured on their quality of diabetes care using performance measures that are inconsistent with diabetes care guidelines and national quality standards. Further, many physicians may not be strongly influenced by diabetes performance measurement. These discrepancies suggest that the impact of physician feedback on quality of diabetes care may be limited.


P. Zarei: None. S.A. Ham: None. A. Nathan: None. L. Sepulveda: None. R.M. Sargis: Advisory Panel; Self; CVS/Caremark. M.T. Quinn: None. E. Huang: None. N. Laiteerapong: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at