Comparing the Quality of Diabetes Care by Generalists and Specialists

Summaries for Patients are a service provided by Annals to help patients better understand the complicated and often mystifying language of modern medicine.

Summaries for Patients are presented for informational purposes only. These summaries are not a substitute for advice from your own medical provider. If you have questions about this material, or need medical advice about your own health or situation, please contact your physician. The summaries may be reproduced for not-for-profit educational purposes only. Any other uses must be approved by the American College of Physicians-American Society of Internal Medicine.

The summary below is from the full report titled "Profiling Care Provided by Different Groups of Physicians: Effects of Patient Case-Mix (Bias) and Physician-Level Clustering on Quality Assessment Results." It is in the 1 January 2002 issue of Annals of Internal Medicine (volume 136, pages 111-121). The authors are S Greenfield, SH Kaplan, R Kahn, J Ninomiya, and JL Griffith.

What is the problem and what is known about it so far?

Comparisons of the quality of care that different types of doctors (for example, primary care doctors and specialists) provide for chronic illnesses, such as diabetes, must be done carefully. First, different types of patients may tend to see different types of doctors. For example, older patients or patients with multiple health problems may seek care from primary care doctors, who will care for all of their problems, instead of from endocrinologists, who would care only for the diabetes. In addition, not all doctors in a specialty practice alike. For this reason, comparisons of quality of care need to consider the fact that patients who see the same doctor are likely to get similar care, a concept known as "physician-level clustering."

Why did the researchers do this particular study?

To compare the quality of diabetes care provided by generalists and by specialists, in this case endocrinologists, taking into account patient factors and physician-level clustering.

Who was studied?

The researchers studied 1750 adult patients with diabetes who received care in 15 endocrinology practices and 14 primary care practices in the United States. The patients were participating in a national study of the quality of diabetes care.

How was the study done?

Using medical records and patient surveys, the researchers collected information on the following components of good diabetes care: measurement of hemoglobin A_1c levels (a measure of sugar control), cholesterol levels, urine protein levels (to look for diabetic kidney disease), blood pressure, foot examinations, and eye examinations. They also collected information about whether blood sugar, blood pressure, and cholesterol levels were in good control and about patients' satisfaction with their health care.

What did the researchers find?

When the researchers did not consider patient factors and physician-level clustering, endocrinologists appeared to provide better care for diabetes than primary care doctors did. However, statistically speaking, the differences between the two types of doctors became insignificant when the researchers considered both patient factors and physician-level clustering. The main reason that the researchers could not detect differences between specialties was that the individual styles of physicians within each specialty varied considerably.

What were the limitations of the study?

The study included only 29 care sites. These findings might not apply to other types of doctors.

What are the implications of the study?

In careful analyses that consider both patient factors and the fact that patients of the same doctor are likely to get similar care, meaningful differences in the quality of diabetes care provided by generalists and endocrinologists will be difficult to see. Comparisons of groups of doctors that do not consider these factors may come to wrong conclusions about the care provided by different types of doctors.