Diabetologie und Stoffwechsel 2006; 1 - A171
DOI: 10.1055/s-2006-943896

Associations between physician gender and treatment in patients with diabetes mellitus type 2: Results from the DUTY register (Diabetes mellitus needs unrestricted evaluation of patient data to yield treatment progress)

HK Berthold 1, I Berthold 2, K Bestehorn 3, M Böhm 4, W Krone 2
  • 1Arzneimittelkommission der deutschen Ärzteschaft, Berlin, Germany
  • 2Klinik II und Poliklinik für Innere Medizin, Universität zu Köln, Köln, Germany
  • 3MSD Sharp und Dohme, Haar, Germany
  • 4Klinik für Innere Medizin III, Universitätsklinikum des Saarlandes, Homburg/Saar, Germany

Purpose of the study: When analyzing gender aspects in medicine, the possible role of the gender of the treating physician has not received the same attention as the gender of the patient.

Methods: In order to examine possible associations between gender of physicians, gender of patients and treatment parameters we evaluated datasets of >53,000 patients with diabetes mellitus type 2 from the DUTY register.

Results: 33% of the physicians were female (F). The proportion of F patients was higher for F than for male (M) physicians (55.5% vs. 49.4%, P<0.0001). Since on average F patients were older than M patients (66.9 vs. 63.4 years, P<0.0001), F doctors had older patients than M doctors (65.6 vs. 65.1 years, P<0.0001). While disease duration differed between M and F patients (6.4 vs. 7.0 years, P<0.0001), physician gender was not associated with disease duration. Mean HbA1c was not different between M and F patients (7.294 vs. 7.298%, P=0.72), but was lower in patients treated by F physicians (7.23 vs. 7.32, P<0.0001). When treated by F physicians, more patients achieved HbA1c values <7% than when treated by M physicians (48.2 vs. 43.3%, P<0.0001). Patients treated by F physicians had lower LDL cholesterol values (135.5 vs. 137.4mg/dl, P<0.0001). Statin prescription rates were not different between M and F physicians, although low in general (<25% of the patients received statins). The incidence of coronary heart disease was higher in M than in F patients (30.8 vs. 24.8%, P<0.0001), but F physicians had more patients with CHD than M physicians (28.9 vs. 27.1%, P<0.0001) although the proportion of women in their clientele was higher. After controlling for age and sex, the incidence of CHD was still higher in patients treated by F physicians (OR 1.08, 95% CI 1.03 to 1.13, P=0.002).

Conclusions: We conclude that physician gender may play a role in the treatment of patients with diabetes mellitus type 2. It is unclear, however, if the association of surrogate endpoints with patient/physician gender is a result of patient preferences or of differential treatment.