Depression Treatment in Primary Care

W. David Robinson, PhD; Jenenne A. Geske, PhD; Layne A. Prest, PhD; Rachel Barnacle, MS


J Am Board Fam Med. 2005;18(2):79-86. 

In This Article


All family medicine (n = 96) and general internal medicine physicians (n = 68) affiliated with a university medical center were initially identified as potential participants. A search of the centralized patient record system identified patients of these physicians who had been diagnosed with depression (based on ICD-9 codes) and who were seen between April 2000 and April 2002. During this study period, 78 of the 164 resident and faculty physicians (53 family medicine and 25 internal medicine) diagnosed patients with a new case of depression. All resident physicians were excluded from the study because of the possible influence of their attending physicians. The final sample of physicians consisted of 44 family medicine (FM) physicians and 23 internal medicine (IM) physicians (27 women and 40 men). The Institutional Review Board of the University of Nebraska Medical Center approved the study in June 2002.

To avoid any chart abstraction bias, we used all the charts of patients diagnosed with depression. A total of 2401 patients were identified. Because we focused only on newly diagnosed patients with uncomplicated depression, we excluded patients who had been treated for depression within the previous 6 months and those with comorbid anxiety and other related disorders.

The research assistant reviewed the charts of the remaining patients (n = 580) and recorded demographic information on each patient (age, sex, race, and marital status). Patient identifying information was not recorded. The research assistant also used the dictated patient record to identify the patients' chief complaint, the presence or absence of symptoms related to depression (sleep, interest, guilt, energy, concentration, appetite, mood, psychomotor changes, suicide), any medications prescribed for mood disorders, the patients' insurance source, the ICD-9 diagnosis code used for the visit, and the recommended treatment. These treatments were grouped into 1 of 7 separate categories (pharmacotherapy, psychotherapy, pharmacotherapy + psychotherapy, counselor-recommended pharmacotherapy, watchful waiting, support group, or other). Because less than 5% of the patients were grouped into the "watchful waiting" and "support group" categories, these groups were combined with the "other" category for analysis.

The research questions for this study concerned the physicians' treatment recommendations and how these recommendations might differ between FM and IM physicians and between female and male physicians. Thus, the primary unit of analysis was the physician rather than the patient. The outcome variable was the percentage of each physician's patients who were recommended each type of treatment.

Mixed method analyses of variance were used to determine whether the percentage of patients who were recommended for each of the treatment types interacted with medical specialty or gender of the physician.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: