OBJECTIVE

The objective of this study was to determine whether factors associated with depression differ between elderly residents of rural and urban areas.

METHODS

The research design was cross-sectional and observational. The study subjects consisted of 926 Medicare primary care patients (650 urban and 276 rural) who were age 65+ and cognitively intact and had enrolled in a randomized, controlled Medicare demonstration. Major depression was identified by the Mini International Neuropsychiatric Interview. A logistic regression model was estimated that included a rural-urban indicator variable, additional independent variables, and interaction terms between the rural-urban indicator and independent variables that were significant at p <0.10.

RESULTS

A total of 8.3% of the rural and 14.8% of the urban patients were identified as having major depression. Reporting 0-1 close friends (odds ratio [OR]: 6.86; 95% confidence interval [CI]: 2.18-21.58), 2+ emergency room visits during the past 6 months (OR: 4.00; 95% CI: 1.19-13.43), and more financial strain (OR: 1.50; 95% CI: 1.01-2.23) were associated with significantly higher likelihood of major depression among rural as compared with urban patients. The SF-36 Physical Component Summary score had a curvilinear relationship with major depression and was higher for urban patients. The predicted probability for major depression is lower for the rural patients when financial strain is low, about the same for rural and urban patients when strain is intermediate, and higher for rural patients when strain is high.

CONCLUSIONS

Clinicians in rural areas should be vigilant for major depression among patients with very few close friends, several recent emergency department visits, and financial strain.