Comparative Effectiveness of Second Generation Antidepressants on Cognition and Dementia in the Elderly

Objectives: The two primary objectives of this study were: 1) to evaluate the comparative effectiveness of second generation antidepressants classes on cognition in elderly nursing home residents with depression, and 2) to evaluate the comparative effectiveness of second generation antidepressants classes on dementia in elderly nursing home residents with depression.

Methods: This study involved retrospective cohort study design conducted using data from Medicare Part D claims and Minimum Data Set (MDS) from 2007-2010. The study population included elderly nursing home residents with depression who initiated treatment with second generation antidepressants classes namely Selective Serotonin Reuptake Inhibitors (SSRIs), Serotonin–Norepinephrine Reuptake Inhibitors (SNRIs) or Tetracyclics. These patients were followed for one year to examine cognition; and two years to evaluate the risk of dementia. Cognition was identified using the MDS Cognition scale. Time to diagnosis of dementia was ascertained using the chronic condition flag for dementia in Medicare Beneficiary Summary File. The differences in covariate distributions between the antidepressant users and non-users were evaluated using chi-squared (χ2) tests for categorical variables and t-tests for continuous variables. A multiple propensity score-adjusted Repeated Measures Mixed Model was used to evaluate the comparative effectiveness of SSRIs, SNRIs and Tetracyclics with respect to cognition. A multiple propensity score-adjusted Cox Proportional Hazards Model was used to examine comparative effectiveness of SSRIs, SNRIs and Tetracyclics on dementia.

Results: For the first objective, the study cohort consisted of 1,518 elderly nursing home residents. Of these, 1,081 received SSRIs (71.21%), 320 received Tetracyclics (21.08%) and 117 received SNRIs (7.71%). After adjusting for multiple propensity scores, the repeated measure mixed model did not find any statistically significant difference in cognition with the use of SSRIs (β = -0.23; 95% Confidence Interval (95% CI), -0.67, 0.22) or Tetracyclics (β = -0.45; 95% CI, -0.96, 0.05) when compared to SNRIs (reference group). Results of multiple sensitivity analyses were consistent with the main findings. For the second objective, the study cohort constituted 13,354 elderly nursing home residents with depression. Of these, 19,952 received SSRIs (79.77%), 2,381 received SNRIs (9.48%) and the rest 2,775 received Tetracyclics (11.05%). The unadjusted incidence of dementia was 8.20% for SSRIs users, 6.01% for the SNRIs users and 7.21% for Tetracyclics users. The propensity score adjusted Cox proportional hazard model did not find any significant difference in the comparative effectiveness of SNRIs [Hazards Ratio, HR, 0.99; 95% CI, 0.84, 1.19] or Tetracyclics [HR, 1.01; 95% CI, 0.87, 1.17] when compared to the SSRIs for the risk of dementia in elderly nursing home residents with depression. Results from the two sensitivity analyses supported the main findings.

Conclusion: This multiple propensity score-adjusted retrospective cohort study did not find any statistically significant difference in the comparative effectiveness of three commonly used second generation antidepressant classes on cognition and risk of dementia. Future studies are required to examine the long-term effectiveness of these antidepressants classes on cognition and dementia.