Longitudinal changes in dose-normalized psychotropic medication exposure around transfer to a specialized older adult mental health service: a retrospective longitudinal study in Peru

Objectives: This research aimed to characterize longitudinal changes in dose-normalized psychotropic exposure around transfer to a specialized older adult mental health service in Peru, with the primary contrast comparing the last eligible pre-transfer consultation (T0) with six months after transfer. We additionally explored characteristics associated with average medication-score levels and potential modification of the T0-to-6-month change. Methods: A retrospective longitudinal single-cohort study was conducted using routine clinical records from the Instituto Nacional de Salud Mental Honorio Delgado-Hideyo Noguchi in Lima, Peru. The primary analytic cohort comprised patients transferred to the older adult service who had valid longitudinal medication data at or before transfer and after transfer. A dose-normalized psychotropic medication score was calculated at approximately -6 months, -3 months, T0 (the last eligible consultation before transfer), +3 months, and +6 months. Longitudinal change was estimated using linear mixed-effects models with categorical time and participant-specific random intercepts; the prespecified primary contrast compared +6 months with T0. Sensitivity analyses evaluated alternative score definitions, age restrictions, influential observations, and participant-level cluster bootstrap resampling. Results: The primary analytic cohort included 213 participants; mean age at T0 was 65.5 years (SD 7.2), 59.2% were female, and the most frequent diagnostic categories were psychotic disorders (52.1%) and depressive disorders (29.1%). The medication score was 17.87 points higher at +6 months than at T0 (95% CI 12.18 to 23.57; p < 0.001), and the exploratory adjusted estimate was similar (17.82 points; 95% CI 12.10 to 23.54; p < 0.001). The primary contrast remained positive across alternative score definitions, age-restricted analyses, influence analyses, and participant-level bootstrap resampling. In exploratory effect-modification analyses, each 10-year increase in age was associated with a 10.94-point greater T0-to-+6-month change (95% CI 2.33 to 19.55; p = 0.013), although this association did not remain statistically significant after false-discovery-rate correction (FDR-adjusted p = 0.091); no other tested modifier showed evidence of interaction after FDR correction. Conclusions: Dose-normalized prescribed psychotropic exposure increased from a relative pre-transfer low point at T0 to 6 months after transfer, with similar estimates across sensitivity analyses. In the absence of a contemporaneous comparison group, this finding should be interpreted as a temporal association around service transfer rather than evidence that transfer itself caused the increase or that the resulting prescribing was clinically beneficial, harmful, or inappropriate.