BACKGROUND:The longitudinal relationship between antipsychotic treatment patterns, such as polypharmacy and use of long-acting injectable (LAI) antipsychotics, and real-world functional outcomes is insufficiently explored.
METHODS:We conducted a single-center 10-year retrospective study including 114 patients with chronic schizophrenia. Monthly Global Assessment of Functioning (GAF) scores were assessed (>122 months), alongside antipsychotic patterns [polypharmacy, LAI, and chlorpromazine-equivalent (CPeq)]. A lagged linear mixed-effects model tested whether prior-month treatment predicted next-month GAF. Ordinary least squares regression assessed cumulative exposure with final GAF and 10-year change. Subgroup analyses were stratified by duration of untreated psychosis (DUP).
RESULTS:Subsequent GAF scores showed significant positive associations with 1-month lagged polypharmacy (estimate = +1.50, standard error (SE) = 0.14, p < .001) and 1-month lagged use of LAI (estimate = +1.89, SE = 0.27, p < .001), and a significant negative association with lagged CPeq dosage (estimate = 3.7 × 10-3, SE = 1.8 × 10-4, p < .001). Hospitalizations were negatively associated with functional outcome (β = -0.24, p = .038). Cumulative polypharmacy, use of LAIs, and CPeq dose were not significantly associated with either final GAF or 10-year GAF change. Using stratified analyses by DUP, all models showed no statistically significant results; however, educational level and number of hospitalizations modulated long-term functional outcomes in the DUP <12 months group.
CONCLUSION:Short-term GAF improvement was linked to polypharmacy and use of LAI, whereas higher doses predicted poorer functioning. DUP-based subgroup models showed no significant results overall.