¹National Taiwan University Hospital, Department of Psychiatry, Taipei, Taiwan, ²National Taiwan University, Department of Psychiatry, College of Medicine, Taipei, Taiwan

Background: The negative impact of antipsychotics on long-term cognition warrants dose reduction, yet rendering patients at higher risk of relapse. Previously, we demonstrated that an exponential tapering method could improve patients' quality of life without increasing the risk of relapse (Liu et al., 2023).

Aims & Objectives: To investigate changes of cognitive function in patients who underwent an exponential antipsychotic dose-reduction trial.

Method: We conducted a 2-year open-label randomized prospective cohort trial, involving schizophrenia-related psychotic patients in remission under a stable antipsychotic dosage. Patients were randomized in a 2:1 ratio into guided dose reduction (GDR) group and maintenance treatment (MT) group. An additional group of naturalistic maintenance controls was also included in MT. The GDR group followed a 2-year dose-reduction protocol, with a 25% reduction in total dosage every six months. Several subtests of Wechsler Adult Intelligence Scale - Third Edition, including Information, Arithmetic, Similarities, Block Design, Digit Symbol, and Digit Span were conducted at baseline, 1 year after the enrollment, and at the end of the study.

Results: Of the 98 participants enrolled, 74 who completed both baseline and at least one follow-up cognitive evaluation were included in the statistical analysis to compare differences in performance between baseline and follow-up. Compared to the MT group (n = 34), the GDR group (n=40) showed significant improvements in the Information and Arithmetic subtests, as well as in the Working Memory Index and the overall intellectual quotient (IQ) (Table 1). Further analysis revealed a positive correlation between the extent of dosage reduction and improvement in Information (r=0.274, p=0.020), Arithmetic (r=0.272, p=0.021), Working Memory (r=0.246, p=0.037) and IQ (r=0.266, p=0.024).

	Maintenance Treatment group N=34	Guided Dose Reduction group N=40	Total N=74	P value
Relapse	4 (11.76%)	2 (5.00%)	6 (8.11%)	0.404
ΔInformation	0.32 (1.20)	1.05 (1.47)	0.72 (1.39)	0.024*
ΔArithmetic	0.00 (1.91)	1.03 (2.01)	0.55 (2.01)	0.028*
ΔSimilarity	0.65 (1.74)	1.15 (1.90)	0.92 (1.83)	0.242
ΔBlock Design	0.21 (2.06)	0.85 (1.66)	0.55 (1.87)	0.140
ΔDigit Symbol	0.76 (2.03)	0.90 (2.18)	0.84 (2.10)	0.785
ΔDigit Span	0.53 (1.38)	0.85 (2.63)	0.70 (2.14)	0.524
ΔWorking Memory	1.24 (7.42)	5.35 (8.41)	3.46 (8.18)	0.0301*
ΔTotal IQ	2.25 (7.44)	6.28 (6.12)	4.43 (7.01)	0.0125*

Table 1. Summary of outcomes. Cognitive improvements were measured as changes in standard scores between baseline and the last observation. IQ, intellectual quotient.
*Denotes parameters that reached statistical significance within the 95% confidence interval.

Discussion & Conclusions: Compared to previous dose-reduction strategies, we implemented a slower and exponential tapering method (Rodolico et al., 2022; Singh et al., 2022). While earlier dose-reduction approaches yielded conflicting results regarding cognitive function (McCutcheon et al., 2024), this study demonstrated positive outcomes. This finding suggests that the dose-reduction process may influence both relapse rates and cognitive improvement.