The following is a summary of the “Predicting psychotic relapse following randomised discontinuation of paliperidone in individuals with schizophrenia or schizoaffective disorder: an individual participant data analysis,” published in the March 2023 issue of Psychiatry by Brandt, et al.

In particular, it is not well-established how to predict relapse in people with psychotic disorders after they have stopped taking antipsychotic medication. Their goal in using machine learning to predict relapse was to identify factors predictive of relapse for all participants (regardless of whether or not treatment was continued) and factors predictive of relapse for those who stopped receiving treatment. For this patient-level meta-analysis, they looked for placebo-controlled, randomized antipsychotic discontinuation trials involving people with schizophrenia or schizoaffective disorder (aged 18 years) in the Yale University Open Data Access Project database. In addition, the research considered if participants were already taking antipsychotic medication and were randomly assigned to either continue or stop taking it altogether and receive a placebo. 

They used univariate and multivariate proportional hazard regression models (including multivariate treatment groups by variable interactions) with machine learning to assess 36 prespecified baseline variables at randomization to predict the time to relapse. The variables were classified as either general prognostic relapse factors or specific predictors of relapse. In total, 414 trials were found, 5 of which included 700 participants (304 women [43%] and 396 men [57%] for the continuation group and 692 people [292 women [42%] and 400 men [58%] for the withdrawal group) (median age 37 [IQR 28–47] years for continuation group and 38 [28–47] years for discontinuation group). Drug-positive urine, paranoid, disorganized, and undifferentiated types of schizophrenia (lower risk for schizoaffective disorder), psychiatric and neurological adverse events, higher severity of akathisia (i.e., difficulty or inability to sit still), antipsychotic discontinuation, lower social performance, younger age, lower glomerular filtration rate, and benzo use were all general prognostic factors of increased risk of relapse for all participants (lower risk for anti-epileptic comedication). 

Increased prolactin concentration, higher hospitalizations, and smoking were the only three of the 36 baseline variables that were predictors of increased risk, specifically after antipsychotic discontinuation. Oral antipsychotic treatment (lower risk for long-acting injectables), higher last dosage of the antipsychotic study drug, shorter duration of antipsychotic treatment, and higher score on the Clinical Global Impression (CGI) severity scale were prognostic factors and predictors with increased risk after discontinuation. For people who weren’t used to training the model, the predictive performance (concordance index) was 0.707 (chance level is 0.5). General prognostic factors of psychotic relapse and predictors specific for treatment discontinuation that is routinely available could be used to facilitate individualized treatment. Individuals at higher risk of relapse, such as those who have been hospitalized frequently, have higher scores on the CGI severity scale and have higher prolactin concentrations, should avoid abruptly discontinuing higher doses of oral antipsychotics.

Source: sciencedirect.com/science/article/abs/pii/S2215036623000081