Risperidone and paliperidone are less effective and more harmful than shown in published studies

Reference

Hodkinson A, Heneghan C, Mahtani KR, Kontopantelis E, Panagioti M. Benefits and harms of risperidone and paliperidone for treatment of patients with schizophrenia or bipolar disorder: a meta-analysis involving individual participant data and clinical study reports. BMC Medicine 2021;19(1):195.

Synopsis

Risperidone and paliperidone are heavily promoted second-generation antipsychotics, often for off-label uses, that have significant harms. However, a network meta-analyses rated risperidone as being only the sixth best treatment option. Have we really seen the totality of the evidence? These researchers identified all published randomized trials of the 2 drugs for the treatment of schizophrenia and bipolar disorder, as well as unpublished clinical trial reports from clinical trial registries. They were able to identify 57 randomized trials, of which 34 had full individual patient data for 12,316 patients. Most of the studies were set in the United States and included patients with schizophrenia. In an individual patient data meta-analysis, individual level data for each study are obtained and definitions and outcomes are standardized as much as possible. The quality of the studies was generally good. The authors found that many outcomes were not reported in the journal publications, including 24% of primary outcomes, 39% of secondary outcomes, and 39% of serious adverse events. The primary outcome was a reduction in the Positive and Negative Schizophrenia Symptom (PANSS) score and was only 5.8 points (risperidone) and 6.0 points (paliperidone) lower than placebo on the 210-point scale. This is lower than the reported improvement in PANSS scores from a Cochrane Review based only on published studies, which reported a 17.8-point reduction for risperidone and a 9-point reduction for paliperidone. The clinical study reports contained information on approximately twice as many adverse events (4434 vs 2296) and 8 times as many serious adverse events (650 vs 82) compared with the journal publications of the same studies. NOTE: For any lack of reported outcomes, I used the converse of the statistics in Table 1 (ie, 100 - 76 = 24).