Objectives: To compare objective and subjective weight gain of second-generation antipsychotics in schizophrenia and bipolar disorder.
Methods: English-language literature published and cited in PubMed (MEDLINE) from January 1966 to December 2015 was searched with the keywords antipsychotic, atypical antipsychotic, or generic/brand name of atypical antipsychotics, and schizophrenia, bipolar disorder, bipolar mania, or bipolar depression, and safety, tolerability, weight gain, and randomized, placebo-controlled clinical trials. The absolute risk increases and the numbers needed to treat to harm (NNH) for ≥7% weight gain (objective) and self-report weight gain (subjective) were estimated.
Results: In schizophrenia, the NNH for 7% or greater weight gain ranged from 5 to 62, and the NNH for self-reported weight gain was from 11 to -224. The ratio of self-reported NNH to 7% or greater NNH was from 1.5 to 8.0. In bipolar mania, the NNH for 7% or greater weight gain ranged from 7 to -101 and the NNH for self-reported weight gain was from 13 to 84. The ratio of self-reported NNH to 7% or greater NNH was from 0.9 to 2.5. In bipolar depression, the NNH for 7% or greater weight gain ranged from 5 to 69, and the NNH for self-reported weight gain was from 8 to 17. The ratio of self-reported NNH to 7% or greater NNH was 1.2 to 1.6.
Conclusions: Subjective reporting underestimated the risk of antipsychotic-related weight gain compared with objectively measured weight change. Self-awareness of antipsychotic-related weight gain was lower in patients with schizophrenia than in patients with bipolar disorder. Measuring weight change during antipsychotic treatment should be a routine practice.