Yasushi Tsujimoto, Yuri Nakamura, Masahiro Banno, Kunihiro Kohmura, Hiraku Tsujimoto, Yuki Kataoka

Humour-based interventions are defined as any intervention that promotes health and wellness by stimulating a playful discovery, expression, or appreciation of the absurdity or incongruity of life's situations. Humour-based interventions can be implemented in different settings, including hospitals, nursing homes and day care centres. They have been posed as an adjunct to usual care for people with schizophrenia, but a summary of the evidence is lacking. To examine the effects of humour-based interventions as an add-on intervention to standard care for people with schizophrenia. On 31 July 2019 and 10 February 2021 we searched the Cochrane Schizophrenia Group's study-based register of trials, which is based on CENTRAL, CINAHL, ClinicalTrials.Gov, Embase, ISRCTN, MEDLINE, PsycINFO, PubMed, and WHO ICTRP. We included all randomised controlled trials comparing humour-based interventions with active controls, other psychological interventions, or standard care for people with schizophrenia. We excluded studies fulfilling our prespecified selection criteria but without useable data from further quantitative synthesis. Two review authors independently inspected citations, selected studies, extracted data and appraised study quality, following the guidance from the Cochrane Handbook for Systematic Reviews of Interventions. For binary outcomes we calculated risk ratios (RRs) and their 95% confidence intervals (CIs). For continuous outcomes we calculated the mean differences (MDs) and their 95% CIs. We assessed risks of bias for included studies and created summary of findings tables using the GRADE approach. We included three studies in this review for qualitative synthesis, although one study did not report any relevant outcomes. We therefore include two studies (n = 96) in our quantitative synthesis. No data were available on the following prespecified primary outcomes: clinically-important change in general mental state, clinically-important change in negative symptoms, clinically-important change in overall quality of life, and adverse effects. As compared with active control, humour-based interventions may not improve the average endpoint score of a general mental state scale (Positive and Negative Syndrome Scale (PANSS) total score: MD -1.70, 95% CI -17.01 to 13.61; 1 study, 30 participants; very low certainty of evidence); positive symptoms (PANSS positive symptom score: MD 0.00, 95% CI -2.58 to 2.58; 1 study, 30 participants; low certainty of evidence), negative symptoms (PANSS negative symptom score: MD -0.70, 95% CI -4.22 to 2.82; 1 study, 30 participants; very low certainty of evidence) and anxiety (State-Trait Anxiety Inventory (STAI): MD -2.60, 95% CI -5.76 to 0.56; 1 study, 30 participants; low certainty of evidence). Due to the small sample size, we remain uncertain about the effect of humour-based interventions on leaving the study early as compared with active control (no event, 1 study, 30 participants; very low certainty of evidence). On the other hand, humour-based interventions may reduce depressive symptoms (Beck Depression Inventory (BDI): MD -6.20, 95% CI -12.08 to -0.32; 1 study, 30 participants; low certainty of evidence). Compared with standard care, humour-based interventions may not improve  depressive symptoms (BDI second edition: MD 0.80, 95% CI -2.64 to 4.24; 1 study, 59 participants; low certainty of evidence). We are uncertain about the effect of humour-based interventions on leaving the study early for any reason compared with standard care (risk ratio 0.38, 95% CI 0.08 to 1.80; 1 study, 66 participants; very low certainty of evidence). We are currently uncertain whether the evidence supports the use of humour-based interventions in people with schizophrenia. Future research with rigorous and transparent methodology investigating clinically important outcomes is warranted.