Abstract
RATIONALE: Parents who are the primary caregivers of people with severe mental illness are at high risk of mental health problems, including stress, depression and anxiety. While growing evidence suggests that psychoeducation may be beneficial, no systematic review has assessed its specific effects on parents. Clarifying this impact is essential to guide mental health services and improve outcomes for both caregivers and patients. OBJECTIVES: To evaluate the benefits and harms of face-to-face psychoeducational interventions for parents of people with severe mental illness compared to inactive or active (pharmacological or non-pharmacological) interventions. SEARCH METHODS: We searched CENTRAL, MEDLINE, Embase, PsycINFO, CINAHL and ProQuest databases, and two trial registries, up to 11 November 2024. We contacted experts in the field, checked references and used forward 'snowballing' to identify additional studies. There were no restrictions on language or date of publication. ELIGIBILITY CRITERIA: We included randomised controlled trials (RCTs) in parents of people with severe mental illness that compared a face-to-face psychoeducational intervention versus either an inactive intervention (i.e. no intervention, placebo, sham intervention, waiting list, or usual or standard care), a pharmacological intervention or another non-pharmacological intervention. We excluded studies that compared different psychoeducational interventions or the same intervention delivered in different modes (e.g. face-to-face versus online). OUTCOMES: Our outcomes were psychosocial well-being, quality of life, adverse events, anxiety, and satisfaction with care (i.e. caregivers' subjective appraisal of their caregiving experience), assessed as clinically important change, any change or average endpoint score. Where possible, we stratified effect estimates into short term (≤ 3 months), medium term (> 3 to ≤ 6 months) and long term (> 6 months). RISK OF BIAS: We used the Cochrane risk of bias tool RoB 2 to assess possible bias in the RCTs. SYNTHESIS METHODS: We synthesised outcome data using random-effects meta-analyses with an inverse-variance approach and the restricted-maximum-likelihood method to estimate between-study variance. For continuous outcomes, we reported the mean difference (MD) or standardised mean difference (SMD), along with the 95% confidence interval (CI) and 95% prediction interval (PI). We assessed the certainty of the evidence for each key outcome using GRADE. INCLUDED STUDIES: We included five RCTs with 304 participants. The studies were conducted in Asia (Iran, Indonesia, Japan and China) and published between 2006 and 2020. Sample sizes ranged from 40 to 84 parents. Most parents were women over the age of 45 years, and their children with severe mental illness primarily had schizophrenia. Interventions lasted 3 to 12 weeks, with 4 to 12 sessions. Four studies reported data for at least one review outcome, mostly in the short term. Outcome data were reported only as average endpoint scores. No study reported outcomes as 'any change' or as 'clinically important change', and no study reported adverse events. Therefore, no data were available for the review's critical outcomes (adverse events and short-term clinically important change in psychosocial well-being and quality of life). SYNTHESIS OF RESULTS: The certainty of the evidence ranged from very low to low, primarily due to imprecision and risk of bias. The risk-of-bias concerns related to inadequate reporting of randomisation and allocation concealment procedures, deviations from the intended interventions and absence of outcome assessor blinding. We also had concerns about selective outcome reporting for all effect estimates. We were not able to assess publication bias. Psychoeducation versus inactive intervention (no intervention, placebo, sham intervention, waiting list, usual or standard care) A meta-analysis of three trials showed that psychoeducation compared with an inactive intervention may lead to a large improvement in psychosocial well-being in the short term (SMD -1.52, 95% CI -2.32 to -0.72; I(2) = 0%; 3 studies, 150 participants; low-certainty evidence; 95% PI -2.32 to -0.72). One trial found that psychoeducation compared to standard care may result in a large improvement in psychosocial well-being in the medium term (MD -19.06; 95% CI -24.99 to -13.13; 1 study, 37 participants; low-certainty evidence). One trial found that the effect of psychoeducation compared to no treatment on quality of life in the short term is very uncertain (MD 1.28, 95% CI -6.70 to 9.26; 1 study, 40 participants; very low-certainty evidence). One trial found that psychoeducation compared to treatment as usual may result in a large improvement in state anxiety (MD -5.4, 95% CI -6.20 to -4.60; 1 study, 73 participants; low-certainty evidence) and trait anxiety in the short term (MD -3.10, 95% CI -3.83 to -2.37; 1 study, 73 participants; low-certainty evidence). One trial found that the effects of psychoeducation compared with a waiting list on satisfaction with the care of their child in the short term are very uncertain: negative symptoms (MD 4.67, 95% CI -13.06 to 22.40; 1 study, 36 participants; very low-certainty evidence); positive symptoms (MD 4.33, 95% CI -0.77 to 9.43; 1 study, 36 participants; very low-certainty evidence). No study presented data for adverse events. Psychoeducation versus pharmacological active intervention No study evaluated this comparison. Psychoeducation versus other non-pharmacological active intervention One trial of 37 participants provided very uncertain evidence about the effect of psychoeducation compared with behavioural family management on psychosocial well-being in the short term (MD -1.60, 95% CI -7.81 to 4.61) and medium term (MD -3.00, 95% CI -9.43 to 3.43) (very low-certainty evidence). No study presented data for our other outcomes. AUTHORS' CONCLUSIONS: Compared to inactive interventions, face-to-face psychoeducation for parents of individuals with severe mental illness may lead to large improvements in parental psychosocial well-being (in the short and medium term) and in parental anxiety (in the short term). However, its impact on parental quality of life and satisfaction with care is very uncertain. Evidence for the effects of psychoeducation compared to other interventions is very limited. No study assessed long-term outcomes or adverse events. Overall, the evidence is limited and of low to very low certainty, mainly due to imprecision and risk of bias. Future trials should be adequately powered, have more diverse samples, clearly report interventions and use a core outcome set with longer follow-up. FUNDING: This Cochrane review had no dedicated funding. REGISTRATION: Protocol available via DOI 10.1002/14651858.CD014532.