Background

Many psychosocial and psychological interventions are used in patients with schizophrenia, but their comparative efficacy in the prevention of relapse is not known. We aimed to evaluate the efficacy, acceptability, and tolerability of psychosocial and psychological interventions for relapse prevention in schizophrenia.

Methods

To conduct this systematic review and network meta-analysis we searched for published and unpublished randomised controlled trials that investigated psychosocial or psychological interventions aimed at preventing relapse in patients with schizophrenia. We searched EMBASE, MEDLINE, PsycINFO, BIOSIS, Cochrane Library, WHO International Clinical Trials Registry Platform, and ClinicalTrials.gov up to Jan 20, 2020, and searched PubMed up to April 14, 2020. We included open and masked studies done in adults with schizophrenia or related disorders. We excluded studies in which all patients were acutely ill, had a concomitant medical or psychiatric disorder, or were prodromal or “at risk of psychosis”. Study selection and data extraction were done by two reviewers independently based on published and unpublished reports, and by contacting study authors. Data were extracted about efficacy, tolerability, and acceptability of the interventions; potential effect moderators; and study quality and characteristics. The primary outcome was relapse measured with operationalised criteria or psychiatric hospital admissions. We did random-effects network meta-analysis to calculate odds ratios (ORs) or standardised mean differences (SMDs) with 95% CIs. The study protocol was registered with PROSPERO, CRD42019147884.

Findings

We identified 27 765 studies through the database search and 330 through references of previous reviews and studies. We screened 28 000 records after duplicates were removed. 24 406 records were excluded by title and abstract screening and 3594 full-text articles were assessed for eligibility. 3350 articles were then excluded for a variety of reasons, and 244 full-text articles corresponding to 85 studies were included in the qualitative synthesis. Of these, 72 studies with 10 364 participants (3939 females and 5716 males with sex indicated) were included in the network meta-analysis. The randomised controlled trials included compared 20 psychological interventions given mainly as add-on to antipsychotics. Ethnicity data were not available. Family interventions (OR 0·35, 95% CI 0·24–0·52), relapse prevention programmes (OR 0·33, 0·14–0·79), cognitive behavioural therapy (OR 0·45, 0·27–0·75), family psychoeducation (OR 0·56, 0·39–0·82), integrated interventions (OR 0·62, 0·44–0·87), and patient psychoeducation (OR 0·63, 0·42–0·94) reduced relapse more than treatment as usual at 1 year. The confidence in the estimates ranged from moderate to very low. We found no indication of publication bias.

Interpretation

We found robust benefits in reducing the risk of relapse for family interventions, family psychoeducation, and cognitive behavioral therapy. These treatments should be the first psychosocial interventions to be considered in the long-term treatment for patients with schizophrenia.

Funding

German Ministry for Education and Research.

中文翻译：

---

精神分裂症复发预防的心理社会和心理干预：系统评价和网络荟萃分析

背景

许多心理社会和心理干预措施用于精神分裂症患者，但它们在预防复发方面的比较效果尚不清楚。我们旨在评估心理社会和心理干预对精神分裂症复发预防的有效性、可接受性和耐受性。

方法

为了进行这项系统评价和网络荟萃分析，我们搜索了已发表和未发表的随机对照试验，这些试验调查了旨在预防精神分裂症患者复发的社会心理或心理干预措施。我们检索了截至 2020 年 1 月 20 日的 EMBASE、MEDLINE、PsycINFO、BIOSIS、Cochrane 图书馆、WHO 国际临床试验注册平台和 ClinicalTrials.gov，并检索了截至 2020 年 4 月 14 日的 PubMed。我们纳入了在成人中进行的公开和隐蔽研究患有精神分裂症或相关疾病。我们排除了所有患者都患有急性病、伴有内科或精神疾病、前驱期或“有精神病风险”的研究。研究选择和数据提取由两名评审员根据已发表和未发表的报告独立完成，并联系研究作者。提取了有关干预措施的有效性、耐受性和可接受性的数据；潜在影响调节剂；和学习质量和特点。主要结果是用可操作的标准或精神病院入院测量复发。我们进行了随机效应网络荟萃分析，以计算具有 95% CI 的优势比 (OR) 或标准化平均差 (SMD)。研究方案在 PROSPERO 注册，CRD42019147884。

发现

我们通过数据库搜索确定了 27 765 项研究，通过参考以前的评论和研究确定了 330 项研究。在删除重复项后，我们筛选了 28 000 条记录。标题和摘要筛选排除了 24 406 条记录，并评估了 3594 篇全文文章的资格。然后由于各种原因排除了 3350 篇文章，定性综合纳入了 85 篇研究对应的 244 篇全文文章。其中，有 10 364 名参与者（3939 名女性和 5716 名男性）的 72 项研究被纳入网络荟萃分析。随机对照试验包括比较 20 种主要作为抗精神病药物附加的心理干预措施。种族数据不可用。家庭干预（OR 0·35，95% CI 0·24-0·52），复发预防计划（OR 0·33，0·14-0·79），认知行为疗法 (OR 0·45, 0·27–0·75), 家庭心理教育 (OR 0·56, 0·39–0·82), 综合干预 (OR 0·62, 0·44–0·87) )，并且患者心理教育 (OR 0·63, 0·42–0·94) 在 1 年时比常规治疗更能减少复发。对估计的置信度从中等到非常低不等。我们没有发现发表偏倚的迹象。

解释

我们发现，家庭干预、家庭心理教育和认知行为疗法在降低复发风险方面有很大的好处。这些治疗应该是精神分裂症患者长期治疗中首先考虑的心理社会干预措施。

资金

德国教育和研究部。