Importance Studies are lacking summarizing how the association between mental disorders and mortality varies by socioeconomic position (SEP), particularly considering different aspects of SEP, specific types of mental disorders, and causes of death. Objective To investigate the role of SEP in the association between mental disorders and mortality and the association between SEP and mortality among people with mental disorders. Data Sources MEDLINE, Embase, PsycINFO, and Web of Science were searched from January 1, 1980, through April 3, 2023, and a snowball search of reference and citation lists was conducted. Study Selection Inclusion criteria were observational studies estimating the associations between different types of mental disorders and mortality, stratified by SEP and between SEP and mortality in people with mental disorders. Data Extraction and Synthesis Pairs of reviewers independently extracted data using a predefined data extraction form and assessed the risk of bias using the adapted Newcastle-Ottawa scale. Graphical analyses of the dose-response associations and random-effects meta-analyses were performed. Heterogeneity was explored through meta-regressions and sensitivity analyses. Main Outcomes and Measures All-cause and cause-specific mortality. Results Of 28 274 articles screened, 71 including more than 4 million people with mental disorders met the inclusion criteria (most of which were conducted in high-income countries). The relative associations between mental disorders and mortality were similar across SEP levels. Among people with mental disorders, belonging to the highest rather than the lowest SEP group was associated with lower all-cause mortality (pooled relative risk [RR], 0.79; 95% CI, 0.73-0.86) and mortality from natural causes (RR, 0.73; 95% CI, 0.62-0.85) and higher mortality from external causes (RR, 1.18; 95% CI, 0.99-1.41). Heterogeneity was high (I2 = 83% to 99%). Results from subgroup, sensitivity, and meta-regression analyses were consistent with those from the main analyses. Evidence on absolute scales, specific diagnoses, and specific causes of death was scarce. Conclusion and Relevance This study did not find a sufficient body of evidence that SEP moderated the relative association between mental disorders and mortality, but the underlying mortality rates may differ by SEP group, despite having scarcely been reported. This information gap, together with our findings related to SEP and a possible differential risk between natural and external causes of death in individuals with specific types of mental disorders, warrants further research.

中文翻译：

---

社会经济地位在精神障碍与死亡率之间关系中的作用：系统回顾和荟萃分析。

缺乏重要性研究总结精神障碍与死亡率之间的关联如何因社会经济地位（SEP）而变化，特别是考虑到 SEP 的不同方面、精神障碍的特定类型和死亡原因。目的 探讨SEP在精神障碍与死亡率的关系中的作用以及SEP与精神障碍患者死亡率的关系。数据源从 1980 年 1 月 1 日到 2023 年 4 月 3 日检索了 MEDLINE、Embase、PsycINFO 和 Web of Science，并对参考文献和引文列表进行了滚雪球式检索。研究选择纳入标准是观察性研究，估计不同类型的精神障碍与死亡率之间的关联，按 SEP 分层以及 SEP 与精神障碍患者的死亡率之间的关系。数据提取和综合 成对的评审员使用预定义的数据提取表格独立提取数据，并使用改编后的纽卡斯尔-渥太华量表评估偏倚风险。对剂量反应关联进行图形分析和随机效应荟萃分析。通过元回归和敏感性分析探讨异质性。主要成果和措施 全因死亡率和特定原因死亡率。结果 在筛选的 28 274 篇文章中，71 篇（其中包括超过 400 万精神障碍患者）符合纳入标准（其中大部分是在高收入国家进行的）。不同 SEP 水平的精神障碍和死亡率之间的相对关联相似。在精神障碍患者中，属于最高而非最低 SEP 组的人与较低的全因死亡率（汇总相对风险 [RR]，0.79；95% CI，0.73-0.86）和自然原因死亡率（RR， 0.73；95% CI，0.62-0.85），外部原因导致的死亡率较高（RR，1.18；95% CI，0.99-1.41）。异质性较高（I2 = 83% 至 99%）。亚组、敏感性和荟萃回归分析的结果与主要分析的结果一致。关于绝对规模、具体诊断和具体死亡原因的证据很少。结论和相关性 本研究没有找到足够的证据表明 SEP 调节精神障碍和死亡率之间的相对关联，但不同 SEP 组的潜在死亡率可能有所不同，尽管几乎没有报道。这种信息差距，加上我们与 SEP 相关的发现以及患有特定类型精神障碍的个体自然死亡和外部原因死亡之间可能存在的差异风险，值得进一步研究。