中国精神科医师对精神疾病误诊率的Meta分析

Misdiagnosis rate of mental disorders in China: a Meta-analysis

目的运用Meta分析综合评价国内精神疾病的误诊情况,为提高精神疾病的诊断水平提供参考。方法计算机检索中国知网、万方数据库、维普数据库、中国生物医学文献数据库、Embase、Cochrane Library、PsycINFO、PubMed,收集公开发表的关于中国精神疾病误诊率的文献。由两位研究者独立进行文献筛选、资料提取和方法学质量评价后,采用R 3.3.3进行Meta分析。结果共10项研究(n=11 053)纳入分析。Meta分析结果显示,中国精神疾病合并误诊率为10.29%(95%CI:6.48%～15.95%,P<0.01)。亚组分析显示:①不同来源的患者误诊率差异有统计学意义(P<0.05);②不同地域的患者误诊率差异有统计学意义(P<0.05);③不同发表年份的研究误诊率差异有统计学意义(P<0.05);④采用不同诊断标准的误诊率差异无统计学意义(P>0.05);⑤不同病程的患者误诊率差异无统计学意义(P>0.05);⑥对治疗合作和不合作患者的误诊率差异有统计学意义(P<0.05)。Egger's线性回归未见明显的发表偏倚(P>0.05)。结论中国精神科医师对精神疾病合并误诊率较高,患者来源、所在地域、研究发表年代及对治疗的态度可能是影响精神疾病误诊的因素。

Objective To evaluate the incidence of misdiagnosis among mental disorders through Meta-analysis, so as to provide references for improving the diagnostic level of mental diseases. Methods Published literatures on misdiagnosis rate of mental diseases in China were retrieved from CNKI, Wanfang, Vip, Chinese biomedical literature database, Embase, Cochrane Library, PsycINFO and PubMed. Literature screening, data extraction and methodological quality evaluation were performed by the two researchers independently. Meta-analysis was performed using R 3.3.3. Results A total of 10 studies were enrolled ( n =11 053). Meta-analysis results showed that the misdiagnosis rate of mental disorders in China was 10.29%(95% CI: 6.48%~15.95%, P <0.01 ). Subgroup analysis showed that,①there was statistically significant difference in misdiagnosis rates among patients from different source ( P <0.05);②the difference in misdiagnosis rates among patients from different regions was statistically significant ( P <0.05 );③the difference in misdiagnosis rates among the studies with different publication years was statistically significant ( P <0.05);④ there was no significant difference in misdiagnosis rate between different diagnostic criteria ( P >0.05);⑤the rate of misdiagnosis showed no difference among patients with different course of disease ( P >0.05);⑥there was a statistically significant difference in the misdiagnosis rate between cooperative and non-cooperative patients ( P <0.05). No obvious publication bias was observed in Egger's linear regression. Conclusion Incidence of misdiagnosis among mental disorders in China is high. The source of patients, location , date of publication and attitude to treatment may affect the misdiagnosis of mental disorders.