Abstract：
The Brief Jail Mental Health Screen (BJMHS) is one of the most well-known and frequently used tools to conduct routine mental health screening at jail intake. In prior research, the BJMHS results typically have been evaluated overall (i.e., yes/no positive screen). However, there is heterogeneity in symptom presentation and treatment histories among people with serious mental illness, and there are potential consequences of this heterogeneity for mental health administration and policy in jails. We conducted a latent class analysis of BJMHS item-level results using administrative data for 37,998 people booked into a southeastern, metropolitan, U.S. county jail over a 3.5-year period. A 4-class solution provided the best fitting and most interpretable model. The largest class (89.5%) comprised people unlikely to report symptoms or treatment histories (limited symptoms). The next class comprised people who were unlikely to report ongoing symptoms but reported medication and hospitalization (managed symptoms). The third class (2.5%) included people likely to report feeling useless/sinful, prior hospitalization, and current psychiatric medication (depressive symptoms). The fourth class (1.0%) comprised people likely to report thought control, paranoia, feeling useless/sinful, medication, and hospitalization (psychotic symptoms). Controlling for sociodemographic and booking characteristics, people in the managed, depressive, and psychotic symptoms classes had significantly longer jail stays compared to those in the limited symptoms class. People in the managed and depressive symptoms classes were at heightened risk of re-arrest compared to the limited symptoms class. Findings can inform case prioritization and the allocation of resources to support efficient and effective jail-based mental health services.
摘要：
简短的监狱心理健康筛查（BJMHS）是最著名和最常用的工具之一，可以在监狱中进行常规的心理健康筛查。在先前的研究中，BJMHS结果通常已经过整体评估(即，是/没有正面屏幕)。然而,严重精神疾病患者的症状表现和治疗史存在异质性，这种异质性对监狱中的心理健康管理和政策有潜在的影响。我们对BJMHS项目级结果进行了潜在的类分析，使用了37,998人预订到东南部，Metropolitan,美国县监狱超过3.5年。4类解决方案提供了最佳拟合和最可解释的模型。最大的类别（89.5％）包括不太可能报告症状或治疗史（有限症状）的人。下一类包括不太可能报告持续症状但报告药物和住院（管理症状）的人。第三类（2.5％）包括可能报告无用/有罪的人，之前住院，和目前的精神病药物（抑郁症状）。第四类（1.0％）包括可能报告思想控制的人，偏执狂,感觉无用/有罪，药物,住院（精神病症状）。控制社会人口统计学和预订特征，被管理的人，抑郁,与有限症状类别相比，精神病症状类别的监狱停留时间明显更长。与有限的症状类别相比，管理和抑郁症状类别的人再次发作的风险更高。研究结果可以为案件的优先次序和资源分配提供信息，以支持有效和有效的基于监狱的精神卫生服务。
