On any given day, 5–15% of men and 20–30% of women jail detainees have a diagnosable serious mental illness, resulting in over 2 million persons with mental disorder admitted to U.S. jails each year. Corrections professionals have a constitutional obligation to provide adequate mental health treatment; thus, they need to immediately triage incoming detainees to determine who is at risk of suicide, who has mental health problems, and who is experiencing typical reactions to incarceration. To assist jails in meeting their constitutional obligation, researchers developed a valid, easy-to-administer screen for mental health disorders in jails: the Brief Jail Mental Health Screen (BJMHS). This article discusses the development, validation, and use of the BJMHS.
Development
The BJMHS was developed with funding from the National Institute of Justice to provide jail staff with a short screen to identify detainees with mental health problems who they could refer for additional assessment and appropriate treatment. The goal was to develop a mental health screen that was simple, free, valid, and decisive and could become part of the routine booking procedures in jails.
The BJMHS is derived from the Referral Decision Scale, a clinical screen with high levels of correctly identifying people with serious mental illness, such as schizophrenia and bipolar disorder, and those who do not have serious mental illness. Questions arose, however, that the Referral Decision Scale was not appropriate for corrections populations because it asks about lifetime experiences rather than current symptoms. The BJMHS was specifically designed for use in corrections settings, in particular to identify individuals currently experiencing mental health symptoms and those with a history of psychiatric hospitalization and medication.
The BJMHS includes 8 yes/no items including 6 questions about current symptoms and 2 about medication and hospitalization history. The BJMHS is decisive, leaving no room for ambiguity as to what the staff person administering the screen does with the results. If an individual answers yes to 2 of the first 6 items, he or she is referred for further assessment. If the individual endorses a history of either hospitalization or medication for emotional or psychological problems, he or she is referred for assessment. The screen can be administered in 3 min by corrections staff with minimal training, and instructions for the screen and for the results are included on the instrument.
Validation of BJMHS
The BJMHS was validated in 2005 and again in 2007. When validating this screen, it was critical to know if the screen correctly identified people who needed further mental health assessment and those who did not. It is imperative that corrections staff identify people who need mental health treatment and those who do not to provide a safe environment and to efficiently allocate jail resources.
In the first validation study, the BJMHS was administered to over 10,000 jail detainees in four U.S. jails and identified 11% who needed further assessment. From this group, a stratified random sample of 125 individuals was selected for assessment. From the 89% who did not need further assessment based on their BJMHS outcome, a stratified random sample of 232 detainees was selected. All 357 individuals were administered the Structured Clinical Interview for DSM-IV (SCID), a diagnostic clinical tool widely used with corrections populations. Based on the SCID results, the BJMHS correctly identified 74% of men and 62% of women. This means that the BJMHS screen outcome agreed with the longer, clinical assessment in 74 of 100 men and 62 of 100 women.
One group of detainees in the validation study was a concern to researchers: those whose BJMHS said no mental illness but whose SCID said mental illness. These are referred to as false negatives. In effect, these detainees would not be identified at booking as needing a referral to mental health but did, in fact, need more services. In the validation study, the BJMHS incorrectly indicated that 15% men and 35% of women had no mental health problem but did when administered the SCID. The opposite outcome—where the screen said yes and the assessment said no—results in misallocated resources through administering unnecessary assessments, but they were not as concerning as missing detainees who need mental health treatment.
Because of the high proportion of women (35%) whose mental disorder was missed by the BJMHS, a second validation study was conducted in 2007. It was posited that the high number of women with mental disorder not identified using the 8-item BJMHS had symptoms and diagnoses not captured by the instrument. Specifically, it was asserted that the 8 items were not sensitive to anxiety and depression resulting from post- traumatic stress disorder, a common diagnosis among female detainees. Four questions were added to the BJMHS to determine if a 12-item scale would more correctly match women’s BJMHS outcome with their SCID. Using an identical research protocol and three of the same four jails plus one new one, the researchers sought to revalidate the BJMHS-Revised.
The BJMHS-Revised was administered to over 10,000 jail detainees, and 464 individuals were sampled to receive the SCID using the same selection procedures as in the first study. Results from the revalidation study showed that the addition of the 4 items did not improve the accuracy of matching the BJMHS and SCID outcomes for women detainees.
A final inquiry into the validity of the BJMHS was to determine if it is equally accurate across racial groups. Black and Hispanic detainees were slightly less likely than Whites to endorse any of the 6 symptom items, and they were much less likely to report past psychiatric hospitalization or medication.
Use of BJMHS
The BJMHS is used in over 200 jails throughout the United States. It has also been studied in England, Australia, New Zealand, and the Netherlands. While the BJMHS can be administered by corrections staff, the researchers encourage it to be administered by corrections and clinical staff who have interviewing skills and a clear understanding of the purpose of the tool. In addition, it is important to be aware that the BJMHS is better at correctly identifying mental health problems in White males than in males of other races or all women.
References:
- Louden, J. E., Skeem, J. L., & Blevins, A. (2013). Comparing the predictive utility of two screening tools for mental disorder among probationers. Psychological Assessment, 25, 405–411. doi:10.1037/ a0031213
- Steadman, H. J., Robbins, P. C., Islam, T., & Osher, F. C. (2007). Revalidating the brief jail mental health screen to improve accuracy for women. Psychiatric Services, 58, 1598–1601.
- Steadman, H. J., Scott, J. E., Osher, F., Agnese, T. K., & Robbins, P. C. (2005). Validation of the brief jail mental health screen. Psychiatric Services, 56, 816–822.