Improving Access to Mental Health Services for New York State Prison Inmates

Abstract

This article describes a multidisciplinary, multiagency approach to improve access to mental health services in individuals incarcerated within the New York State Department of Correctional Services (DOCS). An interagency work group convened focus groups with staff to assess needs and created a brief screening tool for serious mental illnesses. The NYS Brief Screening Tool (NYS BST) was pilot tested at two DOCS maximum security facilities. Results indicate that the NYS BST is easily administered by staff and well tolerated by inmates. Preliminary measures of sensitivity were within acceptable limits. The routine use of the NYS BST during intake at DOCS reception centers is under discussion. It also will be considered for use throughout the system whenever an inmate needs to be screened for potential mental health service needs.

Mental health screening tools in correctional institutions: a systematic review

“…Common concerns with patient selection included sampling from populations with high rates of mental illness such as health care units and substance abuse programs [ 33 , 36 , 37 ], convenience sampling [ 32 , 38 , 39 ] and high refusal and/or drop-out rates [ 20 , 27 , 28 , 40 - 42 ]. In a number of studies [ 30 , 33 , 43 ] index tests were developed by statistically choosing a subset of items that performed best from a larger test battery, and in other studies [ 38 , 44 ], the index test was embedded within the diagnostic assessment.…”

“…Three studies [ 43 - 45 ] received high risk of bias ratings for the administration of the index test due to not having a pre-specified threshold score, which may result in an over-estimation of test performance due to over-fitting [ 35 ]. A number of studies relied on chart information as a reference standard [ 36 , 46 , 47 ], which may result in misclassification. Flow and timing issues were due to the administration of the reference test predominantly [ 20 , 31 , 42 ], or exclusively [ 48 , 49 ] to those who screened positive, without weighting or other statistical adjustment as was done in two studies [ 17 , 19 ].…”

“…In other studies, the timing between the screening test and reference standard was lengthy (e.g. up to one month) [ 33 ], or the reference standard may have been known prior to screening [ 36 , 37 ].…”

“…At the cut-point with the highest overall accuracy, the GHQ-28 had a sensitivity of 65%, 95% CI [54-75%] and a specificity of 69% 95% CI [60-77%]. The The New York State Brief Screening Tool (NYS BST) performed well for women in particular in a small study [ 36 ], with a sensitivity of 88%, 95% CI [60-97%] and a specificity of 84%, 95% CI [58-95%]. Given that a number of tools appear to perform worse among women inmates, this tool may warrant a more rigorous evaluation in a general offender population as opposed to a health care setting.…”

“…While the sensitivity of the NYS BST was approximately 20% higher for women (88%, 95% CI [60-97%]) than for men (67%, 95% CI [21-94%]), there is a lack of statistical power to determine whether this difference is simply the result of sampling error or a true difference in performance of the tool [ 36 ]. The RDS had high sensitivity in two studies with women [ 43 , 47 ], with lower specificity.…”