Abstract
Background
Few mental health screening tools are validated for prisoners. Existing tools do not guide referral between primary and secondary care pathways.
Aims
This study aimed to assess performance of the CORE‐10 in screening any current mental health problem and current severe mental illness (SMI) in prisoners.
Method
CORE‐10 ratings were compared with mini international neuropsychiatric interview (MINI) version 6.0 ratings and current practice outcomes for 150 male prisoners. Receiver operating curve (ROC) analyses were used to estimate the likelihood with which CORE‐10 ratings matched MINI ratings, yielding “area under the curve” statistics.
Results
ROC analyses suggested a strong relationship between CORE‐10 screen scores and the more detailed assessment both in identifying any problem (AUC 0.85) and severe mental illness (AUC 0.76). Sensitivity was 0.88 and 0.83, and specificity was 0.64 and 0.61, respectively. Retest reliability was moderate (ICC = 0.83). The CORE‐10 identified many cases of “any problem” and of severe illness also identified by the MINI who had not been referred to clinical services in routine practice.
Conclusions
This study provides evidence for the potential of the CORE‐10 for improving appropriate referrals for prisoners to mental health services.