Prevalence of mental disorder in remand prisoners: consecutive case study

Introduction

The problems of mentally disordered people in prisons were highlighted in a series of articles in 1984.1 2 Bluglass subsequently drew attention to the lack of improvement despite numerous inquiries into the management of mentally disordered people in prison.3 He pointed out the inadequacy of Home Office and Department of Health and Social Security reports, which relied heavily on the limited information available from censuses of mentally disturbed prisoners carried out by prison medical officers.

The effectiveness of health screening by prison medical staff has also been questioned.4 5 The conditions and time constraints militate against the detection of clinically important information and the screening questionnaires used are of doubtful validity. In addition, some prisoners are difficult to deal with and allegations have been made of poor standards and lack of training in prison medical staff.

A recent national study of convicted prisoners using data collected by psychiatrists reported that 37% of sentenced prisoners have mental disorders.6 The rates in remand prisoners are probably higher (S Dell, personal communication), partly because mentally disordered people are often remanded in custody for psychiatric reports. Evidence from North America suggests that mentally disordered people are more likely to be arrested than those who are not mentally disordered in similar circumstances.7 Factors such as homelessness8 and petty offences that are associated with mental disorder make remand more likely.

British research in the 1980s reported high rates of psychiatric morbidity in remand prisons,9 10 but these studies may have underestimated the problem as they were retrospective case note studies relying on diagnoses by prison medical staff. More recent studies of remand prisoners have confirmed the high rate of mental disorder in this population.11 12 However, it is still unclear how much of the mental disorder is recognised by prison healthcare services.

We conducted this study to determine the prevalence of mental disorder in remand prisoners at the time of their reception into prison and the efficacy of the prison reception screen in detecting this. It is the first part of a longitudinal study of a cohort of new remand prisoners who will be followed throughout the course of their remand.

Results

During the study 606 men were remanded to Durham prison (27 more than once). Thirty seven men returned to court the morning after admission and did not return to the prison. Of the remaining 569 men, 549 (96%) consented to be interviewed, 19 refused, and one was unfit for interview. In all, 528 (96%) interviews were fully completed; 21 were only partly completed because of language barriers, mental state disturbance, and situational constraints. We recorded IQ scores for 441 men. Poor concentration, agitation, language difficulties, or other adverse factors made the testing unreliable in the remainder.

In the 116 jointly rated interviews, 51 lifetime diagnoses of mental disorder and 184 separate substance misuse diagnoses were recorded by either one or both raters. Diagnostic agreement occurred in 216 of these ((kappa) = 0.902). Most disagreements were over diagnoses of personality disorder ((kappa) = 0.761) and adjustment disorder ((kappa) = 0.645). The two researchers agreed on all 15 lifetime diagnoses of psychosis made during inter-rater interviews ((kappa) = 1.0).

Table 1 summarises the characteristics of the population. Most of the men (378, 66%) were white and aged 30 years or under. Almost 80% of the population were unemployed or on sickness benefit. Of the 441 tested, 389 (88%) had an IQ score below the general population mean, and 57 (13%) scored 70 or less.

Discussion

Our findings are based on inmates of one prison with a large catchment area, but there is no reason to believe that Durham prison differs significantly from other remand prisons. The reception screen is used nationally, and the demands on prison health staff are similar. We believe, therefore, that the results of this study can be generalised throughout the remand population.

Twenty six per cent of new prisoners had a current mental disorder, nearly one third of whom had a serious disorder. Most of these disorders were undetected and so untreated. Although these men may have had their mental disorder identified and treated later on in their remand, our impression from our initial follow up is that this is unlikely. In a busy remand prison abnormal behaviour is often tolerated or perceived as a discipline problem and dealt with punitively, while the “quietly mad” are ignored.

Failure to identify psychiatric illness in men entering remand prisons means that the opportunity for treatment is lost. This is important since many of these men might not otherwise come to the attention of healthcare services and the severely psychotic inmates may be at high risk of suicide.

The reception screening undertaken by the prison service is neither sensitive nor specific for detecting mental disorder. Reasons for this may include lack of appropriately experienced or trained staff to conduct the screening, a questionnaire of doubtful validity, time constraints, a highly mobile population, and a perception among prisoners that prison healthcare staff do not have their best interests as a priority.

Health screening on reception into prison should provide an important opportunity to detect mental disorder and provide prompt and appropriate treatment that, particularly in the case of remand prisoners, could be continued in the community. However, to achieve this screening will need to become more effective. If more psychiatric morbidity is identified, the availability of psychiatric bed may become a problem. Beds need to be available at all levels of security to take patients rapidly from prison or as part of a court diversion scheme. Psychiatric beds in the Northern region, as in the rest of the country, are in short supply and often impossible to find at short notice. Our results provide some indication of the extent of the shortfall.