Abstract
The small number of published studies on the use of Section 136 (S136) of the Mental Health Act 1983 may reflect neither the current situation nor the true national picture as most of the research comes from large urban centres and was published before 1997. We studied the use of S136 between 2002 and 2006 in Gloucestershire, a rural English county with a population of approximately 585,000, by analysing data held by the police and local mental health services. On average 192 subjects (range 176–203) were detained each year under S136, equating to a population rate of 32.8 S136 detentions per 100,000 per year in Gloucestershire. In contrast to other published studies, ethnic minorities, particularly Afro-Caribbean males, were over-represented only in a minor way. Of those individuals detained, about a third were admitted, a lower rate than in other studies. Acts or threats of self-harm were common (55%), but acts or threats of violence (28%) and evidence of intoxication (16%) were present in a minority, suggesting that detainees are more likely to pose a risk to themselves than others.
Introduction
In England and Wales, Section 136 (S136) of the Mental Health Act 19831 gives a police constable the power to detain an individual that they find to be mentally disordered in a public place and in need of care and control. The individual is conveyed to a Place of Safety (POS) to allow for further assessment. Thus, for example, a dishevelled and thought-disordered man who is shouting about persecutory delusions and wandering in front of traffic on a public road might be detained by a police officer under S136 and taken to a POS for further assessment. The nature of the POS varies from area to area but is typically either the police cells, a psychiatric hospital or sometimes an emergency department in a general hospital. Usually the police officer will have considered less restrictive options for managing the situation before using S136. As an alternative, if an offence has been committed, the individual might be arrested and taken to the cells where an assessment under the MHA 1 could subsequently be organized, outwith the provisions of S136.
There is a paucity of research on S136. Most of it emanates from large urban centres in England, 2–5 and most of the research was published before 1997. Exceptions are the review conducted by a working party of the Royal College of Psychiatrists, 6 which led to the setting of new guidelines for the use of S136 in England 7 and the recently published Independent Police Complaints Commission (IPCC) report into the use of police cells as a POS. 8 This latter study considered only the use of police cells rather than other POSs and therefore the published research, taken as a whole, may reflect neither the current situation nor the complete national picture. 6 Although professionals' knowledge of details of S136 law have been studied, 9 little is known about how S136 is used in practice.
Official Department of Health (DoH) statistics are limited to S136 detentions where a hospital is used as a POS and fail to capture the situations where police cells are used. The IPCC study, 8 using police data, provided useful information regarding use of police cells as a POS. However, having two separate sources of national data covering the use of the POS is problematic for monitoring and research purposes. This difficulty in obtaining reliable data on the use of S136 has been noted elsewhere. 6
The current Code of Practice for the MHA 1983 10 states that police cells should only be used as a POS on an ‘exceptional basis’ in S136. While hospital settings are increasingly being used, a study in 2005 showed that 34% of localities used police cells exclusively as a POS. 11 The IPCC study 8 compared police and DoH data, showing that two-thirds of S136 detainees were held in police cells with only a third being taken to a hospital POS. Thus the use of police cells as a POS is far from ‘exceptional’. National policy supports the development of alternatives to using police cells as a POS and £130 million for this purpose has been made available from the DoH to mental health services. The Royal College of Psychiatrists recently published updated guidance on standards for the use of S136. 7
To advance knowledge and inform the development of local practice, we studied the use of S136 between 2002 and 2006 in Gloucestershire, a rural English county with a population of approximately 585,000.
Method
This study was part of a larger research project into S136 which included investigating the attitudes and beliefs of professionals involved in the detention process and also the experience and views of S136 detainees and their carers. Favourable opinion for the research was obtained from the local research ethics committee and the mental health trust research governance committee. The methodology was a retrospective analysis of data collected as part of routine monitoring of S136 activity in Gloucestershire.
Two data sources were used. First, data were obtained from the police (Gloucestershire Constabulary) computer systems for the three years 2003–2006. These data are routinely collected by the police when individuals are booked in after arriving at the police cells. Before being supplied to the researchers, the data were anonymized and collated by police staff. Secondly, we had access to data from a paper-based monitoring system on S136 activity held by 2gether NHS Foundation Trust (provider of specialist mental health services) during the period 2002–2004 (1 year 10 months). The Trust Audit Department anonymized and collated the data before they were made available to the researchers.
Data from both sources were entered onto a software package (Microsoft Excel) on a password-protected computer in a locked office in Trust premises. Analysis was by Microsoft Excel and Statistical Package for Social Sciences (SPSS).
Results
Results are shown in Tables 1 –3. For ease of comparison we have, where possible, placed results from the two data sources side by side. Table 1 shows numbers detained and demographics; Table 2 the circumstances of detention; and Table 3 the outcomes of assessment.
Numbers detained under S136 and their demographic characteristics
Where the totals for the analysis are less than the total number of responses there was missing data, i.e. no response. Percentages may not add up to 100% due to rounding of figures
Circumstances of detention under S136
Where the totals for the analysis are less than the total number of responses there was missing data, i.e. no response. Percentages may not add up to 100% due to rounding of figures
Outcome of S136 assessment
Where the totals for the analysis are less than the total number of responses there was missing data, i.e. no response. Percentages may not add up to 100% due to rounding of figures
Police data
In the three years studied, a mean of 192 subjects (range 176–203) were detained each year under S136. This equates to a mean rate of 32.8 S136 detentions per 100,000 per year in Gloucestershire.
People subject to S136 detention were 61% men. The commonest age range for detainees (33%) was 35–44 years, the range being from under 18 years to over 65 years. The ethnic mix of detainees was broadly in keeping with the ethnic mix of Gloucestershire residents: 93% were white, 3% black, 1% Asian, the rest being ‘other’ or ‘unknown’. The 2001 Census showed Gloucestershire's non-white ethnic minority population (2.8%) to be below the national average. 12
Detentions were distributed fairly evenly through the week but with a slight trough midweek on Wednesdays. The busiest time of day for S136 detentions was 18:00 hours to midnight (34%), with the quietest time being 06:00 hours to midday (14%). Detentions were fairly evenly distributed throughout the year, with a slight dip in the first quarter (January to March).
Just over a third (34%) of the detainees were admitted to hospital. Of those not admitted, the vast majority (95%) left the cells with no further action. A minority (5%) were subject to further police action such as being charged or cautioned, or being detained under outstanding warrants.
Mental health services data
This data set comprised paper monitoring forms completed over 22 consecutive months between 2002 and 2004. These forms were completed by the police involved in the detention of each individual along with the Approved Social Worker and/or doctor (usually a senior psychiatrist) who undertook the assessment of the detainee. The monitoring forms collected a combination of demographic and clinical details.
These forms recorded 186 detentions over the 22 months, a mean of 101 detentions per year. The mean age of detainees was 37 years (range 14–91); 55% were male. The ethnicity of detainees was: 95% white, 3% black and 2% other or unknown; 8% of detainees came from addresses outside the county and 5% had no fixed abode or no known address. Repeated detentions were evident, as 9% of detainees had been detained under S136 on at least one previous occasion.
Detentions were fairly evenly distributed throughout the week. About two-thirds of all detentions occurred in the evenings or at night. The mean duration of detention under S136 was four hours 30 minutes (range from less than 1 hour to over 15 hours). Of the detainees, 55% had presented with acts or threats of deliberate self-harm, 28% had presented with actual or threatened violence to others, while 16% were documented to have had evidence of intoxication with drugs or alcohol at the time of arrest.
Broad clinical diagnoses were also recorded on the monitoring forms. Severe mental illnesses such as schizophrenia or bipolar disorder were diagnosed in 13%. More minor affective illnesses and anxiety disorders were diagnosed in 31%. Personality disorder was present in 8% and substance misuse in 11%. Small numbers had learning disability (2%) and medical conditions (1%).
The outcome following assessment was: 61 detainees (33%) were admitted to a psychiatric hospital, 113 (61%) were not admitted and in 12 cases (6%) the outcome was not known. Of the 61 individuals who were admitted, just under half (46%) were detained and admitted under a further section of the MHA, 1 the rest being admitted informally without use of a Section.
Of the 113 individuals not admitted to hospital, no further immediate assistance with mental health or social care needs was recorded for 65 (58%). The remaining 48 (42%) had assistance in a variety of ways: 22% were already known to community psychiatric services and the relevant team was notified, 13% were referred to community psychiatric teams, 4% to specialist drug or alcohol services and 9% to non-statutory agencies (including some drug/alcohol and counselling agencies). A further 13% were given assistance with accommodation or travel.
Discussion
During the period of the study, the police cells were the only designated POS in Gloucestershire. Police procedures for booking in detainees taken to the cells are robust and we conclude, therefore, that the police data showing a mean of 192 subjects detained under S136 per year in the cells are likely to be accurate. Anecdotally, however, it was known to the authors that during the study period the police had, in a few cases, taken S136 detainees to non-designated POSs including local psychiatric hospitals and hospital emergency departments. These detentions, which did not go to the cells, are unlikely to have been included within the police data. However, the numbers were small and in our view they are unlikely to have affected the accuracy of the data in any significant way.
The same cannot be said for the mental health services data, which relies on clinicians and police officers filling out a paper form and sending it to a central collation point. Locally it was already acknowledged, through discussions at inter-agency monitoring group meetings on S136, that the paper form system was failing to capture all S136 activity. Our figures suggest that, compared with the police data, this method only captured about 50% of the S136 detentions as the police data indicated a monthly average of 16 S136 detentions, while the other method gave 8.5. Although the time periods for data collection using the two methods were different (2003–2006 and 2002–2004), they overlapped and we have no reason to think that rates of S136 detention varied during this time sufficiently to explain the wide discrepancy in detention rates shown by the two sets of data. We conclude, therefore, that the mental health services data captured only part of the overall S136 activity, probably about 50%.
The demographic results and admission rates from the two data sets were similar. This supports the argument that the mental health services data were representative of the overall S136 activity. As such, we infer that the other findings from that data set are generalizable to the S136 population in Gloucestershire.
Compared with other published research on S136, recently reviewed, 6 our study found lower admission rates (33–34% versus 82–85%) and only minor over-representation of ethnic minorities. The reasons for these differences are unclear but may include inaccuracies in data collection, geographical differences in clinical/police practice and different population groups.
The prevalence rates of the various categories of mental disorder also warrant some attention. As the categories were not defined, the ‘diagnosis’ was impressionistic and it is difficult, therefore, to draw firm conclusions. However, it is of interest that our finding of a low prevalence of serious mental illness (13%) and minor mental illness (31%) was associated with a lower admission rate than in other studies. 6
Our study also showed a mixture of ‘risk’ behaviours associated with detention. Acts or threats of self-harm were common (55%), but acts or threats of violence (28%) and evidence of intoxication (16%) were only present in a minority. This suggests that those detained under S136 in Gloucestershire were more likely to pose a risk to themselves than to others.
We also found that most S136 activity occurred in the evenings and at night. This raises the question of whether, within working hours, individuals in crisis are dealt with by other methods, with S136 acting as an important ‘safety net’ at other times.
Further research in this area would be useful. Very little is known about what happens to those detained under S136 who are not then admitted to hospital. Our study shows that 42.5% of those not admitted were offered further assistance, but whether they actually went on to receive that assistance is not known. Clinical experience suggests that they are characteristically individuals who find it hard to access mainstream services. The finding that 9% of detainees had previously been detained under S136 was also of interest and raises the issue of whether mental health services and other agencies could do more to support such individuals in the community, such that repeated detention is not necessary.
Since this research was undertaken, a new ‘health-based’ POS has been developed on the site of our local psychiatric hospital. We believe that this facility, which opened in early 2009, will be of benefit to S136 detainees by providing an alternative to the current exclusive use of the local police cells as a POS. We would support the development of a national system for monitoring and recording use of Section 136 in all POSs.
Footnotes
ACKNOWLEDGEMENTS
We gratefully acknowledge the help and support in designing and conducting this research that we received from the following individuals: Sergeant Julie Gardener (Gloucestershire Constabulary), Julie Hapeshi and Chris Foy (Research Development and Support Unit, Gloucestershire Hospitals NHS Trust), Dr Rhys Watkins (General Medical Practitioner and former Forensic Physician, Gloucestershire Constabulary), Dr Delia Parnham-Cope (Consultant in Emergency Medicine, Gloucestershire Hospitals NHS Trust).