The Millan Committee (Reference MillanMillan, 2001) was established in 1999 to undertake a comprehensive review of the Mental Health (Scotland) Act 1984 (MH(S)A). It reported that it was extremely difficult to obtain statistical information concerning many aspects of the Act including diagnosis, duration of short-term detention and what happens to patients who are discharged from detention. The review of the literature commissioned as part of the review process (Reference Aitkinson and PattersonAitkinson & Patterson, 2001) concluded that there remains a ‘general dearth of good research on, or audit of, mental health legislation in Scotland’.

Sections 24 and 25 of the MH(S)A are emergency detentions similar to Sections 4 and 5(2), respectively, of the Mental Health Act, 1983 (MHA). They can be applied by any fully-registered medical practitioner and allow patients to be assessed in hospital for 72 hours, but not treated. In practice the doctor involved is usually a senior house officer or general practitioner (GP); consent is required from an approved social worker or relative. Section 24 applies to patients in the community and Section 25 to patients already admitted to hospital but they are otherwise identical.

Section 26 lasts for 28 days and is similar to Section 2 of the MHA, but can only be applied to patients already on a Section 24 or 25. It allows treatment of patients, but only comes into force when the emergency detention has run its course. Only doctors with special expertise in the diagnosis or treatment of mental disorder approved under Section 20 of the MH(S)A are eligible to apply a Section 26. This is similar to Section 12 approval under the MHA, but approval is usually limited to consultant psychiatrists and specialist registrars in Scotland. Section 26 can be extended after 28 days to allow a Section 18 application to be completed.

Section 18 lasts for up to 6 months in the first instance and is similar to Section 3 of the MHA, except that the application has to be granted by a Sheriff (the Scottish equivalent of a magistrate). This has made it unwieldy in practice and few patients are placed directly onto a Section 18, although the legislation was designed to make this the normal route of detention. Section 18 appeals are made directly to the court and there are no mental health tribunals. Two medical recommendations are required; one must be from an approved doctor and the other is usually from the patient's GP.

This study describes patients detained under the MH(S)A in south Glasgow from 1 April 1997 to 31 March 1998 and discusses the possible impact of the changes proposed by the Millan Committee.

The study

The south sector of the Greater Glasgow Primary Care Trust had a total population of 290 000 at the time of the study. It was an urban area, with above-average deprivation. Data were collected on all patients initially detained under civil sections of the MH(S)A between 1 April 1997 and 31 March 1998. Patients detained under the Criminal Procedure (Scotland) Act (1995) and Part VI of the MH(S)A were excluded. Information from the patient information system was supplemented by a manual search of all section forms. Diagnosis was obtained from the discharge diagnosis on the patient information system. Deprivation was based on the Carstairs scores (Reference Carstairs and MorrisCarstairs & Morris, 1991) of postcode areas. Missing data were obtained from the case notes. Unpaired t-tests were used for parametric data and Mann—Whitney U-tests for non-parametric data.

Results

There were 283 episodes of formal admission involving 240 patients. A total of 30 patients had two episodes, five patients had three episodes and one patient had four episodes. There was no significant difference in the age, gender or diagnoses of patients detained on two or more occasions. The mean age was 44 years (s.d. 20, range 16-93 years) and 20% were 65 or over. A total of 53% (151/283) were male and 2% were homeless. The main disorders were organic F0-9 (33/283; 12%), substance misuse F10-19 (15; 5%), schizophrenia, schizotypal and delusional F20-29 (113; 40%), bipolar F30-31 (52; 18%), depression F32-33 (47; 16%), neurotic, stress-related and somatoform F40-49 (9; 3%), eating disorders F50-59 (3; 1%), adult personality and behaviour F60-69 (6; 2%) and no psychiatric diagnosis (5; 2%) (‘F’ disorders as defined by ICD-10). There were no significant differences with the national statistics produced by the Mental Welfare Commission (1998) for the same period (Table 1).

Table 1. Episodes of detention from 1 April 1997 to 31 March 1998

	South Glasgow n (%)	Scotland n (%)
Section 24/25 to informal	134 (47)	1805 (40)
Section 24/25 to 26	108 (38)	1700 (38)
Section 24/25 to 26 to 18	36 (13)	827 (18)
Direct to Section 18	5 (2)	142 (3)
Total	283 (100)	4475 (100)

The median duration of detention was 6 days (inter-quartile range (IQR) 3-31). A total of 31% were detained for 1 day, 4% for 2 days and 39% for 3 days. After this, the proportion discharged from detention was <2% on any day, apart from 19% on day 31. Only 17% were detained for >31 days and 2% for >1 year. A total of 21 (7%) of the episodes included a period of leave of absence, with a median duration of 155 days (IQR 107-267). There was a non-significant trend towards increased episodes of detention in areas of higher deprivation (P=0.065). Time of detention (279/283) was between 01.00 h and 09.00 h in 7%, 09.00 h and 17.00 h in 52% and 17.00 h and 01.00 h in 41%.

Consent was usually provided by the mental health officer (131/279; 47%) or relatives (116/279; 42%), but was not provided for 11% of detentions (32/379). The main reason given was insufficient time (23/32) and in only four cases had the doctor been unable to contact a mental health officer or relative.

A total of 55% were admitted to hospital under the MH(S)A and 45% were detained after admission. The latter were more likely to be detained for a longer period (median 3 v. 29 days, P<0.001). A total of 59% of Section 24 (community) emergency detentions were not extended after 72 hours, compared with 35% of Section 25 (hospital) detentions (Fig. 1). Only three were admitted under Section 18. Two were detained directly under Section 18 after admission to hospital and two transferred from England after being detained in the community under Section 2 or 3 of the MHA.

Fig. 1. Progress of emergency detentions.

There were 144 28-day Section 26 detentions, including the two patients transferred from England; 25% (36/144) of these were converted to Section 18. The other 108 had a mean duration of detention of 26 days (s.d. 10) and a range 1-54 days. In 50/108 cases the Section 26 was allowed to run for the full 28 days.

The majority of detentions (157/283; 56%) were ended by the responsible medical officer (usually the consultant) but 42% (119) of sections were allowed to lapse. One patient was transferred to England and three died. The Sheriff did not approve the Section 18 application of three patients (3/44; 7%). No patients were discharged by the Sheriff on appeal. There was no information available on the use of the appeal process by patients to the Sheriff court or Mental Welfare Commission.

The median duration of the admission that included the period of detention was 36 days (IQR 10 - 81). A total of 56% of patients were admitted for over 31 days and by this time, the majority of patients were informal (81%).

The point prevalence of patients detained on 1 April 1997 was 52; 38 were detained under Section 18, seven under Section 26 and seven were on leave of absence. There were no patients on emergency detention. The mean age was 42 (s.d. 14, range 24-79) and 62% were male (32/52). They had been detained for a median of 8 months (IQR 2-39 months; range 5 days-192 months). A total of 80% (35/44) had schizophrenia (F20-29), 7% (3/44) organic disorders (F0-9), 7% (3/44) bipolar affective disorders (F30-31) and 7% (3/44) psychotic depression (F33.3).

Discussion

This study provides a detailed description of the use of the Mental Health (Scotland) Act 1984 in a defined geographical area and fills in some of the statistical gaps in the national data identified by the Millan Committee. One of the main limitations of this study is that data might not be representative of patterns of detention across Scotland. This could be because Glasgow is an urban area with above average deprivation (Reference Gruer and MorrisonGruer & Morrison, 1999), or because data were missing or inaccurate (Reference Nemitz and BeanNemitz & Bean, 1995).

The annual incidence of detentions (98 per 100 000) was similar to the reported level in Scotland (Mental Welfare Commission, 1998) and England and Wales (Department of Health, 1998). The use of compulsory powers remained constant between 1945 and 1975 and it was suggested that an annual number of 45 compulsory admissions per 100 000 population represented an inevitable basic level (Reference ElliotElliot, 1979). Since the 1980s, there has been a marked rise in the number of detained patients in both Scotland (Reference Aitkinson and PattersonAitkinson & Patterson, 2001) and England (Reference Hotopf, Wall and BuchananHotopf et al, 2000). The reason for this dramatic increase is unclear, but it does not appear to be linked to changes in legislation (Reference ElliotElliot, 1979; Reference MccreadieMcCreadie, 1989).

This study found that in south Glasgow, only 2% of detentions were initiated by an application to the Sheriff court under Section 18. A total of 98% were 72-hour emergency detentions made by one medical practitioner. This finding alone would support the need for a review of legislation as Section 18 was intended as the normal route into detention in the MH(S)A, but this rarely happened in practice. The reason for this is that it usually takes around 2 weeks to arrange a hearing after the application is submitted to court.

The Millan Committee recognised that changes might increase the workload of mental health professionals and that without adequate services the aspirations underlying the new Act are unlikely to be met. Over half of detentions were outside the standard 9 to 5 working week in our study. This could have major implications as the Millan Committee recommended that approved doctors should detain patients in the community and written reasons would have to be given for the use of emergency detention. The majority of approved doctors in Scotland are consultant psychiatrists. Their involvement in out-of-hours assessment could have major resource and recruitment implications.

There is some evidence from this study that legislation can have an effect on the duration of detention. A total of 39% of detentions lasted for 3 days and 19% for 31 days because these are the maximum duration of the emergency and assessment detentions, respectively. These peaks were not reported in studies in England and Wales (Reference Wall, Churchill and HotopfWall et al, 1999) and appeared after the introduction of the MH(S)A, 1984 (Reference MccreadieMcCreadie, 1989). It is unlikely that the total number of people detained will be changed by the new Act, but it could alter the duration of the detention.

The standard response time for mental health officers in Glasgow is 1 hour and insufficient time was the main reason why patients were detained without consent. Relatives provided consent in 42% of cases. The view of the Millan Committee was that, even if a mental health officer was not involved, there was no substantial benefit to retaining a relative's right to consent as this added to the distress of the family without providing additional safeguards to users. Relatives often accompany patients referred as emergencies (Reference Taylor, Lawrie and GeddesTaylor et al, 1996) and threatening behaviour is more common in patients detained without consent (Reference DeeringDeering, 1994). If relatives were unable to provide consent, you would predict that the work of mental health officers would double. The proportion of cases detained without consent could rise, because the main reason given for lack of consent was lack of time rather than non-availability. This has to be balanced against the potentially detrimental effect on family relationships that relatives providing consent can have. There is limited research in this area but a postal study (Summers, 2002; personal communication) suggested that the detrimental effects might have been overstated. Only one out of 15 relatives said they would not give consent again and seven reported improved family relations compared with three reporting negative consequences.

The Millan Committee described the lack of information on the outcome of patients discharged to informal status after emergency detention as a serious gap in the statistics. Most patients remained in hospital voluntarily when their detention ended but there was a significant difference for patients detained in the community compared with hospital. Patients detained in the community were less likely to be detained for over 72 hours (41% v. 65%) and the average duration of detention was much shorter (3 days v. 29 days). This finding would support the established pattern of mental health legislation differentiating patients detained in the community from those detained after admission to hospital.

The Millan Committee suggested that the large proportion of patients discharged from detention within 72 hours may suggest that insufficient effort was made to secure the patient's admission on a voluntary basis, but no evidence was provided to support this view. The use of Section 4 (72 hour emergency admission) was halved in England and Wales between 1984 and 1996 as services were encouraged by the Mental Health Act Commission to use ‘approved’ doctors to apply 28-day or 6-month sections. Despite this, the total number of formal admissions increased by 63% (Reference Hotopf, Wall and BuchananHotopf et al, 2000). The reasons for this increase remain unclear, but might include increased drug use, a change in the willingness of psychiatrists to tolerate risk and revolving door detentions because of pressure on beds. Our study suggested that the influence of revolving door detentions is unlikely to explain the increase in formal admissions as only 15% were re-detentions.

It is proposed that a new mental health tribunal would replace the role of the Sheriff. The reasons given were that only a minority of patients attended court or were represented and there were wide variations in practice. The Millan Committee reported that very few applications for detentions were refused and the figure in south Glasgow was 7%. Only 17% of patients were detained for over 1 month in our study and it remains to be seen whether a new power of appeal to a tribunal will alter this. There is a risk that alternative systems might involve patients more without increasing their protection. Changes in legislation can have less longer-term impact than anticipated, as psychiatrists and courts tend to use legislation pragmatically (Reference AppelbaumAppelbaum, 1997).

The recommendations of the Millan Committee have been broadly welcomed, but it will be 2004 before any changes in the Mental Health Act will come into effect. Until then, the effect of these changes remains speculative.

Declaration of interest

None.