Intended for healthcare professionals


Why do so few patients appeal against detention under section 2 of the mental health act?

BMJ 1995; 310 doi: (Published 11 February 1995) Cite this as: BMJ 1995;310:364
  1. Caroline Bradley, Wellcome research registrara,
  2. Max Marshall, Wellcome health services research fellowa,
  3. Dennis Gath, clinical readera
  1. a Department of Psychiatry, Oxford University, Warneford Hospital, Oxford OX3 7JX
  1. Correspondence to: Dr Marshall.
  • Accepted 12 December 1994


Objective: To determine why most patients do not exercise their right of appeal against detention under section 2 of the Mental Health Act 1983.

Design: Part one—retrospective analysis of the clinical notes of patients detained under section 2 of the Mental Health Act. Part two—interviews with patients on the penultimate day before the deadline for lodging an appeal.

Setting: In part one, five districts in the Oxfordshire Regional Health Authority. In part two, six hospitals from three districts in the region.

Subjects: In part one all patients detained under section 2 in the five districts in 1993 (n=418). In part two interviews with 40 patients detained under section 2 in the six hospitals.

Results: Patients were more likely to appeal if they were educated to A level standard (odds=2.26; P=0.0014) or had had a previous admission (2.19, P=0.0029). Patients with a diagnosis of depression (0.31; P=0.015) or dementia (0.0003, P=0.0001) were less likely to appeal. Compared with those who appealed (n=12) those who did not (n=28) showed less understanding of their rights (P=0.034) and poorer comprehension of sentences from the booklet describing patients' rights (P=0.057). The main reasons given for not appealing were not being aware of the appeals process and being deterred by having to appeal in writing. After they received a full explanation of their rights 12 of those who did not appeal said that they wished to appeal and four did so within the time remaining before the deadline. Of 40 patients, 39 said there should be an automatic right of appeal.

Conclusions: The appeals procedure against detention under section 2 of the Mental Health Act is not a satisfactory way of protecting the civil liberties of patients. If patients were fully informed of their rights they would probably be much more likely to appeal.

Key messages

  • Key messages

  • Among those who do not appeal about half are unaware that they have the right to do so

  • Factors making patients more likely to appeal are education to A level standard and previous admission to a psychiatric hospital

  • Patients who do not appeal have greater difficulty in understanding the booklet explaining their rights under the act

  • If patients were fully informed of their rights they would probably be much more likely to appeal


The Mental Health Act 1983 permits the compulsory admission to hospital of people with mental disorders in the interests of their own health and safety or for the protection of others.1 Most compulsory admissions to hospital are under section 2 of the act (admission for assessment) or under section 3 (admission for treatment). Of these, section 2 is the most commonly used, accounting for about 9000 admissions a year.2 Section 2 lasts for 28 days and provides for admission for assessment of new patients or of known patients requiring reassessment. Admission to hospital under section 2 requires two written medical recommendations and an application to the hospital managers from an approved social worker (or the nearest relative).

Section 2 may be discharged at any time by the consultant psychiatrist responsible for the patient. Patients have the right of appeal against detention to the mental health review tribunal or to the hospital managers. It is the statutory responsibility of the hospital managers to ensure that on admission all patients are informed of this right of appeal.1 This responsibility entails giving the patient a standard booklet outlining their rights and giving an oral explanation. Most appeals are made to the tribunal. To lodge such an appeal the patients must write to the tribunal within the first 14 days after detention.

There is growing concern about how successfully the tribunal system is protecting patients' civil liberties.3 4 5 So far, most research has focused on the process and outcome of appeal hearings6 7 8 9 10 but not on the fact that only about one in four patients detained under section 2 actually lodges an appeal.2 11 There are two explanations for this low rate of appeal, and neither bodes well for the fairness of the present system. The first explanation is that most detained patients are content to remain in hospital, in which case they should not be detained under a compulsory order. The second explanation is that patients are being deterred from exercising their legal rights. The latter explanation is supported by research carried out after introduction of the Mental Health Act 1983 which showed that many patients did not know their rights.5 Concern over the elective nature of the appeals procedure has led to suggestions that all compulsorily detained patients should have an independent examination of their case as a matter of course rather than on application.12

We investigated why so few patients appeal. Our study was in two parts. Part one determined how far demographic and diagnostic characteristics of patients influence their likelihood of making an appeal. Part two determined, firstly, how far failure to appeal is associated with poor understanding of one's rights under the act and, secondly, patients' reasons for not appealing.

Patients and methods


This study was based on an analysis of case notes. The study included patients from five districts under the Oxford Regional Health Authority. Subjects were included if they met the following criteria: admitted in 1993 under section 2 to a psychiatric unit within the five districts; aged over 18 years; and resident in the United Kingdom. From the case notes, a research psychiatrist extracted data on age, sex, ethnic origin, educational attainment, occupation, social class (registrar general's classification), previous admissions, diagnosis, and whether the patient appealed.

Statistical analysis—The demographic and diagnostic variables were given a code of 0 or 1, 0 indicating the absence and 1 the presence of a characteristic of interest (for example, 0=not a member of an ethnic minority, 1=member of an ethnic minority, see table I for details). Each coded variable was then entered as the independent variable in a logistic regression in which the dependent variable was “whether or not appealed.” This analysis produced an unadjusted odds score for each variable (see table I). Odds of greater than one indicated that the presence of the characteristic of interest increased the chances of appealing whereas odds of less than one indicated that it reduced the chances of appealing. Variables with an odds score significant at less than P=0.2 were then simultaneously entered in a stepwise logistic regression.13 This procedure indicated which variables retained a significant effect on the odds of appealing after adjustment for interaction with all other variables entered into the stepwise regression. The data were analysed with the logistic regression analysis program from SPSS for Windows Advanced Statistics Release 6.14


Odds (unadjusted) of making an appeal against detention under section 2 for each independent variable

View this table:


A consecutive series of patients detained under section 2 was interviewed over a five month period. Recruitment was restricted to three health districts because of the travelling involved. These three districts contained six psychiatric hospitals. Permission to interview patients detained under section 2 was given by 28 out of the 32 consultants who worked in these hospitals.

The interview consisted of questions to assess the patient's knowledge of section 2 (see first part of appendix); a test of the patient's reading skills and comprehension based on sentences taken from the standard booklet given to detained patients (see second part of appendix); and a series of open ended questions about the patient's experiences of the appeals procedure and his or her views on how this procedure might be improved. After responding to the questions on knowledge of section 2 patients were given a full explanation of their rights.

The timing of the interview with patients presented an ethical and methodological dilemma. On the one hand, it would have been unethical after the deadline had passed to inform patients of their right to appeal. On the other hand, it was necessary to interview patients near enough to the deadline to be sure that unprompted they would almost certainly not have appealed. Hence patients were interviewed on day 13 of their detention to allow them one day to appeal if they wished after the interview. At the end of the interview, patients who had not yet appealed were asked if they had changed their minds about appealing. Subsequently the investigator established how many of those interviewed actually lodged an appeal before the deadline.

Statistical analysis—Those who did and did not appeal were compared by using non-parametric tests on three variables: knowledge of section 2; reading and comprehension of sentences from the booklet; and whether they reported any difficulties in understanding the booklet.



The case notes of 418 subjects were analysed. Of these, case notes for 34 (8%) were either missing or contained no useful data; 384 subjects therefore entered the study. For these subjects the numbers of missing values on most demographic variables were small, ranging from 0 to 64 (table I). The mean age of subjects included in the study was 39.6 years (95% confidence interval 37.7 to 41.4). There were 195 women and 190 men. Forty eight subjects were members of ethnic minority groups. The case note diagnoses of subjects were schizophrenia and schizophrenia-like disorders 151 (39.0%); personality disorders 51 (13.2%); depression 56 (14.5%); mania 75 (19.4); dementia 21 (5.4%); others 28 (7%); and missing five (1.3%). Of the 386 subjects, 104 (27%) had appealed to a tribunal or a hospital manager, or both.

The unadjusted analysis revealed 10 variables that affected the likelihood of appealing against detention at a level of P<0.2. These variables were entered together into a stepwise regression from which four (level of education, previous admission, diagnosis of depression, diagnosis of dementia) emerged as significantly affecting the odds of making an appeal (table II). Education to A level standard about doubled the likelihood of making an appeal, as did having had a previous admission. A diagnosis of depression or dementia considerably reduced the likelihood of making an appeal.


Results of stepwise logistic regression carried out on independent variables with unadjusted odds significant at 0.2% level

View this table:


A total of 69 subjects detained under section 2 were approached; 40 were interviewed. Of the subjects who were not interviewed, 16 (55.2%) were discharged from section 2 or transferred before day 13; six (20.6%) were too unwell to be interviewed; and seven (24.2%) declined to participate. No further data were collected on the patients who refused to participate, in accordance with ethical guidelines on research on detained patients.16 Of the 40 subjects who were interviewed, 12 (30%) had already lodged an appeal at the time of interview (day 13). The mean age of subjects was 43 years. There were 23 men and 17 women. All subjects who were interviewed were sufficiently coherent to give informed consent by using the Royal College of Physicians' consent form and were able to complete the research interview as described above.

On the test of knowledge of section 2 of the Mental Health Act, those who did not appeal scored significantly lower than those who did (Mann-Whitney U test=97.5; P=0.034). Those who did and did not were compared on their scores for reading and comprehension of three sentences from the Mental Health Act booklet (see second part of appendix for an explanation of how these scores were derived). There was no significant difference between the scores of the two groups on reading of sentences from the Mental Health Act booklet (Mann-Whitney U test=146.5; P=0.34). There was a trend, closely approaching significance, for those who appealed to have better comprehension of the sentences from the booklet than those who did not appeal (Mann-Whitney U test=109.0; P=0.057). Only 20 subjects (50%) showed full comprehension of the most complex of the three sentences from the booklet.

Subjects were significantly less likely to have appealed if they said that they had difficulty understanding the booklet on their rights or had not received it (three of those who appealed had difficulty understanding and nine did not; 18 who did not appeal had difficulty understanding and 10 did not; χ2=5.20; P=0.023).

When asked their reasons for not appealing, 16 (57%) of the 28 who did not appeal said that they were unaware they could do so; five (17.8%) had not given it any thought; four (14%) said that they were happy to be in hospital; two (7%) said that there was no point in appealing; and one was afraid of upsetting the nurses. After being informed of their rights of appeal, nine (32%) of the 28 said that they would be deterred from appealing by having to write a letter, either because of the act of writing itself or because of the difficulties in obtaining pen, paper, and stamps.

Thirty nine (97.5%) of 40 subjects thought that appeals should be automatic for all patients detained under section 2. When asked to suggest ways in which the current system of appeal could be improved, 10 (25%) wanted more information about their rights; seven (17.5%) suggested an easier way of lodging an appeal, such as making an oral instead of a written request or receiving help from a neutral person; one subject suggested a less formal appeal hearing; and one subject suggested a reduction in the time from application to hearing. Twelve subjects (30%) had no suggestions.

The 12 subjects who appealed were asked if they had met any problems in lodging their appeal. Nine reported no problems, two said that they would have welcomed independent help, and one said that he had been hampered by drowsiness because of drug treatment.

After the interview (during which, as explained above, subjects received a full explanation of their rights under the act) subjects were asked if they had changed their minds about appealing. Twelve (42.8%) of 28 who had not appealed said that they would now like to appeal; four of them lodged an appeal before the deadline. Of these four, one was discharged by the tribunal.


There are two stages in the process by which a patient comes to make an appeal against detention under section 2. We will consider the findings of our study in relation to these stages.

At the first stage the patient must find out how to appeal. At this stage, as shown by the retrospective study (study one), patients with previous experience of the system are at an advantage (many probably know how to appeal already). In the absence of any independent help most new patients must rely on information from the booklet on the Mental Health Act. The findings of the prospective study (study two) that, for example, 57% of those who did not appeal were unaware of the appeals process suggest that understanding the booklet is a difficult task for many patients.

An interesting question is why so many patients find it difficult to understand the booklet, which is not that difficult to read. One explanation is that severely ill patients are unable to understand the booklet because their thinking processes are incoherent. This would explain the finding of the retrospective study (study one) that patients with dementia are much less likely to appeal. It would not, however, explain the findings of the prospective study (study two) as incoherent patients were effectively excluded from this study by the necessity of obtaining informed consent and participating in an interview. A more likely explanation is that many detained patients, though not incoherent, may have reduced powers of concentration and attention, perhaps because they are extremely anxious or depressed or are taking sedative drugs. In this context the findings of study one make sense: that well educated patients are more likely to appeal and that patients who are depressed are less likely.

At the second stage of the process of making an appeal patients must have sufficient motivation and concentration to obtain writing materials and to write and post a letter. This task is likely to be easier for patients who are better educated and for those who are familiar with hospitals. The task is likely to be more difficult for patients who are cognitively impaired—for example, those suffering from dementia—and also for patients who are feeling pessimistic and lacking in energy—for example, those suffering from depression.

Overall our results suggest that the appeals procedure for section 2 of the Mental Health Act is not a satisfactory way of protecting the civil liberties of patients. The procedure has two main flaws: firstly, the patient has to initiate the appeal and, secondly, there is an unsatisfactory method for informing patients of their right of appeal. As a result the procedure favours patients who are well educated or have had previous admissions, but it works against patients suffering from depression or dementia. If patients were fully informed of their rights there might be a considerable increase in the number of appeals.

CB and MM were supported by the Wellcome Trust. We thank all those who helped in the study; particularly patients, nurses, doctors, and staff of medical records departments.


Questions to assess knowledge of section 2

The subject was asked the four questions listed below. The questions relate to the main aspects of section 2. For each correct answer to questions 1 to 3 one point was awarded. If the subject gave both correct answers to question 4 two points were awarded. If the subject gave only one of the two correct answers to question 4 one point was awarded. The points awarded were then summed to give a score ranging from 0–5.

  1. Are you free to leave the hospital at the moment? (yes=0, no=1)

  2. Can you tell me the name of the part of the Mental Health Act under which you are being detained? (section 2=1, other=0)

  3. Can you tell me how long section 2 lasts? (28 days=1, other=0)

  4. Do you know if there is any way that the section can be lifted before 28 days are up? (2=discharge by regional medical officer and after appeal to tribunal; 1=either discharge by regional medical officer or after appeal to a tribunal; 0=did not know or incorrect answer).

Test of reading and comprehension of sentences from the booklet

This test was based on three sentences taken from the Mental Health Act booklet that is given to detained patients to explain their rights. The readability of this booklet was assessed with the grammar checking facilities of Microsoft Word for Windows (version 2). The readability statistic was the Flesch reading ease score, which calculates readability based on the mean number of syllables per word and the mean number of words per sentence. Scores range from 0 to 100. Higher scores indicate that a greater number of people could readily understand the document. The booklet was found to have a Flesch reading ease of 76, corresponding to “fairly easy” on the norms for the scale.

To test reading and comprehension skills three sentences from the booklet were used. The sentences had reading ease scores at, above, and below the mean for the booklet. The sentences were:

  1. “After 28 days you can only be kept in hospital if your doctor thinks you need to stay longer and makes new arrangements (under section 3 of the Mental Health Act).” Reading ease of 60.8 (average)

  2. “You can ask the Tribunal to look at your case by writing to them or sending them a form which the hospital can give you.” Reading ease of 76.6 (fairly easy)

  3. “You must not leave before the end of the 28 days unless a doctor tells you that you can.” Reading ease of 89.6 (very easy).

To test reading and comprehension patients were first asked to read sentence 1 aloud. If subjects could read all the words in the sentence at a conventional speed they were given three points. If the subject had problems reading sentence 1 then the same procedure was repeated for sentence 2, a correct reading of this sentence gaining two points. If the subject had difficulties with sentence 2 the same procedure was repeated for sentence 3, for which a correct reading gained one point. Subjects unable to read any sentences were given no points. To test understanding a similar scoring procedure was used except that subjects were asked to explain, in their own words, the meaning of the sentences.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
View Abstract