Daytime night attire as a therapeutic intervention in an acute adult psychiatric in-patient unit

Aims and Method Dressing in-patients in night attire during daytime is currently practised in many in-patient psychiatric units, despite the lack of evidence to support its benefit in reducing absconding or self-harm. Using a triangulation design, we investigated the prevalence of, attitudes towards and associations of this practice in an acute psychiatric in-patient setting in the Republic of Ireland. Results Case-note review revealed a high prevalence of this practice (57%) and its significant association with involuntary admission. Nursing staff believed that using night attire was effective at reducing absconding and self-harm, and that only voluntary patients should retain the right to choose their clothes. Most patients interviewed were uncomfortable in night clothes and indicated that they should be entitled to choose what to wear. Clinical Implications Night attire is regularly used for risk-management, despite lack of evidence supporting its efficacy and negative attitudes towards it in many patients. This practice and the reasons for its implementation deserve medical documentation.

Dressing in-patients in night attire during daytime is currently practised in many in-patient psychiatric units, despite the lack of evidence to support its benefit in reducing absconding or self-harm. Using a triangulation design, we investigated the prevalence of, attitudes towards and associations of this practice in an acute psychiatric in-patient setting in the Republic of Ireland.

RESULTS
Case-note review revealed a high prevalence of this practice (57%) and its significant association with involuntary admission. Nursing staff believed that using night attire was effective at reducing absconding and self-harm, and that only voluntary patients should retain the right to choose their clothes. Most patients interviewed were uncomfortable in night clothes and indicated that they should be entitled to choose what to wear.

CLINICAL IMPLICATIONS
Night attire is regularly used for risk-management, despite lack of evidence supporting its efficacy and negative attitudes towards it in many patients. This practice and the reasons for its implementation deserve medical documentation.
Restraint and restriction of in-patients with mental illness is not new, as early asylum care depicts patients chained to walls and denied clothing (Belkin, 2002). Although care in Irish psychiatric hospitals is far removed from this, many open acute psychiatric in-patient units maintain a policy of placing newly admitted patients, both voluntary and involuntary, in night attire and withholding their day clothes, apparently for the purposes of reducing absconding and self-harm.
Insistence on night clothes is perceived by some patients as a means of controlling and restricting their freedom (Wallace et al, 1999). It may contribute to institutional neurosis (Barton, 1959) and is arguably contrary to Articles 1, 5 and 9 of the Universal Declaration of Human Rights (United Nations, 1973) and Article 3 of the European Convention on Human Rights (Council of Europe, 1950). More recently, Article 9.1 of the Council of Europe Committee of Ministers recommendations to member states concerning the protection of the human rights and dignity of persons with mental disorder (Council of Europe, 2004), states that the environment and living conditions of persons with a mental disorder should be equitable, in so far as is possible, with those of persons with similar age, gender and culture in the community, a recommendation which is not fully adhered to by the use of night attire in this context. The diagnostic inquiry on the human rights of persons with mental disorders in Central America considered that keeping patients in pyjamas violated their right to quality of life (Levav & Gonza¤ lez Uzca¤ tequi, 2000). The use of night attire to prevent absconding is neither supported by literature reviews (Bowers et al, 1998), nor part of recently implemented anti-absconding interventions (Bowers et al, 2003;Bowers et al, 2005). Gray & Higenbottam (1982) reported widespread placement of patients in pyjamas despite their being ambulatory and not acutely ill, in order to facilitate physical examinations and manage rule infractions. They recommended that the efficacy of this practice requires demonstration in light of associated dehumanisation, demoralisation and negative therapeutic relationships. Such work has not yet appeared in the literature on the subject.
This study aims to identify the prevalence of placing patients in night attire, its clinical and demographic associations, and the attitudes of patients and nursing staff towards it.

Method
We used a triangulation design incorporating data from medical case notes, questionnaires completed by nursing staff and interviews with in-patients in an open acute psychiatric ward in a local general hospital in Ireland.
We reviewed medical and nursing notes relating to all admissions over a 6-month period, generating a sample of 185 patients. Data were collected on age, gender, diagnosis, legal status, length of admission, documentation relating to placement of the patient in night attire, whether this was requested in writing, and whether reasons were documented for making this request.
An anonymous questionnaire was posted to all 35 members of nursing staff working in the unit to ascertain their attitudes towards placement of patients in night clothes. Thirty-five in-patients who had made the transition from night attire to day clothes were also interviewed to assess their attitudes. The patient questionnaire confirmed that they were asked to hand in day clothes on admission, their reactions to this and their beliefs regarding the right to choose their clothes. (Copies of the questionnaire available from the authors.) Ethical approval for the study was granted by the Galway University Hospitals Clinical Research Ethics Committee. Data was analysed using SPSS version 12 for Windows, with w 2 analysis for clinical associations of nominal variables.

Data from admission notes
Of the 185 admissions, 100 were men (54%); 33 patients (17.8%) were admitted involuntarily and the mean age of the sample was 44.4 years (s.d.=14.7). The mean length of stay was 25.4 days (s.d.=34.7).
The ICD-10 diagnoses of all patients were recorded on discharge: 42.7% patients had moderate depression, 13% had manic bipolar disorder, 13% paranoid schizophrenia, 9.2% depressed schizoaffective disorder, 4.9% alcohol dependence, and other ICD^10 diagnoses represented in smaller proportions. Less than half of the patients in the study (42.7%) were referred to the unit from primary care, 18.9% were self-referred, 18.9% were referred by community psychiatrists, 7% by community mental health nurses, 6.5% by general hospital wards, 4.3% by accident and emergency, and 1.6% by police.
As many as 106 patients (57.3%) were placed in night attire at the time of admission, a process recommended by doctors in 59 case notes (55.7%), although only 3 case notes (2.8%) contained a documented reason for doing so. The mean proportion of overall admission time spent in night attire was 41% (s.d.=23).
The associations of being placed in night attire are shown in Table 1. There was a statistically significant association between being admitted involuntarily and being placed in night attire (w 2 =7.58, d.f.=1, P=0.006), but no significant association with gender or a diagnosis of psychotic v. non-psychotic illness.

Survey of nursing staff
Completed questionnaires were returned by 27 staff (a response rate of 77%), of which 81.5% were staff nurses and the rest nurse managers. The majority of respondents were women (n=19, 70.4%). The mean age was 35.7 years (s.d.=9.9) and the mean length of service was 14.3 years (s.d.=9.4).
The perceived reasons by nursing staff for placing patients in night clothes are outlined in Box 1; most of them thought that this would help reduce the risk of absconding or self-harm.
The majority of staff (77.8%) believed that night attire was effective in addressing their chosen reasons. However, 48.1% of nursing staff described the practice as detrimental through loss of independence, enforcement of the sick-role, loss of dignity, increased agitation, punitive associations, institutionalisation, reduced trust in therapeutic relationship and coercive practice. As many as 81.5% of nursing staff believed that choice of attire should be afforded to voluntary patients, whereas only 14.8% thought that involuntary patients should be allowed this choice too. Nursing staff commented that night attire was 'useful for determining a patient's willingness to remain in the unit,' and that it 'allowed time for staff to assess a newly referred patient and to evaluate whether the patient was at risk of absconding'.

Patient survey
Data were obtained from 35 patients, 20 of whom were male; mean age 38.7 years (s.d.=10.0). More than half of the interviewed patients (68.6%) reported feeling uncomfortable in night attire during the daytime. Reasons given included embarrassment and humiliation, lack of privacy, futility of the process, restricted ability to move around the ward, awkwardness, punitive associations and being easily identified as a psychiatric patient. A significant majority (80%) believed that they should retain the right to choose their clothes, 11.4% thought that they should not be given a choice and 8.6% remained unsure.
Patients commented that 'wearing pyjamas marks you out, makes you feel more depressed, and I just don't see the purpose, it wouldn't stop me leaving if I really wanted to,' 'I have no privacy in pyjamas, it is quite humiliating really,' although another patient felt that 'I don't mind it really, it helps me to rest'.

Discussion
This study demonstrates a high prevalence of the use of night attire in an in-patient unit, with medical documentation for this practice rarely provided. Absconding is a significant concern and is linked to risk of harm to self and others (Crammer, 1984;Sheppard, 1996). Patients abscond for several reasons, including feeling bored, frightened, trapped, confined and isolated from family/ friends (Bowers et al, 1999). Studies investigating efficacy of reduced 'restrictive' care practices and subsequent absconding rates report conflicting results (Cancro, 1968;Gudeman et al, 1985;Molnar et al, 1985), although a recent implementation of an anti-absconding package demonstrated reduced absconding rates (Bowers et al, 2003;Bowers et al, 2005). Elements of this package included a sign-in book for patients, supportive breaking of bad news, debriefing of patients regarding ward incidents, multidisciplinary review, identification of patients at high risk of absconding, targeted nursing time daily for those patients and facilitation of social contact. Placing patients in night attire is not documented as an effective part of any anti-absconding package, and may represent a form of conflict-generating preventive containment (Bowers, 2006). In-patient prevention of self-harm has traditionally involved maintenance of physical safety, although the use of no-harm contracts with patients and a one-to-one observation of a patient by a nurse have not had a significant impact here (Drew, 2001;Potter & Dawson, 2001). Data from an Irish qualitative descriptive study investigating the intervention and prevention strategies used by eight psychiatric nurses when working with nonsuicidal self-harming in-patients revealed that the practice of placing patients in their night clothes is followed in order to attempt to ensure safety, although the staff remained unsure of the rationale or efficacy associated with this intervention (O'Donovan, 2007). The National Institute for Health and Clinical Excellence guidelines recommend involvement of people who self-harm in all discussions and decision-making regarding their care (National Institute for Health and Clinical Excellence, 2004), and there is evidence that perceived respect for patient's autonomy is associated with self-reported improvement in mental health (Kjellin et al, 1997;O'Brien & Golding, 2003).
Our findings reveal conflicting views of nursing staff and patients regarding the effects of having to wear night clothes at daytime and the rights of patients to choose their daytime attire. The recently implemented Irish Mental Health Act 2001 states that 'due regard shall be given to the need to respect the right of the person to dignity, bodily integrity, privacy and autonomy' (Office of the Attorney General, 2001), implying that involuntary admission is not synonymous with complete surrender of rights. Furthermore, removal of day clothes from voluntary patients could be considered an insidious form of detention which breaches current mental health legislation.
Strengths of this study include the triangulation design which combined qualitative and quantitative data from a large complete sample of written case notes, a high response rate to questionnaires from nursing staff, and participation by a representative sample of patients. However, these findings may not generalise to closed wards, where different attitudes and usage of interventions may exist.
These findings demonstrate that night attire is used for a substantial proportion of in-patients despite the lack of evidence to support its usefulness as a therapeutic intervention to protect against absconding or self-harm. Its use may fuel conflict between staff members and patients given the apparently disparate attitudes towards its use by the two groups. Staff should consider alternative mechanisms in order to manage risk in a manner that would be more acceptable to patients and more respectful of their dignity and autonomy. We suggest that, when psychiatrists recommend that patients be placed in night attire, due attention should be given to the fact that some patients may perceive this as undermining their autonomy and the decision and reasons for it should be documented in the medical notes. To ascertain the prevalence of epilepsy and understand the differences in the comorbidities of non-epileptic and epileptic patients with learning disabilities. A simple comparative survey was undertaken between the two main groups of patients: nonepileptic and epileptic.

RESULTS
The prevalence of epilepsy in the study group was 30%. A total of 70% of patients with any type of challenging behaviour were in the non-epileptic group compared with 59% in the epileptic group. Depression was the most common diagnosis in both groups, being slightly more in the non-epileptic group.

CLINICAL IMPLICATIONS
Our study suggests that there is no association between epilepsy and the prevalence of challenging behaviour of psychiatric conditions within the learning disabilities population.
The reported prevalence of psychiatric illnesses among adults with learning disabilities varies widely between 10 and 39% (Deb et al, 2001). Estimated prevalence rates for specific conditions are: schizophrenia 3%, bipolar disorder 1.5%, depression 4%, obsessive-compulsive disorder 2.5%, dementia 20%, and autism 7% (Cooper & Bailey, 2001). The prevalence of epilepsy in learning disability is said to be 20-30%. Challenging behaviour has a prevalence of 10-15% and is more common in men than in women. It reaches a peak by 15-34 years of age and increases in severity relative to the level of learning disability. Problem behaviour tends to be long-lasting and more than one type of behaviour is usually found (Smiley, 2005).
A study by Turkistani (2004) looked at the prevalence of epilepsy in learning disabilities and its association with mental illness and behavioural problems. The study found no significant association between epilepsy and behavioural disturbances or mental illness and concluded that epilepsy does not necessarily increase the incidence of mental illness and/or behavioural disturbance. Furthermore, Chung & Cassidy (2001) concluded in their study that there was no significant difference in the degree of learning disabilities between epileptic and nonepileptic groups. In a cross-sectional analysis, Tyrer et al (2006) found no relationship between aggression and the presence of epilepsy or autism. Similarly, Kerr (2002) concluded that behavioural disturbance is independent of epilepsy or its management. On the other hand, Espie et al (2003) found that a third of people with epilepsy and intellectual disability met the criteria for possible psychiatric disorder, particularly affective/neurotic disorder; twice the comparison rates for intellectual disability alone.
The Community Learning Disabilities Team in Waltham Forest caters to a total population of about 218 000. The region has a multi-ethnic population with Afro-Caribbean groups forming at least 14%, and Asian groups forming 12.5% of the overall population (Office for National Statistics, 2003;London Borough of Waltham Forest, 2007). The main aim of the study was to ascertain the prevalence of epilepsy and the difference in morbidities between epileptic and non-epileptic individuals in contact with the learning disability service. We also postulated that there is no association between epilepsy and high prevalence of psychiatric morbidities and/or challenging behaviour in our learning disability population.

Method
At the time of the study there were 353 individuals with learning disabilities in contact with the service. A total of 177 case notes belonging to active clinical cases were included in the study. Information gathered from the case notes was collected in a pre-formatted data collection sheet, where it was placed in the following categories: gender, date of birth, ethnicity, residential status and diagnosis (including epilepsy, psychiatric diagnosis and challenging behaviour). These data were tabulated to