Bibliotherapy for mental health service users Part 1: a systematic review
- Staff Library, Salisbury NHS Foundation Trust
- Department of Information Studies, Aberystwyth University, Aberystwyth, Ceredigion, UK
Deborah Fanner, Staff Library, Salisbury NHS Foundation Trust, Salisbury, Wiltshire, SP2 8BJ, UK. E-mail: email@example.com
Aims and objectives: UK health policy advocates a patient-centred approach to patient care. Library services could serve the rehabilitation needs of mental health service users through bibliotherapy (the use of written, audio, or e-learning materials to provide therapeutic support). Part 1 of a two part article describes a systematic review of the evidence for the effectiveness of bibliotherapy in mental health services.
Methods: The systematic review of the literature used Cochrane guidelines, together with an overview of evaluations of bibliotherapy initiatives, and assessments of the needs of adult mental health service users for rehabilitation support.
Results: The evidence strongly suggests that library-based interventions and the provision of information could be beneficial for service users and economical for the health service in assisting treatment of a range of conditions. At present, public libraries in the UK are developing basic bibliotherapy services.
Conclusions: Librarians, including librarians working for the health service, might provide more sophisticated bibliotherapy services, but the evidence to guide delivery is limited.
There is a growing interest in the provision of library services to patients in the UK, although traditionally the libraries based in hospitals have served health service staff, not patients and the public directly. Mental illness has been estimated to affect at least one in six people at some time in their lives. The National Service Framework (NSF) for Mental Health1 quotes (p. 42) from WHO statistics:
‘Worldwide, mental illness accounts for about 1.4% of all deaths and 28% of years lived with disability. In 1990, five of the 10 leading causes of disability were psychiatric conditions: unipolar depression, alcohol misuse, bipolar affective disorder, schizophrenia and obsessive-compulsive disorder.... People with severe mental illness are also socially excluded, finding it difficult to sustain social and family networks, access education systems and obtain and sustain employment.’
The majority of people with mental disorders are now treated in primary care, with inpatient facilities provided for crisis periods and urgent cases. Services aim to enable the mentally ill to integrate into society wherever possible, and support is provided via Social Services and Community Mental Health Teams. The NSF notes the possible sources of information, and suggests that increasingly information support from health and social care services should be through electronic media. More information is not necessarily good for all mental health service users, and professionals suggest that some service users may be confused by the information supplied, with adverse effects on their compliance with a treatment regime.2 Nevertheless, the NSF states that the written care plan for service users on an enhanced Care Programme Approach should include arrangements to promote independence and sustain social contact, including therapeutic leisure activity, as well as education and training. Libraries could play a role in such support.
Bibliotherapy is the term used in professional literature, although the term ‘reading therapy’3 has been used in the UK, and there was a Reading Therapy subgroup of the Library Association's Medical Health and Welfare Libraries Group with a Reading Therapy newsletter published 1985–1993. At one extreme, bibliotherapy may be unstructured, an informal, but informed recommendation by a librarian or health professional to some reading materials that should have a therapeutic value for the individual. Bibliotherapy in practice is usually more structured than the MeSH definition (a form of supportive psychotherapy in which the patient is given carefully selected material to read). The client often follows a programme of activities set out in a self-help manual (or e-learning package), and there may be a schedule for professional contacts as well.
Methods for systematic review
The review aimed to answer the following questions: (i) is bibliotherapy effective in assisting treatment of mental health conditions (and if so, in which conditions, and how?); and (ii) how can the information needs of mental health service users be assessed? An overview of library provision for mental health service users and bibliotherapy provision provided additional background for the survey (discussed in Part 2).
An extensive literature search was performed using the following databases: PsycINFO; medline; lisa; embase; cinahl; bni; King's Fund; Cochrane Library; AMED. This was supported by hand searching of key psychiatric journals: British Journal of Psychiatry; American Journal of Psychiatry; Journal of Mental Health; Journal of Psychiatric and Mental Health Nursing; Mental Health Practice; Psychological Medicine.
Internet searches were conducted using Google Scholar, Webcrawler metasearch, and the websites of mental health organizations and professional bodies. A later search of the mental health specialist library of the National Library for Health helped to check whether any further guidelines or evidence existed.
Search terms used were patient or ‘service user’; needs; information; education; bibliotherapy; libraries and ‘mental health’ or psych*. No early date restrictions were used as a historical context was required. The latest date included for the review was 2005, and subsequently the literature collection was updated to include recent key documents, using DARE (database of abstracts of reviews of effects) and the Cochrane Library (to 2006). No formats were excluded, but the majority of resources found were journal articles. Citations for theses and dissertations were collected, and abstracts screened carefully to select those directly relevant to the review.
Adhering to the Cochrane Library evidence hierarchy, meta-analyses and systematic reviews are presented first, then RCTs (although few have been conducted). Case studies and original surveys were then examined. Articles discussing the issues without contributing research-based evidence were reviewed for relevant content.
Examination of translated abstracts revealed that a number of articles published in non-English languages were of relevance, and the abstracts reviewed to identify the key articles for translation.
Some of the meta-analyses covered more than one condition but several examined the effectiveness of bibliotherapy in treatment of one type of psychiatric problem. As the clinical literature contained the most rigorous evidence, and was usually organized by condition or problem, the following sections discuss the evidence for the effectiveness of bibliotherapy by condition, before describing the types of delivery for bibliotherapy and how planning for bibliotherapy services might proceed.
Bibliotherapy in unipolar depression
The majority of studies compared groups of patients receiving bibliotherapy and/or psychotherapy with a waiting list control group. Several meta-analyses were identified which integrated results from previous research articles meeting pre-set quality criteria. These studies addressed the question of the effectiveness of bibliotherapy both in general and for specific conditions. Other questions concerned:
- • the difference in outcomes between use of bibliotherapy alone and in combination with other therapies;
- • effect of illness severity on outcomes;
- • influence of self-rating of outcomes; and
- • post-therapy effects and duration of same.
Cuijpers4 studied articles by Brown & Lewinsohn,5 Schmidt & Miller6 (1983); Scogin, Hamblin & Beutler,7 Scogin, Jamison & Davis,8,9 Selmi et al.10 and Wollersheim & Wilson.11 Den Boer, Wiersma & van den Bosch12 included the same studies, but added articles by Al Kubaisy et al.13 Bright, Baker & Neimeyer,14 Landreville & Bissonette,15 McNamee et al.16 Rosner, Beutler & Daldrup,17 Schelver & Gutsch,18 Tyrer et al.19 and White.20
Cuijpers considered the studies to be of high quality, whilst den Boer et al. found the majority to be of moderate quality, with Tyrer et al.'s study achieving the highest rating. In contrast to Cuijpers, den Boer et al. found the studies by Selmi and Wollersheim & Wilson (and the additional study of Schelver & Gutsch) to be of lower quality, failing to ‘adequately report baseline data’ (p. 963).
Both meta-analyses used Meta 5.3 software for data analysis, employing a ‘random effects model’, but den Boer et al. were able to analyse relationships to a more detailed level than Cuijpers, due to the larger number of articles reviewed. When selecting articles they employed the Delphi criteria list to assess quality, whereas Cuijpers employed no such standard, and also stating, without evidence, that books which only contain information are unsuitable for patients and their families (p. 139). A later review by Nelson & Loomis21 supported den Boer et al.'s methodology and conclusions.
Both meta-analyses conclude that bibliotherapy is effective in treating depression. Cuijpers states that bibliotherapy is ‘no less effective than individual or group therapy’ for unipolar depression. Den Boer et al. found bibliotherapy to be ‘significantly more effective than placebos’ and possibly ‘as effective as professional treatment of relatively short duration.’ (p. 967). Both meta-analyses cite the small sample sizes used in the studies as a limitation. They also note the predominant use of volunteers, who may be more compliant with therapy from the outset, and this could have led to positive bias in the results. The emphasis in the Den Boer meta-analysis was on self-help and bibliotherapy groups, but the meta-analysis found little evidence to support any conclusions on the effectiveness of self-help groups.
Riordan & Wilson,22 in an earlier review, found similar difficulties with sample sizes and methodologies. The evidence of the reviewed studies supported the value of bibliotherapy, to the extent that the authors concluded: ‘The real question is not whether bibliotherapy is or is not effective ... but rather how and when it should be used as part of a treatment programme.’ (p. 507).
Marrs23 has conducted the most comprehensive meta-analysis of bibliotherapy studies to date. A literature search found over 400 articles, of which 276 were empirical studies. Marrs excluded irrelevant subjects, studies not meeting quality criteria and articles that were prohibitively expensive, reducing this to 79 useful studies. These were coded to provide data on individual problem types, e.g. depression. Combined results from all studies showed a ‘moderate degree of effectiveness’ (p. 413) for bibliotherapy, with no differences found between the effectiveness of bibliotherapy and the use of psychotherapies. The positive results for certain conditions were more marked, notably anxiety. Moderate effectiveness was found in treating depression but results for alcoholism were less positive (in contrast to Apodaca & Miller's later meta-analysis24 in next section). A possible explanation for this discrepancy is the use of volunteers as opposed to referred patients. Marrs does not provide statistics on the relationship between alcoholism studies and use of volunteers, but as has been shown in other studies25,26 bibliotherapy is not as effective when used by non-volunteer subjects.
Marrs cautions that the results do not reflect the effectiveness of bibliotherapy for patients with more severe clinical problems, as the majority of articles studied featured mild to moderate conditions. Comparison with papers used in den Boer et al.'s meta-analysis show that only seven studies of the 15 are also featured in the Marrs meta-analysis. It is unclear without examining the earlier stages of Marrs's literature search if the eight absent studies were excluded for quality issues, or because the literature search failed to locate them.
Later trials not included in the three meta-analyses include an RCT (n = 57) conducted by McKendree-Smith,27 comparing immediate cognitive bibliotherapy, immediate behavioural bibliotherapy and delayed treatment (control). The cognitive group generally improved most. Floyd et al.28 compared cognitive bibliotherapy and individual cognitive psychotherapy (with delayed treatment controls) for depression in older adults (n = 46). Individual psychotherapy was superior to bibliotherapy at post-treatment on one rating scale, but bibliotherapy patients continued to improve during a follow-up period post-treatment, so that there were no differences at 3 months follow-up. A later 2-year follow up29 of 23 patients showed that treatment gains were mostly sustained, although there were significantly more recurrences of depression among the bibliotherapy group.
Bibliotherapy for other specific conditions
Meta-analyses for specific conditions are presented first, followed by RCTs, and then other evidence.
Apodaca & Miller's meta-analysis24 comprised one set of studies which used volunteers: Abril,30 Cunningham et al.31 Guydish,32 Harris & Miller,33 Heather, Kissoon-Singh & Fenton,34 Heather, Whitton & Robertson,35 Miller & Taylor,36 Miller, Taylor & West,37 Miller, Gribskov & Mortell,38 Sanchez-Craig, Davila & Cooper,39 Savage, Hollin & Hayward,40 Sitharthan, Kavanagh & Sayer,41 Spivak, Sanchez-Craig & Davila.42 The other set used subjects identified via screening programmes for ‘at-risk’ individuals: Anderson & Scott,43 Chick, Lloyd & Crombie,44 Heather et al.45 Heather et al.46 Nilssen,47 Reynolds et al.48 Richmond et al.49 and WHO.50 The meta-analysis by Apodaca and Miller updates the Marrs review,23 which found less positive indications for bibliotherapy.
Each study was assessed using 12 methodological criteria. Eleven of the 12 studies using volunteers found evidence that bibliotherapy aided the reduction of alcohol intake in participants, and in most cases this was a significant reduction. Follow-up studies after 8 years showed that these improvements were maintained. In contrast, Guydish32 found no evidence that any of the three interventions used in his study was effective in reducing alcohol consumption. However, his work used university students rather than alcoholics seeking psychiatric help. Guydish's article scored 2 out of 17, the lowest mark on Apodaca and Miller's methodological quality table. It is therefore reasonable to conclude that the study is of insufficient quality and relevance.
Results from the screening programme group showed more variable outcomes. The majority of studies showed some level of reduction in drinking with participants using bibliotherapy; in cases where drinking was not reduced, other problems were alleviated.44 One notable exception is the WHO study, which found that positive effects of bibliotherapy did not extend to female participants, a finding unsupported by other research. Although one study by Heather et al.45 did not find any difference in reduction levels between three study groups following different therapy regimes, the bibliotherapy element of this research was affected by the low level of compliance with the bibliotherapy regime.
There was no meta-analysis available (Marrs23 had used broader categories of problem treated, Evans et al.26 describes an RCT of bibliotherapy interventions for treatment of deliberate self-harm. Participants were assigned to a group using materials prepared for MACT therapeutic reading, or a control group continuing with their existing treatment. Results showed that the MACT group suffered less incidents of self-harm than the control group. The MACT group also showed a ‘significantly greater reduction in depressive symptoms’ (p. 23). The authors also reported that the average cost of care for MACT participants was 46% lower than usual treatment costs.
As indicated above, the Marrs23 meta-analysis included broad categories (such as anxiety) and therefore individual RCTs need to be examined for panic disorder itself. Gould, Clum & Shapiro51 studied the use of bibliotherapy in an RCT of the treatment of panic disorder in agoraphobics. Results showed bibliotherapy to be effective, with significantly more improvement on four out of six measures of pre-treatment behaviours.
Lidren et al.52 also found evidence of the effectiveness of bibliotherapy in panic disorder, with experimental participants using bibliotherapy showing a significant improvement compared to wait list subjects (control arm). The results were more pronounced than in Gould, Clum and Shapiro's study. This is attributed to the treatment period being 8 weeks, twice as long as Gould et al. allocated.
Wright et al.53 studied the effectiveness of using bibliotherapy in a relapse prevention program for panic attacks. Results in this RCT showed significant reductions in symptoms for the group receiving therapy compared to the wait list control group. A reduction in gains over the second stage of the three-part program was attributed to a lack of telephone contact compared to the first and third stages. The authors recommend further study to confirm this theory.
Febbraro54 compared the use of bibliotherapy alone, bibliotherapy with phone contact, and phone contact alone in an RCT. A significant difference was found between the phone-contact-alone group and the phone-contact-plus-bibliotherapy group, with the latter experiencing a considerable reduction in their fear of having a panic attack. The authors conclude that bibliotherapy can form an effective adjunct to psychotherapies.
In contrast, the RCT by Sharp, Power & Swanson55 compared cognitive behaviour therapy (CBT) interventions and bibliotherapy and found that the latter is the least effective co-therapy for panic disorder. Their conclusions are supported by Persons & Valus56 in their analysis of the research, and the NICE guideline published in 2004 on the management of anxiety57 concurs, with self-help bibliotherapy (based on CBT priniciples) rated the least effective intervention for panic disorder and generalized anxiety disorder.
Van Lankveld58 included 12 studies (including 9 RCTs) in the meta-analysis. The review found that bibliotherapy was an effective treatment for orgasmic disorders, which were the subject of 87% of the studies. It was not possible to reach any conclusions as to the efficacy of bibliotherapy for other sexual dysfunctions as too little original research was available.
The Marrs23 meta-analysis examined adults working on their own problems, and one of the broad categories used was impulse control.
Monti et al.25 studied the effect of bibliotherapy on improving social skills. Results differ from those reported for other disorders, finding that participants in a skills training group showed significant improvement over the bibliotherapy group. The authors conclude that bibliotherapy ‘is not effective in teaching social skills to psychiatric patients’ (p. 191). It should be noted that unlike the majority of studies, subjects were not volunteers. As already noted, bibliotherapy may be less effective when participants are not volunteers. The validity of Monti et al.'s conclusions has yet to be supported by other research.
Delivering bibliotherapy interventions
For the e-learning type of bibliotherapy, a relevant systematic review is that of Kaltenthaler, Parry & Beverly,59 a systematic review of computerized CBT. Sixteen studies were identified, including 11 RCTs of packages such as ‘Beating the Blues’. A meta-analysis proved difficult due to the wide variation of methodologies used and results achieved, so no consensus on the effectiveness or otherwise of this form of therapy could be reached. The research quality of the studies was from poor to moderate, and the authors argue that there needs to be more independent research, as program designers currently undertake most studies. An example of computerized therapy is Kenwright & Marks's study60 into the use of a computer program for the treatment of phobia and panic attack. Preliminary findings showed symptoms were reduced in participants. However, the study should be viewed with caution: the sample size was small, and one of the authors owns intellectual property rights in the program being studied.
A review of the use of self-help books for depression examined studies (11 RCTs) that compared self-help books, with minimum guidance with treatment as usual or a waiting list comparison.61 The review identified one book that was associated with a significant improvement on the Hamilton Rating Scale for Depression at 4 weeks.
Many survey and ‘case study’ report articles have been published by practitioners and clinical advocates of bibliotherapy recommending its use and providing examples of best practice, for example, Adams & Pitre,62 Anstett & Poole,63 Campbell & Smith,64 Cohen,65,66 Frude,67–69 Goldstein,70 Gottschalk,71 Gusich,72 Katz & Watt,73 Lanza,74 and Myers.75 Floyd76 reviewed bibliotherapy (reading a self-help book) as an adjunct to psychotherapy, McKendree-Smith, Floyd & Scogin77 summarized the use of bibliotherapy in depression, and indicated future developments in computer-based treatments. To date there have been insufficient studies on the effectiveness of bibliotherapy in treating clinically severe conditions. The majority of authors recommend further research before definitive conclusions can be made. There is also little research in less formal bibliotherapy in the use of fiction, drama and poetry, despite papers having been published recommending their use.78–80
Assessing information and education needs of mental health service users
Evidence-based studies of user needs commonly employ a questionnaire requiring participants to rate a list of possible needs in order of importance. Researchers aim to answer questions concerning the:
- • range and quality of information provided to patients;
- • needs of patients with different illnesses;
- • needs of carers; and
- • attitude and understanding of clinical staff.
Examination of these studies shows a consistent requirement for educational and leisure facilities—and provision of services would need to consider the specialist requirements of patients and carers.
A survey by Chien, Kam & Fung-Kam Lee81 in the educational needs of mental health patients found that among the most important were information on illness, medications and side-effects; problem-solving strategies; and understanding the mental health system (p. 308).
Perreault et al.82 found that psychiatric outpatients considered information on illness, medication and treatment to be important. Mueser et al.83 reported similar findings, and Tempier84 found that more than 50% of patients surveyed wanted more information. Cleary et al.85 found patients and their carers wanted more information on illness, medication, side-effects and community resources.
Different results were obtained by MacGabhann in a study86 comparing needs of service users and nurse assessment of needs, using focus groups and semi-structured interviews. Nurses believed patient information should focus on illness, psycho-education and medication, whereas patients required more information on their situation and the operation of the hospital (p. 86). Nurses also claimed to provide the information they described, however, patients did not express this as a need (p. 87). One explanation for this unusual result could be the research location, which focused on three wards of the same hospital, suggesting the possibility that the environment could have affected patients’ responses.
The problem of institutional boredom is reported in articles by Riggall,87 Ward,88 and Quirk & Lelliott.89 Armitage90 conducted a 12 year study into the needs of a group of long-term mental patients. He observed at the start of the project that ‘residents were restless and disturbed. Many spent their time pacing up and down the corridor.’ (p. 28). As part of the process of rehabilitation, residents were helped to learn about budgeting, cooking and homemaking, how to seek help, and how to enjoy leisure activities.
Baker91 describes a similar theme in the development of the ‘Tidal Model’ of nursing care. This model places the needs of individual service users at the heart of the care planning process, prioritizing the need for understanding, emotional security and the role of education and leisure in rehabilitation.
A study by Bird92 into advice services for the mentally ill reported that 92% of interviewees needed more information, particularly regarding medication, rights and benefits. Interviews revealed that service users saw it as a ‘... significant advantage to be able to get help with more or less everything in one place, where the individual was known to the staff and did not have to explain themselves or their problems afresh.’ (p. 136). Bird emphasized the need for providers to ‘... recognize the importance of ensuring access to independent advice and information, and the availability of such services in acute hospital and community and day care settings.’ (p. 139).
A study by Hilton, Ghaznavi & Zuberi93 examined the recording of patients’ religion and facilities for following faiths in mental health acute wards. They observed that there was little access to any non-Christian texts, and that copies of the Christian Bible were frequently out of patients’ reach, or left by the beds of patients from other faiths. For the past decade, hospitals have been required to record the ethnicity of all persons admitted. Despite this, a study published in 2000 by Warner et al.94 found that only a third of NHS Trusts surveyed used this information to improve services.
Pollock et al.2 conducted a focus group study examining the information needs of psychiatric patients. Provision of information was found to be poor, with patients and carers forced to seek alternative sources which were often of unreliable quality. Participants provided a comprehensive list of reasons for believing information to be a vital part of care planning (p. 392). The authors conclude that ‘... adequately meeting patient information needs is an important step towards greater involvement of users in their treatment and care.’ (p. 399).
McCann (writing in Campling, Davies & Farquharson,95 as a service user), identified accurate information, preservation of dignity, holistic nursing care, advocacy and advice, involvement in care planning, access to creative therapies, opportunities for occupation and recreation, and help to reintegrate into society as key requirements for inpatients (p. 160). Additionally, Slade et al.96 provide ‘... the first longitudinal evidence that high levels of patient-rated unmet need may actually cause low levels of quality of life.’ (p. 260).
Assessing the evidence for bibliotherapy: conclusions
The literature review shows that the value of using bibliotherapy and associated interventions in the treatment of mental illness is supported by the evidence. Effectiveness is significantly increased when bibliotherapy is used in conjunction with psychotherapies, suggesting that treatment of in-patients could be beneficial, although further studies are required.97 The difficulty is assessing the contribution of bibliotherapy itself. Guirguis-Younger,98 for example, demonstrated that in a complex, but typical care situation of older people with functional impairments leading to depression, a community based intervention programme including bibliotherapy was successful. The programme was supported by community nursing staves who were not psychiatric nursing specialists, but the implication is that bibliotherapy is part of the treatment and requires some professional support. An earlier review3 suggested that reading therapy can be considered as an adjunct to library services, as such therapy seemed effective. Ten years on, the conclusions are little different, but perhaps librarians need to work more closely with the health professionals to devise suitable therapeutic programmes that include bibliotherapy.
Research has been hampered by the inability of researchers to mask the nature of the intervention to trial participants, due to the interactive nature of bibliotherapy, and factors such as the interaction of bibliotherapy with support from nursing contacts, for example. The sampling approach may lead to bias if the participants have already opted for a self-help treatment approach. Studies tend to use small sample sizes, meaning that testing to identify an effect as statistically significant is more difficult. Another difficulty is the duration of the treatment period, and assessing the long-term efficacy of bibliotherapy, or indeed, what type of bibliotherapy works best—a custom made treatment manual or a published self-help book, or a computerized training programme. There are insufficient studies into effectiveness of bibliotherapy for conditions such as self-harm. There is also a lack of evidence relating to more serious mental illness. A common shortcoming is a lack of socio-economic and ethnicity data, without which their significance cannot be assessed.
The need for accessible, quality information for service users and carers is strongly supported by the evidence. The requirement for recreational materials has been a neglected area of research and requires further study, but anecdotal evidence suggests that provision of fiction and poetry would be beneficial. The question, perhaps, is the type of involvement and professional expertise required by the library staff. Perhaps some lessons may be learned from the project ‘There be Monsters’,99 a partnership between the National Archives, Central and NW London Mental Health Trust, and Workshop & Company (supported through Heritage Lottery Funding). The project aimed to interest adults with mental health problems with the devices and designs on the old maps in the National Archives, to inspire them in the creation of a sculpture now in the grounds of the National Archives. This was a successful project—but required considerable commitment from all the partners. The project included an evaluation but the type of impact-measures that are appropriate for such a project are not clear.
Library services for mental health patients—a historical background
There is little information available concerning library services for patients in UK facilities before the 20th century. Existing records reveal that books were viewed as an entertainment and used as a method of diverting patients’ minds away from disturbing thoughts. There were no structured programmes of therapy or acknowledgements of the therapeutic value of reading.100 In the USA, there is documentation. Expansion of facilities for the mentally ill began in 1810, before which there were few such institutions.101 Libraries were a common feature of new asylums, their purpose being a combination of patient therapy and promotion of the hospital to relatives and benefactors. The boredom of life in an asylum was recognized as detrimental to patient recovery, and books were seen as an ideal way to occupy the minds of patients, one of whom is quoted as saying ‘We read when we can do little else, and alas, what would a residence in the asylum be without the occupation of reading!’ (Older102 p. 519). Libraries were commonly funded by donations, with no stock selection other than the approval of donated materials.
Eminent physicians and superintendents such as Benjamin Rush (author of the first key psychiatric text in America) promoted the benefits of library services to patients. Samuel Woodward set aside part of his office to house a library for patients. William Rockwell helped his patients to produce their own newsletter. John Galt was instrumental in the development of bibliotherapy, and published the first essay devoted to the provision of libraries to the mentally ill. Charles Evans recommended the use of reading, encouraging patients to have free access to scientific books and materials. James MacDonald promoted the circulation of periodicals at Bloomingdale asylum and organized the creation of a reading room.102
One consequence of the First World War was a dramatic increase in rates of mental illness. A key aspect of recuperation was the service provided by American Library Association (ALA) librarians. Drake103 detailed her experiences at St. Elizabeth's Hospital for the Insane in Washington, DC, describing cases where the library service helped patients in their recovery. She supervised the creation of a reading room, organized a book trolley to visit the incapacitated, and encouraged patients to undertake voluntary work in the library. One inmate wrote a letter of thanks, noting: ‘... you are doing a very great favour to many unfortunates, and it is sincerely appreciated.’ (p. 138).
Librarians in the USA were responsible for much of the pioneering work carried out in the late-19th and early-20th centuries. Most notable were E. Kathleen Jones of Massachusetts and Miriam Carey of Minnesota. The needs of the Second World War and the post-war expansion of medical education precipitated the development of library services for clinicians. This growth coincided with a series of upheavals amongst the ALA committees responsible for patient libraries, resulting in their gradual decline.104
Bibliotherapy in libraries
Interest in bibliotherapy was at its height between the Wars, and librarians wrote many papers during this period. Schneck's 1945 literature review105 found familiar themes running through the majority of papers: (i) Bibliotherapy was generally found to be an efficacious treatment; (ii) Patients should be assessed as individuals with unique needs; (iii) Material relating to suicidal acts or thoughts should not be permitted in the library; (iv) Fiction was popular, but there were few subjects which did not interest patients; and (v) Librarians frequently employed patients to assist in the library.
Acknowledged as a pioneer bibliotherapist, Sadie Peterson Delaney worked from 1924 to 1958 as chief librarian of the US Veterans Administration Hospital in Tuskegee, Alabama.106 Here she managed both patient and staff libraries, working with clinicians to understand the needs of individual patients. Delaney described the library as ‘... a laboratory and workshop for the improvement and development of the whole individual.’ (Gubert107 p. 129). In 1940 she wrote of the positive feedback she received from patients expressing their appreciation of the library service.
Over the same period of time, a lack of scientific evidence had caused some misgivings regarding the validity of bibliotherapy among librarians and managers, with several critics observing that scientific measurement of effect was virtually impossible given the nature of the intervention.107–109 Bibliotherapy was viewed as a ‘... hit and miss procedure from a scientific point of view.’ (Schneck106 p. 224), but recognized that it had achieved good results in many patients. Schneck noted that few psychiatrists published papers on bibliotherapy, with the result that other clinicians were not inspired to research the subject.
A new enthusiasm for bibliotherapy began in the late-1960s. Again, the work of librarians (notably Mona Going) was instrumental in arousing interest. Professional bibliotherapist Arleen McCarty Hynes published several papers on the value of bibliotherapy, and advocated formal training for practitioners to ensure optimum effectiveness.110 Anderson111 described the use of poetry at St Elizabeth's Hospital from 1965, commenting that ‘The flexibility of poetry therapy makes it highly advantageous as a treatment modality in psychiatric nursing.’ (p. 133). She also commented on the interest patients had in the lives of poets, and how reassured many were that writers often suffered from mental illnesses. In contrast to earlier practices Anderson avoided censorship of suicide-related materials, as patients ‘... strongly disagreed that their readings should be censored ....’ (p. 136).
Wenger112 provides an overview of the evidence-base for bibliotherapy, pointing out the anecdotal nature of most librarian-authored papers in comparison with the scientific approach of clinicians. She recommends the development of bibliotherapy, but refers to it as a ‘... difficult and potentially dangerous practice ...’ (1980, p. 137) This view is provided without explanation or evidence, and thus cannot be assessed for its accuracy.
Armstrong113 observes that bibliotherapy ‘... has been known to result in some cases in harm ...’ (p. 224). Like Wenger, she fails to record specific examples of this. She also discusses the problem of censorship, pointing out legal aspects to be considered such as the civil rights of patients (p. 228). Armstrong also notes that promotion of bibliotherapy is hampered by the disparate organizations and societies involved, and the lack of a structured national (USA) training programme.
Modern library services for the mentally Ill
Services in the UK
The therapeutic value of reading was noted in some accounts published in the 19th century.3
Residents at Springfield Hospital described their days as ‘long and empty’ (Dowton114). A library was introduced, providing fiction, non-fiction and talking books, run by a librarian who emphasized the importance of catering for the specialist needs of ethnic minorities and the physically handicapped. Users frequently commented that they would not consider borrowing from a public library because of the stigma attached to mental illness. The library service ran events and workshops for patients, often involving staff from the public library and museum. Patients also helped operate the library as part of the health authority's work experience project.
The public library service provided the patients’ library at Prestwich Mental Health Unit, under contract to the Health Authority (Kelly115). Ward rounds were made each week to enable confined patients to participate, including those in the Regional Secure Unit. Videos and audiotapes were particularly popular on the wards. Kelly reported that clinicians were ‘... quite liberal in terms of the material they (would) allow inmates to borrow.’ (p. 30), only censoring overtly sexual content.
The public library service in Hertfordshire funded patient and staff libraries at Napsbury Hospital, Barnet, where a resettlement project assisted former service users with their re-integration into the community.116 Discharged service users were encouraged to use the public library service, assisted by hospital library staff. Another initiative was an outreach scheme whereby care assistants were given library materials for housebound service users.
A scheme that has been running successfully in Kirklees since 2002 is RAYS (Reading and You Scheme).117,118 The public library service funds bibliotherapists to provide group and individual therapy sessions with clients referred by health and social care staff.
Public library and archives services have funded several regional projects aimed at providing therapeutic services for disadvantaged residents in the community, including the mentally ill. These include the Northern Ireland Reminiscence Network119 and a similar, earlier archive therapy project in Carmarthenshire.120
Several articles in the German library literature refer to the use of bibliotherapy in psychotherapy (and the role of librarians121), and an integrated approach to library provision and reading groups.122 One identified trial123 discusses the choice of literature for the reading groups, and the therapeutic aims, before comparing the reactions of the patient group with a control group (without diagnosed mental health problems). Different reactions were observed between the patient and control groups concerning aspects of the books discussed. The focus of the discussions often differed as well.
Recently, many public library services have provided a ‘Books on Prescription’ scheme,68–70 partnering with health and social services in several authorities such as Gloucestershire124 and in Wales, through NHS Direct Wales.125 These schemes are aimed at clients in the community, who are not receiving inpatient care, and usually include a range of ‘self-help’ materials. The Devon self-help book prescription scheme is being evaluated to investigate whether client support has an influence (as suggested in other studies, e.g.99), and whether physical activity should be combined with the use of a self-help book.126
Services in the USA
The USA provides more library services for patients than the UK. Some are run in conjunction with staff libraries, while many are run by patients and/or volunteers.127 The role of the library in the care of the mentally ill continues to be well-recognized, and librarians often take part in programmes of rehabilitation and bibliotherapy. Finance continues to be the key barrier to investment; there are no requirements for hospitals to have libraries, therefore services may be targeted when costs need to be reduced.
Russell128 described the advanced, integrated programme of library participation in rehabilitation at the McLean Institute in Massachusetts. Patients were assessed on arrival at the hospital and suitable therapeutic activities recommended. Information was considered key to the process, to ensure patients had an understanding of their illness. In E. K. Jones's time many subjects were censored, such as horror, erotica, and medical or reference books, but it has been made illegal to censor the reading material of patients in many American States. Instead, a system of classification was used whereby contentious materials were kept together under supervision.
Evans129 reported on the success of a project providing computer access to patients at the Florida Mental Health Institute. Patients enjoyed working socially in interactive groups, and evaluation of the project showed a 52% increase in library attendance, with patients responding very positively to the experience. The author concludes by saying that ‘... in the hands of qualified mental health practitioners, computers may have positive social applications in therapy.’ (p. 571). However, the report is too brief to be evaluated comprehensively, and no information is provided on research methodology.
McOuat's summary130 of the benefits of library services emphasizes how important it is ‘... for librarians working with patients to ensure that the library is perceived as a viable source of help in the total process of rehabilitation.’
A different perspective is provided by Carparelli,131 in a piece calling for public library administrators to address the issue of ‘problem patrons.’ The author provides examples of problems experienced by staff, which she attributes to the use of the library as ‘... a shelter for the homeless and a meeting place for the disturbed.’ (p. 212). It is unfortunate that this article is written as an opinion piece rather than as an attempt to provide an evidence-based assessment of risk. The failure of the author to comprehend the nature of psychiatric illness is evident in her use of terms such as ‘real’ patrons. It is difficult to attribute value to this article, which seems to be a biased diatribe resulting from the author's own experience of being assaulted at work.
Lyon (in Johnson128) and Date132 recommend that public library staff should receive training on mental illness in order to reduce the preconceptions held by staff such as Carparelli. Mulvey133 observes that ‘The absolute risk of violence among the mentally ill as a group is very small, and only a small proportion of the violence in our society can be attributed to persons who are mentally ill.’ In England and Wales, on average, 55 people are killed by persons who are judged to be mentally ill at the time of the homicide—compared to 3500 killed on the roads. Of those 55, only five were strangers.134 On the other hand, an audit of violence in mental health and learning disability units135 in England and Wales revealed that 41% of clinical staff had experienced violent or threatening behaviour from service users—the main trigger being substance misuse, although frustration and boredom were also involved.
Value of library services to mental health service users: conclusions
The historical overview suggests that the value of reading and education to mental health service users has long been assumed, although most of the reports are descriptive or anecdotal rather than evidence-based. Moreover, the treatment and understanding of mental illness has progressed considerably since the 19th century, when there were no psychopharmacological or psychoanalytical treatments available. Nevertheless, the considerable amount of historical material and the successful examples of modern services suggest that library services could play a more active part in therapy. The difficulty, perhaps, is the strategic planning involved. Mental health service users may be inpatients, but with the closure of large institutions many of those inpatients are now cared for in the community, in specialist Units or in sheltered accommodation. Some patients with mental health problems will not need referral to a specialist mental health service—and some may simply adopt ‘self-help’ therapies, with or without professional advice and support throughout.
The review of the literature on library services to mental health service users indicates that bibliotherapy may have its roots in library services to assist users in feeling better in themselves through reading that could be viewed as leisure reading, but which also contributes to therapy, perhaps informally more than formally. Reading therapy for children and adolescents appears to help children to develop and cope with problems,3 and librarians have produced guidance for such services.136 For adults, public library services are developing a role in ‘books on prescription’ services for those with minor mental health problems. Self-help books are a popular genre in many public libraries, and a survey137 of lending in Devon libraries indicated that more than half the respondents found the health-related books borrowed had motivated or inspired them, and over a quarter agreed that the book had helped to change their daily life. The current role of libraries sited within mental health services in assisting service users with therapy seems less clear. Several of the studies point to the need to train library staff to deal with the unusual or disturbing behaviour exhibited by some mental health service users, but there are few studies offering or testing new ideas on service development. Part 2 of the article examines attitudes of psychiatric libraries in the UK towards service development and service users.
The authors thank the anonymous referees for their constructive comments. The authors also thank the various library services (NHS libraries, and the Thomas Parry library, Aberystwyth University) for their assistance in obtaining the articles for the studies reviewed.
Implications for Policy
- • Library-based, basic bibliotherapy appears cost-effective in supporting treatment of a wide range of mental health conditions.
- • The evidence to guide development of more sophisticated bibliotherapy services is very limited.
- • Trials of more sophisticated bibliotherapy services should involve librarians who work with mental health service users and librarians who work with mental health professionals.
- • More sophisticated services require more advanced training for librarians.
Implications for Practice
- • Bibliotherapy and reading therapy cover a wide range of services from books on prescription services to supportive psychotherapy with carefully selected learning materials, including e-learning packages.
- • Basic bibliotherapy services such as books on prescription services are relatively easy to set up and manage.
- • Co-operation between NHS and public libraries should help in development and evaluation of bibliotherapy services.
Format AvailableFull text: HTML | PDF
© 2008 The authors. Journal compilation © 2008 Health Libraries Group
- Issue online:
- Version of record online:
- Received 18 November 2007; accepted 19 August 2008
- 1Department of Health (England). A National Service Framework for Mental Health. London: Department of Health, 1999.
- 2Pollock, K., Grime, J., Baker, E. & Mantala, K.Meeting the information needs of psychiatric inpatients: staff and patient perspectives. Journal of Mental Health2004, 13, 389–401.
- 3Forrest, M.Recent developments in reading therapy: a review of the literature. Health Libraries Review1998, 15, 157–64.
- 4Cuijpers, P.Bibliotherapy in unipolar depression: a meta-analysis. Journal of Behavior Therapy and Experimental Psychiatry1997, 28, 139–47.
- 5Brown, R. & Lewinsohn, P.A psychoeducational approach to the treatment of depression: comparison of group, individual and minimal contact procedures.
The Carnegie-Whitney Grant is awarded by the American Library Association and is based on a fund first established by Andrew Carnegie in 1902, “to be applied to the preparation and publication of such reading lists, indexes, and other bibliographical and library aids as will be especially useful in the circulating libraries of this country.” In 1910, the fund was expanded through a merger with another fund established by James Lyman Whitney.
It was another Carnegie grant, awarded in 1938 by the Carnegie Corporation, that funded the Center of Alcohol Studies' own E. M. Jellinek and Mark Keller in their work reviewing the literature on the effects of alcohol on the individual. The results of this work would eventually be published as the Classified Abstract Archive of the Alcohol Literature (CAAAL). While the CAAAL abstracted scholarly and scientific literature relating to alcohol, Reading for Recovery aims to draw attention to popular works that deal with issues relating to alcohol and substance abuse, be they fiction or non-fiction. In a way, however, we see the projects as two sides of the same coin, with both dedicated to increasing the visibility and use of literature relating to addiction. We are thrilled to be following in the footsteps of Jellinek and Keller by creating our own collection of addiction literature, and we are grateful that a Carnegie grant has once again made this possible.