Article Text

Get into Reading as an intervention for common mental health problems: exploring catalysts for change
  1. Christopher Dowrick1,
  2. Josie Billington1,
  3. Jude Robinson2,
  4. Andrew Hamer3,
  5. Clare Williams4
  1. 1Department of Mental Health and Behavioural Sciences, University of Liverpool, Liverpool, UK
  2. 2Department of Sociology, University of Liverpool, Liverpool, UK
  3. 3School of English, University of Liverpool, Liverpool, UK
  4. 4The Reader Organisation, Liverpool, UK
  1. Correspondence to Professor Christopher Dowrick, Department of Mental Health and Behavioural Sciences, 1st Floor Block B Waterhouse Buildings, University of Liverpool, Liverpool L69 3GL, UK; cfd{at}liv.ac.uk

Abstract

There is increasing evidence for the efficacy of non-medical strategies to improve mental health and well-being. Get into Reading is a shared reading intervention which has demonstrable acceptability and feasibility. This paper explores potential catalysts for change resulting from Get into Reading. Two weekly reading groups ran for 12 months, in a GP surgery and a mental health drop-in centre, for people with a GP diagnosis of depression and a validated severity measure. Data collection included quantitative measures at the outset and end of the study, digital recording of sessions, observation and reflective diaries. Qualitative data were analysed thematically and critically compared with digital recordings. The evidence suggested a reduction in depressive symptoms for Get into Reading group participants. Three potential catalysts for change were identified: literary form and content, including the balance between prose and poetry; group facilitation, including social awareness and communicative skills; and group processes, including reflective and syntactic mirroring. This study has generated hypotheses about potential change processes of Get into Reading groups. Evidence of clinical efficacy was limited by small sample size, participant attrition and lack of controls. The focus on depression limited the generalisability of findings to other clinical groups or in non-clinical settings. Further research is needed, including assessment of the social and economic impact and substantial trials of the clinical effectiveness and cost-effectiveness of this intervention.

  • Shared reading
  • depression
  • intervention
  • catalysts for change
  • qualitative
  • mental healthcare
  • literature
  • primary care
  • social science

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Introduction

Population mental health and well-being are high on the international health agenda given the prevalence of depression as a major disabling illness.1 2 Although anti-depressant medication remains the mainstay of treatment in primary care, its effectiveness has been called into question.3 4 There is an expanding evidence base in support of a range of treatment options including psychosocial interventions and comprehensive disease management programmes.5–7 Within these we note increasing interest in narrative8 and bibliotherapeutic9 approaches, which typically emphasise the importance of meaningful social engagement; a sense of value, purpose or comprehensibility in respect of one's self and life; a sense of agency and of manageability in relation to the problems and demands posed by life; and the capacity to “tell a good story about oneself”.10–12 As the last possibility is likely to decrease in the culturally adverse setting of an in-patient ward, or in the context of a diagnosis which offers a passive story of a ‘patient’ who is ill and in need of professional cure, recent initiatives have stressed the importance of preventive interventions which can reach individuals before such adverse personal stories take hold.

Get into Reading

The principal feature of the Get into Reading (GiR) model is shared reading of serious literature within a group setting. ‘Serious’ is used in preference to ‘classic’ which risks confusion with ‘classical’ (ie, Latin/Greek) literature. The term has a long history—The Oxford Professorship of Poetry was established for “the advancement of more serious literature both sacred and human”13—and is widely used by writers and literary critics to distinguish literature that addresses important human issues, such as those concerning social and moral behaviour, from, for example, ephemeral writings (comics, popular magazines) and genre fiction (eg, detective novels, ‘chick lit’).

All material is read aloud in the session itself and open-ended discussion is encouraged by the facilitator. Group members participate voluntarily as they wish and interact in relation to what is happening in the text itself (in terms of narrative, characters, place and setting, themes, description and language, for example) and what may be happening within themselves as individuals (in terms of reflections about personal feelings and thoughts, opinions and experiences, for example) as an articulated and evolved response to the shared reading of the text and wider group discussion.

GiR sessions last for 90 min, and their basic structure is summarised below:

  • Break-In/Re-cap Period (10 min)

  • Prose Reading and Discussion (50–60 min): the facilitator always begins the session by reading aloud from a short story or novel. Group members are free to interrupt during the facilitator's reading if they want to raise questions or express opinions, but if not interrupted the facilitator will usually pause after four pages (approx 8–10-min reading time) of the short story or novel at an appropriate point to encourage discussion and allow time for group reflection. In these pauses, the discussion can range widely, usually starting with issues, characters or situations contained in the material just read and often progressing to personal reflection and the sharing of opinions and experiences. After a time the facilitator will bring the discussion back to the text and ask if anyone would like to take a turn reading. A group member might volunteer and the same pattern will be repeated—the group member pausing or handing over the reading aloud or the facilitator intervening to allow discussion or another group member to take a turn.

  • Poetry Reading and Discussion (20–30 min): the session always concludes with a reading of a poem. The poem is often selected to reflect or develop themes that might have been read about and discussed during the session in response to the prose material. The poem is intended to restore a sense of balance to the mood of a group, which can prove particularly helpful in circumstances where the group has been reading a difficult episode in the story. The poem also provides an opportunity for members who may not have read during the prose section of the session to take a turn at reading aloud. After several readings and discussions, the facilitator will usually ask for one final reading of the poem to conclude the session and to illustrate how far the group members may have come in their understanding of the piece.

  • End of session: group members leave the designated reading area.

Previous research into GiR

The findings from earlier research are that GiR is a feasible and acceptable intervention for a variety of health conditions within both primary care and specialist settings.14 General practitioners (GPs) are supportive of GiR and its benefits, in particular for patients who frequently consult on poor mental health linked to social isolation, and for whom GPs often feel unable to offer conventional medical treatments. Such patients are not just poor in the sense of materially deprived, they are also ‘word poor’ and may lack the resources to be able to communicate their day to day concerns to their GPs.15

Prior to this research, observed and reported outcomes for participants included being ‘taken out of themselves’ via the stimulation of the book or poem; feeling ‘good’, ‘better’ or ‘more positive about things’ after taking part in the group; valuing an opportunity and space to reflect on life experience, via memories or emotions evoked by the story or poem, in a convivial and supportive environment; improved powers of concentration; a sense of common purpose and of a shared ‘journey’; increased confidence and self-esteem; a sense of pride and achievement; valued regular social contact; and improved communication skills.15

These preliminary findings resonated with other innovative research into reading and health, which suggested that the act of reading together a literary text not only harnesses the power of reading as a cognitive process, but also acts as a powerful socially coalescing presence, allowing readers a sense of subjective and shared experience at the same time.16 Related research suggested that the inner neural processing of language when a mind reads a complex line of poetry has the potential to stimulate existing brain pathways and to influence emotion networks and memory function.17

The focus of this paper is an exploration of catalysts of change in GiR groups. We consider evidence to indicate the therapeutic importance (or otherwise) of three potentially influential factors:

  • Literary form and content

  • Facilitation

  • Group processes.

We focus on literary form and content because these are at the heart of the GiR experience. The intervention is predicated on the assumption that serious literature offers a model of, and language for, human thinking and feeling with the potential to ‘find’ and alleviate personal trouble and thus to produce therapeutic benefits. This view of literature, as being health-promoting in the widest human sense, has a long and strong lineage in literature and in literary-philosophical theory: from Aristotelian catharsis, through the Renaissance tradition of poetry's morally redemptive and emotionally alleviating function, to the Victorian commitment to the democratisation of culture as an ennobling and life-supporting influence.18–21 This tradition has been resurrected in recent decades by literary scholars, psychologists and health practitioners who argue for the healing effects of reading fiction and poetry, emphasising literature's capacity to put humans in touch with, and help them articulate, implicit and inchoate aspects of personal experience.22–24 These studies intersect with philosophical and psychoanalytic traditions emphasising the importance of having a language to express complex experience as a means of tolerating and surviving it.25–30

We have already noted experiential evidence to support these theories in relation to GiR as a specific literary intervention.15 Facilitation by someone with expertise both in the key subject matter and in education (using the latter term in its literal sense of ‘bringing out what is within’), and the potential benefits of sharing reflections and responses with others in a group setting, are both procedures with established therapeutic provenance,20 and hence likely to be reproducible in this setting.

Methods

Two open weekly community reading groups were established, in a GP surgery and a mental health drop-in centre, for people with a GP diagnosis of depression and a validated measure of severity (Patient Health Questionnaire (PHQ-9)).31 Participation was voluntary and the criterion for inclusion in the group was a validated diagnosis of depression (ie, a PHQ score of 10 or more); note was taken of other medical/psychiatric diagnoses, but there were no other specific exclusion criteria.

Data collection

Data collection took place over a 12-month period between July 2009 and July 2010.

Quantitative data

Questionnaires for participants were administered before and after the GiR experience. These included PHQ-9 scores and self-report information regarding the use of healthcare services (consultations, prescriptions, referrals to secondary care).

Observation and reflective diaries

To capture the interaction taking place within the reading group sessions, we made digital recordings of all sessions. Two members of the research team (JB and JR) each observed one GIR group per month. They and the facilitator (CW) also completed reflective diaries of the group reading sessions.

Analysis and interpretation

Quantitative data were analysed by CD using the standard descriptive and statistical methods packages contained within SPSS V.15.0 for Windows.32

Digital recordings of each session were transcribed using appropriate annotation to prepare the text for conversation analysis (AH), which included studying speech events, such as turn-taking, interruptions, participation and silences, in relation to the immediate setting and stimuli, the poems presented to the participants and the wider social context of the reading group.33

Other data sources, including the observation notes and reflective diaries of the researchers and group facilitator, were used to enhance, corroborate, question and/or contradict the ongoing analysis of the transcripts and the quantitative data. A detailed thematic analysis of the transcripts was used to identify additional components of the intervention, specifically the social dynamics created by group process and the facilitator's role in promoting communication.34 35

A literary researcher (JB) and a linguist (AH) used the data collected via digital recording/transcription to study the ‘cross-hatch’ between participant responses, as established by conversation analysis, and the literature in terms of: the human-emotional content of the poetry and fiction; the role of the facilitator in guiding emotional and intellectual responses to the literary stimuli; and the role of the group process in encouraging the articulation of personal thoughts and feelings in relation to the fiction or poetry.36

Ethics and safety

The project was approved by the Sefton NHS Research Ethics Committee (ref 09/H1001/45) and conducted according to the principles of good research governance. If any participant displayed suicidal or violent ideation during a meeting, a senior professional was available within the host organisation (a trained psychologist in the drop-in centre, a GP in the primary care setting) for immediate consultation.

Findings

At baseline, 18 GiR group participants provided quantitative information and 14 of these supplied complete information. Most participants were aged between 35 and 64. There were similar numbers of men and women and all considered themselves to be white. Fourteen attended the reading group at the mental health drop-in centre. All participants had been in contact with their GP in the previous 6 months, most commonly between three and six times, while half had hospital contact once or twice during the same period. Everyone reported taking at least one regular medicine. The mean PHQ-9 score was 14.3, which is equivalent to a diagnosis of moderate depression. Despite our formal cut-off of a minimum PHQ-9 score of 10, two participants had initial scores of <10.

At follow-up, eight (44%) participants provided updated information on healthcare use and PHQ-9 scores. There was no significant difference in mean baseline PHQ-9 scores between continuing participants and those who dropped out, and demographic proportions were similar at baseline and at follow-up. The evidence suggested a reduction in depressive symptoms for GiR group participants. Six of the eight follow-up respondents had PHQ-9 scores below 10 (the accepted level for depression caseness) compared with only two of 17 at baseline (Fisher's two-sided exact test, p=0.0036). Analysis of the change in mean scores for those with initial PHQ-9 scores of 10 or above (n=6) indicated that the initial mean (SD) score was 16.0 (4.6) and the follow-up mean (SD) score was 11.2 (9.0) (Student t test, two-tailed p=0.0565). There was no trend towards a reduction in mean numbers of GP or hospital contacts or regular medications over time.

We explored three potential catalysts for change: literary form and content; facilitation; and group processes. We also noted influences of the physical environment.

Literary form and content

GiR groups are exposed to a rich, varied, non-prescriptive diet of serious literature with a mix of fiction and poetry. The former appeared to foster relaxation and calm, while the latter encouraged focused concentration. Both literary forms allowed participants to discover new, and rediscover old or forgotten, modes of thought, feeling and experience. The full set of literary and poetic texts used by both reading groups can be found in supplementary online appendix 1.

The continuous narrative found in works of fiction tended to lead to observed and self-reported outcomes of relaxation, or calming of mental anxiety. One participant (an avid reader—“I used to eat books”—before his wife's long-term illness led to his “not being able to concentrate on anything but TV”) described with surprise, at the close of the first session he attended, how the story had “soothed” him “here” (pointing to his forehead). Such soothing effects of story were particularly visible in another participant who, during the poetry reading at the start and close of the session, had been easily distracted (fidgety in body, eyes and head, and excessively aware of the attention and behaviours of other group members) but became stilled as her absorption in the story overcame other claims on her attention.

The relaxation enabled by story is perhaps specifically related to narrative's mode. Unlike lyric poetry, which exists ‘outside’ of time, narrative moves in a continuous temporal sequence, and this ‘fictional’ time can be picked up, re-joined and eased into again after a break for discussion, or after the week-long gap which separates one session from another. The sense of relaxation and the repeated comment from participants that ‘it takes my mind off other things’ might be related to the fact that, in narrative in temporal terms at least, if not in terms of the reader's ‘production’ of the text,25 the future takes care of itself. Certainly this narrative time overrode any interruption in real time: people entering the room or noises outside were collectively ignored unless attention was directly claimed. The intense absorption was closer to meditation than escapism since group discussion continually touched base in real time and with real personal experience.

Poetry, on the other hand, was demonstrably more exacting regarding levels of concentration and mental effort and elicited more verbal expression of thinking, intensity of focus on individual words and meanings and, interestingly in light of increased difficulty, inclusiveness. There was a tendency for participants to go back to and repeat aloud words, phrases or lines in an effort to understand or mine for meaning and the emphasis in the main was on finding meaning for its own sake rather than relating it to personal experience. This tendency can be seen in the following example of discussion between participants A and L (with facilitator F) on Louis MacNeice's poem Snow (words in italic are text from the poem itself):A: The drunkenness of things being various. It is a strange one.L: Yes. I think it's about Christmas though, as you say, definitely, isn't it?A: Or,L: Don't know what to make of it, what do you make of it?A: And the fire flames and the bubbling sound for words, worlds, bubbling sound for worlds, is more spiteful and gay than one supposes.L: I mean they don't sort of go together really do they? Spiteful and gay.A: NoF: NoL: Sort of opposites.

It was observed that participants who never offered to take a turn reading narrative often accepted the invitation to read a poem. The poems' relative brevity (perhaps not so daunting as a long paragraph of prose) was an influence here. Individual confidence also seemed engendered by the mode of discussion characteristically elicited by the poems, where much more verbalised collective meaning-making was evident—everybody working out the poem together, as if it were a puzzle (or ‘conundrum’ as one participant would often put it).

Facilitation

The group facilitator had roles in expert choice of literature, in making the literature ‘live’ in the room and become accessible to participants through skilful reading aloud, and in sensitively eliciting and guiding discussion of the literature. The facilitator's social awareness and communicative skills were critical in creating individual confidence and group trust and in putting the group's needs above those of the individual where necessary. The facilitator's alert presence in relation to literature, the individual and the dynamics of the group is a complex and crucial element of the intervention.

Literary knowledge and expertise were observed to be essential both in making available suitable choices of reading material and in offering an approachable but credible authority on literature that people could question and query. In addition, the facilitator's skill as an expressive reader was key in making the literature ‘live’ in the room and creating an atmosphere of serious attention. The facilitator's further expertise as an interpreter of narrative and of poetry was crucial in holding, and holding open, key ideas or central concerns, often by returning the discussion to tiny details of the poem or story and repeating individual words, lines or sentences.F … this second bit of the poem, the second stanza, that is interesting though as well isn't it, because it's still going on about this idea of stillness, but as if there are things happening underneath the stillness, so you know this going back, these theme of ice, I don't know, as if there is the ice but then there is this current underneath, I don't know.M: The undercurrent to something, and it's, whatever it is has got to come up to the surface, do you think? While you are talking.F: Because he says we know the current is there, hidden. That is interesting this idea that we know, there is great sort of reassurance in that line would you say, we know, we most probably don't know a lot of things do we, but we know, the current is there, hidden though, but it seems to accept that.M: Wants something to come, to bring it out, yes.

A further aspect of the facilitator's literary expertise was the ability to capture details of participant contributions which helped whole group understanding. In this example the facilitator picks up participant Eddie's instinctive and initially quiet repetition of words from the poem (in italics) and uses it as a tool for the rest of the group to keep in focus a key phrase of RS Thomas' poem The Moor: “It was like a church to me”:F: If we just take the very first, maybe we will work with the very first sort of few lines, when he says, It was like a church to me.E: I entered [it] on soft foot.F: Soft foot. That is interesting, isn't it Eddie?M: You feel when you went in a church you sort of, you know, tiptoe in don't you, you don't just sort of just go marching in like you do into a supermarket or something.D: You have got to go in quiet.M: Yes. You do.F: What's that about that going in softly then?M: WellF: What's that about? Is it going in on tiptoe?M: It's all quietI: Not disturbing anything.F: Not disturbing anything.I: YesF: It was like, so it's not actually a church, It was like a church to me.

Group processes

The role of the group was to offer support and a sense of community and this was supported by the accounts of the participants themselves. The latter was fostered particularly by the shared reading model of GiR which includes everyone together in the reading experience. Likewise the discussion elicited in response to the texts, where personal ideas, feelings, opinions and experiences were mutually shared, was demonstrably critical in ‘knitting’ the group together.

Linguistic analysis of the groups' conversational habits over the 12 months shows a clear increase in reflective mirroring of one another's thought and speech habits, as well as greater cooperation and confidence, as participants took the initiative in supporting one another's comments, in guiding the direction of discussion and in offering to read aloud from the text themselves.

Verbatim or near-verbatim repetition of another participant's words shows that the speaker's attention is closely focused on the conversation, and that they wish to support the view the other has expressed. In the following example, in which the group are discussing Laurie Lee's Cider with Rosie, M repeats two words (“took him”) of A's utterance. A corrects her, and supports her own first utterance, by repeating both her earlier salient points. At the same time, however, she repeats M's “she just took him”. The repetition of M's words acknowledges M's point of view and softens the correction:A: And they used him as a battering ram, and he come back absolutely penniless, and then this girl from the district, Fanny, Fanny took him and married himM: [laughs] Yes, yes, yes, I don't think there was ever nothing said about love, she didn't fall in love with him, she just took him.A: No, she just took him took him and married him.

Syntactic mirroring is another means or indication of a speaker supporting another participant's viewpoint. In the following example, where the group are reading Mitch Albom's novel Five People Who Meet in Heaven, R's syntactic structure is twice mirrored by L.R: Either you are responsible for your fate or you are not.L: Either you believe or you don't believe.F: I wonder if they are two different things there, because partlyL: (interrupt) Either you believe in God or you don't believe in God.

In the next example, again from the discussion of the poem Snow, A ponders the meaning of the poem, then, with the comment “oh, strange”, shifts out of the discussion of the text and creates a new world, warm and pleasant. L immediately joins her in this fantasy world, and mirrors A's syntax as she does so:A: What's going on and maybe Christmas, maybe just a sudden downfall of snow. Oh, strange. Light the fire, get everything nice and warm.L: And sit and look out. Out on the world with the snow coming down.

This discussion was at times intense, bursting with the simultaneity of thoughts, worlds and realities:L: More than one thing happens from. You can just imagine yourself being there can't you?A: Yes.L: The snow coming down.A: Yes. Snow us another tangerine [laughs].L. And the pips.A: Spitting the pips in the fire.L: Sitting there in front of the fire, peeling a tangerine.A: Peeling a tangerine, splitting it up, and the pips in the fire.L: You can just think of that. You can just imagine that really.

Analysis of the rhythm and intonation of the utterance “Snow us another tangerine” shows that this playful literary metaphor can also be set to music (the ‘tune’ represented by musical notation as shown in figure 1), reflecting A's exuberant ‘singing-out’ of the dissolving of boundaries between selves and worlds.

Figure 1

The tune for “Snow us another tangerine”.

At other times, there were signs that group members were free to pursue individual and personal trains of thought, enabled by the protective presence of the group. “Do you think he feels as though it's, there is like a magic to that just that moment, it's magical”, asks L in response to Snow. “You would rather have that place, than a church as well”, was one participant's reflection on The Moor. Separated, divergent lines of thought, or ‘personal’ narratives of meaning often surface and, from time to time, converge with the group discussion. But they are demonstrably continuing intently beneath the surface of the latter and are indicative of subterranean concentration over a long time span on particularised matters.

These factors and processes are known to be influential in enabling therapeutic effectiveness in group settings, and have been described by Yalom in his analysis of interpersonal group psychotherapy. The instillation of hope and the development of cohesiveness are key ingredients here. Syntactic mirroring, for example, is closely linked to Yalom's concept of imitative behaviours.37

Physical environment

The physical environment contributed to the atmosphere, group dynamic and expectation of the utility of the reading group. The group which met at the mental health drop-in centre was easier to recruit, and was much more willing to engage with the literature for its own sake from the very outset of the study. By contrast, the group which met at a GP surgery initially tended to view the literature as something ‘prescribed’ to them in direct relation to their mental health problems. The location of the latter reading group in (often different) doctors' offices may have encouraged this perception, where the former reading group had a designated and more informal space for the group each week.

However, while the environment influenced the group, the collective action of the literature, facilitator and group appeared to supersede that of the environment. The creation of a stimulating, non-pressurised, non-judgemental atmosphere (“not like school”, as one participant emphatically put it) overrode considerations of physical environment.

Discussion

In this study we have found qualitative evidence indicating that three core components—shared reading of literary texts, skilled facilitation and social group processes—may be important for any therapeutic efficacy that GiR possesses. We consider that the focus on literary texts specifically, and the combination of these three components generically, are unique to the GiR model. With regard to the literary content, we found that narrative texts tended to encourage reflection and relaxation, as within this format the future tends to take care of itself, whereas poetry, which is intentionally more immediate, was more likely to evoke participation and active social engagement. The combination of these responses appears to be valuable in reducing key symptoms of depression, such as anhedonia, difficulties with concentration and personal withdrawal.

It was easier to recruit participants in the mental health drop-in centre than in the primary care setting, suggesting that the former may be a more feasible venue for this type of intervention. Our quantitative findings indicated possible therapeutic benefits for participants, but did not indicate any clear pattern of change in health service use: this may suggest that the impact of GiR is mainly at a personal level, in terms of a reduction in negative self-images or stories.

This study has four main limitations. First, its design was such that we are unable to offer more than suggestive evidence of the therapeutic efficacy of GiR groups as an intervention for depression, given the small sample size and the lack of either standardised diagnostic interviews or control groups. Second, less than half of the original participants provided follow-up information: it is possible that those who left the group had more negative experiences of participation than those who remained. Third, we did not collect information on other modalities of psychotherapy in which participants may have engaged. Finally, the focus on participants with depression means that insights into the catalysts for change in GiR groups may not be generalisable to other clinical groups for whom GiR has been offered (eg, people with dementia) or in educational, forensic or other non-clinical settings. In these contexts it is possible that other change processes come into play.

GiR groups are becoming increasingly popular in both clinical and non-clinical settings. There is therefore a pressing need to subject them to rigorous analysis in a number of dimensions, before healthcare policy makers, commissioners and clinicians can have confidence in recommending their introduction into routine healthcare.

We have indicated that proof of concept, feasibility and acceptability studies have already been undertaken. In this study we have identified potential GiR catalysts for change in the context of common mental health problems. Further work is needed to elucidate these processes in greater detail, deploying further methodological approaches in addition to linguistic analysis, and using interview and focus group data gathered from study participants. Important next steps include examination of evidence of the social and economic impact of GiR groups, using before-and-after study designs, and comparison of catalysts for change for patients with other common conditions such as dementia. All of this research should then lead towards substantive effectiveness and cost-effectiveness studies based on randomised controlled trials of GiR groups in comparison with evidence based interventions including adequate attentional controls.

Acknowledgments

We are grateful to the staff of Upstairs at 83 and of the Aintree Park Group Practice for hosting and facilitating the reading groups that form the basis of this study. We are immensely grateful to the people who took part in these reading groups, for sharing their time and experiences with us.

References

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Files in this Data Supplement:

Footnotes

  • Funding This study was funded by a grant from MerseyBeat/Liverpool Primary Care Trust.

  • Competing interests None.

  • Ethics approval Ethics approval was provided by Sefton NHS Research Ethics Committee (ref 09/H1001/45).

  • Provenance and peer review Not commissioned; externally peer reviewed.