Therapeutic writing
Writing as a form of therapy to improve physical or mental health has a long history1 and is widely reported in psychology textbooks as being of therapeutic value.2–5 It can take many formats including those from a psychotherapeutic background, such as therapeutic letter writing,6 specific controlled interventions, such as emotional disclosure/expressive writing,7 to more recent approaches, such as developmental creative writing8 and other epistolary approaches, such as blogging.8 With the development of UK organisations such as Lapidus (Association for Literary Arts in Personal Development), dedicated to the promotion of therapeutic writing (TW) based on the premise that it has health benefits, and an increased interest in the potential of non-pharmacological adjunctive therapies, it is important to evaluate the effectiveness of a variety of different approaches within this field. These include two main categories: written emotional disclosure [or emotional writing (EW)]9 and creative writing, such as poetry.7 Other forms of creative writing include alpha writes/poems (a poetic device in which each successive line of the poem starts with the next letter of the alphabet, or a predetermined word or phrase written vertically down the page);10 writing from published poems/narratives; acrostics; and haiku poems (traditionally a haiku contains three lines of five, seven and then five syllables, making a total of 17 syllables; the tradition is modified so that no more than 17 syllables are arranged in no more than three lines, but the shorter the better);11 autobiographical writing (such as reflective diaries, journaling); descriptive writing; genre writing (e.g. fairy tales); free writing; short stories; drama or fictional narratives; unsent letters; diary/journaling; collaborative writing (workshops); writing accompanying other art forms; life writing or memoirs (such as reminiscence, life review); list writing; redrafting or sentence stems writing; scribing for others; writing from visual/sound stimuli (e.g. writing from mindfulness); writing from the senses; and writing in form and writing from music (as part of a music therapy).10,12 Newer forms of writing include blogging or participating in web-based forums.13,14
Dimensions of therapeutic writing
Within each type of TW there may be significant process variability, for example in the flexibility and number of topics; the dose (frequency and duration); group or individual delivery; computerised compared with handwritten exercises; participant recruitment; and financial compensation. However, a major distinction lies in whether the writing is facilitated or unfacilitated.
Facilitated therapeutic writing
Facilitated TW interventions, when a facilitator is present at some stage before or during the writing, might be delivered in many different ways and contexts: in a health-care centre, as part of a programme in a rehabilitation clinic or within a group of people with common or different chronic conditions, face to face or via the internet. People using these therapeutic tools may receive feedback from someone else, a health-care professional, a group of persons or not receive feedback at all.10
Furthermore, the topic of the writing can be varied, from positive to negative expression of emotions through neutral topics (e.g. childhood/birth, life aims and goals, places, relationships).10 TW can also be used in children, adolescents or adults and among different clients, such as the chronically ill (e.g. cancer, mental health problems, chronic infections) or healthy individuals, and assessed from different angles such as community carers, doctors and nurses, peer training, patient’s family and/or friends or the participant him/herself – the most usual perspective.10
The writing event should not be considered as solely an isolated exercise but as a sequence of exercises, not necessarily all of which encompass a written component. As such, a common practice is that writing starts with (visual) stimuli12 or with mindfulness meditation in order to inspire people and to act as a form of distraction from reality.15 The following examples of the work of facilitated writing practitioners in the UK exemplify the variety of this type of TW (Box 1).
Examples of facilitated therapeutic writing Writing practitioner Carol Ross (CR) facilitates weekly TW groups for inpatients in mental health units in a UK NHS Foundation Trust. CR has developed her own practice, influenced by a number of published research (more...)
Perhaps the last thing to add is that clinicians at Freedom from Torture continue to send more clients than can be coped with, which is some vindication of the practice. As a bridge between intense therapy and the outside world, giving clients a skill they can take with them and a means of self-discovery, self-healing and personal growth, it seems, as the Hippocratic Oath puts it, to do no harm, and a great deal of good.
Unfacilitated emotional writing
Unfacilitated EW means writing, completed without any assistance, feedback, comment or any other form of support. Therefore, in the current review, an unfacilitated writing intervention has been defined as when a facilitator was simply not present in person during the writing exercise, as opposed to facilitated writing (described above). Typically, in unfacilitated writing interventions, participants are instructed to write for 15–30 minutes on 3 to 4 consecutive days (or at weekly intervals). Instructions on these unfacilitated writing assignments can be delivered in writing via leaflets, verbally over the telephone or even via video or the internet. Participants are asked to do their writing unassisted and alone at home or on their own in a given clinic or laboratory setting. In the most commonly evaluated form of unfacilitated EW, there is a single writing topic that can be chosen (usually disease or treatment focused), for which participants are directed to write emotionally and disclose about their deepest thoughts and feelings or about a self-selected trauma. Thereafter, either the writings may be collected by the practitioner without any feedback or the participant can simply decide what to do with the writing. Sometimes, practitioners provide participants the option to make telephone calls during the writing exercise should any concerns emerge; however, this action has not been considered as facilitation in this review.
The Pennebaker writing paradigm: expressive writing or written emotional disclosure
This is the most common form of unfacilitated EW. It is a technique whereby people are encouraged to write (or talk into a tape recorder) in private about a traumatic, stressful or upsetting event, usually from their recent or distant past. They write for 15–30 minutes typically for 3 or 4 days within a relatively short period of time, such as on consecutive days or within 2 weeks. The format has been relatively consistent since the earliest randomised controlled trials (RCTs),1,18 but more recent studies have varied the duration, number of sessions and topic of writing, including positive events and thoughts and feelings about illnesses.19 RCTs of expressive/emotional writing have been conducted in a wide variety of participants, including healthy students, people undergoing psychological stressors, such as bereavement or being in a caregiving role, or in people with long-term physical conditions, such as rheumatoid arthritis (RA) and asthma. Variants of the technique include disclosure in front of a listener, who can be a confederate, a researcher or a doctor. For the purposes of this project, these activities are not considered to be EW and lie outside the scope of this review. The presence of a listener is likely to affect outcomes, as it potentially adds a counselling dimension.
Positive writing
Positive writing can be delivered either as part of a facilitated TW or unfacilitated EW intervention. The exercise involves writing about positive topics only, including positive emotions, typically for 20–30 minutes, three or four times per week if delivered as an unfacilitated EW intervention. Otherwise, the duration and length of the positive writing can be very varied when facilitated (see Appendix 1). Pennebaker et al.19 found in a review published in 1997 that the description of positive emotions could predict improvements in health outcomes. Since then, researchers have studied the positive effects of the positive emotional disclosure, advocating that participants writing about the positive aspects of past traumas (benefit finding by describing any positive outcomes of the disease experience or treatment in detail) or simply about positive life events, could achieve comparable health improvements as those writing about past traumas.20
Bibliotherapy
Although the above require the individual to write as part of therapy, other forms of therapy use existing texts. The most commonly encountered type of bibliotherapy, Reading Bibliotherapy, involves reading material specifically selected for its therapeutic potential for that person.21 In the UK, Books on Prescription Schemes have been running in primary care for several years, and in 2013 a national scheme was launched in England by the Society of Chief Librarians and the Reading Agency.22 Such reading bibliotherapy is not covered by this review.
In contrast, interactive bibliotherapy has been defined as the use of literature to bring about a therapeutic interaction between participant and facilitator.21 The triad of participant, literature and therapist is viewed as critical. In fact, interactive bibliotherapy does not restrict itself to the written word: it can include the spoken word, for example in film or theatre but it must involve the coherent use of language. When interactive bibliotherapy uses poetry, it is synonymous with poetry therapy and they are both encompassed by the term biblio/poetry therapy.21 Sometimes the literature involved in biblio/poetry therapy is new writing generated by the participants themselves. This type of creative writing biblio/poetry therapy is the principal form of facilitated TW included in this review and it is the form of TW used by the practitioner experts collaborating in the current systematic reviews (see Box 1, in which the TW expert practitioners describe their different facilitated biblio/poetry therapy practices).
Nonetheless, little has been published around all the different types of facilitated TW, and literature shows that the most evaluated form of TW is the EW intervention, described by Pennebaker and Beall1 Comprehensive research around the writing paradigm18,23–27 and narrative analysis within the health-care setting28–31 has been performed through the last decades.
Long-term conditions
The prevalence of long-term conditions (LTCs) increases with ageing populations. In 2002, the leading chronic diseases [cardiovascular disease (CVD), cancer, chronic respiratory disease and diabetes] were responsible for 29 million deaths worldwide.32 According to the UK Department of Health (DH),33 more than 15 million people in England (including half of all those aged > 60 years) are living with at least one LTC, and the risk of death is particularly high in those with three or more conditions occurring concurrently.34 LTCs also result in a huge burden on UK NHS resources. Although some are preventable, for most LTCs the only realistic management strategy is continuing care, as biological and psychosocial mechanisms regulating disease progression are not yet fully understood. As LTCs are difficult to improve, especially for elderly populations, health-care programmes, such as self-management support and patient education, often combined with structured clinical follow-up, have been suggested as a way to improve the quality of life (QoL) of such patients.35 New therapeutic approaches, such as TW, have the potential to improve the QoL in people with LTCs.
Possible pathways linking memory, emotions and physical health
There are several potential ways that writing might impact on physical health. For example, cognitive restructuring or behavioural mechanisms (e.g. reflection on health behaviours) may lead to improvement in outcomes. However, many of the types of TW described above engage emotions and memories (both positive or negative) and there are physiological pathways linking memory, emotions, chronic stress and physical health.
Two interdependent memory systems are thought to be associated with remembering events in humans.36 Episodic memory is linked to the hippocampus and this structure is vital for processing events that eventually become long-term memories.37 Emotional memory is linked to the amygdala, part of the limbic system involved with emotions, in particular fear-related responses and general pleasant and unpleasant emotional processing.38 Although the episodic and emotional memory systems are independent, they affect each other in a variety of ways.36 Emotion enhances perception of, and attention to, the memory-provoking stimulus, as well as the long-term storage of the memory.36 Episodic memory also influences emotional memory by, for example, causing the autonomic effects of emotional arousal (e.g. the sweaty palms and dry mouth) when remembering a past situation.36
The limbic system has links with the cerebral cortex, the brainstem and the pituitary gland (part of the hypothalamic–pituitary–adrenal axis). Parts of the cerebral cortex have a role in cognitive appraisal and the conscious awareness of emotional states, and can regulate amygdalar activity.38 Through the brainstem, areas in the limbic system can control many internal conditions of the body, for example cardiovascular regulation. The hypothalamic–pituitary–adrenal axis is both responsive to psychological inputs and has significant influences on the immune system, which, in turn, influences physical health.39 This pathway may be one of the ways chronic stress is linked to poor health.39–42 It is therefore possible that a psychological intervention might improve aspects of physical health, and if modification of such pathways had even a small effect then this could have profound public health significance.
Realist synthesis
In this review, a conventional systematic review of the effectiveness of unfacilitated EW and facilitated TW was conducted, but, in addition, the findings of a realist synthesis are reported. Realist synthesis is a theory-driven interpretive approach to evidence synthesis. Rather than producing a judgement on whether (or not) an intervention works, realist syntheses attempt to explain outcome patterns in data using theory (or theories). It is particularly useful when interventions are complex and evidence is mixed or conflicting and provides little or no clues as to why the intervention worked or did not work when used in different contexts, by different stakeholders or when used for different purposes.43
In brief, realist syntheses ask what works for whom in what circumstances, how and why? To do so, realist syntheses use a particular logic of analysis that deliberately breaks down how an outcome has arisen. An outcome is considered to have occurred because it is caused to do so by a causal process known as a mechanism. In addition, the contexts in which an outcome has occurred are also considered to be important as they cause mechanisms to be activated. This logic of analysis thus provides an approach for understanding how and why it is that context can influence outcomes. In summary, the realist logic of analysis used in a realist synthesis considers the interaction between context, mechanism and outcome (sometimes abbreviated as CMO). That is how particular contexts have triggered (or, conversely, interfered with) mechanisms to generate the observed outcomes.43
To elaborate further, in order to understand how outcomes are generated, the roles of both external reality and human understanding and response need to be incorporated. Realism does this through the concept of mechanisms, whose precise definition is contested but for which a working definition is ‘. . . underlying entities, processes, or structures which operate in particular contexts to generate outcomes of interest.’44 Different contexts interact with different mechanisms to make particular outcomes more or less likely – hence, in general, a realist synthesis produces recommendations of the general format ‘In situations [X], complex intervention [Y], modified in this way and taking account of these contingencies, may be appropriate’. This approach, when done well, is widely recognised as a robust set of methods, which is particularly appropriate when seeking to explore the interaction between CMO in a complex intervention [e.g. see Berwick’s editorial explaining why experimental (RCT/meta-analysis) designs may need to be supplemented (or perhaps in some circumstances replaced) by realist studies aimed at elucidating CMOs].45
The philosophical basis underpinning a realist synthesis is realism. Realism assumes the existence of an external reality (a real world) but one that is filtered (i.e. perceived, interpreted and responded to) through human senses, volitions, language and culture. Such human processing initiates a constant process of self-generated change in all social institutions, a vital process that has to be accommodated in evaluating social programmes. In other words, the way individuals interpret and respond (or not) to the world around them has the potential to cause changes to this world around them. Such changes may then cause additional responses from individuals, potentially leading to a series of feedback loops. Within a realist synthesis, where possible, attempts are made to understand these feedback loops.
A realist approach is particularly useful for this project because TW is a complex intervention that could be useful in a variety of patient groups, and currently it is unclear whether it is effective for all or some, and how and why it might be effective.
Realist syntheses often use input from content experts to help develop the programme theories needed to explain how complex interventions work. In this project, input from practitioner experts was deliberately sought. During the second programme theory-building meeting with practitioner experts, they were asked for their feedback on what their views were on how TW was meant to work, for whom and why (see Chapter 5, Methods, for more details). Two practitioner experts [Carole Ross (CR) and Victoria Field (VF)] provided written responses (see Appendix 1, Tables 76 and 77, respectively) and have been included in this report as they provide an insight into how facilitated TW is used in the NHS and voluntary sector.
Previous systematic reviews on therapeutic writing in long-term conditions
There have been a number of systematic reviews on expressive writing,18,23,46,47 published in psychology journals, that have conducted meta-analyses according to normal practice in psychology, combining different types of participants and outcomes across different conditions, and using Cohen’s d or Hedges’ g statistics. Their results are difficult to interpret because effect sizes for specific populations and interventions are unclear. There have been three recent systematic reviews on TW in LTCs. One concerned post-traumatic stress disorder (PTSD) only and included five studies.27 One of the included studies is on cognitive–behavioural therapy (CBT) rather than TW,48 and another49 is a very small, non-randomised study with students. A second, unpublished systematic review was accessed via the internet.26 This assessed TW for psychological morbidity in people with long-term physical conditions. The review included 14 RCTs and searches were conducted up to May 2011. It is unclear why this review did not include a number of potentially includable studies including Abel et al.50 [human immunodeficiency virus (HIV)], Graham et al.51 (chronic pain), Halpert et al.52 [inflammatory bowel syndrome (irritable bowel syndrome, IBS)], Henry et al.53 (breast cancer), Hughes54 (breast cancer), Kraaij et al.55 (HIV), Petrie et al.56 (HIV), Stark57 [fibromyalgia (FM)] and Theadom et al.58 (asthma), as all of these studies measure psychological morbidity and were published before the search end date. It may be that they did not include some of these because of their definition of long-term physical conditions, as there is no uniform definition as yet. It is unclear how the results might have differed if some of these studies had been included. The third systematic review evaluated the impact of support on the effectiveness of written cognitive–behavioural self-help59 and thus was not really focused on TW per se. It included 38 studies, none of which are included in this project.
Hypotheses tested in the review (research questions)
Overall aims and objectives of this review
What are the different types of TW that have been evaluated in comparative studies? What are their defining characteristics? How are they delivered? What underlying theories have been proposed for their effect(s)?
What is the clinical effectiveness of the different types of TW for LTCs compared with no writing or other suitable comparators?
How is heterogeneity in results of empirical studies accounted for in terms of patient and/or contextual factors, and what are the potential mechanisms responsible for the success, failure or partial success of interventions (i.e. what works for whom in what circumstances and why)?
What is the cost-effectiveness or cost–consequences of one or more types of TW, in one or more representative LTCs, when there is sufficient information on the intervention, comparator and outcomes to conduct an economic evaluation?
