Storytelling is regarded as central to good clinical practice creating trust within the therapeutic encounter  . It is defined as “the effort to communicate events using words (prose or poetry), images, and sounds often including improvisation or embellishment. Stories are reflective, creative and value laden, usually revealing something important about the human condition.” (  ; page 408). In the context of chronic and palliative illnesses they are a well-established medium used e.g.,   . Storytelling interventions have gained a great deal of traction for their ability to enhance health communication and clinical practice  . They have been identified as encouraging positive behaviours associated with diet and physical activity   . From the perspective of people with cancer storytelling may provide a medium through which emotions can be expressed  . Indeed, the process of sharing a story involves a transportation of emotions as the teller relives events  . It is likely that sharing stories enables a positive psycho-emotional and adaptive response to illness  .
Past reviews have considered the benefits of storytelling-interventions for individuals with cancer at the screening point e.g.,  . However, limited studies have been included within review evidence. For instance, there was very limited evidence of storytelling as a benefit for the side effects of cancer treatment  or as a behavioural treatment for paediatric patients  . In a broader review in chronic illness only 2 out of 10 studies considered the benefits of storytelling on individuals with cancer  . A more comprehensive review is justified because of the limited exploration into the value of storytelling. A review needs to be inclusive of the full-range of methodologies used to consider the experience and impact of interventions on people with cancer. Past studies have used qualitative e.g.   , mixed methods e.g.  and quantitative studies e.g.  to reveal the impact and change created by storytelling interventions. Given the above, the aim of the current research is to undertake an integrative review of evidence identifying the impact and outcomes from storytelling interventions for people with cancer.
An integrative review  was undertaken and reported according to three traditional steps; 1) Study eligibility and literature search process, 2) Data evaluation and 3) Data synthesis. A PRISMA checklist and flow diagram was used to aid the documentation of the search techniques and outputs  . For the purposes of this review we assume a subtle realist paradigmatic view-point.
2.1. Protocol and Registration
A protocol was registered with PROSPERO with ID: CRD42018115901.
2.2. Eligibility Criteria
The eligibility criteria are presented according to the PICOS (Participants, Intervention, Comparator, Outcome and Study Design) acronym:
Any participants that have been clinically diagnosed with cancer. Any age group or gender group was allowed. The evaluation of other stakeholder’s view of a storytelling intervention was included. Where other stakeholders (health care professional, family, carer) were included in studies the results section had to devote at least 50% of the results to the participant with cancer to be included. Where multiple populations were used, a separate results section for patients with cancer was required. Screening based interventions were excluded.
A storytelling or narrative intervention had to be reflected on or examined through any means (including face to face, internet based, tele-based). The intervention had to take place within a group setting. The purpose of the study had to include looking at the impact of storytelling on the psychosocial and emotional health or behaviour of participants. The storytelling intervention had to include contact between individuals and involve sharing stories with other patients. For the purpose of this study expressive writing was excluded. Articles that did not include a specified intervention for example a narrative analysis of experiences of services were excluded. Articles that used narrative as a small part of the intervention were excluded.
Any active or inactive control group could be used. No control group was acceptable.
Any form of data collection could be used as long as it was able to capture experiences relating to the impact of storytelling on the individual’s psychosocial or emotional well-being or behaviour.
Any study design was acceptable. All thesis and conference abstracts were excluded. Studies written in English, Spanish, French and Portuguese were included. Other languages were excluded due to the absence of translation services.
2.3. The Search Process
A sensitive topic based search strategy adhering to recent guidelines  was undertaken. The search included electronic databases (until 30 November 2018) including MEDLINE, Psychology and Behavioral Sciences Collection, CINHAL, PsychINFO, SPORT Discuss, EMBASE, PEDro, ZETOC databases. The search was supplemented by selected internet resource sites including: PubMed, Turning Research to Practice, and the first 20 pages of results from the sites Google Scholar and Science Direct. Once initial articles were obtained further searching of the 5 most common journals (identified in the database search) was conducted and citation chasing was undertaken. Key words: Cancer* AND Narrative OR Illness Narrative OR Story OR Storytelling OR Expression OR Emotions AND Intervention OR Experimental OR Qualitative OR Quantitative Or Mixed Methods.
2.4. Study Selection
In the first stage of the selection process authors identified articles that potentially met the criteria for inclusion. The lead author removed duplicates and two independent reviewers considered the abstract of each article. A third independent reviewer was available to establish if an article should be included.
2.5. Data Collection Process
A predefined extraction form was used to identify critical study-design information and demographical details as follows. Participant demographic variables (age, gender, site of cancer, staging, geographical location and setting of the study for both the patient-group and any control-group; design and implementation intervention specifics together with details of trial  , quality appraisal summary, analysis and key results).
2.6. Data Evaluation
Quality assessment of qualitative studies was undertaken using the consolidated criteria for reporting qualitative research (COREQ)  based ona 13-item version  . The studies were also assessed by a more specific checklist that identifies the characteristics of qualitative studies  . Both authors used the checklists to identify the utility of the studies for the synthesis process. The Cochrane risk-of-bias tool allowed the internal validity to be considered for each quantitative and mixed-methods study.
2.7. Data Synthesis
A qualitative-led synthesis was undertaken as recommended  . The synthesis used open-coding, this was followed by mind-mapping of the established themes and tubulisation of results into thematic groups. The table was reduced to focus on the most common themes and the quantitative findings were used to develop the understanding around them through a final process of integration.
Nine hundred and two unique records were screened. Eleven articles were identified as meeting the eligibility criteria. Figure 1 provides a PRISMA diagram representative of the information.
The total number of post-treatment participants included across studies was 493 (49 female, 16 male, 428 not disclosed). There were 168 duplicate participants  and 10 spouses  not included in this number. One study used patients who were terminally ill within a hospice setting  . The most prevalent types of cancer identified included; Head and Neck Cancer (n = 246), breast cancer (n = 173) and not stated (n = 40). Other types with more than one participant included gynaecological (n = 8), Lung (n = 4) and Stomach (n = 2). All studies apart from one  included adults over the age of 18 years, only one study  provided a mean age (55.4 ± 11.0 years). No study provided details of the number of years of living with the illness. Settings included oncology clinics (n = 4), were internet based interventions (n = 3), tele-conferences (n = 3) and a medieval castle (used as a cancer centre that offers courses). Most studies were undertaken in the USA (n = 7), with single studies taking place in Denmark, Brazil, Japan and France. The supplementary file provides a full breakdown of demographics (see Table S1).
3.2. Data Evaluation
All qualitative studies met pre-defined agreed criteria. Full consideration of the quality assessment is provided in the supplementary file. The quantitative studies identified some high risks-of-bias, as; 1) no protocol (n = 8), 2) no allocation concealment (n = 7) or 3) no randomisation procedures (n = 7). A total of four studies     had 5 domains identified as high risk-of-bias. Breakdown of the risks of bias are listed and described in Table 1. The supplementary file provides a full breakdown of data evaluation by design type; see Table S2 and Table S3 for qualitative evaluation and Table S4 for quantitative evaluation.
Figure 1. PRISMA 2009 flow diagram.
Table 1. Summary assessment of the overall risk of bias.
*Trial denoted by first author. Risk of bias criteria: 1, Selection bias = random sequence generation; 2, Selection bias = allocation concealment; 3, Detection/Performance bias = blinding of personnel, assessors and participants; 4 Attrition bias = incomplete outcome data; 5 Reporting Bias = short-term selective outcome reporting; 6, Other bias = potential threats to validity e.g., consideration of a protocol. Levels of risk of bias: H, high risk of bias; U, unclear risk of bias; L, low risk of bias.
3.3. Data Synthesis
Two major themes: content of interaction and outcomes from the interventions and six sub-themes were identified. Full details of data synthesis can be identified in the supplementary file Table S5 onwards.
Major Theme 1: The Content of Interaction
The first major theme identified had two sub-themes: 1) the expression of emotions which validated suffering and 2) the appraisal of life circumstances and experiences.
3.4. Subtheme a: The Expressions of Emotions Which Validated Suffering
This sub-theme identified the importance of the opportunity to express suffering and the ability for each participant to consider and offer true and experienced feelings associated with the illness. Six       of the eleven studies identified content within this theme.
Individuals had an opportunity to discuss fears and worries which were aspects of the situation that were exclusive to the group     . The group environment represented a space where judgement from others about individual’s story/ies was suspended or limited  . Shared topics included fear of suffering, mental exhaustion and a sense of isolation  . Song et al.  identified that around a fifth (22.7%, n = 13/61) of the conversations were focussed on pain or suffering and similar proportion around fears and concerns related to the cancer recurring and/or symptoms worsening (18.2%, n = 11/61). This may vary by type of medium used within the studies. For instance, only 19.7% (n = 12/61) of video postings (n = 12/61) on the internet had a direct reference to cancer.
In a study by Heiney et al.  over half the respondents (64.1%, n = 25/39) identified that having others listen and feeling free to talk was the best aspect of the group-discussion. Allowing individuals to talk openly was identified as healing for the whole group  . There was a very high agreement (4.0 ± 0.2) with the statement (out of a possible score of 4) that “it was good to learn from other patients about their feelings”  . Other studies noted that group interaction didnot have to be serious all of the time. Comedy was used as a way of providing distance from situations discussed and reduced the intensity of the discussion e.g., joking about the need of “shampoo for bald-headed people”  . The group environment could tolerate silences because participants felt at ease during the group sessions  .
3.5. Sub-Theme b: The Appraisal of Life Circumstances and Experiences
Sub-theme b illustrates the importance of the group being more than just focused on emotional expressions and incorporated broader aspects of living to be included within the story-telling groups. Two of the eleven studies   identified content for this theme.
The broader conversation that occurred could include the appraisal of participants’ lives beyond the symptoms of the cancer. One participant in the study by Evans et al.  identified her role within the family as an aspect of life that had been impacted by cancer. She stated her situation around children was affected and required her to project a sense of being in a normal parent role. The study by Song et al.  illustrated that these broader and overarching conversations could include non-pharmacological treatments like nutrition, exercise, alternative medicine (15.0%, n = 9/61) as well as an individual’s interests and hobbies (57.4%, n = 35/61) and unique personal qualities (47.5%, n = 29/61).
Theme 2: Outcomes from groups
The second major theme identified common outcomes evident across the different studies. There were four sub-themes: 1) psychological healing, 2) developing a shared understanding of coping, 3) social well-being and 4) legitimising or rationalising the experiences of illness.
3.6. Sub-Theme (a) Psychological Healing
This sub-theme identifies the impact of the group on psychological well-being as well as an understanding of why the impact may have occurred. The sub-theme was supported by seven        of the eleven studies.
Sharing stories appeared to provide a space to aid the psychological well-being of the individual. This was illustrated by significant decreases in stress (p < 0.05, F = 12.2), fatalism (p = 0.03) and fear (p = 0.02) over time in comparison to a control group  . Another study identified that nine (45%, n = 9/20) participants changed their original pre-group emotion to a happier emotion (p = 0.01)  . Not all changes were significant however. There was no significant change in spiritual well-being or hope identified  , or significant impact on physical self-efficacy pain  or depression   reported by storytelling groups. Although of these non-significant results all studies except Heiney et al.  had at least five aspects which were identified as high on the risk of bias assessment.
Being able to share, be heard by someone else and tell and retell stories within the group setting was identified as a process which eased psychological suffering    . For instance, one participant identified “If a laugh is good medicine then the chat was something that can heal, maybe not our breast cancer but our souls.”  . This was quantified by Heiney et al.  who identified agreement with statements (0 no agreement to 4 complete agreement) that individuals; 1) enjoyed the sharing experience (3.8 ± 0.4), 2) felt better about themselves as they were helped by the group (3.8 ± 0.4), 3) were able to express personal feelings (3.72 ± 0.45) and concerns (3.72 ± 0.45) within the group.
3.7. Sub-Theme (b) Developing a Shared Understanding of Coping
This sub-theme identified that in the process of storytelling individuals were able to describe and understand how to manage their condition because the information was shared by others enabling reflection. Four     of the eleven studies supported the sub-theme.
The group environment provided a space which created the capacity to understand how to cope and manage with both a diagnosis and symptoms of cancer     . One study  quantified this by scoring agreement with the statements out of a complete agreement of 4. It was identified that the group “helped them cope” (3.7 ± 0.44) and “supported others to examine different ways of managing problems associated with cancer” (3.56 ± 0.55.)
Learning to cope included the ability manage difficult aspects of the illness like sexually function  . In the study by Evans et al.  Individuals identified because of the storytelling intervention that they understood how they may not be able to cope now and went on to identify the repercussions in their lives.
3.8. Sub-Theme (c) Impact on Social Well-Being
This sub-theme identified that in the process of storytelling individuals gained new friendships that were a source of comfort and well-being. Four studies     out of eleven supported this sub-theme.
The storytelling intervention counteracted feelings of isolation   . This was identified in a quantitative study with a subsequent significant improvement in social connection  and by another study as agreeing with the statement (out of 4) that individuals “felt supported by other group members” (3.9 ± 0.34)  . Personal stories of those who had survived and found ways to live with cancer were perceived as encouraging  . In one study  the majority (65%, 26/39) of individuals identified that they were more able or empowered to seek support following attendance at the group. One study identified that the internet initially made it easier to discuss and share personal feelings  . However, another  identified a participant who regretted sharing personal information in a group setting. Another study  identified that the process of sharing developed slowly and it was unlikely to happen in the early stages of the group.
3.9. Sub-Theme (d) Legitimising or Rationalising the Experiences of Illness
This sub-theme identified that individuals felt there was an important function of the group through which they could feel able to share their experiences and feel “heard” or listened to by others. Five studies      out of eleven supported this sub-theme.
The following aspects of the storytelling intervention enhanced the experience of being heard and valued. This included; 1) feeling accepted by others  , 2) entering a shared social-world of another person   , and 3) the ability to use and relate to metaphors and conversations with deep personal meaning  . Participants felt able to join in and relate to others   . Relatedness was enhanced by laughter and intimacy 29] or the ability of expressions to reveal vulnerability  . One study identified how this relied on the participants’ ability to believe and trust in the message from the other members of the group. One of the studies  measured “source trust” (the belief and trust in the person giving the message). This quality was higher in the storytelling group compared to comparison group (T = −10.61, p < 0.001). Source trust was also associated with more positive attitudes for instance exercise beliefs and benefits. Another study by Heiney et al.  identified agreement with statements (0 no agreement to 4 complete agreement) that individuals; 1) liked “sharing with other women with breast cancer” (3.8 ± 0.37), 2) were able to “share a thought that would not be shared with most people” (3.63 ± 0.5). Both points illustrated a willingness to share meaningful information within the group.
This is the first integrative review of literature to bring together the value and impact of storytelling interventions for individuals with cancer. Results identified that often the focus and content of discussion within the storytelling group contained elements of suffering, isolation, and emotions or feelings associated with the cancer. The discussion went beyond the immediate impact of cancer. The freedom to choose stories and be open with one’s expressions of concerns and worries was important within the group setting. This ability to be open benefitted patients socially and was optimised through the creation of trusting relationships that developed. The ability for all individuals to contribute to the storytelling intervention if they choose to as peers within an open atmosphere may explain the value of a storytelling intervention  . However, it must be noted that the ability to be free to express views may vary by setting and type of interaction as well as being a reflection of how well individuals know and trust the group.
Health-policy sets patient-centred care well-being and autonomy within the survivorship stage of the disease management. There is increasing recognition of the importance of psychosocial interventions to promote behaviour change, enhance preventative care and to express and process emotion-related responses to cancer e.g.  . Most evidence presented within the review shows an effect of storytelling on an individual’s psycho-emotional well-being through a decrease of negative emotions such as stress or fear. These changes maybe as a direct result of patients being able to share aspects of their lives and reflect upon the impact it has on them.
A person’s mood and emotion may be the most variable aspect related to adjustment and being able to share an illness stories may have a substantial impact on their mood and psychological adaptation. It has been recognised that emotions can be shared through stories which has been identified to bring about physiological changes which are beneficial to the individual   . The current results illustrated a benefit in being able to retell one’s story and this is likely associated with positive changes in psychological well-being  . It may be that through the natural evolution of the teller’s experience the story becomes adapted and modified which enables better psychological adaptation to the situation by the individual  .
Further, it has been identified that the process of sharing information through the medium of a personal story provides a sense of empowerment and that individuals feel liberated by the act of sharing  . Part of the reason for such a positive effect could be the ability to share emotions which influences an individual’s well-being. The current review has highlighted that past quantitative studies may not have considered outcome measures which represent the changes that might be reported by patients participating in storytelling interventions. Storytelling appeared to allow sharing of coping and management strategies between participants, which promotes mental well-being and potentially enhances quality of life. This has been identified in studies of other patients with long term illnesses; including other palliative illnesses  .
・ Listening to stories provides a way of increasing trust between the health care professional and the patient. It allows the health care professional an opportunity to understand the patient rather than to make presumptions based on clinical information in isolation.
・ Storytelling has been identified as a good intervention that can strengthen communities where the intervention takes place and benefit attitudes and behaviours. The current results support this statement and identify a need for further good quality research to develop interventions that are replicable and promote this as good practice.
・ The ability to use stories as a process is readily understood by patients without a requirement to develop new skills at a time when they might have a sense of being overwhelmed.
・ Outcome measures that consider social support, emotions, group social-identity or empathy around interactions may capture changes occurring in the group setting.
・ It may be important to assess the impact of storytelling by considering a simple assessment from practice which can aid the process
・ Heterogeneity of the studies. These included: the cancer type, the geographical location, culture and setting, the type of intervention and ability to identify benefits through different communication means or types of intervention and finally the use of a standardised design and choice of outcome measures all contribute to this.
・ One study Garcia-Schinzari et al.  used children and adolescents aged from 4 - 15 years. However the contribution of this study to the synthesis was minimal except to support the evidence of positive changes in emotions as a result of a story.
・ Focus on the most common findings may limit consideration to specific findings within particular interventions or setting.
・ Restriction on language may focus findings towards particular cultures.
・ Consideration of other stakeholders (carers, spouse, health care professionals) to the process and its impact are not given.
This integrative review has provided an initial indication of the value of storytelling interventions for people with cancer. Whilst there are limitations because of the heterogeneity of the studies, there is consistent evidence that identifies benefit from these interventions and explains a mechanism associated with the process.
Table S1. The demographical and design related information for included studies.
Table S2. The completed 13-item COREQ framework (Soundy et al., 2016) adapted from the 32-item COREQ framework (Tong et al., 2007).
Table S3. Identifying the quality of qualitative research.
Table S4. The summary assessment of the overall risk of bias for quantitative studies.
*Trial denoted by first author. Risk of bias criteria: 1, Selection bias = random sequence generation; 2, selection bias = allocation concealment; 3, Detection/Performance bias = blinding of personnel, assessors and participants; 4, Attrition bias = incomplete outcome data; 5, Reporting Bias = short-term selective outcome reporting; 6, Other bias = potential threats to validity e.g., consideration of a protocol. Levels of risk of bias: H, high risk of bias; U, unclear risk of bias; L, low risk of bias.
Table S5. Synthesis stage 1 showing an example of the qualitative tabulisation.
Table S6. Synthesis stage 2 an example of the descriptive analysis undertaken.
Table S7. Mind mapping.
Table S8. Synthesis stage 3 the mind map.
Table S9. Synthesis stage 4 the thematic development.
Note: 1 = La Cour et al. (2016); 2 = Høybye et al. (2005); 3 = Evans et al. (2008), 4 = Song et al. (2012), 5 = Garcia-Schinzari et al. (2014); 6 = Ando et al, (2018); 7 = Crogan et al. (2008); 8 = Falzon et al. (2015); 9 = Heiney et al. (2012); 10 = Heiney et al. (2013); 11 = Heiney et al. (2015).
Table S10. Synthesis stage 5 the second phase of the thematic development.
Note: 1 = La Cour et al. (2016); 2 = Høybye et al. (2005); 3 = Evans et al. (2008); 4 = Song et al. (2012); 5 = Garcia-Schinzari et al. (2014); 6 = Ando et al., (2018); 7 = Crogan et al. (2008); 8 = Falzon et al. (2015); 9 = Heiney et al. (2012); 10 = Heiney et al. (2015); 11 = Heiney et al. (2013).