A systematic review of psychosocial interventions to improve quality of life of people with cancer and their family caregivers

Abstract Aims To review the characteristics and effectiveness of psychosocial interventions on quality of life of adult people with cancer and their family caregivers. Design A systematic review using PRISMA guidelines. Methods Seven databases were searched from 2009–2019 using key terms. Included studies were assessed using the Quality Assessment Tool for Quantitative Studies. Results 1909 studies were retrieved with 12 studies included, involving 3,390 patients/caregivers. Interventions aimed to improve communication, behaviour change and setting short‐term goals. Duration of interventions varied from 4–17 weeks. Highest benefit was gained from telephone interventions. Interventions based on interpersonal counselling appeared more effective than other approaches. Studies predominantly focused on psychological, physical and social domains of quality of life. Spiritual well‐being received relatively little attention. A paradigm shift is needed to develop psychosocial interventions that incorporate spiritual well‐being. More research is needed in developing countries.

coping. The physical domain relates to health and physical performance (e.g. pain, fatigue, incontinence). The social domain generally relates to the effects of the disease or caregiving on social and personal roles and perceptions of social support (Yanez, Thompson, & Stanton, 2011). Spiritual well-being includes coping with life stressors and attainment of self-transcendence (Counted, Possamai, & Meade, 2018). Spirituality differs from religiosity which usually refers to institutionalized religious beliefs (Counted et al., 2018).
In the last 10 years, researchers have developed various psychosocial interventions to improve the QoL of both adult people with cancer and their family caregivers (Ferrell & Wittenberg, 2017;Hu, Liu, & Li, 2019). However, most research has predominantly focused on psychological, physical and social domains (Gabriel & Mayers, 2019;Titler et al., 2017). Furthermore, several published systematic reviews and meta-analyses (Badr & Krebs, 2013;Fu, Zhao, Tong, & Chi, 2017;Son, Son, Kim, & Lee, 2018)  Spiritual well-being has received relatively little attention despite its influence on QoL (Skalla & Ferrell, 2015). Distress over spiritual concerns has been found to be prevalent among people with cancer (Drummond & Carey, 2019). Badger et al. (2011) who examined the effectiveness of a brief telephone psychosocial intervention with seventy-one prostate cancer survivors and their family partners found spiritual well-being was an important predictor of QoL. Other studies report that spiritual well-being contributes to better health outcomes (Lichter, 2013;MacKinlay & Burns, 2017). Limiting psychosocial interventions to psychological, physical or social domains may provide inadequate information on the applicability and effect of such interventions.
Most intervention studies have been conducted in developed countries such as the United States of America (USA), Australia, Canada or France. Although the overall incidence of cancer is higher in developed countries, total cancer-related mortality is significantly higher in developing (low-and middle-income) countries, where 75% of cancer deaths occur and the number of cancer cases is rising most rapidly (Prager et al., 2018). In African society for example, cancer is still considered incurable (Wallace, Bos, & Noble, 2018). Although people with cancer and their families are reported to seek positive meaning for their circumstances through spiritual endeavours (Kiyancicek & Caydam, 2017), the extent to which this is the case is unknown. To address this apparent gap and inform future work in low-and middle-income countries, there is a need to critical appraise the effectiveness of possible interventions on the four domains of QoL for both people with cancer and their family caregivers.

| AIM
The aim was to review the characteristics and effectiveness of psy-

| ME THODS
The review process was guided by the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) recommended protocols (File S1). The PRISMA checklist is an evidence-based tool for evaluating the title, abstract, methods, results, discussion and findings. It can be used for evaluating randomized controlled trials and reporting systematic reviews for non-heterogeneous research (Liberati et al., 2009). The review protocol is registered with PROSPERO CRD42020144563.

| Study selection
The selection process for eligible studies was based on specified inclusion/exclusion criteria (Table 1). All references were downloaded into Endnote version 9, and titles were screened in Endnote by two reviewers. A second check of all retrieved data was undertaken by the main author and second reviewer. All studies were screened first according to title and then abstract and reviewed by two independent reviewers for inclusion. Disagreements were resolved through consensus.

| Quality assessment of included studies
Included studies were assessed for quality by three researchers using the quality assessment tool for quantitative studies (QATFQS) including randomized and non-randomized designs (Thomas, Ciliska, Dobbins, & Micucci, 2004). The QAFTQS contains eight components of study quality related to sample selection, study design, identification and treatment of confounders, blinding of outcome assessors and of participants, reliability and validity of data collection methods, withdrawals and dropouts, and intervention integrity and analyses.
Each component is rated as strong, moderate or weak. Studies with at least four strong ratings are considered strong, less than four strong ratings and one weak rating are considered moderate and two or more weak ratings are considered weak (Thomas et al., 2004).

| Data extraction
Relevant information was extracted and recorded in a spreadsheet using Microsoft Excel software. A second check of all retrieved data was undertaken by the main author and second reviewer. Twelve full-text articles were extracted and tabulated. Extracted data included general information (first author, place of study, publication year, aims and theoretical approach), methodological information (study design, sample context, response rate, follow-up, retention, therapy type, intervention delivery/dosage and intervention/control group content and measurement tools) and results of the study (main outcomes).

| Data analysis
According to the Cochrane Collaboration (Deeks, Higgins, & Altman, 2008), the analysis of findings can be presented as a narrative such as a summary with a discussion of study characteristics and findings. The high degree of methodological diversity, statistical and clinical heterogeneity of included studies did not afford an opportunity to pool results and conduct a meta-analysis. Thus, a narrative account is presented as an overview of psychosocial interventions for adult people with cancer and family caregivers. Descriptive statistics describe participant characteristics and results.  (Table 1).

| Summary of included studies
Results of the quality appraisal process are presented according to PRISMA reporting guidelines. This includes study selection, study characteristics, risk of bias in individual studies and results of individual studies.
There were no studies from developing countries.

| Risk of bias
As assessed by the QATFQS checklist (see Table 2), risk of bias showed mixed quality: one study (Northouse et al., 2013) was ranked as strong, two studies were evaluated as moderate quality (Clark et al., 2013;Shaw et al., 2016) and nine studies were weak (Badger et al., 2011;Badger, Segrin, Hepworth, et al., 2013;Belgacem et al., 2013;Kayser et al., 2010;Meyers et al., 2011;Mosher et al., 2018;Northouse et al., 2014;Titler et al., 2017). Poor quality related to two issues, selection bias and blinding. Common weaknesses of the studies were confounders such as high withdrawal rate or drop out and lack of a control group (Northouse et al., 2014;Titler et al., 2017). Three studies used randomization but did not describe the process (Badger, Segrin, Hepworth, et al., 2013;Mosher et al., 2018). Although most studies were assessed as weak, their consistent findings contribute to our understanding of research in this area and are included in the results.
Interventions included interpersonal counselling and health education, skills training and coping skills, family connection interventions, the FOCUS (Family involvement, Optimistic attitude, Coping effectiveness, Uncertainty reduction and Symptom management) programme, or COPE (Creativity, Optimism, Planning and Expert information) programme (Meyers et al., 2011) targeting people with cancer and their family caregivers (as described in Table 3).
Duration of interventions ranged from 4-17 weeks, and session length varied from 29-120 min. Most interventions were delivered in eight sessions; however, the number of sessions ranged from three (Meyers et al., 2011)-nine (Kayser et al., 2010). Delivery of the interventions involved telephone, face-to-face and web-based.
Lazarus and Folkman's transactional model of stress and coping evaluates the processes of coping with numerous stressors, such as symptoms, treatment, work stress, family stress and the challenge of caregiving (Lazarus & Folkman, 1987). This has been widely advocated as a useful framework to guide cancer research (Ravindran, Shankar, & Murthy, 2019). Three studies were underpinned by CBT principles. According to Daniels (2015:54), CBT is a "psychotherapeutic approach that emphasizes the significance of how thinking affects the feelings." CBT may include cognitive restructuring, relaxation and skills training among other modalities. CBT focuses on the present and aims to empower individuals to change their responses to circumstances. Interpersonal theory proposes an evidence-based, life event and affect-focused treatment approach based on the premise that distress does not occur in a social vacuum, but is influenced by and affects the patient's psychosocial environment. The goal of interpersonal therapy is to help individuals solve a crisis in role functioning or social environment, which leads to the improvement in QoL (Blanco et al., 2019). This technique has been established as feasible for treating major depressive disorder in patients with breast cancer (Blanco et al., 2014).
The FOCUS programme is a multicomponent intervention that addresses family involvement, family communication and working on problems as a group. Maintaining an optimistic attitude aims

TA B L E 3 Summary of systematic review of studies on psychosocial interventions
Author ( to help the family to keep hope and focus on achieving short-term goals. Coping effectiveness and uncertainty reduction provide strategies on how to get information and live without doubt; symptoms management emphasizes self-care strategies to control symptoms and experiences (Tabrizi & Alizadeh, 2018).
The COPE programme (Meyers et al., 2011) is a supportive educational programme, designed to teach cancer patients and family caregivers problem-solving skills to help manage symptoms and other concerns. It focuses on using creativity (viewing problems as challenges that can be overcome), optimism (focusing on the positive, yet being realistic), planning (developing a sound plan to address problems) and expert information (finding and learning from trustworthy sources) to overcome problems (Tofthagen & Chesak, 2019).
Other psychosocial interventions provided a psycho-educational programme (Belgacem et al., 2013) and a structured multidisciplinary programme (Clark et al., 2013) to assist patients and families to improve their skills in meal support, nursing care, welfare care or symptom management.

| Comparison group
Six studies used randomized controlled trial designs to compare outcomes of the intervention group to those receiving standard or usual care (Belgacem et al., 2013;Clark et al., 2013;Meyers et al., 2011;Shaw et al., 2016), or a standard social work service (Kayser et al., 2010). Northouse et al. (2013) used three groups with two groups exposed to brief or extensive versions of the FOCUS programme and one receiving usual care. Four studies compared two groups exposed to different interventions with no control group.
These studies compared telephone/video interpersonal counselling or telephone health education/health education attention conditions (Badger et al., 2011;Badger, Segrin, Hepworth, et al., 2013;, or a telephone-based coping skills intervention compared with a "peer-assist" plus coping skills intervention (Mosher et al., 2018).

TA B L E 4 Quality-of-life domain outcomes of included studies
Author ( There was no effect noted on the caregivers There was no effect of the intervention noted on patients and caregivers There was no significant effect noted on patients and caregivers Kayser et al. (2010) Dyads had higher means than the control group at time 2 and time 3, but no statistically significant Dyads had higher means than the control group at time 2 and time 3, but no statistically significant Dyads had higher means than the control group at time 2 and time 3, but no statistically significant differences between the two arms Dyads had higher means than the control group at time 2 and time 3, but no statistically significant  (Clark et al., 2013). Domains of QoL were also assessed, with a total of 16 different assessment tools being used to assess specific domains (see Table 3 for full details).

| Psychological/emotional domain
All studies assessed psychological/emotional well-being, with four studies reporting a statistically significant improvement in this domain     TA B L E 4 (Continued)  survivors p < .001, partners p < .01; Badger, Segrin, Hepworth, et al. (2013)

| Spiritual domain
Six studies evaluated changes in spiritual well-being postintervention. A significant improvement in spiritual well-being of both patients and their family caregivers was identified by Badger and colleagues (Badger et al. (2011) p < .01). Other studies identified significant improvement among survivors  p < .01) or partners (Badger, Segrin, Hepworth, et al. (2013) p < .001). No significant change in spirituality was observed in three studies (Clark et al., 2013;Meyers et al., 2011;Mosher et al., 2018).

| D ISCUSS I ON
The catalyst for this review was the necessity to evaluate the char- others. The relevance of interventions based on interpersonal theory is relevant to spirituality, as such approaches can contribute to an improved sense of security and sense of self (Buechler, 2018). A previous study showed that interventions based on theories of interpersonal therapy for cancer patients and their family caregivers contributed to improvements in QoL (Badger, Segrin, Meek, Lopez, & Bonham, 2006). In addition, our findings support the conclusions of a review comparing interpersonal psychotherapy, supportive therapy and CBT by Evans (2009) which showed that interpersonal psychotherapy was most effective.
Although CBT is a beneficial therapy option for patients with various forms of cancer (Brothers, Yang, Strunk, & Andersen, 2011), the current review found either no significant improvement across QoL domains (Clark et al., 2013;Kayser et al., 2010) or a significant decline in psychological and social well-being (Meyers et al., 2011).
These results contradict those from a recent meta-analysis which demonstrated the efficacy of CBT on QoL and psychological health of survivors/patients with breast cancer (Ye et al., 2018). Similarly, Solaimani Khashab, Ghamari Kivi, and Fathi (2017) reported a positive impact of group CBT on improving spiritual well-being of bereaved persons.
In the current review, although three of the eight studies based on FOCUS, CBT and stress-coping measured spiritual well-being, none reported significant improvement. This could be because CBT was developed from empirical studies that did not consider faith as a variable (Carlson & Antonio, 2016). Similarly, it could be that CBT is individually focused and has less applicability than dyad-based, interactional interventions for people with cancer and their family carers.
Regarding methods of delivery, the review found those interventions delivered by telephone had a positive effect compared with other methods of delivery. This result is comparable with that of Cox et al. (2017) who suggested that the impact of telephone interventions on outcomes was far greater than those using Internet delivery methods. In contrast, findings of our review differed from a recent systematic review and meta-analysis of psychosocial interventions on the QoL of patients with colorectal cancer (Son et al., 2018), which highlighted that face-to-face intervention methods as compared with telephone-based approaches had a significant effect on QoL. Given these various findings, further research on delivery of interventions is needed especially in developing countries where there may be limited access to telephone or Internet. It may be argued that face-to-face interventions appear to improve therapeutic relationships, thereby leading to an increase in patients'/caregivers' degree of adherence to treatment protocols and recommendations by healthcare providers (Bombard et al., 2018). In resource-poor clinical environments, it may be prudent to use face-to-face methods as the key delivery component of psychosocial interventions for people with cancer and their family caregivers (Hurt, Walker, Campbell, & Egede, 2016 (Clark et al., 2013). A similar result may occur when studies are conducted in developing countries, whereby there may be little improvement in each domain due to poor medical resources and burden of care, but overall QoL may be significant. These conflicting results suggest the need to measure and report individual domains and overall QoL to ascertain the true effects of interventions.
The cultural context where QoL is measured is also a key issue.
What is considered "a good life" varies between individuals and different societies and cultures. It may be misleading to take QoL concepts developed in one cultural context and apply them to other cultures or even in different ethnic communities (Zhang et al., 2016).

| Strengths and limitations
The strengths of this review include a rigorous literature search, use of a validated methodology and use of two independent reviewers during data evaluation, data extraction and synthesis. It is conceiv- Most studies recruited participants from white ethnicities, ignoring the possibility that race, culture and beliefs may influence individuals' QoL. Therefore, the themes and conclusions are mainly representative of participants from those developed countries and may differ from those of the developing countries. Furthermore, many important studies that aimed to improve aspects relevant to QoL but did not measure this construct specifically were not included in the current systematic review. Finally, only half the included studies gave consideration to the spiritual well-being of people with cancer and their family caregivers. This is an important gap in psychosocial intervention studies. It is possible that over time, more studies measuring spirituality will be available and results of future reviews may differ from this one.

| RELE VAN CE TO PR AC TICE
This systematic review confirms evidence that psychosocial interventions offered to people with cancer and family caregivers can contribute to positive effects on important QoL outcomes. Although effects on QoL domains were mixed, the findings show number of potential implications for clinical practice, research and education.
First, health workers need to be aware that people with cancer and family caregivers tend to respond to cancer and its treatment as a unit; hence, patient/family caregivers should be considered as a dyad when planning care protocols. The findings from this review and those of others (Aubin et al., 2017;Treanor. et al., 2019) suggest that adjustment to cancer is a family affair, not only because both patients and family carers have legitimate need for support, but also because role adjustment problems in the family will negatively affect the long-term adjustment of the patient.
The review concluded that interventions based on interpersonal therapy were more effective than other therapies. It is important for healthcare professionals to promote a therapeutic relationship that encompasses caring and supportive behaviours towards patients and

ACK N OWLED G EM ENTS
The authors would like to thank the health librarians, Griffith University, Queensland, Australia, for their assistance in retrieving relevant studies, which made this review possible.

CO N FLI C T O F I NTE R E S T
The authors declare no conflict of interest.

AUTH O R S ' CO NTR I B UTI O N S
IG, DC and EC: Study design and PRISMA guideline, and manuscript preparation; IG and EC: Database search, study selection, quality assessment and data extraction.

E TH I C A L A PPROVA L
None.