Introduction

Providing physical and emotional support to cancer patients [1] often takes a toll on family caregivers [2]. Because caregivers’ quality of life (QOL) and emotional symptomatology may potentially affect the quality of care of cancer patients [3], it is essential to develop supportive care services for caregivers. Psychosocial interventions for caregivers often fall into three categories: psychoeducation, skills training, and therapeutic counseling [4]. While half of existing interventions consider the caregiver–patient unit [5], it has been suggested that individualized caregiver-specific interventions may be more efficacious [6]. Interventions should also be tailored to specific populations; much research on this, however, has been done exclusively in the United States [1].

The COPE (Caregivers of cancer Outpatients’ Psycho-Education support group therapy) intervention was aimed at family caregivers of patients receiving ambulatory cancer care in Singapore. Developed by an interdisciplinary team of psychiatrists, psychologists, and oncology professionals, and based on the principles of Brief Integrative Psychological Therapy (BIPT) [7, 8], it sought to improve caregivers’ QOL by simultaneously targeting psychoeducation, skills training, and supportive therapy. The content of these four weekly sessions were tailored in response to the concerns of caregivers of oncology patients in Singapore [9]. Facilitated by a clinical psychologist, the first 10–15 minutes of each hour-long session was set aside for didactics (i.e., psychoeducation and skills training), while the remaining time was devoted to supportive group therapy. The specific methods of the COPE intervention and trial protocol are described elsewhere [10, 11].

The present study is thus an evaluation of this pilot four-week brief psycho-education support group intervention using a mixed-methods framework, given the nascence of the COPE intervention. The quantitative component focuses on evaluating the intervention in terms of participant outcomes of QOL, depression, and anxiety. The qualitative component (involving interviews) pertains to the process of the intervention, through obtaining participants’ feedback in order to inform improvements in service delivery.

Methods

Participants and procedures

This study received ethics approval from the National Healthcare Group Domain-specific Review Board (Reference: 2013/00662) and funding from the National University Cancer Institute, Singapore (NCIS) Seed Fund. Participants were recruited at NCIS outpatient clinics if they met inclusion criteria: (a) between 21 and 74 years; (b) willing to attend hour-long weekly programs for four weeks; (c) able to understand, speak, and read English; (d) a family member living with and providing care and support for the patient. No gender, ethnicity, cancer site or type restrictions were imposed given the exploratory nature of the study. All participants provided written informed consent.

Because we did not want to deny care to participants who were interested and able to attend the intervention, a quasi-experimental design was chosen whereby consenting participants were allocated into two study arms based on their availability to attend the intervention (non-randomized groups). Participants interested and able to attend the next intervention session formed the intervention group. Participants interested but unable to attend the next intervention session were put on the waitlist until they could eventually attend, and formed the control group. However, none of these waitlisted participants were able to attend eventually. Therefore, upon study conclusion, there were 56 participants in the intervention group, and 41 participants in the control group. The demography of the intervention and control groups are presented in Table 1. There were, on average, 6 participants per group.

Table 1 Demography of the intervention and control groups
Full size table

Of the participants in the intervention group, 24 (43%) missed at least one session due to their caregiving duties; these participants completed missed sessions in the next run of the intervention. Only 32 (57%) completed the scheduled intervention within four consecutive weeks. The intervention group took an average of 6.06 weeks to complete the program (median = 4; range = 4 to 20). Figure 1 presents the CONSORT diagram for the study.

Fig. 1
figure 1

CONSORT diagram outlining the participant flow

Full size image

Measures

Intervention and control participants completed (a) a sociodemographic questionnaire; (b) Caregiver QOL – Cancer (CQOLC) scale [12], a 35-item instrument on a five-point Likert-type scale (scores ranging 0–140) assessing caregivers’ QOL on the domains of disruptiveness, burden, positive adaptation and financial concerns; and (c) Hospital Anxiety and Depression Scale (HADS) [13], assessing symptoms of depression and anxiety in the past week (maximum score of 21 per anxiety/depression subscale).

Participants completed the questionnaires before the intervention commenced (T1), and immediately post-intervention (T2). A subgroup of 20 participants were invited for an additional 30-minute semi-structured interview at T2 with a research assistant.

Data analysis

Quantitative analyses were conducted using SPSS 22 (Chicago, IL), with significance levels set at .05. As missing data were infrequent (1%) and completely at random (Little’s MCAR test ps > .05), no data imputations were employed [14]. No significant demographic case-mix differences between the intervention (including four-week intervention subgroup) and control groups were found (all chi-square and Fisher’s Exact tests ps > .05). As such, to determine if the four-week intervention group improved significantly over the control group, between subjects repeated measures analyses of variance (ANOVAs) were employed. To determine if there were significant differences in baseline scores between the intervention and control groups, independent samples t tests were conducted. To determine if there were differences across time in the intervention and control groups, one-way repeated measures ANOVAs were employed.

Qualitative analyses were conducted using NVivo 9 (Doncaster, Australia), with post-intervention interviews transcribed and coded independently by two individuals. An inductive, data-driven approach was first used to identify codes arising in the dataset [15], following which iterative coding and constant comparison [16] was used to develop a framework of codes to describe the data. These were reviewed amongst three coders (agreement α = .85) until a consensus was achieved.

Results

Quantitative analyses

There were no significant interactions between the intervention and control groups on all measures of interest (all ps > .05). At baseline, intervention group participants reported impaired QOL and greater depressive and anxious symptoms than control group participants (ps < .05). QOL, depression, and anxiety remained stable in the control group across time (ps > .40), but there were significant post-intervention improvements for intervention participants in overall QOL (p = .053), and specifically in the domain of burden (p = .034). Figure 2 graphically presents the QOL total and domain scores for both the intervention and control groups.

Fig. 2
figure 2

Change in CQOLC scores over time for the control and intervention groups

Full size image

Qualitative analyses

Five main themes surfaced from the post-intervention interviews. Participants highlighted the importance of group processes in creating a central pool of knowledge and a safe space that would allow them to diffuse their emotional turmoil through opportunities for self-expression. Concrete gains were achieved through normalization of such caregiving experiences and the experiential learning of coping skills, which they perceived to have also benefited their care recipients. The group also allowed participants to reflect and reappraise their own behavior and situations through making comparisons with other caregivers. Facilitation by the psychologist was helpful; psychoeducation and skills training were not only informative, but also validated their experiences and feelings. Finally, there was feedback that the support group format could be modified to include the medical aspect of cancer care and accommodate caregivers’ schedules, through employing other media forms such as an online platform or informal meetings. Table 2 presents participants’ quotes from the interviews, and the themes.

Table 2 Themes emerging from the qualitative interviews and the supporting quotes
Full size table

Discussion

Findings from this study suggest that the COPE intervention was particularly helpful in reducing perceived burden. Understandably, QOL domains that are beyond participants’ loci of control, such as disruptiveness and financial concerns, were not amenable to change, and the stability of the domain of positive adaptation may be a result of the short time frame in which the intervention was conducted and evaluated. In addition, participants’ anxiety and depression scores were below subsyndromal cut-offs; the floor effect associated with the HADS may have hampered clear measurement.

However, interviews revealed that participants found the intervention helpful, especially in providing a safe space to ventilate and for social comparisons. Existing literature has described social comparisons as being upward, downward or lateral comparisons with peers, with effects on self-perception, self-evaluation, affect, and behavior [17]. However, because the impact of social comparisons are moderated by personal factors such as levels of self-esteem and perceived control [18], the actual effect among intervention caregivers here may vary across individuals.

Nonetheless, given the setting of an Asian cultural context, support group interventions may be particularly beneficial in addressing the psychosocial needs of caregivers, as compared to individual psychotherapy. Cultural norms may discourage Asian caregivers from seeking external help; there is both a stigma associated with utilizing mental health services [19], as well as a reluctance to discuss personal struggles, such as caregiving difficulties, with non-family members [20]. However, Asian caregivers appear willing to open up and share experiences with other caregivers whom they perceive as being in similar situations [20], making these peer group interventions a suitable strategy to support Asian caregivers.

Limitations

The non-randomization study design resulted in the creation of non-equivalent groups, as reflected by the higher levels of distress at baseline among the intervention group than the control group. This may have compromised the comparability of both groups and the accuracy in evaluating the effects of the intervention. Nonetheless, we opted for this quasi-experimental design as our priority was to ensure that caregivers had access to the program if they were willing to avail themselves to attend.

Future recommendations

Recommendations for similar interventions in the future can be gleaned from the experiences of this pilot study. Firstly, while control participants were never available for the intervention, possibly due to hectic schedules, it is likely, and heartening to note, that participants who recognized that they needed help were willing to make time for the intervention, possibly explaining for their poorer emotional wellbeing at baseline. Hence, an enhanced screening of needs, and greater flexibility and individualization of the program may be required, in order to identify and offer timely support for caregivers who are in greatest distress.

Secondly, while the program was provided at no cost to caregivers, which may have enhanced its appeal, it may not be sustainable in the long run. A possible solution could be the integration of such a program into routine service delivery.

Finally, caregivers had expressed interest for additional sessions; the COPE intervention, which combines psychoeducation, skills training, and therapeutic counseling, could be further developed through transferring didactics to an online platform, with time dedicated to clarifying doubts and supportive expressive therapy.

While further research is essential in replicating and expanding upon the findings of this pilot service in an Asian setting, the hope is that such a program could eventually be effective in Singapore in supplanting sessions with mental health professionals for family caregivers of cancer outpatients.