Northouse, L.L., Katapodi, M.C., Song, L., Zhang, L., & Mood, D.W. (2010). Interventions with family caregivers of cancer patients: Meta-analysis of randomized trials. CA: A Cancer Journal for Clinicians, 60, 317–339.

To provide a meta-analysis that examines interventions delivered to family caregivers of patients with cancer in published randomized controlled trials and their effects on multiple caregiver outcomes

• Databases used in the search were CINAHL, Google Scholar, ISI Web of Knowledge, PsycINFO, and PubMed.
• Keywords were family caregiver, cancer patient, spouse, partner, couple, and interventions used in various combinations. In cases where the query resulted in more than 200 titles, additional refinement was done with use of “random assignment” and “randomization.”
• Hand searches of reference lists of related literature reviews were used to complement computer searches. Limits of English language and involving human subjects were used.
• To be included, (a) the intervention had to involve family caregivers alone or with a patient with cancer, (b) the intervention had to be psychosocially, cognitively, or behaviorally oriented, and (c) participants had to be randomly assigned to either an intervention or control arm of the study.
• Studies involving pediatric patients with cancer and pharmacologic interventions were excluded.

• Initial queries provided 403 articles, in which 160 were potentially relevant. Of these, 124 were excluded as per criteria and 7 were excluded because caregiver outcome data were not reported.
• At least two members of the research team read and analyzed each article. Data were extracted using a standard table, and disagreements were resolved through consensus. Data extracted were organized into three domains: illness appraisal factors, coping resources, and quality of life (QOL).
• Effect sizes were calculated using outcome data from the experimental arm hypothesized by the original researchers to be most effective, in those studies that had more than one experimental arm.
• Extracted data were organized into three time frames: baseline to three months postintervention (T1), follow-up from > three months to six months (T2), and follow-up beyond six months (T3).

The final sample of studies used in the meta-analysis was 29, including a total of 3,495 subjects across all studies.

Intervention Types

• 57.1% of interventions were psychoeducational in which the primary focus was to provide information regarding symptom management and physical aspects of care, as well as to direct some attention to emotional and psychosocial needs of patients, caregivers, and marital and family relationship.
• 25.7% of protocols were focused on skills training: development of coping, communication, and problem-solving skills.
• 17.1% of interventions were therapeutic counseling, focusing primarily on development of a therapeutic relationship to address concerns.
• The most common combination of primary and secondary interventions included both psychoeducational and skills training (68.6%).

Appraisal Domain Results

• Eleven studies assessed caregiving burden during the first three months. Effect sizes ranged from -0.12 to 0.62, with a small but significant effect size (g = 0.22).
• In two studies, interventions had a positive significant effect on appraisal of caregiving benefit between three and six months (g = 0.31). Beyond six months, this appraisal showed a larger effect size, but was not significant. (Caregiving benefit was seen as a rewarding experience and opportunity for personal growth.)
• Three studies looked at whether the intervention was effective in addressing caregivers’ information needs. Overall effect size was large and significant (g = 1.36, range = 0.85–1.87).

Coping Resources Domain

• Coping efforts were evaluated during the first three months in 10 studies. After the intervention, overall effect size was moderate but significant (g = 0.47, range = -0.47–1.46).
• In four studies, changes in coping efforts three to six months postintervention showed a smaller but significant effect size (g = 0.20).
• Two studies evaluated coping efforts beyond six months. There was a persistent moderate effect that was significant (g = 0.35).
• In eight studies that evaluated self-efficacy within three months postintervention, overall effect size was small but significant (g = 0.25, range =  -0.13–0.93). This positive significant effect persisted over time.

Quality of Life Domain

• In seven studies that evaluated physical functioning, during the first three months overall effect size was small and not significant. However, interventions were superior to usual care in improving physical functions at three to six months (g = 0.22) and beyond six months (g = 0.26).
• Interventions were seen to be superior to usual care in reducing distress and anxiety, and the effect appeared to last for at least 12 months. Sixteen studies evaluated changes in mental distress and anxiety during the first three months, 11 evaluated these changes at three to six months, and 6 evaluated distress and anxiety beyond six months.
• Interventions were not successful in reducing caregivers’ depression. In all of the studies that evaluated these changes at various time points, overall effects were small and not significant.
• Among 10 studies that evaluated changes in marital and family relationships in the first three months, overall effect size was small but significant (g = 0.20, range = -0.18–0.47). Evaluation beyond three months did not show significant effects.
• Effects on social functions in two studies that were evaluated beyond six months showed an overall positive moderate effect that was significant (g = 0.39). There were no other significant effects seen in other studies that evaluated changes in social functioning.

Effect of Intervention Characteristics

• Coping was found to be significantly influenced by the use of face-to-face and group methods of intervention delivery than those using mixed methods to deliver the intervention.
• Intervention hours and number of sessions were significantly positively correlated with coping outcomes.
• More sessions were associated with more negative (worse) outcomes in the case of caregiver burden, depression, and marital-family relations.
• Interventions that included caregivers alone reported significantly better outcomes in appraisal of caregiving benefit than interventions that also included patients.

Nurses participated in delivery of the intervention in about half of the studies examined. Interventions focused on three broad areas: patient caregiving, marital/family care, and caregiver care, although there was considerable variability in the emphasis of each area in all studies and many interventions focused mostly on the patient, with a peripheral addressing of caregiver needs. The meta-analysis noted there is a consensus that these are essential content for caregiver-focused intervention protocols. Many of the 29 studies used theoretical models to structure their protocols and had plans to assess integrity of the interventions. Interventions were found to not be effective in decreasing caregiver depression, perhaps a result of the type of cancer or high predominance of females in the 29 studies. On the other hand, interventions aimed at improving caregiver coping were more effective, particularly if the intervention included more sessions and lasted a longer time. Coping outcomes improved more often with face-to-face or group-delivered interventions and if the focus was on promoting active caregiver coping and reducing avoidance and denial behaviors. Although the meta-analysis showed a large intervention effect with decreasing caregivers’ need for knowledge (and most studies had a significant educational component), few studies measured caregiver change in knowledge as an outcome.

This meta-analysis of 29 studies, with variant use of theoretical models, offers insight into effective interventions that improve the life of cancer caregivers and those for whom they care for. Despite the diversity in interventions found in the studies, caregivers reported better outcomes in illness appraisal, coping resources, and quality of life following research study involvement. Positive and long-term intervention effects occurred for caregiver coping, self-efficacy, and distress/anxiety outcomes across all studies. Continued emphasis must be placed on weaving theoretical models throughout studies to generate hypotheses, select components of intervention to ensure integrity, and identify outcomes.