Cameron, J.I., Shin, J.L., Williams, D., & Stewart, D.E. (2004). A brief problem-solving intervention for family caregivers to individuals with advanced cancer. Journal of Psychosomatic Research, 57, 137–143.

To evaluate an intervention that was designed to improve the problem-solving abilities of family caregivers* of patients with advanced cancer**. Authors tested whether the intervention (a) enhanced positive problem-solving skills and decreased negative problem solving, (b) enhanced confidence in caregivers’ ability to perform caregiving activities, and (c) decreased caregivers’ emotional distress.

* Caregivers were defined as those unpaid family members assuming responsibility for homecare needs of an individual with advanced cancer.
** Advanced cancer was defined as patients with estimated survival prognosis of three to six months.

A one-hour intervention used explanations and a slide presentation or flip charts of problem solving. The intervention was given to caregivers only once, at the time of recruitment. The intervention involved:

1. Introducing examples of problem-solving techniques from a Home Care Guide  for Advanced Cancer, which was developed by the American College of Physicians to help caregivers COPE (be Creative, Optimistic, Plan, and obtain Expert information).
2. Encouraging the caregivers to use a five-step approach to address problems. These steps were to (a) define the problem, (b) identify when to seek professional help with the problem, (c) identify what they (caregivers) could do to help, (d) consider possible obstacles to their caregiving plan, and (e) carry out and adjust the plan.

• The sample included 34 family caregiver participants.
• Mean age of caregivers was 53.6 years. No SD or range was mentioned.
• Caregivers were 67.6% (n = 23) female and 32.4% (n = 11) male.
• Caregivers had to be those giving care to patients diagnosed with advanced cancer, defined as having an estimated survival prognosis of three to six months.
• To be included in the study, participants had to be 18 years or older, be able to speak and read English, be able to give written consent, and be clear of severe visual, hearing, or cognitive impairments that would interfere with their ability to participate.

• Single site
• Inpatient setting
• Teaching hospital (University Health Network in Toronto, Ontario, Canada): While the patient attended a clinical visit, the research assistant met separately with the caregiver.

• End-of-life care phase
• Elder care; palliative care

A one-sample pretest/post-test design was used. There was no control group. Survey data were collected at baseline (recruitment) and then by phone four weeks after the intervention was done.

• Social Problem-Solving Inventory–short form (SPSI-SF): Used to assess caregivers’ ability to problem-solve. The scale has five subscales representing: (a) positive problem-solving orientation, (b) negative problem-solving orientation, (c) rational problem-solving orientation, (d) impulsivity/carelessness scale, and (e) avoidance style scale.
• Profile of Mood States–short form: Used to assess emotional well-being. This was an established and reliable tool. The scale includes six subscales: (a) tension-anxiety, (b) depression-dejection, (c) anger-hostility, (d) fatigue-inertia, (e) vigor-activity, and (f) confusion-bewilderment.
• Caregiver Assistance Scale (CAS): Used to assess the level of assistance provided by caregivers to patients in activities of daily living, instrumental activities of daily living, and treatment-related activities. This was an established and reliable scale.
• Caregiver Self-Efficacy Scale: Developed by the authors to assess caregivers’ confidence in their ability to perform each of the caregiving activities. The seven-point scale has ratings ranging from one (not very confident) to seven (very confident). The authors averaged the scores, yielding an overall confidence score ranging between one and seven, with higher scores indicating greater confidence. They calculated an excellent internal consistency of this developed scale in the sample (α = 0.90).

The authors provided a good description of how the original sample went from 112 to 34 caregivers, and the basis for attrition or exclusion were reviewed clearly, giving good insight into factors to consider when researching this group of caregivers of patients with advanced cancer. Reasons that led to shrinking of the sample included (a) some patients who were identified with advanced cancer did not need assistance at home, thus they did not have caregivers (n = 7), (b) some caregivers declined participation (n = 34), (c) some participating caregivers did not complete baseline survey because they changed their mind, their patient died, they were not coming back to the hospital in time to complete the intervention, or they had English language difficulty (n = 13), (d) some caregivers who completed the baseline data did not come back later to receive the intervention (n = 11), (e) some caregivers completing the baseline data and the intervention did not complete the follow-up survey (n = 6), and (f) some caregivers were wrongly identified (patients had advanced cancer but responded to curative treatments, or patients were outliers in terms of the length of time they had been diagnosed with advanced cancer) (n = 7).

Other significant findings include the following.

• One significant difference was identified between participant and nonparticipant characteristics: Participants gave care to patients who had cancer for a shorter period of time.
• The mean time between baseline assessment and follow-up assessment was 4.8 weeks (1.28).
• Caregivers showed significant improvement in their positive problem-solving (mean score at baseline [pretest] was 12.3 [3.56] and at follow-up was 13.7 [3.80], p = 0.054); the authors used a p level of 0.06 to judge significance for this finding.
• Caregivers showed significant improvement in their confidence in caregiving activities (mean at baseline [pretest] was 6.1 [0.65] and at follow-up was 6.3 [0.60], p = 0.059); the authors used a p level of 0.06 to judge significance for this finding.
• Caregivers showed significant improvement in emotional tension (mean at baseline was 1.5 [0.97] and at follow-up was 1.2 (0.85), p = 0.024).
• There was a significant positive correlation between caregivers’ reported levels of fatigue at baseline and their level of impulsivity/carelessness problem-solving subscale score (SPSI-SF) at follow-up (r = 0.40, p < 0.02).

Overall, the findings clearly show potential evidence that the intervention affected caregivers positively. Given that improvements were noted even though participants already showed that they have good levels of confidence in their abilities and problem-solving skills at baseline, the findings strongly suggest that the intervention might have even greater potential among those with lower levels of confidence in their caregiving and problem-solving abilities. The intervention influenced one aspect of emotional well-being (tension subscale) and one aspect of problem-solving (positive problem-solving subscale), suggesting that a modified or more comprehensive or frequent intervention may improve other aspects of emotional well-being or problem-solving. Researchers of caregivers of patients with advanced cancer should consider and address the serious challenges that they may encounter in recruiting this group of caregivers.

• The sample was small, with less than 100 participants.      
• Baseline sample and group differences existed.  
• Risk of bias existed due to no control group, no blinding, and no random assignment.
• Key sample group differences existed that could influence results.*
• Subject withdrawals were 10% or greater.
• No information was provided about whether the research assistant who conducted the intervention with participating caregivers had received or had training or expertise in training others in problem-solving. This is an important piece of information because the value or influence of the intervention would be as good as the training that the caregivers received. 
• The study sample was reduced by 50% of caregivers consented to the study.
• * The findings regarding significant improvements in caregivers' confidence and positive problem-solving are borderline, and one can argue that p levels of 0.059 and 0.054 are bordering significance but not significant as the authors interpreted them. The customary p level is < 0.05; with no correlations shown, it is hard to accept the interpretation as significant as the authors relaxed the p level to 0.06.

Nursing care of patients with advanced cancer should be extended to include care of their caregivers, as they are key in providing care at home. This brief problem-solving intervention significantly improved the emotional tension in caregiver participants, suggesting the value of its use in palliative care settings. The fact that a significant difference existed between participants and nonparticipants in relation to the length of time their patients were diagnosed with cancer suggests that the longer caregivers provide care, the less likely they would be to participate in interventions. This suggests that they (a) may have figured out on their own how to manage the caregiving burden, or that they are too exhausted to participate, or (b) need different types of interventions that meet them where they are, either at home or in their locale, rather than require them to drive to where interventions would be conducted. This findings suggest that researchers may need to focus on caregivers of patients with new onset of disease or those who have had the disease for a short period. Any interventions with these caregivers should be set up in a way that decreases the demand on their time and meet them where they are. Thus, there is a need not only to develop short and effective interventions but also to find ways in which such interventions can be delivered/conducted with caregivers at home.