Hudson, P.L., Aranda, S., & Hayman-White, K. (2005). A psycho-educational intervention for family caregivers of patients receiving palliative care: A randomized controlled trial. Journal of Pain and Symptom Management, 30, 329–341.

To examine the effects of a psycho-educational intervention on the reported levels of preparedness, mastery, self-efficacy, competence, rewards, and anxiety among caregivers of patients receiving home-based palliative care for advanced cancer in comparison to caregivers receiving standard care

The intervention was psycho-educational in nature and was delivered by intervention nurses through two intervention home visits and one intervention follow-up phone call. In addition to the interaction with the nurse, caregivers were given a caregiving guidebook and an audiotape. The guidebook provided easy-to-access information about caring for a dying person, and the audiotape contained a structured relaxation exercise and featured reflections from other carers who also reviewed self-care strategies. The control group received the usual care (24-hour access to phone advice, emergency visits, and prescheduled home visits from nurses, physicians, and other health professionals).

• The sample (N = 106) included 54 intervention group participants and 52 control group participants.
• Mean age of participants was 60.78 years (SD = 13.98), with a range of 21–84 years.
• The sample was 34.9% male and 65.1% female (as calculated by the reviewer).
• All patients were receiving palliative care.
• Eligibility criteria for the caregivers included a minimum age of 18 years and be living with a patient with cancer receiving palliative care, be English speaking, and be free of psychiatric illness that could impede ability to complete the study requirements.

Caregiver profile: The majority of caregivers (74.8%) were Australian-born, and 74.3% self-identified as Christians. Thirty-one (31.7%) of participants had professional or university education, and 34.6% did not complete high school education. The caregivers were giving care to their patient for a mean length of time of 16.3 weeks (SD = 5). The majority (87.6%) of participants expressed that they were involved in the care because they wanted to, not because they had to, and 43.4% had at least three friends or family members assisting them in providing care. About half of participants had to stop or reduce their work in order to be able to provide care for their patient. Caregivers of patients who were identified as Eastern Cooperative Oncology Group level 4 (completely disabled and confined to bed at all times) were excluded from participating when identified at screening.                                                                                                                                                                    

Patient profile: Patients cared for by caregivers had advanced cancer and were admitted to home-based palliative services within a week before recruitment. The mean age of patients was 69.14 years (SD = 13.46), with a range of 31–92 years. About half (48.1%) were confined to bed almost 50% of the time, and 11.3% were completely disabled and confined to bed all the time. The majority of patients (66.7%) were spouses or partners of caregivers; the rest were either parents of caregivers (16%) or their offspring (7.6%).   

• Multisite
• Home setting through palliative care centers
• Two community (home-based) palliative care services in Melbourne, Australia

• End-of-life care phase
• Pediatrics; elder care 

The study design was a controlled trial with random assignment of participants to intervention versus control group. Data were collected at multiple times using self-reported questionnaires.

• Preparedness for Caregiving Scale (8 items): This scale has good reliability (0.86–0.92). The reliability coefficient in the current study was checked and was above 0.70 according to the authors. The tool was used to assess the perceived level of readiness for the caregiving role among participants.
• Caregiver Competence Scale (4 items): This scale has good reliability (0.74). The reliability coefficient in the current study was checked and was above 0.70 according to the authors. The tool was used to assess the perceived level of adequacy to give care among participants.
• Rewards of Caregiving Scale: The authors selected 10 of the 15 original items. This scale has good reliability (0.77–0.94). The reliability coefficient in the current study was checked and was above 0.70 according to the authors. The tool was used to assess the perceived benefits of giving care among participants.
• Hospital Anxiety and Depression Scale (14 items): Cronbach’s alpha value was above 0.70 but was not specifically stated; however, it was mentioned that this tool has been widely used to screen for depression and anxiety among the terminally ill and that it was used previously with caregivers of patients with cancer. The authors chose 11 as the cutoff score to make determination about participants’ anxiety and depression.
• “Mastery” scale: This scale has 6 items from the original Caregiver Appraisal Scale (Lawton et al., 1989), which has 12 items. The 6-item scale has a poor reliability (0.59–0.61); the reliability coefficient in the current study was not reported, but because it was low, the authors mentioned that this scale was excluded.
• Self-efficacy instrument: This tool has two subscales: (a) caregiver self-care efficacy (Cronbach’s alpha 0.76) and (b) problem-solving self-care efficacy (Cronbach’s alpha 0.83). The tool has good reliability for the two subscales, but the reliability coefficients in the current study were above 0.70. The authors reported that they made changes in two  items to make them appropriate for the study.

The intervention and control group were compared to identify any significant differences between them, but none were detected.

Only 12 participants in the intervention group and 15 in the control group filled out the data questionnaires at all three time points. However, the number of participants who completed the data at times 1 and 2 was 75 (35 in the control and 40 in the intervention group), and at times 1 and 3 was 45 (25 in the control and 20 in the intervention group).

Findings of time 1 to time 2 (n = 75):

• No main effect for group was noticed for preparedness, and no effect for interaction between group and time was noticed either. There was a trend (not significant) for main effect within subjects over time F(1,73) = 3.36, p = 0.071 (ƞ2 = 0.044), indicating that preparedness increased over time.
• No main effect for group was noticed for rewards. There was a significant interaction between group and time, F (1,73) = 8.44, p = 0.005 (ƞ2 = 0.10); this indicated that while the control group showed a decreased sense of rewards between time 1 and time 2, the intervention group showed an increased sense of perceived rewards between the two assessment times. There was a trend (not significant) for main effect within subjects over time, F (1, 73) =  2.84, p = 0.096 (ƞ2 = 0.37).
• No significant effects or trends were noticed for self-efficacy, competence, and anxiety.

Findings of time 1 to time 3 (n = 45):

• No significant main effects between groups were noticed for preparedness, rewards, anxiety, or competence between time 1 and time 3.
• There was a significant main effect within-subjects over time, F (1, 42) = 4.94, p = 0.032 (ƞ2 = 0.11) for competence. This means that participants, irrespective of group, had increased overall sense of competence at time 3 compared to time 1.
• There was a main effect trend (not significant) within-subjects for time noted for anxiety F (1, 43) = 2.77, p = 0.10 (ƞ2 = 0.06), indicating that there was decreased anxiety among participants, irrespective of group, between time 1 and time 3.
• The interaction between group and time was significant in rewards F (1, 43) = 4.68, p = 0.036 (ƞ2 = 0.098). Comparing the means showed that there was decreased average rewards scores among control group participants between time 1 and time 3 and an overall increase of these scores among members of the intervention group. Furthermore, there was a main effect trend (not significant) for “time” within-subjects for rewards F (1, 43) = 3.22, p = 0.08 (ƞ2 = 0.07); this indicates that there was increased sense of reward among participants, irrespective of group, between time 1 and time 3.

Although the study showed that the intervention did not show much effect on most of the assessed caregiver outcomes, it did reveal an unexpected finding. That is, participants in the intervention group showed increased perceived sense of reward over time compared to participants in the control group. The findings suggest that the increased sense of perceived reward may play a role as a coping resource and could potentially have a buffering effect from caregiving burden.
 

• The sample was small (< 30 participants).
• Risk of bias existed (sample characteristics).*
• Findings were not generalizable.* 
• The intervention was expensive, impractical, or required training.*
• Subject withdrawals were ≥ 10%.
• Other limitations/*explanation: The study was done in Australia, which runs a different health system than the United States, and this could have implications. It remains unclear whether the lack of evidence for intervention effectiveness is due to the small numbers of participants or due to lack of variability among participants to start with. It is possible that those enrolled in the intervention and control groups were highly functional and independent individuals who did not need help; the fact that they chose to participate could indicate that they were people who were resourceful and willing to learn more about the caregiving experience. It is possible that those who declined participation are the ones who most needed an intervention.

Attrition of participants in caregiving studies seems to be a significant factor to consider and one that is hard to control.

The sense of reward or maintaining positive emotions by caregivers of patients in palliative care could be used as a resource to help caregivers cope with caregiving burden.

Although the study did not show much impact of the intervention, it supports the notion that giving information to caregivers is important, especially when given at selected times that are based on the caregivers’ needs during the caregiving experience. The “proper” timing for interventions, as well as the “proper” intervention with caregivers, are areas open for research. The need remains to identify the nature of interventions that are needed by caregivers. These needs seem to be variable depending on the healthcare system in which they are delivered, the specific characteristics of the caregivers, and the settings in which potential interventions are delivered.