The Effectiveness of Psychosocial Interventions with Cancer Patients: An Integrative Review of the Literature (2006–2011)
Table 4
Supportive interventions. Boesen et al. [39] conducted a regression analysis using an intervention that could be classified as both a supportive and group intervention.
Author (s)
Objective and hypothesis listed
Clear eligibility criteria/date study conducted
Settings and sample specified
Interventions clearly described such that replication is possible
Outcome measures specified
Sample size/power analysis
How randomization was accomplished/who randomized
Blinding
Drop out reasons specified
Limitations discussed
Funding source/trial registry
Findings
Boesen et al., [39]. The title specified randomization.
A hypothesis was not included.
Clients age 18 to 70 with cutaneous malignant melanoma of stage I and II. Recruitment dates were listed
Plastic surgery departments throughout Denmark. 262 patients participated.
Details of the intervention and the control group are described in another publication [43]. The intervention could not be replicated without referring to that article [43].
POMS, dealing with illness coping inventory, the Barret-Lennard Relationship Inventory and the Marlow-Crown scale of Social Desirability.
A power analysis was not included. 51 individuals who declined to participate in the randomized study completed in 2005 [44] were asked to fill out a baseline questionnaire and compared to 259 individuals who participated in that 2005 randomized study.
Details were not provided about the type of randomization or the random allocation sequence.
No mention is made of whether participants, providers, or data collectors were blinded.
Reasons for nonparticipation were given.
Only 40% () of those who declined completed a baseline questionnaire. Who provided the group interventions and how or if treatment fidelity was monitored are not described. Ethnic and racial backgrounds were not reported.
The Danish Cancer Society funded the grant but since it was a secondary analysis, not a randomized trial itself, no mention is made of inclusion in a registry.
Higher socioeconomic status, higher coping, lower social support, and lower mood predicted participation rather than tumor characteristics. Social desirability did not predict participation.
Fukui et al., [40]. The title specified randomization.
A hypothesis was not include.
Clients over 18 with gastric, colorectal, or breast cancer not in an advanced stage. Recruitment dates were provided.
One Japanese cancer center. The background of the nurses was well described. 89 patients (51%) participated.
The intervention is not described in sufficient detail to replicate the study. A waiting list control was used.
Hospital Anxiety and Depression Scale, Mental Adjustment to Cancer Scale administered at 1 week, 1 month, and 3 months after diagnosis.
A power analysis was not included.
The method of randomization is not described in detail. Both nurses in the intervention group who received training and in the control group participated in randomizing patients. Clients were randomly assigned to trained or nontrained nurses and usual care treatment.
Clients were blinded to assignment. Nurses were not blinded.
Reasons for dropouts were discussed. No differences were found based on RN or MD characteristics. RN interviews with intervention clients were longer (22 minutes versus 18 in the control). No mention is made of intent-to-treat analysis.
Only 4 nurses were assigned to the intervention and 4 to the control group. In 24% of interviews nurses did not follow the second step of the intervention, 29% did not follow the 3rd step and 22% did not follow the 4th step indicating poor treatment adherence. Treatment fidelity was determined by audiotaping the intervention group only. Neither interrater reliability nor intrarater reliability was confirmed. The influence of physician communication was not assessed
The Japan Society for the Promotion of Science and a Pfizer Grant supported the study. No mention is made of whether a trial registry was used.
Group by time decreases in psychological distress, increases in fighting spirit, and decreases in fatalism occurred in the intervention group.
Manne et al., [35]. Randomization was not indicated in title.
The effects of coping and communication enhancing interventions on depressive symptoms would be primarily mediated by the process that were encouraged by increasing expression of emotions, increasing attempts to understand emotional reaction and improvements in self-esteem.
Eligible patients were women undergoing active medical treatment diagnosed with primary gynecological cancer, 18 or older with a Karnofsky performance status of 80 or greater living within a 2-hour older that were English speaking. Recruitment dates were listed.
10 hospitals in the northeast United States. 353 women participated.
The Coping and Communication Enhancing Intervention (CCI) was not described in sufficient detail in this publication to replicate that intervention.
The Beck Depression Inventory, COPE, Emotional Expressivity Questionnaire, the Positive Emotion Scale, Emotional Expression Scale, Rosenberg Self-Esteem Inventory evaluated at baseline, 3, 6, and 9 months.
A power analysis was completed
The statistician created the randomization scheme based on the baseline Beck Depression score and the research assistant assigned participants.
Research assistants, participants, and interventionalists were not blinded.
Reasons for drop-outs were not detailed in this publication.
Changes in depressive symptoms may have predated the intervention. The sample consisted primarily of Caucasian women. 58% declined to participate. Therapist background and training and treatment fidelity were not discussed.
No discussion of trail registry although it is likely as the study was funded by the National Cancer Institute.
The coping and supportive counseling interventions both had a beneficial effect on depressive symptoms. There was no impact of any intervention on cancer-specific distress.
Powell, et al., [41]. The title specified randomization.
The study purpose was discussed but the hypothesis was not mentioned.
The only eligibility criteria listed were having attended a gynecological cancer clinic for the first time. It was unclear if the women were actually diagnosed with cancer. Recruitment dates were specified.
A gynecological cancer clinic in San Francisco. 100 women with gynecological cancer were eligible. 43 in the control group completed questionnaires, 21 women received the intervention. The sample was primarily Caucasian (71%).
The intervention, a one-time meeting with a psychologist or a control group was not described in enough detail to replicate.
Functional Assessment of Chronic Illness Therapy, Version 4; POMS, Index of Coping Responses and satisfaction with the clinic.
The sample size was small () and no power analysis was provided.
Randomization (random numbers in sealed envelope) was discussed but not who was responsible for the randomization.
Neither the patients nor the psychologist were blinded.
Reasons for nonparticipation and drop out were discussed. An intent-to-treat analysis was performed.
Only a small sample of 21 women actually received the intervention. Treatment fidelity was not discussed.
No sources of funding or trial registry were mentioned.
Women who received the intervention showed greater decreases in anxiety, depression, and distress as well as increasing physical, emotional, functional, and overall well-being.
