The Effectiveness of Psychosocial Interventions with Cancer Patients: An Integrative Review of the Literature (2006–2011)
Table 6
Telephone Assisted Interventions.
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
Kravitz et al. [34]. Randomization was not indicated in title.
The intervention would improve pain severity, pain-related impairment, and functional status.
English-speaking adults age 18–80 with lung, breast, prostate, head/neck, esophageal, colorectal, bladder, and gynecologic cancer who reported a score of 4 or more on a pain scale that at least moderately interfered with functioning.
3 health systems (Veterans, Kaiser-Permanente, and UCD) and one private practice in Sacramento, Calif. 265 patients
Treatment fidelity was maintained by training, regular reinforcement, and review of audio recordings of patient encounters. The intervention was described in detail and a treatment manual is available from the authors.
Pain severity, Medical Outcomes study pain impairment scale, physical and mental health component of the SF-12, perceived efficacy in patient-physician interaction scale, pain management subscale of the chronic pain self-efficacy scale.
A power analysis was not provided.
Computer-generated lists and a blocked randomization scheme were used to assure balanced assignment within physicians and to preserve concealment.
Patients were blinded to intervention until after signing consent. Research assistants collecting follow-up interviews and physicians were not aware of patient assignment.
Reasons for drop out were included.
Usual care was not used as the control condition rather enhanced usual care was. The study population was heterogeneous in terms of baseline pain and disease status. Because randomization occurred at the patient not physician level, physicians may have applied what they learned with one patient to other patients. Multiple patient outcomes were investigated with no correction for multiple comparisons.
Support was provided by the American Cancer Society Research Scholars Grant and the National Center for Research Resources. No mention is made of a trial registry.
Lay administered tailored education and coaching resulted in increased pain communication self-efficacy and improvement in pain-related impairment at 2 weeks but not at 6 or 12 weeks. No improvement in pain severity was seen.
Aranda et al. [5]. The title specified randomization.
Patients in the intervention group will report a decrease in psychological and informational needs and an increase in quality of life compared to women in the usual care group.
A somewhat vague intake criterion of a new diagnosis of breast cancer at an advanced stage, recurred or progressed in the preceding 12 months was used. No recruitment dates were listed.
In the outpatient clinics of 4 large urban hospitals (3 public, 1 private) in Melbourne, Australia. 105 women.
A clearly described, nurse delivered face-to-face visit (1 hour) followed by a phone call. The FOCUS framework including family involvement, optimistic attitude, coping effectiveness, uncertainty reduction, and symptom management was used. The control group received standard care.
The European Organization of Research and Treatment of Quality of Life version 2 and Supportive Care Needs Survey at months 1 and 3 after recruitment.
A power analysis was done retrospectively only showing a 70% likelihood of finding significance. The authors mention a need for a larger sample size.
Cards were shuffled and placed in consecutively numbered sealed envelopes.
No mention is made of blinding.
No reasons other than death were given for lost patients.
Although the 4 nurses were employed at different hospitals and all received the same training, there was no discussion of how fidelity to the intervention was maintained. More patients in the intervention arm were undergoing radiation. The one-week separation between the meeting and telephone call was too short to address concerns that involved meeting with other professionals.
It is unclear whether the trial was registered. It was funded by the Melbourne Breast Care Consortium, Victorian Department of Human Services, Australia.
No differences were found until the groups were divided into low and high needs. Women in the intervention group with high needs had more of a decrease in psychological needs than those in the lower need group at 1 month. No differences existed at 3 months.
Rose et al. [6]. Randomization was not indicated in title.
No hypothesis was provided.
Stage IV or stage III lung or pancreatic cancer patient (). Recruitment dates were not listed although recruitment is ongoing.
Two ambulatory care cancer clinics providing care to the underserved including a Veteran’s Administration facility.
A coping and communication support intervention that was clearly described was offered by master’s level mental health nurses. It consisted of a home visit that included a family member followed by telephone contact based on participant preference. Contact was available 24 hours/day, 7 days/week. One follow-up call was placed within 2 weeks. If individuals scored 4 or > on the distress thermometer they were called monthly.
Data was collected at 6 weeks using the POMS (profile of mood scale), health information processing style (Miller Monitoring-Blunting Style Scale), the 13-item symptom distress scale, the distress thermometer.
No power analysis was discussed.
The method of randomization and who conducted it were not addressed.
Blinding was not addressed.
Reasons for dropping out were not specified.
Each mental health nurse had a caseload of 80–100 individuals and no access to their medical record. It may have been difficult to keep track of each individual. Most participants were male.
Trial register is likely as the study was funded by the National Cancer Institute and the American Cancer Society.
Three follow-up telephone calls in 6 weeks were the norm with the nurse initiating contact. More middle-aged individuals raised concerns about communicating with family/friends. Older individuals had more comorbidity. Communication preferences between middle and older groups were similar.
Rose et al. [42]. Randomization was not indicated in title.
No hypothesis was listed.
Diagnosed with late-stage cancer (III or IV) in the last year receiving treatment at an ambulatory clinic. No recruitment dates were mentioned.
2 ambulatory cancer clinics in Cleveland (Metro Health Medical Center or Veterans). 109 younger and 101 older patients.
Usual care control or a coping and comm-unication support intervention provided 24/7 by mental health nurses who offered an in-home visit and followup. 75% of patients identified a family caregiver who also participated. Listening, validation, and education were provided in the intervention group. No mention was made of treatment fidelity. Without additional details the intervention would be hard to replicate
Data was collected at 2 months on sociodemographic data including income, well-being, depressed mood, anxiety, health information processing style, and family discord in communication.
A power analysis was not included.
The method of and who was responsible for randomization were not discussed.
No mention was made of blinding.
No reporting was provided about the number of individuals who declined to participate or reasons for declining. Intent-to-treat analysis was not mentioned.
Reliability and validity and names of specific outcome measures were not summarized. The control group was not described in detail.
Trial register is likely as the study was funded by the National Cancer Institute, Veterans Health Affairs Research, the American Cancer Society, and the A National Institute on Aging Grant.
Middle-aged patients averaged more communication support contacts than older clients. African American patients reported more family discord in communication.
Walker et al. [37]. The title specified randomization.
Supplementing usual care with the intervention will improve depressive symptoms, functioning, quality of life, satisfaction with depression care over 8 months.
Lung cancer patients with a diagnosis of major depression of at least 4 weeks. Recruitment timeframes were listed.
Multicenter trial involving outpatient clinics in Scotland. 200 patients participated.
Cancer nurses under the supervision of a psychiatrist were randomized to usual care or depression care. Six to 8 in-person sessions will be provided followed by telephone contact every 4 weeks for those in the depression care group. The usual care group was not described in sufficient detail
Depression severity (Symptom Hopkins Checklist) collected every 4 weeks over 32 weeks, severity of anxiety (Hospital Anxiety and Depression Scale), pain and fatigue, quality of life, cost of care, and satisfaction with depression care.
The study is powered at 90%
A secure computerized central randomization system and a secure web interface will be used for randomization.
Blinded data collection and analysis are planned.
An intent-to-treat analysis is planned.
Only 10% of recordings and treatment notes of nurses are compared to the treatment manual to evaluate nurse adherence to the protocol.
The trial is registered. Funding was obtained from Cancer Research in the United Kingdom.
Outcome data will be collected until June 2011 at 4, 8, 12, 16, 20, 24, 28, and 32 weeks from randomization.
Girgis et al. [33]. Randomization was not indicated in title.
Patients assigned to either intervention group will report decreased levels of anxiety, depression, and unmet supportive care needs over time in combination with improved physical and emotional functioning.
Notification by the New South Wales Central Cancer Registry of nonlocalized breast or colorectal cancer within 6 months of diagnosis, English speaking, age 18 or older. Recruitment dates are not specified.
New South Wales. 356 patients.
Interventions included usual care, a telephone caseworker (oncology nurses), and an oncologist/general practitioner model. The interventions were not described with sufficient detail for replication.
Anxiety, depression (Hospital Anxiety and Depression Scale), quality of life 30-item EORTC QOL questionnaire), perceived needs (Supportive Needs Survey), and perceived improvements needed in comm-unication with health care providers.
A power analysis was completed.
A computer-generated algorithm was used for randomization.
No mention was made of blinding.
Reasons for dropouts and declined were presented.
Participants were recruited 6 months after diagnosis and may have already adjusted psychologically. A large number of individuals declined to participate. Less than 50% of general practitioners returned data collection instruments.
No funding was mentioned. The trial was included in a registry.
No overall intervention effect was observed. Telephone counseling was more likely to have identified issues of need discussed, referrals made and strong agreement the intervention helped improve communication with health providers.
