Berger, A. M., Kuhn, B. R., Farr, L. A., Lynch, J. C., Agrawal, S., Chamberlain, J., & Von Essen, S. G. (2009). Behavioral therapy intervention trial to improve sleep quality and cancer-related fatigue. Psycho-Oncology, 18, 634–646.

To determine the effect of behavioral therapy (BT)—specifically, an individualized sleep promotion plan (ISPP)—on sleep quality and fatigue in patients with breast cancer undergoing adjuvant chemotherapy.

Eligible women who consented to participate were randomized using stratified random sampling to either the BT group or to a healthy eating control (HEC) group prior to adjuvant chemotherapy. Patients completed questionnaires at baseline and wore a wrist actigraph for two days prior to initial treatment. Patients randomized to the BT group developed an ISPP during individual visits with the research nurse two days prior to treatment. Modifications to this plan were made two days prior to each treatment and 30 days after the last treatment. Modifications were based on patients' sleep diary data and treatment adherence. BT plans were reinforced during 15-minute sessions seven days after each revision. Patients in the HEC group received equal time and attention during individual visits and received information on healthy eating topics at each visit. Patients in the HEC group were referred to their treatment clinic for questions about fatigue and sleep.

• The study was comprised of 219 female patients with breast cancer.
• Mean age was 52.13 years (range 29–79) in the BT group and 52.16 years (range 30–83) in the HEC group.
• Patients had an initial diagnosis of stage I to IIIA breast cancer and had undergone either modified radical mastectomy or lumpectomy.
• Of the patients, 70% were partnered and 74% had at least some college education.
   

The study was conducted in 12 oncology clinics in the Midwestern United States.
 

Patients were undergoing the active treatment phase of care.

This was a randomized, controlled trial.

• Symptom Experience Scale (SES):  This scale measures frequency, intensity, and distress associated with six symptoms associated with cancer treatment. Scores for fatigue and sleep were not included in the mean SES score because they were measured in-depth by other study instruments.    
• Hospital Anxiety and Depression Scale (HADS): This scale measures anxiety and depression symptoms.
• Medical Outcomes Study Short-form General Health Survey (MOS-SF 36-v2):  The physical functioning subscale of this measure was used to measure physical functioning prior to cancer diagnosis.
• Piper Fatigue Scale (PFS):  The PFS is a 22-item scale of subjective cancer-related fatigue. The PFS was used at each treatment because fatigue was expected to fluctuate from treatment to treatment. Higher scores on the PFS indicate worse fatigue. Reported internal consistency for this study was 0.93 to 0.98.
• Pittsburgh Sleep Quality Index (PSQI):  The PSQI is a 19-item subjective measure of sleep quality over the past month. Seven component scores are summed to obtain global scores ranging from 0 to 21, with higher scores indicating greater sleep difficulties. In women with breast cancer, Cronbach’s alpha for global PSQI score is 0.80; it was 0.74 to 0.83 for this study.
• Wrist actigraphy and daily diaries were used to measure sleep variables. The study used the Motionlogger® Actigraph with one-minute epochs. Variables included total sleep time after sleep onset, sleep efficiency, number of awakenings, and minutes and percent awake after sleep onset. Actigraphs were recorded 48 hours before the initial treatment, for seven days and nights during treatment and for seven days and nights 30 days after the last treatment.
  
  
  
  
   

Mean PSQI scores in both groups were greater than five, which indicated poor sleep compared to the general population; however, mean scores were not greater than eight, a cutoff score associated with poor sleep quality in patients with breast cancer. Actigraphy and diary data showed normal sleep duration and sleep efficiency in both groups across treatment and follow-up. Number of awakenings after sleep onset measured by both sleep diaries and actigraphy were higher than normal in both groups. Significant differences between sleep diaries and actigraphy were observed for all sleep variables (p < 0.01 for all variables), with lower numbers of awakenings and higher sleep efficiency per diary data in the BT group. A significant group by time interaction was found for changes in the PSQI, with sleep quality improving in the BT group (p < 0.049). Although not significant, there were trends towards improved sleep quality over time in the BT group per actigraphy for total sleep time and number of awakening and per sleep diary for sleep efficiency. Perceived fatigue changed significantly over time in both groups (p < 0.001), with increased fatigue during treatments and decreased fatigue after the end of treatments in both groups. There was no apparent effect of BT on fatigue levels.

Patients in the BT group showed greater improvement in sleep quality over time than those in the the HEC group, although perceptions of improved sleep quality were not consistently associated with objective sleep measures, sleep diaries, or reported fatigue. BT was not shown to have an effect on fatigue.

• There was a lack of true baseline values of sleep and fatigue because patients were enrolled after surgery.
• There was no control of patients’ expectations of treatment.
• There were modifications to stimulus control and restriction therapy within the BT intervention model.
• The study population lacked racial/ethnic diversity.

BT may be used by trained nurses to improve sleep quality in patients with breast cancer receiving adjuvant chemotherapy. Further research is needed to determine the long-term effects of BT on sleep quality and fatigue in this population.