Trajectories of sleep disturbance and daytime sleepiness in women before and after surgery for breast cancer

Abstract

Context: Sleep disturbance is a problem for oncology patients.

Objectives: To evaluate how sleep disturbance and daytime sleepiness (DS) changed from before to six months following surgery and whether certain characteristics predicted initial levels and/or the trajectories of these parameters.

Methods: Patients (n=396) were enrolled prior to surgery and completed monthly assessments for six months following surgery. The General Sleep Disturbance Scale was used to assess sleep disturbance and DS. Using hierarchical linear modeling, demographic, clinical, symptom, and psychosocial adjustment characteristics were evaluated as predictors of initial levels and trajectories of sleep disturbance and DS.

Results: All seven General Sleep Disturbance Scale scores were above the cutoff for clinically meaningful levels of sleep disturbance. Lower performance status; higher comorbidity, attentional fatigue, and physical fatigue; and more severe hot flashes predicted higher preoperative levels of sleep disturbance. Higher levels of education predicted higher sleep disturbance scores over time. Higher levels of depressive symptoms predicted higher preoperative levels of sleep disturbance, which declined over time. Lower performance status; higher body mass index; higher fear of future diagnostic tests; not having had sentinel lymph node biopsy; having had an axillary lymph node dissection; and higher depression, physical fatigue, and attentional fatigue predicted higher DS prior to surgery. Higher levels of education, not working for pay, and not having undergone neo-adjuvant chemotherapy predicted higher DS scores over time.

Conclusion: Sleep disturbance is a persistent problem for patients with breast cancer. The effects of interventions that can address modifiable risk factors need to be evaluated.

Fig. 1

Trajectory of sleep disturbance (A) and daytime sleepiness (DS) (B) over the six months of the study.

Fig. 2

Influence of education (A: less/more education calculated based on 1 standard deviation (SD) above and below the mean years of education) on the slope parameters for sleep disturbance and influence of performance status (B: lower/higher functional status calculated based on 1 SD above and below the mean Karnofsky Performance Status score); comorbidities (C: lower/higher comorbidity calculated based on 1 SD above and below the mean Self-administered Comorbidity Questionnaire score), and receipt of adjuvant chemotherapy (CTX) (D: did or did not receive adjuvant CTX) on interindividual differences in the intercept for sleep disturbance.

Fig. 3

Influence of severity of hot flashes (A: lower/higher severity of hot flashes calculated based on 1 SD above and below the mean hot flash severity score), attentional fatigue (B: lower/higher attentional fatigue calculated based on 1 SD above and below the mean Attentional Function Index score), and physical fatigue (C: lower/higher physical fatigue calculated based on 1 SD above and below the mean Lee Fatigue Scale score) on interindividual differences in the intercept for sleep disturbance; and influences of depressive symptoms (D: i.e., lower/higher depressive symptoms calculated based on 1 SD above and below the mean Center for Epidemiological Studies-Depression score) on interindividual differences in the intercept and slope parameters for sleep disturbance.

Fig. 4

Influence of education (A: less/more education based on 1 SD above and below the mean years of education) and employment status (B: employed or not employed) on the slope parameters for daytime sleepiness.

Fig. 5

Influence of performance status (A: lower/higher functional status calculated based on 1 SD above and below the mean Karnofsky Performance Status score), body mass index (BMI) (B: lower/higher BMI calculated based on 1 SD above and below the mean BMI), underwent a sentinel lymph node biopsy (SLNB) (C: did or did not have SLNB) and underwent an axillary lymph node dissection (ALND) (D: did or did not have an ALND) on interindividual differences in the intercept for daytime sleepiness (DS) and neoadjuvant chemotherapy (CTX) (E: did or did not have neo-adjuvant CTX) and adjuvant CTX (F: did or did not have adjuvant CTX) on the slope parameters for DS.

Fig. 6

Influence of depressive symptoms (A: lower/higher depressive symptoms calculated based on 1 SD above and below the mean Center for Epidemiological Studies-Depression score), physical fatigue (B: lower/higher physical fatigue calculated based on 1 SD above and below the mean Lee Fatigue Scale score), and attentional fatigue (C: lower/higher attentional fatigue calculated based on 1 SD above and below the mean Attentional Function Index score) on interindividual differences in the intercept for daytime sleepiness and fear of future diagnostic tests (D: lower/higher fear calculated based on 1 SD above and below the mean fear of future diagnostic test score) on the slope parameters for DS.