Breast Cancer Collaborative Registry informs understanding of factors predicting sleep quality

Significance: Poor sleep quality is a common and persistent problem reported by women with breast cancer (BC). Empirical evidence identifies many risk factors for self-reported sleep deficiency, but inconsistencies limit translation to practice. Purpose: To increase understanding of risk factors predicting self-reported poor sleep quality in women with BC who completed the Breast Cancer Collaborative Registry (BCCR) questionnaire. Methods: This cross-sectional study recruited women with a first diagnosis of BC (n = 1302) at five sites in Nebraska and South Dakota. Women completed the BCCR that includes numerous variables as well as the Pittsburgh Sleep Quality Index (PSQI) and SF36v2 (n = 1260). Descriptive statistics and non-parametric correlations were used to determine associations and create predictive models of sleep quality with BCCR variables and SF36v2 subscales. Results: Most women were white (93.7%) and married (71.5%); mean age was 60.1 (21–90) years. Poor sleep was self-reported by 53% of women. Seven variables were highly associated with sleep quality (p ≤ 0.001). The first model found younger age, lower physical activity, and higher fatigue were the strongest combined and independent variables predicting poor sleep quality (F = 23.0 (p <.001), R² = 0.103). Participants self-reported lower health status on most SF36v2 subscales [Z = 44.9 (11.6) to 49.1 (10.1)]. A second model found that all subscales were predictors of poor sleep; vitality, mental health, bodily pain, and general health were the strongest predictors (F = 101.3 (p <.001), R² = 0.26). Conclusions: Results confirm previously identified risk factors and reveal inconsistencies in other variables. Clinicians need to routinely screen for the identified risk factors of self-reported poor sleep quality.