Symptoms and survivorship needs differences between "good sleepers" and "bad sleepers" in survivors of breast and gynecologic cancers.

INTRODUCTION:Although 80% of cancer survivors report symptoms of insomnia, only 28-43% meet DSM-5 criteria for this diagnosis. We sought to characterize the association between patient-reported insomnia symptoms, patient outcomes, and supportive care variables, as well as explore clinically meaningful insomnia thresholds in a sample of women diagnosed with breast and gynecologic cancers. METHODS:From July 2018-March 2019, all breast and gynecologic cancer survivors seen at the Stanford Women's Cancer Center were approached and invited to participate in the study (15% declined). Of those who consented, 273 survivors completed an online survey related to their sleep (ISI), quality of life (FACT-G), distress (PHQ-4), supportive care needs (SCNS-SF34), and symptom severity (MDASI). Survivors who scored <8 on ISI were categorized as "good sleepers," survivors with ISI ≥8 were categorized as "bad sleepers." RESULTS:126 (46.2%) of survivors were "good sleepers," 147 (53.8%) were "bad sleepers." Good sleepers were older than bad sleepers (p < .05) but did not differ in any other demographic or any medical variables. Using hierarchical linear regression models, we found that good sleep (ISI <8) was associated with higher quality of life, lower psychological distress, increased social support, lower symptom severity, and lower supportive care needs, after accounting for demographic, medical, and treatment variables. The findings were largely replicated with an ISI cut off of 15. CONCLUSIONS:Among women treated for breast and gynecologic cancers, survivors who were good sleepers had better psychosocial outcomes, fewer supportive care needs, and lower symptom severity compared to those who reported insomnia symptoms. Results also indicate that degree of sleep impairment, whether mild or severe, has similarly poor associations with most aspects of patient functioning and symptomatic burden. Further research is needed to determine causality of these findings.