Graduation Year


Document Type




Degree Granting Department


Major Professor

Cecile A. Lengacher


actigraphy, anxiety, depression, fatigue, pain, women


Breast cancer is a major health problem and comprises the largest population of cancer survivors in the United States, estimated at 2.9 million women, accounting for 22% of all cancer survivors (National Cancer Institute, 2013). The advances in breast cancer screening, diagnosis and treatment has increased the importance of survivorship needs. A major concern among breast cancer survivors (BCS) is sleep disturbances. This study used an innovative approach to examine ethnic and racial disparities in sleep disturbances present in BCS. In addition, this study also explored sleep disturbances across different races/ethnicities. This study was a secondary data analysis of baseline data from the supplement study of the MBSR Symptom Cluster Trial for Breast Cancer Survivors/ 1R01CA131080, conducted by Dr. Lengacher. Sleep was measured using the Pittsburgh Sleep Quality Index, sleep diary (subjective), and sleep actigraphy (objective). The sample consists of 79 women who had been diagnosed with breast cancer (Stage 0, I, II, III), completed lumpectomy and/or mastectomy, and were within 2 weeks to 2 years post radiation and/or chemotherapy treatment. The aims for this study were to: 1) explore racial/ethnic differences in objectively measured sleep patterns among BCS; 2) estimate and compare the correlation between objective and subjective sleep quality by racial/ethnic groups among BCS; 3) examine which sleep actigraphy measure appears to have the strongest relationship with physical and psychological symptoms; and 4) explore whether these relationships (i.e. between objective sleep and self-reported symptoms) appear to be modified by race/ethnicity.

The first aim was conducted using analysis of variance (ANOVA) and analysis of covariance (ANCOVA). Results indicated that white, non Hispanic BCS had improved objective sleep quality compared to minority BCS. The second aim was conducted using Pearson's correlation with significant correlations found between subjective and objective sleep onset latency (r= .310, p= .016), and total sleep time (TST) (r= .328, p= .011) for the white, non-Hispanic group. The third aim was conducted using Pearson's correlation with significant correlations between sleep onset latency and depression (r= .247, p= .029); sleep efficiency and depression (r= -.233, p= .040); sleep efficiency and fatigue (r= -.207, p= .045); and WASO and pain (r= .277, p= .014). There were no significant correlations between the anxiety score and actigraphy parameters.

Using the significant correlations from the results of the third aim, multiple regression analysis was conducted with age as a covariate to test the fourth aim. The main effect of depression on sleep efficiency was significant (p= .044) with less depression associated with higher sleep efficiency. The interaction term, depression by race/ethnicity, had a non-significant effect on sleep efficiency (p= .299). The main effect of pain on WASO was significant (p= .008), and increased pain was associated with longer WASO. The interaction term, race/ethnicity by pain, had a non-significant effect on WASO (p= .148). The main effect of depression predicting sleep onset latency was significant (p= .027), and the interaction term, depression by race/ethnicity, had a trend towards a significant effect (p= .092) on predicting sleep onset latency. The interaction between depression and race/ethnicity predicting sleep onset latency was further decomposed using multiple regression. The average sleep onset latency was longer in the minority group with high depression levels (42 minutes) compared to the white, non-Hispanics with high depression levels (29 minutes). Race modified the effect of depression on sleep onset latency in this sample of BCS.

These finding suggest that the experience of objective sleep disturbances is different among races/ethnicities. Additional research is needed to further explore racial/ethnic differences in subjective and objective sleep disturbances and its impact on physical and psychological symptoms among BCS. As the number of BCS continue to rise, it is becoming increasingly important to recognize sleep disturbances and their potential physical and psychological effects early in BCS, specifically in different races and ethnicities. It is anticipated that these findings may contribute to improved symptom management for women of different races and ethnicities.