Is the Continuity of Care for Comorbidities among Breast Cancer Patients Related to Treatment Adherence and Survival Outcomes?
Type of DegreePhD Dissertation
MetadataShow full item record
Background: Breast cancer patients with comorbidities need to see multiple healthcare providers. These patients often need to take multiple medications, have frequent health care service utilization and high costs, and have a high mortality rate. Objective: To examine the association of continuity of care (COC) and the health outcomes (the risk of mortality, medication adherence, and health care service utilization and costs) for breast cancer patients with comorbidities. Methods: A series of retrospective cohort studies of newly diagnosed female breast cancer patients with comorbidities (hypertension, hyperlipidemia, and diabetes) were identified from the SEER-Medicare data (2006-2014). The primary outcomes were (1) all-cause mortality (assessed annually for up to 5 years); (2) medication adherence (hormone therapy (HT), antihypertensives, hyperlipidemia medications, diabetes medications) which will equal to 1 with proportion of days covered (PDC)≥80% and 0 with PDC<80% (assessed annually for up to 4 years); (3) health care service utilization and costs (assessed annually for up to 4 years). The COC Index, a validated measure accounting for both the frequency and dispersion of healthcare provider visits, was measured yearly with a scale of 0 (dispersed) to 1 (concentrated) and in two ways: Specialty (unique specialist groups: Primary Care Physicians (PCP), Oncologists, and Other specialists (Other)) and Individual (all individual providers regardless of specialty) COC Indices. Cox proportional hazards models estimated the hazard ratio (HR) of all-cause mortality that was associated with the COC Index. Mixed-effects logistic regression models analyzed the associations of the COC Index and medication adherence to hormone therapy, antihypertensives, hyperlipidemia medications, and diabetes medications. Two-stage models including a mixed-effects logistic regression model and a mixed-effects linear regression model (with a log linked and a gamma distribution) were used to determine the association of the COC Index and health care service (emergency department (ED), inpatient) utilization and costs after HT initiation. Results: Patients who received high continuity of care based on the Specialty COC Index (4th vs. 1st quartile, HR 1.34, 95%CI 1.29-1.40) had higher risks of mortality compared with those who received low continuity of care. However, patients who received high continuity of care based on the Individual COC Index (4th vs. 1st quartile, HR 0.53, 95%CI 0.51-0.54) had lower risks of mortality compared to those who received low continuity of care. Patients with higher levels of Specialty COC Index had less medication adherence to antihypertensives (4th vs. 1st quartile, Odd Ratio (OR) 0.86, 95%CI 0.76-0.96) and hyperlipidemia medications (4th vs. 1st quartile, OR 0.81, 95%CI 0.70-0.94). Patients with higher levels of Individual COC Index had higher medication adherence to hyperlipidemia (4th vs. 1st quartile, OR 1.17, 95%CI 1.01-1.36) and diabetes medications (4th vs. 1st quartile, OR 1.51, 95%CI 1.12-2.00). Patients with intermediate levels of Specialty COC Index were more likely to have ED or inpatient visits (significant in 2nd vs. 1st and 3rd vs. 1st quartile) for all conditions. However, patients with high levels of Individual COC Index were less likely to have ED or inpatient visits (significant in 2nd vs. 1st, 3rd vs. 1st, and 4th vs. 1st quartile) for all conditions. Accordingly, high levels of Specialty COC Index were associated with increased inpatient costs (all-cause-, hypertension-, hyperlipidemia-, and diabetes-related p<0.05). High levels of Individual COC Index were iii associated with decreased ED cost (all-cause related p<0.05) and inpatient costs (all-cause-, hypertension-, and diabetes-related p<0.05). Conclusion: This study examined the association between COC Indices and health outcomes for breast cancer patients. We found mixed associations of COC Indices and health outcomes. Higher continuity of care based on visits to individual providers could protect patients from all-cause mortality, medication non-adherence, and health service utilization and costs. In contrast, higher continuity of care based on visits to specialty groups could be associated with a high risk of mortality, non-adherence to medication, and more health service utilization and higher costs in patients. Thus, COC should be considered carefully based on different definitions when building the models of continuity and coordination of care and may help to improve the health outcome for cancer survivorship.
- Dissertation_Ning Cheng_final_July2319.pdf