Date of Graduation


Document Type


Degree Type



School of Pharmacy


Pharmaceutical Systems and Policy

Committee Chair

Usha Sambamoorthi

Committee Member

Kimberly M Kelly

Committee Member

Virginia G Scott

Committee Member

Joanna Kolodney

Committee Member

Chan Shen


With decades of unchanged cancer care with no added survival benefit, immune checkpoint inhibitors (ICI) changed the treatment landscape of late-stage melanoma in 2011. A key factor in determining the use of ICIs is the presence of pre-existing chronic conditions, which can influence the outcome. However, the prevalence of multimorbidity (defined as presence of two or more chronic conditions) among older patients with late-stage melanoma remains unknown. It also remains unknown if the presence of multimorbidity factors into the use of ICIs. Hospital-related factors associated with ICI use have been studied. Yet, patient-level factors, such as age, sex, marital status, which may play a more crucial role in the use of ICIs, remain unknown. Furthermore, ICI use may exacerbate healthcare expenditures for an already expensive condition, i.e., late-stage cancer. There are no studies exploring the effects of ICI on healthcare expenditures among the elderly. Therefore, this study had three main objectives (1) to examine the prevalence and type of pre-existing multimorbidity and the associated risk factors, (2) to assess the association of multimorbidity and other risk factors on ICI use, and (3) to assess the impact of ICI and multimorbidity on healthcare expenditures among older patients with late-stage melanoma. Retrospective cohort studies were conducted using Surveillance, Epidemiology, and End Results cancer registry linked with fee-for-service Medicare claims for older (>66 years) patients with a 12-month pre-index and 12-month post-index period. Index date was the date of incident stage III/stage IV melanoma diagnosis between 2011 and 2015. Logistic regression was used to examine the associations of multimorbidity or ICI use to various patient-level factors. Generalized linear mixed models with gamma distribution and log-link was used to analyze adjusted relationships between ICI/multimorbidity and healthcare expenditures. In the first aim, an overwhelming majority (85%) of older adults with late stage melanoma had pre-existing multimorbidity. The second aim concluded that only 6% of older adults with late-stage melanoma received ICI. Multimorbidity was not significantly associated with ICI use. Factors positively associated with ICI use were lower age, social support in the form of spouse, residence in the Northeastern regions, and had dual eligibility compared to their counterparts. This study concluded that patient-level factors may play a significant role in decisions towards treatment of late-stage melanoma with ICI, regardless of multimorbidity. The third study reported that the average total, outpatient, home health, and inpatient healthcare expenditures in the pre-index period were significantly (p < 0.001) lower than the expenditures in the post-index period. In addition, regardless of multimorbidity, the high expenditures in post-index period were influenced by the use of ICI. Therefore, ICI use was significantly associated with healthcare expenditures. In summary, multimorbidity is a growing concern for oncologists, especially among the elderly. The high prevalence of multimorbidity among elderly in our study points to the fact that multimorbidity should be factored into cancer care. Since multimorbidity is not included in cancer care guidelines, it does not play a role in receipt of ICI among older patients. However, the use of ICI significantly increases healthcare expenditures. To reduce costs, newer payment models that focus on value are being developed and tested. Future studies need to examine whether such models can achieve the triple aims of “better health, better value, and lower costs.”

Embargo Reason

Publication Pending