Frailty Screening and Impairment Patterns in Older Adults with Cancer: A Dissertation Using the Cancer Aging and Resilience Evaluation (CARE) Registry

Public Deposited
Analytics

Downloadable Content

Download PDF
Request Version for Screen Reader
Creator
  • Thai, Sydney T
    • Gillings School of Global Public Health, Department of Epidemiology
Abstract
  • Older adults with cancer often have multiple health impairments. Although geriatric assessments are recommended to identify older adult impairments not captured in oncology assessments, impairments are not often evaluated. Thus, patients may undergo treatment without support for impairments associated with adverse outcomes. The objectives were to evaluate the screening performance of skeletal muscle density (SMD) in classifying frailty in older adults with cancer (Aim 1) and to identify and describe geriatric assessment impairment patterns in older adults with gastrointestinal cancers (Aim 2). We used the Cancer Aging and Resilience Evaluation (CARE) registry (University of Alabama at Birmingham). Patients completed the CARE tool, a self-reported geriatric assessment, and responses were used to calculate frailty scores and categorize impairment indicators. SMD was calculated from computed tomography scans (L3 vertebrae), and performance in classifying frailty was evaluated using diagnostic model methods and compared by gender-diabetes status. Using latent class analysis (LCA) with impairment indicators, we identified impairment classes among patients with gastrointestinal malignancies and described classes using impairment probabilities, patient characteristics, and one-year mortality. SMD performed poorly. Area under the receiver operating curve (AUC) estimates were low for all four gender-diabetes subsets (range: 0.58-0.68). Third quartile gender-specific cut-off points for SMD had high sensitivity (0.76-0.89), but low specificity (0.25-0.34). Positive and negative likelihood ratio results indicated that utility was most promising for men with diabetes compared to other subgroups. We identified 6 impairment latent classes (LC): mild impairment (LC1, 28% prevalence); social support impairment (LC2, 12%); weight loss alone (LC3, 16%); impaired, low anxiety/depression (LC4, 23%); impaired with anxiety/depression (LC5, 11%); global impairment (LC6, 11%). One-year mortality risk estimates ranged from 11% (mild impairment) to 44% (global impairment) compared to 14-34% when using frailty categories. In overall and stratified analyses (high- vs. low-risk cancers), mortality estimates for the 3 impaired classes (LC4, LC5, LC6) were greater than the mild impairment class. Work is needed to improve classification of frailty with SMD and to assess performance in other subgroups and populations. The identified geriatric assessment impairment classes can facilitate awareness of impairment clustering and the planning of support services for older adults with cancer.
Date of publication
Keyword
DOI
Resource type
Rights statement
  • In Copyright - Educational Use Permitted
Advisor
  • Lund, Jennifer L
  • Williams, Grant R
  • Poole, Charles
  • Stürmer, Til
  • Buse, John B
Degree
  • Doctor of Philosophy
Degree granting institution
  • University of North Carolina at Chapel Hill Graduate School
Graduation year
  • 2022
Language

Relations

Parents:

This work has no parents.

Items

Thumbnail Title Date Uploaded Visibility Actions
file details Thai_unc_0153D_21230.pdf 2023-02-02 Public Download