Prognostic scores, and more specifically comorbidity scores, are important and widely used measures in the health care field and in health services research. A comorbidity is an existing disease an individual has in addition to a primary condition of interest, such as cancer. A comorbidity score is a summary score that can be created from these individual comorbidities for prognostic purposes, as well as for confounding adjustment. Despite their widespread use, the properties of and conditions under which comorbidity scores are valid dimension reduction tools in statistical models is largely unknown. This dissertation explores the use of summary comorbidity measures in statistical models. Three particular aspects are examined. First, it is shown that, under standard conditions, the predictive ability of these summary comorbidity measures remains as accurate as the individual comorbidities in regression models, which can include factors such as treatment variables and additional covariates. However, these results are only true when no interaction exists between the individual comorbidities and any additional covariate. The use of summary comorbidity measures in the presence of such interactions leads to biased results. Second, it is shown that these measures are also valid in the causal inference framework through confounding adjustment in estimating treatment effects. Lastly, we introduce a time dependent extension of summary comorbidity scores. This time dependent score can account for changes in patients' health over time and is shown to be a more accurate predictor of patient outcomes. A data example using breast cancer data from the SEER Medicare Database is used throughout this dissertation to illustrate the application of these results to the health care field. / Statistics

OBJECTIVE: Gunshot wounds to the head (GSWH) are devastating injuries with a grim prognosis. Several prognostic scores have been created to estimate mortality and functional outcome, including the so-called Baylor score, an uncomplicated scoring method based on bullet trajectory, patient age, and neurological status on admission. This study aimed to validate the Baylor score within a temporally, institutionally, and geographically distinct patient population. METHODS: Data were obtained from the trauma registry at a level I trauma center in the southeastern US. Patients with a GSWH in which dural penetration occurred were identified from data collected between January 1, 2009, and June 30, 2019. Patient demographics, medical history, bullet trajectory, intent of GSWH (e.g., suicide), admission vital signs, Glasgow Coma Scale score, pupillary response, laboratory studies, and imaging reports were collected. The Baylor score was calculated directly by using its clinical components. The ability of the Baylor score to predict mortality and good functional outcome (Glasgow Outcome Scale score 4 or 5) was assessed using the receiver operating characteristic curve and the area under the curve (AUC) as a measure of performance. RESULTS: A total of 297 patients met inclusion criteria (mean age 38.0 [SD 15.7] years, 73.4% White, 85.2% male). A total of 205 (69.0%) patients died, whereas 69 (23.2%) patients had good functional outcome. Overall, the Baylor score showed excellent discrimination of mortality (AUC = 0.88) and good functional outcome (AUC = 0.90). Baylor scores of 3-5 underestimated mortality. Baylor scores of 0, 1, and 2 underestimated good functional outcome. CONCLUSIONS: The Baylor score is an accurate and easy-to-use prognostic scoring tool that demonstrated relatively stable performance in a distinct cohort between 2009 and 2019. In the current era of trauma management, providers may continue to use the score at the point of admission to guide family counseling and to direct investment of healthcare resources.