▼ Acute appendicitis is the most common cause for emergent surgery among children in the United States. More than 70,000 cases of confirmed appendicitis are diagnosed in the pediatric population in the United States per year, accounting for some $3 billion in hospital charges annually. Potential complications associated with suspected acute appendicitis include misdiagnosed or negative appendectomy (i.e. the surgical removal of a healthy appendix) and perforated appendicitis (i.e. the rupturing of the appendix). The objective of this dissertation was to examine hospital and community characteristics associated with pediatric appendectomy outcomes. The first study examined negative appendectomy in a cross-sectional analysis of pediatric patients, including a secondary analysis involving adult and pediatric appendectomy volumes; the second study focused on the change in negative appendectomy rates over time in a longitudinal study of pediatric patients; and the third study involved perforated appendicitis among pediatric patients, incorporating county-level healthcare resource and socioeconomic variables into the analysis, in a cross-sectional study. Secondary data from the Healthcare Cost and Utilization Project (HCUP) Kids’ Inpatient Database (KID), the Nationwide Inpatient Sample (NIS), and the Area Resource File (ARF) were analyzed in logistic regression models employing generalized estimating equations in order to determine associations between pediatric appendectomy outcomes and hospital and community characteristics. Study one results showed children’s hospitals were associated with lower odds of negative appendectomy, especially for very young patients. Health system membership was not found to be associated with negative appendectomy outcomes generally; however, an exploratory analysis found children’s hospitals have higher odds of negative appendectomy when part of multi-hospital health systems. Appendectomy volume (including adult and pediatric surgical volume) was not found to be associated with negative appendectomy rates. Results of the longitudinal study showed general hospitals achieved the greatest decrease in negative appendectomy rates from 2000 to 2009. Similarly, urban hospitals were found to have greater decreases in the odds of negative appendectomy over the same time period when compared with rural hospitals. Results for hospital teaching status were mixed. Teaching hospitals appear to have had a greater reduction in odds of negative appendectomy from 2000 to 2003 and 2003 to 2006, but both teaching and non-teaching hospitals had meaningful reductions from 2006 to 2009. Perforation rates were not found to be associated with physician density in study three results. However, post-study analysis revealed that increased density of pediatricians is associated with a decrease in the odds of perforated appendicitis. High density of healthcare resource availability, namely, emergency department and operating room density, was found to be associated with increased odds of perforation. Conversely, high median income (regardless of patient income) was found to be associated with decreased odds of perforation. Advisors/Committee Members: Wickizer, Thomas.

▼ This study attempted to determine if select patient and physician demographic variables are predictors of the prescribing of asthma pharmacotherapy in the ambulatory patient population of the United States. Another aim of this study was to evaluate if physician prescribing behavior of asthma pharmacotherapy was in compliance with the recommendations of the National Heart, Lung and Blood Institute’s National Asthma Education and Prevention Program Expert Panel 2 national asthma guidelines. This study was a retrospective cross-sectional study that used data from the National Ambulatory Medical Care Survey (NAMCS) from 1998 through 2004. The weighted population sample size of the study was 82,020,318 patients. There were 1540 observations in this study (pre-weighted sample size) and there were 96 strata, with 446 Population Sampling Units (PSUs). Specific patient demographic variables, physician demographic variables and asthma medications prescribed were extracted from the dataset and analyzed. A series of logistic regression models were created with the choice of asthma pharmacotherapy agent used as the dependent variable and the patient and physician demographic variables as the independent variables. The major finding from the study was that physicians did not seem to be adhering to the EPR-2 pharmacotherapy guidelines. Another major finding from this study was the fact that there were vulnerable sub-populations of asthma patients that were receiving sub-optimal asthma pharmacotherapy. This study did not find a specific physician demographic variable that was strongly and consistently linked to physician prescribing behavior of asthma pharmacotherapy. In conclusion, it appears that a more concerted effort needs to be undertaken to improve physician adherence to the EPR-2 guidelines, especially in the prescribing of asthma pharmacotherapy. Advisors/Committee Members: Balkrishnan, Rajesh.