School
School of Public Health
Document Type
Dissertation
Department (new version)
Public Health
Date of Award
1998
Degree Name by School
Doctor of Public Health (DrPH) School of Public Health
Abstract
Variations in hospital utilization rates among populations have been documented by health services researchers. Diagnosis related groups (DRGs) have been aggregated to reflect these differences. Low variation (LV) DRGs are based upon physician consensus on the need for hospitalization. Ambulatory care sensitive (ACS) DRGs deal with procedures that have shifted from inpatient to outpatient. In avoidable hospital conditions (AHC), inpatient hospitalizations are avoided through an effective outpatient treatment approach. Using the two-county age-specific utilization rates as the standard, expected discharges and indirect age-adjusted rates were calculated for all zip codes. The average rate for all three DRGs was 28.8 discharges per 1,000 population across the 4-year period. AHC DRGs composed the majority of discharges under study, and ACS accounted for the least. No change occurred in the rates for AHC and LV DRGs in the two-county area. Rates for ACS DRGs declined 17% annually. LV DRGs without complication declined with concomitant increases in the same DRG with complication. The rates varied least for ACS DRGs ( CV = 0.24) for most AHC DRGs ( CV = 0.35). Simple correlations found all three categories of DRGs significantly and positively associated with each other. Independent variables were positively correlated to each other and the rate for each DRG category. AHS and LV DRGs were significantly associated with race, education, and income; ACS DRGs were not. In single variable Poisson models, LV and AHC DRGs were significantly and positively associated with each independent variable. Education was a significant and positive predictor of ACS DRGs and explained the most variance across all three categories of DRGs. Using LV DRGs as an adjustment for disease burden, education remained significant for ACS and AHC discharges, and race became insignificant for AHC DRGs. In full Poisson models, only education was significant for ACS DRGs. The study concluded that procedures shifted to the outpatient setting and hospitals increased coding of LV DRGs with complications during the study period. Race was not associated with differences in rates across small areas after adjusting for disease burden. Education appeared to be the single most critical factor in determining variations in population based hospitalization rates.
ProQuest Publication Number
ProQuest ID
9949100
ISBN
978-0-599-51323-5
Recommended Citation
Deshazo, William Floyd IV, "A population-based approach to monitoring inpatient utilization and determining the influence of sociodemographic factors on three theoretical categories of diagnosis -related groups in a two-county area of Alabama" (1998). All ETDs from UAB. 7197.
https://digitalcommons.library.uab.edu/etd-collection/7197
Comments
DrPH