Purpose: To assess across-state variation in inpatient cost of diabetes-related lower extremity amputation (DLEA) in relation to patient, hospital and state factors.
Method: Patient age, gender, race, length of stay, level of amputation, in-patient mortality, co-morbidities index, primary payer, hospital size, hospital volume of DLEA surgery, type of hospital (rural, urban non-teaching, and urban teaching) and cost were were obtained from the 2007 Healthcare Cost and Utilization Project (HCUP). State average personal income, percentage of the population below the poverty line, and population density were obtained from the U.S. Census Bureau. The number of hospital beds per 10,000 residents, number of physicians per 10,000 residents, hospital admissions per 1,000 residents and average malpractice cost were obtained from the Kaiser Family Foundation website. Three-level hierarchical linear regression models were implemented to analyze the association between in-patient cost and independent variables. The clustering effects of patients within hospitals and those of hospitals within states were captured by random effects of the intercepts. The regional patterns of unexplained cost variation were compared with that of the raw cost variation.
Result: There were 9,066 DLEA hospitalizations and thirty-nine states had cost data. The mean cost per in-patient stay was $17,103, which is $6,180 less than the costliest state (NY). Four out of the five most costly states were adjacent coastal states (NY and NJ, CA and OR). In the regression analysis, age, race, length of stay, level of amputation, in-patient mortality, primary payer, co-morbidities and type of hospital were statistically significant and explained 55.3% of the variance. The means of cost unexplained by those factors showed the most costly states were the three west coast states followed by five midwestern states and the least costly states were four southern states and the adjacent Kansas. However, after controlling for the significant state-level variable (hospital beds supply), this pattern became less clear. Since we found that the greater number of hospital beds per 10,000 residents was related to lower inpatient cost, the aforementioned pattern can be partially attributed to the different level of hospital bed availability across states.
Conclusion: We found there were some regional patterns of costs unexplained by patient and hospital factors. Further research is needed to examine whether similar patterns exist for other costly surgical procedures.
See more of: The 33rd Annual Meeting of the Society for Medical Decision Making