Purpose: Simulation models have been widely used to model health care providers’ patient flows and capacity constraints during disasters. These models rely on assumptions about how admission and discharge patterns change during disasters. However, there is little hard data on which to base these assumptions. We examined patient flows during Hurricane Katrina in Louisiana in 2005.
Methods: We have 100% samples of claims data from affected hospitals (n = 45) and skilled nursing facilities (n = 61) in Louisiana. Using these data, we track trends in admissions, discharges, and discharge destination.
Results: Weekly hospital admissions declined from roughly 1,500/patients/week to about 700/patients/week during Hurricane Katrina and remained depressed for the rest of the year. There was a corresponding decline in the number of discharges, but the site of discharge changed. During Hurricane Katrina, there was a large increase, from ~100/patients/week to nearly 250/patients/week discharged to other hospitals. There was a slight decline in the number of patients discharged to skilled nursing facilities. Among skilled nursing facilities, the number of admissions declined during Hurricane Katrina declined from 150/patients/week to less than 50/patients/week. The number of discharges increased from 100/patients/week to almost 300/patients/week. Most were discharged home or to other skilled nursing facilities.
Conclusion: Encouragingly, skilled nursing homes did not transfer patients to hospitals during Katrina. However, our results, which show a decline in hospital-to-skilled nursing home transfers, indicate that simulation models that assume hospitals will be able to create “surge capacity” by discharging patients to nursing homes may rely on unrealistic assumptions about patient flows.
See more of: The 33rd Annual Meeting of the Society for Medical Decision Making