A Cross‐Sectional Statewise Study of Preventable Hospitalization Rates Among Medicaid Recipients

Background:

States in the United States have wide latitude in administering Medicaid, thus creating an opportunity to study interstate differences in quality and outcomes among Medicaid recipients. Studying statewise variations in preventable hospitalization (PH) rates for ambulatory‐sensitive conditions (ASC) can provide insights into the quality and access to health care services in this patient population.

Methods:

We evaluated state variations in PH rates among Medicaid recipients using hospital discharge data from the AHRQ's National Inpatient Sample (NIS) database for 2003. The NIS contains data from 8 million hospital stays, or about 90% of all hospital discharges in the United States. We used hospital discharge data from the NIS to obtain a cross‐sectional sample of hospitalizations for each available state. Hospitalizations with a primary diagnosis that was either ambulatory sensitive or a marker (ambulatory insensitive) were included. Ambulatory‐sensitive, or preventable, hospitalizations are those that can be potentially avoided by access to timely primary care. In contrast, marker conditions are relatively insensitive to primary care access. Because Medicaid enrollment varies throughout the year, the number of marker hospitalizations, which occur at relatively constant rates in populations, was used in the denominator. We calculated the ratio of preventable to marker conditions for 37 U.S. states.

Results:

We calculated ratios of preventable to marker hospitalizations among Medicaid recipients for each state in 2003. Nationally, there was 6.6 times as many preventable hospitalizations as marker hospitalizations, but this ratio varied across states from 3.04 to 11.74. The data were distributed normally, but there was significant regional clustering. States in the Southeast and Southwest had significantly higher PH rates compared with the rest of the country (P < 0.01). All the Southeastern states had rates that were above the U.S. mean, whereas those in the Pacific Northwest had rates below the U.S. mean. Among states with large Medicaid populations, New York, New Jersey, Texas, and Florida had higher rates, whereas Illinois and California had lower rates of preventable hospitalizations.

Conclusions:

States have varying rates of Medicaid population hospitalization for ambulatory‐sensitive conditions. There is nearly a fourfold difference between states with the highest and lowest PH rates, with states in the Southeast having the highest rates. Several factors could explain this varying propensity for hospitalization. Patient factors, such as demographic characteristics, health status, and care preferences, as well as physician supply and practice style could help explain PH variation. Differing Medicaid eligibility, coverage, and physician reimbursement rates could also affect statewise PH variation. Future research to identify the causes of this variation will be useful to policymakers as they craft Medicaid policy that aims to reduce hospitalization costs.

Disclosures:

S. Tumuluri ‐ none; G. Hougham ‐ none; D. Meltzer ‐ none