Hospital Risk-Adjusted Mortality Rates on the Decline

The Patient Protection and Affordable Care Act enacted this year has focused renewed attention on the quality of healthcare in the United States. There is growing concern about the US healthcare system, which is among the highest in the world in annual expenditures but experiences poorer health outcomes than other industrialized countries. Because high mortality in some institutions may be associated with quality of care, measurement of risk-adjusted mortality rates serves as an indicator of quality. Studies have shown decreasing trends in risk-adjusted hospital mortality for conditions including acute myocardial infarction (AMI), congestive heart failure (CHF), pneumonia, and stroke in the hospital setting, but these decreasing trends in the overall population may mask differences in sub-populations. To gain an understanding of trends in mortality in selected subgroups, the Agency for Healthcare Research and Quality (AHRQ) has issued Statistical Brief #98, Trends in Hospital Risk-Adjusted Mortality for Select Diagnoses by Patient Subgroups, 2000-2007. The brief includes data from the Healthcare Cost and Utilization Project on the rates of risk-adjusted inpatient mortality, and utilizes AHRQ’s Inpatient Quality Indicators to develop risk-adjusted in-hospital death rates among adults for AMI, CHF, pneumonia, and stroke. Risk-adjusted mortality rates were evaluated for 1994, 1997, and 2000-2007. The national risk-adjusted mortality rates for the 4 conditions decreased significantly between 1994 and 2007. For CHF patients, the decrease in mortality rate was 60% (from 70 to 28 deaths per 1000 admissions); for pneumonia patients, the rates decreased by 55% (from 91 to 41 deaths per 1000 admissions); for AMI inpatients, the decrease was 47% (from 128 to 67 deaths per 1000 admissions); and for stroke inpatients, the decrease was 35% (from 143 to 92 deaths per 1000 admissions) (P≤.05 for all differences). The brief includes data on changes in inpatient risk-adjusted mortality for the 4 diagnoses by patient subgroups. The decreases between 2000 and 2007 were similar across age groups for stroke and pneumonia; rates for AMI and CHF decreased faster for patients ≥65 years of age compared with patients ages 18 to 44 years, who exhibited overall lower death rates from those conditions in 2000 and 2007 (P≤.05 for all differences). Compared with men, the rates of decrease in inpatient risk-adjusted mortality rates were slower among women with CHF and stroke (−46% vs −52% and −24% vs −29%, respectively; P≤.05 for differences). In general, inpatient mortality rates declined more quickly for patients in suburban locations than for those living in other areas. Inpatient mortality for AMI hospital stays decreased by 37% in suburban areas, compared with 34% in the most rural areas and 32% in large town rural areas. Patients in suburban areas experienced the largest decrease (51%) in inpatient mortality for pneumonia compared with patients living in all other areas. Likewise, inpatient mortality for CHF and stroke decreased more quickly among patients living in suburban areas than it did for patients living in other places. The Midwest had the highest rate of decline for AMI mortality (44%) compared with other regions and went from the highest regional rate in 2000 (112 deaths per 1000 admissions) to the lowest in 2007 (63 deaths per 1000 admissions). In the South, mortality improved for AMI and CHF patients (from 104 to 67 deaths per 1000 admissions and from 55 to 28 deaths per 1000 admissions, respectively; P<.05 for all differences). Mortality among Medicare hospitalizations for AMI decreased by 38% between 2000 and 2007 (from 106 deaths per 1000 admissions to 66 deaths per 1000 admissions); in the same time period, rates for Medicaid AMI hospitalizations decreased by 27% (from 103 deaths per 1000 admissions to 75 deaths per 1000 admissions). Mortality among AMI patients who were privately insured decreased by 32% (from 98 deaths per 1000 admissions to 67 deaths per 1000 admissions). The risk-adjusted mortality for AMI patients who were uninsured also decreased, from 133 to 93 deaths per 1000 admissions from 2000 to 2007 (P≤.05 for ifferences).