SAN DIEGO – There is no apparent relationship between the neighborhood poverty rate, based on patient address, and mortality following critical care, results from a large, 10-year analysis showed.
"Our findings are in contrast to data in other arenas of health care that have established an inverse relationship between socioeconomic status and mortality," Sam Zager said at the annual congress of the Society of Critical Care Medicine. "The few studies that examine economic disparities and mortality in the critically ill are contradictory."
Using 1990 census and hospital administration data, Mr. Zager, a fourth-year student at Harvard Medical School, Boston, and his associates performed an observational study of 38,917 patients aged 18 years and older who received critical care at Brigham and Women’s Hospital and Massachusetts General Hospital, both in Boston, in 1997-2007.
Neighborhood poverty rate was defined as the percentage of each neighborhood’s residents with incomes below the federal poverty line, categorized as 5%-10%, 10%-20%, 20%-40%, or greater than 40%. They used logistic regression to examine death by day 30, 90, and 365 post ICU, as well as in-hospital mortality, and adjusted the data for age, sex, race, admission year, patient type (medical vs. surgical), Charlson-Deyo index, sepsis, CABG, myocardial infarction, hematocrit, white blood cell count, creatinine, and blood urea nitrogen.
The researchers also performed a sensitivity analysis for 1-year postdischarge mortality among patients discharged to home, as well as mortality among patients who lived less than 50 miles from the hospital of care.
The mean age of patients was 62 years, 42% were women, and 78% were white. After multivariable adjustment of the data, Mr. Zager and his associates found no statistically significant relationship between neighborhood poverty rate and all-cause 30-day mortality. The odds ratio was 1.05 for those who resided in neighborhoods in which 5%-10% of residents lived below the federal poverty line (P = .2), 0.96 for those who resided in neighborhoods in which 10%-20% of residents lived below the federal poverty line (P = .5), 1.08 for those who resided in neighborhoods in which 20%-40% of residents lived below the federal poverty line (P = .2), and 1.20 for those who resided in neighborhoods in which more than 40% of residents lived below the federal poverty line (P = .2).
Similar nonsignificant associations were observed for 90-day and 365-day mortality post ICU admission and for in-hospital mortality. In addition, neighborhood poverty rate was not significantly associated with 1-year postdischarge mortality in patients who were discharged to home or in patients who resided less than 50 miles from the hospital of care.
Mr. Zager also reported that patients from neighborhoods in which 20% or more of residents lived below the federal poverty line were more likely to be black, Hispanic, or young; to have a hematocrit of less than 36%; and to live 5 miles or less from the hospital.
He acknowledged certain limitations of the study, including its observational design and the fact that the researchers were unable to fully exclude patients who received critical care only in the emergency department. Also, "our study focuses on neighborhood poverty at the time of critical care initiation, which may not fully reveal the contribution of socioeconomic status to mortality risk," he said.
The study was supported by the National Institutes of Health. The researchers said that they had no relevant financial conflicts to disclose.