Abstract & Commentary: Insurance Status and ICU Outcomes
Abstract & Commentary
Insurance Status and ICU Outcomes
By David J. Pierson, MD, Editor, Professor Emeritus, Pulmonary and Critical Care Medicine, University of Washington, Seattle, is Editor for Critical Care Alert.
SYNOPSIS: In this study of a statewide administrative database, among 138,720 adult patients admitted to an ICU, uninsured patients had a 25% higher likelihood of death within 30 days compared with privately insured patients, differences that persisted after multiple adjustments for demographics, severity of illness, and site of care. Uninsured patients received central venous catheterization, acute hemodialysis, and tracheostomy significantly less often than insured patients.
SOURCE: Lyon SM, et al. The effect of insurance status on mortality and procedural use in critically ill patients. Am J Respir Crit Care Med 2011; 184:809-815.
Lyon and colleagues performed a retrospective cohort study of the relationship between insurance status and 30-day mortality, as well as the use of five common ICU procedures, among 138,720 adult patients admitted to ICUs in Pennsylvania in fiscal years 2005 and 2006. The primary information source was discharge data from the Pennsylvania Health Care Cost Containment Council, a nonprofit government agency that tracks all hospitalization in the state. Patients aged 65 years and older, as well as those with Medicare, military, or VA insurance were excluded, and only initial ICU admissions were used. The authors performed extensive epidemiological and statistical procedures to exclude confounders such as demographics, economic status (from zip codes), admission source and type, primary diagnosis, comorbidities, mechanical ventilation, and severity of illness on presentation. The primary outcome was 30-day mortality, and secondary outcomes were the use of central venous catheterization (CVC), pulmonary artery catheterization (PAC), acute hemodialysis, tracheostomy, and bronchoscopy.
Of the 138,720 qualifying patients, 69.2% were privately insured, 26.6% had Medicaid, and 4.2% (5814 individuals) were uninsured. Uninsured patients, as well as those on Medicaid, were widely distributed among the 169 hospitals and differed from each other. Uninsured patients were more often admitted to small community hospitals, while Medicaid patients were more often admitted to larger academic hospitals.
Absolute 30-day mortality was 4.6% for insured patients, 5.7% for uninsured patients, and 6.4% for Medicaid patients. The unadjusted odds ratio (OR) for death among patients without insurance was 1.26 (95% confidence interval [CI], 1.12-1.41; P < 0.001) compared to patients with insurance. This increased odds of death persisted after adjustment for all the variables used (OR, 1.25; 95% CI, 1.04-1.51; P = 0.020). Medicaid patients also had increased 30-day mortality (OR, 1.42; 95% CI, 1.35-1.50, P < 0.001), but this difference disappeared with adjustment for patient characteristics. The absolute risk difference for death, uninsured vs insured patients, was 0.01 (P = 0.011), meaning that for every 1000 patients admitted to an ICU in Pennsylvania during the study period, there would be 10 more deaths if all those patients were uninsured.
Uninsured patients were significantly less likely to receive CVC (OR, 0.84; 95% CI, 0.72-0.97; P = 0.018), acute hemodialysis (OR, 0.59; 95% CI, 0.39-0.58; P = 0.016), and tracheostomy (OR, 0.43; 95% CI, 0.29-0.64; P < 0.001). Differences for PAC and bronchoscopy were nonsignificant. Medicaid patients were significantly more likely to receive CVC, acute hemodialysis, and tracheostomy, and these differences persisted with the adjustments; differences for PAC and bronchoscopy were nonsignificant.
COMMENTARY
Compared to patients with private insurance, uninsured patients who are hospitalized acutely have higher overall mortality, but whether this is due to patient factors other than the acute illness (such as socioeconomic factors, comorbidities, or higher severity of illness on presentation), or to being treated differently in the hospital, is unclear. Further, whether any differences in outcomes are due to between-hospital differences (that is, the uninsured tending to be admitted to hospitals that provide poorer care) or within-hospital differences (that is, the uninsured tending to be treated differently from insured patients at the same hospital) has not previously been investigated. This study by Lyon et al carefully excluded differences in patient characteristics, accounting for within-hospital care and outcomes — and showed that uninsured ICU patients still had higher mortality. Such was not the case with the Medicaid patients, another group with known poorer outcomes, for whom the mortality differences disappeared when patient characteristics were considered. The authors also demonstrated that, other factors being equal, uninsured patients were less likely to get a CVC or a tracheostomy, or to undergo acute hemodialysis, than their counterparts with private insurance —again, these differences not being due to being less sick or cared for in different hospitals.
The findings of this study are unlikely to please clinicians. Surely uninsured patients have worse outcomes because they neglect their health in addition to not buying health insurance, and have more comorbidities; they wait until they are more seriously ill, and thus less likely to recover, before seeking medical attention; and they tend to be hospitalized in lower-quality institutions that have worse outcomes for everyone. According to this study, which examined those issues and others, none of the above assumptions is true. Uninsured patients, cared for in the same ICUs of the same hospitals as insured patients, undergo at least three common critical care procedures less frequently and are more likely to die. Lyon et al make no attempt to link the last two findings, since proof that ICU procedures improve outcomes is generally lacking. They also do not attempt to explain why, within a given institution, uninsured patients might be treated differently, whether for institutional or health system reasons, or because of the actions of individual clinicians, or for some other reason. That important question must await further research.
Lyon and colleagues performed a retrospective cohort study of the relationship between insurance status and 30-day mortality, as well as the use of five common ICU procedures, among 138,720 adult patients admitted to ICUs in Pennsylvania in fiscal years 2005 and 2006.Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.