Adult patients without health insurance admitted to intensive care units (ICUs) in Pennsylvania hospitals are at a 21 percent increased risk of death compared to similar patients with private insurance, according to researchers from the University of Pennsylvania. The difference in mortality risk was not explained by patient characteristics or differences in care at the hospital level, suggesting that uninsured patients might receive poorer quality care.
The findings will be presented at the ATS 2010 International Conference in New Orleans.
Compared to similar patients with private insurance or Medicaid, uninsured ICU patients were also less likely to receive certain common critical care procedures, including placement of central venous catheters, tracheostomies and acute hemodialysis.
"Previous studies suggested that uninsured critically ill patients may have a higher mortality, and may be less likely to receive certain critical care procedures. But we found that these differences are primarily due to differences in quality within hospitals rather than across hospitals," said Sarah M. Lyon, M.D., pulmonary and critical care fellow at the Hospital of the University of Pennsylvania. "The higher mortality for uninsured patients does not appear to be caused by uninsured patients tending to go to hospitals with poor overall quality. Instead, we found that even when admitted to the same hospitals, and controlling for other differences between patients, critically ill individuals without insurance are less likely to survive than those with private or Medicaid insurance."
Dr. Lyon and colleagues analyzed 30-day mortality, and the use of several key ICU procedures, in all adult patients under 65 admitted to Pennsylvania ICUs from 2005 to2006 using state hospital discharge data. They categorized the 166,995 patients as having private health insurance (67.7 percent), Medicaid (28.5 percent), or being uninsured (3.8 percent.) When the researchers analyzed mortality at 30 days, they found that uninsured patients were 21 percent more likely to die than patients with private insurance; those with Medicaid had a 3 percent greater risk of death. Only the mortality difference between private insurance and uninsured patients was statistically significant.
"Our findings suggest that ICU patients without insurance have a higher risk of death and receive less intense treatment in the ICU. Expanding and standardizing health care coverage through health care reform may improve outcomes in critically ill patients," said Dr. Lyon. "We still do not understand all the reasons for differences in survival between the insured and uninsured. Critically ill patients without insurance may arrive to the hospital in more advanced stages of illness, perhaps in ways we could not control for in our study. Patients without insurance may also have different preferences for intensity of care at the end of life, and may not wish to be kept alive on life support as long as patients with insurance. Another, more concerning explanation is that physicians and hospitals treat patients without insurance differently than those with insurance. More work is needed before we can say with certainty that treatment biases caused these results."