After an ICU Stay, Dual-Eligible Beneficiaries Have More Disability

It is already known that dual-eligible older adults are at an increased risk for death in the year following a stay in the intensive care unit (ICU), but it previously wasn't known whether dual-eligible status is linked with functional decline after an ICU stay.

Being dual eligible—that is, eligible for coverage under both Medicare and Medicaid—was linked with a 30% increase in the number of disabilities after a stay in an intensive care unit (ICU) compared with those who did not have Medicaid.

“This finding has significant implications because disability not only compromises quality of life, but also leads to increased institutionalization, need for long-term support services, and mortality,” said lead author Snigdha Jain, MD, a postdoctoral fellow at the Yale University School of Medicine, in a presentation Wednesday at the ATS 2021 International Conference.

It is already known that dual-eligible older adults are at an increased risk for death in the year following an ICU stay, but it isn’t known whether dual-eligible status is linked with functional decline after an ICU stay.

Jain and fellow researchers used identified older adults in the National Health and Aging Trends Study (NHATS), a nationally representative survey of Medicare beneficiaries aged 65 years and older, who underwent annual assessments of disability in 7 functional activities. These adults lived independently (ie, were not in a nursing home).

ICU stays were identified using critical care revenue codes in linked fee-for-service Medicare claims.

They used a negative binomial Poisson model to evaluate the association between dual eligibility and the count of disabilities (range, 0-7) assessed in the NHATS interview following discharge from the ICU hospitalization. Covariates included age, gender, education, living alone, frailty, hospital length of stay, mechanical ventilation, and the pervious count of disabilities before the ICU stay.

After survey-weighting, they identified 641 participant-ICU stays representing 3,767,695 ICU hospitalizations.

The mean age of the beneficiaries was 79.1 years. Compared with those who were only on Medicare, dual-eligible beneficiaries were more frequently male and:

  • Non-White (38.7% vs 13.1%)
  • Had less education (52% had less than high school vs 17.8%)
  • Lived alone (43.6% vs 32.1%)

The median (interquartile range) post-ICU disability count was 2.18 (0.00-4.83) for dual-eligible beneficiaries and 0.01 (0.00-2.47) for those without Medicaid.

Unadjusted, Medicaid eligibility was strongly associated with post-ICU disability, with a 60% increase in post-ICU disability count compared with those who did not have Medicaid (unadjusted rate ratio [RR], 1.60; 95% CI, 1.29-1.99).

After adjusting for the covariates, Medicaid status was still positively linked with having more disability after an ICU stay compared with those not on Medicaid (adjusted RR, 1.32; 95% CI 1.04-1.67).

Aluko A. Hope, MD, a critical care specialist at Montefiore Medical Center, who moderated the session at which Jain presented, asked her if the team had considered the effect of state policies on this issue.

“Politically in the US, different states are doing different things with Medicaid,” he noted. “I’m wondering whether you had any opportunities to kind of adjust for state effects in this model, and what you then think the policy implications of this work would be in terms of addressing some of the disability outcomes in this study?”

Jain responded that they did not, but that it might be something that should be looked at in the future, as Medicaid is the primary payer for long-term care services that some of these adults ultimately require.

They also do not have the same access to outpatient rehabilitation services or rehabilitation in the home, which might forestall or prevent decline, she said.

Reference

Jain S, Murphy TE, O’Leary JR. The effect of socioeconomic disadvantage on development of functional decline following critical illness among older adults. Presented at: ATS 2021 International Conference; May 14-19, 2021. Abstract A1214