Back to results

The University of Texas Medical Branch at Galveston

Evaluating Comorbidity Indices for Predicting Post-Acute Outcomes

Abstract

dc:description.abstract

Comorbidities are defined as acute or chronic medical conditions that an individual has in addition to his or her primary diagnosis. These comorbid conditions can affect a patient’s prognosis across the continuum of care. Functional decline and the presence of comorbidities are common in older adults. Functional decline can be precipitated by the presence of comorbid conditions during acute hospitalization. Thus, poor management of comorbidities can lead to undesired outcomes such as preventable hospital readmissions. A valid comorbidity index is needed to adjust for adverse effects of comorbidities on post-acute outcomes. However, validation of various comorbidity indices on post-acute health outcomes has not been investigated using administrative data. The overarching goal of this study was to compare the performances of five comorbidity indices, including Charlson Comorbidity Index, Elixhauser Comorbidity Index, the Functional Comorbidity Index, the Hierarchical Condition Category, and Centers for Medicare and Medicaid Services Tier categories for predicting post-acute-relevant health outcomes. The health outcomes studied included functional status, community discharge, and 30-day acute hospital readmission. Secondary analyses were conducted using 100% Medicare data for beneficiaries on fee-for-service plans in calendar year 2011. The Medicare Provider Analysis and Review file was linked to the Inpatient Rehabilitation Facilities-Patient Assessment Instrument file to retrieve admission and discharge Functional Independence Measure ratings. The Hierarchical Condition Category performed relatively better than the other comorbidity indices in predicting functional status at admission to post-acute inpatient rehabilitation, discharge functional gain during rehabilitation, and 30-day acute hospital readmission after discharge from rehabilitation. Our findings provide further evidence that medical diagnosis – including comorbidity burden – cannot be used as a proxy for patient’s functional status or ability to live independently in the community, the two most important patient-centered outcomes in post-acute care.

Degree

thesis:*
Name thesis:degree_name
Population Health Sciences (Doctoral)
Level thesis:degree_level
Doctoral
Grantor
The University of Texas Medical Branch at Galveston

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Kumar, Amit
Advisor dc:contributor.advisor
  • Ottenbacher, Kenneth J
Committee members dc:contributor.committeemember
  • Graham, James E
  • Karmarkar, Amol M
  • Tan, Alai
  • Resnik, Linda
  • Deutsch, Anne

Subjects

dc:subject × 1

Identifiers

dc:identifier.*
Handle dc:identifier.uri
https://hdl.handle.net/2152.3/651
OAI identifier oai:identifier
oai:utmb-ir.tdl.org:2152.3/651

Chain of custody

source
Harvested from
University of Texas Medical Branch
Base URL
utmb-ir.tdl.org/server/oai/request
Last updated
2026-07-24
Source record
OAI-PMH GetRecord
citation

Kumar, Amit. Evaluating Comorbidity Indices for Predicting Post-Acute Outcomes. Doctoral thesis, The University of Texas Medical Branch at Galveston, https://hdl.handle.net/2152.3/651