Treatment Limitation and Risk-Adjusted Mortality Benchmarking in 190 Intensive Care Units in the United States

Introduction: Intensive care units (ICUs) are often ranked using standardized mortality ratios derived from the Acute Physiology and Chronic Health Evaluation (APACHE) IVa model, which does not include a term for treatment limitation.We assessed three primary questions: (1) whether APACHE IVa calibration depended on documented treatment limitation, (2) whether between-unit variation in treatment limitation persisted after balancing measured case mix, and (3) whether accounting for treatment limitation changed unit rankings.Methods: This was a retrospective cohort study of 136,236 adult admissions to 190 US ICUs included in the eICU Collaborative Research Database (2014)(2015).The most restrictive treatment limitation documented within 24 hours and its timing were classified into categories, ranging from full therapy to comfort measures only.Units in the highest and lowest quintiles of documented treatment limitation were compared before and after inverse probability weighting based on 12 characteristics.Units were ranked according to their standardized mortality ratios, with and without modeled treatment limitation.Results: Treatment limitation was documented in 13,496 (9.9%) of 136,236 admissions.The standardized mortality ratio was 0.62 (95% confidence interval: 0.60-0.63)under full therapy and 1.87 (1.73-2.02)under comfort measures, regardless of whether the limitation was documented within four hours (2.03) or later (1.84).The highest quintile documented treatment limitation 3.58 times as often as the lowest quintile before weighting and 2.89 times as often after weighting.Modeling treatment limitation changed the quartile assignment of 38 (23.5%) of 162 units and the outlier status of 13 (8.0%).Excluding the 878 (0.6%) admissions involving comfort measures reclassified 18 (11.1%)units.Conclusions: APACHE IVa calibration was poorer as treatment limitation became more restrictive.Betweenunit variation in documented treatment limitation persisted after balancing measured case mix, and accounting for treatment limitation changed the ranking quartile of 38 (23.5%) of 162 units.Standardized mortality ratios may partly reflect differences in treatment limitation as well as quality of care.

Authors

Institutions

Publication Details

Journal
Cureus
Published
2026-11-12
DOI
https://doi.org/10.7759/cureus.119580
Primary Topic
Palliative Care and End-of-Life Issues
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Treatment Limitation and Risk-Adjusted Mortality Benchmarking in 190 Intensive Care Units in the United States

January G. Msemakweli, Fabrice Tiku Nyambod, Murishi Onesphore, Aja Wagoh
Cureus
Palliative Care and End-of-Life Issues
article

Treatment Limitation and Risk-Adjusted Mortality Benchmarking in 190 Intensive Care Units in the United States

January G. Msemakweli, Fabrice Tiku Nyambod, Murishi Onesphore, Aja Wagoh
article en

Abstract

Introduction: Intensive care units (ICUs) are often ranked using standardized mortality ratios derived from the Acute Physiology and Chronic Health Evaluation (APACHE) IVa model, which does not include a term for treatment limitation.We assessed three primary questions: (1) whether APACHE IVa calibration depended on documented treatment limitation, (2) whether between-unit variation in treatment limitation persisted after balancing measured case mix, and (3) whether accounting for treatment limitation changed unit rankings.Methods: This was a retrospective cohort study of 136,236 adult admissions to 190 US ICUs included in the eICU Collaborative Research Database (2014)(2015).The most restrictive treatment limitation documented within 24 hours and its timing were classified into categories, ranging from full therapy to comfort measures only.Units in the highest and lowest quintiles of documented treatment limitation were compared before and after inverse probability weighting based on 12 characteristics.Units were ranked according to their standardized mortality ratios, with and without modeled treatment limitation.Results: Treatment limitation was documented in 13,496 (9.9%) of 136,236 admissions.The standardized mortality ratio was 0.62 (95% confidence interval: 0.60-0.63)under full therapy and 1.87 (1.73-2.02)under comfort measures, regardless of whether the limitation was documented within four hours (2.03) or later (1.84).The highest quintile documented treatment limitation 3.58 times as often as the lowest quintile before weighting and 2.89 times as often after weighting.Modeling treatment limitation changed the quartile assignment of 38 (23.5%) of 162 units and the outlier status of 13 (8.0%).Excluding the 878 (0.6%) admissions involving comfort measures reclassified 18 (11.1%)units.Conclusions: APACHE IVa calibration was poorer as treatment limitation became more restrictive.Betweenunit variation in documented treatment limitation persisted after balancing measured case mix, and accounting for treatment limitation changed the ranking quartile of 38 (23.5%) of 162 units.Standardized mortality ratios may partly reflect differences in treatment limitation as well as quality of care.

Cureus
University of Buea (CM), Johns Hopkins University (US)
Palliative Care and End-of-Life Issues
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon