AUTHOR=Garland Allan , Marrie Ruth Ann , Wunsch Hannah , Yogendran Marina , Chateau Daniel 

TITLE=Administrative Data Is Insufficient to Identify Near-Future Critical Illness: A Population-Based Retrospective Cohort Study

JOURNAL=Frontiers in Epidemiology

VOLUME=Volume 2 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/epidemiology/articles/10.3389/fepid.2022.944216

DOI=10.3389/fepid.2022.944216

ISSN=2674-1199

ABSTRACT=BACKGROUND: Prediction of future critical illness could render it practical to test interventions seeking to avoid or delay the coming event.
OBJECTIVE: Identify adults having >33% probability of near-future critical illness. 
RESEARCH DESIGN:  Retrospective cohort study, 2013-2015.
SUBJECTS: Community-dwelling residents of Manitoba, Canada, aged 40-89 years.
MEASURES: The outcome was near-future critical illness, defined as intensive care unit admission with invasive mechanical ventilation, or non-palliative death occurring 30-180 days after April 1st of each year.  Dividing data into Training and Test cohorts, Classification and Regression Tree analysis was used to identify subgroups with ≥33% probability of the outcome.  We considered 72 predictors including socio-demographics, chronic conditions, frailty, and health care utilization.  Sensitivity analysis used logistic regression methods.
RESULTS: 0.38% of each yearly cohort experienced near-future critical illness. The optimal Tree identified 2644 mutually exclusive subgroups.  Socioeconomic status was the most influential variable, followed by nursing home residency and frailty; age was sixth.  In the Training data, the model performed well; 41 subgroups containing 493 subjects had ≥33% members who developed the outcome. However, in the Test data those subgroups contained 429 individuals, with 20 (4.7%) experiencing the outcome, comprising 0.98% of all subjects with the outcome.  While logistic regression showed less model overfitting, it likewise failed to achieve the stated objective.
CONCLUSIONS: High-fidelity prediction of near-future critical illness among community-dwelling adults was not successful using population-based administrative data. Additional research is indicated to ascertain whether inclusion of additional types of data can achieve this goal.