Characterizing hospitalization trajectories in the high-need, high-cost population using electronic health record...

Area of Science:

• Health Services Research
• Clinical Epidemiology
• Health Economics

Background:

• High healthcare utilization by a minority of patients significantly impacts overall costs.
• Understanding the dynamics of patient hospitalization patterns is crucial for cost containment.
• Longitudinal analysis of hospitalizations can reveal distinct patient trajectories.

Purpose of the Study:

• To characterize the longitudinal trajectories of hospitalization among adult patients.
• To identify distinct clusters of patient hospitalization patterns over time.
• To determine baseline factors associated with different hospitalization trajectories.

Main Methods:

• Retrospective cohort study of 3404 adult patients at an academic medical center (2017-2023).
• Growth mixture modeling was used to identify clusters of hospitalization trajectories after an initial "rising-risk" period.
• Analysis of baseline clinical and demographic factors associated with identified trajectories.

Main Results:

• Four distinct hospitalization trajectory clusters were identified: no further utilization, low chronic utilization, persistently high with slow increase, and persistently high with fast increase.
• Factors associated with higher utilization trajectories included nonsurgical service admission, full code status, ICU care, and opioid administration.
• Comorbidities such as cardiovascular disease, end-stage kidney or liver disease, and cancer were linked to higher hospitalization rates.

Conclusions:

• Patient hospitalizations follow identifiable longitudinal trajectories.
• Early identification of patients with specific risk factors can predict high-need, high-cost utilization.
• Characterizing these trajectories provides a foundation for targeted interventions and resource allocation.