CCHA predicts health costs and hospitalizations in Cornell study
Cornell researchers say a 38-condition comorbidity score forecast hospital use and health spending for up to five years.
By Priya Raghavan · Science Reporter
3 min read
A Cornell-led study says CCHA can predict health costs and unplanned hospitalizations by scoring the burden of chronic disease across a patient population. The finding matters because unplanned hospital stays are a major driver of medical spending, and health systems often struggle to identify patients who need more intensive primary care before their conditions worsen.
The Charlson Comorbidity Health Analytics measure, developed by Dr. Mary Charlson of Weill Cornell Medicine, assigns scores across 38 chronic health conditions. Charlson is chief of the Division of Clinical Epidemiology and Evaluative Sciences Research and the William Foley Distinguished Professor of Medicine at Weill Cornell Medicine.
In a study published in PLOS One, Charlson and colleagues analyzed six years of anonymized claims data from more than 27,000 Weill Cornell Medicine employees and dependents. Cornell University reported that the tool outperformed other approaches for forecasting unplanned hospital admissions and could estimate future costs over the next five years.
What is CCHA?
CCHA is a population health scoring method for comorbidity, meaning the presence of multiple diseases or health conditions in one patient. Each chronic condition receives a point value based on severity, and a higher total score signals a more complex medical profile.
The researchers applied the score to adults and children covered in the Weill Cornell Medicine beneficiary data from 2016 through 2021. Cornell said the data set was unusually complete because it captured six continuous years of coverage, claims and costs for the participants.
How did CCHA scores relate to hospital stays and costs?
Patients with higher CCHA scores were more likely to be hospitalized, according to the researchers. Among adults with a score of zero in 2016, 1.2% were hospitalized that year; among adults with scores of eight or higher, 61% were hospitalized.
The same pattern appeared in spending. Cornell reported that adults with a score of zero had average costs of $3,835 in 2016, while costs rose with each added CCHA point and reached $53,189 for patients with scores of eight or higher.
A small share of patients accounted for a disproportionate share of spending. People with scores of five or higher made up 4.3% of patients in 2016 but represented 17.5% of total health care spending that year, according to the study.
The team also used earlier CCHA scores to project later annual costs and hospital admissions for as long as five years. Cornell said the measure predicted admissions better than a patient’s prior hospitalization history and performed better than models based only on past costs.
What could health systems do with the score?
The researchers said identifying high-need patients could help health systems target extra primary care and support services to people at elevated risk. Charlson’s team is already testing the approach at scale in several health systems to see whether longer primary care visits can reduce unplanned hospitalizations.
Co-author Martin Wells of Cornell’s Ann S. Bowers College of Computing and Information Science and the ILR School said patients with greater medical complexity often receive the same appointment length as healthier patients. Cornell said the study could support policy recommendations for care models that give more time and services to patients with complicated medical needs.
The paper, “Charlson comorbidity health analytics: A population management strategy to identify risk of hospitalizations, repeated hospitalizations, and resultant high cost,” was authored by Mary Charlson and colleagues and published in PLOS One.
This story draws on original reporting from Medical Xpress.