Researchers Validate Imaging Comorbidity Index Across Insured Populations

Building on a previously published novel risk-adjustment score—the Neiman Imaging Comorbidity Index (NICI)—predictive of patients’ advanced imaging use, researchers have now published more extensive validation of the NICI to test its utility in datasets other than the claims dataset from which it was developed. This multi-pronged validation, published in the Journal of the American College of Radiology, included records from more than 100 million patients across three nationally representative claims datasets. The patient sample was comprised of insured populations, covered by Medicare, Medicaid and private insurance plans.

“Prior to the development of the NICI, there was no validated score or index specifically designed to measure a patient’s likelihood of receiving advanced imaging – or predict their future chances of having a CT, MRI or PET scan. Absent an imaging-centric risk score, other risk scores such as the Charlson Comorbidity Index (CCI) were used,” said Eric Christensen, PhD, Research Director at the Neiman Institute. “In our initial development and internal validation of the NICI in a privately insured population, we found that it outperformed the commonly used CCI, which was designed to predict mortality risk in a cancer population. With this external validation, we sought to test the NICI’s effectiveness in predicting advanced imaging use across insured patient populations, and thus its generalizability to datasets other than the one in which it was developed.”

The results of this external validation showed that the NICI consistently outperformed CCI in predicting advanced imaging among Medicare, Medicaid, and privately insured beneficiaries. Consistent with findings from the initial NICI development, it provided the best performance in predicting advanced imaging outcomes in those aged 65-79 across all three payer datasets.

“NICI was most predictive of advanced imaging in older adults,” said Andrew Rosenkrantz, MD, MPA, professor in the Department of Radiology at NYU Grossman School of Medicine. “While NICI did not perform as well in pediatric patients, it nonetheless performed better than the CCI in this population.”

“Our results show that the NICI is a highly useful, rigorously validated risk-adjustment tool for studying imaging outcomes among insured populations in the US,” remarked Casey Pelzl, lead author and Senior Economics and Health Services Analyst at the Neiman Institute. “Moving forward, our research will employ the NICI in place of the CCI to control for a patient’s comorbidities when studying imaging outcomes.”

The team at HPI has made the NICI algorithm publicly available on the Neiman Institute website for use by other researchers, with the aim of advancing all efforts to study the value of radiology in patient care and promote the effective and efficient use of imaging.