The Neiman Imaging Comorbidity Index (NICI) outperforms the Charlson Comorbidity Index (CCI) for predicting advanced imaging use, according to a team from the Harvey L. Neiman Health Policy Institute (HPI) in Reston, VA.
The findings underscore the efficacy of the index for forecasting advanced imaging across numerous different payers, wrote lead author Casey Pelzl and colleagues.
“[Our study] assessed the broader generalizability of NICI for predicting receipt of advanced imaging in nationally representative populations, including patients insured by Medicare, Medicaid, and private payers,” they wrote.
Radiology health policy and economics research tends to use receipt of advanced imaging as an outcome within statistical models, the team explained, noting that the NICI “was originally developed to provide researchers with a risk-adjustment tool that could accurately predict a patient’s chances of undergoing advanced imaging based on comorbidity burden.” The tool was developed and validated with more than 10 million private payer claims from one insurance provider; it was shown to be better able to predict an individual’s odds of undergoing advanced imaging compared with the CCI (which was developed using data from an all-female population with breast cancer to assess how various comorbidities might affect a woman’s short-term mortality risk after hospital discharge).
For this study, Pelzl’s group investigated the NICI’s efficacy with Centers for Medicare and Medicaid Services (CMS) data from 2018 and 2019 from the Medicare 5% Research Identifiable File, CMS Medicaid 100% Research Identifiable File, and private insurance (commercial and Medicare Advantage) claims from market research firm Inovalon Insights. The research included a total of 108 million beneficiaries, of which 2.5 million were covered by Medicare, 49.6 million by Medicaid, and 56.6 million by private insurance.
The team used 2018 comorbidity data to assign beneficiaries to either the CCI or NICI tool, then calculated the area under the receiver operator characteristic curves (AUCs) for predicting advanced imaging in 2019. The authors also compared AUCs for both the NICI and the CCI across age groups after adjusting for age and sex.
The NICI outperformed the CCI across all types of insurance:
| Comparison of NICI with CCI for predicting use of advanced imaging* | ||
|---|---|---|
| Type of insurance | CCI | NICI |
| Medicare | 0.75 | 0.77 |
| Medicaid | 0.68 | 0.69 |
| Private insurance | 0.65 | 0.67 |
| *All results are statistically significant | ||
The researchers also found that the NICI outperformed the CCI in adjusted models and in nearly all age strata across the three cohorts, they wrote.
The findings are promising, but more study is needed, according to Pelzl and colleagues.
“Validation data support [the] NICI as the preferred index to adjust for patient comorbidities when studying advanced imaging as an outcome, but further investigations are warranted,” they concluded.
The complete study can be found here.