OBJECTIVES

: Predictive models for earlier diagnosis of Alzheimer’s disease and related dementias (ADRD) that rely on variables requiring assessment during an office visit, such as cognitive function, body mass index, or lifestyle factors, may not be broadly applicable as that level of data may be inaccessible. The objective of this study was to build a predictive model for earlier diagnosis of ADRD that would be practical for low-burden implementation for a larger provider such as an integrated delivery system or insurance plan by incorporating prior health care utilization (HCU) information found in administrative claims data.

METHODS

: We conducted a case-control study using data from the OptumLabs™ Data Warehouse. ADRD was defined using ICD-9 codes and prescription fills for anti-dementia medications. We included individuals with mild cognitive impairment and those with a prescription fill for anti-dementia medication but no ADRD diagnosis. Cases ≥18 years with a diagnosis between 2011-2014 were matched to controls without ADRD. HCU variables were incorporated into regression models along with comorbidities and symptoms.

RESULTS

The derivation cohort comprised 24,521 cases and 95,464 controls. Final adjusted models were stratified by age. We obtained moderate accuracy (C-statistic 0.76) for the model among younger (<65 years) adults and poorer discriminatory ability (C-statistic 0.63) for the model among older adults (≥65years). Neurological and psychological disorders had the largest effect estimates.

CONCLUSIONS

: We have created age-stratified predictive models for earlier diagnosis of ADRD using information available in administrative claims. These models could be considered for inclusion in electronic health records to promote cognitive screening and earlier dementia recognition.