Existing evidence points to substantial gaps in detecting mild cognitive impairment in primary care but is based on limited or self-reported data. The recent emergence of disease-modifying treatments for the Alzheimer’s disease, the most common etiology of mild cognitive impairment, calls for a systematic assessment of detection rates in primary care.

The current study aims to examine detection rates for mild cognitive impairment among primary care clinicians and practices in the United States using Medicare claims and encounter data.

Observational study.

Medicare administrative data.

The study sample includes a total of 226,756 primary care clinicians and 54,597 practices that had at least 25 patients aged 65 or older, who were enrolled in Medicare fee-for-service or a Medicare Advantage plan between 2017 and 2019.

The detection rate for mild cognitive impairment is assessed as the ratio between the observed diagnosis rate of a clinician or practice as documented in the data, and the expected rate based on a predictive model.

The average detection rates for mild cognitive impairment is 0.08 (interquartile range=0.00–0.02) for both clinicians and practices, suggesting that only about 8% of expected cases were diagnosed on average. Only 0.1% of clinicians and practices had diagnosis rates within the expected range.

Mild cognitive impairment is vastly underdiagnosed, pointing to an urgent need to improve early detection in primary care.