Functional data analysis of heart rate variability from continuous ECG monitoring in older adults with and without mild cognitive impairment

Background Reduced heart rate variability (HRV) has been associated with cognitive decline in older adults. However, prior research relied on brief in-clinic electrocardiography (ECG) recordings to measure HRV. Using 7-day continuous ambulatory ECG monitoring, we investigated time-specific differences in HRV (i.e., differences in HRV at each time point over the course of a 24-h day) between individuals with mild cognitive impairment (MCI) and those who were cognitively normal (CN) in a cohort of community-dwelling older adults. Methods Analyses included 81 dementia-free participants [mean age = 78, standard deviation (SD) = 5, age range = 72–95; 82% female; 38% non-Hispanic White individuals, 43% non-Hispanic Black individuals]. Among them, 20 met the Jak/Bondi criteria for MCI. Participants were instructed to wear a single-lead ECG monitor continuously for 7 days. Power spectral analyses were used to determine HRV in the high-frequency band (0.15–0.40 Hz, HF-HRV) over consecutive 5-min epochs throughout the recording. Functional additive mixed models were used to analyze participants’ 24-h HF-HRV profiles to examine time-specific HRV differences between MCI and CN, after adjusting for age, sex, ethnicity, and education and further adjusting for depression, history of diabetes, and hypertension. Results Compared to the CN group, the MCI group showed reduced HRV in the early morning (before 7 a.m.) and evening (after 7 p.m.), with the greatest difference occurring around midnight (difference: 0.6, 95% pointwise CI: 0.2, 1.1, Cohen’s d : 0.75). Conclusion Our findings highlight HRV’s dynamic nature and the need to consider the time of day when investigating the relationship between HRV and cognition. Compared to daytime HRV, reduced nighttime HRV may have a stronger association with worse cognition.