Objectives: Variation in repeated blood pressure measurements may represent a decline in homeostatic mechanisms in blood pressure regulation in response to various internal or external stressors, indicating a frail state. We tested this hypothesis by examining the association between variability in repeated blood pressure measurements (BPV) and frailty status, adjusting for other confounding factors.
Design: A longitudinal cohort study.
Setting: Community centres in all three regions of Hong Kong.
Participants: 1156 community-living older adults aged 60 years and over participated in a community geriatric screening program with blood pressure measurements three times a week over one year. Participants were divided into three groups based on variability of repeated blood pressure measurements (low, medium, high) using machine learning methods.
Measurements: Frailty status was assessed using the FRAIL scale. Logistic regression was used to examine cross sectional association between frailty status and BPV adjusting for confounding factors, and also frailty transition with BPV.
Results: In multi-variate models adjusting for co-variates, high BPV was associated with frailty (OR 1.57; 95% CI 1.05-2.37) among all participants; however, this was only significant in women in subgroup analysis. Similar findings were observed when transition to a more frail state was examined over a twelve month period.
Conclusions: The findings of this study support the concept of physiological dysregulation underlying the frail state, and that BPV calculated using machine learning methods may be used as a biomarker of such dysregulation.
Keywords: Frailty; biomarker; blood pressure variability; physiological dysregulation.