Validity of a triaxial accelerometer and simplified physical activity record in older adults aged 64–96 years: a doubly labeled water study

Background The aim was to examine the validity of a triaxial accelerometer (ACCTRI) and a simplified physical activity record (sPAR) in estimating total energy expenditure (TEE) and physical activity level (PAL) in older adults with the doubly labeled water (DLW) method. Methods A total of 44 Japanese elderly individuals (64–96 years), of which 28 were community-dwelling healthy adults with or without sporting habits (S or NS group) and 16 were care home residents with frailty (F group), were included in the study. Basal metabolic rate (BMR) was measured by indirect calorimetry, TEE was obtained by the DLW method, and PAL was calculated as TEE/BMR. Daily step count was monitored by a pedometer (Lifecorder). The 24-h average metabolic equivalent was assessed by ACCTRI and sPAR. Results The TEE DLW in men was 2704 ± 353, 2308 ± 442, and 1795 ± 338 kcal d −1 , and that in women was 2260 ± 208, 1922 ± 285, and 1421 ± 274 kcal d −1 for the S, NS, and F groups, respectively. ACCTRI and sPAR systematically underestimated actual TEE (− 14.2 ± 11.6 and − 15.3 ± 12.3% for ACCTRI and sPAR, respectively). After diet-induced thermogenesis was taken into account for ACCTRI and sPAR, TEE DLW was significantly correlated with TEE ACCTRI ( R 2 = 0.714) and TEE sPAR ( R 2 = 0.668). PAL DLW was also significantly correlated with PAL ACCTRI ( R 2 = 0.438) and PAL sPAR ( R 2 = 0.402). Conclusions Age, living conditions, frailty, and sporting habits contribute to TEE and PAL in the elderly population. ACCTRI and sPAR underestimated TEE and PAL, and adequate corrections are required. The corrected ACCTRI and sPAR are both useful tools to estimate TEE and PAL.