Predictors of Clinically Meaningful Gait Speed Response to Caloric Restriction Among Older Adults Participating in Weight Loss Interventions.

Abstract

The purpose of this study was to examine whether select baseline characteristics influenced the likelihood of an overweight/obese, older adult experiencing a clinically meaningful gait speed response (±0.05 m/s) to caloric restriction (CR). Individual level data from 1 188 older adults participating in 8, 5/6-month, weight loss interventions were pooled, with treatment arms collapsed into CR (n = 667) or no CR (NoCR; n = 521) categories. Exercise assignment was equally distributed across groups (CR: 65.3% vs NoCR: 65.4%) and did not interact with CR (p = .88). Poisson risk ratios (95% confidence interval [CI]) were used to examine whether CR assignment interacted with select baseline characteristic subgroups: age (≥65 years), sex (female/male), race (Black/White), body mass index (BMI; ≥35 kg/m2), comorbidity (diabetes, hypertension, cardiovascular disease) status (yes/no), gait speed (<1.0 m/s), or inflammatory burden (C-reactive protein ≥3 mg/L, interleukin-6 ≥2.5 pg/mL) to influence achievement of ±0.05 m/s fast-paced gait speed change. Main effects were also examined. The study sample (69.5% female, 80.1% White) was 67.6 ± 5.3 years old with a BMI of 33.8 ± 4.4 kg/m2. Average weight loss achieved in the CR versus NoCR group was -8.3 ± 5.9% versus -1.1 ± 3.8%; p < .01. No main effect of CR was observed on the likelihood of achieving a clinically meaningful gait speed improvement (risk ratio [RR]: 1.09 [95% CI: 0.93, 1.27]) or gait speed decrement (RR: 0.77 [95% CI: 0.57, 1.04]). Interaction effects were nonsignificant across all subgroups. The proportion of individuals experiencing a clinically meaningful gait speed change was similar for CR and NoCR conditions. This finding is consistent across several baseline subgroupings.