- Design
- five-year prospective cohort study with Cox models, restricted cubic splines and counterfactual analysis
- Population
- 13,343 community-dwelling Japanese adults aged 65 or over, disability-free at baseline
- Primary outcome
- incident disability, defined as new Long-Term Care Insurance eligibility
- Effect
- lowest versus highest quartile of estimated basal metabolic rate, hazard ratio 2.8 (95% CI 2.4 to 3.1)
Thirteen thousand three hundred and forty-three community-dwelling Japanese adults aged 65 or over, free of disability at baseline, had basal metabolic rate estimated by the revised Harris-Benedict equation — from age, sex, height and weight — and were followed for five years, with incident disability defined as new eligibility for Japan's Long-Term Care Insurance system.
The lowest quartile of estimated basal metabolic rate carried 2.8 times the risk of incident disability compared with the highest (95% CI 2.4 to 3.1). The association was consistent across restricted cubic spline modelling, survived competing risk analysis, multicollinearity assessment and sensitivity analyses excluding stroke and Parkinson's disease, and a counterfactual analysis suggested that a hypothetical 50 kcal/day lower estimated rate was associated with increased risk.
The authors are unusually careful about what their variable is, and the care is the most valuable part of the paper. Estimated basal metabolic rate is not measured metabolism — it is an equation containing age, sex, height and weight, so a low value largely means older, smaller and lighter. That is why it predicts disability: it is a compact composite of the anthropometric vulnerability that geriatricians already recognise, not a window into mitochondrial function.
Read that way, it is still useful. In a clinic with no grip dynamometer, no gait mat and no bioimpedance, a number derived from the height and weight already recorded identifies a group at nearly threefold risk of losing independence within five years. It is a flag for a comprehensive geriatric assessment, not a diagnosis, and it should not displace the direct measures of function — gait speed, grip, chair stands — where those are available.
- Use it as a triage flag where direct functional measures are unavailable, not as a replacement for them
- Remember what drives a low value: older, shorter and lighter — so act on weight loss and sarcopenia, not on metabolism
- Prefer gait speed, grip strength and chair-stand time when you can measure them; they are closer to the outcome
- Record serial height and weight; the trajectory will be more informative than a single value
- Follow a flagged patient with a comprehensive geriatric assessment, nutrition review and resistance exercise advice
The statistics, in plain English
A hazard ratio of 2.8 with a tight interval of 2.4 to 3.1 across 13,343 people is a strong and precisely estimated association — but precision says nothing about causation, and here the exposure is an equation built from age, sex, height and weight. Adjusting for those components would substantially attenuate or eliminate the effect, which is why the authors describe it as an integrated anthropometric marker. The counterfactual analysis of a 50 kcal/day difference is a modelling exercise, not an intervention: nothing was changed in anyone.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for geriatrics, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free