The prevalence of sarcopenia in community-dwelling older adults, an exploration of differences between studies and within definitions: a systematic review and meta-analyses

This is the Mayhew et al. systematic review and meta-analysis of sarcopenia prevalence in community-dwelling older adults, published in Age and Ageing in 2019 alongside the EWGSOP2 consensus statement (Cruz-Jentoft et al. 2019). It is the empirical companion to that conceptual reframing: where EWGSOP2 sets out how sarcopenia should be defined, Mayhew quantifies the consequences of the field's failure to converge on a single operational definition over the preceding decade.

The systematic review pooled 109 articles representing 58 distinct cohorts across 26 countries, applying eight different sarcopenia definitions in use at the time (EWGSOP 2010, AWGS, FNIH, IWGS, ESPEN-SIG, Janssen, Baumgartner, and modifications thereof). The headline finding is the dispersion. Sarcopenia prevalence across community-dwelling older adults ranged from 9.9% to 40.4%, with the position of any individual cohort within that range driven primarily by which diagnostic algorithm was applied rather than by the underlying biology of the population. Composite definitions that combined low muscle mass with low strength or low physical function consistently produced lower prevalence estimates (range 9.9%–18.6%). Mass-only definitions — particularly those using the Baumgartner appendicular lean mass index criterion — produced the highest estimates (24.2%–40.4%). The same individuals could be classified as sarcopenic or non-sarcopenic depending on which framework was applied.

Within definitions, the meta-analysis identified three sources of residual heterogeneity that survived the definitional adjustment. Participant age was the largest driver: prevalence rose steeply with cohort age, with community-dwelling adults aged 80 and older showing prevalence approximately three-fold that of cohorts aged 60–70 under the same diagnostic framework. Muscle-mass measurement modality contributed a non-trivial gap — DXA-derived estimates differed systematically from BIA-derived estimates within the same cohorts, generally with BIA producing higher prevalence figures. Choice of cutoff threshold for muscle mass (sex-specific cutoffs from one reference population versus another) produced additional dispersion within ostensibly the same definition.

The editorial significance for sarcopenia-and-peptides is that cross-trial and cross-population comparisons of sarcopenia treatment response are noisier than they look. Trial A reporting "12% prevalence of sarcopenia" and Trial B reporting "32% prevalence" may be sampling the same underlying population — the gap is methodological. Effect-size estimates for any peptide intervention on sarcopenia outcomes inherit the same dispersion: a 1-kg lean-mass gain has different clinical implications under a strength-first EWGSOP2 framework than under a mass-first 2010 framework. The strength-first reframing in EWGSOP2 was partly a response to the prevalence-dispersion problem Mayhew quantifies — anchoring the definition on strength was an attempt to reduce the methodology-driven variance and make the prevalence estimate more clinically actionable.

The paper also addresses one of the recurring claims in the public-health framing of sarcopenia. The often-cited figure that "approximately 10% of older adults have sarcopenia" maps reasonably well onto the composite-definition estimates (9.9%–18.6%) but is misleading when applied to the mass-only criteria (24.2%–40.4%); the same one-line claim has been used to support both prevalence framings depending on the rhetorical context. The Mayhew analysis recommends standardizing on the composite-definition family for clinical and public-health work, an alignment that EWGSOP2 institutionalized.