Is there agreement across diagnostic instruments in the identification of the impairment in intrinsic capacity later in life? A cross-sectional study with community-dwelling older adults

• Diagnostic instruments substantially impact the detection of impairment on IC domains. • There is moderate agreement across diagnostic instruments in the mobility domain. • There is moderate agreement across diagnostic instruments in the hearing domain. • There is a lack of agreement across diagnostic instruments in some domains of IC. To compare and investigate agreement across diagnostic instruments in detecting impairment in the domains of intrinsic capacity (IC). Cross-sectional descriptive design. Setting and participants: The study was conducted with 205 community-dwelling older adults (ages 60–96; mean: 71.3 ± 7.9 years). IC was assessed across five domains, using multiple diagnostic instruments for each domain: cognition [Mini-Mental State Examination (education-adjusted versus raw score of 26 points) and a cognitive battery]; mobility (Short Physical Performance Battery, gait speed test, chair stand test, and Timed Up and Go test); vitality (Mini Nutritional Assessment, handgrip strength, body mass index, and self-reported unintentional weight loss); psychological well-being (Geriatric Depression Scale, Center for Epidemiologic Studies Depression Scale, and self-reported of mental health diagnosis or use of antidepressant or anxiolytic medications); and sensory function (whisper voice test, Snellen chart, and self-reported hearing or visual impairment). McNemar’s test and Cochran’s Q-test were used to compare the detection of impairment in IC domains. Cohen’s Kappa and Fleiss’ Kappa tests were used to measure the level of agreement across diagnostic instruments. Detection of impairment in each IC domain varied substantially across diagnostic instruments. Moderate agreement was found in the diagnostic instruments for mobility (K = 0.64) and hearing (K = 0.69; agreement = 88.8%). Diagnostic instruments substantially influence the detection of impairment on IC. Moderate agreement was found for mobility and hearing measures, underscoring the urgent need for harmonized assessment protocols and highlighting the critical impact of instrument selection on estimating the prevalence of impairment on IC.