ABSTRACT
ABSTRACT BACKGROUND: Environmental and population characteristics seem to influence the variation in cutoff points of the Short Physical Performance Battery (SPPB) for diagnosing frailty syndrome among older adults. OBJECTIVE: To verify the validity of the SPPB for screening for frailty syndrome among older adults in the Amazonian context. DESIGN AND SETTING: Cross-sectional population-based study on older adults in the urban area of Coari (AM), Brazil. METHODS: In total, 264 older adults (60 years of age or over) were included. Frailty syndrome was defined using the Fried phenotype criteria. The SPPB cutoff points were compared in relation to frailty and validity measurements were calculated for the test. RESULTS: A strong association between poor physical performance and frailty was identified (P < 0.001). The cutoff point of 6 demonstrated the best validity measurements for frailty in the sample studied (sensitivity: 0.28; specificity: 0.94; accuracy: 0.88; area under the receiver operating characteristic curve, AUC-ROC: 0.61; likelihood ratio, LR+: 4.44; LR-: 0.77; prevalence: 8.3%; post-test probability, PTP+: 0.32; PTP-: 0.07), with emphasis on high specificity and the positive likelihood ratio value. CONCLUSION: The SPPB was shown to be useful for screening frail older adults in the Amazon region. The score of 6 demonstrated the best cutoff point for this population. This could be used in healthcare services for diagnostic screening for frailty among older people within the Amazonian context.