Physical performance is an essential component in the clinical assessment among older adults, and its decline as assessed by the Short Physical Performance Battery (SPPB) is associated with increased risk for hospitalization, institutionalization, falls, and disability. Although a SPPB score <10 seems to be predictive of mortality, according to previous studies, the cutoff values are heterogeneous, which makes it difficult to really know the predictive power of SPPB for mortality and the magnitude of the increase in the probability of death as the score decreases.

To analyze the predictive power of SPPB for mortality among older adults due to a systematic review with meta-analysis.

Systematic review with meta-analysis, prepared according to the Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA-P) recommendation, registered in the International Prospective Register of Systematic Reviews - Prospero (CRD42021256040). Prospective and retrospective longitudinal studies conducted with individuals aged 60 years or older were included, considering publications in full text, abstracts, and any identified unpublished data. The search was performed in the following databases with no language or date restrictions: MEDLINE via PubMed, Embase, Latin American and Caribbean Literature on Health Sciences (LILACS), Physiotand herapy Evidence Database (PEDro). The risk of bias was analyzed using the Quality in Prognosis (QUIPS) tool. For the meta-analysis, R software with the "meta" package (version 4.9-6), the "metaprop" function for proportion data and the "metamean" function for continuous data was used. Pooled results of proportion and means (continuous data) with their respective 95% confidence intervals (CI) were obtained using the inverse variance method with a random effects model. Heterogeneity was assessed by calculating i2. Values greater than 50% were considered substantial heterogeneity.

Meta-analysis including 13 studies with 6. 390 participants suggest that elderly with SPPB between 0-3 are more likely to die compared with those with SPPB between 4-12 [Odds Ratio (OR) 2.58; 95% CI (1.93-3.44); moderate certainty of evidence]; elderly with SPPB between 0-6 are more likely to die compared with those with SPPB between 7-12 [Odds Ratio (OR) 2.30; 95% CI (1.94-2.73); moderate certainty of evidence]; and elderly with SPPB between 0-9 are more likely to die compared with those with SPPB between 10-12 [Odds Ratio (OR) 2.17; 95% CI (1.75-2.68); high certainty of evidence].

The chance of death increases as the SPPB score decreases, which reinforces the predictive capacity of this variable. It is suggested the development of further studies with comparative analyses to establish a cutoff point from which SPPB score there is a higher risk of death compared to the general population, especially comparative analyses of interventions to improve the physical performance of older adults and prevent death.

The results may subsidize the development of clinical protocols aimed at improving physical performance, to be used in public health regarding the health management of the elderly population.