Abstract Introduction. Recently studies showed that a persons aged 65 years and older with body mass index (BMI) ≥ 25 kg/ m2 have the same or lower risk of mortality than older adults with a normal BMI.Objective: to investigate the relationship between baseline body mass index (BMI), total skeletal muscle mass (SMM), muscle strength and all-cause mortality in the Russian population.Methods. The Crystal study is a prospective cohort study of a random sample of people aged 65 years and older.The total follow-up period was 9 years. Main parameters: BMI, nutritional status, anemia, C-reactive protein, skeletal muscle mass (SMM), functional status, depression, dementia and non-communicable diseases.Results. Participants with BMI ≥ 25 kg/m2 had a 34,6% lower risk of all-cause mortality during 5 years of follow-up and a 36,8% lower risk during 9 years of follow-up, regardless of nutritional status, chronic comorbidities, and functional status. This association was linked with higher SMM in participants with BMI ≥ 25 kg/m2. After adjusting for all our covariates, higher SMM, but not BMI, was associated with a 7,4% decrease in mortality during the 5-year follow-up.Conclusion. It is necessary to consider qualitative indicators of body composition, such as SMM and fat percentage during assessing the impact of BMI on mortality. The higher SMM is an independent factor associated with a 7,4% decreasing the risk of all-cause mortality over 5 years of follow-up in the population of people aged 65 years and older.