Bidirectional transitions of sarcopenia states in older adults: The longitudinal evidence from CHARLS

Abstract

Background: Sarcopenia, the age-related loss of muscle mass and function, brings multiple adverse outcomes including disability and death. Several sarcopenia consensuses have newly introduced the premorbid concept of possible sarcopenia and recommended early lifestyle interventions. Bidirectional transitions of premorbid states have been revealed in several chronic diseases yet not clarified in sarcopenia. This study aims to investigate the underlying transition patterns of sarcopenia states.

Methods: The study utilized three waves of data from a nationally representative survey, the China Health and Retirement Longitudinal Study (CHARLS), and included community-dwelling individuals aged 60 years and older with at least two sarcopenia states assessments based on the Asian Working Group for Sarcopenia criteria 2019 (AWGS2019) between 2011 and 2015. The estimated transition intensity and probability between non-sarcopenia, possible sarcopenia, sarcopenia, and death were investigated using multi-stage Markov (MSM) models.

Results: The study comprised 4395 individuals (49.2% female, median age 67 years) with a total of 10 778 records of sarcopenia state assessment, and the mean follow-up period was 3.29 years. A total of 24.5% of individuals with a current state of possible sarcopenia returned to non-sarcopenia, 60.3% remained possible sarcopenia, 6.7% progressed to sarcopenia, and 8.5% died by the next follow-up. The transition intensity of recovery to non-sarcopenia (0.252, 95% CI 0.231-0.275) was 2.8 times greater than the deterioration to sarcopenia (0.090, 95% CI 0.080-0.100) for individuals with possible sarcopenia. For individuals with possible sarcopenia, the estimated probabilities of recovering to non-sarcopenia, progressing to sarcopenia, and transitioning to death within a 1-year observation were 0.181, 0.066, and 0.035, respectively. For individuals with sarcopenia, the estimated probabilities of recovering to non-sarcopenia, recovering to possible sarcopenia, and transitioning to death within 1-year observation were 0.016, 0.125, and 0.075, respectively. In covariables analysis, age, sex, body mass index, physical function impairment, smoking, hypertension, and diabetes are important factors influencing bidirectional transitions.

Conclusions: The findings highlight the bidirectional transitions of sarcopenia states among older adults and reveal a notable proportion of possible sarcopenia show potential for recovery in the natural course. Screening and intensifying interventions based on risk factors may facilitate a recovery transition.

Keywords: Multi‐state Markov model; Older adults; Sarcopenia; States transition.

Figure 1

The flow chart of screening for enrolled individuals.

Figure 2

The hypothesis of transitions between sarcopenia states. Non‐sarcopenia, possible sarcopenia, and sarcopenia are designated as transient states while death is the absorbing state. Transient states can move between adjacent states, but once an absorbing state is reached, no further transitions can occur.

Figure 3

Transition diagrams of sarcopenia states within 1, 2, 3, and 5‐year observation intervals. The state transition diagrams for 1‐year (A), 2‐year (B), 3‐year (C), and 5‐year (D) observation intervals based on the MSM model, respectively. To be specific, in (A), for individuals in possible sarcopenia states, the estimated probability of observing maintaining the current state (dashed line from the box of possible sarcopenia), recovery to non‐sarcopenia (blue line from the box of possible sarcopenia) and progression to sarcopenia (red line from the box of possible sarcopenia), or death (grey line from the box of possible sarcopenia) within 1 year was 0.717, 0.181, 0.066, and 0.035, respectively.

Figure 4

The estimated possibility of possible sarcopenia state transitions within different subgroups. The estimated possibility of possible sarcopenia recovering to non‐sarcopenia, remaining as possible sarcopenia, worsening to sarcopenia, or transitioning to death among age subgroups (A), BMI subgroups (B), and physical function impairment subgroups (C) within 5‐year observation interval, respectively. BMI, body mass index; D, death; N, non‐sarcopenia; PS, possible sarcopenia; S, sarcopenia.

Figure 5

The estimated sojourn time and predicted total stay of sarcopenia transient states. (A) the estimated mean sojourn time of each sarcopenia transient state before the next transition for the overall population and subgroups; (B) the predicted total length of stay in each sarcopenia transient state before death for the overall population and subgroups.