The present longitudinal cohort study examined the association between PA, CA, and SA and incident sarcopenia among community-dwelling older adults. Of the participants in this study, the proportion of the development of sarcopenia was 20.8% during the 48month follow-up period. Low PA and SA were both associated with incident sarcopenia, and this association remained significant after adjusting for each activity; however, lowCA was not a risk for incident sarcopenia.
The prevalence of sarcopenia in the first wave was 5.13%; this finding was consistent with a cross-sectional study from another Japanese cohort of older adult participants [11]. In addition, the prevalence of sarcopenia was higher in men than in women; this finding was line with previous studies [12]. At the follow-up assessment, there were 527 participants (20.8%) with incident sarcopenia; after 50times MI, 50,359 participants (23.8%) developed sarcopenia. Reports showing the incidence of sarcopenia in each country showed that it was reported to be between 14–40% [13]. The incidence of sarcopenia obtained in the present study was within the range in the previous report, which indicates consistency with these findings. While the findings on the incidence of sarcopenia in Japan is not well known, we were able to accumulate reports from Asia.
Low levels of PA (i.e., 1.6fold) and SA (i.e., 1.3fold) had an independent impact on incident sarcopenia, even before and after mutual adjustment; a sensitivity analysis using 50times MI showed similar results. Several previous studies have reported that increased PA, CA, and SA contributes to the prevention of sarcopenia [6–8]. The present study showed that low PA and SA are risk factors for the incident sarcopenia. Physical inactivity reportedly increases the odds ratio of probable sarcopenia among older Chinese and United Kingdom community-dwelling adults in cross-sectional studies [14,15]. The main finding of the present study was that insufficient SA was independently associated with the onset of sarcopenia, regardless of PA. A previous study also found that engaging in SA was a protective factor against sarcopenia, but it has been suggested that the underlying mechanism is that SA leads to an increase in PA [8,16]. The present study indicated the possibility that SA may independently affect the onset of sarcopenia, however, regardless of sufficient or insufficient PA; a potential explanation for this is the increase in cFos expressed in the brain as a result of social interactions. A study using rats has shown an increase in cFos expression as a result of social interactions [17], and cFos expression enhances thermogenesis and promotes glucose utilization by increasing insulin function in peripheral tissues such as skeletal muscle [18]. However, this increase in cFos expression was not observed in the isolated rats [17].
In summary, inadequate daily SA may lead to the development of sarcopenia without the skeletal muscle anabolism promoted by cFos. Another potential explanation is that sufficient SA play an important role in maintaining the neural–motor-function system by activating the sympathetic nervous system. Social interaction and social support have been reported to increase sympathetic activity compared to isolates[19, 20], and sympathetic neuron activation has also been reported to rescue age-related impaired motor innervation and neuromuscular junction transmission [21]. In the present study, however, cFos and the other biomarkers including activation of the sympathetic nervous system were not measured. In a future study, there is thus a need to examine the association between biomarkers that arise with SA and the onset of sarcopenia.
Alternatively, older adults with low CA showed no association with incident sarcopenia, even before and after adjusting for PA and SA. One possible explanation for the inconsistency between the findings of a previous study and those of the present study is the different content included in the measured CA. In the previous study, for which 809 older Japanese adults were recruited, engaging in CA was positively associated with gait speed and the chair stand test [7]. Of the 30 leisure activities evaluated, 13 involved PA; in other words, most of the activities measured as leisure activity (i.e., CA) were actually related to PA. In the present study, we considered sedentary activities to be CA, which may explain the lack of association with incident sarcopenia. On the other hand, our participants were asked about sedentary behaviors related to CA (e.g., using the telephone, reading books, playing a shogi). Sedentary behavior ≥ 11 hours per day was associated with higher odds for sarcopenia [22]. In a future study, it will be necessary to classify CA into activities with or without PA, in order to elucidate details about the association between CA and incident sarcopenia.
The strength of the present study is that we were able to examine the association between PA, CA, SA, and incident sarcopenia using longitudinal individual older Japanese cohort data. However, the present study also has several limitations. First, the participants may have been in relatively good health; because the study participants were able to independently participate in this survey at a community center, older people with lower physical function and complications may not have participated in the present study. Second, we were unable to address other potential confounders affecting the progression of sarcopenia, such as dietary intake and medication therapy. Third, some variables were self-reported including each activity (e.g., educational history and medication); future studies should examine these and other factors using a variety of methods.