Dietary Sodium Restriction and Frailty among Middle-Aged and Older Adults: An 8-Year Longitudinal Study

Abstract

Frailty is a common geriatric syndrome. However, there is little information about the relationship between dietary sodium restriction (DSR) and frailty in later life. This study aimed to elucidate the relationship between DSR and frailty in middle-aged and older adults. The 8-year follow-up data from the Taiwan Longitudinal Study on Aging, including 5131 individuals aged ≥50 years, were analyzed using random-effects panel logit models. DSR was evaluated by assessing whether the participants were told by a physician to reduce or avoid sodium intake from food. Three indices were used to measure frailty: the Study of Osteoporotic Fractures (SOF) index, the Fried index, and the Fatigue, Resistance, Ambulation, Illness, and Loss of weight (FRAIL) index. Individuals with DSR were more likely to report frailty compared with those with non-DSR (SOF: adjusted odds ratio [AOR] = 1.82, 95% confidence interval [CI] = 1.46-2.27; Fried: AOR = 2.55, 95% CI = 1.64-3.98; FRAIL: AOR = 2.66, 95% CI = 1.89-3.74). DSR was associated with a higher likelihood of SBF (AOR = 2.61, 95% CI = 1.61-4.22). We identified a temporal trajectory in our study, noting significant participant reactions to both short- and mid-term DSR. Future research should address the balance between frailty risk and cardiovascular risk related to DSR.

Keywords: dietary sodium restriction; frailty; frailty phenotypes; older adults.

Figure 1

The effects of DSR on the probability of frailty among middle-aged and older adults, Taiwan, 1999–2007. Note: All results were obtained using a random-effects panel logit model. The mean and standard error were represented by bar charts with error bars. Pr(frailty) was used to represent the predicted probability of frailty. DSR: dietary sodium restriction; AOR: adjusted odds ratio. Pr(frailty) (SOF): DSR: 0.14 ***, CI = 0.12–0.16; Non-DSR: 0.09 ***, CI = 0.08–0.10; Pr(frailty) (Fried): DSR: 0.06 ***, CI = 0.05–0.07; Non-DSR: 0.03 ***, CI = 0.03–0.04; Pr(frailty) (FRAIL): DSR: 0.10 ***, CI = 0.08–0.11; Non-DSR: 0.05 ***, CI = 0.05–0.06; *** p < 0.001. Source: the author.

Figure 2

The effects of DSR on the probability of frailty phenotypes among middle-aged and older adults, Taiwan, 1999–2007. Note: All results were obtained using a random-effects panel logit model. The mean and standard error were represented by bar charts with error bars. Pr(frailty phenotypes) was used to represent the predicted probability of frailty phenotypes. DSR: dietary sodium restriction; EBF: energy-based frailty; SBF: sarcopenia-based frailty; HBF: hybrid-based frailty; AOR: adjusted odds ratio. Pr(frailty phenotypes) (EBF): DSR: 0.01 **, CI = 0.004–0.02; Non-DSR: 0.01 ***, CI = 0.01–0.01; Pr(frailty phenotypes) (SBF): DSR: 0.07 ***, CI = 0.05–0.08; Non-DSR: 0.04 ***, CI = 0.04–0.05; Pr(frailty phenotypes) (HBF): DSR: 0.01 ***, CI = 0.01–0.02; Non-DSR: 0.01 ***, CI = 0.004–0.01; *** p < 0.001. Source: the author.

Figure 3

The temporal effects of DSR on the probability of frailty among middle-aged and older adults, Taiwan, 1999–2007. Note: All results were obtained using a random-effects panel logit model. The mean and standard error were represented by bar charts with error bars. Pr(frailty) was used to represent the predicted probability of frailty. DSR: dietary sodium restriction; AOR: adjusted odds ratio. Pr(frailty) (SOF): DSR, 1999: 0.13***, CI = 0.10–0.15; Non-DSR, 1999: 0.07 ***, CI = 0.06–0.08; DSR, 2003: 0.16 ***, CI = 0.13–0.19; Non-DSR, 2003: 0.11 ***, CI = 0.09–0.12; DSR, 2007: 0.13 ***, CI = 0.09–0.16; Non-DSR, 2007: 0.11 ***, CI = 0.09–0.12; Pr(frailty) (Fried): DSR, 1999: 0.06 ***, CI = 0.04–0.07; Non-DSR, 1999: 0.03 ***, CI = 0.02–0.04; DSR, 2003: 0.06 ***, CI = 0.04–0.08; Non-DSR, 2003: 0.04 ***, CI = 0.03–0.04; DSR, 2007: 0.06 ***, CI = 0.04–0.08; Non-DSR, 2007: 0.04 ***, CI = 0.03–0.04; Pr(frailty) (FRAIL): DSR, 1999: 0.10 ***, CI = 0.07–0.12; Non-DSR, 1999: 0.04 ***, CI = 0.03–0.05; DSR, 2003: 0.11 ***, CI = 0.08–0.14; Non-DSR, 2003: 0.06 ***, CI = 0.05–0.07; DSR, 2007: 0.08 ***, CI = 0.05–0.11; Non-DSR, 2007: 0.06 ***, CI = 0.05–0.07; * p < 0.05, ** p < 0.01, and *** p < 0.001. Source: the author.

Figure 4

The temporal effects of DSR on the probability of frailty phenotypes among middle-aged and older adults, Taiwan, 1999–2007. Note: All results were obtained using a random-effects panel logit model. The mean and standard error were represented by bar charts with error bars. Pr(frailty phenotypes) was used to represent the predicted probability of frailty phenotypes. DSR: dietary sodium restriction; EBF: energy-based frailty; SBF: sarcopenia-based frailty; HBF: hybrid-based frailty; AOR: adjusted odds ratio. Pr(frailty phenotypes) (SBF): DSR, 1999: 0.07 ***, CI = 0.05–0.09; Non-DSR, 1999: 0.04 ***, CI = 0.04–0.05; DSR, 2003: 0.08 ***, CI = 0.06–0.11; Non-DSR, 2003: 0.04 ***, CI = 0.04–0.05; DSR, 2007: 0.05 ***, CI = 0.03–0.08; Non-DSR, 2007: 0.04 ***, CI = 0.03–0.05; * p < 0.05, ** p < 0.01. Source: the author.

Figure 5

The subgroup effects of DSR on the probability of frailty among middle-aged and older adults, Taiwan, 1999–2007. Note: All results were obtained using a random-effects panel logit model. The mean and standard error were represented by bar charts with error bars. Pr(frailty) was used to represent the predicted probability of frailty. DSR: dietary sodium restriction; AOR: adjusted odds ratio. ⁺ p < 0.1, * p < 0.05, ** p < 0.01, and *** p < 0.001. Source: the author.

Figure 6

The subgroup effects of DSR on the probability of frailty among middle-aged and older adults, Taiwan, 1999–2007. Note: All results were obtained using a random-effects panel logit model. The mean and standard error were represented by bar charts with error bars. Pr(frailty) was used to represent the predicted probability of frailty. Participants were divided into two or three age groups. DSR: dietary sodium restriction; AOR: adjusted odds ratio. ⁺ p < 0.1, * p < 0.05, ** p < 0.01, and *** p < 0.001. Source: the author.