Characteristics of participants in the cross‑sectional and longitudinal study
Table 1 displays the characteristics of the participants categorized according to their SPPB level. The study population had the mean (SD) age of 67.07 ± 5.90, and 2308 individuals (49.78%) were females. Among the 4636 older adults included in the analysis, the prevalence of poor LEF was 18.55% (860/4636), while normal LEF was observed in 17.39% (806/4636) of the participants.
Compared to the population with good LEF, individuals with poor LEF were generally older (mean age: 71.41 vs. 65.45 years) and more likely to be female (67.44% vs. 42.73%). They also had a higher proportion of individuals living alone (31.78% vs. 14.62%) and a higher prevalence of non-smokers (69.65% vs. 55.93%) and non-drinkers (68.60% vs. 53.97%). Additionally, the group with poor LEF exhibited higher systolic blood pressure (140.59 vs. 131.51), higher diastolic blood pressure (76.07 vs. 74.05), and higher prevalence of diabetes (16.75% vs. 12.61%), hypertension (68.03% vs. 49.21%), and cardiovascular disease (25.21% vs. 17.14%) (all P < 0.001). For more detailed information, please refer to Table 1. Table S3 presents the baseline characteristics of the 3699 participants without cardiovascular disease (CVD) in the longitudinal study, stratified by SPPB level.
Table 1
Baseline characteristics of all participants by SPPB level.
Characteristics | Low SPPB (n = 860) | Moderate SPPB (n = 806) | High SPPB (n = 2970) | P |
Age, y | 71.41 ± 6.81 | 68.43 ± 6.02 | 65.45 ± 4.74 | < 0.001 |
Female | 580 (67.44) | 459 (56.95) | 1269 (42.73) | < 0.001 |
Solitary | 273 (31.78) | 201 (24.94) | 434 (14.62) | < 0.001 |
Rural | 583 (67.79) | 547 (67.87) | 1926 (64.85) | 0.122 |
Smoking | 261 (30.35) | 306 (37.97) | 1309 (44.07) | < 0.001 |
Drinking | 270 (31.40) | 309 (38.34) | 1367 (46.03) | < 0.001 |
Region, | | | | 0.113 |
East | 268 (31.16) | 235 (29.16) | 903 (30.40) | |
Middle | 252 (29.30) | 227 (28.16) | 928 (31.25) | |
West | 286 (33.26) | 283 (35.11) | 976 (32.86) | |
Northeast | 54 (6.28) | 61 (7.57) | 163 (5.49) | |
Education | | | | < 0.001 |
0 year | 507 (58.95) | 347 (43.05) | 809 (27.24) | |
0–6 years | 288 (33.49) | 365 (45.29) | 1525 (51.35) | |
Over 6 years | 65 (7.56) | 94 (11.66) | 636 (21.41) | |
Waist, cm | 85.64 ± 12.92 | 84.13 ± 12.66 | 84.04 ± 11.83 | 0.007 |
BMI, kg/m2 | 25.24 ± 61.70 | 22.92 ± 4.23 | 23.57 ± 17.13 | 0.288 |
SBP, mm Hg | 140.59 ± 23.59 | 134.21 ± 22.15 | 131.51 ± 21.37 | < 0.001 |
DBP, mm Hg | 76.07 ± 12.03 | 74.16 ± 12.05 | 74.05 ± 11.48 | < 0.001 |
Metabolic biomarkers | | | | |
FPG, mg/dL | 5.35 ± 0.90 | 5.33 ± 0.86 | 5.26 ± 0.73 | 0.162 |
HbA1c, mg/dL | 3.12 ± 6.88 | 3.08 ± 7.84 | 2.77 ± 7.32 | 0.018 |
CRP, mg/L | 82.43 ± 15.00 | 84.23 ± 13.74 | 86.55 ± 13.19 | 0.447 |
eGFR, mL/min/1.73 m2 | 111.80 ± 29.38 | 111.04 ± 34.80 | 109.27 ± 30.53 | < 0.001 |
Comorbidities | | | | |
Diabetes | 142 (16.75) | 119 (14.93) | 371 (12.61) | 0.005 |
Hypertension | 534 (68.03) | 420 (56.91) | 1331 (49.21) | < 0.001 |
Dyslipidaemia | 104 (12.50) | 97 (12.34) | 353 (12.13) | 0.954 |
History of CVDs | | | | |
Prevalent CVD | 213 (25.21) | 167 (20.93) | 504 (17.14) | < 0.001 |
Prevalent heart disease | 57 (6.72) | 27 (3.37) | 66 (2.24) | 0.002 |
Prevalent stroke | 173 (20.43) | 147 (18.42) | 458 (15.58) | < 0.001 |
Note: Values were described in mean ± SD for continuous variables or frequency (percentages) for categorical variables. |
Abbreviation: BMI, body mass index; SBP, systolic blood pressure; DBP diastolic blood pressure; FPG, fasting plasma glucose; HbA1c, hemoglobin A1c; CRP, C-reactive protein; eGFR, estimated glomerular filtration rate; CVD, cardiovascular disease.
Missing data: 2 for living status, 623 for waist, 640 for BMI, 649 for SBP, 650 for DBP, 1472 for FPG, 1101 for HbA1c, 1125 for CRP, 1130 for eGFR, 50 for diabetes, 408 for hypertension, 107 for dyslipidemia, 53 for CVD, 34 for heart disease, 51 for stroke.
The mean values and percentages in different quintiles were compared using ANOVA analysis and χ2 tests.
Cross-sectional associations of SPPB level with cardiovascular disease
In the cross-sectional study, the prevalence of cardiovascular disease (CVD) was 19.07% (884/4636) in the total population, 16.97% (504/2970) in individuals with good lower extremity function (LEF), 20.72% (167/806) in individuals with normal LEF, and 24.77% (213/860) in individuals with poor LEF (as shown in Table 1 and Table S4). After adjusting for demographic characteristics and health-related factors, individuals with low Short Physical Performance Battery (SPPB) scores had a significantly higher risk of CVD [odds ratio (OR) (95% CI): 1.62 (1.27−2.05), P<0.001] (Figure 2 and Table S4). The cross-sectional association between SPPB level and CVD did not change significantly after further adjusting for plasma biomarkers in the subgroup of 2988 participants (Table S5).
The associations between SPPB level and different components of CVD are presented in Table S4 and Figure 2. After adjusting for covariates, individuals with low SPPB scores were found to have a higher risk of heart disease [1.35 (1.05−1.74), P<0.05] and stroke [3.09 (1.92−4.96), P<0.05] compared to those with high SPPB scores (Table S4). Similar results were observed among the subgroup of 2988 participants with completed plasma biomarkers measurements, where individuals with low SPPB scores remained significantly associated with stroke [2.85 (1.68−4.84), P<0.001] after further adjusting for metabolic biomarkers (Table S5).
Longitudinal association between baseline SPPB level and incident cardiovascular disease at follow-up, 2011−2018
During the 8-year follow-up period, a total of 1354 cases (29.54%) with incident cardiovascular disease (CVD) events were identified. In the longitudinal analysis, the incidence rate of CVD was 33.33 per 1000 person-years among individuals with high SPPB level, 28.77 per 1000 person-years among those with moderate SPPB level, and 45.35 per 1000 person-years among participants with low SPPB level. Table 2 presents the relationship between baseline SPPB level and incident CVD. After adjusting for covariates in Models 1-3, individuals with low SPPB level had a higher likelihood of experiencing CVD incidence [hazard ratio (HR) (95% CI): 1.11 (1.01−1.23)] compared to those with high SPPB level (P<0.05, Figure 3).
Regarding CVD components, individuals with low SPPB level had higher risks of heart disease [1.21 (1.00−1.45), P<0.05] and stroke [1.98 (1.47−2.67), P<0.001] (Table 2 and Figure 3). The longitudinal associations between low SPPB level and CVD were significantly affected after further adjusting for metabolic biomarkers among the subpopulation of 2997 subjects with plasma biomarkers measurements (Table S6). Individuals with low SPPB level had an 11.0% increased risk of incident CVD compared to those with high SPPB level [1.11 (0.99−1.24), P<0.1]. After further adjusting for metabolic biomarkers, individuals with low SPPB level had a higher likelihood of experiencing incident stroke [2.02 (1.46−2.78), P<0.001] compared to those with high SPPB level (Table S6).
Table 2 Incidence of CVD according to baseline SPPB level among all participants, 2011–2018.
Outcome
|
Cases, n
|
Incidence Rate, per 1000 Person-Years
|
HR (95% CI)
|
Model 1a
|
Model 2b
|
Model 3c
|
CVDs
|
|
|
|
|
|
High SPPB
|
792
|
33.33
|
Reference
|
Reference
|
Reference
|
Moderate SPPB
|
250
|
28.77
|
1.05(0.97, 1.14)
|
1.06(0.98, 1.15)
|
1.05(0.96, 1.15)
|
Low SPPB
|
312
|
45.35
|
1.13(1.04, 1.23) **
|
1.14(1.05, 1.24) **
|
1.11(1.01, 1.23) *
|
Heart disease
|
|
|
|
|
|
High SPPB
|
681
|
28.66
|
Reference
|
Reference
|
Reference
|
Moderate SPPB
|
200
|
31.02
|
1.08(0.92, 1.27)
|
1.11(0.95, 1.31)
|
1.11(0.93, 1.32)
|
Low SPPB
|
243
|
35.32
|
1.26(1.07, 1.48) **
|
1.32(1.12, 1.56) **
|
1.21(1.00, 1.45) *
|
Stroke
|
|
|
|
|
|
High SPPB
|
192
|
8.08
|
Reference
|
Reference
|
Reference
|
Moderate SPPB
|
78
|
12.10
|
1.64(1.25, 2.15) ***
|
1.64(1.25, 2.15) ***
|
1.42(1.05, 1.91) *
|
Low SPPB
|
104
|
15.17
|
2.25(1.72, 2.93) ***
|
2.30(1.76, 3.01) ***
|
1.98(1.47, 2.67) ***
|
Note: Abbreviation: HR, hazard ratio; CVD, cardiovascular disease.
a Model 1 was adjusted for age, gender.
b Model 2 was adjusted for age, gender, living status, residence, smoking, drinking, region, education, waist, and body mass index.
c Model 3 was adjusted as model 2 with further adjustment for systolic blood pressure, diastolic blood pressure, and history of diabetes, hypertension, and dyslipidemia.
Significant at †P<0.1, *P<0.05, **P<0.01 and ***P<0.001 compared to high SPPB group.
Longitudinal association between baseline SPPB level with newly onset incident cardiovascular disease at follow-up (sensitivity analysis), 2011−2018
After excluding 937 subjects with a history of CVDs at baseline, the longitudinal analysis included 3699 participants. During the 8-year follow-up, 811 cases (21.92%) of newly onset incident CVD events were identified. The incidence rate of CVD was 25.30 per 1000 person-years among individuals with high SPPB level, 27.34 per 1000 person-years among those with moderate SPPB level, and 35.60 per 1000 person-years among participants with low SPPB level.
Table S7 presents the relationship between baseline SPPB level and newly onset incident CVD. After adjusting for covariates in Models 1-3, individuals with low SPPB level did not show a significantly higher likelihood of developing newly onset CVD compared to those with high SPPB level [hazard ratio (HR) (95% CI): 1.10 (0.98−1.23)] (Table S7). The longitudinal associations between low SPPB level and CVD were not significantly affected after further adjusting for metabolic biomarkers among the subpopulations of 3699 subjects with plasma biomarkers measurements (Table S8).
Regarding CVD components, individuals with low SPPB level had a higher risk of newly onset stroke [1.67 (1.22–2.30), P<0.001], but not newly onset heart disease (Table S7). After further adjusting for metabolic biomarkers, individuals with low SPPB level were more likely to have incident stroke [1.69 (1.20−2.40), P<0.001] compared to those with high SPPB level (Table S8).