Study design and participants
Participants with the biomarker sub-study datasets from the 6th (2011) waves of CLHLS were recruited in this secondary analysis. CLHLS is the first and largest nationwide, community-based, longitudinal prospective cohort survey, concerning older adults in China(38). It provides information on health status, socioeconomic characteristics, and lifestyles of the elderly, including a large percent of the oldest population(37). The in-depth study was launched in the eight longevity areas (Laizhou of Shandong Province, Xiayi of Henan Province, Zhongxiang of Hubei Province, Mayang of Hunan Province, Sanshui of Guangdong Province, Yongfu of Guangxi Autonomous Region, Chengmai of Hainan Province, Rudong of Jiangsu Province) (36). During the in-depth study, the Chinese Center for Disease Control and Prevention (CDC) local network medical doctors conducted physical examinations of the participants, and also collected biomarker datasets contain about 30 indicators on routine blood tests, blood biochemical tests, and urine tests (36). More detailed descriptions have been previously published elsewhere(39-41).
Initially, a total of 2439 elderly participants were included in the study. We excluded those of younger age (less than 80, n=834, 34.2%) and missing data on SOF index components (n=281, 11.5%). Finally, we retained 1324 older adults in this study.
Outcome
Consistent with previous studies of secondary analysis involving CLHLS data(42), frailty was defined by the SOF index in the current study. Three components were included in the index: underweight (defined as body mass index <18.5), low energy level (indicated by a positive response to the question “Over the last 6 months, have you been limited in activities because of a health problem?”), and muscle strengths (inability to stand up from a chair without the assistance of arms). As suggested, participants with two or more of the three components were defined as frailty.
Exposure
Fasting venous blood was collected after an overnight fast from all willing participants. Procedures for the collection and shipment of blood samples were described in detail elsewhere(14). 25(OH)D was assayed by an enzyme-linked immunoassay using Immunodiagnostic Systems Limited (IDS Ltd, Boldon, UK). The 25(OH)D level was expressed as nmol/L.
Covariates
We adjusted for socio-demographic variables, health characteristics and confounding biomarkers in the models. Socio-demographic variables included age, gender (female/male), marital status (married/other), residence (rural/other), education level (no schooling/≥1 years of schooling), co-residence [with family member(s) /other].
Health characteristics included lifestyles and chronic diseases. Lifestyles consisted of smoking (yes/no), drinking (yes/no), and regular exercise (yes/no) at present. Chronic diseases included hypertension (yes/no), diabetes mellitus (yes/no), heart diseases (yes/no), cerebrovascular diseases (yes/no), and respiratory diseases (yes/no). Hypertension was defined as systolic blood pressure≥140mmHg and/or diastolic blood pressure≥90mmHg(43). Diabetes mellitus was diagnosed by fasting plasma glucose≥7.0 mmol/L(14, 44). Other diseases were identified by self-report.
Confounding biomarkers were 11 indicators on blood routine tests and blood biochemistry tests (36). According to the previous relevant studies(19), these 11 indicators which largely investigated in relation with frailty were recruited in this study: 1) inflammatory marker: C reactive protein (CRP); 2) immune marker: counts of leukocytes (WBC); 3)clinical markers: plasma albumin (ALB), total cholesterol (CHO), serum creatinine (CREA), high-density lipoprotein cholesterol (HDLC), low-density lipoprotein cholesterol (LDLC), triglyceride (TG), and hemoglobin (HGB); 4) oxidative stress markers: malondialdehyde (MDA) and superoxide dismutase (SOD). All the standard laboratory techniques were performed by the central clinical lab at Capital Medical University in Beijing.
Overall, few data for most confounding variables were missing (1.05%). For the missing values, we did multiple imputations by chained equations to increase predictive power(45). Distribution of observed data and imputed data were described in Supplementary Table S1 (see Additional file 1). For all the covariates, the distributions of observed and imputed values were similar.
Statistical Analysis
Categorical variables were expressed as numbers and percentages, and continuous data were described as mean (standard deviation, SD) or median (interquartile range, IQR). Characteristics among groups were compared by ANOVA, Kruskal–Wallis test or χ2 test. The IQR of 25(OH)D level was used to divide the data into four categories. The cutoff points were 26.13, 35.89, 50.00 nmol/L.
We used multilayer logistic regression models based on the Likelihood Ratio Test (LRT) to determine the association between 25(OH)D level and the risk of frailty. Box-Tidwell method was used to test the linearity between logit P and all the continuous variables(46). Therefore, we used continuous terms for all the confounding biomarkers and categorized age as subgroups with 80-89, 90-99, ≥100 years. Data were reported as odds ratios (ORs) and 95% confidence intervals (CIs) in both unadjusted and adjusted logistic regression models. The p-value of Hosmer-Lemeshow test >0.05 indicated reasonable goodness of fit(47).
Different from previous studies, in order to examine the linear trend across levels of 25(OH)D, we further performed spline smoothing analysis and threshold effect analysis in the current study, which were relatively novel in studies of examining the respondents’ dose-response relationship between 25(OH)D and frailty. Instead of a priori assumptions, spline smoothing analysis is a form of mixture modeling based on the generalized additive model (GAM)(48), whereby a set of associated items, for example, 25(OH)D and frailty, can visually demonstrate the linear or curvilinear relationship by figures. As for the threshold effect analysis, which based on piece-wise regression model(49), can further examine whether this relationship is segmental or not.
Subgroup analyses and their interactions were tested to explore whether gender and age subgroups would confound the association between 25(OH)D level and frailly. Sensitivity analysis was performed in participants with complete variables, and multiple imputations, separately.
Two-tailed p-value <0.05 was considered the statistical significance in all the analyses. Statistical analyses were conducted by IBM SPSS Statistics Version 22.0, except for the spline smoothing analysis, threshold effect analysis, and multiple imputations were performed by R software Version 3.4.3 (http://www.R-project.org) and Empower® (www.empowerstats.com).
Table 1. Participant characteristics.
|
|
Variables
|
All participants (n=1324)
|
Categories (nmol/L)
|
statistics a
|
Q1 (≤26.13)
|
Q2 (26.13–35.89)
|
Q3 (35.89–50.00)
|
Q4 (>50.00)
|
Socio-demographics, n (%)
|
|
|
|
|
|
|
Age (80-112), M (SD)
|
92.89(7.92)
|
95.63(7.49)
|
93.37(7.72)
|
91.85(7.85)
|
90.70(7.78)
|
25.207***
|
Female
|
844(63.7)
|
251(75.8)
|
236(71.3)
|
200(60.2)
|
157(47.6)
|
68.190***
|
Married
|
294(22.3)
|
44(13.3)
|
60(18.2)
|
83(25.1)
|
107(32.7)
|
40.503***
|
Rural
|
1124(84.9)
|
282(85.2)
|
282(85.2)
|
265(79.8)
|
295(89.4)
|
11.925**
|
No schooling
|
998(76.4)
|
279(85.8)
|
257(78.4)
|
242(74.2)
|
220(67.1)
|
33.410***
|
With household member(s)
|
950(73.2)
|
263(82.2)
|
236(72.2)
|
225(69.7)
|
226(69.1)
|
25.873***
|
Health characteristics, n (%)
|
|
|
|
|
|
|
Smoking
|
148(11.3)
|
28(8.5)
|
39(11.8)
|
37(11.2)
|
44(13.5)
|
4.337
|
Drinking
|
167(12.7)
|
32(9.7)
|
41(12.5)
|
45(13.6)
|
49(14.9)
|
0.218
|
Regular exercise
|
178(13.9)
|
27(8.4)
|
38(11.9)
|
54(16.7)
|
59(18.6)
|
17.317***
|
Hypertension
|
785(62.2)
|
197(61.9)
|
195(62.1)
|
192(61.5)
|
201(63.0)
|
0.155
|
Diabetes mellitus
|
98(7.4)
|
28(8.5)
|
25(7.6)
|
24(7.3)
|
21(6.4)
|
1.057
|
Heart diseases
|
91(7.0)
|
24(7.4)
|
24(7.4)
|
27(8.4)
|
16(4.9)
|
3.398
|
Cerebrovascular diseases
|
102(7.8)
|
34(10.4)
|
31(9.5)
|
18(5.5)
|
19(5.8)
|
8.630*
|
Respiratory diseases
|
116(8.9)
|
29(9.0)
|
32(9.8)
|
23(7.0)
|
32(9.8)
|
2.080
|
Biomarkers, M (IQR)
|
|
|
|
|
|
|
CRP (mg/L)
|
1.01(0.41,2.93)
|
1.12(0.38,3.35)
|
0.93(0.43,3.05)
|
0.96(0.41,2.54)
|
1.09(0.39,2.75)
|
1.491
|
ALB (g/L)
|
39.10(35.90,42.40)
|
37.90(35.30,41.40)
|
38.60(35.48,42.12)
|
39.70(36.70,42.93)
|
39.90(37.20,42.80)
|
29.923***
|
CHO (mmol/L)
|
4.16(3.52,4.79)
|
4.03(3.49,4.72)
|
4.21(3.51,4.79)
|
4.21(3.47,4.97)
|
4.20(3.70,4.78)
|
4.186
|
CREA (mmol/L)
|
78(65,96)
|
69(60,85)
|
77(63,93)
|
82(69,98)
|
87(71,102)
|
76.765***
|
HDLC (mmol/L)
|
1.23(1.03,1.49)
|
1.20(1.01,1.45)
|
1.25(1.04,1.51)
|
1.27(1.03,1.55)
|
1.23(1.04,1.46)
|
5.065
|
LDLC (mmol/L)
|
2.45(1.94,3.02)
|
2.40(1.92,2.97)
|
2.42(1.89,3.05)
|
2.41(1.86,3.08)
|
2.54(2.04,3.00)
|
5.147
|
TG (mmol/L)
|
0.79(0.59,1.10)
|
0.78(0.59,1.07)
|
0.79(0.58,1.09)
|
0.82(0.61,1.16)
|
0.77(0.57,1.07)
|
6.429
|
SOD (IU/mL)
|
58.53(53.43,63.24)
|
56.75(51.75,62.97)
|
58.18(53.49,63.20)
|
58.75(53.33,63.06)
|
59.39(55.39,64.24)
|
18.975***
|
MDA (μmol/L)
|
4.71(3.73,5.79)
|
4.81(3.93,5.89)
|
4.87(3.88,5.91)
|
4.84(3.82,5.83)
|
4.33(3.25,5.55)
|
27.303***
|
WBC (109/L)
|
5.30(4.30,6.40)
|
4.80(4.00,6.00)
|
5.10(4.10,6.10)
|
5.60(4.57,6.60)
|
5.60(5.60,6.80)
|
34.983***
|
HGB (g/L)
|
118(106,131)
|
121(110,133)
|
120(107,132)
|
116(105,129)
|
117(105,131)
|
11.618***
|
Frailty, n (%)
|
426(33.2)
|
162(48.9)
|
112(33.8)
|
93(28.0)
|
59(17.9)
|
76.606***
|
M (SD), mean (standard variance); M (IQR), median (interquartile range).
a Coefficient of ANOVA, Kruskal–Wallis test or χ2 test among categories of plasma 25(OH)D level.
* <0.05, **<0.01, *** <0.001.
Abbreviations: CRP, C reactive protein; ALB, plasma albumin; CHO, total cholesterol; CREA, plasma creatine; HDLC, high-density lipoprotein cholesterol; LDLC, low-density lipoprotein cholesterol; SOD, superoxide dismutase; TG, triglyceride; SOD, superoxide dismutase; MDA, malondialdehyde; WBC, white blood cell count; HGB, hemoglobin.
|
|
|
|
|
|
|
|
|
|
|