Participant characteristics
The baseline characteristics by sex are presented in Table 1. The average time spent in MVPA was 27.4 ± 20.7 minutes/day and 52.7% of participants met the criteria of the 2018 PAG (MVPA ≥ 150 minutes/week). The average scores of the DGAI-2015 were 61.1 (52.1 [DGAI-2015 < median, poor adherence], and 70.0 [DGAI-2015 ≥ median, higher adherence]. Only 28% of participants where characterized as both meeting the 2018 PAG and having higher adherence to the 2015 DGA. However, 47% of participants either met the 2018 PAG (MVPA ≥ 150 minutes/week) or had higher adherence to the 2015 DGA. The characteristics of participants excluded from the analysis are presented in Supplementary Table 1. Adherence to the 2018 PAG was associated with higher adherence to the 2015 DGA (DGAI-2015 ≥ median; odds ratio 0.61, 95% CI, 0.51–0.72; p < .001). For the cross-sectional analysis, 66% of participants provided 7 days of valid accelerometry data (23% [6 days of valid accelerometry data], 8% [5 days of valid accelerometry data], and 3% [4 days of valid accelerometry data]. Similarly, 66%, 22%, 8%, and 2% of participants provided 7, 6, 5, and 4 days of valid accelerometry data, respectively, at baseline for the longitudinal analysis.
Table 1. Sample Characteristics.
|
Men (n=1,086)
|
Women (n=1,293)
|
Age (years)
|
47±8
|
47±9
|
Body mass index (kg/m2)
|
28.9±4.7
|
26.8±6.0
|
Waist circumference (inches)
|
40.2±4.8
|
36.1±5.9
|
SBP (mm Hg)
|
121±13
|
113±14
|
DBP (mm Hg)
|
78±8
|
72±9
|
Antihypertensive medication, n (%)
|
222 (20.5)
|
173 (13.4)
|
Hypertension, n (%)
|
303 (28.0)
|
222 (17.2)
|
Fasting blood glucose (mg/dL)
|
100.3±19.7
|
92.4±13.7
|
Diabetes, n (%)
|
63 (5.8)
|
43 (3.3)
|
Total cholesterol (mg/dL)
|
187±35
|
187±35
|
Triglycerides (mg/dL)
|
129±83
|
95±59
|
HDL-C (mg/dL)
|
51±14
|
67±18
|
LDL-C (mg/dL)
|
110±30
|
101±30
|
Lipid-lowering medication, n (%)
|
247 (22.7)
|
130 (10.1)
|
Smoking, n (%)
|
99 (9.1)
|
102 (7.9)
|
Total caloric intake (kcal/day)
|
2,106±660
|
1,913±594
|
DGAI-2015 (0-100)
|
58.0±10.7
|
65.3±9.7
|
Accelerometer wear time (minutes/day)
|
934.0±102.7
|
911.8±88.4
|
Sedentary time (minutes /day)
|
657.8±70.1
|
668.9±67.1
|
Light intensity PA (minutes /day)
|
129.7±49.6
|
123.2±42.2
|
MVPA (minutes /day)
|
29.9±21.5
|
25.3±19.9
|
Total PA (minutes /day)
|
159.6±59.9
|
148.5±51.0
|
MVPA ≥150 minutes/week, n (%)
|
641 (59.0)
|
613 (47.4)
|
MetS, n (%)
|
361 (33.2)
|
223 (17.3)
|
Abbreviations: SBP, systolic blood pressure; DBP, diastolic blood pressure; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; DGAI, dietary guidelines for Americans adherence index; MVPA, moderate to vigorous physical activity; PA, physical activity; MetS, metabolic syndrome
Note: Values are mean±SD unless otherwise indicated; Hypertension was defined by SBP/DBP ≥ 140/90 mmHg or the use of antihypertensive medications; Diabetes was defined as a fasting blood glucose ≥126 mg/dL or use of insulin or oral hypoglycemic agents.
Cross-sectional associations of adherence to the 2018 PAG and 2015 DGA with the prevalence of MetS
The individual and conjoint cross-sectional associations of adherence to the 2018 PAG and 2015 DGA with the prevalence of MetS are shown in Tables 2 and Fig. 1 (Additional file 1), respectively. The prevalence of MetS was 24.6% (n = 584/2,379). After adjusting for covariates, participants who met the 2018 PAG had lower odds of MetS, compared to those who did not. Likewise, participants with higher adherence to the DGA (DGAI-2015 ≥ median) had lower odds of MetS, compared to those with poor adherence (DGAI-2015 < median). Results were similar in analyses examining the MVPA and DGAI-2015 as continuous variables (Table 2). Every 10 minutes per day increase in MVPA or 10-point increase in DGAI-2015 was associated with 18% or 19% lower odds of prevalent MetS, respectively (Table 2).
Table 2. Cross-sectional associations of adherence to the 2018 PAG and 2015 DGA with the presence of MetS.
2018 PAG Adherence
|
#MetS/#total
|
OR
|
95% CI
|
P value
|
MVPA <150 min/week
|
358/1,125
|
Referent
|
|
|
MVPA ≥150 minutes/week
|
226/1,254
|
0.49
|
0.40-0.60
|
<.001
|
MVPA (per 10 minutes increment)
|
584/2,379
|
0.82
|
0.76-0.88
|
<.001
|
2015 DGA Adherence
|
#MetS/#total
|
OR
|
95% CI
|
P value
|
DGAI_2015 <median
|
342/1186
|
Referent
|
--
|
|
DGAI_2015 <median
|
242/1193
|
0.67
|
0.54-0.83
|
<.001
|
DGAI-2015 (per 10 point increment)
|
584/2,379
|
0.81
|
0.74-0.89
|
<.001
|
Conjoint association
|
#MetS/#Total
|
OR
|
95% CI
|
P value
|
MVPA <150 min/week & DGAI-2015 <median
|
210/600
|
Referent
|
|
|
MVPA ≥150 min/week & DGAI-2015 <median
|
132/586
|
0.57
|
0.43-0.75
|
<.001
|
MVPA <150 min/week & DGAI-2015 ≥median
|
148/525
|
0.80
|
0.60-1.06
|
.12
|
MVPA ≥150 min/week & DGAI-2015 ≥median
|
94/668
|
0.35
|
0.26-0.47
|
<.001
|
Abbreviations: PAG, Physical Activity Guidelines; DGA, Dietary Guidelines for Americans; MetS, metabolic syndrome; OR, odds ratio; CI, confidence interval; MVPA, moderate to vigorous physical activity; DGAI, dietary guidelines for Americans adherence index.
Note: Models adjusted for age, sex, accelerometer wear time, number of cigarettes smoked per day, total calorie intake per day, and prevalence of CVD at exam 2; Accelerometer wear time was excluded in the model evaluating the association between adherence to the 2015 DGA and presence of MetS; Family relatedness was further adjusted as a random variance-covariance factor in the generalized linear models; median DGAI-2015 was 62.1/100; Both PA and diet quality were measured at the second examination cycle (2008-2011); Bold P-values indicate statistical significance.
We observed a significant multiplicative interaction between MVPA and DGAI-2015 on the association of each with odds of MetS (Pinteraction =0.02). In the conjoint analysis, we observed that participants who met the 2018 PAG (MVPA ≥ 150 minutes/week) had lower odds of MetS with higher adherence to the 2015 DGA. Participants who met the 2018 PAG and had higher adherence to the 2015 DGA were less likely to have MetS compared to the referent group (i.e., participants who did not meet the 2018 PAG and had poor adherence to the 2015 DGA; Table 2 and Fig. 1 [Additional file 1]).
Longitudinal associations of adherence to the 2018 PAG and 2015 DGA with the incidence of MetS
In our sample of participants who were free of MetS at baseline (Sample 2, n = 1,562), 287 (18.4%) individuals developed new-onset MetS over an average follow-up of 8 years. After adjusting for covariates, adherence to the 2018 PAG was associated with a lower risk of developing MetS. Similarly, higher adherence to the 2015 DGA (DGAI-2015 ≥ median) was associated with a lower risk of MetS compared to those with poor adherence to the 2015 DGA (DGAI-2015 < median). Additionally, every 10 minutes per day increase in MVPA and each 10-point increase in DGAI-2015 were associated with 8% and 13% lower risk of developing MetS, respectively (Table 3).
Table 3. Longitudinal associations of adherence to the 2018 PAG and 2015 DGA with the incidence of MetS.
2018 PAG Adherence
|
#Events/#At risk
|
HR
|
95% CI
|
P value
|
MVPA <150 min/week
|
141/648
|
Referent
|
|
MVPA ≥150 min/week
|
146/914
|
0.66
|
0.50-0.88
|
.004
|
MVPA (per 10 minutes increment/day)
|
287/1,562
|
0.92
|
0.86-0.99
|
.03
|
2015 DGA Adherence
|
#Events/#At risk
|
HR
|
95% CI
|
P value
|
DGAI-2015 <median
|
174/781
|
Referent
|
|
DGAI-2015 ≥median
|
113/781
|
0.68
|
0.51-0.90
|
.008
|
DGAI-2015 (per 10 point increment)
|
287/1,562
|
0.87
|
0.76-0.99
|
.03
|
Conjoint association
|
#Events/#At risk
|
HR
|
95% CI
|
P value
|
MVPA <150 min/week & DGAI-2015 <medain
|
86/339
|
Referent
|
|
|
MVPA ≥150 min/week & DGAI-2015 <median
|
88/442
|
0.74
|
0.51-1.06
|
.10
|
MVPA <150 min/week & DGAI-2015 ≥median
|
55/309
|
0.78
|
0.52-1.17
|
.22
|
MVPA ≥150 min/week & DGAI-2015 ≥median
|
58/472
|
0.48
|
0.32-0.70
|
<.001
|
Abbreviations: PAG, Physical Activity Guidelines; DGA, dietary guidelines for Americans; MetS, metabolic syndrome; HR, hazards ratio; CI, confidence interval; MVPA, moderate to vigorous physical activity; DGAI, diet guidelines for Americans adherence index;
Note: Models adjusted for age, sex, accelerometer wear time, number of cigarettes smoked per day, total calorie intake per day, and prevalence of CVD at baseline; Accelerometer wear time was excluded in the model evaluating the association between adherence to the 2015 DGA and incidence of MetS; median DGAI-2015 was 63.0/100; Both PA and diet quality were measured at the second examination cycle (2008-2011); Bold P-values indicate statistical significance.
We also observed a significant multiplicative interaction between MVPA and DGAI-2015 on the association of each with the risk of MetS (Pinteraction =0.03). In our longitudinal analysis, participants who met the 2018 PAG and had higher adherence to the 2015 DGA had a lower risk of MetS when compared to the referent group (MVPA < 150 minutes/week and DGAI-2015 < median; Table 3 and Fig. 1 [Additional file 1]).