3.1 Descriptive statistics
The basic characteristics of the study participants by HUA status are summarized in Table 1. The mean age of total participants was 55.56 years (SD =12.21 years), and more than half of the subjects were females (60.44%). The mean serum uric acid level was 286.49μmol/L (SD =79.64μmol/L) and the prevalence of HUA was 10.24% among total participants. Participants with HUA were younger than Non-HUA population (54.04 vs. 55.74 years), had higher education level and average monthly income, were more likely to be smoker, drinker, and with low PA and high ST (all P<0.001). Moreover, they also had higher BMI and serum uric acid level. (both P<0.001).
Table 1. Basic characteristics of study participants by with and without hyperuricemia (n=38855).
Characteristics
|
Total (n=38855)
|
|
Non-HUA (n=34877)
|
HUA (n=3978)
|
P-value
|
Age (year, mean ± SD)
|
55.56 ± 12.21
|
|
55.74 ± 12.05
|
54.04 ± 13.43
|
<0.001a
|
Gender (n, %)
|
|
|
|
|
<0.001b
|
Male
|
15371 (39.56)
|
|
13403 (38.43)
|
1968 (49.47)
|
|
Female
|
23484 (60.44)
|
|
21474 (61.57)
|
2010 (50.53)
|
|
Education level (n, %)
|
|
|
|
|
<0.001b
|
≤Primary school
|
17385 (44.74)
|
|
15748 (45.15)
|
1637 (41.15)
|
|
≥Middle school
|
21470 (55.26)
|
|
19129 (54.85)
|
2341 (58.85)
|
|
Marital status (n, %)
|
|
|
|
|
0.690 b
|
Married/living together
|
34872 (89.75)
|
|
31309 (89.77)
|
3563 (89.57)
|
|
Divorced/widowed/separated/ unmarried
|
3983 (10.25)
|
|
3568 (10.23)
|
415 (10.43)
|
|
Average monthly income (n, %)
|
|
|
|
|
<0.001b
|
<500 RMB
|
13837 (35.61)
|
|
12432 (35.65)
|
1405 (35.32)
|
|
500-1000 RMB
|
12795 (32.93)
|
|
11608 (33.28)
|
1187 (29.84)
|
|
>1000 RMB
|
12223 (31.46)
|
|
10837 (31.07)
|
1386 (34.84)
|
|
Smoking status (n, %)
|
|
|
|
|
<0.001b
|
Never
|
28260 (72.73)
|
|
25628 (73.48)
|
2632 (66.16)
|
|
Former
|
3134 (8.07)
|
|
2739 (7.85)
|
395 (9.93)
|
|
Current
|
7461 (19.20)
|
|
6510 (18.67)
|
951 (23.91)
|
|
Drinking status (n, %)
|
|
|
|
|
<0.001b
|
Never
|
30018 (77.26)
|
|
27323 (78.34)
|
2695 (67.74)
|
|
Former
|
1796 (4.62)
|
|
1610 (4.62)
|
186 (4.68)
|
|
Current
|
7041 (18.12)
|
|
5944 (17.04)
|
1097 (27.58)
|
|
Physical activity, n (%)
|
|
|
|
|
<0.001b
|
Light
|
12563 (32.33)
|
|
11077 (31.76)
|
1486 (37.36)
|
|
Moderate
|
14644 (37.69)
|
|
13222 (37.91)
|
1422 (35.75)
|
|
Vigorous
|
11648 (29.98)
|
|
10578 (30.33)
|
1070 (26.90)
|
|
MET-Hour/day, (mean ± SD)
|
18.09±10.11
|
|
18.09±10.11
|
16.81±10.14
|
<0.001a
|
Sitting time
|
|
|
|
|
<0.001b
|
<4h/d
|
11863 (30.53)
|
|
10846 (31.10)
|
1017 (25.57)
|
|
4-6h/d
|
10377 (26.71)
|
|
9386 (26.91)
|
991 (24.91)
|
|
6-8h/d
|
6498 (26.04)
|
|
5798 (16.62)
|
700 (17.60)
|
|
≥8h/d
|
10117 ()
|
|
8847 (25.37)
|
1270 (31.93)
|
|
Sitting time (h/day, mean ± SD)
|
5.59±3.29
|
|
5.53±3.26
|
6.09±3.50
|
<0.001a
|
Dietary pattern
|
|
|
|
|
<0.001b
|
Pattern I
|
7088 (23.90)
|
|
7088 (23.90)
|
1050 (30.02)
|
|
Pattern Ⅱ
|
7820 (26.36)
|
|
7820 (26.36)
|
803 (22.96)
|
|
Pattern Ⅲ
|
8362 (28.19)
|
|
8362 (28.19)
|
964 (27.56)
|
|
Pattern Ⅳ
|
6393 (21.55)
|
|
6393 (21.55)
|
681 (19.47)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
BMI (kg/m2, mean ± SD)
|
24.84 ± 3.57
|
|
24.63 ± 3.49
|
26.66 ±3.73
|
<0.001a
|
Serum uric acid level (μmol/L, mean ± SD)
|
286.49 ± 79.64
|
|
269.20 ± 60.69
|
438.14 ± 63.57
|
<0.001a
|
SD, standard deviation; T2DM, type 2 diabetes mellitus.
a Student's t-test was used to compare normal distributed continuous variables among the groups; b A Chi-square test was used to test the distributions of categorical variables among the groups.
3.2 Association of PA or ST with serum uric acid level and HUA
Table 2 summarizes the linear relationship between PA level or ST with serum uric acid level. As a whole, the serum uric acid level increased with decreasing PA level or increasing ST. In model 2, compared with light PA or ST of <4h/d, the βs and 95% CIs for moderate and vigorous PA, ST of 4-6, 6-8 and ≥8h/d were 0.76 (-0.97, 2.48) and -4.27 (-6.10, -2.45), 2.74 (0.89, 4.59), 8.12 (5.99, 10.26) and 17.60 (15.70, 19.51), respectively. The βs (95% CIs) for MET-hour/d and per hour increased were -0.15 (-0.22, -0.07) (P for trend<0.001) and 2.12 (1.90, 2.34) (P for trend<0.001). In multivariable-adjusted logistic regression analysis, PA level was inversely associated with the prevalence of HUA, while ST was positively associated with prevalence of HUA (Table 3 and Figure 1). After adjusting for potential covariates in model 2, the ORs (95% CIs) for HUA comparing vigorous and moderate PA to light group were 0.94 (0.87, 1.02), 0.84 (0.77, 0.91), respectively (P for trend <0.001); the ORs (95% CIs) for HUA comparing ST 4-6, 6-8 and ≥8h/d to <4h/d were 1.11 (1.01, 1.22), 1.28 (1.15, 1.42) and 1.50 (1.36, 1.64) (P for trend <0.001); when HUA was analyzed per MET-hour/day and per hour for ST increment, the adjusted ORs (95% CIs) were 0.97 (0.96, 0.99) and 1.05 (1.04, 1.06), respectively (Table 3). In the sex-stratified analyses (Additional file 1: Table S1 and Table S2), similar associations were observed in both males and females. Meantime, in Figure 1, the restricted cubic spline showed that there was a linear dose–response relationship between ST and HUA (P for non-linearity > 0.05), and a non-linear dose–response relationship for PA (P for non-linearity < 0.05). Finally, when the interaction term was included, a significant interaction effect of ST and PA on HUA was observed (Table 3, P<0.001).
Table 2. Association of physical activity level and sitting time with serum uric acid level.
|
Number
|
Serum uric acid level
(μmol/L, mean ± SD)
|
β (95% CI)
|
Model 1
|
Model 2
|
|
Physical activity level
|
|
|
|
|
|
Light
|
12563
|
293.03±82.57
|
0 (Ref.)
|
0 (Ref.)
|
|
Moderate
|
14644
|
280.70±77.92
|
-9.53 (-11.46, -7.60)
|
0.76 (-0.97, 2.48)
|
|
Vigorous
|
11648
|
286.73±78.00
|
-1.75 (-3.80, 0.30)
|
-4.27 (-6.10, -2.45)
|
|
MET-hour/d
|
|
|
-0.29 (-0.37, -0.21)
|
-0.15 (-0.22, -0.07)
|
|
P value for trend
|
|
|
<0.001
|
<0.001
|
|
Sitting time
|
|
|
|
|
|
<4h/d
|
11863
|
278.27±77.77
|
0 (Ref.)
|
0 (Ref.)
|
|
4-6h/d
|
10377
|
282.30±78.25
|
4.85 (2.76, 6.94)
|
2.74 (0.89, 4.59)
|
|
6-8h/d
|
6498
|
287.84±80.75
|
10.75 (8.34, 13.16)
|
8.12 (5.99, 10.26)
|
|
≥8h/d
|
10117
|
299.57±80.81
|
20.63 (18.48, 22.77)
|
17.60 (15.70, 19.51)
|
|
Per hour increased
|
|
|
2.44 (2.19, 2.68)
|
2.12 (1.90, 2.34)
|
|
P value for trend
|
|
|
<0.001
|
<0.001
|
|
Multivariable model 1 adjusted for physical activity and sitting time level where applicable; model 2 included sitting time and physical activity as well as age, gender, education level, marital status, average monthly income, smoking status, drinking status, dietary pattern and BMI SD, standard deviation; CI, confidence interval.
Table 3. Association of physical activity level and sitting time with prevalence of hyperuricemia.
|
Cases/Number
|
Prevalence (95% CI)
|
OR (95%CI)
|
Model 1
|
Model 2
|
|
Physical activity level
|
|
|
|
|
|
Light
|
1486/12563
|
11.83 (11.26, 12.39)
|
1 (Ref.)
|
1 (Ref.)
|
|
Moderate
|
1422/14644
|
9.71 (9.23, 10.19)
|
0.84 (0.78, 0.91)
|
0.94 (0.87, 1.02)
|
|
Vigorous
|
1070/11648
|
9.19 (8.66, 9.71)
|
0.82 (0.75, 0.89)
|
0.84 (0.77, 0.91)
|
|
MET-hour/day
|
|
|
0.95 (0.93, 0.97)
|
0.97 (0.96, 0.99)
|
|
P value for trend
|
|
|
<0.001
|
<0.001
|
|
Sitting time
|
|
|
|
|
|
<4h/d
|
1017/11863
|
8.57 (8.07, 9.08)
|
1 (Ref.)
|
1 (Ref.)
|
|
4-6h/d
|
991/10377
|
9.55 (8.98, 10.12)
|
1.13 (1.03, 1.24)
|
1.11 (1.01, 1.22)
|
|
6-8h/d
|
700/6498
|
10.77 (10.02, 11.53)
|
1.29 (1.16, 1.42)
|
1.28 (1.15, 1.42)
|
|
≥8h/d
|
1270/10117
|
12.55 (11.91, 13.20)
|
1.46 (1.34, 1.60)
|
1.50 (1.36, 1.64)
|
|
Per hour increased
|
|
|
1.04 (1.03, 1.05)
|
1.05 (1.04, 1.06)
|
|
P value for trend
|
|
|
<0.001
|
<0.001
|
|
P Sitting time × Physical activity
|
|
|
0.006
|
<0.001
|
|
Multivariable model 1 adjusted for physical activity and sitting time level where applicable; model 2 included sitting time and physical activity as well as age, gender, education level, marital status, average monthly income, smoking status, drinking status, dietary pattern and BMI. OR, odds ratio; CI, confidence interval.
3.3 Stratified analysis and joint analysis
The results from stratified analysis by ST groups were present in Figure 2. After adjustment for confounders, only in ST <4h/d and 4-6h/d groups, compared with the reference group (light PA), vigorous group had significantly decreased odds of T2DM, but in longer ST groups, no significant associations can be found (all P>0.05). Meanwhile, in the joint analysis (Additional file 1: Table S3 and Figure S1), combinations of the light PA and the highest ST (≥8h/d) were associated with a 75% increased risk for HUA compared with the reference group (OR:1.75, 95% CI: 1.52, 2.01).
3.4 Interaction of PA and ST on HUA
Furthermore, in order to explore the interaction of PA and ST, the effect of MET-hour/day on risk of HUA was plotted as a function of ST. Interaction plot was used to visualize the changes of effects of MET-hour/day on HUA along with increasing ST as shown in Figure 3. In these models, the protective effects of PA-MET (hour/day) on HUA were attenuated with increasing ST (h/day).