U-shaped association between 25-hydroxyvitamin D concentration and the prevalence of asthma in the overweight and obese U.S. population

Figures & data

Figure 1. Flow chart of this study. This study included 59,842 participants who completed the interviews. Of these, 34,770 participants aged >20 years were included for analysis. Exclusion criteria included pregnant women (n = 251), BMI < 25 kg/m², missing variables (25(OH)D (n = 3945), asthma (n = 27), BMI (n = 81), various covariates (n = 24,900)). Ultimately, the cross-sectional analysis comprised 3889 participants from six NHANES cycles conducted between 2007 and 2018.

Table 1. Demographic characteristics of participants by 25-hydroxyvitamin D concentration category.

Variables	25-Hydroxyvitamin D concentration (nmol/l)
	Total (n = 3889)	Group1 (n = 405) (≤46.6)	Group2 (n = 675) (46.7–62)	Group3 (n = 1005) (62.1–78.2)	Group4 (n = 1804) (≥78.3)	p-Value
Sex, n (%)						<0.001
Male	1765 (45.4)	169 (41.7)	335 (49.6)	510 (50.7)	751 (41.6)
Female	2124 (54.6)	236 (58.3)	340 (50.4)	495 (49.3)	1053 (58.4)
Age, mean ± SD	56.0 ± 16.0	46.3 ± 14.5	50.6 ± 15.5	54.8 ± 15.7	61.0 ± 14.9	<0.001
Race, n (%)						<0.001
Mexican American	485 (12.5)	72 (17.8)	129 (19.1)	144 (14.3)	140 (7.8)
Other Hispanic	650 (16.7)	75 (18.5)	134 (19.9)	200 (19.9)	241 (13.4)
Non-Hispanic white	2010 (51.7)	87 (21.5)	257 (38.1)	497 (49.5)	1169 (64.8)
Non-Hispanic black	744 (19.1)	171 (42.2)	155 (23)	164 (16.3)	254 (14.1)
Education level, (year) n (%)						0.03
Below high school	625 (16.1)	54 (13.3)	131 (19.4)	176 (17.5)	264 (14.6)
High school	815 (21.0)	78 (19.3)	142 (21)	209 (20.8)	386 (21.4)
College or above	2449 (63.0)	273 (67.4)	402 (59.6)	620 (61.7)	1154 (64)
PIR, n (%)						0.005
<1.35	909 (23.4)	98 (24.2)	177 (26.2)	251 (25)	383 (21.2)
1.35–3.5	1400 (36.0)	162 (40)	253 (37.5)	343 (34.1)	642 (35.6)
>3.5	1580 (40.6)	145 (35.8)	245 (36.3)	411 (40.9)	779 (43.2)
BMI, n (%)						<0.001
Overweight	1922 (49.4)	147 (36.3)	300 (44.4)	499 (49.7)	976 (54.1)
Obesity	1967 (50.6)	258 (63.7)	375 (55.6)	506 (50.3)	828 (45.9)
Alcohol, n (%)						0.971
No	953 (24.5)	103 (25.4)	163 (24.1)	246 (24.5)	441 (24.4)
Yes	2936 (75.5)	302 (74.6)	512 (75.9)	759 (75.5)	1363 (75.6)
Diabetes, n (%)						<0.001
No	3210 (82.5)	350 (86.4)	582 (86.2)	837 (83.3)	1441 (79.9)
Yes	679 (17.5)	55 (13.6)	93 (13.8)	168 (16.7)	363 (20.1)
Asthma, n (%)						0.001
No	3258 (83.8)	317 (78.3)	574 (85)	869 (86.5)	1498 (83)
Yes	631 (16.2)	88 (21.7)	101 (15)	136 (13.5)	306 (17)
Physical activity, n (%)						0.241
Sedentary	2261 (58.1)	237 (58.5)	377 (55.9)	569 (56.6)	1078 (59.8)
Moderate	903 (23.2)	94 (23.2)	154 (22.8)	236 (23.5)	419 (23.2)
Vigorous	725 (18.6)	74 (18.3)	144 (21.3)	200 (19.9)	307 (17)
Kidney stones, n (%)						0.001
No	3451 (88.7)	376 (92.8)	605 (89.6)	904 (90)	1566 (86.8)
Yes	438 (11.3)	29 (7.2)	70 (10.4)	101 (10)	238 (13.2)
Sleep disorder, (%)						0.004
No	3076 (79.1)	327 (80.7)	539 (79.9)	827 (82.3)	1383 (76.7)
Yes	813 (20.9)	78 (19.3)	136 (20.1)	178 (17.7)	421 (23.3)
Hypertension, n (%)						<0.001
No	2285 (58.8)	272 (67.2)	430 (63.7)	645 (64.2)	938 (52)
Yes	1604 (41.2)	133 (32.8)	245 (36.3)	360 (35.8)	866 (48)
Smoking status, n (%)						<0.001
Never	2183 (56.1)	248 (61.2)	403 (59.7)	558 (55.5)	974 (54)
Current	436 (11.2)	63 (15.6)	85 (12.6)	123 (12.2)	165 (9.1)
Former	1270 (32.7)	94 (23.2)	187 (27.7)	324 (32.2)	665 (36.9)
Vitamin D intake, (mcg) (%)						< 0.001
Q1 (≤4.8)	476 (12.2)	134 (33.1)	128 (19)	112 (11.1)	102 (5.7)
Q2 (5.0–9.98)	452 (11.6)	83 (20.5)	113 (16.7)	129 (12.8)	127 (7)
Q3 (10–24.8)	1521 (39.1)	139 (34.3)	316 (46.8)	474 (47.2)	592 (32.8)
Q4 (≥25)	1440 (37.0)	49 (12.1)	118 (17.5)	290 (28.9)	983 (54.5)

Abbreviations: PIR: ratio of family income to poverty; BMI: body mass index; Q: quartiles.

Baseline characteristics.

Table 1: Baseline characteristics stratified by the 25-hydroxyvitamin D group levels.From the cohort of 3889 participants, the average age was 56.0 ± 16.0 years, with 1765 (45.4%) women and 2124 (54.6%) men. Among them, 631 (16.2%) reported having asthma. Higher levels of 25(OH)D were observed in females, non-Hispanic whites, those with higher school education, and individuals from higher-income families. Conversely, lower 25(OH)D levels were observed in participants with diabetes, kidney stones, sleep disturbance, and hypertension. Participants who had never smoked and those with a higher dietary vitamin D intake in the previous 30 days had higher 25-hydroxyvitamin D concentrations.

Table 2. Associations between covariates and asthma incidence.

Variable	OR_95CI	p_Value	Variable	OR_95 CI	p_Value
Sex, n (%)			Alcohol, n (%)
Male	1 (Reference)		No	1 (Reference)
Female	1.54 (1.29 ∼ 1.84)	<0.001	Yes	1.26 (1.02 ∼ 1.55)	0.029
Age, (years)	0.99 (0.98 ∼ 0.99)	<0.001	Diabetes, N (%)
Race, n (%)			No	1 (Reference)
Mexican American	1 (Reference)		Yes	1.42 (1.15 ∼ 1.75)	0.001
Other Hispanic	1.43 (1.02 ∼ 2.01)	0.037	Kidney, n (%)
Non-Hispanic White	1.36 (1.01 ∼ 1.82)	0.042	No	1 (Reference)
Non-Hispanic Black	1.49 (1.07 ∼ 2.06)	0.018	Yes	1.32 (1.03 ∼ 1.7)	0.029
Education level, (year) n(%)			Sleep disorder, n (%)
Below high school	1 (Reference)		No	1 (Reference)
High school	1.42 (1.05 ∼ 1.9)	0.021	Yes	1.95 (1.61 ∼ 2.35)	<0.001
High school or above	1.34 (1.04 ∼ 1.74)	0.024	Hypertension, n (%)
PIR, n (%)			No	1 (Reference)
<1.35	1 (Reference)		Yes	1.17 (0.98 ∼ 1.38)	0.081
1.35–3.5	0.77 (0.62 ∼ 0.97)	0.024	Vitamin D intake (mcg)
>3.5	0.8 (0.65 ∼ 0.99)	0.041	Q1 (≤4.8)	1 (Reference)
25(OH)D (nmol/l), (IQR)			Q2 (5.0–9.98)	0.94 (0.67 ∼ 1.33)	0.737
Group1 (10.2–46.6)	1(Reference)		Q3 (10–24.8)	0.74 (0.56 ∼ 0.97)	0.03
Group2 (46.7–62)	0.63 (0.46 ∼ 0.87)	0.005	Q4 (≥25)	1.05 (0.8 ∼ 1.37)	0.736
Group3 (62.1–78.2)	0.56 (0.42 ∼ 0.76)	<0.001	Activity n (%)
Group4 (78.3–257)	0.74 (0.56 ∼ 0.96)	0.024	Sedentary	1 (Reference)
Smoking status, n (%)			Moderate	1.05 (0.85 ∼ 1.29)	0.656
Never	1 (Reference)		Vigorous	1.04 (0.83 ∼ 1.31)	0.709
Current	1.49 (1.15 ∼ 1.93)	0.003	BMI n (%)
Former	1.22 (1.01 ∼ 1.47)	0.036	Overweight	1 (Reference)
			Obesity	1.52 (1.28 ∼ 1.81)	<0.001

Abbreviations: OR: odds ratio; CI: confidence interval; Q: quartiles.

Association between 25-Hydroxyvitamin D and asthma.

Table 2: In the univariate analysis, lower incidence of asthma was significantly associated with higher levels of 25(OH)D. Likewise, high income households exhibited decreased incidence of asthma compared to low income households. Compared to the Q1 group (vitamin D intake ≤4.8 mcg), there was a decrease in the incidence of asthma as dietary supplement vitamin D doses increased (10–24.8 mcg); however, with further dose increases, asthma incidence no longer decreased.

Table 3. Association between 25-hydroxyvitamin D concentration and asthma.

Variable	n.Total	n. Event_%	Crude model		Model-I		Model-II		Model-III
			OR_95CI	p Value	OR_95 CI	p Value	OR_95 CI	p Value	OR_95 CI	p Value
25(OH)D (nmol/l)
Group1 (10.2–46.6)	405	88(21.7)	1 (Ref)		1 (Ref)		1 (Ref)		1 (Ref)
Group2 (46.7–62)	675	101(15)	0.63 (0.46 ∼ 0.87)	0.005	0.68 (0.49 ∼ 0.94)	0.021	0.69 (0.5 ∼ 0.95)	0.024	0.68 (0.49 ∼ 0.94)	0.021
Group3 (62.1–78.2)	1005	136(13.5)	0.56 (0.42 ∼ 0.76)	<0.001	0.63 (0.46 ∼ 0.86)	0.003	0.64 (0.47 ∼ 0.87)	0.004	0.65 (0.47 ∼ 0.88)	0.006
Group4 (78.3–257)	1804	306(17)	0.74 (0.56 ∼ 0.96)	0.024	0.83 (0.62 ∼ 1.11)	0.214	0.84 (0.63 ∼ 1.13)	0.256	0.83 (0.61 ∼ 1.12)	0.22

Abbreviations: OR: odds ratio; CI: confidence interval; Ref: reference.

Crude model was the non-adjusted model.

Model 1 was adjusted for sociodemographic variables (age, sex, race/ethnicity).

Model 2 was adjusted for sociodemographic (age, sex, race/ethnicity, education level, family income).

Model 3 was adjusted for sociodemographic (age, sex, race/ethnicity, education level, household income), alcohol use, diabetes, physical activity, kidney stones, hypertension, sleep disturbance, smoking status, and dietary vitamin D intake.

Adjustment for confounders.

Table 3: Comparison of asthma ORs for different quartiles of 25(OH)D after adjusted for confounders.Relative with reference group 1 (≤46.6 nmol/L), ORs for groups 2, 3 and 4 were 0.68 (95% CI: 0.49–0.94, p = 0.021), 0.65 (95% CI: 0.47–0.88, p = 0.006) and 0.83 (95% CI: 0.61–1.12, p = 0.22).

Threshold effect analysis of the associations between 25(OH) D Concentration and asthma.

Table 4: In the threshold analysis, the OR for developing asthma was 0.905 (95%CI: 0.838–0.976, p = 0.010), indicating a 9.5% decrease in asthma risk with every 10 nmol/L increase in 25(OH)D levels. However, no significant association was found when the levels exceeded 84.95 nmol/L, suggesting that further increments in 25(OH)D levels did not reduce the incidence of asthma.

Figure 2. Curve fitting. Using a fully adjusted model, RCS (restricted cubic spline) regression analysis revealed a non-linear association (p = 0.017) between 25(OH)D levels and asthma, with an inflection point at approximately 84.95 nmol/L. Odds ratio between 25-hydroxyvitamin D level and asthma. Solid and dashed lines represent predicted values and 95% confidence intervals. Red represents people with asthma, and blue represents people without asthma. Adjusted for age, sex, race/ethnicity, education level, household income, smoking status, physical activity, hypertension, diabetes, kidney stones, sleep disturbance, and dietary vitamin D intake.

Table 4. Threshold effect analysis of the associations between 25(OH) D concentration and asthma.

25(OH)D 10 nmol/l	Adjusted model
	OR (95%CI)	p-Value
<84.95	0.905 (0.838 ∼ 0.976)	0.010
≥84.95	1.069 (0.981 ∼ 1.165)	0.127
Likelihood ratio test		0.015

Abbreviations: OR: odds ratio; CI: confidence interval.

Adjustments were made for demographics (age, sex, race/ethnicity, education level, household income), smoking status, physical activity, hypertension, diabetes, kidney stones, sleep disturbance, and vitamin D intake. Only 99% of the data is shown.

Figure 3. Stratified analyses based on additional variables. Subgroup analyses by sex, age, diabetes, sleep disturbance, and smoking status showed no significant interactions, confirming the robustness of the results. The association between 25-hydroxyvitamin D level and asthma according to basic features. Except for the stratification component itself, each stratification factor was adjusted for all other variables (age, sex, race/ethnicity, education level, household income, smoking status, physical activity, hypertension, diabetes, kidney stones, sleep disturbance, vitamin D intake).

Data availability statement

Publicly available datasets for this study can be found online.The repository/repositories name and accession numbers are available online at http://www.cdc.gov/ nchs/nhanes.htm.