Basic characteristics of participants
Out of the NHANES 1999-2004 data, a total of 31,126 participants were obtained. Exclusions were made for 27,025 participants who were pregnant (968), participants aged below 20 years (15,659), and participants with missing data (10,398). Consequently, the present study encompassed 4,101 participants. The exclusion process was detailed in Figure 1. In the study, a total of 4,101 participants were included, among whom 702 (17.12%) reported experiencing migraine. The mean age of the participants was 52.98 ± 16.92, with 2,127 (51.87%) of them being men. It was observed that individuals who were men, younger, other race, better educated, married, had higher income, non-drinkers, and never smokers, tended to consume higher amounts of vitamin C. The basic characteristics of the participants were listed in Table 1.
Table 1 Basic characteristics of the included participants
Variables
|
ln (vitamin C intake), mg
|
P-value
|
Total
|
Q1 (<3.424 )
|
Q2 (3.424-4.267 )
|
Q3 (4.268-4.932)
|
Q4 (>4.932)
|
No.
|
4101
|
1023
|
1027
|
1025
|
1026
|
|
Gender, n (%)
|
|
|
|
|
|
< 0.001
|
Male
|
2127 (51.87)
|
512 (50.05)
|
507 (49.37)
|
520 (50.73)
|
588 (57.31)
|
|
Female
|
1974 (48.13)
|
511 (49.95)
|
520 (50.63)
|
505 (49.27)
|
438 (42.69)
|
|
Age, Mean ± SD
|
52.98 ± 16.92
|
50.37 ± 16.57
|
52.55 ± 16.51
|
55.57 ± 17.25
|
53.41 ± 16.96
|
< 0.001
|
Race, n (%)
|
|
|
|
|
|
0.314
|
Mexican American
|
601 (14.65)
|
152 (14.86)
|
150 (14.61)
|
145 (14.15)
|
154 (15.01)
|
|
Other Hispanic
|
143 ( 3.49)
|
31 (3.03)
|
42 (4.09)
|
33 (3.22)
|
37 (3.61)
|
|
Non-Hispanic white
|
2608 (63.59)
|
639 (62.46)
|
663 (64.56)
|
664 (64.78)
|
642 (62.57)
|
|
Non-Hispanic black
|
606 (14.78)
|
173 (16.91)
|
141 (13.73)
|
147 (14.34)
|
145 (14.13)
|
|
Other race
|
143 ( 3.49)
|
28 (2.74)
|
31 (3.02)
|
36 (3.51)
|
48 (4.68)
|
|
Education, n (%)
|
|
|
|
|
|
< 0.001
|
Less than high school
|
711 (17.34)
|
223 (21.8)
|
169 (16.46)
|
171 (16.68)
|
148 (14.42)
|
|
High school diploma
|
906 (22.09)
|
261 (25.51)
|
205 (19.96)
|
241 (23.51)
|
199 (19.4)
|
|
More than high school
|
2484 (60.57)
|
539 (52.69)
|
653 (63.58)
|
613 (59.8)
|
679 (66.18)
|
|
Marital status, n (%)
|
|
|
|
|
|
0.034
|
Married
|
2649 (64.59)
|
621 (60.7)
|
673 (65.53)
|
694 (67.71)
|
661 (64.42)
|
|
Living with a partner
|
1279 (31.19)
|
348 (34.02)
|
312 (30.38)
|
298 (29.07)
|
321 (31.29)
|
|
Living alone
|
173 ( 4.22)
|
54 (5.28)
|
42 (4.09)
|
33 (3.22)
|
44 (4.29)
|
|
Family income, n (%)
|
|
|
|
|
|
< 0.001
|
Low (PIR ≤ 1.3)
|
648 (15.80)
|
203 (19.84)
|
164 (15.97)
|
141 (13.76)
|
140 (13.65)
|
|
Medium (PIR > 1.3 to 3.5)
|
1445 (35.24)
|
381 (37.24)
|
349 (33.98)
|
388 (37.85)
|
327 (31.87)
|
|
High (PIR > 3.5)
|
2008 (48.96)
|
439 (42.91)
|
514 (50.05)
|
496 (48.39)
|
559 (54.48)
|
|
BMI, Mean ± SD
|
28.26 ± 5.94
|
28.58 ± 6.07
|
28.74 ± 6.09
|
27.95 ± 5.43
|
27.75 ± 6.08
|
< 0.001
|
CRP, Median (IQR)
|
0.21 (0.08, 0.45)
|
0.23 (0.10, 0.48)
|
0.23 (0.09, 0.49)
|
0.20 (0.08, 0.41)
|
0.18 (0.07, 0.42)
|
< 0.001
|
Diabetes, n (%)
|
|
|
|
|
|
0.04
|
No
|
3686 (89.88)
|
917 (89.64)
|
923 (89.87)
|
903 (88.1)
|
943 (91.91)
|
|
Yes
|
415 (10.12)
|
106 (10.36)
|
104 (10.13)
|
122 (11.9)
|
83 (8.09)
|
|
Migraine, n (%)
|
|
|
|
|
|
< 0.001
|
No
|
3399 (82.88)
|
805 (78.69)
|
835 (81.3)
|
867 (84.59)
|
892 (86.94)
|
|
Yes
|
702 (17.12)
|
218 (21.31)
|
192 (18.7)
|
158 (15.41)
|
134 (13.06)
|
|
Physical activity, n (%)
|
|
|
|
|
|
0.841
|
Moderate activities
|
3434 (83.74)
|
852 (83.28)
|
854 (83.15)
|
865 (84.39)
|
863 (84.11)
|
|
Vigorous activities
|
667 (16.26)
|
171 (16.72)
|
173 (16.85)
|
160 (15.61)
|
163 (15.89)
|
|
Drinking status, n (%)
|
|
|
|
|
|
0.937
|
Non-drinker
|
1123 (27.38)
|
274 (26.78)
|
288 (28.04)
|
281 (27.41)
|
280 (27.29)
|
|
Drinker
|
2978 (72.62)
|
749 (73.22)
|
739 (71.96)
|
744 (72.59)
|
746 (72.71)
|
|
Smoking status, n (%)
|
|
|
|
|
|
< 0.001
|
Never smokers
|
2156 (52.57)
|
475 (46.43)
|
533 (51.9)
|
561 (54.73)
|
587 (57.21)
|
|
Current smokers
|
627 (15.29)
|
250 (24.44)
|
157 (15.29)
|
107 (10.44)
|
113 (11.01)
|
|
Former smokers
|
1318 (32.14)
|
298 (29.13)
|
337 (32.81)
|
357 (34.83)
|
326 (31.77)
|
|
Coronary heart disease, n (%)
|
|
|
|
|
|
0.099
|
No
|
3855 (94.00)
|
976 (95.41)
|
953 (92.79)
|
962 (93.85)
|
964 (93.96)
|
|
Yes
|
246 ( 6.00)
|
47 (4.59)
|
74 (7.21)
|
63 (6.15)
|
62 (6.04)
|
|
Stroke, n (%)
|
|
|
|
|
|
0.405
|
No
|
3975 (96.93)
|
988 (96.58)
|
991 (96.49)
|
1001 (97.66)
|
995 (96.98)
|
|
Yes
|
126 ( 3.07)
|
35 (3.42)
|
36 (3.51)
|
24 (2.34)
|
31 (3.02)
|
|
Hypertension, n (%)
|
|
|
|
|
|
0.405
|
No
|
2632 (64.18)
|
663 (64.81)
|
655 (63.78)
|
639 (62.34)
|
675 (65.79)
|
|
Yes
|
1469 (35.82)
|
360 (35.19)
|
372 (36.22)
|
386 (37.66)
|
351 (34.21)
|
|
Hypercholesterolaemia, n (%)
|
|
|
|
|
|
0.194
|
No
|
2470 (60.23)
|
633 (61.88)
|
596 (58.03)
|
607 (59.22)
|
634 (61.79)
|
|
Yes
|
1631 (39.77)
|
390 (38.12)
|
431 (41.97)
|
418 (40.78)
|
392 (38.21)
|
|
Energy, Mean ± SD
|
2125.14 ± 994.71
|
1809.18 ± 817.47
|
2084.32 ± 967.54
|
2132.89 ± 960.47
|
2473.28 ± 1100.06
|
< 0.001
|
Protein, Mean ± SD
|
81.50 ± 40.40
|
68.53 ± 32.90
|
79.78 ± 37.92
|
83.87 ± 39.22
|
93.80 ± 46.32
|
< 0.001
|
Carbohydrate, Mean ± SD
|
260.68 ± 128.32
|
210.44 ± 105.75
|
249.65 ± 119.34
|
262.22 ± 120.26
|
320.28 ± 140.75
|
< 0.001
|
Fat, Median (IQR)
|
71.03 (48.42, 102.47)
|
64.74 (44.44, 91.20)
|
71.43 (50.49, 101.98)
|
70.94 (47.36, 103.69)
|
77.70 (53.01, 111.91)
|
< 0.001
|
Abbreviations: PIR, poverty income ratio.
Association between dietary vitamin C intake and migraine
The univariate analysis showed that gender, age, marital status, race/ethnicity, family income, BMI, physical activity, drinking status, smoking status, and coronary heart disease are influential factors associated with the risk of migraine (Table 2). Dietary vitamin C intake was assessed both as continuous and categorical variables using multivariable logistic regression analysis. The study findings indicated a negative association between vitamin C intake and migraine after adjusting for all covariates (OR=0.89, 95%CI: 0.83-0.96, P=0.002). The association remained similar when categorizing vitamin C intake. After adjusting for all covariates, the ORs (95% CI) of Q2, Q3, and Q4 were 0.92 (95% CI: 0.73-1.16, P=0.485), 0.80 (95% CI: 0.63-1.03, P=0.082), and 0.64 (95% CI: 0.49-0.84, P=0.001) compared with Q1 (Table 3). Additionally, the RCS analysis discovered a linear negative relationship between dietary vitamin C intake and the risk of migraine (P for non-linearity=0.449) (Figure 2).
Table 2 Association of covariates and migraine risk
Variable
|
OR (95%CI)
|
P-value
|
Gender, n (%)
|
|
|
Male
|
1 (Ref)
|
|
Female
|
2.13 (1.80-2.52)
|
<0.001
|
Age (year)
|
0.97 (0.96-0.97)
|
<0.001
|
Race/ethnicity, n (%)
|
|
|
Mexican American
|
1 (Ref)
|
|
Other Hispanic
|
1.26 (0.82-1.95)
|
0.291
|
Non-Hispanic White
|
0.74 (0.59-0.93)
|
0.010
|
Non-Hispanic Black
|
0.95 (0.71-1.26)
|
0.718
|
Other Race
|
1.08 (0.69-1.69)
|
0.752
|
Education, n (%)
|
|
|
Less than high school
|
1 (Ref)
|
|
High school diploma
|
1.00 (0.77-1.28)
|
0.970
|
More than high school
|
0.88 (0.70-1.09)
|
0.233
|
Marital status, n (%)
|
|
|
Married
|
1 (Ref)
|
|
Living with a partner
|
1.08 (0.91-1.29)
|
0.374
|
Living alone
|
1.58 (1.09-2.27)
|
0.015
|
PIR, n (%)
|
|
|
Low (PIR ≤ 1.3)
|
1 (Ref)
|
|
Medium (PIR > 1.3 to 3.5)
|
0.86 (0.68-1.08)
|
0.188
|
High (PIR > 3.5)
|
0.63 (0.50-0.79)
|
<0.001
|
BMI (kg/m2)
|
1.02 (1.01-1.04)
|
<0.001
|
CRP (mg/dl)
|
1.07 (0.98-1.16)
|
0.140
|
Diabetes, n (%)
|
|
|
No
|
1 (Ref)
|
|
Yes
|
0.94 (0.72-1.24)
|
0.676
|
Physical activity, n (%)
|
|
|
Moderate activities
|
1 (Ref)
|
|
Vigorous activities
|
1.30 (1.05-1.6)
|
0.014
|
Drinking status, n (%)
|
|
|
No
|
1 (Ref)
|
|
Yes
|
0.70 (0.59-0.83)
|
<0.001
|
Smoking status, n (%)
|
|
|
Never smokers
|
1 (Ref)
|
|
Current smokers
|
1.19 (0.96-1.49)
|
0.113
|
Former smokers
|
0.62 (0.51-0.75)
|
<0.001
|
Coronary heart disease, n (%)
|
|
|
No
|
1 (Ref)
|
|
Yes
|
0.53 (0.35-0.81)
|
0.003
|
Stroke, n (%)
|
|
|
No
|
1 (Ref)
|
|
Yes
|
1.14 (0.73-1.80)
|
0.559
|
Hypertension, n (%)
|
|
|
No
|
1 (Ref)
|
|
Yes
|
0.94 (0.79-1.11)
|
0.465
|
Hypercholesterolaemia, n (%)
|
|
|
No
|
1 (Ref)
|
|
Yes
|
0.99 (0.84-1.17)
|
0.920
|
Energy (kcal)
|
1.00 (1.00-1.00)
|
0.389
|
Protein (g)
|
1.00 (1.00-1.00)
|
0.100
|
Carbohydrate (g)
|
1.00 (1.00-1.00)
|
0.731
|
Fat (g)
|
1.00 (1.00-1.00)
|
0.752
|
Abbreviations: CI, confidence interval; OR, odds ratio.
Table 3 Association between dietary vitamin C intake and the risk of migraine
Dietary vitamin C intake (mg)
|
Event (%)
|
Non-adjusted Model
|
Model 1
|
Model 2
|
Model 3
|
OR (95%CI)
|
P-value
|
OR (95%CI)
|
P-value
|
OR (95%CI)
|
P-value
|
OR (95%CI)
|
P-value
|
ln (vitamin C)
|
704 (17.1)
|
0.86 (0.8-0.91)
|
<0.001
|
0.90 (0.84-0.96)
|
0.002
|
0.90 (0.84-0.97)
|
0.003
|
0.89 (0.83-0.96)
|
0.002
|
Quartiles, ln (vitamin C)
|
|
|
|
|
|
|
|
|
|
Q1 (<3.424)
|
218 (21.3)
|
1 (Ref)
|
|
1 (Ref)
|
|
1 (Ref)
|
|
1 (Ref)
|
|
Q2 (3.424-4.267)
|
192 (18.7)
|
0.85 (0.68-1.05)
|
0.139
|
0.94 (0.75-1.18)
|
0.606
|
0.92 (0.73-1.16)
|
0.496
|
0.92 (0.73-1.16)
|
0.485
|
Q3 (4.268-4.932)
|
158 (15.4)
|
0.67 (0.54-0.84)
|
0.001
|
0.81 (0.64-1.02)
|
0.072
|
0.82 (0.65-1.04)
|
0.105
|
0.80 (0.63-1.03)
|
0.082
|
Q4 ( >4.932)
|
134 (13.1)
|
0.55 (0.44-0.70)
|
<0.001
|
0.66 (0.52-0.84)
|
0.001
|
0.67 (0.52-0.85)
|
0.001
|
0.64 (0.49-0.84)
|
0.001
|
Trend test
|
|
|
<0.001
|
|
<0.001
|
|
0.001
|
|
0.001
|
Abbreviations: CI, confidence interval; OR, odds ratio; ln (vitamin C), vitamin C intake values were natural logarithmised.
Model 1 was adjusted for demographic characteristics, including age, gender, race/ethnicity, marital status, education level, and family income.
Model 2 was adjusted for smoking status, drinking status, BMI, physical activity, CRP, hypertension, hypercholesterolaemia, stroke, diabetes and coronary heart disease in addition to the variables from model 1.
Model 3 was was fully adjusted, including nutrients (energy, proteins, carbohydrates, and fat) and variables from model 2.
Adjustment factors included age, gender, race/ethnicity, marital status, education level, family income, smoking status, drinking status, BMI, physical activity, CRP, hypertension, hypercholesterolaemia, stroke, diabetes, coronary heart disease, energy, proteins, carbohydrates, and fat.
Stratified and interaction analyses
The results consistently showed a negative relationship between vitamin C intake and the risk of migraine in stratified analyses (Figure 3). There were no significant interactions observed in any of the subgroups (P>0.05).