Vitamin D Status,Fiber Intake,and Type 2 Diabetes in U.S. Adults

Abstract

Vitamin D and fiber intake are nutritional factors that could affect the development of type 2 diabetes (T2D), potentially by reducing insulin resistance. Therefore, we hypothesized that the influence of vitamin D on T2D might depend on fiber intake. This study investigated the association between vitamin D status and T2D according to fiber intake. The present study analyzed data from 9,656 American adults (≥20 years old) who participated in the National Health and Nutrition Examination Survey (NHANES) 2007–2010. The serum concentration of 25-hydroxyvitamin D [25(OH)D] was used as a biomarker for vitamin D status. The T2D classification was based on two criteria: T2D^a was identified using only self-reported questionnaire data and T2D^b was identified based on both survey and laboratory data. The deficient vitamin D status (<50 nmol/L) was used as the reference group. After controlling for sociodemographic, behavioral, and dietary factors, the odds ratios (ORs) were 0.72 (95% confidence interval [CI]: 0.58, 0.90) for T2D^a and 0.60 (0.50, 0.80) for T2D^b in the sufficient vitamin D status (≥75 nmol/L). Furthermore, the total vitamin D concentration exhibited dose-dependent associations with lower OR values for T2D^a (P for trend = .005) and T2D^b (P for trend <.001). Among participants with high-fiber intake, the OR values for T2D^b were 0.60 (95% CI: 0.42, 0.90) at suboptimal vitamin D status and 0.49 (95% CI: 0.31, 0.77) at sufficient vitamin D status. Moreover, the significant dose-dependent association persisted in the high-fiber-intake subgroup (P for trend = .004). Therefore, combining vitamin D plus high-fiber intake would help reduce the prevalence of diabetes, although the interaction analysis results were not statistically significant.

Introduction

Based on U.S. survey data, the prevalence of diabetes in adults is 12–14%,¹ which may lead to growth in related morbidity, mortality, and economic costs.² Therefore, it is essential to manage and treat type 2 diabetes (T2D), for preventing the development of cardiovascular disease,³ premature death, and all-cause mortality.⁴

Fiber intake is a well-known dietary factor that could reduce the risk of T2D⁵ and insulin resistance.⁶ One meta-analysis also revealed that dietary fiber intake may help reduce the prevalence of T2D.⁷ Despite these results, most Americans consume less than the recommended daily amount of dietary fiber (14 g of total fiber per 1000 kcal, women; 25 g/day, men; 38 g/day).⁸

The National Health and Nutrition Examination Survey (NHANES) 2011–2014 revealed a risk of vitamin D insufficiency in 18% of the U.S. population that was ≥1 year old.⁹ Vitamin D has also been investigated as a relevant factor for preventing the development of T2D.¹⁰ For example, Bourlon et al. reported that the active form of vitamin D is involved in promoting the biosynthesis of beta cells.¹¹ Also, Kjalarsdottir et al. suggested that vitamin D could increase insulin secretion in the pancreas.¹² A recent review also indicated that sufficient vitamin D concentrations have a positive effect on T2D.¹³

Previous studies have revealed that sufficient vitamin D concentrations reduce insulin resistance,^14,15 and dietary fiber has also been reported to be involved in reducing insulin resistance.^16,17 However, previous studies have only evaluated these as separate factors. Therefore, we hypothesized that the coexistence of sufficient vitamin D concentrations and sufficient fiber intake might have a synergistic effect on reducing insulin resistance and thereby reducing the prevalence of T2D. The present study conducted the joined effects of vitamin D concentration and fiber intake on the prevalence of diabetes among American adults.

Methods

Study population

Data in this study includes participants ≥20 years of age from the NHANES 2007–2010, conducted by the Centers for Disease Control and Prevention (CDC). The NHANES is a health-related survey program to assess the health and nutritional status from a representative sample of the civilian noninstitutionalized U.S. population. The survey uses a complex, multistaged, stratified, and clustered design and included household interview such as demographic, socioeconomic, behavioral, and dietary questionnaires. Oversampled participants included Hispanics, non-Hispanic blacks, low-income Whites, and persons who were ≥80 years old. Demographic and health history information were collected through extensive household interviews, and a mobile examination center (MEC) was used to collect blood and urine samples and to perform standardized physical examinations.

The present study included two cycles from NHANES (2007–2008, 2009–2010), and NHANES data from 1999 to 2006 were omitted because the serum 25-hydroxyvitamin D [25(OH)D] concentrations were measured using a different method (i.e., radioimmunoassay). During 2007–2010, the NHANES enrolled 20,686 participants, 11,766 adults (≥20 years old) had completed data on both interview and examination. The present study excluded participants with missing information regarding serum 25(OH)D concentrations (n = 1,435) or other covariates (n = 672), and a total of 9,659 participants were included to our analysis (Fig. 1).

FIG. 1.

Description of the study population.

The NCHS Research Ethics committee reviewed and approved the study protocol (continuation of protocol #2005-06). Written informed consent was obtained from all NHANES participants. The NHANES dataset is publicly available, and additional details regarding the study procedures, data documentation, and questionnaires are available elsewhere.^18,19

Measurement of 25(OH)D concentrations

Blood samples for measurement of vitamin D status were collected from each participant at the MEC. Those blood samples were immediately frozen at −30°C and shipped to the National Center for Environmental Health (CDC, Atlanta, GA) for analysis. Serum samples were subjected to ultra high-performance liquid chromatography–tandem mass spectrometry (ThermoElectron Corp., West Palm Beach, FL) for the quantitative detection of 25-hydroxyvitamin D₃ [25(OH)D₃], 25-hydroxyvitamin D₂ [25(OH)D₂], and 3-epi-25-hydroxyvitamin D₃ [3-epi-25(OH)D₃] concentrations.^20,21

The total vitamin D status was defined as the sum of the 25(OH)D₃ and 25(OH)D₂ concentrations in SI units of nmol/L, but did not include the 3-epi-25(OH)D₃ concentration, which is the predominant form in infants (<1 year old).²² For the statistical analyses, the total vitamin D status was classified into three groups based on the cutoff points suggested by the U.S. Endocrine Society²³: <50 nmol/L (deficient), 50–74.99 nmol/L (suboptimal), and ≥75 nmol/L (sufficient).

Fiber intake

Information regarding fiber intake for the previous 24-hour period was collected via a dietary recall interview at the MEC. The daily fiber intake was adjusted by the total energy intake using the residual method,²⁴ and categorized as high or low intake based on the average value (15.3 g/day).

Identification and classification of T2D

T2D was identified based on two sets of criteria, with positive results classified as T2D^a and T2D^b. T2D^a was identified based only on self-reported questionnaire data indicating (1) a physician diagnosis of diabetes, (2) the current use of antidiabetes medication, or (3) the current use of insulin. T2D^b was identified based on the subject fulfilling the criteria for T2D^a plus one or more of following laboratory criteria: (1) glycosylated hemoglobin A1c concentration of ≥6.5%, (2) fasting glucose concentration of ≥126 mg/dL, or (3) an oral glucose tolerance test result of ≥200 mg/dL.

Covariates

As potential confounders, we considered the participants' age, sex, race/ethnicity, education, season of examination, physical activity, smoking status, dietary intakes (total food energy and energy-adjusted protein, vitamin D, and calcium values), body mass index (BMI), household income, family history, and supplement use. Age was categorized as 20–29 years, 30–39 years, 40–49 years, 50–59 years, 60–69 years, and ≥70 years. Race/ethnicity was categorized as non-Hispanic white (reference), Mexican American, non-Hispanic black, and other. We categorized education as completion of less than 9th grade (reference), high school graduate or less, and more than high school graduate. Smoking status was categorized as never (reference), former, and current smoker. Season of examination was classified as November–April (reference) or May–October. Physical activity was self-reported for the three levels of activity intensity (walking, moderate, and vigorous activity) and computed using metabolic equivalent (MET) value.²⁵

Information regarding dietary intakes (total food energy, protein, vitamin D, and calcium) was collected for the previous 24-hour period through a dietary recall interview immediately before the interview at the MEC. Dietary factors (protein, vitamin D, and calcium) were adjusted based on total energy intake using the residual method.

Household income was calculated based on the poverty-to-income ratio (PIR) with family size-specific thresholds, and was categorized as <1 (below the poverty line), 1 to <3, and ≥3. Use of supplements within the previous 30 days (vitamins, minerals, herbal supplements, or other dietary supplements) was self-reported and categorized as user (reference) or nonuser. The risk factors for diabetes were classified as yes (reference) or no based on the following question: “Have you ever been told by a doctor or other health professional that you have health conditions or a medical or family history that increases your risk of diabetes?” HOMA-IR was calculated using the following formula: HOMA-IR = [fasting serum insulin (mU/L) × fasting plasma glucose (mmol/L)]/22.5.

Statistical analyses

Data analyses were performed using the SAS survey procedure (version 9.4; SAS Institute, Inc., Cary, NC) to account for the complex survey design and the sample weights of the 2007–2010 NHANES. We calculated the 4-year sample weights from the 2-year weights for 2007–2008 and the 2-year weights for 2009–2010 according to the NHANES analysis guideline. P-values were two sided and considered statistically significant at <.05. In the descriptive analyses, we presented the weighted means and standard error of the mean for continuous variables, and the weighted percentages (%) for categorical variables. Continuous variables were analyzed using the survey t-test and categorical variables were analyzed using the survey (Rao-Scott) χ ² test.

The association between the prevalence of T2D and vitamin D concentrations was analyzed for a self-reported diagnosis of T2D^a and for a confirmed diagnosis of T2D^b. The analysis was also performed according to fiber intake. Four survey logistic regression models (PROC SURVEYLOGISTIC) were used to assess the influence of potential confounding factors. Model A considered age, sex, race/ethnicity, education, smoking status, season of examination, and physical activity. Model B considered the covariates from Model A plus BMI. Model C considered the covariates from Model B plus energy-adjusted dietary covariates (i.e., intakes of total energy, protein, vitamin D, and calcium). In Supplementary Tables 3 and 4, model D considered the covariates from Model C plus household income, supplement use, and family history. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated to estimate the association between total 25(OH)D concentration and the prevalence of T2D, with the deficient group (serum 25(OH)D of <50 nmol/L) defined as the reference group. Tests of trend were conducted for total vitamin D concentration as the ordinal variable in logistic regression models using integer values (1–3). In addition, we examined the multiplicative interaction between total vitamin D concentration and fiber intake on the prevalence of T2D, and calculated the interaction P-value using Wald tests. We confirmed the multicollinearity, and except for the factors (intake of carbohydrate, fat, and fiber) that show multicollinearity (variance inflation factor >10).

Table 3.

Adjusted Odds Ratio and 95% Confidence Interval of Type 2 Diabetes Stratified According to Total 25(OH)D Concentration

Variables	No. of T2D /no. of participants	Model A	Model B	Model C
Type 2 diabetes^a (n = 9,659)
Total 25(OH)D (nmol/L)
<50 (deficient)	484/3209	1 (reference)	1 (reference)	1 (reference)
50 ∼ 74.99 (suboptimal)	466/3652	0.80 (0.64, 1.00)^*	0.90 (0.72, 1.12)	0.90 (0.72, 1.13)
≥75 (sufficient)	294/2798	0.55 (0.44, 0.70)^***	0.72 (0.57, 0.89)^**	0.72 (0.58, 0.90)^**
P-trend		<.0001	.003	.005
Type 2 diabetes^b (n = 4,625)
Total 25(OH)D (nmol/L)
<50 (deficient)	559/1553	1 (reference)	1 (reference)	1 (reference)
50 ∼ 74.99 (suboptimal)	558/1794	0.67 (0.53, 0.85)^***	0.77 (0.61, 0.96)^*	0.76 (0.60, 0.97)^*
≥75 (sufficient)	316/1278	0.45 (0.34, 0.59)^***	0.59 (0.45, 0.78)^***	0.60 (0.50, 0.80)^***
P-trend		<.0001	<.001	<.001

***

P < .001, ^** P < .01, ^* P < .05 compared with reference group (vitamin levels <50 nmol/L).

Diabetes was diagnosed based on survey only.

Diabetes was diagnosed based on survey and lab data.

Model A was adjusted for age, sex, race/ethnicity, education level, smoking status, season of examination, physical activity.

Model B: model A+ further adjusted for BMI.

Model C: model B+ further adjusted for intake of total energy and energy adjusted dietary factor (protein, calcium, vitamin D).

Results

Descriptive characteristics

Table 1 summarizes the participants' physical and dietary characteristics according to their total vitamin D status. Relative to the deficient, participants with sufficient total vitamin D were more likely to be older, taller, have a lower body weight, and have a higher physical activity. For diet-related factors, intakes of total energy, energy-adjusted protein, fiber, vitamin D, and calcium were higher in participants with sufficient total vitamin D. Table 2 presents the participants' demographic characteristics stratified by total vitamin D. Participants with sufficient total vitamin D compared with the deficient were more high likely to be non-Hispanic white, former smokers, highly educated, evaluated during May–October, have a higher household income, and use supplements. In contrast, participants with deficient total vitamin D were to have a BMI of ≥30 kg/m² and to have a high risk of T2D.

Table 1.

Physical and Dietary Characteristic Stratified According to Total 25(OH)D Concentration

Characteristic	Total 25(OH)D concentration (nmol/L)
Characteristic	<50 (n = 3,209)	50 ∼ 74.99 (n = 3,652)	≥75 (n = 2,798)	P for trend
Age (year)	44.74 ± 0.56	46.48 ± 0.34	49.46 ± 0.58	<.0001
Height (cm)	167.20 ± 0.28	169.64 ± 0.23	169.04 ± 0.26	.0001
Weight (kg)	86.68 ± 0.76	84.09 ± 0.63	78.56 ± 0.40	<.0001
Physical activity (METs-hour)	51.45 ± 2.04	60.83 ± 1.96	67.27 ± 3.27	.001
Dietary factor
Total energy intake (kcal)	2067.4 ± 20.4	2233.9 ± 30.3	2156.6 ± 22.2	.015
Protein intake (g)	77.84 ± 0.88	86.57 ± 1.50	83.35 ± 1.03	.002
Energy adjusted protein (g)	71.70 ± 0.62	74.34 ± 0.67	73.97 ± 0.67	.020
Carbohydrate intake (g)	252.77 ± 2.69	269.14 ± 3.37	257.82 ± 2.68	.352
Energy adjusted carbohydrate (g)	231.93 ± 1.25	229.59 ± 1.59	227.54 ± 1.39	.016
Fiber intake (g)	15.03 ± 0.28	17.13 ± 0.37	17.09 ± 0.39	<.001
Energy adjusted fiber (g)	14.19 ± 0.24	15.46 ± 0.26	15.87 ± 0.36	.001
Total fat intake (g)	77.47 ± 0.99	84.82 ± 1.49	81.61 ± 0.93	.013
Energy adjusted fat (g)	68.54 ± 0.54	68.51 ± 0.45	69.20 ± 0.45	.340
Total vitamin D intake (mcg)	3.65 ± 0.07	5.02 ± 0.15	5.25 ± 0.15	<.0001
Energy adjusted vitamin D (mcg)	3.52 ± 0.07	4.44 ± 0.12	4.80 ± 0.12	<.0001
Calcium intake (mg)	847.16 ± 10.81	1024.6 ± 19.9	1030.0 ± 14.7	<.0001
Energy adjusted calcium (g)	792.86 ± 7.82	887.09 ± 9.49	935.01 ± 11.92	<.0001

Survey regression for continuous variable were used.

25(OH)D, 25-hydroxyvitamin D; MET, metabolic equivalent.

Table 2.

General Characteristic Stratified According to Total 25(OH)D Concentration

Characteristic	Total 25(OH)D concentrations (nmol/L)
Characteristic	<50 (n = 3,209)	50 ∼ 74.99 (n = 3,652)	≥75 (n = 2,798)	P
Sex, n (%) of participant				<.0001
Men	1446 (44.9)	1985 (55.3)	1284 (43.6)
Women	1763 (55.1)	1667 (44.7)	1514 (56.4)
Race/ethnicity, n (%) of participant				<.0001
Non-Hispanic white	770 (41.3)	1811 (71.5)	2124 (89.0)
Non-Hispanic black	1113 (27.3)	442 (6.8)	157 (2.4)
Mexican American	760 (14.7)	773 (9.8)	244 (3.2)
Other ethnicity	566 (16.7)	626 (11.9)	273 (5.4)
Smoking status, n (%) of participant				<.0001
Never	1816 (58.1)	1935 (54.0)	1422 (52.3)
Former	606 (17.2)	967 (25.5)	869 (29.9)
Current	787 (24.8)	750 (20.4)	507 (17.8)
Season of examination, n (%) of participant				<.0001
November to April	1858 (54.3)	1611 (37.8)	862 (27.5)
May to October	1351 (45.7)	2041 (62.2)	1936 (72.5)
Education (%), n (%) of participant				<.0001
<9th grade	457 (8.5)	570 (7.7)	230 (3.9)
≤High school graduate	1374 (39.9)	1436 (36.4)	1052 (34.0)
>High school graduate	1378 (51.6)	1646 (55.9)	1516 (62.1)
BMI (%), n (%) of participant				<.0001
18.5–24.99 kg/m²	698 (22.8)	927 (27.2)	963 (37.0)
25.0–29.99 kg/m²	982 (28.9)	1325 (35.8)	1070 (37.7)
≥30 kg/m²	1529 (48.3)	1400 (36.9)	765 (25.3)
Household income [n (%)]				<.0001
PIR <1	1028 (26.1)	1006 (18.6)	624 (15.3)
PIR 1 to <3	1344 (38.1)	1465 (34.1)	995 (29.2)
PIR ≥3	837 (35.8)	1181 (47.2)	1179 (55.4)
Supplement use, n (%) of participant (n = 9,655)				<.0001
Yes	979 (31.7)	1732 (47.7)	1876 (66.8)
No	2228 (35.3)	1918 (52.3)	922 (33.2)
Risk factors for diabetes, n (%) of participant (n = 8,458)				.018
Yes	378 (14.0)	374 (11.6)	259 (10.4)
No	2363 (86.0)	2829 (88.4)	2255 (89.6)

Survey (Rao-Scott) χ ² test for categorical variable were used.

BMI, body mass index; PIR, poverty-to-income ratio.

Supplementary Tables S1 and S2 summarize the participants' characteristics according to fiber intake. Participants with high-fiber intake were more likely to be older, shorter, have a lower body weight, have high total vitamin D, have lower physical activity, have lower intake of total energy, and have higher energy-adjusted values for carbohydrate, protein, fiber, vitamin D, and calcium intake. In addition, significant differences according to fiber intake were observed according to sex, race/ethnicity, smoking status, education level, and BMI category.

Total 25(OH)D and T2D

Table 3 shows the ORs for T2D according to total vitamin D status in the various models. Total vitamin D was significantly associated with the ORs of T2D in all models. In the fully adjusted model (model C), and relative to participants with deficient total vitamin D, the ORs for T2D^a were 0.90 (95% CI: 0.72, 1.13) for participants with suboptimal total vitamin D and 0.72 (95% CI: 0.58, 0.90) for participants with sufficient total vitamin D. A significant dose-dependent association was observed between total vitamin D and T2D^a (P for trend = .005). Furthermore, the ORs for T2D^b were 0.76 (95% CI: 0.60, 0.97) for participants with suboptimal total vitamin D and 0.60 (95% CI: 0.50, 0.80) for participants with sufficient total vitamin D. A significant dose-dependent association was also found between total vitamin D and T2D^b (P for trend <.001).

Total vitamin D, fiber intake, and T2D^b

Table 4 presents the ORs for T2D according to total vitamin D and fiber intake in the various models. Among the group with low-fiber intake, the ORs for T2D^b were 0.89 (95% CI: 0.65, 1.21) for participants with suboptimal total vitamin D and 0.70 (95% CI: 0.48, 1.01) for participants with sufficient vitamin D. The dose-dependent association showed a marginal trend in the fully adjusted model (P for trend = .056). Similarly, among the group with high-fiber intake, the ORs for T2D^b were 0.60 (95% CI: 0.40, 0.90) for participants with suboptimal vitamin D and 0.49 (95% CI: 0.31, 0.77) for participants with sufficient total 25(OH)D. These dose-dependent association was significant in the fully adjusted model (P for trend = .004). However, there was no significance in the result from the interaction analysis (P = .197).

Table 4.

Adjusted Odds Ratio and 95% Confidence Interval for Type 2 Diabetes Diagnosed Based on Survey and Laboratory Data with Stratification According to Fiber Intake

Variables	No. of T2D/no. of participants	Model A	Model B	Model C
Low-fiber intake (n = 2,574)
Total 25(OH)D (nmol/L)
<50 (deficient)	308/936	1 (reference)	1 (reference)	1 (reference)
50 ∼ 74.99 (suboptimal)	293/965	0.77 (0.57, 1.06)	0.91 (0.67, 1.22)	0.89 (0.65, 1.21)
≥75 (sufficient)	161/673	0.51 (0.36, 0.71)^***	0.70 (0.49, 0.99)^*	0.70 (0.48, 1.01)^†
P-trend		<.001	.041	.056
High-fiber intake (n = 2,051)
Total 25(OH)D (nmol/L)
<50 (deficient)	251/617	1 (reference)	1 (reference)	1 (reference)
50 ∼ 74.99 (suboptimal)	266/829	0.53 (0.36, 0.79)^**	0.59 (0.39, 0.88)^*	0.60 (0.40, 0.90)^*
≥75 (sufficient)	155/605	0.37 (0.24, 0.58)^***	0.47 (0.30, 0.75)^**	0.49 (0.31, 0.77)^**
P-trend		<.001	.003	.004
Interaction term		0.276	0.168	0.197

***

P < .001, ^** P < .01, ^* P < .05, ^† P < .1 compared with reference group (vitamin levels <50 nmol/L).

Diabetes was diagnosed based on survey and lab data.

Model A was adjusted for age, sex, race/ethnicity, education level, smoking status, season of examination, physical activity.

Model B: model A+ further adjusted for BMI.

Model C: model B+ further adjusted for intake of total energy and energy adjusted dietary factor (protein, calcium, vitamin D).

Total 25(OH)D and HOMA-IR

Supplementary Table S3 shows the results for the associations between total vitamin D and the HOMA-IR findings in the various models. In the Model C, relative to participants with deficient total vitamin D, the HOMA-IR values were significantly lower for participants with suboptimal total vitamin D (−0.59, 95% CI: −0.95, −0.23) and participants with sufficient total vitamin D (−1.03, 95% CI: −1.36, −0.66). These dose-dependent associations were statistically significant (P for trend <.001). When we controlled for additional confounders (household income, supplement use, and family history), these associations became less significant but remained generally consistent with the unadjusted results.

Discussion

The present study evaluated a representative sample of American adults and the findings suggest that vitamin D concentrations (based on total 25(OH)D as a surrogate biomarker) and fiber intake were related to the prevalence of T2D, even after adjusting for potential confounding factors. Thus, strategies that involve vitamin D supplementation and high-fiber intake may more effectively reduce the prevalence of diabetes. In the fully adjusted models, the ORs for diabetes in participants with sufficient total 25(OH)D were 0.49 (95% CI: 0.31, 0.77) for participants with high-fiber intake and 0.70 (95% CI: 0.48, 1.01) for participants with low-fiber intake, although the interaction analysis results were not significant.

An in vitro study has indicated that of vitamin D is involved in promoting the biosynthesis of beta cells¹¹ and increased insulin secretion in the pancreas.¹² A recent review also noted that epidemiology studies support an association between vitamin D deficiency and the risk of T2D, whereas most randomized intervention studies could not prove the effect of vitamin D supplementation in terms of glucose metabolism among subjects who were at risk of developing diabetes.¹³ These discrepancies may be related to differences in the duration and dose of vitamin D supplementation, as well as the subjects' characteristics. Moreover, a recent in vivo study indicated that vitamin D supplementation reduced blood glucose concentrations, increased insulin secretion, and helped prevent diabetes-related complications.²⁶ Park et al. suggested that the sufficient vitamin D (>50 nmol/L) was necessary to observe the maximum reduction in the risk of T2D in a prospective study.²⁷ Similarly, the present study revealed that the prevalence of diabetes was significantly lower when the total vitamin D concentrations were >50 nmol/L.

Fiber intake is a well-known dietary factor that may improve the risk of T2D⁵ and insulin resistance.⁶ The relationship of fiber intake and T2D has been explained by several mechanisms. First, high-fiber intake may improve the risk of overweight/obesity, and indirectly influence the risk of developing T2D.²⁸ Second, fiber intake may postpone gastric emptying and reduce the absorption of carbohydrates, as a result, the concentrations of postprandial blood glucose decrease.²⁹ One meta-analysis has suggested that high-fiber intake (≥25 g/day) may prevent T2D.⁷

Insulin resistance is a pathological status in which cells or a whole organism is unable to respond normally to a given insulin, and this condition is associated with obesity and T2D.³⁰ Previous studies demonstrated that sufficient vitamin D status reduces the risk of insulin resistance.^14,15 The present study included an insulin resistance marker (HOMA-IR) to evaluate how vitamin D might reduce the prevalence of T2D. Similar to previously reported results, the present study revealed that insulin resistance was decreased at sufficient vitamin D concentrations. Furthermore, the present study revealed that the OR value for T2D was lower in the group with high-fiber intake (≥15.3 g/day) and sufficient vitamin D status (≥50 nmol/L), relative to the group with low-fiber intake (<15.3 g/day) and deficient vitamin D status (<50 nmol/L). Nevertheless, the results of the interaction analysis were not significant. While our observational findings suggest that these two factors have a synergistic beneficial effect on T2D, further experimental studies are required to understand these related mechanisms.

When we controlled for additional confounders (household income, supplement use, and family history) in model D, the association between vitamin D concentration and T2D^b became less significant. Nevertheless, the results remained generally consistent with the results from the unadjusted models (Supplementary Table S4). Moreover, analysis according to fiber intake revealed consistent results relative to the unadjusted results.

The major strengths of this study include a large dataset from a representative sample of American adults. Second, we adjusted the models for a variety of potential confounding factors, including dietary factors, and our major results persisted after the adjustment for those factors. Third, this study used serum 25(OH)D concentration, which is a biomarker that reflects vitamin D derived from sun exposure and obtained from food intake. Nevertheless, this study also has several limitations. First, the cross-sectional design precludes a definitive conclusion regarding the causality of the relationships we observed. Second, residual confounding is possible despite the fact that we controlled for a variety of potential confounders. Third, we were unable to classify fiber as water soluble or insoluble based on the NHANES data. Finally, although total 25(OH)D is a commonly used biomarker for estimating vitamin D status, it may not capture the body's stores of bioavailable vitamin D.

In conclusion, the present study's findings suggest that vitamin D intake (food rich in vitamin D, supplementation, and outdoor activities in sunny days) combined with high-fiber intake may more effectively reduce the prevalence of diabetes. Further studies are needed to explain the potential mechanisms underlying the beneficial effects of dietary fiber intake and vitamin D status on the prevention of T2D.

Ethical Standards Disclosure

This study was conducted in accordance with the Declaration of Helsinki. All procedures in the NHANES protocol were reviewed and approved by the National Center for Health Statistics' Research Ethics Review Board and all participants provided informed consent.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

Funding Information

This work was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF), which is funded by the Korea Ministry of Education (2017R1A6A3A11034504).

Supplementary Material

Supplementary Table S1

Supplementary Table S2

Supplementary Table S3

Supplementary Table S4

References

Menke

, Casagrande

, Geiss

, Cowie

: Prevalence of and trends in diabetes among adults in the United States, 1988–2012. JAMA, 2015; 314:1021–1029.

Centers for Disease Control and Prevention: National Diabetes Fact Ssheet: National Estimates and General Information on Diabetes and Prediabetes in the United States, 2011. US Department of Health and Human Services, Centers for Disease Control and Prevention, Atlanta, GA, 2011, 201, pp. 2568–2569.

Dunlay

, Givertz

, Aguilar

, et al.: Type 2 diabetes mellitus and heart failure: A scientific statement from the American Heart Association and the Heart Failure Society of America: This statement does not represent an update of the 2017 ACC/AHA/HFSA heart failure guideline update. Circulation, 2019; 140:e294–e324.

Xuan

, Gao

, Anusruti

, et al.: Association of serum markers of oxidative stress with incident major cardiovascular events, cancer incidence and all-cause mortality in type 2 diabetes patients: Pooled results from two cohort studies. Diabetes Care, 2019; 42:1436–1445.

Silva

, Kramer

, de Almeida

, Steemburgo

, Gross

, Azevedo

: Fiber intake and glycemic control in patients with type 2 diabetes mellitus: A systematic review with meta-analysis of randomized controlled trials. Nutr Rev, 2013; 71:790–801.

Tucker

: Fiber intake and insulin resistance in 6374 adults: The role of abdominal obesity. Nutrients, 2018; 10:E237.

Yao

, Fang

, Xu

, et al.: Dietary fiber intake and risk of type 2 diabetes: A dose-response analysis of prospective studies. Eur J Epidemiol, 2014; 29:79–88.

Casagrande

, Cowie

: Trends in dietary intake among adults with type 2 diabetes: NHANES 1988–2012. J Hum Nutr Diet, 2017; 30:479–489.

Herrick

, Storandt

, Afful

, et al.: Vitamin D status in the United States, 2011–2014. Am J Clin Nutr, 2019; 110:150–157.

10.

Maddaloni

, Cavallari

, Napoli

, Conte

: Vitamin D and diabetes mellitus. Front Horm Res, 2018; 50:161–176.

11.

Bourlon

, Billaudel

, Faure-Dussert

: Influence of vitamin D3 deficiency and 1,25 dihydroxyvitamin D3 on de novo insulin biosynthesis in the islets of the rat endocrine pancreas. J Endocrinol, 1999; 160:87–95.

12.

Kjalarsdottir

, Tersey

, Vishwanath

, et al.: 1,25-Dihydroxyvitamin D3 enhances glucose-stimulated insulin secretion in mouse and human islets: A role for transcriptional regulation of voltage-gated calcium channels by the vitamin D receptor. J Steroid Biochem Mol Biol, 2019; 185:17–26.

13.

Munoz-Garach

, Garcia-Fontana

, Munoz-Torres

: Vitamin D status, calcium intake and risk of developing type 2 diabetes: An unresolved issue. Nutrients, 2019; 11:E642.

14.

Chung

, Lee

, Hong

, et al.: Inverse relationship between vitamin D status and insulin resistance and the risk of impaired fasting glucose in Korean children and adolescents: The Korean National Health and Nutrition Examination Survey (KNHANES) 2009–2010. Public Health Nutr, 2014; 17:795–802.

15.

Kabadi

, Lee

, Liu

: Joint effects of obesity and vitamin D insufficiency on insulin resistance and type 2 diabetes: Results from the NHANES 2001–2006. Diabetes Care, 2012; 35:2048–2054.

16.

Dong

, Chen

, Gutin

, Zhu

: Total, insoluble, and soluble dietary fiber intake and insulin resistance and blood pressure in adolescents. Eur J Clin Nutr, 2019; 73:1172–1178.

17.

Morimoto

, Kasuga

, Tanaka

, et al.: Association between dietary fibre:carbohydrate intake ratio and insulin resistance in Japanese adults without type 2 diabetes. Br J Nutr, 2018; 119:620–628.

18.

Centers for Disease Control and Prevention: National Health and Nutrition Examination Survey 2007–. Data Documentation, Codebooks SAS Code. http://wwwn.cdc.gov/Nchs/Nhanes/Search/nhanes07_08.aspx (accessed November 15, 2017 ).

19.

Centers for Disease Control and Prevention: National Health and Nutrition Examination Survey 2009–. Data Documentation, Codebooks SAS Code. http://wwwn.cdc.gov/Nchs/Nhanes/Search/nhanes09_10.aspx (accessed November 15, 2017 ).

20.

Centers for Disease Control and Prevention: National Health and Nutrition Examination Survey 2007–. Laboratory procedure manual: 25-hydroxivitamin D, C3-epi-25-hydroxyvitamin D, and 25-hydroxyvitamin D2 in serum by ultra high performance liquid chromatography-tandem mass spectrometry. www.cdc.gov/nchs/data/nhanes/nhanes_07_08/VID_E_met_Vitamin_D.pdf (accessed January 20, 2018 ).

21.

Centers for Disease Control and Prevention: National Health and Nutrition Examination Survey 2009–.Laboratory procedure manual: 25-hydroxivitamin D, C3-epi-25-hydroxyvitamin D, and 25-hydroxyvitamin D2 in serum by ultra high performance liquid chromatography-tandem mass spectrometry. www.cdc.gov/nchs/data/nhanes/nhanes_09_10/VID_F_met_Vitamin_D.pdf (accessed January 20, 2018 ).

22.

Singh

, Taylor

, Reddy

, Grebe

: C-3 epimers can account for a significant proportion of total circulating 25-hydroxyvitamin D in infants, complicating accurate measurement and interpretation of vitamin D status. J Clin Endocrinol Metab, 2006; 91:3055–3061.

23.

Holick

, Binkley

, Bischoff-Ferrari

, et al.: Evaluation, treatment, and prevention of vitamin D deficiency: An Endocrine Society clinical practice guideline. J Clin Endocrinol Metab, 2011; 96:1911–1930.

24.

, Stampfer

, Rimm

, et al.: Dietary fat and coronary heart disease: A comparison of approaches for adjusting for total energy intake and modeling repeated dietary measurements. Am J Epidemiol, 1999; 149:531–540.

25.

Armstrong

, Bull

: Development of the World Health Organization Global Physical Activity Questionnaire (GPAQ). J Public Health, 2006; 14:66–70.

26.

Derakhshanian

, Djazayery

, Javanbakht

, et al.: Vitamin D downregulates key genes of diabetes complications in cardiomyocyte. J Cell Physiol, 2019; 234:21352–21358.

27.

Park

, Garland

, Gorham

, BuDoff

, Barrett-Connor

: Plasma 25-hydroxyvitamin D concentration and risk of type 2 diabetes and pre-diabetes: 12-year cohort study. PLoS One, 2018; 13:e0193070.

28.

Thompson

, Hannon

, An

, Holscher

: Effects of isolated soluble fiber supplementation on body weight, glycemia, and insulinemia in adults with overweight and obesity: A systematic review and meta-analysis of randomized controlled trials. Am J Clin Nutr, 2017; 106:1514–1528.

29.

de Carvalho

, de Paula

, Viana

, Machado

, de Almeida

, Azevedo

: Plasma glucose and insulin responses after consumption of breakfasts with different sources of soluble fiber in type 2 diabetes patients: A randomized crossover clinical trial. Am J Clin Nutr, 2017; 106:1238–1245.

30.

Shanik

, Xu

, Skrha

, Dankner

, Zick

, Roth

: Insulin resistance and hyperinsulinemia: Is hyperinsulinemia the cart or the horse?. Diabetes Care, 2008; 31:S262–S268.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.03 MB

0.04 MB

Vitamin D Status,Fiber Intake,and Type 2 Diabetes in U.S. Adults

Abstract

Introduction

Methods

Study population

Measurement of 25(OH)D concentrations

Fiber intake

Identification and classification of T2D

Covariates

Statistical analyses

Results

Descriptive characteristics

Total 25(OH)D and T2D

Total vitamin D, fiber intake, and T2D b

Total 25(OH)D and HOMA-IR

Discussion

Ethical Standards Disclosure

Footnotes

Author Disclosure Statement

Funding Information

Supplementary Material

References

Supplementary Material

Total vitamin D, fiber intake, and T2D^b