Prevalence of Hypovitaminosis D and Its Association with Comorbidities of Childhood Obesity
Perm J 2014 Fall; 18(4):32-39 [Full Citation]
Purpose: Our study sought to further delineate the prevalence of hypovitaminosis D and its relationship to comorbidities of childhood obesity.
Methods: We conducted a retrospective chart review from 155 obese children aged 5 to 19 years who attended the Penn State Children's Hospital Pediatric Multidisciplinary Weight Loss Program from November 2009 through November 2010. We determined the incidence of hypovitaminosis D and examined its association with comorbidities including elevated blood pressure, diabetes, acanthosis nigricans, depression, hyperlipidemia, hyperinsulinemia, and abnormal liver function test results, as well as age, sex, and geographic location.
Results: Under the latest Institute of Medicine definitions, vitamin D deficiency (< 20 ng/mL) and insufficiency (20-29 ng/mL) was present in 40% and 38% of children, respectively. The prevalence of vitamin D deficiency was 27.8% in children aged 5 to 9 years, 35.4% in children aged 10 to 14 years, and 50.9% of children aged 15 years or older. Older age, African-American race, winter/spring season, higher insulin level, total number of comorbidities, and polycystic ovary syndrome (in girls) were significantly associated with vitamin D deficiency. African-American race, winter/spring season, hyperinsulinemia, elevated systolic blood pressure, urban location, and total numbers of comorbidities were significantly associated with hypovitaminosis D (< 30 ng/mL).
Conclusions: Hypovitaminosis D is associated with several medical comorbidities in obese children. Given the large percentage of children, even in our youngest age group, who are vitamin D deficient, obese children should be considered for routine vitamin D screening.
Obesity among children and adolescents has continued to rise in epidemic proportions since the late 1970s. The prevalence of obesity among children and adolescents in the US has tripled between 1976 and 2008,1 with a 2012 prevalence of 19.7% (females) and 17.2% (males) among those aged 6 to 12 years, and 20.4% (females) and 21.4% (males) among those aged 12 to 19 years.2 Obesity rates in general have plateaued since 2008, but rates for obesity Class 2 (120% of the 95th percentile or a body mass index [BMI] > 35 kg/m2) and Class 3 (140% of 95th percentile or BMI > 40 kg/m2) continue to increase.2 The obesity prevalence among children is particularly alarming because obesity-related diseases, rarely seen in children in the past, are increasingly being diagnosed in pediatric patients; these obesity-associated conditions include obstructive sleep apnea, nonalcoholic fatty liver disease with resultant cirrhosis, and type 2 diabetes.3 In addition, obesity is related to hypovitaminosis D (vitamin D deficiency and insufficiency).4
Vitamin D is a fat-soluble vitamin and is naturally present in only a small number of foods. Vitamin D can be added to food, and it can be ingested as a dietary supplement. Vitamin D is produced endogenously when ultraviolet rays from sunlight strike the skin and trigger vitamin D synthesis. Groups of people at risk of vitamin D inadequacy include breastfed infants and older adults, as well as people with limited sun exposure, dark skin, or fat malabsorption, and those who are obese.5
The serum concentration of 25-hydroxy-vitamin D (25[OH]D) is the best indicator of vitamin D status. The 25(OH)D that is produced cutaneously or obtained from food and supplements has a fairly long circulating half-life of 15 days.6 Vitamin D insufficiency is defined as a 25(OH)D level of 21 to 29 ng/mL and vitamin D deficiency as a 25(OH)D level below 20 ng/mL, according to an Institute of Medicine 2011 report.7
Since the mid-1990s, mean serum 25(OH)D concentrations in the US have slightly declined among males but not females.8 This decline is probably caused by reduced milk intake, inadequate sun exposure, greater use of sun protection when outdoors, and simultaneous increases in body weight.8 Obesity does not affect the skin's capacity to synthesize vitamin D. Rather, greater amounts of subcutaneous fat sequester more of the vitamin, making it less bioavailable to the body.9
Hypovitaminosis D has been considered a risk factor for glucose intolerance and decreased insulin sensitivity.10 Serum 25(OH)D levels below 20 ng/mL have been associated with decreased pancreatic b-cell function.11 It has also been found that insulin sensitivity is as much as 60% higher in individuals with serum 25(OH)D levels of 30 ng/mL vs those with levels of 10 ng/mL.11
Along with glucose intolerance, vitamin D deficiency and insufficiency are associated with obesity-related health diseases.12 For instance, low vitamin D status may increase the risk of diabetes mellitus, hypertension, cardiovascular disease,13 and certain types of cancer14,15 and has been associated with greater severity of critical illness.16 Vitamin D also plays a role in the liver, skeletal muscles,10 and the immune system.10,17,18 In the present study, we report the prevalence of vitamin D insufficiency and deficiency in obese pediatric patients at Penn State Hershey Children's Hospital in Hershey, PA, and the association of hypovitaminosis D with comorbidities of childhood obesity.
A retrospective chart review was conducted of 155 obese children and teenagers aged 5 to 19 years who attended the Penn State Hershey Children's Hospital Pediatric Multidisciplinary Weight Loss Program from November 2009 through November 2010. Five children with missing or invalid birth dates were excluded from the study, leaving 150 children for analysis. We did not have a control group for comparison. This multidisciplinary program has physicians, nurses, and a dietitian who see children and adolescents referred by their primary care provider who are obese as defined by BMI. In children, obesity is defined as having a BMI at or above the 95th percentile for age and sex. Most of our study population was at or above the 99th percentile.
Each patient underwent a complete history and physical examination, and data from the medical history were recorded. Screening laboratory tests were ordered on the basis of a full medical evaluation, which included a 25(OH)D level. Parents and/or patients provided the medical history by form and systematic questioning by the physician. The history included presence of asthma, attention-deficit/hyperactivity disorder, constipation, depression, eating disorders, gastroesophageal reflux disease, hypertension, polycystic ovary syndrome [PCOS], and snoring. Pulse and blood pressure (BP) were measured in the office and were deemed high if above the 95th percentile for age and sex. Laboratory values were obtained that included fasting lipid profile, insulin, glucose, aspartate aminotransferase, alanine aminotransferase, hemoglobin A1c (HbA1c), and 25(OH)D; reference values for age were used to determine if the results were abnormal. Rural-urban commuting area code data were used to categorize the patient's residence as either urban or rural on the basis of zip codes.19 These data were analyzed to determine the prevalence of vitamin D insufficiency/deficiency and to assess for its relationship to comorbidities. In addition to those from the history, comorbidities included acanthosis nigricans, hyperlipidemia, and abnormal liver function tests. Also assessed was the relationship between vitamin D insufficiency/deficiency and the total number of comorbidities, insulin level, sex, race, rural vs urban homestead, and season of the year. None of the patients were taking more than 400 IU/day of vitamin D (the recommended daily intake of vitamin D per the American Academy of Pediatrics20). The study was conducted with the permission of the Penn State Hershey institutional review board.
All analyses were carried out using SAS statistical software Version 9.3 (SAS Institute, Cary, NC). Descriptive statistics were generated for all variables using means, medians, and standard deviations for continuous variables and frequency tables for categorical variables. The season was defined by the laboratory date using the meteorologic definitions of the seasons by months. The outcome variables were defined in binary form for vitamin D insufficiency (< 30 ng/mL) and vitamin D deficiency (< 20 ng/mL). A bivariate analysis was performed to assess the association of these outcome variables with the various independent variables of interest using a logistic regression. This same analysis was stratified by categories: age, sex, race, and season to determine if any of the associations seen overall were manifested within 1 group more than in another. Because of a small number of children from a rural location, we were unable to stratify by location.
On the basis of results of the overall bivariate analysis, a subset of clinically and statistically significant independent variables was chosen to be considered in a multivariate logistic regression for each outcome. The starting subset of variables to be considered included age, sex, race, location, season, systolic BP, insulin level, hyperlipidemia, and total comorbidities as a surrogate for the collection of individual comorbidities. Before doing any modeling, this subset of independent variables was tested for multicollinearity using variance inflation factor. From this subset, a final reduced set of significant independent variables was chosen for each outcome using a backward process of elimination with an inclusion criterion of p = 0.10. Two-way interactions were tested for significance between all independent variables remaining as significant in the model after the backward elimination process, but none were found to be significant to the model for either outcome. The fit of the final logistic regression model was assessed using the Hosmer-Lemeshow goodness-of-fit test. Odds ratios were used to quantify the direction and magnitude of the association with the outcome of the independent variables remaining in the final model.
The characteristics of our study sample are shown in Table 1. Our population was 65% female, 91% urban, 61% white, 19% African American, and 8% Hispanic. Their most common medical comorbidities included acanthosis nigricans (59%), hyperinsulinemia (42%), elevated systolic BP at first visit (35%), PCOS (20%), asthma (16%), attention-deficit/hyperactivity disorder (11%), gastroesophageal reflux disease (11%), constipation (7%), autism (7%), and binge eating disorder (6%).
Vitamin D levels ranged from 5 to 60 ng/mL with a mean of 23 ng/mL. The prevalence of vitamin D deficiency (< 20 ng/mL) was 40% and insufficiency (20-30 ng/mL) was 38%; only 22% had a normal vitamin D level above 30 ng/mL (Figure 1). Stratifying these results by age group, 27.8% aged 5 to 9 years, 35.4% aged 10 to 14 years, and 50.9% aged 15 and older were vitamin D deficient.
In bivariate analyses, older age, African-American race, winter/spring season (Figures 2 and 3), higher insulin level, hyperlipidemia (elevated total cholesterol, low-density lipoprotein, or triglycerides levels), and PCOS (female only) were significantly associated with vitamin D deficiency (Table 1). Some of these associations were also found with overall hypovitaminosis D, including African-American race, winter/spring season, and higher insulin level. Higher systolic BP was significantly associated with hypovitaminosis D but not with vitamin D deficiency. Urban location was significantly associated with hypovitaminosis D and trended toward significance for vitamin D deficiency despite having a limited number of children from rural locations. In an unadjusted comparison, the total number of comorbidities was not associated with a low vitamin D level. However, when controlling for all other variables in our model, there was a significant association with vitamin D deficiency, and leaning toward a significant association with hypovitaminosis D. Sex; heart rate; diastolic BP; aspartate aminotransferase, alanine aminotransferase, and HbA1c levels; attention-deficit/hyperactivity disorder; asthma; depression; gastroesophageal reflux disease; and snoring were not associated with vitamin D deficiency or hypovitaminosis D.
In the adjusted analysis, older age, nonwhite race, rural location, winter/spring season, insulin, and total comorbidities remained significantly associated with vitamin D deficiency (Table 2). These same associations with hypovitaminosis D were sustained in a multivariate model together except for older age. Elevated systolic BP was also associated with vitamin D insufficiency.
When the analysis was stratified by age category, sex, race, or season, several significant associations were stronger in 1 group than in another. For children age 10 to 14 years, girls were more likely to be vitamin D deficient than boys (43.4% vs 19.2%, p = 0.040). For children age 15 years and older, a higher insulin level was significantly associated with being vitamin D deficient (p = 0.007) or with having hypovitaminosis D (p = 0.026). A stronger association was seen in female subjects between urban location and vitamin D deficiency (p = 0.009) or hypovitaminosis D (p = 0.017), as well as between winter/spring season and vitamin D deficiency (p = 0.029) or hypovitaminosis D (p = 0.013). Male subjects tended to have a greater association between a higher insulin level and vitamin D deficiency (p = 0.004) or hypovitaminosis D (p = 0.043). In whites, female subjects were more likely to be vitamin D deficient than male subjects (31.7% vs 10.7%, p = 0.043), and winter/spring season had a more significant association with both vitamin D deficiency (p = 0.038) and hypovitaminosis D (p = 0.028). There was a more pronounced association between higher systolic BP and vitamin D deficiency (p = 0.038) or hypovitaminosis D (p = 0.021) in the summer/fall seasons.
The etiology of hypovitaminosis D is likely multifactorial because it has been associated with several dietary factors21 as well as decreased sunlight exposure and poor vitamin D intake.22 Our Pennsylvania prevalence of 77.8% (insufficient and deficient) is less than that of a study in Texas that showed a prevalence of 92% of vitamin D insufficiency and deficiency in 6- to 16-year-old obese patients,21 and more than 90% in 17-year-old obese teens in Rhode Island.9 Our study did not look into milk intake or specific dietary sources of vitamin D consumption or sunlight exposure. We found that nonwhite race, elevated systolic BP, hyperinsulinemia, multiple comorbidities, urban location, and winter/spring collection of blood samples were associated with hypovitaminosis D.
Our findings suggest that older children and teenagers are most likely to be vitamin D deficient. Across age groups, urban patients are more likely than rural patients to be vitamin D deficient. In multivariate analysis, this was still significant for overall hypovitaminosis D. Diet and sunlight exposure probably played a role, but these data were not collected.
After multivariate analysis, we found a significant association with total number of comorbidities and vitamin D deficiency, and the association with hypovitaminosis D almost reached statistical significance. This may be an overall reflection of general health status. Vitamin D is part of many physiologic pathways, and its role in many illnesses is still not completely understood. These results may imply its cumulative involvement in obesity-related problems. The exact reason for this is unknown, and further study in this area may be warranted.
A few sex-based differences were noted. Female subjects showed a seasonal variation of being more likely to have low vitamin D in winter/spring. Male subjects showed no seasonal variation. Male subjects showed higher mean systolic BP and insulin levels as vitamin D levels dropped. This was not seen in female subjects. Obese 10- to 14-year-old girls are at higher risk of vitamin D deficiency compared with boys the same age. The exact relationship here is unknown. It may be related to more outdoor activity for the boys, but this was not directly measured.
Race played an important factor. African Americans had the lowest vitamin D levels, with 100% being insufficient and 79.3% being deficient. Whites had much lower prevalence. For those who were vitamin D sufficient, 85% were white. This is consistent with findings of previous studies looking at race.10,23 African Americans have skin pigmentation that reduces vitamin D production, and from about puberty onward, vitamin D intake is also generally below recommended levels.23 It has been recently shown that African Americans can have lower vitamin D and vitamin D binding protein levels, which may actually lead to similar levels of bioavailable vitamin D compared with whites.24 Thus, our current "normal" values for vitamin D may need to be adjusted for this population. Further study in this area is needed.
We found a strong link with seasonality in the winter and spring collections. There was a 3.7-fold greater likelihood of being vitamin D deficient and a 4.2-fold greater likelihood of having hypovitaminosis D in these seasons compared with summer and fall collections. This is probably secondary to decreasing sun exposure in late fall into winter and poor bioavailable Vitamin D stores. Vitamin D is fat soluble and is readily stored in adipose tissue, and may be sequestered in a larger body pool of fat of obese individuals.4 People obtain most of their vitamin D requirement from casual exposure to sunlight. It has been shown that there is more than 50% decreased bioavailability of cutaneous synthesized vitamin D3 in obese adult subjects, and this likely accounts for the consistent observation that obesity is associated with vitamin D deficiency.4 Insulin levels have been shown to have seasonal variation with the highest levels in spring compared with fall.25 This may explain why all of the vitamin D sufficient subjects had normal insulin levels in the winter/spring collection. Even with this seasonal variation, 69.6% had hypovitaminosis D and 30.4% were deficient in summer/fall collections. This may lead the practitioner to consider the season with borderline vitamin D results before supplementation recommendations are given. For example, if it is fall and the number is borderline, it will likely decrease over the winter/spring months.
Low vitamin D levels have been associated with risk factors for Type 2 diabetes mellitus (specifically insulin resistance but not HbA1c) in obese children.21 We also found this to be true in our population, as shown by its associations with overall BMI, hyperinsulinemia, and PCOS. Hyperinsulinemia was particularly predictive in the age group 15 years and older, in which 80% were vitamin D deficient (odds ratio = 10, p < 0.001). Insulin resistance is also part of PCOS, and we found associations with both. In another study, 72.8% of women with PCOS were vitamin D insufficient.26
Serum triglyceride levels and systolic BPs have been linked to vitamin D deficiency.27 Our study overall showed that elevated systolic BP was predictive for hypovitaminosis D, but not deficiency. For male subjects, however, it was predictive of both, with higher systolic BPs in the deficiency group. We did not find an association with fasting triglycerides or cholesterol levels, but we did find an association with hyperlipidemia in general for any elevation in total cholesterol, LDL, or triglycerides levels.
It is difficult to precisely predict the exact amount of sunlight required to produce enough vitamin D for each individual, and there are some health concerns relating excessive sun exposure to skin cancer risks in adulthood.10 Therefore, it is important to encourage protected sun exposure for children and teenagers. However, this practice may translate into insufficient endogenous vitamin D production. It is then necessary to encourage dietary supplementation of vitamin D. This is needed to normalize vitamin D levels and may temper insulin resistance, thus possibly delaying the onset of Type 2 diabetes mellitus (although this has not been demonstrated). Because most of our patients had hypovitaminosis D and vitamin D supplementation is relatively safe, the argument could be made to supplement all obese children with vitamin D and forego lab testing. This, however, would not pick up the vitamin D deficient children and would not adequately meet their replacement needs. In a study by Heaney et al28 of response to oral dosing of vitamin D, it can be extrapolated from data in men older than age 20 years that to raise the serum vitamin D concentration by 1 ng/mL would require roughly 140 IU of cholecalciferol daily. There may be different responses to supplementation based on race, with whites normalizing quicker than Hispanics or African Americans.9
Limitations of the present study are small sample sizes for age group comparisons and from rural homes. Data on dietary vitamin D (especially milk intake), sun exposure, and sunscreen use were not obtained as this was a retrospective study and these data were not routinely collected. In addition, data were not collected on parameters that could affect insulin levels, such as amount of physical activity, medication use, hours of sleep, or a family history of diabetes and PCOS. Our subjects were obese children from central Pennsylvania who were referred for weight management, which may not be representative of all obese children. Despite these limitations, these results show that there are multiple associations for hypovitaminosis D, and these findings are likely applicable to all obese children in the US. Our study looked only at associations and did not show causality.
Hypovitaminosis D has many extraskeletal associations, including cancer, cardiovascular disease, diabetes, and autoimmune disorders. Previous evidence has been inconsistent, inconclusive as to causality, and insufficient to inform nutritional requirements.7
In this study, African-American race, winter/spring season, urban location, and higher insulin level were significantly associated with vitamin D deficiency and overall hypovitaminosis D. These, along with elevated systolic BP, urban location, total number of comorbidities, and PCOS (in female subjects only) were also associated with hypovitaminosis D.
The American Academy of Pediatrics Expert Committee, regarding the assessment, prevention, and treatment of child and adolescent overweight and obesity, does not currently recommend assessing vitamin D status in obese children.29 Given the large percentage of children, even in our youngest age group who are vitamin D deficient, routine screening of vitamin D should be considered in obese children and supplementation when needed. Supplementation has been found helpful in patients with systemic lupus erythematosus and hypovitaminosis D, as it improved inflammatory markers and disease activity.17 Vitamin D screening is more likely to have abnormal results with any of the following: older age, nonwhite race, elevated systolic BP, hyperinsulinemia, multiple comorbidities, PCOS, urban location, or winter/spring collection of blood samples. Seasonal variations of vitamin D levels should also be considered when recommending supplementation. Further study is needed to see if vitamin D supplementation will have an impact on preventing further weight gain or preventing possible comorbidities such as hyperinsulinemia, diabetes, and PCOS.
The author has no conflicts of interest to disclose.
Kathleen Louden, ELS, of Louden Health Communications provided editorial assistance.
1. Ogden C, Carroll M. Prevalence of obesity among children and adolescents: United States, trends 1963-1965 through 2007-2008 [Internet]. Atlanta, GA: Centers for Disease Control and Prevention; 2010 Jun [cited 2014 May 27]. Available from: www.cdc.gov/nchs/data/hestat/obesity_child_07_08/obesity_child_07_08.pdf.
2. Skinner AC, Skelton JA. Prevalence and trends in obesity and severe obesity among children in the United States, 1999-2012. JAMA Pediatr 2014 Jun;168(6):561-6. DOI: https://doi.org/10.1001/jamapediatrics.2014.21.
3. Crocker MK, Yanovski JA. Pediatric obesity: etiology and treatment. Endocrinol Metab Clin North Am 2009 Sep;38(3):525-48. DOI: https://doi.org/10.1016/j.ecl.2009.06.007.
4. Wortsman J, Matsuoka LY, Chen TC, Lu Z, Holick MF. Decreased bioavailability of vitamin D in obesity. Am J Clin Nutr 2000 Sep;72(3):690-3. Erratum in: Am J Clin Nutr 2003 May;77(5):1342.
5. Vitamin D [Internet]. Bethesda, MD: Office of Dietary Supplements, National Institutes of Health; c2014 [cited 2014 May 27]. Available from: http://ods.od.nih.gov/factsheets/list-all/VitaminD/.
6. Jones G. Pharmacokinetics of vitamin D toxicity. Am J Clin Nutr 2008 Aug;88(2):582S-586S.
7. Ross AC, Manson JE, Abrams SA, et al. The 2011 report on dietary reference intakes for calcium and vitamin D from the Institute of Medicine: what clinicians need to know. J Clin Endocrinol Metab 2011 Jan;96(1):53-8. DOI: https://doi.org/10.1210/jc.2010-2704.
8. Looker AC, Pfeiffer CM, Lacher DA, Schleicher RL, Picciano MF, Yetley EA. Serum 25-hydroxyvitamin D status of the US population: 1988-1994 compared with 2000-2004. Am J Clin Nutr 2008 Dec;88(6):1519-27. DOI: https://doi.org/10.3945/ajcn.2008.26182.
9. Harel Z, Flanagan P, Forcier M, Harel D. Low vitamin D status among obese adolescents: prevalence and response to treatment. J Adolesc Health 2011 May;48(5):448-52. DOI: https://doi.org/10.1016/j.jadohealth.2011.01.011.
10. Alemzadeh R, Kichler J, Babar G, Calhoun M. Hypovitaminosis D in obese children and adolescents: relationship with adiposity, insulin sensitivity, ethnicity, and season. Metabolism 2008 Feb;57(2):183-91. DOI: https://doi.org/10.1016/j.metabol.2007.08.023.
11. Chiu KC, Chu A, Go VL, Saad MF. Hypovitaminosis D is associated with insulin resistance and beta cell dysfunction. Am J Clin Nutr 2004 May;79(5):820-5.
12. Holick MF. Vitamin D: importance in the prevention of cancers, type 1 diabetes, heart disease, and osteoporosis. Am J Clin Nutr 2004 Mar;79(3):362-71.
13. Freedman DS, Dietz WH, Srinivasan SR, Berenson GS. The relation of overweight to cardiovascular risk factors among children and adolescents: the Bogalusa Heart Study. Pediatrics 1999 Jun;103(6 Pt 1):1175-82. DOI: https://doi.org/10.1542/peds.103.6.1175.
14. Giovannucci E, Liu Y, Rimm EB, et al. Prospective study of predictors of vitamin D status and cancer incidence and mortality in men. J Natl Cancer Inst 2006 Apr 5;98(7):451-9. DOI: https://doi.org/10.1093/jnci/djj101.
15. Abbas S, Linseisen J, Chang-Claude J. Dietary vitamin D and calcium intake and premenopausal breast cancer risk in a German case-control study. Nutr Cancer 2007;59(1):54-61. DOI: https://doi.org/10.1080/01635580701390223.
16. McNally JD, Menon K, Chakraborty P, et al; Canadian Critical Care Trials Group. The association of vitamin D status with pediatric critical illness. Pediatrics 2012 Sep;130(3):
17. Abou-Raya A, Abou-Raya S, Helmii M. The effect of vitamin D supplementation on inflammatory and hemostatic markers and disease activity in patients with systemic lupus erythematosus: a randomized placebo-controlled trial. J Rheumatol 2013 Mar;40(3):265-72. DOI: https://doi.org/10.3899/jrheum.111594.
18. Science M, Maguire JL, Russell ML, Smieja M, Walter SD, Loeb M. Low serum 25-hydroxy-
19. RUCA Data [Internet]. Seattle, WA: WWAMI Rural Health Research Center; 2004 [cited 2014 Jul 17]. Available from: http://depts.washington.edu/uwruca/ruca-download.php.
20. Wagner CL, Greer FR; American Academy of Pediatrics Section on Breastfeeding; American Academy of Pediatrics Committee on Nutrition. Prevention of rickets and vitamin D deficiency in infants, children, and adolescents. Pediatrics 2008 Nov;122(5):1142-52. DOI: https://doi.org/10.1542/peds.2008-1862. Erratum in: Pediatrics 2009 Jan;123(1):197. DOI: https://doi.org/10.1542/peds.2008-3365.
21. Olson ML, Maalouf NM, Oden JD, White PC, Hutchison MR. Vitamin D deficiency in obese children and its relationship to glucose homeostasis. J Clin Endocrinol Metab 2012 Jan;97(1):279-85. DOI: https://doi.org/10.1210/jc.2011-1507.
22. Palacios C, Gil K, Pérez CM, Joshipura K. Determinants of vitamin D status among overweight and obese Puerto Rican adults. Ann Nutr Metab 2012;60(1):35-43. DOI: https://doi.org/10.1159/000335282.
23. Harris SS. Vitamin D and African Americans. J Nutr 2006 Apr;136(4):1126-9.
24. Powe CE, Evans MK, Wenger J, et al. Vitamin D-binding protein and vitamin D status of black Americans and white Americans. N Engl J Med 2013 Nov 21;369(21):1991-2000. DOI: https://doi.org/10.1056/NEJMoa1306357.
25. Behall KM, Scholfield DJ, Hallfrisch JG, Kelsay JL, Reiser S. Seasonal variation in plasma glucose and hormone levels in adult men and women. Am J Clin Nutr 1984 Dec;40(6 Suppl):1352-6.
26. Wehr E, Pilz S, Schweighofer N, et al. Association of hypovitaminosis D with metabolic disturbances in polycystic ovary syndrome. Eur J Endocrinol 2009 Oct;161(4):575-82. DOI: https://doi.org/10.1530/EJE-09-0432.
27. Zhou P, Schechter C, Cai Z, Markowitz M. Determinants of 25(OH)D sufficiency in obese minority children: selecting outcome measures and analytic approaches. J Pediatr 2011 Jun;158(6):930-4.e1. DOI: https://doi.org/10.1016/j.jpeds.2010.11.034.
28. Heaney RP, Davies KM, Chen TC, Holick MF, Barger-Lux MJ. Human serum 25-hydroxycholecalciferol response to extended oral dosing with cholecalciferol. Am J Clin Nutr 2003 Jan;77(1):204-10. Erratum in: Am J Clin Nutr 2003 Nov;78(5):1047.
29. Barlow SE; Expert Committee. Expert committee recommendations regarding the prevention, assessment, and treatment of child and adolescent overweight and obesity: summary report. Pediatrics 2007 Dec;120 Suppl 4:S164-92. DOI: https://doi.org/10.1542/peds.2007-2329C.