Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
J Am Geriatr Soc. Author manuscript; available in PMC 2012 February 20.
Published in final edited form as:
PMCID: PMC3282467

Correlates and prevalence of insufficient 25-hydroxyvitamin D status in black and white older adults: The Health ABC Study



Vitamin D insufficiency is associated with increased risk for age-related disease and mortality, leading to increased interest in measuring circulating 25-hydroxyvitamin D [25(OH)D]. Although blacks are more likely to have lower 25(OH)D than whites, whether the predictors of 25(OH)D are similar in blacks and whites is unclear. Therefore, we determined the prevalence and correlates of vitamin D insufficiency in black and white older adults.




Health, Aging and Body Composition (Health ABC) Study.


977 black and 1604 white adults aged 70–81 years.


Logistic regression and classification and regression tree analysis were used to identify correlates of vitamin D insufficiency (25(OH)D <30 ng/mL) in blacks and whites separately.


The prevalence of 25(OH)D insufficiency was 84% in blacks and 57% in whites. Seventy-six percent of blacks and 56% of whites did not take a multivitamin; those who did not take a multivitamin were more likely to be vitamin D insufficient [OR (95%CIs): blacks 5.17 (3.47–7.70); whites 2.56 (2.05–3.19)]. Additional risk factors for vitamin D insufficiency in blacks included vitamin D-containing supplement use, female gender and obesity while in whites included winter season, low dietary vitamin D intake, obesity, type 2 diabetes, and female gender.


Vitamin D insufficiency was more prevalent in blacks than whites. Not consuming a multivitamin increased the odds of vitamin D insufficiency in both blacks and whites. Knowledge of additional risk factors, such as dietary intake and co-morbid conditions, may help identify older adults who are likely to be vitamin D insufficient.

Keywords: vitamin D, insufficiency, race, prevalence, determinants


Vitamin D insufficiency has been associated with increased risk for several age-related chronic diseases, impaired physical function, and mortality in older age 13. Vitamin D status is measured clinically using the circulating concentration of 25-hydroxyvitamin D [25(OH)D]. It is generally accepted that 25(OH)D concentrations of at least 20 ng/mL (50 nmol/L; to convert to nmol/L, multiply by 2.496) are needed to prevent an elevation in parathyroid hormone (PTH) and maintain skeletal health, as recommended recently by the Institute of Medicine (IOM) 4 and others 57. Since vitamin D has many physiological functions beyond bone, experts have suggested 25(OH)D concentrations higher than 30 ng/mL (75 nmol/L) should be maintained to ensure vitamin D adequacy to meet both the calcemic and non-calcemic requirements 5;810, which is supported by several reports that serum 25(OH)D ≥30 ng/ml 25(OH)D is associated with improved cardiometabolic and functional outcomes 912. Whether vitamin D insufficiency is defined as <30 ng/mL or <20 ng/mL of circulating 25(OH)D, the prevalence is increasing among older Americans 13;14 and is more common in blacks than in whites.

Sunlight, the primary source of vitamin D, is synthesized in the skin upon exposure to UVB light. However, the ability to synthesize vitamin D cutaneously decreases with age, and is also reduced in African Americans 15;16. In addition to age and race, inadequate vitamin D intake, increased adiposity, and chronic disease also contribute to vitamin D insufficiency 9. However, few studies have examined the determinants of vitamin D status in blacks separately from whites 1719 and none have specifically examined the determinants of vitamin D insufficiency in older black and white adults. Furthermore, although epidemiological evidence of the association between vitamin D insufficiency and chronic disease, physical function and mortality has accumulated 13;17;2022, some of these associations appear to differ by race 15;2325. Thus, it is important to identify the extent to which modifiable and non-modifiable factors contribute to vitamin D insufficiency in order to effectively design and implement supplementation trials to more definitively determine vitamin D’s effect on health outcomes in older black and white adults and to help determine for whom 25(OH)D screening and/or vitamin D supplementation might be appropriate.

Therefore, the purpose of these analyses was to determine the prevalence and determinants of vitamin D insufficiency in community-dwelling older black and white adults separately, according to the 30 ng/mL 25(OH)D threshold of insufficiency. We also examined the prevalence and determinants of vitamin D insufficiency using 20 ng/mL 25(OH)D as the threshold, since the definition of sufficient 25(OH)D remains to be clarified, and persons with 25(OH)D below this threshold may be at even greater risk for vitamin D-related skeletal and non-skeletal co-morbidities.


Study Population

The Health ABC cohort consists of 3075 well-functioning community-dwelling black and white older adults who were recruited between April 1997 and June 1998 from a random sample of white and all black Medicare-eligible residents of Pittsburgh, PA (40.4°N latitude) and Memphis, TN (35.2°N latitude). At the time of enrollment, all Health ABC participants were 70–79 yrs old, reported no difficulty walking ¼ mile, climbing 10 stairs, or performing activities of daily living, were free of life-threatening illness, and planned to remain in the geographic area for at least 3 years. Unless indicated otherwise, data for these cross-sectional analyses were obtained at the 12-month follow-up visit (year 2) when dietary intake and serum 25(OH)D were measured. Participants who did not return for the year 2 visit (n=77) or were missing data on serum 25(OH)D (n= 205), dietary intake (n=103) and additional pertinent covariates (n=109) were excluded from these analyses. The final sample included 2581 participants (1604 white and 977 black). All participants provided written informed consent and all protocols were approved by the institutional review boards at both study sites.

Vitamin D status

Blood samples were taken at the year 2 clinic visit (1998–99) after an overnight fast and stored at −70°C until time of analysis. Serum 25(OH)D concentrations were measured using a 2-step radioimmunoassay (25-Hydroxyvitamin D 125I RIA Kit, DiaSorin, Stillwater, Minn., USA) at the University of Georgia. The inter-assay coefficient of variation was 6.78% for log-transformed values. For our primary analysis, 25(OH)D <30 ng/mL was considered insufficient based on the most recent expert consensus opinion 5;9. We also examined 25(OH)D <20 ng/mL.

Potential correlates of vitamin D status

At the year 2 clinic visit, dietary intake over the previous year was estimated using an interviewer-administered 108-item Food Frequency Questionnaire (FFQ), which was developed specifically for Health ABC. The FFQ food list was derived using 24-hour recall data obtained from NHANES III for non-hispanic black and white adults >65 yrs old, residing in the Northeast or South. The Health ABC FFQ was analyzed for micronutrient and macronutrient content using the Block Dietary Data System 26. To determine dietary supplement and medication use, participants were asked to bring to their clinic visit all medications and supplements they were currently taking. If a supplement had more than 3 vitamin or mineral ingredients, it was considered a multivitamin. If a supplement contained 3 or fewer ingredients, the individual nutrient ingredients were recorded. Vitamin D-containing supplements were defined as supplements containing 3 or fewer ingredients, one of which was vitamin D. Calcium-containing supplements were defined as supplements containing 3 or fewer ingredients, one of which was calcium. A supplement containing both calcium and vitamin D was considered as both a calcium supplement and a vitamin D supplement. Body weight and height were measured using a standard balance-beam scale and Harpenden stadiometer (Holtain Ltd., Crosswell UK), and BMI was calculated as weight (kg)/(height (m))2. Physical activity was self-reported as the number of minutes per week spent walking. Information on prevalent cardiovascular disease (CVD), diabetes, osteoporosis, hypertension, and knee osteoarthritis (knee pain) was obtained from self-report, clinic data, and medication use. At the baseline clinic visit, demographic and lifestyle characteristics, including age, gender, race, education, smoking status, and alcohol intake, were ascertained using an interviewer-administered questionnaire.

Biochemical measures

Serum C-reactive protein (CRP) was measured using ELISA (Calbiochem, San Diego, CA). Serum creatinine and total cholesterol were measured using a colorimetric technique on a Johnson & Johnson Vitrols 950 Analyzer (Johnson & Johnson, New Brunswick, NJ) and used to calculate the estimated glomerular filtration rate (eGFR) using the abbreviated Modification of Diet in Renal Disease equation.

Statistical analyses

Differences in vitamin D insufficiency by race were compared using a t-test for continuous (natural log) 25(OH)D. Initially we calculated the odds ratios (95% confidence intervals) for serum 25(OH)D <30ng/mL and <20ng/mL for blacks compared to whites, adjusted for season. Because our overall aim was to determine race-specific determinants of vitamin D status and the distribution of vitamin D insufficiency in blacks and whites were disparate (Figure 1), subsequent analyses were stratified by race. The relationship between categories of serum 25(OH)D (<20 ng/mL, 20-<30 ng/mL, ≥30 ng/mL) and participant characteristics were analyzed using chi-square tests for categorical variables. Differences in continuous exposures were also examined across categories of serum 25(OH)D using general linear models. Logistic regression with backward selection was used to identify predictors of vitamin D insufficiency defined as 25(OH)D <30 ng/mL and as <20 ng/mL separately in whites and blacks. Initially, the full models contained the following predictors: age, gender, vitamin D supplement use, multivitamin use, calcium supplement use, calcium intake (<1000 mg/d or ≥1000 mg/day), vitamin D intake (<400 IU/day or ≥400 IU/day), obesity (BMI ≥30 kg/m2 or <30 kg/m2), physical activity participation (none or any), current drinking (none, <1/week, 1–7/week, >7/week), smoking status (current, former, never), site (Pittsburgh, Memphis), season (Dec–Feb, Mar–May, June–Aug, Sept–Nov), kidney function (eGFR <60 ml/min or ≥60 ml/min), prevalent diabetes, cardiovascular disease, hypertension, osteoporosis, or knee osteoarthritis (knee pain), and inflammation (CRP >3 mg/L). The proportion of European ancestry (admixture %) was also included in the full model for blacks 27. Potential determinants were retained in the model if they were associated with vitamin D insufficiency at p<0.20, and the retained predictors were included in a second backward selection logistic regression analysis. The odds ratios (95% confidence intervals) for being vitamin D insufficient associated with the final predictors were calculated. The discriminative power of each logistic regression model was evaluated using the c-statistic. Multivariable linear regression with backward elimination was also used to determine predictors of serum 25(OH)D as a continuous (natural-log transformed) outcome in black and white adults.

Figure 1
Distribution of vitamin D status in older black and white adults in Health ABC.

The predictors that were retained in the final logistic regression models for blacks and whites were entered into classification tree analysis for model verification and to explore hierarchical relationships among predictor variables. Classification and Regression Tree (CART, version 6.0, Salford Systems, San Diego, CA) is an analysis software that uses a non-parametric technique and recursive partitioning to identify optimal cut-points in the data (referred to as nodes) and generates a classification tree that identifies those at greater risk for the outcome [25(OH)D <30 ng/mL] 28;29. Blacks and whites were analyzed separately. For each analysis, generally the tree with the smallest misclassification rate and highest area under the curve (AUC) was considered optimum. However, a tree with a misclassification rate statistically indistinguishable from the minimum and near maximum AUC was considered optimum due to a considerable reduction in nodes used in the tree in rare cases.


The mean (SD) serum 25(OH)D concentration was lower in blacks compared to whites overall (21.0(10.4)ng/mL vs 29.2(10.9)ng/mL, p<0.001) and during each season [March–May: blacks=21.2(12.3)ng/mL, whites=28.4(10.5)ng/mL; June–August: blacks=22.1(8.0)ng/mL, whites=30.5(8.7)ng/mL; Sept.–Nov.: blacks=21.2(8.5)ng/mL, whites=31.6(13.0)ng/mL; Dec.–Feb.: blacks=19.2(9.8)ng/mL, whites=26.6(10.4)ng/mL; all p<0.001]. The distribution of serum 25(OH)D by race is shown in Figure 1. Among blacks, 84% had serum 25(OH)D <30 ng/mL, and 54% had serum 25(OH)D <20 ng/mL. Among whites, 57% had serum 25(OH)D <30 ng/mL and 18% had serum 25(OH)D <20 ng/mL. Blacks were over four times more likely to have 25(OH)D <30 ng/mL [OR(95%CI): 4.36 (3.57–5.33), adjusted for season] and almost six times more likely to have 25(OH)D <20 ng/mL [OR(95%CI): 5.78 (4.79–6.92), adjusted for season], compared to whites.

The descriptive characteristics of black and white study participants according to serum 25(OH)D categories are shown in Tables 1 and and2.2. Overall, 73% of blacks and 51% of whites did not report taking multivitamins or dietary supplements that contained vitamin D. Over 90% of both black and white participants reported low dietary vitamin D intake (<400 IU/day). Following adjustment for age, gender, BMI, and season, among blacks there was a significant trend for higher dietary vitamin D intakes across the increasing 25(OH)D categories, but dietary calcium intake did not differ across 25(OH)D categories (Table 1). Among whites there were significant trends for higher vitamin D and calcium dietary intakes across increasing serum 25(OH)D categories (Table 2). Vitamin D status was associated with vitamin D-containing supplement use, season, obesity and prevalence of type 2 diabetes among blacks and whites, and with physical activity, kidney function and prevalence of CVD among whites in unadjusted analyses.

Table 1
Characteristics of black participants (n=977) by serum 25(OH)D concentration [n (%)*, unless indicated otherwise]
Table 2
Characteristics of white participants (n=1604) by serum 25(OH)D concentration [n (%)*, unless indicated otherwise]*

Participants who were excluded from these analyses due to missing covariate data (n=212) were more likely to be black, not take dietary supplements or participate in physical activity, and have type 2 diabetes, less than a high school education and lower mean serum 25(OH)D concentrations (23.2±10.9ng/mL vs. 26.1±11.5ng/mL, p=0.001), compared to those included in our sample.

Regression analyses

The odds ratios associated with identified predictors of vitamin D insufficiency in blacks are shown in Table 3. Among blacks, the odds of having 25(OH)D <30 ng/mL for those who did not report taking a multivitamin were more than five-fold greater than the odds for those who reported taking a multivitamin. Similarly, the odds for 25(OH)D <30 ng/mL were approximately three times higher for blacks who did not report taking a vitamin D-containing supplement compared to blacks who did. Being female, obese or living in Pittsburgh were all associated with greater odds of having 25(OH)D <30 ng/mL in blacks. Blacks were also more likely to have 25(OH)D <20 ng/mL during the winter.

Table 3
Correlates of 25(OH)D insufficiency among black participants (n=977)*

The odds ratios associated with identified predictors of vitamin D insufficiency in whites are shown in Table 4. Similar to blacks, multivitamin use was identified as a primary determinant of vitamin D insufficiency, defined according to either threshold, in whites. Obesity and prevalent CVD were also associated with greater odds of being vitamin D insufficient. Being female, having diabetes, and normal kidney function were associated with a higher odds of having 25(OH)D <30 ng/mL, while prevalent osteoporosis was associated with lower odds of 25(OH)D <30 ng/mL. Vitamin D intake <400 IU/day, lack of physical activity, and having 25(OH)D measured in the winter and spring season were associated with higher odds of having 25(OH)D <30 ng/mL and <20 ng/mL.

Table 4
Correlates of 25(OH)D insufficiency among white participants (n=1604)*

The results of linear stepwise models were comparable to those of the logistic stepwise models. Among blacks, vitamin D-containing supplements (unstandardized beta-coefficient (unstd β)=0.24), calcium-containing supplements (unstd β=0.17), and/or multivitamin (unstd β=0.31) were positively associated with (natural log) 25(OH)D (all p<0.001), and female gender (unstd β=−0.10, p<0.001), obese BMI (unstd β=−0.06 p=0.018), winter (unstd β=−0.14, p<0.001, compared to summer), spring (unstd β=−0.05, p=0.049, compared to summer), and living in Pittsburgh (unstd β=−0.09, p<0.001) were inversely associated 25(OH)D. Among whites, use of vitamin D-containing supplements (unstd β=0.09, p<0.001), calcium-containing supplements (unstd β=0.07, p<0.001), multivitamin (unstd β=0.21, p<0.001), prevalent osteoporosis (unstd β=0.06, p=0.01), and GFR < 60 (unstd β=0.08, p<0.001) were positively associated with (natural log) serum 25(OH)D, and female gender (unstd β=−0.12, p<0.001), obese BMI (unstd β=−0.07, p=0.002), calcium intake < 1000 mg/day (unstd β=−0.07, p=0.01), vitamin D intake < 400 IU/day (unstd β=−0.05, p=0.06), no participation in physical activity (unstd β=−0.05, p=0.003), prevalent CVD (unstd β=−0.06, p=0.002), prevalent diabetes (unstd β=−0.06, p=0.01), high cholesterol (unstd β=−0.04, p=0.01), winter (unstd β=−0.16, p<0.001, compared to summer), spring (unstd β=−0.09, p<0.001, compared to summer), and age (unstd β=−0.01, p=0.04) were inversely associated with 25(OH)D.

CART Analysis

Overall, the predictors of vitamin D insufficiency identified by our logistic regression analyses were confirmed by CART analysis, with multivitamin use being identified as the main predictor in both blacks and whites. Among blacks, multivitamin and vitamin D-containing supplement use were identified as the main predictors of serum 25(OH)D <30 ng/mL. Ninety percent of blacks who did not take a multivitamin had 25(OH)D <30 ng/mL. Even among blacks who did take a multivitamin and/or a vitamin D-containing supplement, the prevalence of 25(OH)D <30 ng/mL was 66% and 60%, respectively. Among whites, the prevalence of serum 25(OH)D <30 ng/mL increased to 66% among those who did not take a multivitamin. During winter, 83% of whites who did not take a multivitamin were vitamin D insufficient, based on 25(OH)D <30 ng/mL. Season, obesity, dietary intake of vitamin D, kidney function, and physical activity were associated with vitamin D insufficiency among whites using CART analysis. The classification trees for 25(OH)D <30 ng/mL for black and white participants are shown as supplementary on-line Figures S1 and S2.


In our sample of well-functioning men and women aged 70–81 years, 84% of blacks had 25(OH)D <30 ng/mL, while the prevalence among whites was 57%. Furthermore, blacks were more than four times as likely to have 25(OH)D <30 ng/mL compared to whites. The prevalence of 25(OH)D <30 ng/mL was lower in Health ABC compared to black and white men and women ≥70 years old in NHANES 2001–2004, in which the prevalence of 25(OH)D <30 ng/mL was approximately 93% in blacks and 73% in whites30. This may be due, in part, to the improved functional status and overall health of Health ABC participants.

Among blacks who reported taking a multi-vitamin or vitamin D-containing supplement, more than 60% had serum 25(OH)D <30ng/mL, while 38% of whites who reported taking these supplements had serum 25(OH)D <30ng/mL. Dietary supplement use did not appear to have been sufficient enough to increase serum 25(OH)D levels ≥30 ng/mL, especially among blacks It is important to note that supplement use was queried in 1998, when most multivitamins contained 400 IU or less of vitamin D; these levels have increased and many multivitamins currently contain 800 IU. Dose-response studies in blacks and whites suggest an intake of 400 IU/day (10 µg/d) vitamin D3 will increase circulating 25(OH)D by 2.8 ng/mL in the absence of sun exposure 31;32. It was recently reported that 800 IU/day of vitamin D3 increased circulating 25(OH)D to 35.4 ng/mL (from a mean baseline value of 12.1 ng/mL) in older adults 32. However, outcomes of other studies suggest higher doses may be necessary to achieve 25(OH)D ≥ 30 ng/mL in older adults 33, and higher doses of vitamin D given daily, monthly, or as a single bolus have been shown to be safe and effective for repletion of 25(OH)D in this age-group 3437. The baseline level of 25(OH)D, which is usually lower in blacks, is an important determinant of the response to supplementation, so it is plausible higher supplemental doses would have been required to raise serum levels to 30 ng/mL in the blacks in our cohort, as demonstrated by Aloia et al 38. Considering a recent report that a single annual dose of 500,000 IU of vitamin D3 increased hip fracture risk in older women 39, the optimal dosing regimen of vitamin D supplementation for older adults merits investigation.

Consistent with younger age groups, being female and/or obese increased the odds of being vitamin D insufficient in older black adults in Health ABC 17;19;40. We also identified several additional risk factors for vitamin D insufficiency among whites. Whites who reported a dietary intake <400IU/day were more likely to have 25(OH)D <30 ng/mL. Season and participation in physical activity were also predictive of vitamin D insufficiency among whites. Because the cutaneous production of vitamin D is greater in whites compared to blacks, whites tend to be more susceptible to seasonal variation in vitamin D status 41. Participation in physical activity was defined according to self-reported walking; thus, it is likely that those who reported walking spent more time outdoors which could influence the cutaneous synthesis of vitamin D. Being female, obese, or having type 2 diabetes was also associated with vitamin D insufficiency among whites, so these characteristics may help discriminate white older adults who are at greater risk for low vitamin D status.

Low dietary vitamin D intake (<400 IU/day) was identified as an important determinant of 25(OH)D <30 ng/mL among whites in Health ABC, but not as a primary determinant among blacks, although dietary intake has been identified as determinant of 25(OH)D in blacks by others 17;19;40. In our analyses, we dichotomized vitamin D intake as <400 or ≥400 IU/day because only 6 black and 8 white participants reported dietary intakes ≥800 IU/day and less than 2% of Health ABC participants (15 black and 30 white) reported dietary intakes ≥600 IU/day. In NHANES III (1988–1994), no more than 2% of adults over age 70 met the requirement for vitamin D intake from food alone, which at the time was 600IU/day 42. The prevalence of dietary vitamin D intake <400 IU/day was similar among blacks (90%) and whites (92%). The IOM recently recommended adults over 70 years old consume 800IU/day of vitamin D and 1200 mg/day of calcium4. The overall median (interquartile range) of vitamin D intake in Health ABC was 198(164) IU/day in whites and 159(157) IU/day in blacks and of calcium intakes was 739(444) mg/day in whites and 681(480) mg/day in blacks. Furthermore, 63% of blacks and 51% of whites in Health ABC reported a dietary vitamin D intake <200IU/day. Therefore, to meet the 800IU/day recommendation it is likely black and white older adults will require dietary supplementation.

Whites with type 2 diabetes and CVD were more likely to have 25(OH)D <30ng/ml. These results are consistent with previous analyses of NHANES III that found vitamin D status to be associated with prevalence of type 2 diabetes among whites and Hispanics, but not in blacks, which the authors attributed to blacks being less sensitive to the effects of vitamin D and related hormones 25, as has been suggested by others 15. It is plausible the associations between vitamin D status and disease outcomes differ by race because low vitamin D levels are reported to contribute to the increased risk for certain cancers (such as colorectal) and cancer-related mortality in blacks, but not in whites 24;43;44. In Health ABC, both black and white participants with osteoporosis were less likely to have insufficient vitamin D. The majority (70%) of participants diagnosed with osteoporosis were on medication to treat osteoporosis, which likely included advisement to take a vitamin D-containing supplement.

Although several have suggested ≥30 ng/mL are needed to meet skeletal and non-skeletal needs for vitamin D 5;810, ≥20 ng/mL have been considered sufficient to maintain skeletal health 4, and the concentration of 25(OH)D that is considered sufficient remains controversial5;8;9. We also examined the prevalence and correlates of 25(OH)D <20 ng/mL. Over half of black participants in Health ABC had 25(OH)D <20 ng/mL, while the prevalence among the whites was 18%. Overall, the determinants of 25(OH)D <20 ng/mL were similar to those of 25(OH)D <30 ng/mL. However, some differences were noted. Among blacks, season was associated with 25(OH)D <20 ng/mL, but not <30 ng/mL. Those measured during winter were over twice as likely to have 25(OH)D <20 ng/mL, supporting others who have suggested that winter season contributes to more severe vitamin D insufficiency among blacks 40. Among whites, the odds of having 25(OH)D <20 ng/mL were 85% higher among those who did not participate in any physical activity and among those who consumed <1000 mg/day of calcium. Although 25(OH)D <20 ng/mL is not considered to be frank vitamin D deficiency, 9% of black and 2% of white participants had 25(OH)D <10ng/mL which has been considered deficient 13. Approximately 80% of blacks and whites with 25(OH)D <10 ng/mL were female (Chi square p<0.001). Sixty-eight percent of blacks with 25(OH)D <10ng/mL were measured in spring or winter (Chi square p=0.01) and 96% of whites below 10 ng/mL reported calcium intakes < 1000 mg/day (Chi square p=0.02). Although Health ABC participants with 25(OH)D <10 ng/mL did not significantly from those with ≥10ng/mL in any other way, it is plausible additional differences would have been detected with a larger number of participants with 25(OH)D below this threshold. Since the evidence of vitamin D’s role in age-related disease continues to mount, identifying older adults at risk for more severe vitamin D insufficiency or deficiency may be important clinically.

This study has several notable strengths and certain limitations to consider. Importantly, we were able to describe the prevalence and determinants of vitamin D insufficiency in a large sample of black and white adults aged ≥70 years to provide insight into which older adults are at risk for vitamin D insufficiency. Although the determinants of vitamin D insufficiency among older adults have been previously described 17;45;46, the factors that are associated with specific thresholds of serum 25(OH)D in black and white adults aged ≥ 70 years has not yet been reported. Our results may assist geriatricians in identifying older adults to screen and treat for vitamin D insufficiency as well as researchers who need to identify vitamin D insufficient participants for supplementation trials. In Health ABC, dietary supplement use was not based on self-report. Rather, all participants were asked to bring all medications and supplements they were currently taking to the yearly clinic visit, which were then inventoried and recorded by study personnel. However, the quantity of vitamin D and/or calcium contained in the supplements is not available, so we are unable to determine associations between supplement dose and vitamin D insufficiency. It is likely that the high prevalence of vitamin D insufficiency even among those who reported taking a multivitamin or vitamin D-containing supplement may be because the supplement dose was not high enough to improve 25(OH)D concentrations to ≥30 ng/mL. However, we were unable to test this formally. Although we were able to adjust for latitude/location (Memphis vs Pittsburgh) and participation in physical activity, we did not have more specific measures of sunlight exposure and/or use of sunscreen, which have been identified as determinants of 25(OH)D in older adults in other population-based studies 45;46. Such measures would have likely explained more of the variability in 25(OH)D in Health ABC and would provide additional pertinent information with respect to additional risk factors for vitamin D insufficiency in older adults. Also, fewer blacks were recruited during the summer (June–August) at the Memphis field center which may have contributed to the association between season and vitamin D insufficiency we observed in whites compared to blacks. Substantial variation in the measurement of serum 25(OH)D, because of different assay methodology, laboratory experience, and differences between assays to recognize the vitamin D3 and D2 form equally 47 make it difficult to make comparisons across studies.

The importance of vitamin D’s role in health and disease is becoming more evident, and older adults have been identified as a group for whom the prolonged effects of vitamin D insufficiency may be exacerbated 14;14;48. We have identified that the determinants of vitamin D insufficiency differ in black and white older adults. Knowledge of supplement use, dietary intake, physical activity and/or prevalence of co-morbidities such as obesity, type 2 diabetes and/or cardiovascular disease may provide additional pertinent information to identify older adults who are at risk for vitamin D insufficiency and, thus, for whom 25(OH)D screening and/or vitamin D supplementation might be appropriate.

Supplementary Material

Supplementary Figures 1 and 2


Funding: The Health, Aging, and Body Composition Study was funded by the National Institute on Aging (N01-AG-6-2101, N01-AG-6-2103, and N01-AG-6-2106). This work was also supported by the National Institute of Aging (R01-AG029364), the Wake Forest University Claude D. Pepper Older Americans Independence Center (P30-AG21332), by the Intramural Research Program of the NIH, National Institute on Aging.

DKH receives speaker honorarium from Abbott Nutrition Health Institute

JAC receives grant support from Novartis (to study Zoledronic acid treatment)

Sponsor's Role: The funders were not involved in any aspect of study design, data acquisition, data analysis, interpretation of data, or manuscript preparation



Author Contributions: MKS, DKH, JAT, SKB study design, data interpretation, and manuscript preparation; CCD data analysis and manuscript preparation; MAJ and DAH laboratory analysis, data interpretation and manuscript preparation; JAC, DCB, FT, TBH data interpretation and manuscript preparation and editing.

Reference List

1. Ginde AA, Scragg R, Schwartz RS, Camargo CA., Jr Prospective study of serum 25-hydroxyvitamin D level, cardiovascular disease mortality, and all-cause mortality in older U.S. adults. J Am Geriatr Soc. 2009;57:1595–1603. [PubMed]
2. Houston DK, Cesari M, Ferrucci L, et al. Association between vitamin D status and physical performance: the InCHIANTI study. J Gerontol A Biol Sci Med Sci. 2007;62:440–446. [PMC free article] [PubMed]
3. Virtanen JK, Nurmi T, Voutilainen S, Mursu J, Tuomainen TP. Association of serum 25-hydroxyvitamin D with the risk of death in a general older population in Finland. Eur J Nutr. 2010 [PubMed]
4. Institute of Medicine. Dietary Reference Intakes for Calcium and Vitamin D 2010. Washington DC: National Academy Press; 2010.
5. Dawson-Hughes B, Heaney RP, Holick MF, Lips P, Meunier PJ, Vieth R. Estimates of optimal vitamin D status. Osteoporos Int. 2005;16:713–716. [PubMed]
6. Heaney RP, Dowell MS, Hale CA, Bendich A. Calcium absorption varies within the reference range for serum 25-hydroxyvitamin D. J Am Coll Nutr. 2003;22:142–146. [PubMed]
7. Malabanan A, Veronikis IE, Holick MF. Redefining vitamin D insufficiency. Lancet. 1998;351:805–806. [PubMed]
8. Bischoff-Ferrari HA, Giovannucci E, Willett WC, Dietrich T, Dawson-Hughes B. Estimation of optimal serum concentrations of 25-hydroxyvitamin D for multiple health outcomes. Am J Clin Nutr. 2006;84:18–28. [PubMed]
9. Holick MF. Vitamin D deficiency. N Engl J Med. 2007;357:266–281. [PubMed]
10. Bischoff-Ferrari HA, Shao A, Dawson-Hughes B, Hathcock J, Giovannucci E, Willett WC. Benefit-risk assessment of vitamin D supplementation. Osteoporos Int. 2010;21:1121–1132. [PMC free article] [PubMed]
11. Bischoff-Ferrari HA, Dietrich T, Orav EJ, et al. Higher 25-hydroxyvitamin D concentrations are associated with better lower-extremity function in both active and inactive persons aged > or =60 y. Am J Clin Nutr. 2004;80:752–758. [PubMed]
12. Forman JP, Giovannucci E, Holmes MD, et al. Plasma 25-hydroxyvitamin D levels and risk of incident hypertension. Hypertension. 2007;49:1063–1069. [PubMed]
13. Ginde AA, Liu MC, Camargo CA., Jr Demographic differences and trends of vitamin D insufficiency in the US population, 1988–2004. Arch Intern Med. 2009;169:626–632. [PMC free article] [PubMed]
14. Mosekilde L. Vitamin D and the elderly. Clin Endocrinol (Oxf) 2005;62:265–281. [PubMed]
15. Aloia JF. African Americans, 25-hydroxyvitamin D, osteoporosis: a paradox. Am J Clin Nutr. 2008;88:545S–550S. [PMC free article] [PubMed]
16. Harris SS, Soteriades E, Coolidge JA, Mudgal S, Dawson-Hughes B. Vitamin D insufficiency and hyperparathyroidism in a low income, multiracial, elderly population. J Clin Endocrinol Metab. 2000;85:4125–4130. [PubMed]
17. Benjamin A, Moriakova A, Akhter N, et al. Determinants of 25-hydroxyvitamin D levels in African-American and Caucasian male veterans. Osteoporos Int. 2009 [PubMed]
18. Egan KM, Signorello LB, Munro HM, Hargreaves MK, Hollis BW, Blot WJ. Vitamin D insufficiency among African-Americans in the southeastern United States: implications for cancer disparities (United States) Cancer Causes Control. 2008;19:527–535. [PubMed]
19. Nesby-O'Dell S, Scanlon KS, Cogswell ME, et al. Hypovitaminosis D prevalence and determinants among African American and white women of reproductive age: third National Health and Nutrition Examination Survey, 1988–1994. Am J Clin Nutr. 2002;76:187–192. [PubMed]
20. Giovannucci E, Liu Y, Hollis BW, Rimm EB. 25-hydroxyvitamin D and risk of myocardial infarction in men: a prospective study. Arch Intern Med. 2008;168:1174–1180. [PubMed]
21. Pilz S, Marz W, Wellnitz B, et al. Association of vitamin D deficiency with heart failure and sudden cardiac death in a large cross-sectional study of patients referred for coronary angiography. J Clin Endocrinol Metab. 2008;93:3927–3935. [PubMed]
22. Reis JP, von MD, Kritz-Silverstein D, Wingard DL, Barrett-Connor E. Vitamin D, parathyroid hormone levels, and the prevalence of metabolic syndrome in community-dwelling older adults. Diabetes Care. 2007;30:1549–1555. [PubMed]
23. Fiscella K, Winters P, Tancredi D, Hendren S, Franks P. Racial disparity in death from colorectal cancer: does vitamin D deficiency contribute? Cancer. 2010 [PubMed]
24. Giovannucci E, Liu Y, Willett WC. Cancer incidence and mortality and vitamin D in black and white male health professionals. Cancer Epidemiol Biomarkers Prev. 2006;15:2467–2472. [PubMed]
25. Scragg R, Sowers M, Bell C. Serum 25-hydroxyvitamin D, diabetes, and ethnicity in the Third National Health and Nutrition Examination Survey. Diabetes Care. 2004;27:2813–2818. [PubMed]
26. Houston DK, Nicklas BJ, Ding J, et al. Dietary protein intake is associated with lean mass change in older, community-dwelling adults: the Health, Aging, and Body Composition (Health ABC) Study. Am J Clin Nutr. 2008;87:150–155. [PubMed]
27. Shaffer JR, Kammerer CM, Reich D, et al. Genetic markers for ancestry are correlated with body composition traits in older African Americans. Osteoporos Int. 2007;18:733–741. [PubMed]
28. Breiman L, Freidman JH, Olshen RA, Stone CJ, et al. Classification and Regression Trees. 2 ed. Pacific Grove, CA: Wadsworth; 1984.
29. Lemon SC, Roy J, Clark MA, Friedmann PD, Rakowski W. Classification and regression tree analysis in public health: methodological review and comparison with logistic regression. Ann Behav Med. 2003;26:172–181. [PubMed]
30. 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;88:1519–1527. [PMC free article] [PubMed]
31. 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;77:204–210. [PubMed]
32. Talwar SA, Aloia JF, Pollack S, Yeh JK. Dose response to vitamin D supplementation among postmenopausal African American women. Am J Clin Nutr. 2007;86:1657–1662. [PMC free article] [PubMed]
33. Harris SS, Dawson-Hughes B. Plasma vitamin D and 25OHD responses of young and old men to supplementation with vitamin D3. J Am Coll Nutr. 2002;21:357–362. [PubMed]
34. Bischoff-Ferrari HA, Dawson-Hughes B, Platz A, et al. Effect of high-dosage cholecalciferol and extended physiotherapy on complications after hip fracture: a randomized controlled trial. Arch Intern Med. 2010;170:813–820. [PubMed]
35. Lips P, Binkley N, Pfeifer M, et al. Once-weekly dose of 8400 IU vitamin D(3) compared with placebo: effects on neuromuscular function and tolerability in older adults with vitamin D insufficiency. Am J Clin Nutr. 2010;91:985–991. [PubMed]
36. Mocanu V, Stitt PA, Costan AR, et al. Long-term effects of giving nursing home residents bread fortified with 125 microg (5000 IU) vitamin D(3) per daily serving. Am J Clin Nutr. 2009;89:1132–1137. [PubMed]
37. Bacon CJ, Gamble GD, Horne AM, Scott MA, Reid IR. High-dose oral vitamin D3 supplementation in the elderly. Osteoporos Int. 2009;20:1407–1415. [PubMed]
38. Aloia JF, Patel M, Dimaano R, et al. Vitamin D intake to attain a desired serum 25-hydroxyvitamin D concentration. Am J Clin Nutr. 2008;87:1952–1958. [PubMed]
39. Sanders KM, Stuart AL, Williamson EJ, et al. Annual high-dose oral vitamin D and falls and fractures in older women: a randomized controlled trial. JAMA. 2010;303:1815–1822. [PubMed]
40. Tseng M, Giri V, Bruner DW, Giovannucci E. Prevalence and correlates of vitamin D status in African American men. BMC Public Health. 2009;9:191. [PMC free article] [PubMed]
41. Armas LA, Dowell S, Akhter M, et al. Ultraviolet-B radiation increases serum 25-hydroxyvitamin D levels: the effect of UVB dose and skin color. J Am Acad Dermatol. 2007;57:588–593. [PubMed]
42. Moore C, Murphy MM, Keast DR, Holick MF. Vitamin D intake in the United States. J Am Diet Assoc. 2004;104:980–983. [PubMed]
43. Fiscella K, Franks P. Vitamin D, race, and cardiovascular mortality: findings from a national US sample. Ann Fam Med. 2010;8:11–18. [PubMed]
44. Erber E, Maskarinec G, Lim U, Kolonel LN. Dietary vitamin D and risk of non-Hodgkin lymphoma: the multiethnic cohort. Br J Nutr. 2010;103:581–584. [PMC free article] [PubMed]
45. Jacques PF, Felson DT, Tucker KL, et al. Plasma 25-hydroxyvitamin D and its determinants in an elderly population sample. Am J Clin Nutr. 1997;66:929–936. [PubMed]
46. van Dam RM, Snijder MB, Dekker JM, et al. Potentially modifiable determinants of vitamin D status in an older population in the Netherlands: the Hoorn Study. Am J Clin Nutr. 2007;85:755–761. [PubMed]
47. Binkley N, Krueger D, Cowgill CS, et al. Assay variation confounds the diagnosis of hypovitaminosis D: a call for standardization. J Clin Endocrinol Metab. 2004;89:3152–3157. [PubMed]
48. Oudshoorn C, van der Cammen TJ, McMurdo ME, van Leeuwen JP, Colin EM. Ageing and vitamin D deficiency: effects on calcium homeostasis and considerations for vitamin D supplementation. Br J Nutr. 2009;101:1597–1606. [PubMed]