Vitamin D deficiency among older women with and without disability

¹ From The Johns Hopkins Medical Institutions, Baltimore, and the Epidemiology, Demography, and Biometry Program, National Institute on Aging, Bethesda, MD.

² Supported by the National Institutes of Health (grants RO1 AG11703, NO1 AG122112, and RO1 AI41956) and the Johns Hopkins General Clinic Research Center (grant RR00722).

³ Address reprint requests to RD Semba, 550 N Broadway, Suite 700, Baltimore, MD 21205. E-mail: rdsemba{at}jhmi.edu.

ABSTRACT

Background: Vitamin D deficiency is associated with bone loss and bone fractures, and the identification of vulnerable populations is important to clinical practice and public health.

Objective: The objectives of this study were to determine the prevalence of vitamin D deficiency and to examine associated risk factors for vitamin D deficiency in older women.

Design: We measured serum concentrations of 25-hydroxyvitamin D [25(OH)D], 1,25-dihydroxyvitamin D [1,25(OH)₂D], intact parathyroid hormone (PTH), osteocalcin, and ionized calcium in women aged 65 y who were participating in the Women's Health and Aging Study I, an observational study of women representing the approximately one-third most disabled women living in the community, and women aged 70–80 y who were participating in the Women's Health and Aging Study II, an observational study of women among the two-thirds least disabled women living in the community in Baltimore.

Results: The women were classified into 4 domains of physical disability. Among 371 women with 0 or 1 domain of disability and 682 women with 2 domains of disability, 6.2% and 12.6%, respectively, had vitamin D deficiency [serum concentrations of 25(OH)D < 25 nmol/L]. In univariate analyses, risk factors for vitamin D deficiency included increasing age, black race, low educational level, high body mass index, high triceps skinfold thickness, increasing level of disability, winter season, and elevated creatinine concentration. In multivariate models, black race had a strong association with vitamin D deficiency when other risk factors were adjusted for.

Conclusions: Vitamin D deficiency, a preventable disorder, is a common and important public health problem for older disabled women living in the community; black women are at higher risk than are white women.

Key Words: Vitamin D deficiency • calcium • bone fractures • disability • women • aging • Baltimore

INTRODUCTION

Vitamin D is essential for normal calcium metabolism and maintenance of bone density. Vitamin D deficiency is associated with bone loss and bone fractures (1). The risk of deficiency increases with age (2), as skinfold thickness increases (3), during the winter season (4), with obesity (5, 6), and with renal disease (7). Vitamin D supplementation was shown previously to reduce bone loss from the femoral neck (8, 9); calcium supplementation also reduces bone loss in postmenopausal women (10). Supplementation with both calcium and vitamin D was shown to reduce bone loss and the incidence of nonvertebral fractures in men and women aged 65 y and living in the community (6) and to reduce nonvertebral fractures and hip fractures in older, healthy, ambulatory women (11).

Sunlight and diet are the 2 major sources of vitamin D in humans. Vitamin D₃ is produced in the human skin after photoconversion of 7-hydrocholesterol to previtamin D₃, which then isomerizes to vitamin D₃ (12). Ultraviolet exposure studies showed that adults with heavy skin pigmentation have lower cutaneous synthesis of vitamin D₃ than do adults with lighter skin pigmentation (13). Despite public health measures in the United States to fortify dairy products and cereals with vitamin D, a high prevalence of vitamin D deficiency has been described among older community-dwelling adults and those who are homebound (14–16). A preliminary study suggested that older African American women are at higher risk of vitamin D deficiency than are other racial groups (17). Identification of groups at high risk of vitamin D deficiency is important because vitamin D deficiency is readily amenable to dietary vitamin D and calcium supplementation. We hypothesized that vitamin D deficiency would be prevalent among community-dwelling, older women with disability and would have the highest prevalence among African American women. To address these hypotheses, we measured vitamin D concentrations and examined epidemiologic risk factors in older women with and without disability who were living in the community.

SUBJECTS AND METHODS

Subjects
The population of the Women's Health and Aging Study (WHAS) I and WHAS II consisted of a random sample from the Health Care Finance Administration's Medicare enrollment file for 32538 women residing in 12 contiguous ZIP-code areas in Baltimore; details of the sample were reported previously (18). In brief, for WHAS I an age-stratified (65–74, 75–84, and 95 y) random sample of women was selected; 5316 were eligible for screening, 4137 were screened in the home, 1409 met the study criteria, and 1002 agreed to participate in the study. The screening interview was designed to identify approximately the one-third most disabled older women living in the community. Physical disability was categorized into 4 domains of physical function based on self-report of difficulty in any of several related tasks primarily associated with 1) mobility and exercise tolerance, 2) upper extremity function, 3) higher functioning tasks, and 4) basic self-care tasks (19). Domains 0, 1, 2, 3, and 4 were defined as no reported disability, 1 reported disability, 2 reported disabilities, 3 reported disabilities, or 4 reported disabilities with the activities of daily living, respectively. Details of the methods and questionnaires used in determining the domains of disability can be found elsewhere (18). After providing written, informed consent, the study participants received an extensive interview and examination in their homes.

For WHAS II, an age-stratified (70–74 and 75–79 y) random sample of women was selected; 3541 women were eligible for screening, 1630 were screened (primarily by telephone), 880 met the study criteria, and 436 agreed to participate in the study. To be eligible for screening, subjects had to be able to be contacted by telephone and to have sufficient hearing and English-language proficiency to be interviewed. Eligibility for participation in the study was determined on the basis of self-report of difficulties in no tasks (out of 15 assessed) or in only one domain of physical function among mobility tasks, upper-extremity tasks, household management tasks, and self-care tasks (18). These criteria were shown previously to identify the higher functioning two-thirds of older women (18). After providing written, informed consent, the women received an extensive physical examination at the Johns Hopkins Functional Status Laboratory; answered standardized questionnaires that ascertained demographic characteristics, chronic diseases, and self-report of difficulty with tasks of daily life; and had blood drawn for laboratory studies. WHAS I was conducted entirely in the participants' homes; blood was drawn on a separate visit with a separate consent form. WHAS II was conducted at the Johns Hopkins Functional Status Laboratory with methods rigorously standardized to WHAS I. The study protocol was approved by the Joint Committee on Clinical Investigation at the Johns Hopkins School of Medicine.

Laboratory analysis
Nonfasting blood samples were obtained by venipuncture, and the samples were processed, placed on ice, and sent the same day to the central laboratory of Quest Diagnostics Laboratories (formerly Corning Clinical Laboratories and MedPath) in Teterboro, NJ, for analysis. Serum 25-hydroxyvitamin D [25(OH)D] was measured with use of a radioreceptor assay (Nichols Institute Diagnostics, San Juan Capistrano, CA) (20); the interassay and intraassay CVs were 9.6% and 7.5%, respectively. Serum 1,25-dihydroxyvitamin D [1,25(OH)₂D] was measured with use of extraction, chromatography, and radioreceptor assay (21); the interassay and intraassay CVs were 10.9% and 7.5%, respectively. Intact serum parathyroid hormone (PTH) was measured by using chemiluminescence (22); the interassay and intraassay CVs were 6.7% and 5.7%, respectively. Serum osteocalcin was measured by using radioimmunoassay (23); the interassay and intraassay CVs were 5.2% and 4.0%, respectively. Serum ionized calcium was measured with ion selective electrodes (Nova 8; Nova Biomedical, Waltham, MA) (24), with an SD of 0.05 mmol/L for standards.

Data analysis
The data for WHAS I and WHAS II were analyzed separately because of differences in sampling and eligibility criteria and recruitment. The mean values of each measurement were calculated by severity (number of domains of disability). In WHAS I, there were 2–4 domains in which participants were disabled; in WHAS II, there were 0 or 1 domains of disability. Vitamin D deficiency (12) was defined as a serum 25(OH)D concentration <25 nmol/L (10 ng/mL). The prevalence of vitamin D deficiency in both studies was also examined by using an alternative cutoff of 50 nmol/L (25). Chi-square and trend tests were used to examine the associations between vitamin D deficiency and dichotomized covariates as follows. Education level was classified as 12 or <12 y of education. A body mass index (BMI; in kg/m²) 30 was used to define obesity according to the criteria of the National Institutes of Health (26). A triceps skinfold thickness of >30.8 mm was also used to define obesity, because this was the 85th percentile of triceps skinfold thickness of reference data for obesity (27). Season was classified as winter (November through March) or nonwinter (April through October). Creatinine was considered abnormal if it was >14 mg/L (>124 µmol/L) (28). Logistic regression models were fitted with vitamin D deficiency as the outcome variable. Multivariate logistic regression models were fitted to determine the risk factors that were significant and independently predictive of vitamin D deficiency after adjustment for other factors. Lowess smoothing was used to calculate a smooth line in scatter plots (29). Significance was set at P < 0.05.

RESULTS

Laboratory studies were performed in 682 of 1002 women (68.0%) enrolled in WHAS I because not all women agreed to undergo phlebotomy. In WHAS II, laboratory studies were performed in 369 of 436 (84.6%) of women enrolled. The women who participated in the blood drawing in WHAS I and WHAS II were not significantly different from the women who did not participate by race, season, or BMI. There were significant differences between participants and nonparticipants by domain of disability and by age category. In domain 4, some 66% of women participated, and in women aged 85 y, 62.4% participated, compared with an overall participation rate of 77%.

The serum concentrations of 25(OH)D, 1,25(OH)₂D, PTH, osteocalcin, and ionized calcium are shown according to the number of domains of disability (0–4) in Table 1. When an alternative serum 25(OH)D concentration cutoff of 50 nmol/L (25) was used, 344 of 682 women (50.4%) in WHAS I and 165 of 371 women (44.5%) in WHAS II had vitamin D deficiency. Plasma concentrations of PTH are shown plotted against plasma 25(OH)D concentrations for WHAS I and WHAS II in Figure 1. A smoothing line was calculated (29). A significant linear relation was noted between PTH and 25(OH)D in both studies. No significant relation was found between serum 25(OH)D concentrations and serum osteocalcin concentrations or between serum 25(OH)D concentrations and ionized calcium concentrations (data not shown). No women in either WHAS I or WHAS II had elevated plasma PTH concentrations and hypercalcemia consistent with primary hyperparathyroidism.

View this table:
TABLE 1. Serum vitamin D, intact parathyroid hormone (PTH), ionized calcium, and osteocalcin in the study subjects by domain of disability¹  

View larger version (28K):
FIGURE 1. . Plasma concentrations of intact parathyroid hormone versus 25-hydroxyvitamin D [25(OH)D] in the Women's Health and Aging Study I (left) and plasma concentrations of PTH versus 25(OH)D in the Women's Health and Aging Study II (right).

 
In WHAS I, which included women with 2 domains of disability, significant risk factors for vitamin D deficiency included increasing age and black race (Table 2). There was no significant relation between season, BMI, high triceps skinfold thickness, and vitamin D deficiency in WHAS I, a moderately to severely disabled population. There was a marginal association between elevated creatinine concentration and vitamin D deficiency (P < 0.08). In WHAS II, which included women with 0 or 1 domain of disability, black race, high BMI, high triceps skinfold thickness, winter season, and elevated creatinine concentration were significant risk factors for vitamin D deficiency. Level of education was not a significant risk factor for vitamin D deficiency in WHAS II.

View this table:
TABLE 2. Risk factors for hypovitaminosis D in the study subjects  
Logistic regression models were fitted with vitamin D deficiency as the outcome variable in univariate analyses. Risk factors that were investigated included age, race, education, BMI, triceps skinfold thickness, hip fracture, season, creatinine concentration, walking outdoors, and outdoor chores. The data for WHAS I and WHAS II were analyzed separately. Age, race, education, creatinine concentration, and outdoor chores were significant at the 0.10 level in univariate analyses in WHAS I. Race, BMI, triceps skinfold thickness, season, and creatinine concentration were significant at the 0.10 level in univariate analyses in WHAS II. Multiple logistic regression models were then fitted to determine which covariates remained significant after adjustment for other factors. Variables that showed little association with vitamin D deficiency after adjustment for other variables were excluded from the model. As shown in Table 3, in WHAS I, increasing age and black race were significantly associated with increased risk of vitamin D deficiency. In WHAS II, black race, BMI, and winter season were independent predictors of vitamin D deficiency.

View this table:
TABLE 3. Multivariate models for vitamin D deficiency  
DISCUSSION

The results of this study suggest that vitamin D deficiency is common in older women living in the community and is associated with higher levels of disability. It is notable that even among the two-thirds least disabled older women in the community (WHAS II), 6% had vitamin D deficiency, and among the one-third most disabled older women in the community (WHAS I), >12% had vitamin D deficiency. Overall, the prevalence of vitamin D deficiency in WHAS I and II was higher than in the Framingham Heart Study cohort, in which only 4% of elderly women had plasma 25(OH)D concentrations consistent with deficiency (25 nmol/L) (16). No vitamin D deficiency was reported in the Baltimore Longitudinal Study of Aging, an investigation involving highly selected, self-recruited, mostly white, educated professionals without disability or major disease (30). Mean serum 25(OH) concentrations of nondisabled older women in WHAS II were similar to mean concentrations found in nondisabled, elderly white women in rural Iowa (31). The overall prevalence of vitamin D deficiency in WHAS I was similar to the 12–15% reported in noninstitutionalized adults in the third National Health and Nutrition Examination Survey (NHANES III, 1988–1994) (32).

The estimates of vitamin D deficiency in WHAS are probably conservative because older and more disabled women did not participate in the blood screening as much as did younger, less disabled women. It was suggested recently that older adults may require a serum 25(OH)D concentration of 50 nmol/L to achieve optimal PTH concentrations (25). Data from both WHAS I and WHAS II support this idea, showing an inflection of PTH with 25(OH)D at 40–50 nmol 25(OH)D/L. Although a cutoff of 25 nmol/L for plasma 25(OH)D is most commonly used to define vitamin D deficiency (12), these data suggest that a higher cutoff for vitamin D deficiency may be needed to reduce the risk of secondary hyperparathyroidism in older adults.

About 15% and 20% of black women had vitamin D deficiency in WHAS II and WHAS I, respectively, and the association between black race and vitamin D deficiency was significant after adjustment for other risk factors. This study corroborates the findings of a previous preliminary study that suggested that older African American women are at higher risk of vitamin D deficiency than are older white women (17). These findings are also consistent with the results of studies that suggested that there is lower cutaneous synthesis of vitamin D₃ with increasing skin pigmentation (13). Another recent study of women aged 20–40 y suggested that black women have significantly lower plasma 25(OH)D concentrations than do white women (33). In general, black persons are known to have denser bones and lower fracture rates than do whites (34), and there may be a relative skeletal resistance to elevated PTH concentrations in black women (35). Further studies are needed to determine whether the risk of occult vitamin D deficiency is high in black women who develop fractures.

The results of this study corroborate previous observations that the risk of vitamin D deficiency increases with age (2, 3). Age was a significant risk factor in the WHAS I study after adjustment for race and level of education in multivariate analyses. In the WHAS II study, age was not a significant risk factor in the multivariate analyses, but this may have been the result of a more limited age range in WHAS II. In the Framingham Heart Study cohort, a weak association was found between low concentrations of 25(OH)D and age among elderly women but not men (16), and plasma 25(OH)D concentrations were inversely related to age among women aged 20–80 y in Iowa (31). Aging has been shown to decrease the capacity of human skin to produce vitamin D₃ (36) and the decline in skin capacity may be linked to an age-related decrease in skin thickness (3).

The association between obesity, as indicated by high BMI and high triceps skinfold thickness, and vitamin D deficiency was found in WHAS II but not in WHAS I. The women in WHAS II had minimal or no disability; in this group, obesity may have been more of a defining factor in determining outdoor activity and exposure to sunlight. In WHAS I, obesity may have been less important in determining outside activity and sunlight exposure because of the level of disability in this group. Obesity has been associated with altered vitamin D metabolism (5), and other studies showed an inverse relation between BMI and plasma 25(OH)D concentrations (3, 6, 16). No relation was found between self-reported hip fracture and vitamin D deficiency. The number of women with reported hip fracture was small, and the questionnaire was limited in that it did not differentiate whether the reported hip fracture was recent or in the remote past.

This study confirmed the relation between season and plasma or serum 25(OH)D concentrations, as reported previously in different populations (37–40). Winter season was significantly associated with vitamin D deficiency in WHAS II but not in WHAS I, and the lack of association in WHAS I may have been due to the degree of disability among the women. The more disabled older women probably showed no seasonality in their serum 25(OH)D concentrations because of limited outdoor activity and sunlight exposure. Seasonal changes in 25(OH)D concentrations were shown to be greatest in free-living subjects and less with decreasing mobility (4).

In WHAS I, a low level of education was associated with vitamin D deficiency, but in WHAS II, which included the least disabled women in the community, such an association was not found. The reasons for this association in WHAS I are unclear, but low education level may represent lower income and poorer dietary intake of vitamin D–containing foods. A limitation of this study was that detailed dietary information was not collected.

This study showed that vitamin D deficiency, a preventable disorder, is an important public health problem among older women living in the community. Women with vitamin D deficiency are at significant risk for bone loss, bone fractures, and disability, and the results of this study raise the issue of whether clinicians should routinely screen older women for vitamin D deficiency. A recent case series showed that a high proportion of postmenopausal women with hip fracture had vitamin D deficiency (40). Vitamin D deficiency is readily alleviated with dietary supplementation, but amounts beyond that available in milk or cereal are probably needed (41). Calcium and vitamin D supplementation may help slow bone resorption and increase bone formation (6). Such inexpensive interventions should be encouraged to reduce osteoporosis and fractures in older adults.

REFERENCES

1. Lips P, van Ginkel FC, Jongen MJM, Rubertus F, van der Vijgh WJF, Netelenbos JC. Determinants of vitamin D status in patients with hip fracture and in elderly control subjects. Am J Clin Nutr 1987;46:1005–10.
2. van der Wielen RPJ, Löwik MRH, van der Berg H, et al. Serum vitamin D concentrations among elderly people in Europe. Lancet 1995; 346:207–10.
3. Need AG, Morris HA, Horowitz M, Nordin BEC. Effects of skin thickness, age, body fat, and sunlight on serum 25-hydroxyvitamin D. Am J Clin Nutr 1993;58:882–5.
4. Webb AR, Pilbeam C, Hanafin N, Holick MF. An evaluation of the relative contributions of exposure to sunlight and of diet to the circulating concentrations of 25-hydroxyvitamin D in an elderly nursing home population in Boston. Am J Clin Nutr 1990;51:1075–81.
5. Bell NH, Epstein S, Greene A, Shary J, Oexmann MJ, Shaw S. Evidence for alteration of the vitamin D-endocrine system in obese subjects. J Clin Invest 1985;76:370–3.
6. Dawson-Hughes B, Harris SS, Krall EA, Dallal GE. Effect of calcium and vitamin D supplementation on bone density in men and women 65 years of age or older. N Engl J Med 1997;337:670–6.
7. Korkor AB. Reduced binding of [³H]1,25-dihydroxyvitamin D₃ in the parathyroid glands of patients with renal failure. N Engl J Med 1987;316:1573–7.
8. Dawson-Hughes B, Harris SS, Krall EA, Dallal GE, Falconer G, Green CL. Rates of bone loss in postmenopausal women randomly assigned to one of two dosages of vitamin D. Am J Clin Nutr 1995; 61:1140–5.
9. Ooms ME, Roos JC, Bezemer PD, van der Vijgh WJF, Bouter LM, Lips P. Prevention of bone loss by vitamin D supplementation in elderly women: a randomized double-blind trial. J Clin Endocrinol Metab 1995;80:1052–8.
10. Dawson-Hughes B, Dallal GE, Krall EA, Sadowski L, Sahyoun N, Tannenbaum S. A controlled trial of the effect of calcium supplementation on bone density in postmenopausal women. N Engl J Med 1990;323:878–83.
11. Chapuy MC, Arlot ME, Duboeuf F, et al. Vitamin D3 and calcium to prevent hip fractures in the elderly women. N Engl J Med 1992; 327:1637–42.
12. Holick MF. The use and interpretation of assays for vitamin D and its metabolites. J Nutr 1990;120:1464–9.
13. Clemens TL, Adams JS, Henderson SL, Holick MF. Increased skin pigment reduces the capacity of skin to synthesise vitamin D₃. Lancet 1982;1:74–6.
14. Omdahl JL, Garry PJ, Hunsaker LA, Hunt WC, Goodwin JS. Nutritional status in a healthy elderly population: vitamin D. Am J Clin Nutr 1982;36:1125–33.
15. Gloth FM, Gundberg CM, Hollis BW, Haddad JG, Tobin JD. Vitamin D deficiency in homebound elderly persons. JAMA 1995;274:1683–6.
16. 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;66:929–36.
17. Perry HM III, Miller DK, Morley JE, et al. A preliminary report of vitamin D and calcium metabolism in older African Americans. J Am Geriatr Soc 1993;41:612–6.
18. Fried LP, Kasper JD, Guralnik JM, Simonsick EM. The Women's Health and Aging Study: an introduction. In: Guralnik JM, Fried LP, Simonsick EM, Kasper KD, Lafferty ME, eds. The Women's Health and Aging Study: health and social characteristics of old women with disability. Bethesda, MD: National Institute on Aging, 1995:1–8. (NIH publication no. 95-4009.)
19. Guralnik JM, Fried LP, Simonsick EM, Bandeen-Roche KJ, Kasper JD. Screening the community for disability. In: Guralnik JM, Fried LP, Simonsick EM, Kasper KD, Lafferty ME, eds. The Women's Health and Aging Study: health and social characteristics of old women with disability. Bethesda, MD: National Institute on Aging, 1995:9–27. (NIH publication no. 95-4009.)
20. Hadda JG, Chyu KJ. Competitive protein-binding radioassay for 25-hydroxycholecalciferol. J Clin Endocrinol Metab 1971;33:992–5.
21. Reinhardt TA, Horst RL, Orf JW, Hollis BW. A microassay for 1,25 dihydroxy vitamin D not requiring high pressure liquid chromatography: applications in clinical studies. J Clin Endocrinol Metab 1984;58:91–8.
22. Nussbaum SR, Zahradnik RJ, Lavigne JR, et al. High sensitive two-site immunoradiometric assay of parathyrin, and its clinical utility in evaluating patients with hypercalcemia. Clin Chem 1987;33:1364–7.
23. Delmas PD, Christiansen C, Mann KG, Price PA. Bone Gla protein (osteocalcin) assay standardization report. J Bone Miner Res 1990;5:5–11.
24. Bowers GN Jr, Brassard C, Sena SF. Measurement of ionized calcium in serum with ion-selective electrodes: a mature technology that can meet the daily service needs. Clin Chem 1986;32:1437–47.
25. Malabanan A, Veronikis IE, Holick MF. Redefining vitamin D insufficiency. Lancet 1998;351:805–6.
26. National Institutes of Health. Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults—the evidence report. Obes Res 1998;6(suppl): 51S–209S.
27. Must A, Dallal GE, Dietz WH. Reference data for obesity: 85th and 95th percentiles of body mass index (wt/ht²) and triceps skinfold thickness. Am J Clin Nutr 1991;53:839–46. (Published erratum appears in Am J Clin Nutr 1995;54:773.)
28. Williamson JD, Maloney P, Skinner EA, et al. Hematologic, biochemical, and hormonal characteristics. In: Guralnik JM, Fried LP, Simonsick EM, Kasper KD, Lafferty ME, eds. The Women's Health and Aging Study: health and social characteristics of old women with disability. Bethesda, MD: National Institute on Aging, 1995:169–75. (NIH publication no. 95-4009.)
29. Cleveland WS. Robust locally weighted regression and smoothing scatterplots. J Am Stat Assoc 1979;74:829–36.
30. Sherman SS, Hollis BW, Tobin JD. Vitamin D status and related parameters in a healthy population: the effects of age, sex, and season. J Clin Endocrinol Metab 1990;71:405–13.
31. Sowers MFR, Wallace RB, Hollis BW, Lemke JH. Parameters related to 25-OH-D levels in a population-based study of women. Am J Clin Nutr 1986;43:621–8.
32. Looker AC, Gunter EW. Hypovitaminosis D in medical inpatients. N Engl J Med 1998;339:344–5 (letter).
33. Harris SS, Dawson-Hughes B. Seasonal changes in plasma 25-hydroxyvitamin D concentrations of young American black and white women. Am J Clin Nutr 1998;67:1232–6.
34. Meier DE, Luckey MM, Wallenstein S, Lapinski RH, Catherwood B. Racial differences in pre- and postmenopausal bone homeostasis: association with bone density. J Bone Miner Res 1992;7:1181–9.
35. Perry HM III, Horowitz M, Morley JE, et al. Aging and bone metabolism in African American and Caucasian women. J Clin Endocrinol Metab 1996;81:1108–17.
36. MacLaughlin J, Holick MF. Aging decreases the capacity of human skin to produce vitamin D₃. J Clin Invest 1985;76:1536–8.
37. Lester E, Skinner RK, Wills MR. Seasonal variation in serum-25-hydroxyvitamin-D in the elderly in Britain. Lancet 1977;1:979–80.
38. Bouillon RA, Auwerx JH, Lissens WD, Pelemans WK. Vitamin D status in the elderly: seasonal substrate deficiency causes 1,25-dihydroxycholecalciferol deficiency. Am J Clin Nutr 1987;45:755–63.
39. Krall EA, Sahyoun N, Tannenbaum S, Dallal GE, Dawson-Hughes B. Effect of vitamin D intake on season variations in parathyroid hormone secretion in postmenopausal women. N Engl J Med 1989; 321:1777–83.
40. LeBoff MS, Kohlmeier L, Hurwitz S, Franklin J, Wright J, Glowacki J. Occult vitamin A deficiency in postmenopausal US women with acute hip fracture. JAMA 1999;281:1505–11.
41. Utiger RD. The need for more vitamin D. N Engl J Med 1998; 338:828–9.