Vitamin D status and health correlates among German adults



To analyze vitamin D status based on serum 25-hydroxyvitamin D (25OHD) measurements, its determinants and health correlates in a representative sample of German adults.


A total of 1763 men and 2267 women, 18- to 79-year old, who participated in the representative German National Health Interview and Examination Survey 1998 and the integrated German Nutrition Survey.


The median vitamin D intake from both diet and supplements did not meet the recommended level of 5 μg/day, in either men (2.8 μg/day) or women (2.3 μg/day). Altogether 80.9% of men and 88.5% of women had vitamin D intakes below this level. Moderate (12.5–25 nmol/l serum 25OHD levels) and mild (25–50 nmol/l) vitamin D deficiency was prevalent in the adult population in Germany, even in younger age groups. Overall, 57% of men and 58% of women had vitamin D levels below 50 nmol/l. Among 65- to 79-year-old women, the proportion amounted to 75%, even during the sunnier half of the year. In sex-specific multiple linear regression models, independent determinants of serum 25OHD levels consistently included season, vitamin D intake from both diet and supplements, physical activity and living in a partnership. In addition, age and current menopausal hormone use contributed to the model among women, as opposed to time of day of blood sampling and body mass index (marginally) among men. Significantly lower serum 25OHD levels were observed in women with hypertension, cardiovascular diseases and noninsulin-treated diabetes mellitus as well as in men with insulin-treated diabetes mellitus compared with nonaffected participants.


Vitamin D deficiency is a public health issue in Germany. We identified a number of determinants with potential for primary prevention of vitamin D deficiency. Risk and benefits of preventive actions need to be examined in further studies.


Vitamin D plays a vital role in maintenance of calcium homeostasis. Currently, in Europe and North America, rickets and osteomalacia, the clinical outcomes of a severe vitamin D deficiency, are rarely observed (Lips, 2004). Milder states of vitamin D deficiency are more common and may contribute to the risk of fragility fractures in the elderly on the basis of secondary hyperparathyroidism and high bone turnover (Avenell et al., 2005) as well as muscle weakness and an increased propensity to falls (Bischoff-Ferrari et al., 2004). Beyond these musculoskeletal effects, long-standing suboptimal vitamin D states as defined by serum 25-hydroxyvitamin D (25OHD) levels below 50 nmol/l may contribute to the pathogenesis of chronic conditions involving the immune system and low-grade inflammation, such as multiple sclerosis, type 1 and 2 diabetes, cardiovascular diseases and specific types of cancer (Chiu et al., 2004; Holick, 2004b; Zittermann et al., 2005; Christakos et al., 2006; Garland et al., 2006; Vieth and Kimball, 2006). A recent study suggests that lower serum 25OHD levels are also associated with a greater risk of future nursing home admission and possibly with mortality in older persons (Visser et al., 2006).

It is well known that vitamin D status shows a marked seasonal variation as ultraviolet-B radiation is not strong enough during winter time to assure sufficient dermal synthesis of vitamin D. Vitamin D deficiency states appear to be prevalent in populations of different latitudes and ethnic backgrounds (Mc Kenna, 1992; Ovesen et al., 2003; Hashemipour et al., 2004; Heaney, 2004; Meyer et al., 2004; Raiten and Picciano, 2004; Gannage-Yared et al., 2005; Hanley and Davision, 2005; Lucas et al., 2005; Meddeb et al., 2005; Park and Johnson, 2005; Whiting and Calvo, 2005). Specific risk factors of vitamin D deficiency mediated by lack of sunshine exposure include residence at high latitudes and in large cities, high skin pigmentation, cultural dress code or immobility (Semba et al., 2000; Hashemipour et al., 2004; Meyer et al., 2004; Meyer, 2004; Gannage-Yared et al., 2005). Insufficient dietary compensation (Calvo et al., 2005), and diseases interfering with vitamin D and calcium absorption (Lips, 2004) have a role as well.

Previous studies among the adult population in Germany showed that the average dietary vitamin D intake does not meet the recommended level (Mensink et al., 2002; DGE, 2004). However, information regarding vitamin D status and potential health correlates in Germany is scarce. Data available so far relate to vitamin D deficiency among residents of older people's homes (Mc Kenna, 1992) and to the association between serum 25OHD levels and bone mineral density or bone metabolism in a small regional sample of older adults (Scharla et al., 1996; Woitge et al., 1998). The present study aimed to assess the prevalence of vitamin D deficiency and replete states in a representative sample of the German noninstitutionalized adult population and to examine some independent determinants of vitamin D status as well as the cross-sectional relationship between serum 25OHD levels and several common chronic conditions previously related to vitamin D deficiency.

Subjects and methods


The representative German National Health Interview and Examination Survey 1998 (GNHIES) was conducted from October 1997 to March 1999. The population sample included 7124 men and women, aged 18–79 years, reflecting the noninstitutionalized adult population in Germany. Participants were selected by a random sampling method based on a two-staged clustered design, stratified by age, community size and federal state. The response rate was 61.4% (Bellach et al., 1998; Thefeld et al., 1999). The sampling procedure has been previously described in detail (Mensink and Beitz, 2004). The present analysis is based on a random subsample of 2267 women and 1763 men who also participated in an extended dietary assessment, the German Nutrition Survey 1998. As budget and time restrictions precluded to include all persons who responded to the main survey, participants were recruited consecutively according to interviewer availability without any systematic selection. As an exception, we attempted to cover the largest possible proportion of women of childbearing age assuring a sufficient number of participants for the allied folic acid study. To correct for this disproportionality as well as for discrepancies due to nonresponse and noncoverage between the net sample and the situation of the German population according to age, gender, community size and residence in West or East Germany in 1998, a weighting factor was computed (Mensink and Beitz, 2004). Surveys were approved by the federal data protection office. Respondents were informed in detail about the study objectives, interview and examination procedures as well as pseudonymized record-keeping and data analyses. All participants provided written informed consent before the interview and examination.

Data collection

Participants filled in a questionnaire on health and lifestyle issues, among them leisure-time physical activity, income, education and profession. A computer-assisted personal interview (CAPI) was conducted by specifically trained physicians to obtain a detailed medical history and information on any medication used within 7 days before the interview. To assess the lifetime diagnosis of particular health problems, including various cardiovascular conditions as well as diabetes mellitus, participants were asked ‘Has a doctor ever told you that you have one or more of the following diseases?’. Diabetes mellitus was further differentiated as noninsulin-treated and insulin-treated diabetes mellitus. Standardized measures of blood pressure, height and weight as well as blood samples were taken as part of a medical examination. Blood pressure was measured by a physician 3 times after at least 3 min rest, using a mercury sphygmomanometer (Erkameter 3000, Erka, Bad Jölz, Germany). Mean systolic and diastolic blood pressure was calculated from the second and the third measurements.

Laboratory analyses

The participants were asked to fast for at least 3 h, whereupon venous blood samples were drawn for biochemical analyses. Blood samples were immediately processed and separated. Whole blood and the larger part of serum were immediately used for analysis of hematologic parameters and blood chemistry. Extra serum was aliquoted and stored at −40°C. Total serum calcium was measured by cresolphthalein complexon method (Merck, Darmstadt, Germany) on a Mega-Analyzer. Calcium levels were corrected for serum albumin (Rustad et al., 2004). The best indicator of vitamin D status is serum 25OHD. Serum 25OHD levels and serum intact parathyroid hormone levels (iPTH) were both measured from June to September 2005 in the Epidemiological Research Laboratory of the Robert Koch-Institute, using LIAISON chemiluminescence immunoassay (CLIA) (DiaSorin Inc., Stillwater, MN, USA). Inter- and intraassay coefficients of variation for serum 25OHD were 11.7 and 9.9% and for serum iPTH 7.2 and 3.7%, respectively. The lower detection limits of the assays were 5 nmol/l for serum 25OHD and 0.106 pmol/L for serum iPTH.

Dietary assessment

Trained nutritionists interviewed the participants about their dietary habits in the past 4 weeks before the examination, using the computerized modified dietary history instrument DISHES 98 (Dietary Interview Software for Health Examination Studies) with documented validity (Mensink et al., 2001). Special tableware and drawings of food portions helped to facilitate the estimations of portion sizes. The individual nutrient intake from the information on food consumption was calculated using the German Food Consumption Table, Version 2.3 (Dehne et al., 1999). As part of the dietary interview, the participants were asked about their use of vitamin B complex, C, E, folic acid, multivitamin and mineral supplements during the last year. This included the frequency of intake as well as the corresponding brand names of the supplements. Besides, there was the possibility to provide open-ended information about the use of other vitamins, for example vitamin D. Based on the brand names, the quantities of supplement ingredients were calculated, using an updated supplement composition database, developed by the GSF – National Research Centre for Environment and Health (Beitz et al., 2002).

Operationalization of variables

Vitamin D status was categorized into groups, according to Lips as serum 25OHD levels <12.5 nmol/l, 12.5–25 nmol/l, 25–50 nmol/l and >50 nmol/l, indicating severe, moderate and mild vitamin D deficiency as well as the desirable safe reference limit, respectively (Lips, 2004). For each individual, daily intakes of vitamin D (in μg/day) and calcium (in mg/day) from both diet and supplements were calculated from dietary assessment data. In addition, supplement use of vitamin D and calcium was indicated by dichotomous variables (yes vs no). Estimates of resting energy expenditure (REE) were computed from new standard equations based on height, weight, age and sex, according to specific body mass index (BMI) groups derived from German individuals (Müller et al., 2004). Energy intake (EI, in MJ/day) from diet was calculated and the ratio between EI and REE (EI/REE) was assessed to examine the potential influence of EI on vitamin D status. For some analyses, the study group was classified into age groups of 18–34, 35–64 and 65–79 years. Information on current month of examination was categorized according to calendar seasons: spring (March–May), summer (June–August), autumn (September–November) and winter (December–February). Time of day of blood sampling was categorized as morning (0800–1300), noon/afternoon (1300–1800) and evening (1800–2200). Information on physical activity was defined using the question ‘How often do you engage in sports?’ as inactive (no engagement in sports), moderately active (1–2 h/week) and active (2 h/week). However, survey questions on physical activity were not designed to permit any further differentiation between indoor and outdoor physical activity. BMI was calculated by dividing weight (in kg) by squared height (in meters). BMI measures were categorized in four groups: underweight (<20 kg/m2), normal weight (20 to <25 kg/m2), overweight (25 to <30 kg/m2) and obese (30 kg/m2). In addition, information on social and professional status was included in the analyses as the dichotomous variables ‘outdoor working’ as well as ‘married or living with a partner’.

A lifetime diagnosis of any cardiovascular disease was based on reported lifetime diagnosis of any one or more of the following conditions: previous myocardial infarction or stroke, angina, cerebral ischemia, congestive heart failure and intermittent claudication of the leg. Hypertension was defined as 90 mm Hg diastolic or 140 mm Hg systolic blood pressures according to Joint National Committtee criteria for stage 1 and 2 hypertension (Chobanian et al., 2003). In addition, participants with current antihypertensive medication were labeled as hypertensives irrespective of blood pressure measurements. With respect to other medication use, current uses of ‘antiepileptic therapy’ as well as ‘menopausal hormone therapy’ among women were considered in the analyses as potential confounders due to their known effect on calcium and bone metabolism.

Statistical analyses

All statistical analyses were performed using the SAS-software package Version 9.1. (SAS Institute Inc., Carry, NC, USA). They were performed sex specifically, and a P-value of 0.05 was considered statistically significant based on two-sided tests. No adjustment for multiple testing was applied. The weighting factor was used for all analyses. The distribution of serum 25OHD levels was skewed and required logarithmical transformation for the multiple linear regression models. Values for seasonal variation were given as individual data points for serum 25OHD levels and serum iPTH levels. Both curves were fitted using a periodic function: F (month)=αx1+βx2+C, where x1=cos(month*π/6) and x2=sin(month*π/6). These variations were similar for both sexes and were therefore not shown separately. Unadjusted associations between serum 25OHD levels and explanatory variables were first tested using χ-square statistics, simple linear regression and Pearson's as well as Spearman's correlation analyses, suitable for the nature and distribution of the variables. Differences in serum 25OHD levels between BMI groups were tested using the Kruskal–Wallis test. In addition, a multiple linear regression model with log serum 25OHD levels as the dependent variable was constructed to identify the importance of determinants. For this purpose, we excluded participants with antiepileptic therapy (n=31) to make sure that the observed association was not biased by this prevailing treatment. Antiepileptic drugs are inducers of the cytochrome P450 system in the liver, promoting the metabolism of 25OHD to less biologically active analogs (Kulak et al., 2004; Mintzer et al., 2006). We calculated prevalence rates across sex-specific quartiles of serum 25OHD levels to examine the association between serum 25OHD levels and hypertension as well as a lifetime diagnosis of diabetes mellitus and cardiovascular disease. The association between serum 25OHD levels and these conditions was further examined in multiple logistic regression analysis. Separate models were fitted for the various dependent variables including serum 25OHD levels as a continuous variable as well as the following covariates: age (continuous), calendar season and time of day of blood sampling (ordinate variables). Reported odds ratios, 95% confidence intervals and P-values relate to increase in serum 25OHD levels per 10 nmol/l.


Characteristics of the study population are summarized in Table 1 for men and women separately. Median serum 25OHD levels were 45.2 nmol/l for men and 44.7 nmol/l for women. We observed that median vitamin D intake from both diet and supplements did not meet the recommended level, in either men (2.81 μg/day) or women (2.31 μg/day). Overall, 80.9% of men and 88.5% of women had vitamin D intakes below the recommended 5 μg/day (data not shown). Median calcium intake of both men (1181 mg/day) and women (1082 mg/day) reached the recommended 1000 mg/day. A small amount of 1.5% of men and 3.8% of women used vitamin D supplements, whereas 12.1% of men and 19.3% of women reported use of calcium supplements. There was no age gradient in Vitamin D supplement use but a smaller amount in summer compared to winter (data not shown).

Table 1 Participant characteristics by gendera

Seasonal variation in serum 25OHD levels is demonstrated in Figure 1a. There were disparities throughout the year in vitamin D status with the highest median observed in summer (56.6 nmol/l in June) and the lowest at the end of winter (33.2 nmol/l in March). Furthermore, the largest variability occurred during winter while a smaller variability was seen in summer. Seasonal declines in serum 25OHD levels were closely mirrored by increases in serum iPTH levels (Figure 1b).

Figure 1

(a) Values for seasonal variation are given as individual data points for serum 25OHD levels. The curve is fitted using a periodic function: F (month)=αx1+βx2+C, where x1=cos (month π/6) and x2=sin (month π/6), P<0.001. The median 25OHD levels were lower in winter than in summer. The lowest median was observed at the end of winter (March), the highest median in summer (June). (b) Values for seasonal variation are given as individual data points for serum iPTH levels. The curve is fitted using a periodic function: F (month)=αx1+βx2+C, where x1=cos (month π/6) and x2=sin (month π/6), P<0.001. Serum iPTH levels showed an inverse seasonal variation in comparison with serum 25OHD levels, with lowest median values in September and October.

Sex- and age-specific prevalence rates of vitamin D status for the sunnier (May–October) and sun-deprived (November–April) half of the year are shown in Table 2. In the total sample, 56.8% of men and 57.8% of women showed serum 25OHD levels of 50 nmol/l and below. Vitamin D deficiency and replete states were strongly related to season with the highest prevalence of replete states during sunnier months compared to sun-deprived months. From May to October replete vitamin D states were generally more prevalent in men than in women, whereas the reverse was true from November to April except for the highest age group. There was a strong age dependency of vitamin D status in women but not in men. Even from May–October, nearly 75% of the 65- to 79-year-old women had serum 25OHD levels below 50 nmol/l.

Table 2 Prevalence of vitamin D-deficient and replete states by gender, season and age groupa

We found several significant associations between vitamin D status and potential determinants (as shown in Table 1) in bivariate analyses. Calcium intake from both diet and supplements as well as EI/REE was not significantly correlated with serum 25OHD levels. EI/REE was strongly and inversely correlated with BMI. In both sexes, serum 25OHD levels varied significantly between different BMI groups with the lowest levels in both low and high BMI, indicating a curvilinear, U-shaped association. Box plots of serum 25OHD levels by BMI groups for men and women, respectively, showed this relationship (Figure 2a and b).

Figure 2

(a) The center in the box indicates the median. Boxes represent values within the twenty-fifth and seventy-fifth percentiles and lines represent values within the tenth and ninetieth percentiles. Serum 25OHD levels and BMI showed a curvilinear, U-shaped association with the lowest 25OHD levels observed in both low- and high-BMI groups (P=0.0034). (b) The center in the box indicates the median. Boxes represent values within the twenty-fifth and seventy-fifth percentiles and lines represent values within the tenth and ninetieth percentiles. Serum 25OHD levels and BMI showed a curvilinear, U-shaped association with the lowest 25OHD levels observed in both low- and high-BMI groups (P<0.001).

In full multiple linear regression models, season, vitamin D intake, married or living with a partner and being physically active were significantly and positively associated with serum 25OHD levels in both sexes (Table 3). In addition, age and current menopausal hormone therapy were independent determinants of serum 25OHD levels in women, whereas time of day of blood sampling independently contributed to the model in men. BMI at the lower and upper end of the distribution (U-shaped relationship as reflected by the squared term) was marginally significant in men.

Table 3 Independent determinants of log-serum 25OHD levels: multiple linear regressiona

Calcium intake did not contribute to the multiple models in either sex, estimates in Table 3 are therefore based on the models without calcium intake. Due to the strong correlation between BMI and EI/REE, the potential effect of EI/REE was examined in separate models replacing BMI by EI/REE. We observed no significant effect on serum 25OHD levels.

Men with lower serum 25OHD levels were more likely to have insulin-treated diabetes mellitus. However, the number of observations was rather small (1.1% in men and 1.2% in women). Among women, hypertension and a reported lifetime diagnosis of cardiovascular disease and noninsulin-treated diabetes mellitus were associated with lower serum 25OHD levels. Prevalence rates showed a significant stepwise increase from highest to lowest 25OHD quartiles (Table 4a). These associations persisted in multiple logistic models adjusted for age, season and time of day of blood sampling (Table 4b). In addition, the relationship between hypertension and serum 25OHD levels was also significant in men.

Table 4 Prevalence (%) of chronic diseases according to quartiles of vitamin D levelsa
Table 5 Relationship between serum 25OHD levels and chronic diseases (odds ratio and 95% confidence intervals)a


Prevalence of vitamin D deficiency and seasonal variation

We observed a high prevalence of vitamin D deficiency among German adults. Regardless of season, 15.6% of men and 17.0% of women had severe to moderate vitamin D deficiency (<25 nmol/l). Over half of the population (56.8% of men and 57.8% of women) generally had serum 25OHD levels below 50 nmol/l. Vitamin D-deficient states were most prevalent among women 65 years and older. Nearly 75% of women in this age group had serum 25OHD levels below 50 nmol/l, even during sunnier months. Nevertheless, vitamin D deficiency was prevalent among younger persons. For example depending on seasonal changes 20.5% of men and 19.1% of women, 18 to 34-year old, showed severe to moderate vitamin D deficiency during sun-deprived months.

To our knowledge, this is the first survey to examine the vitamin D status in a representative sample of the German noninstitutionalized adult population. As there is no consensus on an optimal vitamin D status (Lips, 2004; Dawson-Hughes et al., 2005; Aloia et al., 2006; Bischoff-Ferrari et al., 2006) and no standardization of vitamin D measurement (Lips et al., 1999; Binkley et al., 2004), comparison between different studies and results is rather difficult. However, our data are similar to results of other European studies concerning the prevalence of vitamin D deficiency from various countries in Europe (Burnand et al., 1992; Scharla et al., 1996; Chapuy et al., 1997; Woitge et al., 1998; Lamberg-Allardt et al., 2001; Kudlacek et al., 2003; Ovesen et al., 2003; MacFarlane et al., 2004; Meyer et al., 2004) and other continents (Sherman et al., 1990; Mc Kenna, 1992; Jacques et al., 1997; Looker et al., 2002; Rapuri et al., 2002; Hashemipour et al., 2004; Hanley and Davision, 2005), including countries located at low latitudes (Levis et al., 2005; Lucas et al., 2005). Moreover, studies considerably vary with respect to the recruitment and the age range of the included sample, and only few studies have been based on nationally representative samples (Chapuy et al., 1997; Finch et al., 1998; Looker et al., 2002; Hirani and Primatesta, 2005).

The median vitamin D intake from both diet and supplements did not meet the recommended level of 5 μg/day, in either men or women. Altogether, 80.9% of men and 88.5% of women had vitamin D intakes below the recommended level. In comparison with studies from northern countries concerning elderly women (Andersen et al., 2005), median vitamin D intakes seem to be rather small.

In accordance with previous reports from a number of regional cross-sectional studies (Scharla et al., 1996; Woitge et al., 1998; Rapuri et al., 2002; Bhattoa et al., 2004), we observed considerable seasonal changes in serum 25OHD levels, with lowest and highest median values among participants studied in late winter (March) and summer (June), respectively. Serum iPTH also showed the expected inverse seasonal variation, with lowest median values in September and October (Woitge et al., 1998; Rapuri et al., 2002). Lack of fit between seasonal variation predicted from models based on the individual iPTH measurements and the observed variation based on median values was evident between July and September. The rise in median iPTH from July to August is likely to be explained by a smaller number of study participants (since August is the typical holiday month in Germany, the fieldwork was paused for some weeks) and the coincidental relatively high proportion of elderly women participating between mid July and mid September. This may, at least in part, also explain why the variability in serum 25OHD levels was observed to be larger during winter than summer. However, there are also a number of biochemical explanations. First, it may be that people generally have a certain degree of sunlight exposure during summer, partly explaining the smaller variability. Second, a larger variability in outdoor activities during the sun-deprived season is likely, as some participants may have traveled to sunny districts or skiing regions within several weeks before the survey examination. Finally, an increase in supplement use during winter as observed in our study population is also possible.

Determinants of vitamin D status

Given the fact that vitamin D deficiency has been implicated in many disorders, it is of high public health relevance to identify risk groups. We therefore examined determinants of serum 25OHD levels. According to our observations, Jacques et al. (1997) found an inverse relationship between serum 25OHD levels and age among elderly women but not in men in the Framingham Heart Study. While there is evidence that cutaneous vitamin D biosynthesis (Holick et al., 1989) may become less efficient with advancing age, this would not explain the observed sex differences.

Clinical as well as population-based studies have consistently found that obesity is associated with lower serum 25OHD levels (Jacques et al., 1997; Wortsman et al., 2000; Snijder et al., 2006). In contrast, there is only one population-based study conducted in the framework of the Health Survey for England 2000, that has reported a significant association between low BMI (<25 kg/m2) and vitamin D status (Hirani and Primatesta, 2005). In extension of these previous reports, we observed not only a high but also a low BMI to be associated with low serum 25OHD levels. A recent study suggests that decreased bioavailability due to its deposition in body fat is responsible for low serum 25OHD levels in obese subjects (Wortsman et al., 2000). It is also possible that obese as well as very lean persons may be less exposed to sunlight as they are less often outdoor or wear more covering clothing. Residual confounding of sunlight exposure is possible, since we had to rely on proxy variables to control for UV exposure. As underweight persons may be prone to vitamin D deficiency due to inadequate food consumption, we examined the association between serum 25OHD levels and the ratio of EI to REE (EI/REE) (Wortsman et al., 2000) as measure of adequate food consumption. It is often observed that EI/REE correlates strongly with BMI (Mensink et al., 2001). Low-energy reporting as determined by a low ratio of EI/REE is positively associated with BMI. In the regression model we included this ratio to analyze the impact of low-energy reporters. Since the influence was not relevant and EI/REE showed collinearity with BMI, it was not included in the presented models.

We did not observe a significant relationship between vitamin D supplement use and serum 25OHD levels in either gender. Andersen et al. (2005) reported that vitamin D supplement use was a significant determinant of serum 25 OHD levels in postmenopausal women living in northern Europe. Our study is a representative of nationwide survey including adults from 18 to 79 years of age. In this group, the observed proportion of vitamin D supplement users was very small compared to postmenopausal women in the previous study. We can, however, not exclude misclassification due to imprecise questions concerning vitamin D supplement use.

Persons who are married or live with a partner are known to be generally healthier than singles (Joutsenniemi et al., 2006) and more active in their leisure time (Pettee et al., 2006). We therefore hypothesized that these groups may spend more time outdoors and have a more balanced diet than singles. While the observed significant and independent positive association between vitamin D status and living in a partnership does not prove any of the particular mechanisms, it is likely to serve as a proxy for differences in health-related behavior that was not captured by the constructs of ‘physical activity’ or ‘vitamin D intake’.

To exclude possible effects of diurnal variation of serum 25OHD levels, we included time of day of blood sampling in our analyses. In men, but not in women, we found a significant relationship with serum 25OHD levels, increasing from morning to noon/afternoon and strongly declining until evening. The sex difference disappeared when we confined the analyses to younger women up to the age of 50 years, which may indicate an interrelationship with sex hormone metabolism. To our knowledge, this is the first observation of diurnal variation in serum 25-OHD. Diurnal rhythm of 1,25-dihydroxyvitamin D and other markers of bone mineral metabolism was shown previously (Greenspan et al., 1997; Rejnmark et al., 2002).

Multiple regression models explained only a small fraction of the variance in vitamin D status. It is likely that this is largely attributable to difficulties in the estimation of sunlight exposure, as we had no information on time spent outdoors, traveling to sunny regions, clothing habits, use of sunscreen or grade of skin pigmentation. We could also not consider genetic factors such as vitamin D receptor gene polymorphism (Zajickova et al., 2006).

Role of diseases

Due to the increasing number of studies implicating vitamin D deficiency in the pathogenesis of various chronic disorders, we also examined the cross-sectional relationship between serum 25OHD levels and chronic conditions.

There is strong evidence for a role of hypovitaminosis D in type 1 diabetes mellitus. Hyppönen et al. (2001) found that children, studied at 1 year of age, who took a recommended dose of vitamin D during infancy, were only one-fifth likely to develop type 1 diabetes mellitus compared to those who received less than the recommended amount. We were not able to precisely distinguish between type 1 and type 2 diabetes mellitus. Based on medical history regarding the first onset of disease and insulin treatment, the vast majority of our diabetics can be assumed to have had type 2 diabetes mellitus.

An association of vitamin D with type 2 diabetes is more controversial, and may be more pronounced in insulin-treated type 2 diabetics (Scragg et al., 2004; Suzuki et al., 2006). This may explain why we observed a stronger association with insulin-treated than noninsulin-treated diabetes, at least among men. There is potential for misclassification bias due to undiagnosed diabetes, which would have weakened the association between serum 25OHD and noninsulin-treated diabetes mellitus. It is well possible that the observed sex differences result from a small number of insulin-treated diabetics as well as a higher proportion of undiagnosed disease among men than women (Scheidt-Nave et al., 1990; Rathmann et al., 2003).

Recent reviews suggest that suboptimal vitamin D status may also play a role in the development of other chronic diseases including hypertension and cardiovascular diseases (Zittermann, 2003; Holick, 2004a, 2004b). In adjusted models, hypertension was related to 25OHD in both sexes, whereas a significant association with cardiovascular disease was also restricted to women. At any rate, our study is cross sectional with the consequence that causality cannot be inferred. Only longitudinal studies are capable to identify disorders due to long-standing vitamin D deficiency.

Study limitations

Based on blood specimens and highly standardized interview methods including assessment of diet and supplement use, the German National Health Interview and Examination Survey provided a unique source of data to assess vitamin D status at the population level. Results were weighted with a factor adjusting for deviations in demographic characteristics between the study populations and official population statistics at the time of the survey. Therefore, the reported results can be extended to the noninstitutionalized adult population residing in Germany in 1998.

Our study also has several major limitations. Mainly, a detailed history of sunlight exposure was not assessed at the time of the survey. Synthesis of vitamin D in the skin under UV exposure essentially contributes to maintain vitamin D supplies, and an effect of ultraviolet radiation on serum 25OHD concentrations can be expected within 4–8 weeks of exposure (Chel et al., 1998; Lucas et al., 2005). In the present study, season, outdoor working as well as leisure-time physical activity served as proxy measures of this effect. We can therefore not exclude misclassification with possible bias of effect estimates in the regression model. A further limitation may be the possible measurement errors in the dietary assessment. While it has been shown previously that dietary vitamin D in general only contributes in a small part to vitamin D status in comparison with UV radiation, the influence may increase in winter (Ovesen et al., 2003). Our survey explicitly excluded persons who were not fluently speaking German. Thus, persons with a migrant background, who are known to be at particularly high risk of vitamin D deficiency (Meyer et al., 2004), are underrepresented.

The presence or absence of chronic disease was based on self-report (ever diagnosed by a physician), except for hypertension. Information on morbidity was obtained from various sources, including self-administered questionnaires prompting study participants with a list of conditions as well as detailed computer-assisted health interviews regarding a lifetime diagnosis by a physician, year of first onset, presence of the condition within the past 12 months or 4 weeks and current medication use. Using this information, a self-reported diagnosis of diabetes mellitus and several cardiovascular conditions have previously been found highly reliable and reasonably valid in German national health surveys (Bormann et al., 1990; Thefeld, 1999). Nevertheless, misclassification bias remains an issue, as persons who are yet undiagnosed as well as those who may have not participated due to advanced illness or hospitalization will be missed in health surveys.


Our results demonstrate that moderate and mild vitamin D deficiency affects a large proportion of the adult population in Germany, and is already prevalent in younger age groups. Specific risk groups include women 65 years of age and older, obese as well as very lean persons. Together with the observed demographic shift toward an elderly population in Germany, this is a serious public health problem. Attention should be paid to optimizing vitamin D status, in both young and elderly persons. Regular outdoor activities (while paying attention to the risk of skin cancer due to unprotected or prolonged sun exposure) should be considered as well as optimizing vitamin D intake, in particular during winter. To clarify in how far our observations have implications on health and health care responsibilities, in particular preventive actions, prospective observational and interventional studies are urgently needed.


  1. Aloia JF, Talwar SA, Pollack S, Feuerman M, Yeh JK (2006). Optimal vitamin D status and serum parathyroid hormone concentration in African American women. Am J Clin Nutr 84, 602–609.

  2. Andersen R, Molgaard C, Skovgaard LT, Brot C, Cashman KD, Chabros E et al. (2005). Teenage girls and elderly women living in northern Europe have low winter vitamin D status. Eur J Clin Nutr 59, 533–541.

  3. Avenell A, Gillespie WJ, Gillespie LD, O'Connell DL (2005). Vitamin D and vitamin D analogues for preventing fractures associated with involutional and post-menopausal osteoporosis. Cochrane Database Syst Rev 20, CD000227.

  4. Beitz R, Mensink GBM, Fischer B, Thamm M (2002). Vitamins – dietary intake and intake from dietary supplements in Germany. Eur J Clin Nutr 56, 539–545.

  5. Bellach BM, Knopf H, Thefeld W (1998). Der Bundes-Gesundheitssurvey 1997/98. Das Gesundheitswesen Sonderheft 2, 59–68.

  6. Bhattoa HP, Bettembuk P, Ganacharya S, Balogh A (2004). Prevalence and seasonal variation of hypovitaminosis D and its relationship to bone metabolism in community dwelling postmeopausal Hungarian women. Osteoporos Int 15, 447–451.

  7. Binkley N, Krueger D, Cowgill CS, Plum L, Lake E, Hansen KE et al. (2004). Assay variation confounds the diagnosis of hypovitaminosis D: a call for standardization. J Clin Endocrinol Metab 89, 3152–3157.

  8. Bischoff-Ferrari HA, Dawson-Hughes B, Willet WC, Staehelin HB, Bazemore MG, Zee RY et al. (2004). Effect of vitamin D on falls – a meta-analysis. JAMA 291, 1999–2006.

  9. Bischoff-Ferrari HA, Giovannucci E, Willet WC, Dietrich T, Dawson-Hughes B (2006). Estimation of optimal serum concentrations of 25-hydroxyvitamin D for multiple health outcomes. Am J Clin Nutr 84, 18–28.

  10. Bormann C, Hoeltz J, Hoffmeister H, Klaes L, Kreuter H, Lopez H et al. (1990). Subjektive Morbidität – Prävalenz, Reliabilität und Validität von Angaben über Herz-Kreislaufkrankheiten, Diabetes und Risikofaktoren im Nationalen Untersuchungs-und Befragungs-Survey. Beiträge des Bundesgesundheitsamtes zur Gesundheitsberichterstattung 2 4, 25–129.

  11. Burnand B, Sloutskis D, Gianoli F, Cornuz J, Rickenbach M, Paccaud F et al. (1992). Serum 25-hydroxyvitamin D: distribution and determinants in the Swiss population. Am J Clin Nutr 56, 537–542.

  12. Calvo MS, Whiting SJ, Barton CN (2005). Vitamin D intake: a global perspective of current status. J Nutr 135, 310–316.

  13. Chapuy MC, Preziosi P, Maamer M, Arnaus S, Galan P, Hercberg S et al. (1997). Prevalence of vitamin D insufficiency in an adult normal population. Osteoporos Int 7, 439–443.

  14. Chel VG, Ooms ME, Popp-Snijders C, Peavel S, Schothorst AA, Meulemans CC et al. (1998). Ultraviolet irradiation corrects vitamin D deficiency and suppresses secondary hyperparathyroidism in the elderly. J Bone Miner Res 13, 1238–1242.

  15. Chiu KC, Chu A, Go VLW, Saad MF (2004). Hypovitaminosis D is associated with insulin resistance and beta cell dysfunction. Am J Clin Nutr 79, 820–825.

  16. Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, Izzo JL et al. (2003). Seventh report of the joint national committee on prevention, detection, evaluation, and treatment of high blood pressure. Hypertension 42, 1206–1252.

  17. Christakos S, Dhawan P, Shen Q, Peng X, Benn B, Zhong Y (2006). New insights into the mechanism involved in the pleiotropic actions of 1,25 dihydroxyvitamin D3. Ann NY Acad Sci 1068, 194–203.

  18. Dawson-Hughes B, Heaney RP, Holick MF, Lips P, Meunier PJ, Vieth R (2005). Estimates of optimal vitamin D status. Osteoporos Int 16, 713–716.

  19. Dehne LI, Klemm C, Henseler G, Hermann-Kunz E (1999). The German Food Code and Nutrient Data Base (BLS 2.2). Eur J Epidemiol 15, 355–359.

  20. DGE (2004). Vitaminversorgung in Deutschland. Ernährungsumschau 51, 51–55.

  21. Finch S, Doyle W, Lowe C (1998). National Diet and Nutrition Survey: People Aged 65 years and Older. Volume 1: Report of the Diet and Nutrition Survey. The Stationery Office: London.

  22. Gannage-Yared MH, Chemali R, Sfeir C, Maalouf G, Halaby G (2005). Dietary calcium and vitamin D intake in an adult Middle Eastern population: food sources and relation to lifestyle and PTH. Int J Vitam Nutr Res 75, 281–289.

  23. Garland CF, Garland FC, Gorham ED, Lipkin M, Newmark H (2006). The role of Vitamin D in cancer prevention. Am J Public Health 96, 252–261.

  24. Greenspan SL, Dresner-Pollak R, Parker RA, London D, Ferguson L (1997). Diurnal variation of bone mineral turnover in elderly men and women. Calcif Tissue Int 60, 419–423.

  25. Hanley DA, Davision KS (2005). Vitamin D insufficiency in North America. J Nutr 135, 332–337.

  26. Hashemipour S, Larijani B, Adibi H, Javadi E, Sedaghat M, Pajouhi M (2004). Vitamin D deficiency and causative factors in the population of Teheran. BMC Public Health 4, 38.

  27. Heaney RP (2004). Functional indices of vitamin D status and ramifications of vitamin D deficiency. Am J Clin Nutr 80, 1706–1709.

  28. Hirani V, Primatesta P (2005). Vitamin D concentrations among people aged 65 years and over living in private households and institutions in England: population survey. Age and Aging 34, 485–491.

  29. Holick MF (2004a). Sunlight and vitamin D for bone health and prevention of autoimmune disease, cancers, and cardiovascular disease. Am J Clin Nutr 80, 1678–1688.

  30. Holick MF (2004b). Vitamin D: importance in the prevention of cancers, type 1 diabetes, heart disease, and osteoporosis. Am J Clin Nutr 79, 362–371.

  31. Holick MF, Matsuoka LY, Wortsman J (1989). Age, vitamin D, and solar ultraviolet. Lancet 4, 1104–1105.

  32. Hyppönen E, Läärä E, Reunanen A, Järvelin M-R, Virtanen SM (2001). Intake of vitamin D and risk of type 1 diabetes: a birth-cohort study. Lancet 358, 1500–1503.

  33. Jacques PF, Felson DT, Tucker KL, Mahnken B, Wilson PWF, Rosenberg IH (1997). Plasma 25-hydroxyvitamin D and its determinants in an elderly population sample. Am J Clin Nutr 66, 929–936.

  34. Joutsenniemi KE, Martelin TP, Koskinen SV, Martikainen PT, Härkänen TT, Luoto RM et al. (2006). Official marital status, cohabiting, and self-rated health-time trends in Finland, 1978–2001. Eur J Public Health 16, 476–483.

  35. Kudlacek S, Schneider B, Peterlik M, Leb G, Klaushofer K, Weber K (2003). Assessment of vitamin D and calcium status in healthy adult Austrians. Eur J Clin Invest 33, 323–331.

  36. Kulak CAM, Borba VZC, Bilezikian JP, Silvado CE, Paoloa L, Boguszewski CL (2004). Bone mineral density and serum levels of 25OH vitamin D in chronic users of antiepileptic drugs. Arq Neuropsiquiatr 62, 940–948.

  37. Lamberg-Allardt CJE, Outila TA, Kärkkäinen MuM, Rita HJ, Valsta LM (2001). Vitamin D deficiency and bone health in healthy adults in Finland: could this be a concern in other parts of Europe? J Bone Miner Res 16, 2066–2073.

  38. Levis S, Gomez A, Jimenez C, Veras L, Ma F, Lai S et al. (2005). Vitamin D deficiency and seasonal variation in an adult South Florida population. J Clin Endocrinol Metab 90, 1557–1562.

  39. Lips P (2004). Which circulating level of 25-hydroxyvitamin D is appropriate? J Steroid Biochem Mol Biol 89-90, 611–614.

  40. Lips P, Chapuy MC, Dawson-Hughes B, Pols HAP, Holick MF (1999). An international comparison of serum 25-hydroxyvitamin D measurements. Osteoporos Int 9, 394–397.

  41. Looker AC, Dawson-Hughes B, Calvo MS, Gunter EW, Sahyoun NR (2002). Serum 25-hydroxyvitamin D status of adolescents and adults in two seasonal subpopulations from NHANES 3. Bone 30, 771–777.

  42. Lucas JA, Bolland MJ, Grey AB (2005). Determinants of vitamin D status in older women living in a subtropical climate. Osteoporos Int 16, 1641–1648.

  43. MacFarlane GD, Sackrison JL, Body JJ, Ersfeld DL, Fenske JS, Miller AB (2004). Hypovitaminosis D in a normal, apparently healthy urban European population. J Steroid Biochem Mol Biol 89-90, 621–622.

  44. Mc Kenna MJ (1992). Differences in vitamin D status between countries in young adults and the elderly. Am J Med 93, 69–77.

  45. Meddeb N, Sahli H, Chahed M (2005). Vitamin D deficiency in Tunesia. Osteoporos Int 16, 180–183.

  46. Mensink GBM, Beitz R (2004). Food and nutrient intake in East and West Germany, 8 years after the reunification – the German Nutrition Survey 1998. Eur J Clin Nutr 58, 1000–1010.

  47. Mensink GBM, Burger M, Beitz R, Henschel Y, Hintzpeter B (2002). Was essen wir heute? Ernährungsverhalten in Deutschland. Gesundheitsberichterstattung des Bundes, Berlin: Robert Koch-Institut.

  48. Mensink GBM, Haftenberger M, Thamm M (2001). Validity of DISHES 98, a computerized dietary history interview: energy and micronutrient intake. Eur J Clin Nutr 55, 409–417.

  49. Meyer C (2004). Scientists probe role of Vitamin D: deficiency a significant problem, experts say. JAMA 292, 1416–1418.

  50. Meyer HE, Falch JA, Sogaard AJ, Haug E (2004). Vitamin D deficiency and secondary hyperparathyroidism and the association with bone mineral density in persons with Pakistani and Norwegian background in Oslo, Norway, The Oslo Health Study. Bone 35, 412–417.

  51. Mintzer S, Boppana P, Toguri J, DeSantis A (2006). Vitamin D levels and bone turnover in epilepsy patients taking carbamazepine or oxcarbazepine. Epilepsia 47, 510–515.

  52. Müller MJ, Bosy-Westphal A, Klaus S, Kreymann G, Lührmann PM, Neuhäuser-Berthold M et al. (2004). World Health Organization equations have shortcomings for predicting resting energy expenditure in persons from a modern, affluent population: generation of a new reference standard from a retrospective analysis of a German database of resting energy expenditure. Am J Clin Nutr 80, 1379–1390.

  53. Ovesen L, Andersen R, Jakobsen J (2003). Geographical differences in vitamin D status, with particular reference to European countries. Proc Nutr Soc 62, 813–821.

  54. Park S, Johnson MA (2005). Living in low-latitude regions in the United States does not prevent poor vitamin D status. Nutr Rev 63, 203–209.

  55. Pettee KK, Brach JS, Kriska AM, Boudreau R, Richardson CR, Colbert LH et al. (2006). Influence of marital status on physical activity levels among older adults. Med Sci Sports Exerc 38, 541–546.

  56. Raiten DJ, Picciano MF (2004). Vitamin D and health in the 21st century: bone and beyond. Executive summary. Am J Clin Nutr 80, 1673–1677.

  57. Rapuri PB, Kinyamu HK, Gallagher JC, Haynatzka V (2002). Seasonal changes in calciotropic hormones, bone markers, and bone mineral density in elderly women. J Clin Endocrinol Metab 87, 2024–2032.

  58. Rathmann W, Haaster B, Icks A, Lowel H, Meisinger C, Holle R (2003). High prevalence of undiagnosed diabetes mellitus in Southern Germany: target populations for efficient screening. The KORA Survey 2000. Diabetologia 46, 182–189.

  59. Rejnmark L, Lauridsen AL, Vestergaard P, Heickendorff L, Andreasen F, Mosekilde L (2002). Diurnal rhythm of plasma 1,25-dihydroxyvitamin D and vitamin D-binding protein in postmenopausal women: relationship to plasma parathyroid hormone and calcium and phosphate metabolism. Eur J Endocrinol 146, 635–642.

  60. Rustad P, Simonsson P, Felding P, Pedersen M (2004). Nordic Reference Interval Project Bio-bank and Database (NOBIDA): a source for future estimation and retrospective evaluation of reference intervals. Scand J Clin Lab Invest 64, 431–438.

  61. Scharla SH, Scheidt-Nave C, Leidig G, Woitge HW, Seibel MJ, Ziegler R (1996). Lower serum 25-hydroxyvitamin D is associated with increased bone resorption marker and lower bone density at the proximal femur in normal females: A population-based study. Exp Clin Endocrinol Diabetes 104, 289–292.

  62. Scheidt-Nave C, Barrett-Connor E, Wingard D (1990). Resting ECG abnormalities suggestive of asymptomatic ischemic heart disease associated with non-insulin-dependent diabetes mellitus in a defined population. Circulation 81, 899–906.

  63. Scragg R, Sowers MF, Bell C (2004). Serum 25-hydroxyvitamin D, diabetes, and ethnicity in the Third National Health and Nutrition Examination Survey. Diabetes Care 27, 2813–2818.

  64. Semba RD, Garret E, Johnson BA, Guralnik JM, Fried LP (2000). Vitamin D deficiency among older women with and without disability. Am J Clin Nutr 72, 1529–1534.

  65. Sherman SS, Hollis BW, Tobin JD (1990). Vitamin D status and related parameters in a healthy population: the effects of age, sex, and season. J Clin Endocrinol Metab 71, 405–413.

  66. Snijder MB, van Schoor NM, Pluijm SM, van Damm RM, Visser M, Lips P (2006). Vitamin D status in relation to one-year risk of recurrent falling in older men and women. J Clin Endocrinol Metab 91, 2980–2985.

  67. Suzuki A, Kotake M, Ono Y, Kato T, Oda N, Hayakawa N et al. (2006). Hypovitaminosis D in type 2 diabetes mellitus: association with microvascular complications and type of treatment. Endocr J 53, 503–510.

  68. Thefeld W (1999). Prävalenz des Diabetes mellitus in der erwachsenen Bevölkerung Deutschlands. Das Gesundheitswesen Sonderheft 2, 85–89.

  69. Thefeld W, Stolzenberg H, Bellach BM (1999). Bundes-Gesundheitssurvey: Response, Zusammensetzung der Teilnehmer und Non-Responder-Analyse. Das Gesundheitswesen Sonderheft 2, 57–61.

  70. Vieth R, Kimball S (2006). Vitamin D in congestive heart failure. Am J Clin Nutr 83, 731–732.

  71. Visser M, Deeg DJH, Puts MTE, Seidell JC, Lips P (2006). Low serum concentrations of 25-hydroxyvitamin D in older persons and the risk of nursing home admission. Am J Clin Nutr 84, 616–622.

  72. Whiting SJ, Calvo MS (2005). Dietary recommendations to meet both endocrine and autocrine needs of vitamin D. J Steroid Biochem Mol Biol 97, 7–12.

  73. Woitge HW, Scheidt-Nave C, Kissling C, Leidig-Bruckner G, Meyer K, Grauer A et al. (1998). Seasonal variation of biochemical indexes of bone turnover: results of a population-based study. J Clin Endocrinol Metab 83, 68–75.

  74. Wortsman J, Matsuoka LY, Chen TC, Lu Z, Holick MF (2000). Decreased bioavailability of vitamin D in obesity. Am J Clin Nutr 72, 690–693.

  75. Zajickova K, Hill M, Vankova H, Zofkova I (2006). Low-density lipoprotein receptor-related protein 5 and vitamin D receptor gene polymorphisms in relation to vitamin D levels in menopause. Clin Chem Lab Med 44, 1066–1069.

  76. Zittermann A (2003). Vitamin D in preventive medicine: are we ignoring the evidence? Br J Nutr 89, 552–572.

  77. Zittermann A, Schleithoff SS, Koerfer R (2005). Putting cardiovascular disease and vitamin D insufficiency into perspective. Br J Nutr 94, 483–492.

Download references

Author information

Correspondence to B Hintzpeter.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Hintzpeter, B., Mensink, G., Thierfelder, W. et al. Vitamin D status and health correlates among German adults. Eur J Clin Nutr 62, 1079–1089 (2008).

Download citation


  • vitamin D
  • deficiency
  • serum 25OHD level
  • epidemiology
  • population survey
  • Germany

Further reading