Serum vitamin D levels in Berliners of Turkish descent –a cross-sectional study

Vitamin D is a steroid hormone with regulatory effects on the calcium and phosphate homeostasis, parathormone (PTH) and thyroxine secretion [1]. Vitamin D also affects the cardiovascular system, diabetes, the immune system, cancer, and overall mortality [2–4]. It is usually measured in serum as 25(OH)-D_3. Levels > 75 nmol/L have been defined as vitamin D sufficiency, while 50–75 nmol/L have been considered as vitamin D insufficiency and < 50 nmol/L as vitamin D deficiency [5]. Vitamin D deficiency over a longer period can cause a softening of the bones, termed rickets in children and osteomalacia in adults. Less severe vitamin D insufficiency may lead to an insufficient mineralization of the bone due to an increase of PTH, which can cause osteoporosis and fractures, muscle weakness and muscle pain [2, 3].

The main source of vitamin D is sunlight exposure inducing vitamin D production in the skin through ultraviolet B radiation (UVBR). Only a few foods such as fatty fish or cod liver oil contain vitamin D in a relevant amount. Vitamin D can be ingested as supplement, and in some countries foods are fortified with vitamin D [6]. The main risk factor for a poor vitamin D status is therefore a lack of exposure to sunlight. This can be caused either by sun avoidance, sunscreen, clothing habits, or an indoor lifestyle. Latitude, season and time of the day influence the UVBR [6].

Vitamin D has become a worldwide research issue and vitamin D deficiency has been described as pandemic by several authors; the lowest population-based vitamin D levels have been assessed in Asian countries such as China, Mongolia, India and the Middle East [6–9]. In Germany, a high prevalence of vitamin D deficiency, especially in winter, has been described [10, 11]. Since Germany is at high latitude (47–55° north, Berlin: 52° 31′ N, 13° 24′ E), sun exposure from October to March is not sufficient to maintain an adequate vitamin D supply leading to decreased 25(OH) D levels during winter if no supplements are used [12]. For subjects with darker skin pigmentation, sun-avoiding behavior or a long clothing style this geographical location can be an aggravating factor [13–15]. Already in 1961, Dunnigan et al. described cases of rickets and osteomalacia in the Pakistani Community in Glasgow, Scotland [16]. In the 1970s, the first German case reports about osteomalacia in Turkish guest workers in Germany as consequence of a vitamin D deficiency were published [17, 18].

It is known that non-western immigrants migrating to a country with limited sun exposure are at higher risk for vitamin D deficiency, however there are still considerable knowledge gaps regarding factors other than darker skin or veiling behavior [15, 19–23]. The aim of the present study was therefore to examine and compare the vitamin D status and potentially associated sociodemographic, lifestyle and medical factors in Berliners with and without Turkish roots.

Our study showed that vitamin D levels were significantly lower among Berliners of Turkish descent, and especially in veiled women, compared to a control sample of the local Berlin population without Turkish roots.

This observation is in line with previous studies, where the proportion of low vitamin D levels and vitamin D deficiency in children, adolescents and adults with migration background was found to be considerably higher than in the existing population of the country that they migrated to [15, 18, 30–32]. Erkal et al. reported average 25(OH)D-values of 38.1 nmol/L for persons with Turkish background and 68.5 nmol/L for a German control group [15], while our results, which were even lower, confirm this gap. If differentiated in low (< 50 nmol/L) and very low (< 25 nmol/L) vitamin D concentrations, our results emphasize that migrant populations show more often very low vitamin D levels (often defined as severe deficiency) than the general local non-migrant population as also reported by Tarner et al. [21].

Besides the known risk factor of darker skin among many migrants with Turkish background, the fact that Berlin is a metropolitan area with limited access to green spaces in large parts of the inner city, could also be a reason for the high percentage of low and very low vitamin D levels [19]. Our results are in line with several studies showing that wearing long clothing is an important risk factor for low vitamin D levels among Turkish women in Turkey [13, 33] and especially in northern and central Europe, as shown in a previous German study that was conducted in the Ruhr region [34].

Other risk factors besides migration background also play a role in vitamin D levels. However, study results regarding the determinants vary [19]. When comparing our results with previously published studies, only some of the determinants had similar effects on vitamin D levels, namely low physical activity, high BMI, and sex. This was shown by other authors, who also found that education and age were associated with low vitamin D levels, which is in contrast to our findings [10, 35, 36]. We found relatively similar associations in both study samples. Among both samples, a high BMI was significantly associated with lower vitamin D levels, which can partly be explained by the repository effects of adipose tissue. A further explanation for reduced vitamin D levels is the higher sedentary behavior and more indoor activities among persons with higher BMI [37]. In the Turkish sample, a low physical activity was associated with lower vitamin D levels, which is in line with other study results; physical activity has been shown to be associated with vitamin D levels, the mechanism however is not clear. While some studies interpret physical activity as a proxy of sunlight exposure, other studies report effects of physical activity irrespective of whether it is performed indoors or outdoors [35, 37]. A factor that was associated with vitamin D in the unselected sample was smoking. In contrast to other studies, we observed a higher percentage of vitamin D deficiency, defined as concentrations <50nmo/L, in non- or ex-smokers than in smokers. This association may be explained partly by the fact that smokers often go outside to smoke (and are therefore more exposed to sunlight) since the introduction of a smoking ban that prohibits smoking inside public buildings [35]. In the linear regression analyses however, we did not find an association of smoking with the vitamin D concentration.

Only in the unselected sample was female sex associated with lower vitamin D levels. This is in contrast to other German studies, where no sex-specific differences were observed [10, 22]. Although the findings of the rather small sample should not be over interpreted, one possible explanation could be more distinct lifestyle differences between men and women in that sample than in the Turkish sample. For instance men engaged more often in physical activity and were more often smokers. Previous studies showed correlations between vitamin D deficiency and cardiovascular diseases [3, 38]. We assessed self-reported diagnoses of several major chronic diseases but were not able to confirm any associations between cardiovascular or other chronic diseases such as diabetes or osteoporosis with the vitamin D concentration.

One of the strengths of our study was the considerable effort including home visits that was made to recruit participants with a Turkish migration background. We reached a participation rate of over 10% which is much higher than in other population-based studies without any sub-group-specific activities to increase the response [25, 39]. Another strength was the use of stringent epidemiological assessment methods by using questions from the national German health surveys with translations into Turkish. Furthermore, the bilingual questionnaires and staff in the recruitment office, during home visits and in the study center contributed to the relative recruitment success by allowing participants to use the language they felt most comfortable with, thus increasing general acceptance and trust in the aims of the study.

However, certain limitations of our study need to be considered such as the relatively small control sample of unselected Berliners without Turkish roots, which may have reduced the statistical power for rather rare exposures. Another limitation is that “home leave” was not assessed among the Turkish sample. Holidays in a sunny country could have increased the annual mean vitamin D concentration and low vitamin D levels during winter could have been compensated by that. Further, no eating habits were assessed and therefore nutritional information of the participants are lacking. A third limitation is the relatively short assessment period which allowed us to collect blood only in winter, spring and early summer where vitamin D levels are generally lower than at the end of summer. So we cannot present the course of vitamin D levels over a full year and all 4 seasons. We observed small seasonal changes with the lowest vitamin D concentrations from January through March which is in line with another German study of Turkish migrants [40]. However, these seasonal changes during the assessment period are in our opinion unlikely to have produced a seasonal bias or distorted our study results.

Serum vitamin D levels were considerably low in Berliners of Turkish descent, and especially in veiled women. Factors associated with low vitamin D levels were high BMI and low physical activity in the sample with Turkish background and high BMI and female sex in the unselected sample.

Possible future public health actions include weight reduction programs and physical activity promotion for the development of preventive strategies regarding a sufficient vitamin D status, especially in Berliners with Turkish background. Further research is needed to identify other potential public health strategies.