Photochemistry and Photobiology, 2014, 90: 1455–1461
Sun Exposure and Vitamin D Status as Northeast Asian Migrants Become Acculturated to Life in Australia Shuyu Guo*1, Peter Gies2, Kerryn King2 and Robyn M Lucas1,3 1
National Centre for Epidemiology and Population Health, Research School of Population Health, The Australian National University, Canberra, ACT, Australia 2 Australian Radiation Protection and Nuclear Safety Agency, Melbourne, VIC, Australia 3 Telethon Kids Institute, University of Western Australia, Perth, WA, Australia Received 26 June 2014, accepted 13 September 2014, DOI: 10.1111/php.12349
ABSTRACT Vitamin D deficiency is more common in Northeast-Asian immigrants to western countries than in the local population; prevalence equalizes as immigrants adopt the host country’s culture. In a community-based study of 100 Northeast-Asian immigrants in Canberra, Australia, we examined predictors of vitamin D status, its association with indicators of acculturation (English language use; time since migration) and mediators of that association. Participants completed a sun and physical activity diary and wore an electronic ultraviolet radiation (UVR) dosimeter for 7 days. Skin colour was measured by reflectance spectrophotometry. Serum concentrations of 25-hydroxyvitamin D (25(OH)D) and cardio-metabolic biomarkers were measured on fasting blood. In a multiple linear regression model, predictors for 25(OH)D concentration were season of blood collection, vitamin D supplementation, UVR exposure, body mass index, physical activity and having private health insurance (R2 = 0.57). Greater acculturation was associated with lower risk of vitamin D deficiency (de-seasonalized 25(OH)D level 5.0 mmol L1) (AOR: 7.48 [95%CI 1.51–37.0]). Targeted public health approaches are required to manage the high prevalence of vitamin D deficiency in migrants retaining a traditional lifestyle.
to the host population, have been implicated as major risk factors for these discrepancies. Australia is among one of the most multicultural societies in the world: 27% of Australians are born overseas, and a high proportion of these are from Asia (4). As duration of residence increases, migrants tend to adopt the lifestyle of the host country, a process known as acculturation. This process changes the health risk profile, typically with increasing prevalence of obesity, hypertension and diabetes (5–7), but also higher physical activity levels (5,8) and higher vitamin D status (2). In a recent cross-sectional study in Chinese and Korean women in Sydney, Australia greater acculturation was associated with higher vitamin D status. However, lack of information on several key determinants of vitamin D status, such as skin type and body mass index, restriction of blood sampling to the spring and summer months, and the availability only of recalled time in the sun in each season, limited a detailed assessment of the determinants of vitamin D status and possible mediators of the observed association with acculturation in these migrant women. Here we extend these analyses using objective measures of skin type and sun exposure, measuring acculturation using a validated scale, in both men and women from a broader range of Northeast Asian origins living in Canberra, Australia, with data collected across a full year. We explore: (1) the determinants of vitamin D status (measured as the serum 25-hydroxyvitamin D [25(OH)D] level); (2) the association between acculturation and vitamin D status, and its potential mediators; and (3) the inter-relationships between vitamin D status, cardio-metabolic health, and acculturation.
MATERIALS AND METHODS
INTRODUCTION Vitamin D deficiency is common in migrants from Asia, the Middle East and Africa who relocate to Australia (1,2). In Australia, the most recent national health survey reported that the overall prevalence of vitamin D deficiency (serum 25-hydroxyvitamin D [25(OH)D] concentration 25 kg m2 (obese). We dichotomised the markers of cardio-metabolic Table 1. Descriptive characteristics of Northeast Asian Australian study participants and their association with 25(OH)D concentration N (%)*
25(OH)D Mean (SD) nmol L1
Age group 18–29 43 58.6 (23.6) 30–39 17 46.3 (21.2) 40–49 21 58.7 (18.9) ≥50 19 76.8 (21.0) Sex Male 46 56.2 (22.5) Female 54 56.0 (27.1) University degree or higher No 14 53.0 (30.2) Yes 86 56.6 (27.4) Private health insurance No 42 51.8 (23.1) Yes 58 65.6 (22.3) Acculturation Traditional 42 50.5 (20.5) Bicultural 32 66.2 (24.5) Acculturated 26 67.5 (22.0) Season of interview Spring 33 48.2 (13.0) Summer 21 80.3 (17.7) Autumn 29 62.5 (26.0) Winter 17 55.4 (25.7) Daily UVR exposure (SEDs)† 1.0 mmol L1. Acculturation was recorded according to the total score over the following variables: language preference: 2 = only English/mostly English, 1 = Asian language and English equally well, 0 = Asian only/mostly Asian language; location raised: 2 = in Australia only/mostly in Australia, 1 = equally in Asia and Australia, 0 = in Asia only/mostly in Asia; self-rated acculturation:
1457
2 = mostly or very westernized, 1 = bicultural, 0 = mostly or very Asian. Categorisation into three groups was as follows: Traditional (score4). Statistical analysis. Descriptive statistics (mean and standard deviation, number and proportion) were used to describe socio-demographic characteristics and relevant health behaviours. Fisher’s exact tests were used to compare characteristics of acculturation groups. For seasonally varying variables such as SEDs, physical activity and time spent outdoors, we used Cochran-Mantel-Haenszel Statistics to adjust for season. We categorised the following variables into tertiles: melanin density, fish intake and cheese intake. Smoking status was not used in the analyses as only two of the participants were current smokers. We used backwards stepwise regression and then purposeful selection of covariates (17) to generate the model for the determinants of vitamin D status, using the likelihood ratio test to assess whether addition of previously removed variables significantly improved the model fit. The relative contribution of each factor was estimated using the R2 contribution averaged over orderings among regressors calculated using the “relaimpo” package for R Software (18). Because of the small sample size, we used exact logistic regression for the remaining analyses: the association between vitamin D deficiency and acculturation level and the possible (statistical) mediators of that association; the association between cardio-metabolic risk factors and vitamin D deficiency, and between cardio-metabolic risk factors and acculturation, including the possible mediating role of vitamin D status (tested by adjusting for vitamin D status and examining the effect on the odds ratio). Adjusted odds ratios (AOR, adjusted for age, sex, university education, private health insurance, with additional adjustment for BMI, intake of supplements, duration of physical activity and time spent outdoors as indicated for specific analyses) with 95% confidence intervals (CI) were calculated. We additionally adjusted for season where seasonally varying variables (i.e. time spent outdoors, physical activity) were included as explanatory factors in the model. Except where previously noted, analyses were performed in SAS version 9.3 (SAS Institute, Inc, Cary, NC).This study was approved by the Human Research Ethics Committee of the Australian National University. All participants gave written informed consent.
RESULTS General findings
Figure 1. (A) Distribution of skin melanin density (of inner upper arm); (B) Box-plots of melanin density (of inner upper arm) by self-reported “untanned skin colour”.
Descriptive data are provided in Table 1. Of the 100 participants, the majority (96%) completed the sun dairy and wore the UVR dosimeter as instructed. Diary data were available for 694 of 700 days; dosimeter data were available for 616 days (12% of data were lost due to dosimeter malfunction). Most (94%) participants provided a blood sample.
Table 2. Predictors of 25(OH)D concentrations among Northeast Asian Australians living in Canberra, Australia Predictors Season (vs. Spring) Summer Autumn Winter Intake of vitamin D supplements in previous 4 weeks (vs. No) Yes UVR exposure (SEDs, standardized to the average ambient UVR over the previous 28 days) BMI (vs.
Sun Exposure and Vitamin D Status as Northeast Asian Migrants Become Acculturated to Life in Australia Shuyu Guo*1, Peter Gies2, Kerryn King2 and Robyn M Lucas1,3 1
National Centre for Epidemiology and Population Health, Research School of Population Health, The Australian National University, Canberra, ACT, Australia 2 Australian Radiation Protection and Nuclear Safety Agency, Melbourne, VIC, Australia 3 Telethon Kids Institute, University of Western Australia, Perth, WA, Australia Received 26 June 2014, accepted 13 September 2014, DOI: 10.1111/php.12349
ABSTRACT Vitamin D deficiency is more common in Northeast-Asian immigrants to western countries than in the local population; prevalence equalizes as immigrants adopt the host country’s culture. In a community-based study of 100 Northeast-Asian immigrants in Canberra, Australia, we examined predictors of vitamin D status, its association with indicators of acculturation (English language use; time since migration) and mediators of that association. Participants completed a sun and physical activity diary and wore an electronic ultraviolet radiation (UVR) dosimeter for 7 days. Skin colour was measured by reflectance spectrophotometry. Serum concentrations of 25-hydroxyvitamin D (25(OH)D) and cardio-metabolic biomarkers were measured on fasting blood. In a multiple linear regression model, predictors for 25(OH)D concentration were season of blood collection, vitamin D supplementation, UVR exposure, body mass index, physical activity and having private health insurance (R2 = 0.57). Greater acculturation was associated with lower risk of vitamin D deficiency (de-seasonalized 25(OH)D level 5.0 mmol L1) (AOR: 7.48 [95%CI 1.51–37.0]). Targeted public health approaches are required to manage the high prevalence of vitamin D deficiency in migrants retaining a traditional lifestyle.
to the host population, have been implicated as major risk factors for these discrepancies. Australia is among one of the most multicultural societies in the world: 27% of Australians are born overseas, and a high proportion of these are from Asia (4). As duration of residence increases, migrants tend to adopt the lifestyle of the host country, a process known as acculturation. This process changes the health risk profile, typically with increasing prevalence of obesity, hypertension and diabetes (5–7), but also higher physical activity levels (5,8) and higher vitamin D status (2). In a recent cross-sectional study in Chinese and Korean women in Sydney, Australia greater acculturation was associated with higher vitamin D status. However, lack of information on several key determinants of vitamin D status, such as skin type and body mass index, restriction of blood sampling to the spring and summer months, and the availability only of recalled time in the sun in each season, limited a detailed assessment of the determinants of vitamin D status and possible mediators of the observed association with acculturation in these migrant women. Here we extend these analyses using objective measures of skin type and sun exposure, measuring acculturation using a validated scale, in both men and women from a broader range of Northeast Asian origins living in Canberra, Australia, with data collected across a full year. We explore: (1) the determinants of vitamin D status (measured as the serum 25-hydroxyvitamin D [25(OH)D] level); (2) the association between acculturation and vitamin D status, and its potential mediators; and (3) the inter-relationships between vitamin D status, cardio-metabolic health, and acculturation.
MATERIALS AND METHODS
INTRODUCTION Vitamin D deficiency is common in migrants from Asia, the Middle East and Africa who relocate to Australia (1,2). In Australia, the most recent national health survey reported that the overall prevalence of vitamin D deficiency (serum 25-hydroxyvitamin D [25(OH)D] concentration 25 kg m2 (obese). We dichotomised the markers of cardio-metabolic Table 1. Descriptive characteristics of Northeast Asian Australian study participants and their association with 25(OH)D concentration N (%)*
25(OH)D Mean (SD) nmol L1
Age group 18–29 43 58.6 (23.6) 30–39 17 46.3 (21.2) 40–49 21 58.7 (18.9) ≥50 19 76.8 (21.0) Sex Male 46 56.2 (22.5) Female 54 56.0 (27.1) University degree or higher No 14 53.0 (30.2) Yes 86 56.6 (27.4) Private health insurance No 42 51.8 (23.1) Yes 58 65.6 (22.3) Acculturation Traditional 42 50.5 (20.5) Bicultural 32 66.2 (24.5) Acculturated 26 67.5 (22.0) Season of interview Spring 33 48.2 (13.0) Summer 21 80.3 (17.7) Autumn 29 62.5 (26.0) Winter 17 55.4 (25.7) Daily UVR exposure (SEDs)† 1.0 mmol L1. Acculturation was recorded according to the total score over the following variables: language preference: 2 = only English/mostly English, 1 = Asian language and English equally well, 0 = Asian only/mostly Asian language; location raised: 2 = in Australia only/mostly in Australia, 1 = equally in Asia and Australia, 0 = in Asia only/mostly in Asia; self-rated acculturation:
1457
2 = mostly or very westernized, 1 = bicultural, 0 = mostly or very Asian. Categorisation into three groups was as follows: Traditional (score4). Statistical analysis. Descriptive statistics (mean and standard deviation, number and proportion) were used to describe socio-demographic characteristics and relevant health behaviours. Fisher’s exact tests were used to compare characteristics of acculturation groups. For seasonally varying variables such as SEDs, physical activity and time spent outdoors, we used Cochran-Mantel-Haenszel Statistics to adjust for season. We categorised the following variables into tertiles: melanin density, fish intake and cheese intake. Smoking status was not used in the analyses as only two of the participants were current smokers. We used backwards stepwise regression and then purposeful selection of covariates (17) to generate the model for the determinants of vitamin D status, using the likelihood ratio test to assess whether addition of previously removed variables significantly improved the model fit. The relative contribution of each factor was estimated using the R2 contribution averaged over orderings among regressors calculated using the “relaimpo” package for R Software (18). Because of the small sample size, we used exact logistic regression for the remaining analyses: the association between vitamin D deficiency and acculturation level and the possible (statistical) mediators of that association; the association between cardio-metabolic risk factors and vitamin D deficiency, and between cardio-metabolic risk factors and acculturation, including the possible mediating role of vitamin D status (tested by adjusting for vitamin D status and examining the effect on the odds ratio). Adjusted odds ratios (AOR, adjusted for age, sex, university education, private health insurance, with additional adjustment for BMI, intake of supplements, duration of physical activity and time spent outdoors as indicated for specific analyses) with 95% confidence intervals (CI) were calculated. We additionally adjusted for season where seasonally varying variables (i.e. time spent outdoors, physical activity) were included as explanatory factors in the model. Except where previously noted, analyses were performed in SAS version 9.3 (SAS Institute, Inc, Cary, NC).This study was approved by the Human Research Ethics Committee of the Australian National University. All participants gave written informed consent.
RESULTS General findings
Figure 1. (A) Distribution of skin melanin density (of inner upper arm); (B) Box-plots of melanin density (of inner upper arm) by self-reported “untanned skin colour”.
Descriptive data are provided in Table 1. Of the 100 participants, the majority (96%) completed the sun dairy and wore the UVR dosimeter as instructed. Diary data were available for 694 of 700 days; dosimeter data were available for 616 days (12% of data were lost due to dosimeter malfunction). Most (94%) participants provided a blood sample.
Table 2. Predictors of 25(OH)D concentrations among Northeast Asian Australians living in Canberra, Australia Predictors Season (vs. Spring) Summer Autumn Winter Intake of vitamin D supplements in previous 4 weeks (vs. No) Yes UVR exposure (SEDs, standardized to the average ambient UVR over the previous 28 days) BMI (vs.
