Overweight and Obesity among Russian, Somali, and Kurdish Origin Populations in Finland

Overweight and Obesity among Russian, Somali, and Kurdish Origin Populations in Finland NATALIA SKOGBERG1 ADAM ADAM1 TARJA KINNUNEN2 EERO LILJA1 ANU E. CASTANEDA1,3 1 National Institute for Health and Welfare (THL), Helsinki, Finland 2 University of Tampere, Finland 3 University of Helsinki, Finland Abstract Previous studies have shown the prevalence of overweight and obesity to be higher in foreign-born populations than their native counterparts. This study aimed at assessing the prevalence of overweight and obesity in Russian, Somali and Kurdish origin populations in Finland in comparison with the general population, and determining which background factors are associated with overweight and obesity. Data from the Finnish Migrant Health and Wellbeing Study (Maamu) were used. Prevalence of overweight and obesity was as high as 73% in Somali and 65% in Kurdish origin women. In contrast, Somali origin men had a significantly lower prevalence of overweight and obesity than men in the general population. Health promotion measures should aim at decreasing the prevalence of overweight and obesity particularly among Somali and Kurdish origin women. Foreign-born people need to be taken into account when planning and implementing national obesity prevention programs. Keywords: Overweight; Obesity; Migrant; Population-based study; Survey; Maamu 74 Introduction The alarming increase in the prevalence of overweight and obesity in the last decades has made them a global public health problem that threatens peoples’ health and wellbeing universally (Caballero 2007; Flegal et al 2012; Groves 2006). Overweight and obesity are associated with older age, female sex, lower socioeconomic status, poor diet and low levels of physical activity (Everson et al. 2002; Lahti-Koski et al. 2000; WHO 2015). Likewise, parity has been repeatedly demonstrated to be associated with overweight and obesity among women (Davis et al. 2009; Kim, Stein and Martorell 2007). Several studies have demonstrated that the prevalence of overweight and obesity tends to be higher among migrant origin populations than the general population in the country of migration (Gele and Mbalilaki 2013; Oza-Frank and Cunningham 2010; Sussner et al. 2008). Previous studies also indicate that prevalence of overweight and obesity among migrants increases with longer length of stay in the host country (Ro and Bostean 2015; Singh et al. 2011). Some studies indicate that migrants’ high obesity rates can be attributed to lifestyle and environmental changes (Delavari et al. 2013; Franzen and Smith 2009). A change from a traditional diet to a more industrial one, accompanied with decreased energy expenditure due to the urbanization, have been suggested to be the contributing factors to the observed high prevalence rates of overweight and obesity among migrant origin populations (Toselli et al. 2014). Hence, some studies emphasise that acculturation plays a key role in explaining why migrants living in high-income countries have a high prevalence of overweight and obesity. For example, Toselli et al. noted that migrants’ disease patterns change rapidly after migration and become similar to the disease pattern of the inhabitants of the host country. A recent health examination survey in Finland, FINRISK 2012, illustrated that the prevalence of overweight and obesity in Finland has increased among men since the 1970s and among women since the 1980s (Männistö et al. 2015). Prevalence of overweight or obesity was 66% in men and 46% in women. Among men and women in Finland, one in five was obese. High prevalence of obesity among Somali and Kurdish origin women aged 30 years and older living in Finland has been previously reported (Skogberg et al. 2016). Despite these findings, overweight and obesity among foreign origin populations in Finland has not yet been studied extensively. The foreign origin population in Finland is increasing and it is projected that migrants will constitute 10% of the population within the next few years (Ministry of the Interior 2016). As the share of foreign origin population is growing, there is an enormous need to study the health, including overweight and obesity, among migrants residing in Finland. Despite extensive information on the high prevalence rates of overweight and obesity among some migrant origin population on an international level, there is insufficient information on factors associated with overweight and obesity among different migrant groups living in the Nordic countries. 75 Therefore, we firstly aim to compare the prevalence of overweight and obesity in adult foreign-born populations of Russian, Somali and Kurdish origin in Finland to that in the general Finnish population. Secondly, we aim to determine which background factors (age, gender, education, employment status, years lived in Finland, physical activity, diet, and parity) are associated with overweight and obesity in migrants of Russian, Somali and Kurdish origin and the Finnish general population. Findings of this study will facilitate planning of health interventions and obesity prevention programs that are targeted to meet the needs of these migrant origin populations in Finland. Methods Design and study populations The present study uses data from the cross-sectional Migrant Health and Wellbeing Study (Maamu), conducted by the National Institute for Health and Welfare (THL) in Finland between 2010 and 2012. Detailed information about the methods of the Maamu Study can be found elsewhere (Castaneda et al. submitted for publication). Briefly, a random sample of 3,000 persons of Russian, Somali and Kurdish origin (1,000 for each migrant group) was drawn from the National Population Registry. Inclusion criteria were: age (18–64 years), country of birth (former Soviet Union or Russia, Iraq or Iran and Somalia), mother tongue (Russian or Finnish for the Russian origin group and Kurdish for the Kurdish origin group), city of residence (Helsinki, Vantaa, Espoo, Turku, Tampere and Vaasa) and minimum one year of residence in Finland. Mother tongue (Kurdish) was used as an additional criteria to identify specifically the Kurdish origin population originating from Iran and Iraq. Similarly, Russian or Finnish language was used to identify the Russian-speaking population originating from the broad geographical region of the former Soviet Union and Ingrian Finns. The Maamu Study consisted of a face-to-face interview and a health examination, conducted by trained multilingual fieldwork personnel. A brief interview was available for those not able to participate in the full study. Participation rate in at least one part of the study was 70% for Russian, 51% for Somali and 63% for the Kurdish group. The sample of the present study is limited to 464 Russian (46%), 372 Somali (37%) and 504 Kurdish (50%) participants who took part in the health examination and the face-to-face interview (or the brief interview) of the study and thus for whom anthropometric measurement were available. Reference data for the general population in Finland were obtained from the Health 2011 Survey (Koskinen, Lundqvist and Ristiluoma 2012), conducted by the National Institute for Health and Welfare between 2011 and 2012. Health 2011 Survey followed a similar standardized protocol as the Maamu Study. The reference group of the present study (n=953) included participants of the same age range and living in the same cities as in the Maamu study. Both the Maamu Study and the Health 2011 Survey were ap- 76 proved by the Coordinating Ethical Committee of the Helsinki and Uusimaa Hospital Region, Finland. Participants gave written informed consent. Measures Standardised measurements of weight and height were taken during the health examination. For the measurements, participants were asked to remove their outer clothing, including shoes. Height was measured standing upright and looking straight ahead with the Seca 213 stand-alone stadiometer. Weigh was measured wearing light clothing with the Seca 709 beam scale. In the Health 2011 survey, weight was measured as part of the bioimpedance body composition analysis (Seca 514). Weight was not measured if the participant was over 20 weeks pregnant. Body mass index (BMI) was calculated based on measured weight and height kg/m2 and classified according to WHO recommendations: underweight (≤18.5 kg/m2), normal weight (18.5–24.9 kg/m2), overweight (25.0–29.9 kg/m2), obesity (30.0–34.9 kg/m2), severe obesity (35.0–39.9 kg/m2) and morbid obesity (≥40 kg/m2). In this study, also a dichotomized variable of BMI for overweight and obesity was used, with the cut-off of ≥25 kg/m2. Taylor et al. (2006) have studied the relationship between selfreported and clinical measurements for height and weight in adults and concluded that prevalence estimates obtained by self-reports are likely to be underestimations for weight and overestimations for height. This seemed to be true also in the sample of the present study. Therefore, BMI was calculated only based on measured data even though self-report data was also gathered in the face-to-face interview. The interview included socio-demographic and health-related questions, including education, occupation, years of stay in Finland, physical activity, and parity. These variables were selected based on literature demonstrating their associations with overweight and obesity used to examine factors associated with overweight and obesity among the studied groups. Statistical analysis Analyses were conducted using SAS 9.3/SUDAAN 11.0.1 software. All analyses were stratified by sex and conducted using inverse probability weights (IPWs). IPWs were calculated to reduce bias due to non-response. Age-adjusted prevalence rates of overweight and obesity in the studied population groups were calculated by sex using predictive margins. The association between overweight and obesity with background factors were calculated using logistic regression analysis. Statistical significance was calculated using the Satterthwaite F-statistic. The association between population group and overweight and obesity were examined adjusting for age and for socio-demographic variables considered to be confounding variables. P-value of <0.05 was considered statistically significant. Results are presented as age-adjusted prevalence rates or odds ratios (OR) with 95% confidence intervals (CI). 77 Results Descriptive characteristics of the study groups are presented in Table 1. Nearly half of the Somali origin women (40%) reported no formal education. A large proportion of persons of migrant origin were unemployed (24–30%). Most of the study participants had lived in Finland for more than five years. Only a small proportion of men and women of Somali origin reported consuming fresh fruits and vegetables daily (0.2– 2%). As many as 65% of the Somali origin women and 41% of the Kurdish origin women had given birth three times or more, whereas the respective prevalence was 12% for women belonging to the general Finnish population and 8% for women of Russian origin. More than half (54%) of women belonging to the general Finnish population had never given birth. Table1. Age-adjusted characteristics of the study groups (%)1. Men Women Russian Somali Kurdish General Russian Somali Kurdish General population population N=168 N=155 N=277 N=923 N=300 N=222 N=238 N=1163 Age1 18–29 years 35.0 43.5 37.2 25.8 26.7 36.6 30.6 35.7 30–44 years 34.7 38.9 41.1 31.9 28.9 41.1 47.8 28.8 45–64 years 30.2 17.6 21.6 42.3 44.5 22.4 21.6 35.5 Education None 0.0 13.2 7.9 0.0 0.0 40.2 20.2 0.0 Primary school 25.0 43.2 49.7 50.3 15.9 45.2 40.3 31.4 High school 74.9 43.7 42.4 49.7 84.1 14.6 39.4 68.6 Employment status Employed 53.7 30.6 45.3 66.5 45.5 18.0 29.6 61.9 Student 17.8 25.7 17.7 19.8 19.6 22.4 25.4 22.9 Domestic work 0.4 0.0 0.9 0.1 9.7 32.2 13.1 4.6 Unemployed 26.6 32.9 30.1 4.5 23.5 24.3 27.1 3.4 Pensioner 0.1 10.9 6.0 9.0 1.8 3.1 4.8 7.3 Time lived in Finland ≤5 years 22.3 21.5 21.7 NA223.2 21.1 17.0 NA2 6–14 years 47.2 36.9 53.0 38.0 39.7 54.7 ≥15 years 30.5 41.6 25.2 38.8 39.2 28.0 Physical activity Low 27.4 19.2 36.8 23.3 33.2 18.7 40.7 25.2 Moderate 36.0 63.0 32.1 37.6 37.8 63.1 41.1 44.7 High 36.6 17.8 31.1 39.1 29.1 18.3 18.2 30.0 Vegetables or fruits in diet (days per week) <1 day 3.0 3.0 5.6 1.6 3.4 16.6 4.6 1.6 1–2 days 16.0 92.2 27.4 11.4 10.6 78.8 21.1 10.3 3–5 days 25.0 4.6 27.7 27.3 24.4 2.5 22.18 26.6 6–7 days 56.0 0.2 39.9 59.7 61.5 2.0 51.5 61.4 Parity 0 births 33.2 16.5 22.5 53.5 1–2 births 58.8 18.1 36.9 34.5 ≥3 births 8.0 65.4 40.6 12.0 1Unadjusted NA2= Data not available BMI distributions among the study groups are presented in Figure 1. Adjusted odds ratios for overweight and obesity by migrant group and sex in comparison with the general Finnish population are presented in Table 2. Age-adjusted prevalence of overweight and obesity was higher among Somali (73.2%; OR 4.31; 95% CI 2.57–7.23) and Kurdish origin women (65.4%; OR 2.86; 95% CI 1.77–4.62) compared with women in the general Finnish population (43.4%). In contrast, Somali origin men (37.7%; OR 0.41; 95% CI 0.24–0.72) had a lower prevalence of overweight and obesity compared with men in the general Finnish population (56.9%). Prevalence of overweight and obesity did not differ significantly between Russian and Kurdish origin men and Russian origin women compared with the general Finnish population. Differences in the prevalence of overweight and obesity among Somali and Kurdish origin women remained statistically significant following additional adjustment for socio-demographic factors (education, employment status; Model 2) and lifestyle factors (physical activity, diet; Models 3–4). Age-adjusted differences remained significant also when adjusted for parity alone (Model 5). Adjustment for age, parity and socio-demographic factors, on the other hand, accounted for the differences in the prevalence of overweight and obesity among Somali (OR 1.45; 95% CI 0.82–2.54), but not among Kurdish origin women (OR 1.74; 95% CI 1.11–2.74). Figure 1. BMI distributions by population groups and gender (%)1. 1Age-adjusted and weighted prevalences *** Statistically significant difference in BMI distribution between the study population and the general population 78 79 Tables 3 and 4 present age-adjusted odds ratios for overweight and obesity according to background characteristics within migrant groups in men and women respectively. Prevalence of overweight and obesity increased with age among women in all of the study groups. Among men, older age was associated with overweight and obesity among Russian and Kurdish origin men and the association was borderline statistically significant among men in the general Finnish population. In Somali men, age was not associated with overweight and obesity. Lower education was associated with overweight and obesity in women belonging to the general population, but not in migrant origin women. No association between age and education was found among men. In Kurdish women, unemployment and economic inactivity were associated with overweight and obesity. In contrast, employment was associated with a higher prevalence of overweight and obesity than unemployment among Russian origin men and economical inactivity among Kurdish origin men and men belonging to the general Finnish population. Table 2. Adjusted odds ratios for overweight and obesity by migrant group and sex in comparison with the general Finnish population. Russian Somali Kurdish General population OR (95% CI) OR (95% CI) OR (95% CI) OR (95% CI) MEN Model 1: Age 0.75 (0.45–1.26) 0.41 (0.24–0.72) 1.37 (0.86–2.20) 1.00 Model 2: Age, education, employment status 0.67 (0.42–1.07) 0.41 (0.24–0.71) 1.33 (0.87–2.05) 1.00 Model 3: Age, education, employment status, physical activity 0.66 (0.42–1.06) 0.40 (0.23–0.70) 1.28 (0.83–1.98) 1.00 Model 4: Age, education, employment status, diet 0.67 (0.42–1.06) 0.39 (0.22–0.70) 1.37 (0.89–2.10) 1.00 WOMEN Model 1: Age 0.90 (0.60–1.34) 4.31 (2.57–7.23) 2.86 (1.77–4.62) 1.00 Model 2: Age, education, employment status 0.85 (0.59–1.22) 2.88 (1.77–4.69) 2.20 (1.45–3.35) 1.00 Model 3: Age, education, employment status, physical activity 0.82 (0.57–1.18) 3.21 (1.90–5.42) 1.96 (1.29–2.96) 1.00 Model 4: Age, education, employment status, diet 0.82 (0.57–1.19) 2.11 (1.24–3.57) 2.16 (1.40–3.34) 1.00 Model 5: Age and parity 0.86 (0.57–1.29) 1.90 (1.06–3.40) 2.10 (1.27–3.47) 1.00 Model 6: Age, education, employment status, parity 0.83 (0.57–1.19) 1.45 (0.82–2.54) 1.74 (1.11–2.74) 1.00 OR = Odds ratio 95% CI = 95% confidence interval 80 Somali origin men who had lived in Finland for 6–14 years were more likely to be overweight or obese than those who had lived in Finland for longer. Low levels of physical activity were associated with overweight and obesity among Russian origin women and women belonging to the general Finnish population. No statistically significant associations between physical activity levels and overweight and obesity were found among men. Eating fresh fruits and vegetables less than daily was associated with Table 3. Age-adjusted odds ratios for overweight and obesity according to background characteristics within migrant groups in men. Russian Somali Kurdish General population OR (95% CI) OR (95% CI) OR (95% CI) OR (95% CI) Age1 18–39 years 1.00 1.00 1.00 1.00 40–64 years 4.48 (2.12–9.48) 1.28 (0.59–2.78) 4.03 (2.32–6.97) 2.00 (0.99–4.04) Education Less than primary school NA2 1.54 (0.54–4.38) 0.63 (0.23–1.73) NA2 Primary school 0.70 (0.28–1.76) 0.68 (0.25–1.83) 1.04 (0.60–1.79) 1.49 (0.86–2.59) High school 1.00 1.00 1.00 1.00 Employment status Economically inactive30.35 (0.10–1.16) 0.42 (0.12–1.40) 0.50 (0.27–0.91) 0.32 (0.13–0.78) Unemployed 0.33 (0.13–0.81) 1.82 (0.60–5.45) 1.24 (0.65–2.35) 0.82 (0.28–2.42) Employed 1.00 1.00 1.00 1.00 Time lived in Finland ≤5 years 1.93 (0.68–5.49) 1.32 (0.41–4.21) 1.00 (0.47–2.10) NA4 6–14 years 1.68 (0.65–4.32) 3.00 (1.09–8.20) 0.88 (0.48–1.63) ≥15 years 1.00 1.00 1.00 Physical activity Low 2.53 (0.95–6.75) 0.43 (0.13–1.39) 1.04 (0.56–1.95) 0.95 (0.37–2.44) Moderate 2.17 (0.86–5.49) 0.80 (0.30–2.11) 0.94 (0.47–1.86) 1.04 (0.42–2.56) High 1.00 1.00 1.00 1.00 Vegetables or fruits in diet Less than 6 days a week 0.68 (0.31–1.49) NA20.64 (0.37–1.08) 2.71 (1.24–5.93) 6–7 days a week 1.00 1.00 1.00 1.00 OR = Odds ratio 95% CI = 95% confidence interval 1Unadjusted NA2= Too few observations for statistical analysis 3Economically inactive includes housekeepers, students and pensioners NA4= Data not available for Finnish reference group 81 higher prevalence of overweight and obesity among men in the general Finnish population only. Statistically significant associations between parity and overweight and obesity were found among Kurdish origin women only. Table 4. Age-adjusted odds ratios for overweight and obesity according to background characteristics within migrant groups in women. Russian Somali Kurdish General population OR (95% CI) OR (95% CI) OR (95% CI) OR (95% CI) Age1 18–39 years 1.00 1.00 1.00 1.00 40–64 years 6.48 (3.71–11.29) 14.31 (5.47–37.43) 3.69 (2.08–6.54) 4.05 (2.14–7.65) Education Less than primary school NA20.54 (0.18–1.65) 2.33 (0.94–5.77) NA2 Primary school 1.57 (0.75–3.29) 0.85 (0.32–2.26) 0.83 (0.46–1.49) 1.90 (1.222.94) High school 1.00 1.00 1.00 1.00 Employment status Economically inactive31.16 (0.56–2.42) 1.18 (0.41–3.40) 2.75 (1.42–5.33) 1.20 (0.50–2.90) Unemployed 1.15 (0.59–2.23) 3.67 (0.87–15.51) 3.81( 1.62–8.97) 0.86 (0.29–2.52) Employed 1.00 1.00 1.00 1.00 Time lived in Finland ≤5 years 0.87 (0.38–1.98) 0.78 (0.24–2.46) 0.94 (0.41–2.18) NA4 6–14 years 1.03 (0.56–1.91) 1.16 (0.44–3.08) 0.88 (0.45–1.69) ≥15 years 1.00 1.00 1.00 Physical activity Low 2.49 (1.17–5.30) 0.83 (0.20–3.40) 1.89 (0.92–3.88) 2.79 (1.12–6.96) Moderate 1.30 (0.63–2.72) 1.74 (0.51–5.91) 1.78 (0.88–3.61) 1.81 (0.81–4.03) High 1.00 1.00 1.00 1.00 Vegetables or fruits in diet Less than 6 days a week 1.53 (0.82–2.87) NA21.09 (0.62–1.93) 1.61 (0.79–3.29) 6–7 days a week 1.00 1.00 1.00 1.00 Parity 0 births 1.00 1.00 1.00 1.00 1–2 births 1.05 (0.41–2.67) 1.80 (0.55–5.87) 2.92 (1.29–6.61) 0.93 (0.56–1.54) ≥3 births 2.07 (0.67–6.40) 2.44 (0.82–7.29) 4.95 (1.95–12.5) 0.93 (0.48–1.81) OR = Odds ratio 95% CI = 95% confidence interval 1Unadjusted NA2= Too few observations for statistical analysis 3Economically inactive includes housekeepers, students and pensioners NA4= Data not available for Finnish reference group