Is acculturation associated with physical activity among female immigrants in Sweden?

Journal of Public Health, Jun 2013

Background Immigrant women in Sweden have lower levels of leisure time physical activity (LTPA) than Swedish-born women. The reasons are unclear, although acculturation has been suggested to play a role.

A PDF file should load here. If you do not see its contents the file may be temporarily unavailable at the journal website or you do not have a PDF plug-in installed and enabled in your browser.

Alternatively, you can download the file locally and open with any standalone PDF reader:

http://jpubhealth.oxfordjournals.org/content/35/2/270.full.pdf

Is acculturation associated with physical activity among female immigrants in Sweden?

Journal of Public Health | Is acculturation associated with physical activity among female immigrants in Sweden? Lena S. Jo¨ nsson 2 3 Karolina Palm e´r 2 3 Henrik Ohlsson 2 3 Jan Sundquist 1 2 3 Kristina Sundquist 1 2 3 0 The Author 2012, Published by Oxford University Press on behalf of Faculty of Public Health. All rights reserved 1 Visiting Professor, Stanford Prevention Research Center, Stanford University, School of Medicine, Medical School Office Building , 1265 Welch Road, Stanford, CA 94305-5411 , USA 2 Lena S. Jo ̈nsson, Postdoctoral Researcher Karolina Palme ́r, Statistician Henrik Ohlsson, Statistician Jan Sundquist , Professor, Family physician Kristina Sundquist, Professor, Family physician 3 Center for Primary Health Care Research, Department of Clinical Science, Lund University , Malmo ̈ SE-205 02 , Sweden A B S T R AC T Background Immigrant women in Sweden have lower levels of leisure time physical activity (LTPA) than Swedish-born women. The reasons are unclear, although acculturation has been suggested to play a role. Methods We used a cross-sectional study design to investigate the association between LTPA and two indicators of acculturation: (i) language proficiency (ability to understand news reports on the radio and television) and (ii) age at the time of migration, and if there existed a modifying effect on these hypothesized associations. The study sample consisted of 1651 women, aged 18 - 65, living in Sweden and born in Finland, Chile or Iraq. A postal questionnaire (translated into the women's native language) was used to collect the variables. The International Physical Activity Questionnaire was used to assess LTPA. Data were collected in 2002- 05 and analyzed in 2009- 10. A partial-proportional odds model was used for the analysis. Results Increased language proficiency was associated with increased LTPA [odds ratio (OR) ¼ 2.31, 95% confidence interval (CI) ¼ 1.57 - 3.41]. Country of birth modified the association. Furthermore, younger age at migration was associated with increased LTPA (OR ¼ 1.44, 95% CI ¼ 1.01- 2.03). Conclusions Increased language proficiency has the potential to be an important health-promoting factor among immigrant women. physical activity; public health Introduction Regular leisure time physical activity (LTPA) is associated with numerous physical and psychological benefits. For example, LTPA is associated with a decreased risk of myocardial infarction or coronary heart disease1,2 and the metabolic syndrome.3 Several studies have shown that minority women are more sedentary or less physically active during leisure time than women in the majority population.4 – 8 The difference in LTPA between women in different racial/ethnical groups persists even after adjustment for socioeconomic status.9,10 Thus, minority women may be exposed to an increased risk of chronic diseases unless they undertake regular physical activity through other means. It is, however, likely that the low levels of LTPA in minority women may contribute to disparities in health and increased costs in the health-care budget. The possible causes behind the low levels of LTPA among minority women are relatively unknown. It is important to examine the possible factors that are associated with low levels of LTPA in order to plan for appropriate interventions and improve the health status among minority women. Many minority women are poorly integrated in the host country11 and it is possible that their lack of integration could contribute to their low levels of LTPA. One way to examine integration is to measure acculturation, which is described as changes in cultural patterns when groups of individuals from different cultures come into first-hand continuous contact.12 Acculturation could be regarded as a complex, multidimensional process that takes place in the immigrant over time, either continuously or stepwise.13,14 Acculturation is partly related to individual behavior and it includes the process of social and cultural integration to the host country,13 such as being employed and having good language and communicative skills.15 The association between acculturation and LTPA is, however, ambiguous; most studies suggest a positive association,8,16 – 18 although others do not.19 Most studies on the association between acculturation and LTPA have been conducted in the USA. To our knowledge, only one such Swedish study has been conducted previously.17 Furthermore, few studies have been conducted on multiethnic populations making it difficult to ascertain whether the association between acculturation and LTPA might differ from one immigrant group to another.17,20 This is important to determine as interventions are often conducted in multiethnic populations. In 2008, 14% of the Swedish population was foreign born21 and .150 countries were represented among the immigrants. Therefore, Sweden is a suitable country for studies on multiethnic populations. More knowledge will provide a better basis for tailored interventions in the future. The aims of this study were to analyze the association between LTPA and two indicators of acculturation, i.e. language proficiency and age at the time of migration, in a multiethnic female population in Sweden, and to analyze if country of birth, season, age, self-rated health and body mass index (BMI) had a modifying effect on these hypothesized associations. Materials and Methods Study subjects The study was based on a cross-sectional survey of female, multiethnic, immigrants conducted in 2002 – 05. Each group of women (see below) comprised a simple random sample drawn from the local, Swedish government population registers in the municipalities of Stockholm and Botkyrka. These municipalities were chosen because of their high immigrant density. The random samples of women were first-generation immigrants, aged 18 – 65 and born in Finland (n ¼ 1506), Chile (n ¼ 1168) or Iraq (n ¼ 1192). These countries were chosen because they represent large minority groups in Sweden stemming from different parts of the world, including one neighboring country where the inhabitants have a mother tongue unrelated to Swedish (no Finnish-born women with Swedish as their mother tongue were included in the study). A total of 2031 women (53% response rate in average) answered the questionnaire. Non-respondents were approached by telephone in their native language and they responded to five questions. In most cases the reason for non-participation was refusal. There were no statistically significant differences between respondents and non-respondents in age, educational status, employment status, sick leave and self-rated health.22 Eighty-six women (4.2%) were excluded from the analyses due to outlier values or missing data for weight, height or physical activity, resulting in 1945 women. Of these, 13% also had missing data on language proficiency and 2.1% had missing data on age at migration. Thus, 1651 women were included in the analyses. They represented 42.7% of the original sample. Eighty-three percent of the women with missing data on language proficiency were from Finland. Written informed consent of the study subjects was obtained. The study was approved by the Regional Ethical Committee of Karolinska Institute, Stockholm. Variables Information on all variables, except season, was collected by a postal questionnaire. Outcome variable LTPA was used as the outcome variable and was defined as physical activity, e.g. walking, swimming or aerobics, which was done in the study subject’s spare time. Other types of physical activity, such as occupational or household-related physical activity, were not included. Self-reported LTPA was collected using the self-administered long format of the International Physical Activity Questionnaire (IPAQ long).23 It was translated into the women’s native language (Finnish, Spanish or Arabic).22 All participants were also offered the questionnaire in Swedish. The self-reported minutes per week of walking and moderate and vigorous intensity of physical activity during leisure time was calculated. Only activities lasting 10 min were included. To avoid self-report bias in the form of over-reporting, truncation was set to 120 min per activity type per day.22 The summed minutes of each type of activity were then weighted by a metabolic equivalent (MET). The used MET weights were 3.3 for walking, 4.0 for moderate intensity (e.g. bicycling or swimming at a regular pace or double tennis) and 8.0 for vigorous intensity (e.g. aerobics, running, fast bicycling or swimming).24 The continuous score, MET-min/week, was then categorized into three approximately equally sized groups: (i) 0 MET-min/week, (ii) 1 – 599 MET-min/week and (iii) 600 MET-min/week. Exposure variables Language proficiency. Information about the participants’ knowledge in Swedish was examined in five questions: (i) ability to understand news reports on the radio and television, (ii) speaking Swedish at meetings, (iii) communicating with authorities over the telephone, (iv) reading books in Swedish and (v) completing a written application for employment. The questions had four possible response alternatives, with 1 characterizing the lowest degree and 4 the highest degree of knowledge in Swedish. A Goodman – Kruskal gamma test showed at least a 90% positive association between the questions. Therefore, only one of the questions (i.e. ability to understand news reports on the radio and television) was chosen to characterize the women’s knowledge in Swedish. Response alternatives 1 ( poor) and 2 (average) for this question were merged, due to relatively few individual responses for these alternatives, resulting in the following groups: poor, good and very good language proficiency. Poor language proficiency was used as reference group. Age at migration. This variable was dichotomized as younger (0 – 12 years) or older (13 – 65 years) at migration. Initial analyses showed similar effect for different subgroups within the category 13 – 65 years and therefore we decided to use this broad categorization. Younger age at migration was used as the reference group. Sociodemographic variables The sociodemographic variables were chosen based on the past literature8,10 and established risk factors regarding physical inactivity. The groups noted as 0 were used as reference groups. Country of birth was categorized as: (0) born in Chile, (1) born in Finland and (2) born in Iraq. Season was dichotomized as (0) from November to March and (1) from April to October. Age was categorized in equally sized groups as: (0) 18 – 34 years, (1) 35 – 44 years, (2) 45 – 54 years and (3) 55 – 65 years. Self-rated health: each respondent was asked, ‘How would you describe your general health?’ Answers were dichotomized as: (0) very poor, poor, fair and (1) good, very good. Body mass index (BMI) was categorized as: (0) ,25 kg/m2 (normal weight or underweight), (1) 25 – 30 kg/m2 (overweight) and (2) .30 kg/m2 (obese). Educational level was categorized into two approximately equally sized categories: (0) 12 years of education and (1) .12 years of education. Employment status was dichotomized as: (0) unemployed and (1) employed (full-time or part-time). Financial resources: each respondent was asked, ‘If you were in economic difficulties, could you raise 14 000 Swedish crowns (¼1460 Euro) within a week?’ Answers were dichotomized as: (0) no and (1) yes. Marital status was dichotomized as: (0) not cohabiting with husband/partner and (1) cohabiting with husband/partner. Living with children was dichotomized as: (0) no and (1) yes. Statistical analysis A partial proportional odds model was fitted to investigate the association between the three-level LTPA measure (0 MET-min/week, 1 – 599 MET-min/week, 600 MET-min/ week) and different acculturation groups. The partial proportional odds model is an extended version of the binary logistic regression for an ordinal outcome variable and it relaxes the assumption of identical log-odds over all levels of the outcome. In the model, women performing no LTPA were compared with women performing low or high LTPA, and women performing no or low LTPA were compared with women performing high LTPA. We constructed several different models in order to investigate our research question. Model A included only the exposure variable of interest (language proficiency (A1) and age at migration (A2)). Model B1 and B2 also included the sociodemographic variables, i.e. country of birth, season, age, self-rated health, BMI, educational level, employment status, financial resources, marital status and living with children. Model C included both exposure variables in the same model and in Model D the sociodemographic variables were added. Model E included the sociodemographic variables and an interaction term between each of the sociodemographic variables and the exposure variable of interest [language proficiency (E1) and age at migration (E2)]. The interaction between language proficiency and country of birth was statistically significant (P-value ,0.05), but no other interaction terms were significant. For the exposure variables in the models we calculated odds ratios (ORs) with 95% confidence intervals (95% CI). The goodness of fit for the models was evaluated by comparing the likelihood value of the partial-proportional odds model with the likelihood obtained by fitting the same model with multinomial logistic regression, which fits maximum likelihood models for outcomes that have no natural ordering. A likelihood ratio test was then performed where a small value can be taken as evidence of a good fit.25 Based on this test we found no reason to reject the models. The analyses were conducted in 2009 – 10 and we used STATA version 11 for the analyses (Stata Corporation, College Station TX, 2003). Results The immigrant groups (here defined as a group of women born in the same country and now living in Sweden) were different regarding many of the sociodemographic variables (Table 1). For example, the percentage of women with no LTPA was 40% among Chilean, 16% among Finnish and 54% among Iraqi women. The distribution of the Finnish women in the different LTPA categories corresponded to that of the native Swedish women (data not shown). Finnish women had in general higher language proficiency, were older at migration, had better self-rated health, better socioeconomic status and were to a lesser extent living with children, compared with the Chilean and Iraqi women. The Chilean women were the youngest at migration, but otherwise they were intermediate between Finnish and Iraqi women regarding the other sociodemographic variables. In Models A1 and A2, we illustrate that increased language proficiency, or younger age at migration, was associated with increased LTPA among female, multiethnic, immigrants in Sweden (Tables 2 and 3). Comparing the unadjusted and the adjusted models showed that the sociodemographic variables included in Model B attenuated the association between language proficiency and LTPA, but the results remained conclusive in most cases. The adjustment had only a minor effect on the association between age at migration and LTPA (Tables 2 and 3). When including both language proficiency and age at migration in the same model (Model D), the ORs concerning language proficiency were almost the same compared with Models B1 and B2, but the OR concerning age at migration was no longer conclusive (OR ¼ 1.23, 95% CI ¼ 0.86 – 1.77). As the language proficiency increased from poor to very good, the odds of performing low or high LTPA (compared with no LTPA) was 2.31 times higher. On the other hand, the odds of performing high LTPA compared with low or no LTPA was only 1.79 times higher for the same increase in language proficiency (Tables 2 and 3). We observed an interaction between country of birth and language proficiency, but not between any other variables (Tables 4 and 5). That is, the strength of the association between language proficiency and LTPA was dependent on the women’s country of birth. For example, the association between language proficiency and LTPA was stronger among Chilean women than among Iraqi women (Tables 4 and 5). Furthermore, language proficiency did not have as strong association with LTPA among Finnish women increasing from no or low LTPA to high LTPA as among Chilean women with the same increase in language proficiency (Table 5). Discussion Main findings of this study Acculturation is strongly associated with LTPA, as evidenced by the main findings of this study showing that an increase in language proficiency, or a younger age at migration, is associated with increased levels of LTPA among multiethnic immigrant women in Sweden. The level of LTPA among the immigrant women increased towards the level of LTPA among native Swedish women. Interestingly, as the language proficiency increased from poor to very good, the odds of performing low or high LTPA (compared with no LTPA) was higher than the odds of performing high LTPA compared with low or no LTPA for the same increase in language proficiency. Furthermore, language proficiency showed an interaction with country of birth and thus, increased language proficiency could have a greater impact on LTPA, and hence, by implication, health, in some immigrant groups. What is already known on this topic In earlier studies, the percentage of Mexican and Central or South American women with no LTPA ranged from 47 to 53%,9,10 the percentage of Caucasians and Non-Hispanic white women with no LTPA ranged from 22 to 38%9,10 and the percentage of first and second generation female immigrants from Turkey with no LTPA was 60%.18 In the present study, the percentage of women with no LTPA was somewhat lower than in these studies. The reason for the difference in results could be that the studies by Crespo et al.9 and Neighbors et al.10 was conducted in the USA and that LTPA might be lower in the USA in general compared with Sweden. Moreover, earlier studies have shown an association between LTPA and preferred language at home,16 cultural orientation towards the host country and social contact with native inhabitants.18 However, studies examining the association between LTPA and length of residency in the host country have been inconsistent.17,19 The difference between our results and those of Yang et al.19 could be due to the difference in number of study subjects: we examined 1651 women, whereas Yang et al. were only able to include 152 women. Furthermore, Wolin et al.20 showed no modifying effect of race when examining the association between Table 1 Description of the study population Chilean (n ¼ 481) Finnish (n ¼ 645) Iraqi (n ¼ 525) Total (n ¼ 1651) language preferences and LTPA in low-income multiethnic populations in the USA. This is in contrast to the results of the present study, where we found an interaction between country of birth and language proficiency. For example, the association between language proficiency and LTPA was stronger among Chilean women than among Iraqi women. Table 2 Results from the partial-proportional odds model (low or high LTPA versus no LTPA) Table 3 Results from the partial-proportional odds model (high LTPA versus low or no LTPA) Language proficiency and age at migration Poor language proficiency Good language proficiency Very good language proficiency Older age at migration Younger age at migration 1.44 (1.08 – 1.92) 1.44 (1.01 – 2.03) LTPA, denotes leisure time physical activity; OR, denotes odds ratio; CI, denotes confidence intervals. aThe ORs are adjusted for country of birth, season, age, self-rated health, body mass index, educational level, employment status, financial resources, marital status and living with children. What this study adds To our knowledge, this is one of the first studies in Sweden and Europe to study the association between acculturation and LTPA. Earlier studies have mainly been conducted in the USA.8,9,19,20,26 Furthermore, only two earlier studies on acculturation and LTPA have been conducted in a multiethnic, adult population, which is a shortcoming as interventions are often conducted in multiethnic populations.17,20 In the present study, we show that the strength of the association between language proficiency and LTPA is dependent on the women’s country of birth, which is a novel contribution.17,20 Moreover, the present study adds support to earlier studies showing an association between acculturation and LTPA,16 – 18 and diminishes the negative results of an earlier, smaller study.19 Our results indicate that poor language proficiency may lead to a poorer health profile in minority women, independently of a comprehensive set of sociodemographic characteristics. Limitations of this study The present study had some limitations: data on LTPA were based on self-reported physical activity, which includes a recall bias. Moreover, not all types of physical activity were LTPA, denotes Leisure Time Physical Activity; OR, denotes Odds Ratio; CI, denotes Confidence Intervals. aThe ORs are adjusted for country of birth, season, age, self-rated health, BMI, educational level, employment status, financial resources, marital status and living with children. taken into account in this study. For example, we did not take into account physical activity in other domains, such as active traveling and having a manual job, which may be common among some immigrant groups because of their relative socioeconomic disadvantage at the individual as well as at the community level. In addition, our study did not disentangle whether our findings are a consequence of isolation and low levels of LTPA within communities (group effects) or a direct consequence of lack of individual language skills. Further qualitative studies may help to uncover the most appropriate level (group or individual) for public health interventions. It is possible that the recall bias is weaker during leisure time than during the rest of the day due to more organized activities during leisure time. Besides, earlier studies have shown an association between low physical activity during leisure time and deteriorated health.1 – 3 A further limitation was the limited response rate and the missing values in some variables, which may have led to a non-response bias. However, imputing data using univariate multiple imputation (the ordered logistic method with five sets of simulated values) only had a minor effect on the results (data not shown). Furthermore, there is no defined cutoff between low and high LTPA and therefore we used Table 4 Results from the partial-proportional odds model with an interaction term between language proficiency and country of birth included (low or high LTPA versus no LTPA) Table 5 Results from the partial-proportional odds model with an interaction term between language proficiency and country of birth included (high LTPA versus low or no LTPA) OR (95% CI)a Poor Chile Good Finland Good Iraq Very good Finland Very good Iraq OR (95% CI)a LTPA, denotes leisure time physical activity; OR, denotes odds ratio; CI, denotes confidence intervals. aThe ORs are adjusted for season, age, self-rated health, body mass index, educational level, employment status, financial resources, marital status and living with children. the cutoff for total physical activity;23 a different categorizing of the outcome variable, however, did not seem to change the results (data not shown). The cross-sectional design of this study did not allow us to infer causality. However, a possible explanation to the results is that women with improved language proficiency, or younger age at migration, were able to navigate their new environment more easily and were in a better position to make use of leisure centers, or more prone to a changed life style behavior because of fewer perceived expectations from the family and the society regarding the duties of women. This has been demonstrated in an earlier study examining the reasons why non-English speaking migrant Muslim women do or do not attend English language classes.27 In addition, Muslim women may perceive physical activity differently than women from Western countries. Berge et al.28 studied physical activity behavior of Muslim women in the Emirates and found several perceived barriers to daily exercise. However, the association between acculturation and LTPA may also be explained by a possible differential reporting, i.e. the conception of leisure time, physical activity or language proficiency may differ between ethnic groups. It is also possible that residual confounding factors for the association between language proficiency or age at the time of LTPA, denotes leisure time physical activity; OR, denotes odds ratio; CI, denotes confidence intervals. aThe ORs are adjusted for country of birth, season, age, self-rated health, body mass index, educational level, employment status, financial resources, marital status and living with children. migration and LTPA were not taken into account and that the positive association would be attenuated if these unknown confounders were included. Conclusions and future research Our results, like the results from other similar-sized studies, suggest that there is an association between acculturation and LTPA. We conclude that the strength of the association between acculturation and LTPA varies between different levels of LTPA as well as between different immigrant groups. The best way to examine acculturation is an area for future research as well as is the question on how to best increase the acculturation among immigrants. Also, it would be suitable to study the effect of the women’s total healthpromoting physical activity in future research. Our conclusion is that increased language proficiency could be an important health-promoting factor among multiethnic, immigrant women in Sweden, especially among women with no LTPA. We suggest that future studies include a language class RCT in minority women with LTPA as an outcome. Such language classes may have positive individual as well as community effects. A S S O C I AT I O N B E T W E E N AC C U LT U R AT I O N A N D P H Y S I CA L AC T I V I T Y 1 2 3 4 5 6 7 8 9 Lovasi GS , Lemaitre RN , Siscovick DS et al. Amount of leisuretime physical activity and risk of nonfatal myocardial infarction . Ann Epidemiol 2007 ; 17 : 410 - 6 . Sundquist K , Qvist J , Johansson SE et al. The long-term effect of physical activity on incidence of coronary heart disease: a 12-year follow-up study . Prev Med 2005 ; 41 : 219 - 25 . Laaksonen DE , Lakka HM , Salonen JT et al. Low levels of leisuretime physical activity and cardiorespiratory fitness predict development of the metabolic syndrome . Diab Care 2002 ; 25 : 1612 - 8 . Bennett SA . Inequalities in risk factors and cardiovascular mortality among Australia's immigrants . Aust J Public Health 1993 ; 17 : 251 - 61 . Crespo CJ , Keteyian SJ , Heath GW et al. Leisure-time physical activity among US adults. Results from the Third National Health and Nutrition Examination Survey . Arch Intern Med 1996 ; 156 : 93 - 8 . Lindstro¨m M , Sundquist J. Immigration and leisure-time physical inactivity: a population-based study . Ethn Health 2001 ; 6 : 77 - 85 . Gadd M , Sundquist J , Johansson SE et al. Do immigrants have an increased prevalence of unhealthy behaviours and risk factors for coronary heart disease ? Eur J Cardiovasc Prev Rehabil 2005 ; 12 : 535 - 41 . Kandula NR , Lauderdale DS . Leisure time, non-leisure time, and occupational physical activity in Asian Americans . Ann Epidemiol 2005 ; 15 : 257 - 65 . Crespo CJ , Smit E , Andersen RE et al. Race/ethnicity, social class and their relation to physical inactivity during leisure time: results from the Third National Health and Nutrition Examination Survey , 1988 - 1994 . Am J Prev Med 2000 ; 18 : 46 - 53 . 10 Neighbors CJ , Marquez DX , Marcus BH . Leisure-time physical activity disparities among Hispanic subgroups in the United States . Am J Public Health 2008 ; 98 : 1460 - 4 . 11 Reitzel LR , Mazas CA , Cofta-Woerpel L et al. Acculturative and neighborhood influences on subjective social status among Spanishspeaking Latino immigrant smokers . Soc Sci Med 2010 ; 70 : 677 - 83 . 12 Berry J. Conceptual approaches to acculturation . In Chun K, Organista P , Marin G (eds). Acculturation: Advances in Theory, Measurement, and Applied Research. Washington, DC: American Psychological Association , 2003 , 17 - 37 . 13 Ryder AG , Alden LE , Paulhus DL . Is acculturation unidimensional or bidimensional? A head-to-head comparison in the prediction of personality, self-identity, and adjustment . J Pers Soc Psychol 2000 ; 79 : 49 - 65 . 14 Ritsner M , Ponizovsky A. Psychological distress through immigration: the two-phase temporal pattern? Int J Soc Psychiatry 1999 ; 45 : 125 - 39 . 15 Lieber E , Chin D , Nihira K et al. Holding on and letting go: identity and acculturation among Chinese immigrants . Cultur Divers Ethnic Minor Psychol 2001 ; 7 : 247 - 61 . 16 Crespo CJ , Smit E , Carter-Pokras O et al. Acculturation and leisuretime physical inactivity in Mexican American adults: results from NHANES III , 1988 - 1994 . Am J Public Health 2001 ; 91 : 1254 - 7 . 17 Dawson AJ , Sundquist J , Johansson SE . The influence of ethnicity and length of time since immigration on physical activity . Ethn Health 2005 ; 10 : 293 - 309 . 18 Hosper K , Klazinga NS , Stronks K. Acculturation does not necessarily lead to increased physical activity during leisure time: a crosssectional study among Turkish young people in the Netherlands . BMC Public Health 2007 ; 7 : 230 . 19 Yang K , Laffrey SC , Stuifbergen A et al. Leisure-time physical activity among midlife Korean immigrant women in the US . J Immigr Minor Health 2007 ; 9 : 291 - 8 . 20 Wolin KY , Colditz G , Stoddard AM et al. Acculturation and physical activity in a working class multiethnic population . Prev Med 2006 ; 42 : 266 - 72 . 21 Statistics Sweden. Description of the population in Sweden 2008 . O¨ rebro, Sweden: SCB-Tryck , 2009 . ISBN: 978 - 91 - 618 - 1484 -8. 22 So¨ dergren M , Sundquist K , Johansson SE et al. Associations between health-enhancing physical activity and country of birth among women . J Phys Act Health , 2010 ; 7 : 613 - 21 . 23 Craig CL , Marshall AL , Sj o¨stro¨ m M et al. International physical activity questionnaire: 12-country reliability and validity . Med Sci Sports Exerc 2003 ; 35 : 1381 - 95 . 24 Ainsworth BE , Haskell WL , Whitt MC et al. Compendium of physical activities: an update of activity codes and MET intensities . Med Sci Sports Exerc 2000 ; 32 : S498 - 504 . 25 StataCorp . Stata: Release 11. Statistical Software. College Station, TX: StataCorp LP , 2009 . 26 Marquez DX , McAuley E. Gender and acculturation influences on physical activity in Latino adults . Ann Behav Med 2006 ; 31 : 138 - 44 . 27 Rida A , Milton M. The non-joiners: why migrant Muslim women aren't accessing English language classes . Prospect 2001 ; 16 : 35 - 48 . 28 Berger G , Peerson A. Giving young Emirati women a voice: participatory action research on physical activity . Health Place 2009 ; 15 : 117 - 24 .


This is a preview of a remote PDF: http://jpubhealth.oxfordjournals.org/content/35/2/270.full.pdf

Lena S. Jönsson, Karolina Palmér, Henrik Ohlsson, Jan Sundquist, Kristina Sundquist. Is acculturation associated with physical activity among female immigrants in Sweden?, Journal of Public Health, 2013, 270-277, DOI: 10.1093/pubmed/fds091