Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Insulin Resistance in Chileans of European and Indigenous Descent: Evidence for an Ethnicity x Environment Interaction

  • Carlos A. Celis-Morales,

    Affiliations Institute of Cardiovascular and Medical Sciences, College of Medical, Veterinary and Life Sciences, University of Glasgow, Glasgow, United Kingdom, School of Life Sciences, College of Medical, Veterinary and Life Sciences, University of Glasgow, Glasgow, United Kingdom

  • Francisco Perez-Bravo,

    Affiliation Laboratory of Nutritional Genomics, Department of Nutrition, Faculty of Medicine, University of Chile, Santiago, Chile

  • Luis Ibañes,

    Affiliation Center of Nutrition, Metabolism and Physical Activity (CNEF), Concepcion, Chile

  • Ruth Sanzana,

    Affiliation Center of Nutrition, Metabolism and Physical Activity (CNEF), Concepcion, Chile

  • Edison Hormazabal,

    Affiliation Center of Nutrition, Metabolism and Physical Activity (CNEF), Concepcion, Chile

  • Natalia Ulloa,

    Affiliation Department of Clinical Biochemistry and Immunology, Universidad de Concepción, Concepción, Chile

  • Carlos Calvo,

    Affiliation Department of Clinical Biochemistry and Immunology, Universidad de Concepción, Concepción, Chile

  • Mark E. S. Bailey,

    Affiliation School of Life Sciences, College of Medical, Veterinary and Life Sciences, University of Glasgow, Glasgow, United Kingdom

  • Jason M. R. Gill

    jason.gill@glasgow.ac.uk

    Affiliation Institute of Cardiovascular and Medical Sciences, College of Medical, Veterinary and Life Sciences, University of Glasgow, Glasgow, United Kingdom

Insulin Resistance in Chileans of European and Indigenous Descent: Evidence for an Ethnicity x Environment Interaction

  • Carlos A. Celis-Morales, 
  • Francisco Perez-Bravo, 
  • Luis Ibañes, 
  • Ruth Sanzana, 
  • Edison Hormazabal, 
  • Natalia Ulloa, 
  • Carlos Calvo, 
  • Mark E. S. Bailey, 
  • Jason M. R. Gill
PLOS
x

Abstract

Background

Effects of urbanisation on diabetes risk appear to be greater in indigenous populations worldwide than in populations of European origin, but the reasons are unclear. This cross-sectional study aimed to determine whether the effects of environment (Rural vs. Urban), adiposity, fitness and lifestyle variables on insulin resistance differed between individuals of indigenous Mapuche origin compared to those of European origin in Chile.

Methodology/Principal Findings

123 Rural Mapuche, 124 Urban Mapuche, 91 Rural European and 134 Urban European Chilean adults had blood taken for determination of HOMA-estimated insulin resistance (HOMAIR) and underwent assessment of physical activity/sedentary behaviour (using accelerometry), cardiorespiratory fitness, dietary intake and body composition. General linear models were used to determine interactions with ethnicity for key variables. There was a significant “ethnicity x environment” interaction for HOMAIR (Mean±SD; Rural Mapuche: 1.65±2.03, Urban Mapuche: 4.90±3.05, Rural European: 0.82±0.61, Urban European: 1.55±1.34, p(interaction) = 0.0003), such that the effect of urbanisation on HOMAIR was greater in Mapuches than Europeans. In addition, there were significant interactions (all p<0.004) with ethnicity for effects of adiposity, sedentary time and physical activity on HOMAIR, with greater effects seen in Mapuches compared to Europeans, an observation that persisted after adjustment for potential confounders.

Conclusions/Significance

Urbanisation, adiposity, physical activity and sedentary behaviour influence insulin resistance to a greater extent in Chilean Mapuches than Chileans of European descent. These findings have implications for the design and implementation of lifestyle strategies to reduce metabolic risk in different ethnic groups, and for understanding of the mechanisms underpinning human insulin resistance.

Introduction

Prevalence of diabetes is increasing worldwide, but differs widely between regions [1], which can be partly attributed to differences in urbanisation and obesity [2]. However, differences in environment alone do not appear to tell the whole story, particularly in elucidating why certain populations and ethnic groups experience a disproportionately high prevalence of type 2 diabetes when they adopt a Western lifestyle. The classic example of this is the Pima Indians, who when living a traditional rural lifestyle in Mexico are lean, active and have low diabetes prevalence, but when living in the US, are generally obese and have a diabetes prevalence in the adult population of ∼40% [3]. This pattern is also evident in other indigenous populations throughout the Americas and Australasia [4].

Mapuches – an indigenous Native American population from Chile – appear to follow this pattern of a disproportionate increase in risk of diabetes (compared to Chileans of White European descent) when they transition from a traditional rural lifestyle to an urban one. Limited data, based on observations from small sample groups, suggest that diabetes prevalence for Mapuches living in traditional rural environments is low, at between 1% and 4% of the adult population [5], [6], but rises to 6.3–8.2% amongst urban-dwelling Mapuches [7], [8]. In contrast, Chileans of European descent (∼92% of the Chilean population [9]) have a smaller differential between the two environments, with higher rates of diabetes than Mapuches in rural areas (4.5%) and lower rates in urban (5.8%) settings [10]. Interestingly, differences in BMI do not explain the difference in diabetes prevalence between rural and urban Mapuches [5], [8], which contrasts with observations from other indigenous groups where the transition to an urban environment is associated with a substantial increase in obesity prevalence [3], [11], [12]. It is likely that, in common with other indigenous groups, Mapuches have innate factors that predispose them to increased insulin resistance and diabetes risk, whose effects are revealed on adoption of an urbanised lifestyle. However, it is also possible that Mapuches have a greater lifestyle shift compared to European Chileans when moving from rural to urban settings, which could explain the larger apparent effect of environment on diabetes risk in this group. For example, physical activity or cardiorespiratory fitness levels both predict diabetes risk independently of obesity [13], [14]: larger differences in these factors or in aspects of body composition not reflected in BMI measurements may be evident between urban and rural Mapuches compared to Europeans living between these two environments. Thus, in order to determine whether Mapuches are indeed more susceptible to the adverse metabolic effects of an urban environment than European Chileans, detailed and objective measures of physical and lifestyle variables of both Mapuches and European Chileans living in rural and urban environments are required.

The aims of the present study were therefore twofold. The first aim was to determine cross-sectionally, using objective measures, whether urbanisation influenced metabolic risk factors, cardiorespiratory fitness, body composition and lifestyle variables differently in Chileans of Mapuche compared to European descent. The second aim was to determine whether physical activity, sedentary behaviour, cardiorespiratory fitness, dietary intake and body composition affected insulin resistance differently between Mapuche and European Chilean populations.

Methods

Ethics statement

All participants gave written informed consent prior to inclusion in this study, which was approved by Research Ethics Committees at the University of Glasgow, University of Chile, and University of Concepcion.

Participants and study design

The sampling strategy was cross-sectional in design and involved the planned collection of data from four groups of participants: Mapuches living in Rural and Urban environments and Chileans of European descent living in Rural and Urban environments. Participants of both sexes, aged 20–60 years, were recruited, from towns/cities and from rural villages and communities in three regions of Chile – Los Ríos, Bio-Bio and Metropolitana – between February and June 2008. The study was advertised using posters displayed in prominent locations (in the smaller communities), using local radio advertisements, and by open invitations to the membership of local community organisations, social and sports clubs. These methods were employed in an attempt to ensure that all groups within each society had an approximately equal chance of seeing the study and of signing up to it, thereby minimising potential recruitment bias. In the smaller communities, particularly those rural communities dominated by Mapuches, almost all adults in the community were sampled after agreement from local community leaders to let us advertise the study locally. Sampling rates were lower in larger communities.

To ensure that Mapuche and European population groups with as little admixture as possible were studied, the following inclusion criteria were applied: i) Mapuche participants were included if they had; a) both maternal and paternal last names of Mapuche origin; these names are identifiably different from European names, b) both parents of Mapuche origin, c) type O blood group in the ABO blood group system (Mapuches have a very high frequency (95%) of this blood group [8], thus possession of A or B blood groups is likely to indicate mixed ancestry). ii) Europeans were included if they had a) both maternal and paternal last names of European origin, b) both parents of European origin. Individuals with a known history of cardiovascular disease or taking anti-hypertensive or diabetes medications were excluded from participation. All participants were told that personal health-related phenotypic data would be communicated back to them via their general practitioner, but they received no other incentive to participate.

Physical, biochemical and behavioural measurements

All phenotype measurements were conducted ‘in the field’ at the locations where the study populations resided. Facilities for centrifuging blood samples and for cooling/freezing blood and plasma were transported to all locations including remote rural communities. Height, body mass, waist and hip circumferences and skinfolds at four sites (biceps, triceps, subscapular, suprailiac) were measured using standard protocols [15]. Body composition was calculated from skinfold measures [16]. Blood pressure was measured on the right arm after at least 10 minutes of seated rest using an automated blood pressure monitor (Omron HEM705 CP, Omron Healthcare UK Limited, Milton Keynes, UK) which has been validated according to the European Society of Hypertension International Protocol [17]. The mean of two blood pressure readings was used in analysis.

The Chester step-test, a validated 3-stage incremental stepping test, was used to estimate maximal oxygen uptake (VO2max) [18] as a measure of fitness. Participants wore accelerometers (ActiTrainer; ActiGraph LLC, Pensacola, FL, USA) on the left hip at all times, except when showering, swimming and sleeping, for seven consecutive days to objectively assess physical activity levels. Dietary intake was assessed by 7-day weighed food record and analyzed using the Chilean Food Composition Database (Software MINUTA, University of Concepcion, Chile).

Venous blood samples were drawn after an overnight fast and collected into potassium EDTA tubes and placed on ice. Plasma was separated within 10 minutes of collection and frozen at −20°C until analysis. Glucose, triglyceride (TG), total cholesterol, HDL cholesterol, γ- glutamyltransferase (GGT), alanine aminotransferase (ALT) and C-reactive protein (hsCRP) concentrations (using a high sensitivity assay) were determined by enzymatic colorimetric methods using commercially available kits (Roche Diagnostics Gmbh, Mannheim, Germany; Randox Laboratories Ltd., Co. Antrim, Ireland; and Kamiya Biomedical, Seattle, USA). LDL cholesterol was calculated using the Friedewald equation [19]. Insulin and leptin concentrations were determined by radioimmunoassay (Diagnostic System Labs, TX, USA and Linco Research Inc, St. Louis MO, USA). Coefficients of variation were <3.0% for all enzymatic colorimetric assays, 5.0% for insulin and 3.1% for leptin.

Participants' health history, including smoking status and family health history, was determined by questionnaire. Socioeconomic status was determined with the European Society for Opinion and Marketing Research (ESOMAR) questionnaire validated in the Chilean population [20]. The original 6 ESOMAR socioeconomic classes were re-grouped into three classes by combining the two lower, two middle and two higher classes for analysis. Demographic and cultural data (age, attained education, most recent occupation, and ethnicity) were determined using the Chilean Socioeconomic Characterisation Questionnaire [21]. All questionnaire data were collected during in-person interviews.

Data and statistical analysis

Data on which to base a sample size calculation for this study were limited as there have been few previous investigations of insulin resistance in Mapuches. The available data, based on small samples, indicated that the population SD for HOMAIR was ∼3–4 units in Mapuches [8], and <1 unit in the non-obese urban Chilean general population [22]. Based on this, we estimated an overall population SD for HOMAIR of ∼3 units, and predicted that the study would need power to detect mean group differences of ∼2 HOMAIR units. Power calculations using Minitab (version 14, Minitab Inc., State College, PA, USA) indicated that 65 participants in each of the four study groups would provide 80% power to detect such differences in HOMAIR between groups at the p<0.004 level (see below for adjustment of the critical p-value to account for multiple testing). This was therefore set as our minimum recruitment target. The upper limit for participant numbers was determined by participant availability within the data collection time-frame.

Accelerometer readings were summarized in 60-second epochs and Freedson cut-points used to define intensity domains [23]. Data from participants with at least 10 hours of daily accelerometer wear time for 4 days were included in the analysis. Non-wear was defined by intervals of at least 60 minutes of zero activity counts [24]. Wear time was calculated by subtracting non-wear time from 24 hours. Activity count values of <100 count.min−1 were defined as sedentary behaviour [25]. Insulin resistance was assessed using the Homeostasis Model Assessment of Insulin Resistance (HOMAIR) [26].

Data were analyzed using Statistica (version 8.0; StataSoft, Tulsa, USA) and Minitab. Quantitative data were tested for normality using the Anderson–Darling normality test, subjected to Box-Cox analysis, and transformed as appropriate. Accordingly, statistical analysis for insulin, HOMAIR, GGT, ALT and hsCRP was performed using logarithmically transformed data.

To determine whether the effects of urbanisation differed between Chileans of European and Mapuche descent, anthropometric, metabolic, fitness, physical activity and dietary variables were compared between the Rural Mapuche, Urban Mapuche, Rural European and Urban European groups using General Linear Models, considering ethnicity and environment main effects and the ethnicity x environment interaction. In cases where this model revealed no significant ethnicity x environment interaction, the ethnicity and environment main effects were subsequently determined in a model without the interaction term. Men and women were combined in these analyses; all analyses (except age) were undertaken on age- and sex-adjusted data. Missing data values were not imputed. Numbers of observations for individual variables are presented as supplementary data (Table S1).

To determine whether the effects of adiposity, physical activity, fitness and dietary intake on insulin resistance differed between Mapuches and Europeans, participants were divided into sex-specific tertiles for adiposity, physical activity, fitness and dietary intake variables (see Table S2 for tertile cut-points), and General Linear Models were used to determine main effects of increasing ‘tertile’ of each variable on HOMAIR, and any ethnicity x tertile interactions. Tertiles for men and women were combined in this analysis (i.e. Tertile 1 in the analyses contained the lowest third of women and the lowest third of men for any given variable). Rural and Urban groups were combined in these analyses. All models were adjusted for age and sex and further models adjusted for potential confounding variables were undertaken as appropriate.

The critical (α-value) for p, at which statistical significance was accepted, was corrected for multiple testing using the Bonferroni method as implemented at the Simple Interactive Statistical Analysis website (http://www.quantitativeskills.com/sisa/), taking into account the number of comparisons made (n = 36) and the mean correlation between the variables tested (r = 0.28). Accordingly, statistical significance was accepted at p<0.004 for all analyses.

Results

A total of 873 individuals (247 Rural Mapuche, 187 Urban Mapuche, 216 Rural European, 223 Urban European) responded to our call for volunteers. Of these, 472 individuals (123 Rural Mapuche, 124 Urban Mapuche, 91 Rural European, 134 Urban European) fulfilled the inclusion/exclusion criteria and agreed to participate in the study. Of the 401 individuals who responded but were not included, 291 did not meet the inclusion/exclusion criteria and 110 chose not to participate. The number of eligible non-participants was higher in Mapuches (62 Rural, 11 Urban) than Europeans (28 Rural, 9 Urban): this was largely the consequence of religious or ethnically-influenced political concerns about the study in some members of the Rural Mapuche population. Three Urban Europeans, 1 Rural European, 11 Urban Mapuches and 4 Rural Mapuches had fasting glucose concentrations >7.0 mmol.l−1 in the blood sample collected, indicating possible undiagnosed diabetes. These participants were included in the data analyses reported below, however excluding these individuals from the analyses did not alter any of the study findings (data not shown). Demographic data, by sub-group (Rural Mapuche, Urban Mapuche, Rural European and Urban European) are shown in Table 1. In the sample taken for this study, Mapuches were slightly younger overall than the Europeans (by 3.5 years, p = 0.003). Because of this difference, all subsequent statistical analyses were performed on age-adjusted (and sex-adjusted) data.

thumbnail
Table 1. Demographic and anthropometric variables by ethnic group and environment.

https://doi.org/10.1371/journal.pone.0024690.t001

Anthropometric variables for Mapuches and Europeans living in Rural and Urban Environments

Table 1 displays anthropometric variables for the four sub-groups along with the results of tests for differences based on ethnicity and/or environment. No significant ethnicity x environment interactions were observed for any of these variables. However, significant main effects of environment were evident for BMI and percentage body fat, with the Rural participants having higher mean BMI (by 1.6 kg.m−2, p = 0.002) but lower mean body fat measurements (by 2.2 percentage points, p = 0.0001) than their Urban counterparts. This indicates that the higher BMI observed in the Rural participants was a consequence of greater fat-free mass.

Metabolic variables for Mapuches and Europeans living in Rural and Urban Environments

Table 2 shows metabolic variables for the four sub-groups. There was a significant ‘ethnicity x environment’ interaction for HOMAIR (p = 0.0003), such that HOMAIR was 3-fold higher in Urban compared to Rural Mapuches but only 1.9-fold higher in Urban compared to Rural Europeans (Table 2, Figure 1). In addition to this interaction effect, HOMAIR was higher in Mapuches than Europeans (p = 0.0001) and was higher in Urban compared to Rural groups (p = 0.0001). A similar pattern was observed for fasting insulin concentrations.

thumbnail
Figure 1. Effect of ethnicity and environment on HOMAIR in European and Mapuche participants.

Bars show mean±SEM for all groups. p(interaction) describes the ‘ethnicity x environment’ interaction for HOMAIR in age- and sex-adjusted analysis.

https://doi.org/10.1371/journal.pone.0024690.g001

No significant ‘ethnicity x environment’ interactions were observed for any of the other measured metabolic variables, however main effects for ethnicity or environment were observed for some factors. Systolic blood pressure, and glucose, triglyceride, leptin, GGT and hsCRP concentrations were higher, and HDL cholesterol concentration was lower, in Urban compared to Rural groups (all p<0.004). In addition, leptin and GGT concentrations were higher in Mapuches compared to Europeans (both p<0.004).

Fitness and physical activity for Mapuches and Europeans living in Rural and Urban Environments

Fitness and physical activity variables for the four sub-groups are shown in Table 3. A significant ethnicity x environment interaction was evident for VO2max (p = 0.0001), such that the difference in VO2max between Rural and Urban Mapuches (8.6 ml.kg−1.min−1 higher in Rural) was greater than the difference between Rural and Urban Europeans (4.0 ml.kg−1.min−1 higher in Rural). In addition to this interaction effect, VO2max was higher in Rural compared to Urban groups (p = 0.0001).

thumbnail
Table 3. Fitness, physical activity and dietary intake variables by ethnic group and environment.

https://doi.org/10.1371/journal.pone.0024690.t003

No significant ethnicity x environment interactions were observed for any of the activity variables, but sedentary time was higher (p = 0.0001) and moderate-to-vigorous activity time was lower (p = 0.0001) in Urban compared to Rural groups.

Dietary variables for Mapuches and Europeans living in Rural and Urban Environments

Dietary factors differed noticeably between sub-groups (Table 3). Significant ethnicity x environment interactions were observed for energy, fat, dietary fibre and alcohol intakes (all p<0.004), indicating that ethnic differences exist for the effects of urbanisation on diet. For all these dietary components the differences between Rural and Urban populations were greater in Mapuches than Europeans. In addition to these interaction effects, Rural participants had higher intakes than Urban participants for all dietary components, and Mapuches had higher intakes than Europeans for all dietary components, except protein (all p<0.004).

Effects of adiposity on HOMAIR in Mapuches and Europeans

Figure 2 shows the effect of increases in BMI, waist circumference and percentage body fat on HOMAIR in Europeans and Mapuches. There was a significant trend overall for HOMAIR to increase with increasing BMI, waist circumference and percentage body fat (all p<0.004, Table S3). In age- and sex-adjusted analyses there were significant ‘ethnicity x BMI tertile’ and ‘ethnicity x body fat tertile’ interactions for HOMAIR (p<0.004) with Mapuches experiencing significantly greater increases in HOMAIR with increasing BMI or body fat than Europeans (Figure 2, Table S3). The ‘ethnicity x waist tertile’ interaction followed the same pattern but only achieved borderline statistical significance (p = 0.007). Further adjustment for smoking status, environment (rural or urban), socio-economic level and education level did not alter the statistical significance of any of these findings (Table S3).

thumbnail
Figure 2. Effects of BMI, waist circumference and body fat on HOMAIR in European and Mapuche participants.

Bars show mean ± SEM for all groups. p(interaction) describes the ‘tertile x environment’ interaction for HOMAIR after adjustment for potential confounding covariates (tertile cut-points shown in Table S2, see text or Table S3 for list of covariates).

https://doi.org/10.1371/journal.pone.0024690.g002

Effects of physical activity, sedentary time and fitness on HOMAIR in Mapuches and Europeans

Figure 3 shows the effects of increasing amounts of sedentary time, moderate-to-vigorous physical activity (MVPA) and fitness (as measured using VO2max) on HOMAIR in Europeans and Mapuches. As for the anthropometric variables described above, there were significant trends for HOMAIR to increase with increasing tertiles of sedentary time, and with decreasing tertiles of MVPA and decreasing tertiles of VO2max (all p<0.004, Table S4). Significant ‘ethnicity x sedentary time tertile’ and ‘ethnicity x MVPA tertile’ interactions were also evident, such that Mapuches experienced greater increases in HOMAIR with increasing sedentary time and decreasing MVPA than Europeans in age- and sex-adjusted analyses (all p<0.004, Figure 3, Table S4). The ‘ethnicity x sedentary time tertile’ and ‘ethnicity x MVPA tertile’ interactions with HOMAIR were unchanged by further adjustment for: accelerometer wear time, VO2max, BMI, waist circumference, percentage body fat, energy intake, smoking status, environment, socio-economic level, education level and MVPA (for the sedentary time model) or sedentary time (for the MVPA model) (both p<0.004, Figure 3, Table S4).

thumbnail
Figure 3. Effects of sedentary time, moderate-to-vigorous physical activity and fitness on HOMAIR in European and Mapuche participants.

Bars show mean ± SEM for all groups. p(interaction) describes the ‘tertile x environment’ interaction for HOMAIR after adjustment for potential confounding covariates (tertile cut-points shown in Table S2, see text or Table S4 for list of covariates).

https://doi.org/10.1371/journal.pone.0024690.g003

Effects of dietary intake on HOMAIR in Mapuches and Europeans

There was no significant influence of any component of dietary intake on HOMAIR.

Discussion

The main findings of this study were: 1) that Chilean Mapuches are more insulin resistant (as assessed by HOMAIR) than Chileans of European descent; 2) urbanisation appears to have a greater effect on insulin resistance in Mapuches than European Chileans – there was a three-fold difference in HOMAIR between Urban and Rural Mapuches, compared to less than a two-fold difference between Urban and Rural European Chileans, with a significant ‘ethnicity x environment’ interaction for this effect; 3) the differences between Urban and Rural populations for physical activity and sedentary time, and percentage body fat were similar between the two ethnic groups, suggesting that the greater difference in HOMAIR between Urban and Rural Mapuches compared to Europeans is not simply the consequence of a larger lifestyle shift between rural and urban environments in the former group; and 4) differences in adiposity, physical activity level and sedentary time had a greater influence on insulin resistance in Mapuches than Europeans.

The findings in relation to insulin resistance are consistent with the pattern observed in Pima Indians. A recent study reported that Pima Indians in the USA with normal glucose tolerance have higher HOMAIR values than Mexican Pimas even after adjustment for differences in BMI (4-fold higher in the unadjusted analysis) or in a BMI-matched sample [27]. In addition, US Pima Indians have been shown to be more insulin resistant than age- and adiposity-matched Americans of European descent [28]. In the present study the greater insulin resistance in the Urban, compared to Rural, Mapuche occurred in the absence of a higher BMI, although percentage body fat was greater in the Urban Mapuche group. However, although percentage body fat was greater in the Urban compared to Rural groups in our sample, the extent of the difference did not differ between the Mapuche and Chileans of European descent, indicating that increasing adiposity, in itself, cannot explain the disproportionately large increase in insulin resistance associated with urbanisation in the Mapuche population.

The present data indicate that increasing adiposity is associated with larger increases in insulin resistance in Mapuches than European Chileans. Moving from the lowest to the highest percentage body fat tertile was associated with an approximate doubling in HOMAIR in European Chileans, but a greater than 3.5-fold increase in HOMAIR in Mapuches. Thus, the data suggest that Mapuche are particularly susceptible to the adverse metabolic effects of obesity. This susceptibility may contribute to the disproportionately large difference in insulin resistance (and reported diabetes prevalence) between Mapuches living in Rural and Urban environments, despite the relatively modest difference in adiposity between these two groups. Similar observations have been made when considering other groups at increased risk of type 2 diabetes. For example, children [29] of South Asian origin exhibit larger increases in insulin resistance with increasing adiposity than White Europeans. Our data do not allow conclusions to be drawn about why adiposity had a more potent effect on insulin resistance in the Mapuche group. Speculatively, this could be due to differences in ectopic and regional fat distribution, in adipocyte size, or in adipose tissue signalling, between the groups, but further study is needed to test this hypothesis.

It is well established that high levels of physical activity [13] and cardiorespiratory fitness [14] are protective against type 2 diabetes and are associated with a favourable metabolic risk profile [30]. Recently, it has also become evident that increasing sedentary time is associated with increased diabetes risk [31], [32] and an adverse metabolic risk profile [33], [34] and that these effects may be independent of time spent in physical activity [32], [34]. Reports have shown that low levels of self-reported physical activity or low cardiorespiratory fitness are associated with insulin resistance in other indigenous groups [35]. The findings here are consistent with these observations, and extend them by using objective measures of physical activity and sedentary time and by revealing that the effects on insulin resistance of a low level of moderate-to-vigorous physical activity, or a high level of sedentary time was greater in the Mapuche than the European Chileans. For example, moving from the lowest to the highest tertile for sedentary time was associated with a 5.5-fold increase in insulin resistance in Chileans of European descent, but a 12.2-fold increase in Mapuches. Thus, the adverse effects of sedentariness and low physical activity appear particularly large in this indigenous group. Again, we cannot draw conclusions about why Mapuches are particularly susceptible to the adverse effects of low physical activity and high sedentary time on insulin resistance. However, it is clear that this effect is not simply mediated by differences in adiposity, fitness, dietary intake, smoking, or socioeconomic level or education between the Mapuche and European groups: statistically adjusting for these potential confounders did not influence the findings. Thus, further studies are needed to address the underlying mechanisms.

There was no significant influence of any dietary component on insulin resistance in either the Mapuche or European populations. This is consistent with other observational [36], [37] and intervention trial [38] data, in populations of European descent, showing no effect of diet on insulin resistance or diabetes risk once the effects of diet on adiposity are accounted for. Thus, despite the difference in a number of dietary variables between rural and urban populations being greater in Mapuches than Europeans, the present data indicate this larger dietary difference does not explain the larger effect of environment on insulin sensitivity in the Mapuches.

Data collection in this study presented a number of logistical challenges. The rural Mapuches generally live in remote areas far from road access or mains electricity. Data collection from some of these villages required transport of all equipment (including generators to provide electricity and centrifuges to spin blood) by horseback from site to site. Given these issues, the study is relatively large, with the final sample including 247 Mapuches and 225 European Chileans. However, we recognise that with a sample of this size our ability to draw strong conclusions about population-level group differences is somewhat limited. A particular strength of the study is the detailed phenotyping of the study populations, particularly with respect to lifestyle measures. Fitness, physical activity and sedentary time were objectively measured, providing greater validity in these measures than would be obtained from self-report questionnaire [39]. Dietary intake was also objectively measured by 7-day weighed food records – the best available method. Although underreporting is a common problem in the measurement of dietary intake [40], [41], this does not appear to have been a major factor here as reported energy intakes (means of ∼2900 kcal per day in men and ∼2500 kcal per day in women) were relatively high. Due to the data collection challenges and sample size, we used HOMAIR, rather than a euglycaemic clamp or intra-venous glucose tolerance test, to assess insulin sensitivity. It will be important to seek verification that our conclusions remain valid when insulin sensitivity is assessed using more sophisticated techniques. Because the study is cross-sectional in nature, it is not possible to draw firm conclusions about the causality of associations observed. However, our comprehensive study design, with relatively precise measures of a number of relevant exposure variables, allowed us to control for a number of potential confounding variables in our analyses. Although potential residual confounding effects cannot be excluded, the robustness of our findings to these adjustments supports our conclusions.

While the present findings have been adjusted for a number of potential confounders, it was not possible to collect data on birthweight in our cohort, so we were unable to control for potential effects of early life environment. There is a growing body of evidence that fetal environmental effects – which are reflected in size at birth – can influence adult health, potentially through epigenetic programming mechanisms [42]. These early life effects may interact with adult environment to determine disease risk: for example, it has been reported that low birth size was associated with increased risk of the metabolic syndrome in inactive, but not in active, middle-aged Finnish men [43]. However, a recent study found no differences in birthweights between Chileans of Mapuche and non-Mapuche origin born between 2000 and 2004 [44], although no such data are available for the generations represented in the present cohort. Thus, while it is possible that differences in early life environment could have contributed to the ethnicity x (adult) environment interactions observed in the present study, the data are not available to permit any firm conclusions to be drawn on this issue.

In conclusion, the present study reveals that urbanisation, adiposity, physical activity and sedentary behaviour influence insulin resistance to a greater extent in Chilean Mapuches than in Chileans of European descent. These associations persist after adjustment for a comprehensive range of potential confounding factors. The cross-sectional nature of these data does not allow firm conclusions to be drawn about causality, but the findings highlight the fact that environmental and lifestyle effects on metabolic risk differ between ethnic groups and suggest that further studies into the mechanisms underpinning this effect are needed. This has potential implications both for the design and implementation of lifestyle strategies to reduce metabolic risk in different ethnic groups, and for advancing the understanding of the mechanisms underpinning human insulin resistance.

Supporting Information

Table S1.

Number of participants with valid data by data category.

https://doi.org/10.1371/journal.pone.0024690.s001

(DOC)

Table S2.

Tertile cut-points for obesity-related phenotypes, physical activity and fitness for men and women.

https://doi.org/10.1371/journal.pone.0024690.s002

(DOC)

Table S3.

Effects of BMI, waist circumference and body fat on HOMAIR in European and Mapuche men and women in age-sex and fully-adjusted models.

https://doi.org/10.1371/journal.pone.0024690.s003

(DOC)

Table S4.

Effects of sedentary time, moderate-to-vigorous physical activity and fitness on HOMAIR in European and Mapuche men and women in age-sex and fully-adjusted models.

https://doi.org/10.1371/journal.pone.0024690.s004

(DOC)

Acknowledgments

We thank all participants for their cooperation, and CNEF-Concepcion, the Alhualef Comolai community, CESFAM-Ralco, and the Hospital of Panguipulli for invaluable assistance in organising the phenotyping and sample collection. We would especially like to thank the Mapuche community leaders for facilitating this study and Alison Teyhan and Richard Wilson for their assistance with data analysis.

Author Contributions

Conceived and designed the experiments: CAC-M FP-B NU CC MESB JMRG. Performed the experiments: CAC-M FP-B LI RS EH. Analyzed the data: CAC-M MESB JMRG. Wrote the paper: CAC-M MESB JMRG.

References

  1. 1. International Diabetes Federation (2009) IDF Diabetes Atlas. Brussels: International Diabetes Federation. International Diabetes Federation2009IDF Diabetes Atlas.BrusselsInternational Diabetes Federation
  2. 2. Zimmet P (2000) Globalization, coca-colonization and the chronic disease epidemic: can the Doomsday scenario be averted? J Intern Med 247: 301–310. jim625 [pii].P. Zimmet2000Globalization, coca-colonization and the chronic disease epidemic: can the Doomsday scenario be averted?J Intern Med247301310. jim625 [pii]
  3. 3. Schulz LO, Bennett PH, Ravussin E, Kidd JR, Kidd KK, et al. (2006) Effects of traditional and western environments on prevalence of type 2 diabetes in Pima Indians in Mexico and the U.S. Diabetes Care. 29. : 1866–1871.LO SchulzPH BennettE. RavussinJR KiddKK Kidd2006Effects of traditional and western environments on prevalence of type 2 diabetes in Pima Indians in Mexico and the U.S.Diabetes Care291866187129/8/1866 [pii];10.2337/dc06-0138 [doi]. 29/8/1866 [pii];10.2337/dc06-0138 [doi].
  4. 4. Yu CH, Zinman B (2007) Type 2 diabetes and impaired glucose tolerance in aboriginal populations: a global perspective. Diabetes Res Clin Pract. 78. : 159–170.CH YuB. Zinman2007Type 2 diabetes and impaired glucose tolerance in aboriginal populations: a global perspective.Diabetes Res Clin Pract78159170S0168-8227(07)00249-5 [pii];10.1016/j.diabres.2007.03.022 [doi]. S0168-8227(07)00249-5 [pii];10.1016/j.diabres.2007.03.022 [doi].
  5. 5. Perez-Bravo F, Carrasco E, Santos JL, Calvillan M, Larenas G, et al. (2001) Prevalence of type 2 diabetes and obesity in rural Mapuche population from Chile. Nutrition. 17. : 236–238.F. Perez-BravoE. CarrascoJL SantosM. CalvillanG. Larenas2001Prevalence of type 2 diabetes and obesity in rural Mapuche population from Chile.Nutrition17236238S0899900700005505 [pii]. S0899900700005505 [pii].
  6. 6. Larenas G, Arias G, Espinoza O, Charles M, Landaeta O, et al. (1985) [Prevalence of diabetes mellitus in a Mapuche community of Region IX, Chile]. Rev Med Chil 113: 1121–1125.G. LarenasG. AriasO. EspinozaM. CharlesO. Landaeta1985[Prevalence of diabetes mellitus in a Mapuche community of Region IX, Chile].Rev Med Chil11311211125
  7. 7. Carrasco EP, Perez FB, Angel BB, Albala CB, Santos JL, et al. (2004) [Prevalence of type 2 diabetes and obesity in two Chilean aboriginal populations living in urban zones]. Rev Med Chil 132: 1189–1197.EP CarrascoFB PerezBB AngelCB AlbalaJL Santos2004[Prevalence of type 2 diabetes and obesity in two Chilean aboriginal populations living in urban zones].Rev Med Chil13211891197
  8. 8. Perez-Bravo F, Fuentes M, Angel B, Sanchez H, Carrasco E, et al. (2006) Lack of association between the fatty acid binding protein 2 (FABP2) polymorphism with obesity and insulin resistance in two aboriginal populations from Chile. Acta Diabetol. 43. : 93–98.F. Perez-BravoM. FuentesB. AngelH. SanchezE. Carrasco2006Lack of association between the fatty acid binding protein 2 (FABP2) polymorphism with obesity and insulin resistance in two aboriginal populations from Chile.Acta Diabetol43939810.1007/s00592-006-0221-7 [doi]. 10.1007/s00592-006-0221-7 [doi].
  9. 9. Instituto Nacional de Estadísticas Chile (2004) Resultados Oficiales Censo de Población 2002. Instituto Nacional de Estadísticas Chile2004Resultados Oficiales Censo de Población 2002
  10. 10. Baechler R, Mujica V, Aqueveque X, Ramos L, Soto A (2002) [Prevalence of diabetes mellitus in the Seventh Region of Chile]. Rev Med Chil 130: 1257–1264.R. BaechlerV. MujicaX. AquevequeL. RamosA. Soto2002[Prevalence of diabetes mellitus in the Seventh Region of Chile].Rev Med Chil13012571264
  11. 11. King H, Taylor R, Zimmet P, Pargeter K, Raper LR, et al. (1984) Non-insulin-dependent diabetes (NIDDM) in a newly independent Pacific nation: the Republic of Kiribati. Diabetes Care 7: 409–415.H. KingR. TaylorP. ZimmetK. PargeterLR Raper1984Non-insulin-dependent diabetes (NIDDM) in a newly independent Pacific nation: the Republic of Kiribati.Diabetes Care7409415
  12. 12. Collins VR, Dowse GK, Toelupe PM, Imo TT, Aloaina FL, et al. (1994) Increasing prevalence of NIDDM in the Pacific island population of Western Samoa over a 13-year period. Diabetes Care 17: 288–296.VR CollinsGK DowsePM ToelupeTT ImoFL Aloaina1994Increasing prevalence of NIDDM in the Pacific island population of Western Samoa over a 13-year period.Diabetes Care17288296
  13. 13. Gill JM, Cooper AR (2008) Physical activity and prevention of type 2 diabetes mellitus. Sports Med 38: 807–824.JM GillAR Cooper2008Physical activity and prevention of type 2 diabetes mellitus.Sports Med38807824
  14. 14. Wei M, Gibbons LW, Mitchell TL, Kampert JB, Lee CD, et al. (1999) The association between cardiorespiratory fitness and impaired fasting glucose and type 2 diabetes mellitus in men. Ann Intern Med 130: 89–96.M. WeiLW GibbonsTL MitchellJB KampertCD Lee1999The association between cardiorespiratory fitness and impaired fasting glucose and type 2 diabetes mellitus in men.Ann Intern Med1308996
  15. 15. Marfell-Jones M, Olds T, Stewart A, Carter L (2006) International standards for anthropometric assessment. Potchefstroom, South Africa: ISAK. M. Marfell-JonesT. OldsA. StewartL. Carter2006International standards for anthropometric assessment.Potchefstroom, South AfricaISAK
  16. 16. Durnin JVGA, Womersley J (1974) Body fat assessed from total body density and its estimation from skinfold thickness: measurements on 481 men and women aged from 16 to 72 years. British Journal of Nutrition 32: 77–97.JVGA DurninJ. Womersley1974Body fat assessed from total body density and its estimation from skinfold thickness: measurements on 481 men and women aged from 16 to 72 years.British Journal of Nutrition327797
  17. 17. El Assaad MA, Topouchian JA, Asmar RG (2003) Evaluation of two devices for self-measurement of blood pressure according to the international protocol: the Omron M5-I and the Omron 705IT. Blood Press Monit. 8. : 127–133.MA El AssaadJA TopouchianRG Asmar2003Evaluation of two devices for self-measurement of blood pressure according to the international protocol: the Omron M5-I and the Omron 705IT.Blood Press Monit812713310.1097/01.mbp.0000087393.96145.b1 [doi]. 10.1097/01.mbp.0000087393.96145.b1 [doi].
  18. 18. Buckley JP, Sim J, Eston RG, Hession R, Fox R (2004) Reliability and validity of measures taken during the Chester step test to predict aerobic power and to prescribe aerobic exercise. Br J Sports Med 38: 197–205.JP BuckleyJ. SimRG EstonR. HessionR. Fox2004Reliability and validity of measures taken during the Chester step test to predict aerobic power and to prescribe aerobic exercise.Br J Sports Med38197205
  19. 19. Friedewald WT, Levy RI, Fredrickson DS (1972) Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem 18: 499–502.WT FriedewaldRI LevyDS Fredrickson1972Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge.Clin Chem18499502
  20. 20. Méndez R (1997) El nivel socioeconómico Esomar. Santiago: VIII Congreso Chileno de Marketing de Icare. R. Méndez1997El nivel socioeconómico Esomar.SantiagoVIII Congreso Chileno de Marketing de Icare
  21. 21. MIDEPLAN (2006) Encuesta de Caracterizacion Socioeconomica Nacional (CASEN). Santiago: Gobierno de Chile. MIDEPLAN2006Encuesta de Caracterizacion Socioeconomica Nacional (CASEN).SantiagoGobierno de Chile
  22. 22. Acosta AM, Escalona M, Maiz A, Pollak F, Leighton F (2002) [Determination of the insulin resistance index by the Homeostasis Model Assessment in a population of Metropolitan Region in Chile]. Rev Med Chil 130: 1227–1231.AM AcostaM. EscalonaA. MaizF. PollakF. Leighton2002[Determination of the insulin resistance index by the Homeostasis Model Assessment in a population of Metropolitan Region in Chile].Rev Med Chil13012271231
  23. 23. Freedson PS, Melanson E, Sirard J (1998) Calibration of the Computer Science and Applications, Inc. accelerometer. Med Sci Sports Exerc 30: 777–781.PS FreedsonE. MelansonJ. Sirard1998Calibration of the Computer Science and Applications, Inc. accelerometer.Med Sci Sports Exerc30777781
  24. 24. Troiano RP, Berrigan D, Dodd KW, Masse LC, Tilert T, et al. (2008) Physical activity in the United States measured by accelerometer. Med Sci Sports Exerc. 40. : 181–188.RP TroianoD. BerriganKW DoddLC MasseT. Tilert2008Physical activity in the United States measured by accelerometer.Med Sci Sports Exerc4018118810.1249/mss.0b013e31815a51b3 [doi]. 10.1249/mss.0b013e31815a51b3 [doi].
  25. 25. Hagstromer M, Oja P, Sjostrom M (2007) Physical activity and inactivity in an adult population assessed by accelerometry. Med Sci Sports Exerc. 39. : 1502–1508.M. HagstromerP. OjaM. Sjostrom2007Physical activity and inactivity in an adult population assessed by accelerometry.Med Sci Sports Exerc391502150810.1249/mss.0b013e3180a76de5 [doi];00005768-200709000-00010 [pii]. 10.1249/mss.0b013e3180a76de5 [doi];00005768-200709000-00010 [pii].
  26. 26. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, et al. (1985) Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 28: 412–419.DR MatthewsJP HoskerAS RudenskiBA NaylorDF Treacher1985Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man.Diabetologia28412419
  27. 27. Esparza-Romero J, Valencia ME, Martinez ME, Ravussin E, Schulz LO, et al. (2010) Differences in Insulin Resistance in Mexican and U.S. Pima Indians with Normal Glucose Tolerance. J Clin Endocrinol Metab. pp. jc.2010–0297.J. Esparza-RomeroME ValenciaME MartinezE. RavussinLO Schulz2010Differences in Insulin Resistance in Mexican and U.S. Pima Indians with Normal Glucose Tolerance.J Clin Endocrinol Metabjc.20100297[pii];10.1210/jc.2010-0297 [doi]. [pii];10.1210/jc.2010-0297 [doi].
  28. 28. Stefan N, Stumvoll M, Weyer C, Bogardus C, Tataranni PA, et al. (2004) Exaggerated insulin secretion in Pima Indians and African-Americans but higher insulin resistance in Pima Indians compared to African-Americans and Caucasians. Diabet Med. 21. : 1090–1095.N. StefanM. StumvollC. WeyerC. BogardusPA Tataranni2004Exaggerated insulin secretion in Pima Indians and African-Americans but higher insulin resistance in Pima Indians compared to African-Americans and Caucasians.Diabet Med211090109510.1111/j.1464-5491.2004.01290.x [doi];DME1290 [pii]. 10.1111/j.1464-5491.2004.01290.x [doi];DME1290 [pii].
  29. 29. Whincup PH, Gilg JA, Papacosta O, Seymour C, Miller GJ, et al. (2002) Early evidence of ethnic differences in cardiovascular risk: cross sectional comparison of British South Asian and white children. BMJ 324: 635.PH WhincupJA GilgO. PapacostaC. SeymourGJ Miller2002Early evidence of ethnic differences in cardiovascular risk: cross sectional comparison of British South Asian and white children.BMJ324635
  30. 30. Gill JM, Malkova D (2006) Physical activity, fitness and cardiovascular disease risk in adults: interactions with insulin resistance and obesity. Clin Sci (Lond) 110: 409–425.JM GillD. Malkova2006Physical activity, fitness and cardiovascular disease risk in adults: interactions with insulin resistance and obesity.Clin Sci (Lond)110409425
  31. 31. Dunstan DW, Salmon J, Owen N, Armstrong T, Zimmet PZ, et al. (2004) Physical activity and television viewing in relation to risk of undiagnosed abnormal glucose metabolism in adults. Diabetes Care. 27. : 2603–2609.DW DunstanJ. SalmonN. OwenT. ArmstrongPZ Zimmet2004Physical activity and television viewing in relation to risk of undiagnosed abnormal glucose metabolism in adults.Diabetes Care272603260927/11/2603 [pii]. 27/11/2603 [pii].
  32. 32. Hu FB, Li TY, Colditz GA, Willett WC, Manson JE (2003) Television watching and other sedentary behaviors in relation to risk of obesity and type 2 diabetes mellitus in women. JAMA. 289. : 1785–1791.FB HuTY LiGA ColditzWC WillettJE Manson2003Television watching and other sedentary behaviors in relation to risk of obesity and type 2 diabetes mellitus in women.JAMA2891785179110.1001/jama.289.14.1785 [doi];289/14/1785 [pii]. 10.1001/jama.289.14.1785 [doi];289/14/1785 [pii].
  33. 33. Healy GN, Dunstan DW, Salmon J, Shaw JE, Zimmet PZ, et al. (2008) Television time and continuous metabolic risk in physically active adults. Med Sci Sports Exerc. 40. : 639–645.GN HealyDW DunstanJ. SalmonJE ShawPZ Zimmet2008Television time and continuous metabolic risk in physically active adults.Med Sci Sports Exerc4063964510.1249/MSS.0b013e3181607421 [doi]. 10.1249/MSS.0b013e3181607421 [doi].
  34. 34. Gill JM, Bhopal R, Douglas A, Wallia S, Bhopal R, et al. (2011) Sitting Time and Waist Circumference Are Associated With Glycemia in U.K. South Asians: Data from 1,228 adults screened for the PODOSA trial. Diabetes Care. 34. : 1214–1218.JM GillR. BhopalA. DouglasS. WalliaR. Bhopal2011Sitting Time and Waist Circumference Are Associated With Glycemia in U.K. South Asians: Data from 1,228 adults screened for the PODOSA trial.Diabetes Care3412141218dc10-2313 [pii];10.2337/dc10-2313 [doi]. dc10-2313 [pii];10.2337/dc10-2313 [doi].
  35. 35. Kriska AM, Hanley AJ, Harris SB, Zinman B (2001) Physical activity, physical fitness, and insulin and glucose concentrations in an isolated Native Canadian population experiencing rapid lifestyle change. Diabetes Care 24: 1787–1792.AM KriskaAJ HanleySB HarrisB. Zinman2001Physical activity, physical fitness, and insulin and glucose concentrations in an isolated Native Canadian population experiencing rapid lifestyle change.Diabetes Care2417871792
  36. 36. Schulze MB, Liu S, Rimm EB, Manson JE, Willett WC, et al. (2004) Glycemic index, glycemic load, and dietary fiber intake and incidence of type 2 diabetes in younger and middle-aged women. Am J Clin Nutr. 80. : 348–356.MB SchulzeS. LiuEB RimmJE MansonWC Willett2004Glycemic index, glycemic load, and dietary fiber intake and incidence of type 2 diabetes in younger and middle-aged women.Am J Clin Nutr8034835680/2/348 [pii]. 80/2/348 [pii].
  37. 37. van Dam RM, Willett WC, Rimm EB, Stampfer MJ, Hu FB (2002) Dietary fat and meat intake in relation to risk of type 2 diabetes in men. Diabetes Care 25: 417–424.RM van DamWC WillettEB RimmMJ StampferFB Hu2002Dietary fat and meat intake in relation to risk of type 2 diabetes in men.Diabetes Care25417424
  38. 38. Jebb SA, Lovegrove JA, Griffin BA, Frost GS, Moore CS, et al. (2010) Effect of changing the amount and type of fat and carbohydrate on insulin sensitivity and cardiovascular risk: the RISCK (Reading, Imperial, Surrey, Cambridge, and Kings) trial. Am J Clin Nutr. 92. : 748–758.SA JebbJA LovegroveBA GriffinGS FrostCS Moore2010Effect of changing the amount and type of fat and carbohydrate on insulin sensitivity and cardiovascular risk: the RISCK (Reading, Imperial, Surrey, Cambridge, and Kings) trial.Am J Clin Nutr92748758ajcn.2009.29096 [pii];10.3945/ajcn.2009.29096 [doi]. ajcn.2009.29096 [pii];10.3945/ajcn.2009.29096 [doi].
  39. 39. Shephard RJ (2003) Limits to the measurement of habitual physical activity by questionnaires. Br J Sports Med 37: 197–206.RJ Shephard2003Limits to the measurement of habitual physical activity by questionnaires.Br J Sports Med37197206
  40. 40. Westerterp KR, Goris AH (2002) Validity of the assessment of dietary intake: problems of misreporting. Curr Opin Clin Nutr Metab Care 5: 489–493.KR WesterterpAH Goris2002Validity of the assessment of dietary intake: problems of misreporting.Curr Opin Clin Nutr Metab Care5489493
  41. 41. Black AE, Cole TJ (2001) Biased over- or under-reporting is characteristic of individuals whether over time or by different assessment methods. J Am Diet Assoc 101: 70–80.AE BlackTJ Cole2001Biased over- or under-reporting is characteristic of individuals whether over time or by different assessment methods.J Am Diet Assoc1017080
  42. 42. Wadhwa PD, Buss C, Entringer S, Swanson JM (2009) Developmental origins of health and disease: brief history of the approach and current focus on epigenetic mechanisms. Semin Reprod Med. 27. : 358–368.PD WadhwaC. BussS. EntringerJM Swanson2009Developmental origins of health and disease: brief history of the approach and current focus on epigenetic mechanisms.Semin Reprod Med2735836810.1055/s-0029-1237424 [doi]. 10.1055/s-0029-1237424 [doi].
  43. 43. Laaksonen DE, Lakka HM, Lynch J, Lakka TA, Niskanen L, et al. (2003) Cardiorespiratory fitness and vigorous leisure-time physical activity modify the association of small size at birth with the metabolic syndrome. Diabetes Care 26: 2156–2164.DE LaaksonenHM LakkaJ. LynchTA LakkaL. Niskanen2003Cardiorespiratory fitness and vigorous leisure-time physical activity modify the association of small size at birth with the metabolic syndrome.Diabetes Care2621562164
  44. 44. Amigo H, Bustos P, Kaufman JS (2010) Absence of disparities in anthropometric measures among Chilean indigenous and non-indigenous newborns. BMC Public Health. 10. 392 p.H. AmigoP. BustosJS Kaufman2010Absence of disparities in anthropometric measures among Chilean indigenous and non-indigenous newborns.BMC Public Health101471-2458-10-392 [pii];10.1186/1471-2458-10-392 [doi]. 392 1471-2458-10-392 [pii];10.1186/1471-2458-10-392 [doi].