The role of childhood social position in adult type 2 diabetes: evidence from the English Longitudinal Study of Ageing
© Pikhartova et al.; licensee BioMed Central Ltd. 2014
Received: 26 September 2013
Accepted: 20 May 2014
Published: 26 May 2014
Socioeconomic circumstances in childhood and early adulthood may influence the later onset of chronic disease, although such research is limited for type 2 diabetes and its risk factors at the different stages of life. The main aim of the present study is to examine the role of childhood social position and later inflammatory markers and health behaviours in developing type 2 diabetes at older ages using a pathway analytic approach.
Data on childhood and adult life circumstances of 2,994 men and 4,021 women from English Longitudinal Study of Ageing (ELSA) were used to evaluate their association with diabetes at age 50 years and more. The cases of diabetes were based on having increased blood levels of glycated haemoglobin and/or self-reported medication for diabetes and/or being diagnosed with type 2 diabetes. Father’s job when ELSA participants were aged 14 years was used as the measure of childhood social position. Current social characteristics, health behaviours and inflammatory biomarkers were used as potential mediators in the statistical analysis to assess direct and indirect effects of childhood circumstances on diabetes in later life.
12.6 per cent of participants were classified as having diabetes. A disadvantaged social position in childhood, as measured by father’s manual occupation, was associated at conventional levels of statistical significance with an increased risk of type 2 diabetes in adulthood, both directly and indirectly through inflammation, adulthood social position and a risk score constructed from adult health behaviours including tobacco smoking and limited physical activity. The direct effect of childhood social position was reduced by mediation analysis (standardised coefficient decreased from 0.089 to 0.043) but remained statistically significant (p = 0.035). All three indirect pathways made a statistically significantly contribution to the overall effect of childhood social position on adulthood type 2 diabetes.
Childhood social position influences adult diabetes directly and indirectly through inflammatory markers, adulthood social position and adult health behaviours.
KeywordsELSA Diabetes Longitudinal Glycated haemoglobin Inflammation
Early life socio-demographic characteristics and their influence on the onset of adult chronic disease have become recently the focus of more intensive research , with the findings showing that a disadvantaged social position in childhood is associated with higher morbidity and mortality later in life [2, 3].
Diabetes mellitus is a serious disease affecting a significant proportion of the population ; and its incidence and prevalence are increasing (in Britain, for example, its prevalence increased between 1995 and 2006 from 2.8% to 4.5%) . A wide range of risk factors for diabetes have been proposed, including age, ethnicity, family history, intra-uterine growth, childhood health and obesity and, later in life, BMI or waist circumference and physical activity, although the results sometimes have been inconsistent [5–11]. The role of tobacco smoking is unclear , with some studies finding an indirect effect of smoking on the development of glucose intolerance .
The association between adulthood socioeconomic position and diabetes morbidity and mortality is well-known , although it may differ by gender, and some studies did not find such association at older ages . The role of adverse childhood social environment in development of increased insulin resistance or diabetes is less clear. Direct association between social position in childhood and increased insulin resistance has been found only very rarely in the previous research  and some studies have questioned such association mainly because the association between childhood social conditions and diabetes was usually explained by socioeconomic position in adulthood [10, 17, 18]. Few studies proposed pathways through which adverse childhood environment influenced later onset of type 2 diabetes indirectly, such as through stress originating from poorer living conditions in childhood, earlier health problems, financial security or family stability, or through unhealthy behaviours and overweight [8, 19–23]. Inflammation has also been proposed as possible link between low social status and type 2 diabetes . Inflammatory markers were found elevated in those with risky health behaviours which in turn are closely associated with lower socioeconomic position and stressful life and also with type 2 diabetes . Stress can stimulate HPA axis which can stimulate higher production of cortisol and consequently influence levels of inflammatory markers, production of insulin or adipose tissues gain [24, 25]. Several studies showed the association between inflammatory markers and type 2 diabetes [26, 27].
The concepts of life-course epidemiology can be used to investigate the influence of factors in childhood and adulthood on the later onset of diabetes and other diseases. It can be used to identify particularly the sensitive periods where socioeconomic factors play a role in the development of diabetes. It can be useful also for evaluating the magnitude of any direct association between childhood circumstances and the later onset of diabetes; as well as to describe a series of life stages and help to assess whether the role of socioeconomic exposures accumulates over time and across the life-cycle. This life course methodology has been used by many researchers to evaluate the effect of adversity in childhood and other life stages on the risk of later disease onset [28, 29], but it has been used little in diabetes research.
In the present paper we use data from the English Longitudinal Study of Ageing (ELSA). The main aim is to examine the role of childhood and adulthood socioeconomic position, inflammatory markers and health behaviours in developing type 2 diabetes at older ages using a life course analytical approach in a sample of men and women aged 50 years and more. The life course approach has been used little in diabetes research and has not been used previously to analyse diabetes outcomes in ELSA. We use multiple imputations to maximise the effective sample size and path analysis to assess direct and indirect effects.
Study population and study sample
Data for this analysis were derived from the English Longitudinal Study of Ageing (ELSA). The ELSA sample consists of people aged 50 years and more who were selected from participants in the Health Surveys for England (HSE) in 1998, 1999 and 2001, constructed to represent the English population over 50 years of age living in private addresses. Those who agreed to participate were invited to wave 1 of ELSA in 2002–03, with subsequent follow-up every two years. Wave 1 data were collected by interviews alone, but Waves 2 and 4 included physical examinations and blood samples collection. Ethical approval for all the ELSA waves was granted by the National Research and Ethics Committee. All participants gave informed consent. More information on ELSA can be found at http://www.ifs.org.uk/elsa/documentation.php. This analysis includes data from all four waves (including nurse’s part of wave 4). Data were accessed through the Economic and Social Data Service.
The present study sample consists of the 7,015 men and women, from the original 11,392 ELSA members, who were screened at Waves 1 and 4 and had valid outcome data (described in the next section). Loss to follow-up in ELSA is associated with male gender, morbidity, older age and a disadvantaged social position . Nurses invited some three-quarters of Wave 4 participants to give a blood sample, excluding those who refused informed consent or who had a history of fits or convulsion, bleeding or clotting disorders or were prescribed anticoagulants. Glycated haemoglobin (HbA1c) samples were finally available for 4,245 individuals of 7,015 individuals in the analytical sample. All participants from English Longitudinal Study of Ageing signed full informed consent to participate in the study.
Construction of diabetes diagnosis variable
The outcome variable was based on combination of the following four conditions. Individuals were classified as having diabetes if the level of glycated haemoglobin (Hb1Ac) exceeded 6.5% (>48 mmol/mol, following International Diabetes Federation recommendations) when measured in wave 4 and/or if they answered positively at least one of the three following questions: “Has a doctor ever told you that you have diabetes or high blood sugar?”, “Do you currently inject insulin for diabetes?”, and “Are you currently taking any tablets, pills or other medicines that you swallow for diabetes?”. Fasting plasma glucose was not used as there was insufficient response to the fasting glucose blood test compared to HbA1c blood test, which did not require fasting. This self-reported information was derived from all four waves of ELSA.
The ascertainment of diabetes cases in ELSA in this analysis is as complete as possible with the available data. Using only the self-reported doctor diagnosis, one case in three would be missed.
Explanatory variables characterizing childhood and adulthood
A range of variables characterizing childhood and adulthood (including current life) was used in the analysis. The core part of the ELSA dataset supplied the father’s occupation when the participant was aged 14 years, used as a dichotomous measure of the manual employment of the father. Other explanatory variables used to obtain information about adulthood and current life and health status included current risk factor score and recent social position.
The current risk factor score was constructed from: waist circumference measured during nurse examination at wave 4; information about level of physical activity collected at wave 4; and tobacco smoking assessed across all four waves. Guided by the literature that fat tissue behaves as an independent organ, producing hormones that contribute to systemic inflammation and the development of insulin-resistance, and that central obesity is sufficient instrument to measure the risk of being overweight [31, 32], only waist measurement was used in our analysis. The current risk factor score was constructed as gender and age specific.
C-reactive protein and blood fibrinogen from wave 4 were grouped into deciles, coded 0–9. The inflammatory score was constructed combining C-reactive protein and blood fibrinogen decile scores.
Descriptive characteristics were obtained for an analytical sample including missing data. For the level of missingness in these data, 20 imputations are considered sufficient, so 20 imputed datasets were produced using Bayesian imputation methods in MPlus version 6.12. Afterwards a sensitivity analysis was performed on both the original and imputed datasets, to check that the imputation had not introduced bias. Logistic regression was used to assess bivariate associations between diabetes and the socioeconomic and health variables from childhood and adulthood. Age and sex- adjusted estimates of odds ratios (OR) were calculated together with 95% confidence intervals (CI). Sex interaction was tested in all steps of the analysis, and as there was not any statistically significant differences between the results for men and women, our results are being presented as sex- adjusted rather than stratified by sex. First, the association between childhood socioeconomic position and type 2 diabetes at older ages was assessed. Then, the association between the adult factors (inflammatory score, risk factor score, and socioeconomic position) and both childhood socioeconomic position and type 2 diabetes was assessed. All descriptive and regression analysis was done in STATA 11/12. The possible mediating role of inflammatory markers, current risk factors and current social position was evaluated in pathway analysis in MPlus version 6.12. Pathway analysis allowed assessing the direct and indirect relationship between childhood socioeconomic position and type 2 diabetes using several mediators simultaneously.
Data were available for 7,015 study participants after applying all eliminating processes. In comparison with the overall sample, the subgroup used in the present analyses was slightly older (68.9 years compared to 65.2 in the Wave 4 core sample); had slightly less males (42.7 per cent in our sample compared to 44.9%); the distribution of current social position was similar to the main dataset, with slightly higher proportion of males in the most disadvantaged social class (21.1 per cent compared to 19.3%) and a lower proportion of females in the same class (17.7 per cent compared to 20.9%). The prevalence of type 2 diabetes in the Wave 4 core sample, and in our sub-sample, and the prevalence in England in 2009 did not differ substantially (males in England respectively 14.2 per cent, 14.8% and 14.1%; in females 10.6%, 11.0% and 11.1%).
Distribution of social and demographic characteristics in the analytical sample, English Longitudinal Study of Ageing 2002-2008
Childhood social position
Non-manual father’s occupation
Manual father’s occupation
Current social position
High waist circumference (cm)
>102 cm (M)
>88 cm (F)
<102 cm (M)
<88 cm (F)
Yes (>3.00 g/l)
Bivariate associations between variables used in pathway model (Age-sex adjusted OR, 95% CI and level of significance)
OR (95% CI)
Childhood social position at age 14
Type 2 diabetes
Current social position1
Current risk factor score
Inflammatory markers score
Current social position1
Type 2 diabetes
Current risk factors score
Type 2 diabetes
Inflammatory markers score
Type 2 diabetes
Model estimates and model fit statistics for direct and indirect associations between childhood SEP and diabetes mellitus in older age
Indirect association – mediation through inflammation
Indirect Association – Mediation through Current SEP, Current Risk Factors and Inflammation
Type 2 diabetes on:
Type 2 diabetes on:
Current SEP on:
Current risk factors on:
Type 2 diabetes on:
Current risk factors
In a national representative sample of men and women aged 50 years and more our results show that a disadvantaged social position in childhood, measured as a father’s manual occupation, is associated with an increased risk of type 2 diabetes in adulthood. It has been shown also that a disadvantaged social position in childhood increases the odds of diabetes both directly and indirectly through an inflammatory score, adulthood social position and an adulthood risk score constructed from waist circumference, tobacco smoking and physical activity.
These findings need to be discussed further in the light of the literature reviewed earlier, but first several methodological issues must be addressed. Loss to follow-up of individuals between the waves of ELSA data collection might have introduced selection bias. Recent articles using ELSA data suggest that sample attrition is greater among those who were in a disadvantaged socioeconomic position at the start of the study. Demakakos et al. , meaning that our results are likely to underestimate the size of the real association between childhood social position and type 2 diabetes ( the comparison of the whole ELSA sample and our analytical sample suggests that any such underestimation may be small). In addition to attrition, the ELSA sample excludes those residing in institutions such as nursing homes. Although the proportion of such individuals in the population is small, we cannot claim that the ELSA sample is entirely nationally representative. On the other hand, ELSA’s large sample size is a major strength of the present study. Data imputation further maximised the effective sample size, and as such is major advantage of this paper in comparison to a previously published ELSA analysis on a similar topic . Next, the diabetes variable was constructed mainly from self-reported information. Participants were asked to self-report whether they had been told by a doctor that they had diabetes mellitus or high blood sugar, whether they use pills or other medicines because of diabetes or high blood sugar, and whether they used insulin injections. Reliance on self-reports may have introduced reporting bias. Recall bias is also an issue in this analysis. As ELSA is a study of middle aged and older people, questions relating to childhood were asked several decades after the event, although any recall bias hopefully was minimised by the use of a computerised life-grid to collect the retrospective information. Further, residual confounding might be an issue. Although we have controlled for a range of relevant variables, there are still some variables, particularly data on diet, which would further improve the analytical model. The major methodological strength of this analysis, however, is the use of a pathway analytical approach to assess the links between childhood social position and type 2 diabetes in later life.
A recent paper reported, the association between socioeconomic characteristics in early life and diabetes mellitus in the English Longitudinal Study of Ageing using standard regression modelling . Standard regression modelling techniques cannot precisely assess the mediating role of variables assumed to be on the pathway between exposures and outcomes; nor can they explicitly distinguish between the direct and indirect effects of such exposures. Our approach to the analysis, allowing evaluation of potential pathways, gives more insight into the relationship between childhood social circumstances and later diabetes. Finally, intermediary pathways between current social position and current risk factor score and inflammatory markers score could be correlated and, when tested, the effects of some pathways, such as via the inflammatory markers score, could be moderated and could become non-significant.
Despite some methodological limitations this analysis brings important findings. The estimated effect of childhood SEP is stronger than in most previous studies. It is possible that the use of data imputation and pathway analysis produced more precise measures, with smaller selection and recall biases, than previous studies which did not use such methods. For the population in the age range of interest to this analysis, this approach might be the best available in terms of data availability (the birth cohorts studies, with their prospective data from childhood, at present are younger than the ELSA respondents).
Our findings confirm the relationship between childhood socioeconomic position and adult type 2 diabetes mellitus and previously hypothesised pathways through current risk factors, current socioeconomic position and levels of inflammatory markers. The association between various measures of childhood socioeconomic position (education; income) and diabetes has been reported in some previous longitudinal studies of women [33, 34], although not all ; and among men, such associations are reported as either weak or non-existent [10, 33, 35–37]. The previous analysis of ELSA similarly reported a weak association between childhood socioeconomic position and incident diabetes in women but no association in men . However, their study focused on incident cases of diabetes within the study duration, and it is possible that diabetes cases in late life may be less strongly related to early childhood social circumstances than all cases including those at younger ages used in our analysis. Additionally, the number of incident cases of diabetes is limited in the ELSA dataset making estimates of the association between childhood socioeconomic position and incident diabetes relatively imprecise, while our analysis by using all cases and data imputation allowed more precise estimates.
Future replication of our findings in different cohorts is needed using a broader range of variables and focusing on other alternative pathways. The evidence from this study showing that childhood social circumstances influence health status in older age both directly and indirectly could provide strong arguments for policy makers to focus on socio-economic conditions of families with children aiming to reduce health inequalities across the entire life course.
In conclusion, a disadvantaged social position in childhood is associated with an increased risk of diabetes in people aged 50 years and over, with social position in adulthood, health behaviours and inflammatory markers mediating this relationship. The direct role of childhood social conditions is reduced after taking account of these mediators, but it does not disappear entirely. Diabetes prevention programmes thus should concentrate on social disadvantage from early stages of life in addition to the established risk factors.
Body mass index
English Longitudinal Study of Ageing
Economic and Social Data Service
This research was supported by the Economic and Social Research Council-funded International Centre for Life Course Studies in Society and Health (RES-596-28-0001). The authors thank ELSA participants and researchers; and ESDS for enabling use of ELSA data for this analysis.
- Everson SA, Maty SC, Lynch JW, Kaplan GA: Epidemiologic evidence for the relation between socioeconomic status and depresion, obesity, and diabetes. J Psychosomat Res. 2002, 53: 891-895. 10.1016/S0022-3999(02)00303-3.View ArticleGoogle Scholar
- Lawlor DA, Sterne JAC, Tynelius P, Davey Smith G, Rasmussen F: Association of childhood socioeconomic position with cause-specific mortality in a prospective record linkage study of 1,839,384 individuals. Am J Epidemiol. 2006, 164: 907-915. 10.1093/aje/kwj319.View ArticlePubMedGoogle Scholar
- Galobardes B, Davey Smith G, Lynch JW: System review of the influence of childhood socioeconomic circumstances on risk for cardiovascular disease in adulthood. Ann Epidemiol. 2006, 16: 91-104. 10.1016/j.annepidem.2005.06.053.View ArticlePubMedGoogle Scholar
- Gonzales ELM, Johansson S, Wallander MA, Rodriguez LAG: Trends in the prevalence and incidence of diabetes in the UK: 1996–2005. J Epidemiol Commun Health. 2009, 63: 332-336. 10.1136/jech.2008.080382.View ArticleGoogle Scholar
- Game FL, Jones AF: Ethnicity and risk factors for coronary heart disease in diabetes mellitus. Diabetes Obes Metab. 2000, 2: 91-97. 10.1046/j.1463-1326.2000.00063.x.View ArticlePubMedGoogle Scholar
- Yki-Järvinen H: Pathogenesis of non-insulin-dependent diabetes mellitus. Lancet. 1994, 343: 91-95. 10.1016/S0140-6736(94)90821-4.View ArticlePubMedGoogle Scholar
- Kohler IV, Soldo BJ: Childhood predictors of late-life diabetes: The case of Mexico. Biodemography Soc Biol. 2005, 52: 112-131. 10.1080/19485565.2005.9989105.View ArticleGoogle Scholar
- Case A, Fertig A, Paxson C: The lasting impact of childhood health and circumstance. J Health Econ. 2005, 24: 365-389. 10.1016/j.jhealeco.2004.09.008.View ArticlePubMedGoogle Scholar
- Eriksson JG, Forsen TJ, Osmond C, Barker DJP: Pathways of infant and childhood growth that lead to type 2 diabetes. Diabetes Care. 2003, 26: 3006-3010. 10.2337/diacare.26.11.3006.View ArticlePubMedGoogle Scholar
- Best LE, Hayward MD, Hidajat MM: Life course pathways to adult-onset diabetes. Biodemography Soc Biol. 2005, 52: 94-111. 10.1080/19485565.2005.9989104.View ArticleGoogle Scholar
- Helmrich SP, Ragland DR, Leung RW, Paffengarger RS: Physical activity and reduced occurence of non-insulin-dependent diabetes mellitus. New Engl J Med. 1991, 325: 147-152. 10.1056/NEJM199107183250302.View ArticlePubMedGoogle Scholar
- Perry I, Wannamethee SG, Walker MK, Thomson AG, Whincup PH, Shaper AG: Prospective study of risk factors for development of non-insulin dependent diabetes in middle aged British men. BMJ. 1995, 310: 560-10.1136/bmj.310.6979.560.View ArticlePubMedPubMed CentralGoogle Scholar
- Sairenchi T, Iso H, Nishimura A, Hosoda T, Irie F, Saito Y, Murakami A, Fukutomi H: Cigarette smoking and risk of type 2 diabetes mellitus among middle-aged and elderly Japanese men and women. Am J Epidemiol. 2004, 160: 158-162. 10.1093/aje/kwh183.View ArticlePubMedGoogle Scholar
- Demakakos P, Marmot M, Steptoe A: Socioeconomic position and the incidence of type 2 diabetes: the ELSA study. Eur J Epidemiol. 2012, 27: 367-378. 10.1007/s10654-012-9688-4.View ArticlePubMedGoogle Scholar
- Kowall B, Rathmann W, Strassburger K, Meisinger C, Holle R, Mielck A: Socioeconomic status is not associated with type 2 diabetes incidence in an elderly population in Germany: KORA S4/F4 Cohort Study. J Epidemiol Commun Health. 2011, 65: 606-612. 10.1136/jech.2009.094086.View ArticleGoogle Scholar
- Lawlor DA, Ebrahim S, Davey Smith G: Socioeconomic position in childhood and adulthood and insulin resistance: cross sectional survey using data from British women’s heart and health study. BMJ. 2002, 326: 488-Google Scholar
- Agardh EE, Sidorchuk A, Hallquist J, Ljung R, Peterson S, Moradi T, Allebeck P: Burden of type 2 diabetes attributed to lower educational levels in Sweden. Popul Health Metrics. 2011, 9: 60-10.1186/1478-7954-9-60.View ArticleGoogle Scholar
- Agardh E, Ahlbom A, Andersson T, Efendic S, Grill V, Hallquist J, Ostenson CG: Socio-economic position at three points in life association with type 2 diabetes and impaired glucose tolerance in middle-aged Swedish men and women. Int J Epidemiol. 2007, 36: 84-92. 10.1093/ije/dyl269.View ArticlePubMedGoogle Scholar
- Lynch JW, Kaplan GA, Shema SJ: Cumulative impact of sustained economic hardship on physical, cognitive, psychological, and social functioning. N Engl J Med. 1997, 337: 1889-1895. 10.1056/NEJM199712253372606.View ArticlePubMedGoogle Scholar
- Agardh E, Allebeck P, Hallquist J, Moradi T, Sidorchuk A: Type 2 diabetic incidence and socio-economic position: a systematic review and meta-analysis. Int J Epidemiol. 2012, 40: 804-818.View ArticleGoogle Scholar
- Rabi DM, Edwards AL, Southern DA, Svenson LW, Sargious PM, Norton P, Larsen ET, Ghali WA: Association of socio-economic status with diabetes prevalence and utilization of diabetes care services. BMC Health Serv Res. 2006, 6: 124-10.1186/1472-6963-6-124.View ArticlePubMedPubMed CentralGoogle Scholar
- Strand BH, Murray ET, Guralnik J, Hardy R, Kuh D: Childhood social class and adult adiposity and blood pressure trajectories 36–53 years: gender-specific results from a British birth cohort. J Epidemiol Commun Health. 2010, 66: 512-518.View ArticleGoogle Scholar
- Lehman BJ, Taylor SE, Kiefe CI, Seeman TE: Relation of childhood socioeconomic status and family environment to adult metabolic functioning in the CARDIA Study. Psychosom Med. 2005, 67: 849-854.View ArticleGoogle Scholar
- Rathmann W, Haastert B, Giani G, Koenig W, Imhof A, Herder C, Holle R, Mielck A: Is inflammation a causal chain between low socioeconomic status and type 2 diabetes? Results from the KORA survey 2000. Eur J Epidemiol. 2005, 21: 55-60.View ArticleGoogle Scholar
- Pyykkonen AJ, Raikkonen K, Tuomi T, Eriksson JG, Groop L, Isomaa B: Stressful life events and the metabolic syndrome: the prevalence, prediction and prevention of diabetes (PPP)- botnia study. Diabetes Care. 2010, 33: 378-384. 10.2337/dc09-1027.View ArticlePubMedGoogle Scholar
- Hu FB, Meigs JB, Li TY, RIfai N, Manson J: Inflammatory markers and risk of developing type 2 diabetes in women. Diabetes. 2004, 53: 693-700. 10.2337/diabetes.53.3.693.View ArticlePubMedGoogle Scholar
- Pickup JC: Inflammation and activated innate immunity in the pathogenesis of type 2 diabetes. Diabetes Care. 2004, 27: 813-823. 10.2337/diacare.27.3.813.View ArticlePubMedGoogle Scholar
- Mishra G, Nitsch D, Black S, de Stavola B, Kuh D, Hardy R: A structured approach to modelling the effects of binary exposure variables over the life course. Int J Epidemiol. 2009, 38: 528-537. 10.1093/ije/dyn229.View ArticlePubMedGoogle Scholar
- Marmot M, Shipley M, Brunner E, Hemingway H: Relative contribution of early life and adult socioeconomic factors to adult morbidity in the Whitehall II study. J Epidemiol Commun Health. 2001, 55: 301-307. 10.1136/jech.55.5.301.View ArticleGoogle Scholar
- Demakakos P, Hammer M, Stamatakis E, Steptoe A: Low-intensity physical activity is associated with reduced risk of incident type 2 diabetes in older adults: evidence from the English Longitudinal Study of Ageing. Diabetologia. 2010, 53: 1877-1885. 10.1007/s00125-010-1785-x.View ArticlePubMedGoogle Scholar
- Wei M, Gaskill SP, Haffner SM, Stern MP: Waist circumference as the best predictor of noninsulin dependent diabetes mellitus (NIDDM) compared to body mass index, waist/hip ratio and other anthropometric measurements in Mexican Americans–a 7-year prospective study. Obes Res. 1997, 5: 16-23. 10.1002/j.1550-8528.1997.tb00278.x.View ArticlePubMedGoogle Scholar
- McKeigue PM, Shah B, Marmot M: Relation of central obesity and insulin resistance with high diabetes prevalence and cardiovascular risk in South Asian. Lancet. 1991, 337: 382-386. 10.1016/0140-6736(91)91164-P.View ArticlePubMedGoogle Scholar
- Maty SC, Lynch JW, Raghunathan TE, Kaplan GA: Childhood socioeconomic position, gender, adult body mass index, and incidence of type 2 diabetes mellitus over 34 years in the Alameda County Study. Am J Public Health. 2008, 98: 1486-1494. 10.2105/AJPH.2007.123653.View ArticlePubMedPubMed CentralGoogle Scholar
- Lidfeldt J, Li TY, Manson JE, Kawachi I: A prospective study of childhood and adult socioeconomic status and incidence of type 2 diabetes in women. Am J Epidemiol. 2007, 165: 882-889. 10.1093/aje/kwk078.View ArticlePubMedGoogle Scholar
- Smith BT, Lynch JW, Fox CS, Harper S, Abrahamowicz M, Almeida ND, Loucks EB: Life-course socioeconomic position and type 2 diabetes mellitus. Am J Epidemiol. 2011, 173: 438-447. 10.1093/aje/kwq379.View ArticlePubMedPubMed CentralGoogle Scholar
- Robbins JM, Vaccarino V, Zhang H, Kasl SV: Socioeconomic status and type 2 diabetes in African American and non-Hispanic white women and men: Evidence from the Third National Health and Nutrition Examination Survey. Am J Public Health. 2001, 91: 76-83.View ArticlePubMedPubMed CentralGoogle Scholar
- Ross NA, Gilmour H, Dasgupta K: 14-year diabetic incidence: the role of socio-economic status. Health Reports. 2010, 21: 19-28.PubMedGoogle Scholar
- The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1471-2458/14/505/prepub
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.