Socio-economic position at three points in life in association with type 2 diabetes and impaired glucose tolerance in middle-aged Swedish men and women
1Department of Molecular Medicine and Surgery, Endocrine and Diabetes Unit, Karolinska Institutet, Stockholm, Sweden.
2Department of Epidemiology, Institute of Environmental Medicine, Karolinska Institutet, Stockholm, Sweden.
3Department of Epidemiology, Stockholm Center of Public Health, Stockholm, Sweden.
4Department of Public Health Science, Division of Social Medicine, Karolinska Institutet, Stockholm, Sweden.
* Corresponding author. Department of Molecular Medicine and Surgery, Endocrine and Diabetes Unit, Karolinska Institutet, SE 171 76 Stockholm, Sweden. E-mail: claes-goran.ostenson{at}karolinska.se
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Background It has been suggested that low socio-economic position (SEP) during childhood and adolescence predicts risk of adult type 2 diabetes. We investigated the associations between type 2 diabetes and childhood SEP (fathers’ occupational position), participants’ education and adult SEP (participants’ occupational position). To determine possible independent associations between early SEP (fathers’ occupational position and participants’ education) and disease, we adjusted for adult SEP and factors present in adult life associated with type 2 diabetes.
Methods This cross-sectional study comprised 3128 men and 4821 women aged 35–56 years. All subjects have gone through a health examination and answered a questionnaire on lifestyle factors. At the health centre, an oral glucose tolerance test was administered and identified 55 men and 52 women with previously undiagnosed type 2 diabetes. Relative risks (RRs) with 95% CIs were calculated in multiple logistic regression analyses.
Results The age-adjusted RRs of type 2 diabetes if having a father with middle occupational position were 2.3 [Confidence interval (CI:1.0–5.1) for women and, 2.0 (CI:0.7–5.6) for men]. Moreover, low education was associated with type 2 diabetes in women, RR = 2.5 (CI:1.2–4.9). Low occupational position in adulthood was associated with type 2 diabetes in women, RR = 2.7 (CI:1.3–5.9) and men, RR = 2.9 (CI:1.5–5.7). The associations between early SEP and type 2 diabetes disappeared after adjustment for adult SEP and factors associated with type 2 diabetes.
Conclusion The association between type 2 diabetes and low SEP during childhood and adolescence in middle-aged Swedish subjects disappeared after adjustment for adult SEP and adult risk factors of diabetes.
Keywords Diabetes Mellitus, type 2, epidemiology, socioeconomic factors, lifestyle, childhood
Accepted 6 November 2006
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Several studies have recognized the influence of low socio-economic position (SEP) throughout the life on morbidity and mortality.1 Some,2–5 but not all6,7 have found an independent association between lower SEP in childhood and increased risk of type 2 diabetes and cardiovascular disease in adult life. The underlying processes are not yet fully understood, although associations between lower SEP in childhood and adult risk factors such as obesity, smoking and leisure time physical inactivity are of importance.8,9 In type 2 diabetes research there are few studies examining the possible influence of childhood and adolescence SEP, although it has been identified as an important issue.10 One approach to determine associations between early socio-economic circumstances and morbidity in adulthood is to adjust for conditions in adulthood.5 We first aimed to investigate SEP at three points in life, i.e. childhood and adolescence (fathers’ occupational position), education, and adult occupational position, in association with type 2 diabetes and impaired glucose tolerance (IGT). Second, to determine possible independent associations between early SEP (fathers’ occupational position and participants’ education) and disease, we adjusted for adult SEP, family history of diabetes (FHD) and factors in adult life associated with type 2 diabetes risk, such as physical inactivity, overweight, smoking, low sense of coherence (SOC) and work stress.
|
Subjects and methods |
|---|
|
TopAbstractIntroductionSubjects and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Study design
The design of this cross-sectional study, from the baseline study of Stockholm Diabetes Prevention Programme has been described in detail previously.11,12 The study includes 7949 subjects (3128 men and 4821 women) aged 35–56 years at time of examination (1992–1994 and 1996–98, respectively), residing in five municipalities in Stockholm County Council (Figure 1). All study individuals were identified by the Stockholm County Council register.
|
The collection of data was performed in two steps. In the first step, a questionnaire was sent by mail to all identified men and women aged 35–56 years, living within the study area (n = 12 952 and 19 416, respectively). The questionnaire contained questions on country of birth, previous diabetes diagnosis and FHD. Completed questionnaires were returned from 79% (10 236/12 952) of the men and 85% (16 481/19 416) of the women. At this stage 4801 (47%) men and 8178 (50%) women were excluded from further study, due to already diagnosed diabetes (type 1 or type 2), incomplete answers, foreign origin, moving out of municipality, death and mental retardation and not fulfilling the criteria of FHD i.e. having at least one first (parents or siblings) or two second-degree relatives (grandparents, uncles or aunts) with diabetes.
In the second step, all men (2106) and women (3583) with FHD were invited to participate in a physical health examination, along with 2424 randomly selected men and 3497 women without FHD but who were matched to the first group according to age and municipality. Moreover, 424 women reporting previous gestational diabetes were invited.
Totally 3162 (70%) men and 4946 (70%) women agreed to participate in the health examination. The others were not available, had no wish to participate or had died. At the health care centre, additionally 33 (1%) of the men were excluded due to insufficient FHD and 129 (2%) of the women were excluded due to intake of medication, pregnancy or breast-feeding. Thus, the final study sample consisted of 3128 men and 4821 women.
During the health examination, an oral glucose tolerance test (OGTT) was administered in the final study sample and weight, height and blood pressure were measured. In addition, the participants responded to a detailed questionnaire on lifestyle factors including physical activity, smoking, psychosocial factors and SEP (occupation, education and their fathers’ occupation).
Classification of type 2 diabetes and impaired glucose tolerance
The 75 g OGTT was performed according to the World Health Organisation criteria.13 A person was diagnosed as having type 2 diabetes when the 2-h serum glucose level was 
11.1 mmol/l and IGT with 2-h serum glucose levels between 7.8 and 11.0 mmol/l. The OGTT test identified 55 (1.8%) men and 52 (1.1%) women with previously undiagnosed type 2 diabetes and 172 (5.5%) men and 167 (3.5%) women with IGT. Women and men with normal glucose tolerance formed the control groups.
Classification of socio-economic position
Childhood and adolescence SEP
SEP was classified according to a system elaborated by the Central Bureau of Statistics in Sweden and used in Swedish censuses.14 This classification system is based on occupational titles grouped according to educational demands and prestige. Childhood and adolescence SEP was based on the father's occupational title reported by the participants. The fathers’ occupational position was divided into three groups: those with high position, including high and medium level non-manual employees, those with middle position, including low level non-manual employees, and finally those with low position, including unskilled and skilled manual workers. In the male sample a total of 2642 (84%) were included in the analyses of father's occupational title (39 with diabetes and 143 with IGT). The others were not included due to the following reasons: 64 were missing occupational titles, (five participants with diabetes and five with IGT), 229 were self-employed (7 with diabetes and 12 with IGT) and 194 were farmers (4 with diabetes and 12 with IGT).
In the female sample a total of 3979 (83%) were included in the analyses of father's occupational title (42 with diabetes and 140 with IGT). The others were not included due to the following reasons: 27 were missing occupational titles, (one with diabetes and one with IGT), 449 were self-employed (5 with diabetes and 17 with IGT), 275 were farmers (two with diabetes and seven with IGT) and 91 were unclassified workers (two with diabetes and two with IGT).
Education
Educational level was based on the question, ‘What education do you have?’, and the responders were asked to state their highest level of education. The response alternatives were; ‘elementary school or nine-year compulsory school’, ‘junior secondary school’, ‘2-year secondary high school’, ‘3–4 year secondary high school’, ‘university or college of higher learning’ and ‘other education’. We formed three groups, high (university), middle (‘3–4 year secondary high school’ and ‘2 year secondary high school’) and low (‘elementary school or nine-year compulsory school’ and ‘junior secondary school’). In the male sample a total of 3053 (98%) men were included (52 with type 2 diabetes and 169 with IGT) and in the female sample 4680 (97%) were included, 49 with diabetes and 160 with IGT. Those who stated ‘other education’ (n = 60, three with diabetes and two with IGT in the male sample and n = 136, three with diabetes and seven with IGT in the female sample) were excluded from further analysis.
Adult SEP
Adult SEP was measured by occupational position, based on self-reported occupational titles and classified the same way as the fathers’ position. A total of 2864 men (92%) (49 with type 2 diabetes and 154 with IGT) and 4509 women (94%) (48 with type 2 diabetes and 152 with IGT) were included in the analysis. Reasons for exclusion have been described in detail previously.15
Classification of factors in adult life associated with type 2 diabetes risk
Physical activity, measured from the question on physical activity during leisure hours, was based on four response alternatives and categorized into three groups; low (sedentary), middle (moderately active) and high (regular exercise and training). Body mass index (BMI, kg/m2) was used to assess overweight and divided into three groups (<25, 25–29.9 and 30). Smoking was divided into non-smokers (non-smokers included former smokers), 1–15 and 16 cigarettes per day. FHD was categorized as positive, including at least one first (parent or sibling) or two second-degree relatives (grandparents, uncles or aunts) or negative i.e. no close relative with diabetes.
We have previously found that self-reported psychosocial factors such as low decision latitude at work and low SOC was associated with type 2 diabetes in women.12 However, no obvious associations were observed in men.15 Decision latitude at work derives from the demand-decision latitude model introduced by Karasek and Theorell16,17 and describes person's skills and ability to master their work activities.18 We analysed five out of six questions from the original questionnaire. Each question contained four response options; ‘almost always’, ‘often’, ‘seldom’ and ‘never’. From these responses each question was given a point ranging from 1–4, and then we calculated a summed score from all questions. From the distribution of scores among all respondents three groups were created from the upper, median and lower quartile, representing high, middle and low decision latitude.
SOC, a paradigm developed by Antonovsky,19 was based on three questions on comprehensibility, meaningfulness and manageability. We used a simplified way of measuring SOC that has been recommended previously.20 The original questionnaire consists of 29 items that are based on three dimensions (comprehensibility, meaningfulness and manageability). Each question consisted of three response alternatives, ‘yes, most often’, ‘yes, sometimes’ and ‘no’. From the distribution of scores among all respondents four groups of SOC were created, i.e. high, upper middle, lower middle and low.
Statistical analysis
Prevalence odds ratios together with 95% confidence intervals (CI), were calculated with logistic regression analysis using SAS statistical package version 8.2 for Windows.21 Due to the relatively low prevalence of cases in this study, odds ratios were interpreted as relative risks (RR).22 Analyses were performed on men and women separately. First, we estimated age-adjusted RR for childhood and adolescence SEP (fathers’ occupational position), participants’ education and adult SEP in association with type 2 diabetes and IGT. Then, to investigate possible independent associations between disease and early SEP (fathers’ occupational position and participants’ education), we adjusted for adult SEP, established risk factors and psychosocial factors in adult life, as recommended previously.5 We regarded these factors as having an influence on the association if they changed the effect estimates more than 10–15%.23
|
Results |
|---|
|
TopAbstractIntroductionSubjects and methodsResultsDiscussionAcknowledgementsReferences |
|---|
SEP at three points in life
In both men and women, increased RRs of type 2 diabetes were found if their fathers had middle compared with high occupational position (RR = 2.0; CI 0.7–5.6) and (RR = 2.3; CI 1.0–5.1), respectively (Table 1). No apparent associations with type 2 diabetes were observed in men if their fathers had low occupational position, although the point estimates were somehow increased. In women, type 2 diabetes and IGT were not associated with father's low occupational position.
|
There was an association between low education and IGT in men, (RR = 1.5; CI 1.0–2.4), as well as between low education and IGT and type 2 diabetes in women (RR = 1.4; CI 1.0–2.1) and (RR = 2.5; CI 1.2–4.9), respectively (Table 1). However, lower (middle and low) education was not obviously associated with type 2 diabetes in men.
Adult SEP in association with IGT and type 2 diabetes have previously been reported in these men and women,15 showing an increased risk in all groups except for IGT in men with lower occupational position (Table 1).
Early SEP after adjustment for adult risk factors
To determine possible independent associations between early SEP (fathers’ occupational position and participants’ education) we adjusted for adult factors (adult SEP, established and psychosocial) associated with type 2 diabetes risk. In men, the associations between type 2 diabetes and fathers’ lower (middle and low) occupational position seemed to be influenced by adult SEP (Table 2). Although low educational level was not apparently associated with type 2 diabetes, the association was influenced by adult SEP, physical inactivity, BMI, smoking and FHD. In women, the association between father's middle occupational position and type 2 diabetes decreased after adjustment for BMI and physical inactivity (Table 2). In addition, the association between low education and type 2 diabetes in women, was diluted after adjustment by adult factors such as; adult SEP, BMI, smoking, low decision latitude at work and low SOC.
|
In men, the associations between IGT and low educational level decreased after adjustment for BMI (Table 3). In women, the association between low educational level and IGT disappeared after adjustment for adult SEP and BMI in women (Table 3).
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and methodsResultsDiscussionAcknowledgementsReferences |
|---|
When we investigated SEP at three points (fathers’ occupational position, participants’ education and adult occupational position) in the life course, we first found that fathers’ middle occupational position was associated with type 2 diabetes in women. Second, low educational level was associated with type 2 diabetes and IGT in women as well as with IGT in men. Third, as published previously, middle and low adult occupational position was associated with type 2 diabetes in both men and women, and with IGT in women.15 When we adjusted for adult SEP and adult factors associated with type 2 diabetes risk, the associations between early SEP (fathers’ occupational position and participants’ education) and disease disappeared.
Few studies have assessed the association between childhood SEP and type 2 diabetes, although one study found that parents’ education was associated with type 2 diabetes in adult men.2 In addition, two studies demonstrated that childhood SEP, measured by fathers’ occupational position, was associated with insulin resistance in women.24,25 Type 2 diabetes and cardiovascular disease (CVD) share some causes26 hence it is also of interest to discuss epidemiological findings in type 2 diabetes in relation to CVD. Numerous studies have examined the effects of life course SEP in relation to cardiovascular and coronary heart disease (CHD).1 For example, one study performed in English men showed that fathers’ lower occupational position was associated with CVD mortality, and this association persisted after adjustment for systolic blood pressure and smoking.3 Another study in men from the west of Scotland found that childhood SEP was independently related to CHD after adjustment for adult social position and adult risk factors associated with CVD risk.5 In a Swedish cohort, childhood SEP was independently related to ischaemic heart disease, although adult SEP decreased the association to some extent.4 Additionally, two studies performed in women found greater risks of CVD mortality with poorer childhood SEP.27,28 In the first study, the associations persisted after adjustment of adult SEP and other risk factors, while in the second study adjustment was made only for adult SEP.
In our study, fathers’ middle occupational position in association with type 2 diabetes in women and to some extent in men, as well as low educational level in association with IGT in men and type 2 diabetes and IGT in women, disappeared after adjustment for adult SEP and adult factors associated with type 2 diabetes risk. There are a few previous studies on CVD showing that childhood SEP is not independently associated with the disease. For example, in one Finnish study, the association of fathers’ lower occupational SEP with CVD mortality was explained by adult SEP.29 In another Finnish study no effect of childhood SEP was found in relation to CVD after adjustment for adult SEP.6 Moreover, one study in women found that childhood SEP was not apparently associated with CVD, either before or after adjustment for risk factors.30 In addition, in the Nurses Health Study there was no effect of fathers’ occupational position on the risk of CHD in women.31
We found that, fathers’ middle occupational position and lower educational levels were more obviously associated with type 2 diabetes in women than in men. It has been suggested that the social position is more persistently influenced by the family background in women than in men.32 Hence, it is possible that fathers’ lower SEP is a stronger predictor for adult type 2 diabetes in women than in men. On the other hand, associations of early SEP with type 2 diabetes disappeared after adjustment for adult SEP and other risk factors. It remains unclear, however, why only fathers’ middle and not low occupational position was associated with type 2 diabetes. It may be hypothesized that cases excluded prior to this study had more severe diabetes due to poorer socio-economic circumstances, and thus the influence of fathers’ low SEP was underestimated.
SEP was investigated at three stages in the life course, where education has been suggested to belong to early rather than adult life.33 Although we could identify some effects from early SEP before adjustment for risk factors, adult SEP influenced these associations substantially, thus suggested to be an important confounding factor. Hence, we cannot exclude the possibility that adult SEP is a summary measure reflecting SEP throughout the life course. Moreover, it may also be possible that adult SEP and adult risk factors are in the pathway between childhood SEP and the outcome, thus not being confounders. Adult SEP has been found to be more important for CHD risk in men7 and women34 than early SEP. Moreover, the two Finnish studies on CVD,6,29 suggested adult SEP to be an important confounding factor when analysing fathers’ occupational position. In contrast, one study investigating SEP at three points in the life course, found that mortality risk was similar at each stage of life.35
When education was analysed in association with type 2 diabetes in men, adult risk factors as well as adult SEP seemed to be important confounding factors. In addition, as published previously,14 established risk factors made contributions to some extent in explaining inequalities in adult SEP in men. In women, overweight seemed to be an important confounding factor for the association between fathers’ middle occupational position and type 2 diabetes. Also in women, however, adult risk factors seemed to be important confounding factors as we analysed education and adult SEP.15 This may reflect that risk factors implicated in the development of type 2 diabetes accumulate as a result of socio-economic disadvantage over the life course.
We need to account for some methodological issues when interpreting the results of the present study. First, the participants answered questions about their former lifestyles, which may introduce recall bias, especially since the respondents were asked to recall family and childhood conditions. However, all participants responded to the questionnaire prior to their diagnosis. Hence, possible misclassification should not overestimate the RRs. In addition, recall of childhood SEP has been shown previously to be reliable in a middle-aged population in the US.36 Second, it is possible that occupational titles classified according to the Swedish classification system may result in too broad categories to be homogenous as for a single socio-economic group, and therefore some subjects may not be coded in the correct group. However, this misclassification will be the same for cases and controls and should not affect the results. Third, our cross-sectional study design entails that the direction of causation cannot be known with certainty, hence, having lower SEP may be a result of type 2 diabetes. However, this seems rather unlikely since we measured childhood and adolescence SEP which are factors that most likely preceded the participants’ disease. Moreover, those who were diagnosed with diabetes prior to the study were not included and, as mentioned previously, the participants did not know about their disease status when answering the questions. Fourth, it may also be questioned, as discussed previously, if those who were diagnosed with diabetes before this study had lower SEP throughout the life course. However, we do not have any information on excluded cases. Nevertheless, to investigate if participants differed from non-participants with regard to SEP we collected information through record linkage to the 1990 Swedish census for the total study population. Both men and women participating in the study appeared to have slightly higher SEP than non-participants. We can only speculate how this may influence our results. However, since type 2 diabetes is more prevalent in lower socio-economic groups, and if the excluded cases had the same or lower SEP as the non-participating study-population, we may have underestimated the associations in the study.
In conclusion, in middle-aged Swedish men and women, the associations between lower early SEP (fathers’ occupational position and participants’ education), type 2 diabetes and IGT disappeared after adjustment for adult SEP and adult risk factors associated with type 2 diabetes risk. It seemed as if risk factors associated with type 2 diabetes among these men and women became increasingly important when analysing education and adult SEP as compared with childhood SEP. It is possible that factors associated with type 2 diabetes risk accumulate as a result of socio-economic disadvantage over the life course.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionSubjects and methodsResultsDiscussionAcknowledgementsReferences |
|---|
This study was supported by grants from the Stockholm County Council, Swedish Council for Working Life and Social Research, Vårdalstiftelsen and Glaxo Smithkline, Sweden.
Conflict of interest: None declared.
Key Messages
|
|
References |
|---|
|
TopAbstractIntroductionSubjects and methodsResultsDiscussionAcknowledgementsReferences |
|---|
1 Galobardes B, Lynch JW, Davey Smith G. Childhood socioeconomic circumstances and cause-specific mortality in adulthood: systematic review and interpretation. Epidemiol Rev (2004) 26:7–21.
2 Leonetti DL, Fujimoto WY, Wahl PW. Early-life background and the development of non-insulin-dependent diabetes mellitus. Am J Phys Anthropol (1989) 3:345–55.
3 Davey Smith G, McCarron P, Okasha M, McEwen J. Social circumstances in childhood and cardiovascular disease mortality: prospective observational study of Glasgow University students. J Epidemiol Community Health (2001) 55:340–41.
4 Vagero D, Leon D. Effect of social class in childhood and adulthood on adult mortality. Lancet (1994) 343:1224–25.[Web of Science][Medline]
5 Davey Smith G, Hart C, Blane D, Hole D. Adverse socioeconomic conditions in childhood and cause specific adult mortality: prospective observational study. Br Med J (1998) 316:1631–35.
6 Lynch JW, Kaplan GA, Cohen RD, et al. Childhood and adult socioeconomic status as predictors of mortality in Finland. Lancet (1994) 343:524–27.[CrossRef][Web of Science][Medline]
7 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 Community Health (2001) 55:310–7.
8 Wannamethee SG, Whincup PH, Shaper G, Walker M. Influence of fathers’ social class on cardiovascular disease in middle-aged men. Lancet (1996) 348:1259–63.[CrossRef][Web of Science][Medline]
9 Brunner E, Shipley MJ, Blane D, Davey Smith G, Marmot MG. When does cardiovascular risk start? Past and present socioeconomic circumstances and risk factors in adulthood. J Epidemiol Community Health (1999) 53:757–64.[Abstract]
10 Kuh D, Ben-Shlomo Y. A Life Course Approach to Chronic Disease Epidemiology (2004) Chapter 7, 2nd edn. Oxford, UK: Oxford University Press.
11 Carlsson S, Persson PG, Alvarsson M, et al. Weight history, glucose intolerance, and insulin levels in middle-aged Swedish men. Am J Epidemiol (1998) 148:539–45.
12 Agardh EE, Ahlbom A, Andersson T, et al. Work stress and low sense of coherence is associated with type 2 diabetes in middle-aged Swedish women. Diabetes Care (2003) 26:719–23.
13 WHO Expert Committee on Diabetes Mellitus. Diabetes Mellitus: Report of a WHO Study Group (1985) Geneva, Switzerland: World Health Organization. WHO technical series no. 727.
14 Statistics Sweden. Swedish Socioeconomic Classification (1989) Stockholm: Statistics Sweden.
15 Agardh EE, Ahlbom A, Andersson T, et al. Explanations of socioeconomic differences in excess risk of type 2 diabetes in Swedish men and women. Diabetes Care (2004) 27:716–21.
16 Karasek RA. Job demands, job decision latitude, and mental strain: implications for job redesign. Admin Sci Q (1979) 24:285–307.[CrossRef]
17 Karasek R, Baker D, Marxer F, Ahlbom A, Theorell T. Job decision latitude, job demands and cardiovascular disease: a prospective study of Swedish men. Am J Public Health (1981) 71:694–705.
18 Karasek R, Theorell T. Healthy Work: Stress, Productivity and Reconstruction of Working Life (1990) New York: Basic Books.
19 Antonovsky A. Unrevealing the Mystery of Health: How People Manage Stress and Stay Well (1987) San Francisco, CA: Jossey-Bass.
20 Lundberg O, Nyström PM. A simplified way of measuring sense of coherence: experiences from a population survey in Sweden. Eur J Public Health (1995) 5:56–9.
21 SAS Institute Inc. The logistic procedure. In: SAS/STAT Software: User's Guide, release 8.2 (2001) Cary, NC: SAS Institute, Inc.
22 Rothman KJ, Greenland S. Modern Epidemiology (1998) 2nd edn. Philadelphia: Lippincott Williams & Wilkins.
23 Greenland S. Modelling and variable selection in epidemiological analysis. Am J Public Health (1989) 79:340–49.
24 Lawlor DA, Davey Smith G, Ebrahim S. Life course influences on insulin resistance. Findings from the British Women's Heart and Health Study. Diabetes Care (2003) 26:97–103.
25 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. Br Med J (2002) 325:805–9.
26 Perry IJ, 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. Br Med J (1995) 310:560–64.
27 Beebe-Dimmer J, Lynch JW, Turrell G, Lustgarten S, Raghunathan T, Kaplan GA. Childhood and adult socioeconomic conditions and 31-year mortality risk in women. Am J Epidemiol (2004) 159:481–90.
28 Claussen B, Davey Smith G, Thelle D. Impact of childhood and adulthood socioeconomic position on cause specific mortality: the Oslo Mortality Study. J Epidemiol Community Health (2003) 57:40–45.
29 Pensola TH, Matikainen P. Cumulative social class and mortality from various causes of adult men. J Epidemiol Community Health (2001) 57:745–51.[CrossRef]
30 Heslop P, Davey Smith G, Macleod J, Hart C. The socioeconomic position of employed women, risk factors and mortality. Soc Sci Med (2001) 53:477–85.[CrossRef][Web of Science][Medline]
31 Gliksman MD, Kawachi I, Hunter D, et al. Childhood socioeconomic status and risk of cardiovascular disease in middle aged US women: a prospective study. J Epidemiol Community Health (1995) 49:10–15.
32 Roberts R, Brunner EJ, White I, Marmot M. Gender differences in occupational mobility and structure of employment in the British civil service. Soc Sci Med (1993) 37:1415–25.[CrossRef][Web of Science][Medline]
33 Davey Smith G, Hart C, Hole D, et al. Education and occupational social class: which is the more important indicator of mortality risk? J Epidemiol Community Health (1998) 52:153–60.[Abstract]
34 Wamala SP, Lynch J, Kaplan GA. Women's exposure to early and later life socioeconomic disadvantage and coronary disease risk: The Stockholm Female Coronary Risk Study. Int J Epidemiol (2001) 30:275–84.
35 Hart CL, Davey Smith G, Blane D. Inequalities in mortality by social class measured at 3 stages of the lifecourse. Am J Public Health (1998) 88:471–74.
36 Krieger N, Okamoto A, Selby JV. Twins recall of childhood social class and father's education: a validation study for public health research. Am J Epidemiol (1998) 147:704–8.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  E. J. Mayer-Davis, J. Beyer, R. A. Bell, D. Dabelea, R. D'Agostino Jr., G. Imperatore, J. M. Lawrence, A. D. Liese, L. Liu, S. Marcovina, et al. Diabetes in African American Youth: Prevalence, incidence, and clinical characteristics: the SEARCH for Diabetes in Youth Study Diabetes Care, March 1, 2009; 32(Supplement_2): S112 - S122. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||