IJE Advance Access originally published online on January 25, 2007
International Journal of Epidemiology 2007 36(3):522-531; doi:10.1093/ije/dyl309
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Lifecourse influences on health among British adults: Effects of region of residence in childhood and adulthood
1 Division of Community Health Sciences St George's University of London, Cranmer Terrace, London SW17 0RE, UK.
2 Centre for Paediatric Epidemiology and Biostatistics, Institute of Child Health, UCL, 30 Guilford Street, London WC1N 1EH, UK.
3 Centre for Longitudinal Studies, Bedford Group for Lifecourse and Statistical Studies, Institute of Education, 20 Bedford Way, London, WC1A 0AL, UK.
4 National Centre for Social Research, 35 Northampton Square, London, EC1V 0AX, UK.
5 MRC Hearing and Communication Group, University of Manchester, Oxford Road, Manchester, M13 9PL, UK.
6 Department of Clinical Biochemistry, Royal Victoria Infirmary, Newcastle, NE1 4LP, UK.
7 Department of Epidemiology and Public Health, University College London, 1–19 Torrington Place, London, WC1E 6BT, UK.
8 Division of Cardiovascular and Medical Sciences, University of Glasgow, Royal Infirmary, 10 Alexandra Parade, Glasgow G31 2ER, UK.
9 Aberdeen Pain Research Collaboration, Epidemiology Group, University of Aberdeen, Foresterhill, Aberdeen, AB25 2ZD, UK.
10 National Centre for Epidemiology and Population Health, Mills Road, The Australian National University, Canberra, ACT 0200, Australia.
11 Centre for Psychiatry, Queen Mary's School of Medicine and Dentistry, Charterhouse Square, London, EC1M 6BQ, UK.
* Corresponding author: Division of Community Health Sciences, St George's, University of London, Cranmer Terrace, London, SW17 0RE, UK. E-mail: d.strachan{at}sgul.ac.uk
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionConclusionSupplementry materialSources of fundingAcknowledgementsReferences |
|---|
Background It has been suggested that early life exposures are important determinants of geographical variations in adult diseases. We examined inter-regional migrants in Britain to evaluate the relative importance of early and recent exposures for adult cardiorespiratory risk factors, mental ill-health and sensory function.
Methods A total of 9023 persons born throughout England, Scotland and Wales during 1 week in 1958 were followed periodically through childhood into adulthood. At 44–45 years, height, body mass index (BMI), blood pressure (BP), glycosylated haemoglobin, total and high-density lipoprotein (HDL) cholesterol, triglycerides, fibrinogen, total immunoglobulin E (IgE), one-second forced expiratory volume (FEV1), hearing threshold at 4 kHz, visual impairment, symptoms of depression and anxiety, and chronic widespread pain were measured. Analysis of migration between 12 regions included 3125 cohort members who were examined in a region different to their birthplace.
Results Height, BMI, diastolic BP (DBP), FEV1, log-transformed IgE and hearing threshold varied by region among non-migrants (each P < 0.05). Among inter-regional migrants, the spatial associations with current region, independent of birthplace, followed closely the geographical pattern shown among non-migrants for BMI, DBP and FEV1 (each P < 0.001). In contrast, of the 15 outcomes, only adult height was related to region of birth, after adjustment for region of examination (P = 0.002)
Conclusions Although individual disease risk is predicted by early life factors, early exposures do not explain regional variations in cardiovascular and respiratory risk factors among middle-aged adults in Britain. Geographical inequalities in cardiorespiratory health are more strongly related to factors associated with region of examination that influence obesity, BP and ventilatory function.
Keywords Birth cohort, health inequalities, geographical variation, migration
Accepted 2 November 2006
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionSupplementry materialSources of fundingAcknowledgementsReferences |
|---|
In an influential series of papers in the mid-1980s, Barker and Osmond noted that geographical variations in mortality from ischaemic heart disease1, stroke2 and chronic bronchitis3 within England and Wales during 1968–78 were correlated with the spatial patterns of infant and maternal mortality five decades earlier. They hypothesised that these ecological associations arose from ‘programming’ of cardio-respiratory risk in pre-natal or early post-natal life, stimulating two decades of research into lifecourse epidemiology.4 An alternative explanation is that these geographical correlations arise from ecological confounding, due to historical continuities in the variations of social and environmental conditions across Britain.5
Studies of migrants can help to distinguish between these alternative explanations. If geographical variations in mortality risk are determined to a substantial degree early in life (or by genetic factors), then the patterns of death, cardiorespiratory disease and associated risk factors should be related more closely to place of birth than to place of current residence, whereas birthplace would be less influential if recent social circumstances, environmental exposures or medical care were the main contributors to geographical variations in disease risk and mortality.
There have been no migrant studies within the UK for cardiovascular risk factors other than blood pressure (BP),6 or for other common diseases and risk indicators. In this article, we report migration analyses of cardiorespiratory risk factors, sensory function and mental health that were ascertained during biomedical examination of a national British birth cohort in early middle age. In an extension of the statistical methods applied earlier to case-control data,7 we use the measurements on non-migrants to score each geographical region, for use in a quantitative assessment of the independent effects of birthplace and current area of residence among inter-regional migrants. This complements the companion paper that analyses socio-economic inequalities in disease risk by social position in childhood and adulthood.8
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionSupplementry materialSources of fundingAcknowledgementsReferences |
|---|
Subjects and measurements
The British 1958 birth cohort (also known as the National Child Development Study) is a longitudinal study of persons born in England, Scotland and Wales during 1 week in March 1958 who were originally recruited for the Perinatal Mortality Survey.9 The cohort was followed up at ages 7, 11 and 16 years by parental interviews and examinations conducted by school medical officers, and at ages 16, 23, 33 and 42 years by means of interviews with participants. Immigrants of the same dates of birth were identified at ages 7, 11 and 16, and followed into adulthood, but adult immigrants (after age 16) have not been included.
All cohort members who were still in contact with the cohort study team, and at age 42, had not required a proxy interview (e.g. due to learning disability), were invited to participate in a clinical examination in their home at the age of 44–45 years. The home visits were carried out from September 2002 to March 2004 by a team of 122 trained nurses from the National Centre for Social Research, who conduct the annual Health Surveys of England and Scotland. Visits occurred in every region in each month of fieldwork. From a target sample of 12 069 persons, 9377 (78%) cohort members were visited (Figure 1), of whom 9023 were born in England, Scotland and Wales.
|
Standing height was measured to the nearest millimetre using a Leicester portable stadiometer placed on a hard floor. Weight was measured to the nearest 0.1 kg in light clothing with shoes removed. Where it was not possible to obtain an accurate measurement, or consent was not provided, self-reported weight (n = 100) or height (n = 84) was recorded. Body mass index (BMI) was calculated as kg/m2 excluding two participants who were pregnant at the time of interview.
BP and pulse rate were measured three times in the seated position after a period of 5 min rest, using the Omron 705CP automated sphygmomanometer (Omron, Tokyo, Japan), with a large cuff for subjects with a mid-upper arm circumference of 
32 cm. The mean BP was determined for readings that the nurse considered to be reliable.
Spirometry was performed in the standing position without noseclips, using Vitalograph Micro spirometers, checked daily for calibration drift. At least three blows were recorded, and up to five were performed if the best-test variation [assessed by the sum of one-second forced expiratory volume (FEV1) and forced vital capacity] was > 5%. In the analysis, all readings with a best-test variation > 10% were excluded. FEV1 measurements were adjusted for height and gender, using separate linear regression models for males and females. Values with standardized residuals greater than + 3 SD units were excluded from subsequent analyses.
Venous blood samples were obtained without prior fasting, and posted to a central laboratory. Glycosylated haemoglobin (HbA1c) was assayed by ion exchange high performance liquid chromatography on whole blood. Triglycerides, total and high-density lipoprotein (HDL) cholesterol were measured by autoanlyser. Fibrinogen was assayed by the Clauss method on citrated plasma samples (MDA 180 coagulometer, Biomerieux, Basingstoke, UK). Total circulating immunoglobulin E (IgE) was measured using the HYTEC enzyme immunoassay; values > 2000 were recoded to 2000 and zero values to 0.5.
Pure tone audiometry was performed in the home by air conduction in both ears, with test tones at 1 kHz and 4 kHz according to the British Society of Audiology's recommended procedure.10 MA25 portable audiometers with TDH 49 earphones in audiocups were used, calibrated to British Standard BS EN ISO 389–1 (2000). We used hearing level at 4 kHz in the better ear, if hearing at 1 kHz was better than 30 dB. Distance visual acuity in each eye was measured using the 3 m Keeler crowded logarithm of the minimum angle of resolution (LogMAR) test booklet with distance correction (glasses or contact lenses) if prescribed for distance vision, and pinhole. Vision impairment was defined as logMAR > 0.2 (
6/12 or worse) in the better eye.
Symptoms of depression and anxiety in the previous week were measured by the revised Clinical Interview Schedule11 administered by the nurse. Participants who reported two or more items for either the depressive ideas or anxiety symptom modules were classified as high scorers for the respective symptom.
Chronic widespread pain, ascertained by self-completed questionnaire including a manikin (shaded by participants), was defined, as proposed by the American College of Rheumatology in their classification of fibromyalgia,12 as pain occurring for longer than 3 months: both above and below the waist; on both the left and right sides of the body; and in the axial skeleton (upper spine, low back or sternum).
Socio-economic position in childhood is based on father's occupation recorded in 1958, (or at the age 7 years if data were unavailable at birth; n = 422). Adult socio-economic position is based on the participant's current or most recent occupation at age 42 (or 33 if data were unavailable at 42; n = 1558). Occupation was categorized using the Registrar General's Classification into six classes: professional (I), managerial/technical (II), other non-manual (IIInm), skilled manual (IIIm), partly skilled (IV) and unskilled manual (V); those with no male head of household in childhood were grouped with class V.
Statistical analysis
Regression analyses were performed using STATA version 9.1 (STATA Corporation, Texas, USA). Continuous outcome measures were log-transformed if necessary to normalize their distribution (HDL cholesterol, triglycerides, fibrinogen, total IgE and hearing threshold).
For all measurements in the field, probability plots of the means by nurse and by instrument were inspected for outliers. This identified outlying hearing data from one nurse which was excluded from subsequent analyses.
Place of birth and place of residence at examination were both coded to the regional boundaries prevailing in 1958 with Greater London (as defined in 1965) further distinguished. Twelve regions were thus defined (Table 1). Migrants were identified as those whose birth region differed from their region at examination, no account being taken of intervening moves. The regional distribution of each outcome was assessed for the non-migrants and migrants separately.
|
In contrast to variables such as socio-economic position, where there is a conventional ranking imposed on the classification, the ranking of regions is not defined a priori. We, therefore, used the non-migrant data to generate quantitative scores for each region. For each outcome, this score variable was calculated as the mean value among non-migrants born in that region (for quantitative outcomes) or the logodds among non-migrants born in that region (for dichotomous outcomes). The regional score was then introduced as an explanatory variable in models based only on inter-regional migrants, assigned, in turn, to the region of birth and to the region of examination. This is an efficient method of evaluating early (place of birth) and later (place of examination) effects, because it allows more epidemiologically focused and statistically powerful (1df) trend tests to be applied to the migrant data, rather than drawing conclusions based solely on (11df) tests of heterogeneity across regions of birth or examination.
In modelling the data for migrants, we first fitted a linear regression model including birth region and current region simultaneously, both as categorical variables. F-tests for heterogeneity for region in 1958 and 2002–04 (each on 11df) were used as an overall estimate of the independent effects of region of birth and adulthood on each continuous outcome. A similar approach was adopted for binary outcome measures using logistic regression, where the likelihood ratio test (on 11df) was used to test for the independent effects of region at birth and in adulthood.
Two models were then fitted for each outcome in migrants only. The first model included current region (as 12 categories) and the score variable assigned to region of birth. In this model, the parameter estimate and associated (1df) significance test for the score variable evaluates the extent to which geographical variations in each outcome among migrants are determined by their region of birth, independent of region of examination. If region of birth totally explained the geographical variation, then the expected value of the regression coefficient for the score variable would be 1.0. For illustration, persons born in regions where the population is generally taller would be expected to be taller as adults regardless of their region of residence in adulthood. As shown in Table 2, adult height is associated with region of birth independent of region of examination (P = 0.002, 1df), but the regression coefficient for the score variable differs both from zero and from unity (0.55, 95%CI 0.20 – 0.91), indicating some influence of later place of residence in addition to the birthplace effect, plausibly due to geographical mobility during childhood.
|
The second model included region of birth (as 12 categories) and the score variable assigned to region of examination. In this model, the score variable evaluates the extent to which geographical variations in each outcome among migrants are determined by their region of examination, independent of region of birth. If region of examination totally explained the geographical variation, then the expected value of the regression coefficient for the score variable would be 1.0. As shown in Table 2, adult BMI is associated with region of examination independent of region of birth (P = 0.001, 1df), and the regression coefficient for the score variable (0.93, 95%CI 0.36 –1.50) is very close to unity. This implies a dominant effect of current place of residence, independent of region of residence early in life. This is represented visually by the superimposed regression line in the lower left-hand portion of Figure 1, relating BMI in migrants to the mean BMI of non-migrants in their region of birth.
All regression models in migrants included adjustment for gender. Regression models were further elaborated by adjusting for socio-economic position in childhood and adulthood as categorical variables. Where the predominant influence was region of examination, possible artifacts arising from observer, instrument or laboratory batch were evaluated by random effects modelling.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionSupplementry materialSources of fundingAcknowledgementsReferences |
|---|
Table 1 shows the distribution of all 18 558 original cohort members and the 9377 participants in the biomedical survey, by region of birth. The proportion of the full cohort who were examined in 2002–04 varied by birthplace from 46.5% in Scotland to 56.7% in Eastern England, but the resulting regional distribution for participants was similar to that for the full cohort (Table 1).
From our analyses, 354 participants born in Northern Ireland or outside the UK were excluded, leaving 9023 (4487 men, 4536 women) participants born in England, Scotland or Wales, of whom 3125 (1534 men, 1591 women) were examined in a region different to their region of birth. About one-quarter of these migrants moved from Greater London, mainly to the surrounding Eastern, Southeastern and Southern regions of England (Table 1).
Among the 5898 non-migrants, there was unadjusted regional variation in height, BMI, diastolic BP (DBP), FEV1, total IgE and hearing thresholds but not in levels of total or HDL cholesterol, glycosylated haemoglobin, triglycerides or fibrinogen. Visual acuity, anxiety, depression and chronic widespread pain did not show heterogeneity by region (Table 2).
Among the 3125 inter-regional migrants, 11df tests for heterogeneity showed independent effects of current region for BMI, DBP, FEV1, triglycerides and anxiety after adjustment for birth region and gender; and birth region for adult height after adjustment for current region and gender (Table 2).
When the average values for non-migrants were used to score each region of examination, there were positive trends (P < 0.001, 1df) for BMI, DBP and FEV1, suggesting that the association of these cardiorespiratory risk factors with current region (independent of birth region and gender) follows the geographical pattern found in non-migrants (Figure 2). These trends persisted after adjustment for social position in childhood and adulthood (P < 0.002, 1df). After further adjustment for potential confounders, including observer and instrument effects and factors given in Table 3, the associations for place of current residence remained for DBP (P < 0.0001), BMI (P = 0.002) and FEV1 (P = 0.02). The current region variation in triglycerides was not so clearly related to the average value in non-migrants (P = 0.08 after adjustment). The corresponding graphical presentations for each health outcome appear in the web supplement as Figure 3W and supporting data as Table 4W.
|
|
For anxiety symptoms, there was evidence of heterogeneity among the migrants in relation to current region (P = 0.02, 11df), and a trend (P = 0.02, 1df) in relation to current region, scored by the a verage values for non-migrants. In contrast, height showed a trend for birth region (independent of current region and gender, P = 0.002) that persisted after additional adjustment for childhood and adulthood social class (P = 0.003). For two outcomes, a weak correlation was observed between birth region effect and non-migrant score: this relationship was positive for triglycerides (P = 0.06) but negative for log-transformed IgE (P = 0.06). Given the lack of overall heterogeneity for these risk factors in relation to birthplace (Table 2), these results may be type 1 errors.
A comparison between migrants and non-migrants showed migrants to be taller and to have lower levels of BMI, DBP, HbA1c, HDL cholesterol, triglycerides, fibrinogen, hearing loss, and a lower prevalence of depressive symptoms. These effects were attenuated after adjustment for all potential confounders (listed in Table 3), but persisted (P < 0.05) for height, BMI, HDL cholesterol and triglycerides. In general, differences between migrants and non-migrants are considerably smaller than the regional variations in these outcomes.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionSupplementry materialSources of fundingAcknowledgementsReferences |
|---|
The ‘early origins’ hypothesis was stimulated by ecological (area-based) correlations,1–3 but it has been investigated mainly by studies of individuals.4 Our migrant study provides the most comprehensive test, to date, of the importance of early life factors as determinants of the geographical variations in risk of common diseases and risk factors for premature death in British adults. Our results generally implicate region of current residence as much more important than region of birth, particularly for BMI, BP and ventilatory function. However, the effect of region of current residence on ventilatory function was attenuated in analyses stratified by gender. None of the other outcomes exhibited this attenuation in gender stratified analyses. Some important risk factors, including cholesterol, glucose intolerance and fibrinogen, as well as indicators of sensory function and mental ill-health, showed little evidence of regional inequalities. Adult height provides a counter example, and as expected is influenced by region of residence early in life.
Few studies of disease risk among migrants within the UK have been reported. Among a representative 1% sample, enumerated in 1939 and at census in 1971, regional variations in mortality from ischaemic heart disease and stroke were related to both place of enumeration in 1939 and place of residence in 1971.7 Among men followed for 22 years in the British Regional Heart Study, zone of examination was more strongly associated with incident ischaemic heart disease (fatal and non-fatal) than was zone of birth, even after adjusting for major risk factors.13 BP levels in the same study were also influenced primarily by zone of examination rather than zone of birth,6 consistent with our results and also with studies of international migrants.14 We are not aware of previous migrant studies of other cardiovascular risk factors or the other outcomes we tested.
We found evidence of a more favourable cardiovascular risk profile among migrants, which is consistent with the lower levels of cardiovascular mortality among inter-regional migrants in Britain.7 However, not all measurements showed this pattern: interesting exceptions being total cholesterol, total IgE and anxiety symptoms. This suggests that it may be unwise to make generalized assumptions about ‘healthy migrant’ effects.
Over the past two decades, ample evidence has accumulated that individual risk of disease is predicted by measures of growth and health in the first few years of life.4 What is much less clear is the extent to which these individual-level associations explain the level of disease risk in different populations.15 The companion paper8 shows that for many of the outcomes studied, adult health is related to social inequalities in childhood after adjustment for adult social position. Comparing that article with this one, three features are noteworthy.
First, outcomes that show social inequalities do not necessarily vary by region, even among non-migrants. Secondly, all outcomes that were independently related to region of examination (obesity, BP and lung function) were also independently related to socio-economic position in adulthood, but the regional variations were independent of social position and vice versa. Thirdly, while all cardiovascular risk factors studied were related independently to childhood social class (along with hearing threshold, depression and chronic pain), there was no evidence of a strong influence of birthplace on any of these outcomes. This implies that different causal pathways and biological mechanisms underlie social and geographical distribution of disease risk in this cohort.
|
Conclusion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionSupplementry materialSources of fundingAcknowledgementsReferences |
|---|
Our results suggest that early life factors, including genetic variation, do not explain, to any great extent, the regional variation in cardiovascular and respiratory risk factors, sensory function and mental ill-health among middle-aged adults in Britain. Geographical inequalities in cardiovascular health are more strongly related to factors associated with region of residence that influence obesity, and BP, but not circulating lipids or plasma fibrinogen. The correlations observed between adult mortality rates and past infant mortality rates at the population (area) level1–3 are thus more likely to arise from historical continuities in lifestyle and environmental exposures5, than from ‘programming’ of cardiorespiratory risk, as originally implied.
Ethical approval for the medical examination of the British 1958 Birth Cohort was obtained from South East MREC (ref: 01/1/44).
|
Supplementry material |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionSupplementry materialSources of fundingAcknowledgementsReferences |
|---|
Supplementry material can be foung at IJE online (http://ije.oxfordjournals.org).
|
Sources of funding |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionSupplementry materialSources of fundingAcknowledgementsReferences |
|---|
The biomedical examination and related statistical analyses were funded by Medical Research Council grant G0000934, awarded under the Health of the Public initiative. The MRC played no role in study design, collection, analysis and interpretation of data, writing the report, or submitting the paper for publication.
Conflicts of Interest: None to declare.
KEY MESSAGES
|
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionSupplementry materialSources of fundingAcknowledgementsReferences |
|---|
We are especially grateful to the cohort members who participated in the 2002–2004 biomedical follow-up, and to the nurses, office and laboratory staff who contributed to the successful completion of the nationwide fieldwork. Bryan Rodgers was supported by Research Fellowships Nos 148948 and 366758 and by Program Grant No. 179805 from the National Health and Medical Research Council of Australia. Research at the Institute of Child Health and Great Ormond Street Hospitals for Children NHS Trust benefits from R&D funding received from the NHS Executive.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionSupplementry materialSources of fundingAcknowledgementsReferences |
|---|
1 Barker DJP, Osmond C. Infant mortality, childhood nutrition and ischaemic heart disease in England and Wales. Lancet (1986) i:1077–81.
2 Barker DJP, Osmond C. Death rates from stroke in England and Wales predicted from past maternal mortality. Brit Med J (1987) 295:83–86.
3 Barker DJP, Osmond C. Childhood respiratory infection and adult chronic bronchitis in England and Wales. Brit Med J (1986) 293:1271–75.
4 Kuh D, Ben Shlomo Y, eds. A life course approach to chronic disease epidemiology (2004) 2nd. Oxford: Oxford University Press.
5 Ben-Shlomo Y, Davey Smith G. Deprivation in infancy or adult life: which is more important for mortality risk? Lancet (1991) 337:530–34.[CrossRef][Web of Science][Medline]
6 Elford H, Phillips AN, Thomson AG, Shaper AG. Migration and geographic variations in blood pressure in Britain. Brit Med J (1990) 300:291–95.
7 Strachan DP, Leon DA, Dodgeon B. Mortality from cardiovascular diseases among interregional migrants in England and Wales. Brit Med J (1995) 310:423–427.
8 Power C, Atherton K, Strachan DP, et al. Lifecourse influences on health among British adults: effects of socioeconomic position in childhood and adulthoood. Int J Epidemiol (2007) 36:532–539.
9 Power C, Elliott J. Cohort profile: 1958 British birth cohort (National Child Development Study). Int J Epidemiol (2006) 35:34–41.
10 Recommended procedures for pure-tone audiometry using a manually operated instrument. Br J Audiol (1981) 15:213–16.[Medline]
11 Lewis G, Pelosi AJ, Araya R, Dunn G. Measuring psychiatric disorder in the community: a standardized assessment for use by lay interviewers. Psychol Med (1992) 22:465–86.[Web of Science][Medline]
12 Wolfe F, Smythe HA, Yunus MB, et al. The American College of Rheumatology 1990 Criteria for the Classification of Fibromyalgia. Report of the Multicenter Criteria Committee. Arthritis Rheum (1990) 33:160–72.[Web of Science][Medline]
13 Wannamethee SG, Shaper AG, Whincup PH, Walker M. Migration within Great Britain and cardiovascular disease: early life and adult environmental factors. Int J Epidemiol (2002) 31:1054–60.
14 Elford J, Ben-Shlomo Y. Geography and migration with special reference to cardiovascular disease. In: A life course approach to chronic disease epidemiology—Kuh D, Ben Shlomo Y, eds. (2004) 2nd. Oxford: oxford university press. 144–64.
15 Sick individuals and sick populations. Int J Epidemiol (1985) 14:32–38.[Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  G J Macfarlane, G Norrie, K Atherton, C Power, and G T Jones The influence of socioeconomic status on the reporting of regional and widespread musculoskeletal pain: results from the 1958 British Birth Cohort Study Ann Rheum Dis, October 1, 2009; 68(10): 1591 - 1595. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Dehghan, Q. Yang, A. Peters, S. Basu, J. C. Bis, A. R. Rudnicka, M. Kavousi, M.-H. Chen, J. Baumert, G. D.O. Lowe, et al. Association of Novel Genetic Loci With Circulating Fibrinogen Levels: A Genome-Wide Association Study in 6 Population-Based Cohorts Circ Cardiovasc Genet, April 1, 2009; 2(2): 125 - 133. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Tabassum, M. Kumari, A. Rumley, G. Lowe, C. Power, and D. P. Strachan Effects of Socioeconomic Position on Inflammatory and Hemostatic Markers: A Life-Course Analysis in the 1958 British Birth Cohort Am. J. Epidemiol., June 1, 2008; 167(11): 1332 - 1341. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K Atherton, E Fuller, P Shepherd, D P Strachan, and C Power Loss and representativeness in a biomedical survey at age 45 years: 1958 British birth cohort J Epidemiol Community Health, March 1, 2008; 62(3): 216 - 223. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Power, K. Atherton, D. P Strachan, P. Shepherd, E. Fuller, A. Davis, I. Gibb, M. Kumari, G. Lowe, G. J Macfarlane, et al. Life-course influences on health in British adults: effects of socio-economic position in childhood and adulthood Int. J. Epidemiol., June 1, 2007; 36(3): 532 - 539. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A.-M. N. Andersen Commentary: Life-course and social epidemiology, the biological fig leaf and Bob Dylan Int. J. Epidemiol., June 1, 2007; 36(3): 540 - 541. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||