Skip Navigation


American Journal of Epidemiology Advance Access originally published online on May 7, 2007
American Journal of Epidemiology 2007 166(2):160-169; doi:10.1093/aje/kwm054
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
166/2/160    most recent
kwm054v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Davey Smith, G.
Right arrow Articles by Lawlor, D. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Davey Smith, G.
Right arrow Articles by Lawlor, D. A.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

American Journal of Epidemiology © The Author 2007. Published by the Johns Hopkins Bloomberg School of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oxfordjournals.org.

ORIGINAL CONTRIBUTIONS

Offspring Birth Weight and Parental Mortality: Prospective Observational Study and Meta-Analysis

George Davey Smith1, Elina Hyppönen2, Chris Power2 and Debbie A. Lawlor1

1 Department of Social Medicine, University of Bristol, Bristol, United Kingdom
2 Centre for Paediatric Epidemiology and Biostatistics, Institute of Child Health, London, United Kingdom

Correspondence to George Davey Smith, Department of Social Medicine, University of Bristol, Canynge Hall, Whiteladies Road, Clifton, Bristol BS8 2PR, United Kingdom (e-mail: George.Davey-Smith{at}bristol.ac.uk).

Received for publication November 4, 2005. Accepted for publication January 11, 2007.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
The authors have investigated associations between offspring size at birth and parental cardiovascular disease mortality among 12,086 mothers and 6,936 fathers of participants in the British 1958 birth cohort. Birth weight was inversely associated with all-cause mortality and cardiovascular mortality in both mothers and fathers. The adjusted hazard ratio of cardiovascular disease mortality for a 1-standard deviation increase in offspring birth weight in mothers was 0.87 (95% confidence interval (CI): 0.82, 0.93) and in fathers was 0.94 (95% CI: 0.89, 0.99). The association was not specific for cardiovascular disease. In fathers, similar weak associations with violent and accidental deaths, stomach cancer, and alcohol- and smoking-related outcomes were found. Weak associations for these outcomes were also found for mothers, but the magnitude of the association with cardiovascular disease was greater than with any other outcomes. In a meta-analysis pooling results from this study with six others, the adjusted hazard ratio of cardiovascular disease mortality among mothers was 0.75 (95% CI: 0.67, 0.84) and that among fathers was 0.93 (95% CI: 0.91, 0.95), with evidence that the difference in effect between mothers and fathers was not due to chance (p < 0.001). The weak association of offspring birth weight with cardiovascular disease in fathers may be due to residual confounding by factors such as socioeconomic position and smoking that they share with the offspring's mother and that would therefore be associated with low offspring birth weight as well as adverse outcomes in the father. The stronger association in mothers is consistent with intergenerational effects on intrauterine growth and with the fetal origins hypothesis.

birth weight; infant, newborn; meta-analysis; mortality; parents; prospective studies

Abbreviations: CI, confidence interval


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
Fetal growth (assessed by birth weight or birth weight standardized for gestational age) has been associated with subsequent blood pressure, insulin resistance, and cardiovascular disease in a large number of studies carried out in various contexts (1–4). These associations have been interpreted as demonstrating the establishment, during fetal development, of processes with long-term consequences for metabolic and cardiovascular function (5). Fetal undernutrition is considered to "program" future disease risk, with risk particularly increased if there are overnutrition and accelerated growth in the postnatal period. If intrauterine nutrition is an important determinant of later cardiovascular disease risk, this would have important implications for policies aimed at reducing, or preventing an increase in, disease burden. However, it has also been hypothesized that common genetic factors could underlie both restricted fetal growth and later insulin resistance and cardiovascular disease risk, leading to an association that would not be reversible by environmental manipulation to improve fetal nutrition and growth (6). Distinguishing between these two potential mechanisms underlying the association between fetal growth and later disease is clearly of considerable importance for understanding the etiology and prevention of cardiovascular disease.

One approach to investigating this issue is to examine cardiovascular disease risk in parents in relation to the birth characteristics of their offspring (7, 8). If the birth weight–cardiovascular disease association seen within individuals is due to a genetic mechanism from both maternal and paternal genes, then one would predict that offspring birth weight would be inversely associated with cardiovascular disease in both mothers and fathers. Associations in fathers are of particular importance, since many factors that could influence maternal cardiovascular disease risk (e.g., maternal diet and smoking) could also influence the intrauterine environment—and thus birth weight—of her offspring (9), whereas an association between offspring birth weight and the biologic father's cardiovascular disease risk is more likely to be explained by genetic factors.

Six previous studies have investigated this association with respect to mothers (10–14), with three of these also examining the association with respect to fathers (7, 13, 14). In these studies, offspring birth weight is inversely associated with both maternal and paternal cardiovascular disease, with some evidence that this association is weaker in fathers compared with mothers. The possibly weaker association in men has been interpreted as suggesting that the association between birth weight and cardiovascular disease is influenced by both (maternal) environmental factors and genetic factors (8). However, with respect to paternal cardiovascular disease mortality in particular (the crucial test of the hypothesis), data are currently sparse and there are important limitations to the studies examining this association to date. Self-report of offspring birth weight was used in one study and, in others, it was not possible to adjust for gestational age or other potentially important confounding factors (refer to table 4). Further, individually none of these studies had the statistical power to detect an important maternal/paternal difference in the effect of offspring birth weight on cardiovascular disease. We have therefore investigated the associations between offspring birth characteristics and parental mortality in the 1958 British birth cohort and undertaken a systematic review and meta-analysis of all studies to date that have examined the association between offspring birth weight and parental cardiovascular disease.


View this table:
[in this window]
[in a new window]

  		TABLE 4. Published studies of the association between parental cardiovascular disease risk and offspring birth weight

 

   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 References
 
All births in England, Wales, and Scotland, during 1 week in March 1958, were included in the Perinatal Mortality Survey (coverage: 98 percent) (15, 16). When the children born to the cohort were aged 7 years, they were recontacted, and the study continued as the 1958 cohort (also known as the National Child Development Study). Participants in the cohort have been followed up regularly thereafter, and detailed information has been collected on growth, health, social, and behavioral indicators for the cohort members, their parents, and offspring (16, 17). We have recently described in detail the methods used to trace the biologic parents for all cohort members participating in the first follow-up survey in 1965 (target sample: 15,888 parent pairs) (18, 19). Parents for English and Welsh cohort members were traced and flagged at the National Health Service Central Register at Southport, and parents of Scottish residents were traced through the equivalent register in Edinburgh. The overall tracing rate for the mothers was 94.9 percent (n = 15,070) and for the fathers was 90.2 percent (n = 14,323) (18). Ethical approval for the tracing study was obtained from the Local Research Ethics Committee (Great Ormond Street Hospital/Institute of Child Health, London).

Birth data for the cohort members (including birth weight in pounds and ounces, as well as gestational age in days from the first day of the last menstrual period) were collected in 1958 largely by midwives in charge of the deliveries; supplementary information was obtained from obstetric records and interviews with the mother (15). Information on gestational blood pressure was collected from the obstetrics records, and "preeclampsia" was defined as increased blood pressure (diastolic: >90 mmHg) with proteinuria in the absence of urinary tract infection (20).

We used data on maternal smoking reported in 1958 (none, before pregnancy, up to the fifth month, and after the fifth month). Information on paternal smoking was collected in 1974 (none, 1–10, 11–20, >20 cigarettes per day). Maternal parity (including stillbirths and miscarriages after the 28th week of gestation) was determined up to the birth of the index child in 1958. Prepregnancy weights for the mothers were self-reported, and heights were measured in 1958; heights and weights of fathers were reported in 1969. The body mass index of parents was determined as weight (kg)/height (m)2. Social class in 1958 was based on the father's occupation and grouped according to the Registrar General's Classification (I–II, III nonmanual, III manual, and IV–V). Individuals with no male head of household (i.e., unmarried mothers) were coded as class IV–V because numbers were too small for separate analysis and, with respect to numerous characteristics, they closely resembled class IV–V.

Individuals in the current analysis were restricted to mothers (n = 14,585) and fathers (n = 13,861) with data on offspring birth weight. Final analyses were further restricted to parents with full information on background factors (12,086 mothers and 6,936 fathers). Data for the fathers were more strongly reduced, as the availability of background paternal information was dependent upon participation of their child in the 1969 and 1974 surveys, whereas maternal background information was collected at the cohort's initiation. Mortality rates were slightly lower for those mothers and fathers with full information on background factors (6.2 and 11.0 per 1,000, respectively) compared with those for whom these details were missing (7.9 and 13.0 per 1,000). Individuals with full information were somewhat more likely to be from higher social classes compared with those with missing data (18 percent vs. 12 percent of mothers and 19 percent vs. 17 percent of fathers in class I–II). Birth weight was somewhat higher for offspring of mothers with complete data (3.35 kg vs. 3.29 kg, respectively), but there was no marked difference in offspring birth weight by availability of full information on paternal background factors (3.35 kg vs. 3.34 kg).

Data on the cause of death were coded according to the International Classification of Diseases, Tenth Revision. The main endpoints used in the study were all-cause mortality (including unknown causes) and mortality from circulatory disease (codes I00–I99X, G45, G46), coronary heart disease (codes I20–I25), and stroke (codes I61–I69, G45, G46; subarachinoid subtype excluded). In addition, we assessed association of offspring birth weight with mortality from respiratory disease (codes J00–J998A), chronic obstructive pulmonary disease (codes J40–J47), cancer (codes C00–C97X), stomach cancer (code C16), lung cancer (code C34), breast cancer (code C50), prostate cancer (code C61), accidents and violence (codes S00–Z999), and suicide (codes X60–X84 and Y10–Y34). We further grouped the causes of mortality by smoking-related (codes C00–C159, C25–C259, C30–C349, C64X–C689) and alcohol-related (C01–C06, C10, C13–C15, C22, C32, F10, K70, K74.6, S1–Y98) causes. Our main hypotheses concerned the association of offspring birth weight with parental cardiovascular disease. Other outcomes were assessed as a test of specificity to determine whether associations with cardiovascular disease might be explained by residual confounding (21). For example, associations between offspring birth weight and accidents and violence and smoking- or alcohol-related mortality, in addition to an association with cardiovascular disease, would indicate residual confounding by socioeconomic position and smoking as an explanation for the association.

Statistical analysis
The relation between offspring birth weight and parental mortality was evaluated by use of single- and multiple-term Cox proportional hazards models. Calendar time originating from the approximate time of offspring conception (July 5, 1957) was used as the main time scale in all analyses. The follow-up for each parent lasted up to death, emigration (245 fathers and 190 mothers), event cancellation (152 fathers and 260 mothers), or the end date for this study (December 31, 2003) (18, 19). All models were adjusted for birth year and stratified by offspring sex. Offspring birth weight was categorized to gender-specific fifths. Tests for trend were done with birth weight standardized for gestational age and sex. There was a suggestion of curvature in the association between offspring birth weight and maternal mortality from all causes, circulatory diseases, and cancer. However, for all outcomes, the quadratic term was not significant after adjustment for smoking, social, and anthropometric background factors, and therefore only results for the linear measure are presented.

Systematic review
Papers published prior to September 2005 were identified through a search of MEDLINE (www.ncbi.nlm.nih.gov), ISI Web of Knowledge (http//:wok.mimas.ac.uk), and Google (www.google.com) using the following search terms (indexed and text word searches were used): "birthweight or birth weight," "fetal growth or foetal growth," "intrauterine growth," "parental mortality," "parental," "cardiovascular disease," "transgenerational or trans-generational," and "intergenerational or inter-generational." A cited-reference search of retrieved articles was carried out, and publications were also identified by review of the bibliographies of retrieved articles. Information on characteristics of the study population; outcome, exposure, and covariate assessment; results; and conclusions were abstracted for each paper. Where necessary, additional information was obtained from the authors, and a meta-analysis of all studies examining the association of offspring birth weight with parental cardiovascular disease was conducted (one study in which the outcome was a continuous measure of carotid intimal-medial thickness was not included). We undertook separate random effects meta-analyses (22) for mothers and fathers, including all studies where appropriate, irrespective of whether there were both mothers and fathers included in the individual study. Evidence for heterogeneity between studies in these meta-analyses was assessed by computing a Q statistic (22). The effects in mothers and fathers were compared by computing a z test from the standard errors of each pooled effect in the meta-analyses. Tests referred to as "Egger and Begg tests" were computed to examine small-study bias (indicative of publication bias) (22). All analyses were conducted in STATA, version 9, software (23).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 References
 
In table 1, we show maternal and paternal characteristics by fifth of the birth weight distribution for their offspring. Mothers who had larger babies were older, had higher body mass index, were taller, were less likely to smoke during pregnancy, and were less likely to suffer from preeclampsia. For fathers, the results were in the same direction, although weaker for body mass index, height, and smoking. For fathers, the proportion who were in manual social classes was lower among those who had higher offspring birth weights.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Maternal and paternal characteristics by fifths of offspring birth weight, 1958 British birth cohort (National Child Development Study)

 
In table 2, we present maternal and paternal mortality according to offspring birth weight and demonstrate inverse associations for coronary heart disease, stroke, chronic obstructive pulmonary disease, and lung cancer among mothers. Stomach cancer and suicide also showed strong inverse associations for mothers, but with considerable imprecision around the effect estimates. For fathers, the associations were weaker although in the same direction as for mothers, with the exception of accidental and violent deaths, which showed a stronger association in fathers than in mothers.


View this table:
[in this window]
[in a new window]

  		TABLE 2. Parental mortality (per 1,000 person-years) by fifths of offspring birth weight, 1958 British birth cohort (National Child Development Study)

 
In table 3, we present results adjusted for age and gender of offspring and then additionally for offspring gestational age, social class of the father, parity of the mother, and preeclampsia during pregnancy. The mother's mortality was additionally adjusted for her height and body mass index in 1958 and smoking during pregnancy. For fathers, we additionally adjusted for height and body mass index in 1969 and smoking habits in 1974. The adjustments attenuated, but did not abolish, the associations of offspring birth weight with maternal mortality from coronary heart disease and stroke. There was considerably greater attenuation of the associations of offspring birth weight and mortality from chronic obstructive pulmonary disease and lung cancer after adjustment. A similar picture was seen for fathers, although there was no evidence of an inverse association with stroke after age adjustment.


View this table:
[in this window]
[in a new window]

  		TABLE 3. Hazard ratio (with confidence intervals) for mortality for 1-standard deviation higher birth weight, before and after adjustments, 1958 British birth cohort (National Child Development Study)

 
The association of offspring birth weight and coronary heart disease mortality among mothers is weaker in this study than in previous ones, and we hypothesized that this could be due to the longer follow-up period in this study. If the higher coronary heart disease mortality of mothers was in part at least due to disturbances evident during pregnancy that would influence both offspring birth weight and subsequent coronary heart disease risk, it would be expected that this would be most evident in the period relatively soon after birth. Our results suggested that this may be the case. Over the first 25 years following the birth of the child whose birth weight is utilized as the exposure measure in this study, each standard deviation higher birth weight was associated with an age-adjusted hazard ratio of 0.65 (95 percent confidence interval (CI): 0.50, 0.85) whereas, in the follow-up period subsequent to this, the hazard ratio was 0.82 (95 percent CI: 0.75, 0.90). A formal test of difference of these hazard ratios gave a p value of 0.13. However, there was no suggestion that the association of birth weight with all circulatory diseases (in the first 25 years of follow-up: hazard ratio = 0.83, 95 percent CI: 0.71, 0.97; in subsequent years of follow-up: hazard ratio = 0.86, 95 percent CI: 0.81, 0.91; p for difference = 0.72) or with total mortality (in the first 25 years of follow-up: hazard ratio = 0.90, 95 percent CI: 0.82, 0.98; in subsequent years of follow-up: hazard ratio = 0.89, 95 percent CI: 0.86, 0.93; p for difference = 0.94) varied by length of time since follow-up.

For the systematic review, we identified six published studies of the association, and these are summarized in table 4. Three of these studies included information on both mothers and fathers, but in one of these three the outcome was carotid intimal-medial thickness, and we were therefore unable to include this in our meta-analysis. We pooled the adjusted hazard ratios of cardiovascular disease mortality risk per standard deviation of birth weight from six studies of mothers (five previously published and the results from the study presented here) and three studies of fathers (two previously published and results from the study presented here). Among mothers, the pooled, adjusted (for the factors listed in table 4) hazard ratio of cardiovascular disease mortality for a 1-standard deviation increase in birth weight was 0.75 (95 percent CI: 0.67, 0.84) and, among fathers, the equivalent association was 0.93 (95 percent CI: 0.91, 0.95), with statistical evidence of a difference between these two effects (p < 0.001). In the meta-analysis of associations with cardiovascular disease mortality among mothers, there was strong statistical evidence of heterogeneity (p < 0.001) that was largely due to a weaker effect in mothers in the study presented here compared with previously published studies. With this study removed, there was no evidence of heterogeneity in the pooled estimate for mothers (p = 0.4), and the association of offspring birth weight with cardiovascular disease mortality was slightly strengthened compared with that with this study included (hazard ratio = 0.72 vs. 0.75). There were no evidence of heterogeneity among the three studies of fathers (p = 0.9) and no strong statistical evidence of small-study bias (indicating publication bias) in either meta-analysis (p = 0.6 for the Egger test and 0.4 for the Beggs test in a meta-analysis of association in mothers; p = 0.3 for the Egger test and 0.1 for the Beggs test in that of association in fathers). When we restricted the meta-analyses only to those three studies that had assessed associations in both mothers and fathers, the pooled effect in mothers remained stronger than that in fathers: hazard ratio = 0.75 (95 percent CI: 0.67, 0.84) versus hazard ratio = 0.93 (95 percent CI: 0.91, 0.95; p = 0.001 for difference).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
References
 
In this study, we found an inverse association between offspring birth weight and cardiovascular disease mortality in mothers and fathers that persisted after adjustment for potential confounding factors and was somewhat weaker for fathers compared with mothers. These findings are consistent with those from two previous studies that have examined this association in both mothers and fathers (7, 14).

The association between offspring birth weight and maternal cardiovascular disease may reflect maternal/fetal nutritional factors and intrauterine programming, as women who themselves had poor fetal growth and low birth weight tend to have offspring who are small for their gestational age (24, 25). This effect may be mediated via maternal pelvic restriction, poor placental growth, and hence a programming effect of intrauterine nutrition (26) or via shared environmental exposures, for example, cigarette smoking, across generations. The spousal correlation for birth weights is remarkably low (r = 0.02) (27) so that, despite the possibility of assortative mating by factors that reflect each parent's birth weight, it is unlikely that any observed increased risk for fathers reflects paternal low birth weight.

In addition to genetic factors, shared environmental and behavioral risk factors between fathers and mothers are a plausible reason for fathers showing an increased cardiovascular disease risk if their offspring birth weight is low. For example, mothers who smoke are more likely to have children with men who smoke. Maternal smoking will be associated with lower offspring birth weight, and a greater likelihood that the father will smoke if the mother does will result in his increased risk of cardiovascular disease. Similarly, low family socioeconomic position could link low offspring birth weight and increased cardiovascular disease risk in both mothers and fathers. Some support that parental shared socioeconomic and behavioral risk factors explain the offspring birth weight–cardiovascular disease risk in fathers comes from the nonspecific nature of the weak associations in fathers, with inverse associations between birth weight and other smoking-related outcomes (smoking-related cancers and respiratory disease) and with outcomes that are known to be associated with low socioeconomic position (stomach cancer and accidents and violence) of a similar magnitude to those for cardiovascular disease in this cohort. Similar nonspecificity of the association was also found in a large record linkage study of Swedish women and men (14). Thus, the weak association between offspring birth weight and paternal cardiovascular disease that remains in adjusted analyses may reflect residual confounding due to measurement error in the assessment of smoking and the fact that we have taken account of socioeconomic position at only one life stage and using only one domain (occupational social class), which may be insufficient to fully capture the effect of socioeconomic position on both offspring birth weight and parental cardiovascular disease risk.

There is a stronger association between offspring birth weight and maternal cardiovascular disease risk compared with that with paternal risk. Possible explanations include paternal misclassification (as some fathers in these studies will not be the biologic parent, and this would be expected to dilute the paternal association); direct effects of maternal health-related behaviors, such as smoking, heavy alcohol consumption, and poor diet, on both offspring birth weight and maternal mortality risk; poor maternal nutrition in her childhood, resulting in her own reduced vitality and therefore increased cardiovascular disease risk and low offspring birth weight (resulting in stronger associations in mothers due to both maternal-specific environmental as well as possible genetic factors and environmental factors that are shared among mothers, fathers, and their offspring); and maternal imprinting (i.e., the phenotype resulting from a particular gene locus is differentially modified by the sex of the parent contributing that particular allele) or other epigenetic effects (i.e., a change in the outcome of a gene that is controlled by nongenetic factors so that the phenotype resulting from a particular gene is modified by environmental exposures) (28). It is also possible that some of the maternal effect reflects reduced offspring birth weight due to health-compromising factors (including illness itself) during pregnancy, which is in turn associated with increased maternal mortality risk. In this case, the expectation would be that the elevated mortality risk would be greater for periods closer in time to pregnancy and, as with other cases of health-related selection, this effect would reduce over time. For coronary heart disease mortality, we found weak evidence in support of this, and this could also explain the rather weaker associations found in our study than in other studies of this issue, since our study had considerably longer follow-up than did previous studies. This issue should be investigated in other studies with long-term follow-up.

Study strengths and limitations
The results from this large cohort study add to the scant evidence examining the association of offspring birth weight with paternal cardiovascular disease risk. We have examined other disease outcomes as a test of specificity, and this provides some evidence that residual confounding by smoking and socioeconomic position may explain the remaining weak inverse association between offspring birth weight and father's cardiovascular disease risk. Our meta-analysis provided sufficient power to determine whether there was statistical evidence of a difference in effect between fathers and mothers.

In the 1958 birth cohort, we were able to include fathers in the adjusted analysis only if their child survived and participated in the 1969 and 1974 surveys, whereas similar restrictions were not applied to the mothers. When we repeated our analyses including only those mothers whose children participated in these surveys, the results did not differ substantively from those presented here. We had greater difficulty tracing fathers of cohort participants than mothers (18). However, the validity of the trace has been demonstrated by the strong association between smoking and lung cancer (observed for both fathers and mothers) (19). Further, the effect estimate found for fathers in this study is very similar to that found in a very large study from Sweden, in which linkage of family members was uncomplicated and had a very high level of accuracy.

Conclusion
Taking the results of our cohort study and the meta-analysis together, we have found evidence for an inverse association of offspring birth weight with cardiovascular disease mortality in both mothers and fathers. The effect in fathers is weak and may be explained by residual confounding due to shared socioeconomic position and smoking habits between mothers and fathers that will affect intrauterine growth and both parents' risk of cardiovascular disease. The stronger effect in mothers suggests specific maternal factors (over and above the mechanisms—shared environmental factors between mothers and fathers or a possible genetic mechanism—that could explain the association in both parents) and is consistent with a fetal programming hypothesis. Thus, while this work is unable to provide a definitive answer on the mechanism explaining the association between birth weight and cardiovascular disease, of the two proposed hypotheses it provides more support for the importance of fetal programming than of direct genetic mechanisms.


   ACKNOWLEDGMENTS
 
The study of parental mortality was funded by the Wellcome Trust (grant 059480/99/A). Research at the Institute of Child Health and Great Ormond Street Hospital for Children National Health Service Trust benefits from research and development funding received from the National Health Service Executive. E. H. and D. A. L. are funded by United Kingdom Department of Health career scientist awards.

The authors thank G. C. Smith, University of Cambridge (12), and Dr. C. Hart, University of Glasgow (7), who provided additional information concerning their studies, thus enabling inclusion of these studies in the present meta-analysis. The Office for National Statistics Medical Research at Southport and General Register Office (Scotland) are acknowledged for the tracing and flagging of study subjects.

Conflict of interest: none declared.


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 

  1. Barker DJP. Mothers, babies and health in later life (1998) London, United Kingdom: Churchill Livingstone.
  2. McKeigue PM, Lithell HO, Leon DA. Glucose tolerance and resistance to insulin-stimulated glucose uptake in men aged 70 years in relation to size at birth. Diabetologia (1998) 41:1133–8.[CrossRef][Web of Science][Medline]
  3. Leon DA, Koupilová I. Birth weight, blood pressure and hypertension: epidemiological studies. In: Fetal origins of cardiovascular and lung disease—Barker DJP, ed. (2000) New York, NY: Marcel Dekker, Inc. 23–48.
  4. Lawlor DA, Ronalds G, Clark H, et al. Birthweight is inversely associated with incident coronary heart disease and stroke among individuals born in the 1950s: findings from the Aberdeen Children of the 1950s prospective cohort study. Circulation (2005) 112:1416–20.
  5. Bateson P, Barker D, Clutton-Brock T, et al. Developmental plasticity and human health. Nature (2004) 430:419–21.[CrossRef][Medline]
  6. Hattersley AT, Tooke JE. The fetal insulin hypothesis: an alternative explanation of the association of low birthweight with diabetes and vascular disease. Lancet (1999) 353:1789–92.[CrossRef][Web of Science][Medline]
  7. Davey Smith G, Hart C, Ferrell C, et al. Birth weight of offspring and mortality in the Renfrew and Paisley study: prospective observational study. BMJ (1997) 315:1189–93.[Abstract/Free Full Text]
  8. Lawlor DA, Ben-Shlomo Y, Leon DA. Pre-adult influences on cardiovascular disease. In: A life course approach to chronic disease epidemiology—Kuh D, Ben-Shlomo Y, eds. (2004) 2nd ed. Oxford, United Kimgdom: Oxford University Press. 41–76.
  9. Sattar N, Greer IA. Pregnancy complications and maternal cardiovascular risk: opportunities for intervention and screening? BMJ (2002) 325:157–60.[Free Full Text]
  10. Davey Smith G, Whitley E, Gissler M, et al. Birth dimensions of offspring, premature birth, and the mortality of mothers. Lancet (2000) 356:2066–7.[CrossRef][Web of Science][Medline]
  11. Davey Smith G, Harding S, Rosato M. Relation between infants' birth weight and mothers' mortality: prospective observational study. BMJ (2000) 320:839–40.[Free Full Text]
  12. Smith GC, Pell JP, Walsh D. Pregnancy complications and maternal risk of ischaemic heart disease: a retrospective cohort study of 129,290 births. Lancet (2001) 357:2002–6.[CrossRef][Web of Science][Medline]
  13. Lawlor DA, Davey Smith G, Whincup P, et al. The association between offspring birth weight and atherosclerosis in middle aged men and women: British Regional Heart Study. J Epidemiol Community Health (2003) 57:462–3.[Free Full Text]
  14. Davey Smith G, Sterne J, Tynelius P, et al. Birth weight of offspring and parental cardiovascular mortality: record linkage study. Epidemiology (2005) 16:563–9.[CrossRef][Web of Science][Medline]
  15. Butler NR, Bonham DG. Perinatal mortality (1963) Edinburgh, Scotland: Livingstone.
  16. Power C, Elliott J. Cohort profile: 1958 British birth cohort (National Child Development Study). Int J Epidemiol (2006) 35:34–41.[Free Full Text]
  17. National Birthday Trust Fund, National Children's Bureau, City University Social Statistics Research Unit. National Child Development Study. Composite file including selected perinatal data and sweeps one to five. (Computer file) (1994) Colchester, Essex, United Kingdom: The Data Archive. (Study no. 3148).
  18. Hyppönen E, Davey Smith G, Shepherd P, et al. An intergenerational and lifecourse study of health and mortality risk in parents of the 1958 birth cohort (I): methods and tracing. Public Health (2005) 119:599–607.[CrossRef][Web of Science][Medline]
  19. Hyppönen E, Davey Smith G, Shepherd P, et al. An intergenerational and lifecourse study of health and mortality risk in parents of the 1958 birth cohort (II): mortality rates and study representativeness. Public Health (2005) 119:608–15.[CrossRef][Web of Science][Medline]
  20. Butler NR, Alberman E. Perinatal problems (1969) Edinburgh, Scotland: Livingstone.
  21. Lawlor DA, Davey Smith G, Bruckdorker KR, et al. Those confounded vitamins: what can we learn from the differences between observational versus randomised trial evidence? Lancet (2004) 363:1724–7.[CrossRef][Web of Science][Medline]
  22. Sterne JAC, Bradburn MJ, Egger M. Meta-analysis in STATA. In: Systematic reviews in health care: meta-analysis in context—Egger M, Davey Smith G, Altman D, eds. (2001) 2nd ed. London, United Kingdom: BMJ Publishing Group. 347–69.
  23. StataCorp LP. Intercooled STATA, version 9.1 for Windows (2005) College Station, TX: StataCorp LP.
  24. Klebanoff MA, Schulsinger C, Mednick BR, et al. Preterm and small-for-gestational-age birth across generations. Am J Obstet Gynecol (1997) 176:521–6.[CrossRef][Web of Science][Medline]
  25. Ramakrishnan U, Martorell R, Schroeder DG, et al. Role of intergenerational effects on linear growth. J Nutr (1999) 129(suppl):544S–9S.[Web of Science][Medline]
  26. Martyn CN, Barker DJ, Osmond C. Mothers' pelvic size, fetal growth, and death from stroke and coronary heart disease in men in the UK. Lancet (1996) 348:1264–8.[CrossRef][Web of Science][Medline]
  27. Magnus P, Gjessing HK, Skrondal A, et al. Paternal contribution to birth weight. J Epidemiol Community Health (2001) 55:873–7.[Abstract/Free Full Text]
  28. Davey Smith G. Genetic risk factors in mothers and offspring. (Letter). Lancet (2001) 358:1268.[Web of Science][Medline]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Epidemiol RevHome page
M. R. Kramer and C. R. Hogue
What Causes Racial Disparities in Very Preterm Birth? A Biosocial Perspective
Epidemiol. Rev., November 1, 2009; 31(1): 84 - 98.
[Abstract] [Full Text] [PDF]

Home page
 Int J EpidemiolHome page
K. R Risnes, T. I L Nilsen, P. R Romundstad, and L. J Vatten
Head size at birth and long-term mortality from coronary heart disease
Int. J. Epidemiol., August 1, 2009; 38(4): 955 - 962.
[Abstract] [Full Text] [PDF]

Home page
 J. Epidemiol. Community HealthHome page
L H Mortensen, F Diderichsen, G D. Smith, and A M N. Andersen
Time is on whose side? Time trends in the association between maternal social disadvantage and offspring fetal growth. A study of 1 409 339 births in Denmark, 1981-2004
J Epidemiol Community Health, April 1, 2009; 63(4): 281 - 285.
[Abstract] [Full Text] [PDF]

Home page
 Int J EpidemiolHome page
N. Bergvall, A. Lindam, Y. Pawitan, P. Lichtenstein, S. Cnattingius, and A. Iliadou
Importance of familial factors in associations between offspring birth weight and parental risk of type-2 diabetes
Int. J. Epidemiol., February 1, 2008; 37(1): 185 - 192.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
166/2/160    most recent
kwm054v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Davey Smith, G.
Right arrow Articles by Lawlor, D. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Davey Smith, G.
Right arrow Articles by Lawlor, D. A.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?