New from the International Journal of Epidemiology
Press release issued: 24 March 2004
Studies on the reproductive health of Gulf War veterans, breast screening and low birth weight are published in the March edition of the International Journal of Epidemiology, edited in the Department of Social Medicine.
The first epidemiological study to be conducted in the UK on reproduction following service in the Gulf War in the early ‘90s is published today in the International Journal of Epidemiology (IJE), edited in the Department of Social Medicine at the University of Bristol.
The study, funded by the Ministry of Defence, was administered by the Medical Research Council (MRC) and produced by the London School of Hygiene & Tropical Medicine (LSHTM). It sought to determine whether the children of UK Gulf War veterans were more likely to suffer congenital malformations or for the pregnancies to end in miscarriage or stillbirth.
The findings reveal no strong evidence for a link between a father’s deployment to the Gulf War and increased risk of stillbirth or of chromosomal malformations, syndromes or certain specific malformations in offspring. Female veterans, meanwhile, were found to be at no greater risk of miscarriage than the comparison group, but there were too few stillbirths or malformations reported to allow for meaningful analysis.
However, the research findings did reveal some areas of concern. There was some evidence for an increased risk of malformations of the genital and renal system, and of other non-specific defects, but these associations were weakened when the analysis was restricted to clinically confirmed conditions (evidence from doctors).
The researchers also found a 40% increased risk of miscarriage among the pregnancies of Gulf War veterans. According to Dr Pat Doyle, Head of the Department of Epidemiology and Population Health at LSHTM, and one of the report’s authors, these figures need to be interpreted with caution.
Dr Doyle said: “Overall, we found no evidence that service in the Gulf is linked to fathering children with most major defects and this will be reassuring to veterans and their families. Although we did find a 40% increased risk of miscarriage among pregnancies reported by male Gulf War veterans, this may be the result of underreporting of early miscarriages by the control group, and we cannot at this stage conclude that there is a real link between miscarriage and a father’s service in the Gulf War. We are now studying this in more depth.”
The important issue of the effectiveness of breast screening programmes is addressed in a paper and commentaries in the March issue of the IJE.
In recent years, a series of papers published by two researchers at the Cochrane institute in Copenhagen, Peter Gøtzsche and Ole Olsen, controversially concluded that mammography does not save lives but rather exposes women to unnecessary surgical procedures.
In a paper in this month’s IJE, researchers in the United States argue that the statistical analysis on which this opinion was based is flawed and that breast screening does save lives.
Gøtzsche and Olsen analysed the results of existing trials but chose to exclude two studies that had found breast screening to have a positive effect – the Health Insurance Plan (HIP) trial in the United States and the Two-County trial in Sweden – arguing that they were of poor quality.
David Freedman, Diana Petitti and James Robins of the University of California, Berkeley, re-examined the findings of these two positive studies and also considered the best-known of the negative studies, the Canadian National Breast Screening Study (CNBSS), which was included in Gøtzsche and Olsen’s analysis.
They point to a number of misunderstandings on the part of Gøtzsche and Olsen regarding the statistical analysis of the data, and argue that they were wrong to ignore the findings of the HIP and Two-County studies. They also question the methodology of the CNBSS, concluding: “Clinical trials of mammography have led to substantial advances in understanding breast cancer, and a substantial reduction in mortality from this disease.”
However, four commentaries on the paper argue that many questions still remain. These questions are addressed in a rejoinder by Freedman and his co-authors.
Babies with low birth size have a greater risk of mortality in early adulthood according to new research published in the March issue of the IJE.
The study by Anne-Marie Nybo Andersen and Merete Osler of the University of Copenhagen looks at the relationship between birth dimensions and mortality.
It found that low birth weight and especially short birth length were strongly associated with earlier death as an adult. The link remained even after adjustment for parental lifespan and indicators of social position, although these do have some effect.
The report studied 10,753 male children born in the metropolitan area of Copenhagen in 1953. This cohort was followed from 1968 to 2002, and information on deaths of members of the cohort and deaths of their parents was tracked using the Danish Civil Registration System. Cause-of-death data was based on official death certificates and categorized into death from cardiovascular disease, death from cirrhosis, accident and suicides, and other causes.
The inverse association between size at birth and mortality was found irrespective of cause of death, but seemed to be strongest for cirrhosis of the liver. This was an unexpected finding, though some studies have indicated that liver function in adult life might be influenced by experiences in the womb. As death from liver cirrhosis among men in the cohort’s age group is clearly related to alcohol consumption, this could suggest an increased susceptibility to alcohol among those born small (organ programming). However, it may be due to an individual’s social circumstances, which have an influence on both rates of alcohol consumption and birth size.
The study also found that low birth size is inversely associated with parental mortality and that maternal mortality is much more strongly associated with offspring birth size than paternal mortality. This suggests that rather than common genetic factors, inherited equally from both mother and father, it is maternal health characteristics (eg hypertension) and/or behaviour (eg smoking, poor nutrition) influencing both offspring birth size and maternal mortality which play a part in the associations found.Professor Nybo Andersen said: “These findings indicate that confounding by factors related to social position, specific organ programming, and common genetic factors might all play a role as mechanisms behind the association between birth size and mortality. Gestational age at birth may play an independent role, which calls for further investigation. Our findings disprove the idea that disproportionate fetal growth retardation in general, resulting in thinness at birth, is causally related to adult mortality.”
The International Journal of Epidemiology
The International Journal of Epidemiology is a key journal in the field of epidemiology and public health, published six times per year by Oxford University Press. It is edited at the Department of Social Medicine, University of Bristol, which is a leading centre for epidemiology, health services research and public health in the UK and was one of only three to be awarded the top 5* grade in the 2001 Research Assessment Exercise.