A study published last week in the American Journal of Epidemiology reported that children born to obese mothers were 57% more likely to develop leukaemia before the age of 14. The study was picked up by the medical and mainstream press – with the Daily Mail take here a fair representation of the general coverage.
The coverage highlights three common issues concerning reporting of research relating to risk in pregnancy, each of which is being considered as part of the WRISK research.
Reporting based on relative rather than absolute risk
Media reporting of the study limited their reporting to relative risks, which can seem to exaggerate risks when there is a low baseline risk. A 57% increase in the risk of leukaemia sounds very high. However, childhood leukaemia is thankfully relatively rare. Of the more than 1.8 million births analysed by the researchers from the University of Pittsburgh over 13 years, there were just under 750 cases. Risk was statistically elevated in the highest BMI groups (40 or more), with proportionately more children receiving a diagnosis of leukaemia compared to women with a ‘normal’ BMI (18.5 – 24.9), yet the overall risk in both groups remains low: 0.056% of births in the high BMI group v 0.039% of births in the ‘normal’ BMI group were affected.
This journey of the risk message was a topic we looked at in this previous WRISK blog here. It’s important to note that the media cannot be held entirely responsible for alarmist headlines: headlines and top-line findings are often taken directly from academic press releases, and/or highlighted within the studies themselves. In the case of this study, the ‘57% increased risk’ figure is lifted from the study abstract, where the absolute risk is not expressed.
Under-emphasising that association does not equal causation
While treatments and survival rates for Childhood leukemia have improved dramatically, this remains a condition whose causes we do not properly understand. There is a need to identify and explore associations and mechanisms. It may be that the finding of an association between maternal obesity and childhood leukaemia will help further the understanding of potential origins of this disease and such epidemiological studies can inform where further work may be indicated. However, the difference between association and causation needs to be clear to clinicians and to the general public, including to the women and their families on the receiving end of research findings. Given the inability to control for all possible confounders the paper cannot exclude that the link was not related to other factors associated with maternal obesity.
A jump to solutions based on individual behaviour-change
The initial WRISK project stakeholder meeting identified a concern that individual behaviour change by expectant mothers is increasingly seen as the solution to complex societal issues (such as population level obesity). It is notable that despite the low overall risks, the authors for this study concluded that if their findings are confirmed, they may provide “rationale for weight control of would-be mothers as a possible preventive measure against the risk of cancer in children“.
Expectant mothers with high BMI are already known to have an increased chance of a number of potentially more negative outcomes for their child, even if some of the raised risks remain relatively small. A side-effect of an emphasis on individual-level solutions (e.g. weight control), even where risks are low, may be that some groups of women are targets for extensive risk messaging about a range of conditions – common and rare – without having fully explored the impact of multiple risk-messages on recipients. The WRISK project seeks to understand how risks can be effectively and respectfully communicated, how to present clear information about absolute chances, complexities and uncertainties and ways to take account of limits to individual-level control of the risk context.
WRISK Survey of experiences
More than 7,000 women have now completed the WRISK project survey about experiences of risk communication in pregnancy. To share your experience, click here, there’s still time before the survey closes on 31 July.
If you are a bigger Mum or Mum-to-Be who is looking for information and support, please get in touch with Big Birthas.