University of Southampton, MRC EEU, Highfield, Southampton SO17 1BJ, UK.
The presence of atherosclerotic lesions in the coronary arteries of young soldiers killed in the Korean War was greeted with some surprise.1 Surely, the age of onset of clinically apparent disease, and the emerging findings of epidemiological studies such as Framingham2 pointed to middle age as the time when coronary heart disease (CHD) was determined and therefore could be prevented. In his review,3 Beaglehole sets out the arguments for the pre-eminence of the traditional adult risk factors for CHDhigh serum cholesterol, high blood pressure, and smoking. Studies of men and women from middle age onwards show that these risk factors are associated with a large proportion of CHD, and that, in many countries, much of the population can be classified as being at high risk.
But estimating the importance of these risk factors by calculating their attributable risk fraction distracts from how mature men and women come to have high levels of risk factors in the first place. Consideration of the development of cardiovascular risk over the lifetime of individuals, sometimes called the life-course approach, offers a different perspective. During life, there are periods when we are more or less sensitive to different events, and our responses to these events alter the developmental pathways that we follow.4 Rather than looking for new risk factors, many life-course researchers are trying to understand the biological, social, and behavioural pathways that lead to CHD and its risk factors, and the sensitive periods when they might be influenced.57 This is a broader approach than simply adding up risk that occurs throughout life, which does not match well with biological theory8 or human experience.9
Each of the traditional adult risk factors is linked to early life. Serum cholesterol and blood pressure track (persist in rank order) from childhood into adult life,10 and children whose parents smoke are more likely to become smokers themselves.11 The post-mortem observations on young soldiers dying in Korea stimulated other necropsy studies. These showed that the extent of both fatty streaks and raised lesions (fibrous plaques and other advanced atherosclerotic lesions) was greater in young adults who in life had had higher levels of blood pressure and non-high density lipoprotein (HDL) cholesterol, and who had been smokers.12
Research linking early life with adult CHD suffers from inevitable difficulties in establishing causation and, often, in loss to follow-up, and the long lag times make direct testing of interventions on disease outcomes impractical. In technical terms, evidence on the relationships of CHD with blood pressure, serum cholesterol and smoking from cohort studies or clinical trials in middle-aged or older people is almost bound to appear superior. But ignoring the evidence that part of CHD risk is established in early life leads to inequity, denying children and young adults the opportunity for primary prevention. It devalues their lives because their deaths are distant.
Few would disagree with Beaglehole that we need to use the evidence we already have better. However, primary prevention of CHD in children and young people should be considered carefully and cautiously. Indicators of early health that predict CHD may or may not be part of the causal pathway. For example, though accelerated weight gain in childhood is associated with higher CHD risk measured as adult blood pressure13 or obesity,7 and with CHD itself,14 it is not clear whether the association is due to the accumulation of excess fat, the rate of change of weight, factors related to weight gain (for instance the components of diet, or metabolic correlates of physical inactivity), or other mechanisms. Some argue that until mechanisms are understood then policy recommendations are dangerous.15 Certainly, well-intentioned interventions have harmed their recipients in the past. The placing of infants to sleep in the prone position turned out to increase, not decrease, the risk of sudden infant death, though at the time a rational mechanism of protection was advanced.16 But understanding of the mechanisms which operate in early life to set up CHD risk may be impossible given the long lag periods and the changing pattern of early exposures across successive birth cohorts. Intervention studies may still be ethical and feasible if preceded by a rigorous assessment of the theoretical and empirical evidence as would occur for a new medicine.17
The views of children, young people and their parents are an important component of the evidence. The United Nations Convention on the Rights of the Child, to which all but two of the worlds nations are signatories, upholds the right of the child to express an opinion on matters affecting him or her, and to have that opinion heard. Children are unlikely to regard childhood as a training ground for a particular type of adult life required to meet public health mortality targets, though they deserve to be asked if this is true. The recent fall in rates of immunization against measles, mumps and rubella in the UK as a result of parental uncertainty about the safety of the triple vaccine is demonstration, if it were needed, that parents views cannot be forced to coincide with those of public health experts.18
A positivist model of science assumes that research evidence can and does influence public policy. The move towards evidence-based policy embraces such a model. Weiss described an alternative enlightenment model, in which research is one of several knowledge sources, all of which create concern and set an agenda.19 In discussing the application of this model to health policy, Black concludes that researchers (as well as policymakers and research funders) need to change if they are to be more influential. They need to understand that there are many sorts of evidence, (and) that sensible decisions may not reflect scientific rationality... They must also resist simplistic payback models, recognise the difficulty of identifying and quantifying the contribution of research, and be prepared to defend the unmeasurable.20 Such change may be uncomfortable for epidemiologists who value the methodological purity and quantitative power of their discipline. But responding to these challenges may allow us to make a greater impact on public health than searching for the missing percentage of the attributable risk fraction.
References
1 Enos WF, Holmes RH, Beyer J. Coronary disease among United States soldiers killed in action in Korea. JAMA 1953;152:109093.[ISI]
2 Kannel WB. The Framingham Study: 50-year legacy and future promise. J Atheroscl Thromb 2000;6:6066.
3 Beaglehole R, Magnus P. The search for new risk factors for coronary heart disease: occupational therapy for epidemiologists? Int J Epidemiol 2002;31:117722.
4 Kuh D, Ben-Shlomo Y (eds). A Life Course Approach to Chronic Disease. Oxford: Oxford University Press, 1997.
5 Barker DJP, Forsen T, Uutela A, Osmond C, Eriksson JG. Size at birth and resilience to the effects of poor living conditions in adult life: longitudinal study. BMJ 2001;323:127377.
6 Blane D, Berney L, Davey Smith G, Gunnell DJ, Holland P. Reconstructing the life course: health during early old age in a follow-up study based on the Boyd Orr cohort. Public Health 1999; 113:11724.[CrossRef][ISI][Medline]
7 Parsons TJ, Power C, Manor O. Fetal and early life growth and body mass index from birth to early adulthood in 1958 British cohort. BMJ 2001;323:133135.
8 Bateson P. Fetal experience and good adult design. Int J Epidemiol 2001;30:92834.
9 Independent Inquiry into Inequalities in Health. Report of the Independent Inquiry into Inequalities in Health. London: The Stationery Office, 1998.
10 Labarthe D, Eissen M, Varas C. Childhood precursors of high blood pressure and elevated cholesterol. Ann Rev Public Health 1991; 12:51941.[CrossRef][ISI][Medline]
11 Jarvis L. Teenage Smoking Attitudes in 1996. A Survey of the Smoking Behaviour, Knowledge, and Attitudes of 11 to 15 Year Olds in England. London: The Stationery Office, 1997.
12 McGill HC, McMahan CA, Herderick EE, Malcolm GT, Tracy RE, Strong JP. Origin of atherosclerosis in childhood and adolescence. Am J Clin Nutr 2000;72(Suppl.):1307S15S.
13 Law CM, Shiell AW, Newsome CA et al. Fetal, infant and childhood growth and adult blood pressure: a longitudinal study from birth to 22 years. Circulation 2002;105:108892.
14 Eriksson JG, Forsen T, Tuomilehto J, Osmond C, Barker DJP. Early growth and coronary heart disease in later life: longitudinal study. BMJ 2001;322:94953.
15 Gillman MW. Epidemiological challenges in studying the fetal origins of adult chronic disease. Int J Epidemiol 2002;31:29499.
16 McKee M, Fulop N, Bouvier P et al. Preventing sudden deathsthe slow diffusion of an idea. Health Policy 1996;37:11735.[CrossRef][ISI][Medline]
17 Campbell M, Fitzpatrick R, Haines A et al. Framework for design and evaluation of complex interventions to improve health. BMJ 2000; 321:69496.
18 Kmietowicz Z. Government launches intensive media campaign on MMR. BMJ 2002;324:38383.
19 Weiss CH. Research for policys sake: the enlightenment function of social research. Policy Analysis 1977;3:53147.[ISI]
20 Black N. Evidence based policy: proceed with care. BMJ 2001;323: 27579.