Causation

The debate surrounding the ‘fetal origin hypothesis’ focuses primarily on whether or not there exist sufficiently confounding factors to undermine the conclusion that there is a relationship between fetal growth/nourishment (measured through birth weight) and several mid-life diseases (including coronary heart disease, hypertension, diabetes, and depression). However, even if it can be established that such a relationship exists, there will be no possibility of preemptive prenatal treatment unless the relationship between birth weight and subsequent disease is determined to be a //causal// relationship. Further, this type of relationship is much more difficult to establish. Researchers must not merely eliminate confounding variables, they must demonstrate that their efforts will actually yield results – that is, they must show that //if// fetal undernourishment is corrected, //then// there will be a corresponding drop-off in subsequent disease.

This type of relationship (one expressible as an ‘//if//…//then//’ statement) is often referred to as a //counter-factual conditional//. The ‘counter-factual’ simply means that the antecedent condition (the ‘//if// ’ clause) doesn’t //in fact// have to obtain in order for the relationship to be true. Causal relationships have been defined primarily in terms of counter-factual conditionals such as these. That is, ‘x causes y’ is interpreted as: if x occurs, then y occurs, //and// if x doesn’t occur, then y will not occur.

While this terse definition may serve the interests of physics and most everyday instances of causation, it may yet be worried that the broad, abstract causal relationships with which epidemiology concerns itself are far too comprehensive, with far too many individual, potentially mitigating differences between patients to be adequately defined in deterministic language. Additionally, various In fact, Parascandola and Weed,[1] following the lead of philosophers of science such as Marta Bunge, have recently argued in the Journal of Epidemiology and Community Health that the strict, deterministic causal definitions of Hume are ill-suited to characterizing health causes and effects in large populaces. They argue, instead, for a probabilistic account of causation, whereby certain trends in society at large can be used to assign a //chance// of recovery or a //chance// of relapse to an individual patient.

Such a definition would substantially ease the burden of proof for the proponent of prenatal-treatment. Additionally, it would allow policy makers to gauge the potential return of any investment in treatment.

Nevertheless, as it is unlikely that there already exist data sufficiently reliable to establish either sort of causal relationship, a forty to sixty year study would be necessary in order to decide whether and to what extent preemptive prenatal treatment could successfully curb the rate of coronary heart disease, hypertension, diabetes, depression, etc.

From a policy perspective, however, the potential benefits of such preemptive prenatal treatment are advantageous enough to warrant the adoption of preventative action before such a causal relationship can be conclusively established. In the absence of any indicators to the contrary, then, a strong correlative relationship would be sufficient to justify at least some minimal preemptory action.

A good way to non-experimentally evaluate the hypothesis that fetal development and subsequent disease are causally related is to consider what possible contravening factors could undermine such a relationship – that is, to search for a root cause which could be causally responsible for //both// fetal development //and// subsequent susceptibility to disease. If such a root cause exists, then prenatal intervention would not yield the desired effect on adult health.

Adult disease Fetal Development Adult Disease vs. Fetal Development Root Cause

A likely root cause in this case is that of genetics. That is, genetics may be responsible for //both// fetal development //and// adult disease.

If those diseases which correlate with fetal origins all have genetic causes, then this would undermine the claim that fetal origins cause these diseases in later life, and thus challenges the corresponding prospect that prenatal intervention has the potential to effect health changes later in life.

In fact, this seems to be the case, in varying degrees, for each of the putatively correlated diseases. In a study on the probability of death from coronary heart disease published in The New England Journal of Medicine, it was concluded that “death from coronary heart disease is influenced by genetic factors in both women and men”[2]

In another twin study, researches found that “genetic factors may play an important role in the association between birth weight and blood pressure.”[3]

Similarly, “[t]he concordance rate for developing IDDM [diabetes] between monozygotic twins approaches 50%, suggesting that genetic factors are necessary”.[4]

And, according to the National Institute of Mental Health, “major depression is thought to be 40-70 percent heritable”[5]

These data strongly indicate that genetic factors may be responsible for both fetal development and later-life disease. These indicators must be borne in mind when considering the possible allocation of funds for research and development, and may warrant restraint pending further research.

[1] Parascandola, M. and Weed, D.L. “Causation in epidemiology.” __Journal of__ Epidemiology and Community Health. Dec 2001; 55: 905 - 912. Online at: http://jech.bmjjournals.com/cgi/content/full/55/12/905#B73 [2] Marenberg, Marjorie E., et al. “Genetic Susceptibility to Death from Coronary Heart Disease in a Study of Twins.” __The New England Journal of Medicine__. Volume 330:1041-1046, April 14, 1994, Number 15. Online at: http://content.nejm.org/cgi/content/abstract/330/15/1041?ck=nck [3] Richard G. IJzerman, Coen D. A. Stehouwer, and Dorret I. Boomsma “Evidence for Genetic Factors Explaining the Birth Weight–Blood Pressure Relation : Analysis in Twins” __Hypertension__, Dec 2000; 36: 1008 - 1012. Online at: http://hyper.ahajournals.org/cgi/content/abstract/hypertensionaha;36/6/1008 [4] Yoon JW. “The role of viruses and environmental factors in the induction of diabetes.” __Current Topics Microbiological Immunology__. 1990;164:95-123. Online at: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=2073786&dopt=Citation [5] National Institute of Mental Health. “Press Release: Mutant Gene Linked to Treatment-Resistant Depression.” http://www.nimh.nih.gov/press/tryptophangene.cfm

(ss)