Send to

Choose Destination
J Child Psychol Psychiatry. 2013 Aug;54(8):813-5. doi: 10.1111/jcpp.12122.

Editorial: from correlations to causation: the value of preventive interventions in studying pathogenic mechanisms in childhood psychiatric disorders.


The ultimate goal of all research in childhood psychiatric disorders is to provide knowledge that will be useful in treating or preventing illness. To be useful in treating or preventing illness, research studies must provide valid information about the causes of illness, because only by targeting those causes can we prevent or counter their effects in a truly rational way. Until now, the discovery of interventions has largely been by serendipity, yielding medications and behavioral interventions that produce too little therapeutic response, in too few people, and with too many side effects. The selection of which specific intervention to use for which child is based on trial-and-error guesswork and the personal preferences and idiosyncrasies of the treating clinician, with little or no empirical support, and at great cost to children, families, and health care systems. Guesswork in treatment will decline only when we have more detailed knowledge of the differing causal pathways that produce differing subtypes of the same disease phenotype. Identifying more homogeneous subtypes of disease that have more uniform natural histories and treatment responses will permit development of more genuinely rational and individualized interventions, or more truly personalized medicine (Insel, 2009). Identifying causal pathways in human disease, however, is especially challenging because the classic method of experimental investigation - isolating and manipulating at will a variable that is hypothesized to exert a causal influence on the illness in order to understand its effects on other variables - is not possible when studying the causes of human illness, for obvious ethical reasons. We are therefore left with a variety of methods for identifying probable causal pathways in human illness, all of which have varying strengths and limitations in the weight of evidence that they provide that a variable has a true causal influence on the disorder.

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Wiley
Loading ...
Support Center