COVID-19 Show message

Skip to content

Post by Former NIMH Director Thomas Insel: What Caused This to Happen? – Part 2

By on

We are getting into the season of scientific meetings.  April, May, and June are prime time for national conferences and a good time to take stock of progress and continuing challenges in mental health research. One of the concepts that is likely to be debated across these meetings is the notion of “cause.”  The purported causes of psychopathology range from early trauma to the microbiome, from genes to prenatal exposures. My blog earlier this year, entitled “What caused this to happen?” was inspired by a study of cancer that attributed most cancers to bad luck.  Mental disorders are usually attributed to an ill-defined mixture of genetics and environment, but the nature of cause itself deserves careful consideration.

The cognitive neuroscientist Jonathan Roiser from University College London has just written a very thoughtful essay on cause with the provocative, Monty Pythonish title “What has neuroscience ever done for us?” 1 Roiser argues for two kinds of cause. Proximal cause is the mechanism of disease. In cancer, a proximal cause is a mutation in a tumor suppressor gene that leads to unregulated cell replication. In AIDS, the proximal cause is infection with HIV. Distal cause, by contrast, is about risk. For cancer, a distal cause might be smoking; for AIDS, this might be risky sexual behavior. If proximal causes are mechanisms, distal causes influence probabilities.

These two versions of cause are frequently conflated when we think about mental disorders. The proximal causes of depression or schizophrenia are almost certainly neural events that lead to altered regulation of mood or thinking. While we have not identified these specific neural events, science is closing in on the circuits and molecular pathways involved. This search is critical because, based on experience with cancer and AIDS, defining these pathways will lead to new, targeted treatments. Proximal causes can be the basis of precision medicine.

The distal causes are quite different. Early childhood adversity or stress during adulthood are distal causes that increase the probability for many forms of psychopathology. Genome-wide associations are, like family history, almost certainly distal causes that increase risk for mental disorders but may not be tied directly or simplistically to the mechanism of disease. The distinction between distal and proximal cause is important because distal causes rarely point to treatment targets. Instead, like smoking and risky sexual behavior, they point to prevention targets. And rather than predisposing someone to a specific disease, they increase risk for multiple negative outcomes. This point is frequently misunderstood, whether by the psychotherapist focusing on early childhood adversity or the genomic scientist focusing on the recent tsunami of common variants. Distal causes are not really causes in the traditional sense; they are risk factors. Risk factors are vital for public health interventions and for understanding population-level patterns of disease. But risk factors may not point to treatment targets and they may not be actionable for an individual who has already developed the disease.

This season of scientific meetings will almost certainly be even more exciting than last year. There are spectacular findings emerging at every level of our science, from molecular biology to social factors. As NIH Director Francis Collins recently noted, in the past decade the cost of sequencing a human genome has dropped from $22M to $2K, and from requiring 2 years to less than 24 hours. In this same decade, the tools for mapping brain structure and function and for manipulating neural circuits have transformed how we study the neural basis of behavior. The social sciences are leveraging the power of smart phone technology to yield new insights into social networks. With this traction, we can expect spectacular new findings to be reported over the next few weeks. When the notion of cause arises, best to consider precisely what cause means and its value for either treatment or prevention.


1 Roiser J. What has neuroscience ever done for us? The Psychologist. 2015 April;28(4)284-287.