How Does Depression Start?

For all the vast research we have on depression, with millions spent on genotyping, brain scans, and intervention studies, I have discovered a massive and most embarrassing hole in our knowledge.

Usually when we want to understand a serious problem, say, an airline accident, we start at the beginning. An investigator will want to inspect maintenance records and analyze data from the flight recorder. He or she will establish a detailed timeline. A critical mark on this timeline is the first sign of something unusual. Often the eventual outcome --the plane slamming into a mountainside-- comes at the end of a chain of unfortunate events (i.e., the weather is bad, the pilots are unfamiliar with new traffic control rules, the tower miscommunicates at a critical moment in the flight). To understand what happened, one needs to go to the beginning of the chain.

It seems reasonable to apply the methods of accident investigation to depression. How does it start? What are its warning signs and first symptoms? A specialized term from medicine to denote the idea of first symptoms is a prodrome.

It turns out that we know almost nothing about the depressive prodrome. The amount of research is paltry, and nearly all of it is retrospective, meaning people who have already become depressed are asked to reflect on what they can remember about what was going wrong before they got depressed. This is a poor method for the task at hand because human memory lacks the fidelity of a flight data recorder. Memory is a fallible record at best and depression only makes it more so!

I can dramatize how little we know about the depressive prodrome: Estimates of how long early symptoms typically last before becoming full-scale depression vary from 6 weeks to 23 months. In other words, the uncertainty about how long the prodrome lasts spans more than a factor of 10. Tremendous uncertainty!

To understand the prodrome we need prospective studies that collect comprehensive data on symptoms BEFORE the emergence of a full-scale depression. My recent search found a grand total of one such study, published in the year 2010. This study, by Iacoviello and colleagues, first collected data on 29 different kinds of symptoms that could serve as depressive warning signs and then followed people over time to see who become fully depressed. The critical comparison was between 60 people who would go on to have a full depression and 60 people who would not become depressed. There were 7 warning signs that distinguished the groups and were most diagnostic of an impending depression: depressed mood, decreased interest in or pleasure from activities, decreased concentration, hopelessness, worrying/ brooding, decreased self esteem, and irritability. What is most interesting about this study is that none of the somatic or physical symptoms, things like sleep problems, weight loss, or fatigue, heralded the coming of a full depression. It was the mood symptoms -- problems in what the person is feeling -- that were king!.

This study offers promising leads for future work. We face an epidemic of depression and we need to get to the bottom of it fast. My hunch is that the reason we have so much deep depression is that we have more low grade depression than ever before, with millions of people making the transition from prodrome to syndrome. We need a good map of how the prodrome becomes the syndrome.

Historically, treatments for depression have focused on the syndrome--what should be done once the horse is out of the barn. And now there's a host of therapies to help people once they become acutely depressed. Recently, there has been greater emphasis on trying to keep the horse in the barn, ideally, to stop an emerging episode before it gets up and running. But we cannot prevent depression intelligently unless we understand the nature of the prodrome --the warning signs that the horse is going to make a break for it. And we can't understand the nature of the prodrome unless we have real research.

-----------------------------------------------------------------------------------------------------------------------------------

Follow Jonathan Rottenberg on Facebook