Depressing Comparisons

An August post on a sociology blog began, “For the last several decades, depression rates have been on the rise at a rapid pace.”

Source: Andrew Mason via flickr

That assertion has appeared in many places over recent years. The blogger provided no reference for the assertion. I think I know the initial source of the claim; most writers who declare that depression has been rising probably read it in – of course – The New York Times.

Research indicates, however, that there was no rise in depression rates over the last several decades. The key studies relied on for the claim that depression rose had one or more important  flaws. Understanding those flaws helps us understand the difficulties of discovering and making historical claims.

Sounding the Alarm

In 1992, a large, international group of medical researchers reported data suggesting that “ more recent birth cohorts are at increased risk for major depression.” In several surveys conducted around the world, they explained, younger and older respondents were about equally likely to report having experienced episodes of distress some time during their lives. However, because younger respondents had been “at risk” for depression for far fewer years than the older ones — they simply hadn’t lived as long — that must mean that their underlying lifetime propensity for depression must have beeen much higher than the propensity of people born earlier. For example, in 1980s data, a much higher percentage of 35-year-olds than of 75-year-olds recalled ever being seriously depressed. Ergo, the researchers reasoned, incidents of depression must have increased over the century: every birth cohort must have had more episodes of major depression and experienced them earlier in life than had members of the cohort born before them. (This conclusion, by the way, implied that Europeans who lived through World War II were less likely to have ever been depressed than born afterwards.) A later article estimated that Americans born around the late 1970s had three times the chances of ever having a major depressive disorder in their lifetimes than those born around the early 1950s.

The 1992 report led to a story in The New York Times on “the dawn of the Age of Melancholy.” The phrase “epidemic of depression” became common.

Cautions

Some critics of the “epidemic” (for example, these) doubt the accuracy and validity of the questions surveys use to measure depression and some are skeptical of even clinical diagnoses. Even if the questions are appropriate, however, assuming that people’s recall of depressive episodes over their lifetimes reveals actual historical trends has been questioned, and probably discredited.

One problem with the procedure is the way memory works. The older people are and the longer ago an upsetting experience, the more likely they are to forget it.  In one study, for example, most of the people who had been hospitalized for major depression could not, 25 years later, recall the experience accurately enough to have it diagnosed as depression at all. (Also, when people remember such traumas, they tend to remember the episodes as happening more recently than they actually did – a systematic memory error called “telescoping.”) Thus, older respondents in a survey tend to seriously under-report episodes of depression – or other traumas – compared to the younger people. And that makes it falsely look as if those born longer ago had fewer episodes at a particular age than did those born more recently – leading to the questionable conclusion that rates of depression have increased over the generations.

Wikimedia Commons:Bundesarchiv

Another problem with the original studies is that Americans — particularly younger ones — have become more informed about psychological issues, more self-aware, and more forthcoming about depression. (See studies such as this and this.) Consider the increasing millions who have taken Psychology 101; consider the increasing tens of millions who have watched television psychologists like Joyce Brothers and Dr. Phil; and consider the social movements that urge people with psychological issues to come forward. (For example, in May, the “PostpartumProgress.com” web site publicized an L.A. County mental health brochure entitled “Speak Up When You’re Down.”)

The sorts of studies that avoid most of the distortions from look-back surveys are ones that compare samples of people interviewed in different years. There are few of these and they cover only the last half-century at best. Still, they suggest no net change in rates of depression; people, in say, the mid-1970s were about as likely to display symptoms as people in the mid-1950s. (For examples, see here and here.) The fairest conclusion is that any substantive change in depression  over the last roughly 50 years has yet to be demonstrated.

(More details and more references on these points can be found at p. 343,n.89 in Made in America.)

Perils of Looking Back

Aside from the specific lessons this account has for understanding trends – or non-trends – in mental health, it has implications for how we generally understand the past. Simply looking back, asking ourselves or older people about how things used to be way back when, yields fascinating glimpses of an earlier time. But we see these glimpses through a lens – a telescope, we might say – that is often distorting. Memory loss and memory tricks, nostalgia, embarrassment, tall-tale telling, and other biases reshape, even “photo-shop” the picture of the past.

UPDATE (January 25, 2012):

Analyzing a high-quality, longitudinal survey of American adults (here), Phillipa Clarke and colleagues find a consistent trend: Taking into account respondents’ ages, the later in the 20th century the respondents had been born, the less likely they were to be depressed. This was due, only in part, to recent generations’ higher levels of education.