About 25-years ago Rita and I published an article titled, “Efficacy of antidepressant medication with depressed youth: what psychologists should know.” Although the article targeted psychologists and was published in the journal, Professional Psychology, the content was relevant to all mental health professionals as well as anyone who works closely with children.
Yesterday, when teaching my research class to a fantastic group of Master’s students in the Department of Counseling at UM, I had a moment of reminiscence. Not surprisingly, along with the reminiscence, came a resurgence of emotion and passion. I was sharing about how it’s possible to find an area of interest that hooks so much passion, that you might end up tracking down, literally everything ever published on that topic (as long as the topic is small enough!).
The motivation behind my interest in the efficacy of antidepressants with youth came about because of a confluence of factors. First, I was working with youth every day, many of whom were prescribed antidepressant medications. Second, I was in a sort of professional limbo—working in full-time private practice—but wishing to be in academia. Third, out of virtual nowhere, in 1994, Bob Deaton, a professor of social work at the University of Montana, asked Rita and I to do an all-day presentation for the Montana Chapter of the National Association of Social Workers. Bob’s offer was not to be refused, and I’ve been in Bob Deaton’s debt ever since. If you’re out there reading this, thanks again Bob, for your confidence and the opportunity.
To prep, Rita and I split up the content. One of my tasks was to dive into all things related to antidepressant medications. Before embarking on the journey into the literature, I expected there would be modest evidence supporting the efficacy of antidepressants in treating depression in youth.
My expectations were completely wrong. Much to my shock, I discovered that not only was there not much “out there,” but the prevailing research was riddled with methodological problems and, bottom line, there had NEVER been a published study indicating that antidepressants were more effective in treating depression in youth than placebo. I was gob smacked.
Just to give you a taste, here’s the abstract:
Pharmacologic treatments for mental or emotional disorders are becoming increasingly popular, especially in managed care environments. Consequently, psychologists must remain cognizant of medication efficacy concerning specific mental disorders. This article reviews all double-blind, placebo- controlled efficacy trials of tricyclic antidepressants (TCAs) with depressed youth that were published in 1985-1994. Also, all group-treatment studies of depressed youth using fluoxetine, a serotonin-specific reuptake inhibitor (SSRI), are summarized. Results indicate that neither TCAs nor SSRIs have demonstrated greater efficacy than placebo in alleviating depressive symptoms in children and adolescents, despite the use of research strategies designed to give antidepressants an advantage over placebo. The implications of these findings for research and practice are discussed.
Early in my research class this semester, an astute young woman asked about the “rule” she had heard about that you shouldn’t cite research that’s more than 10-years-old. It was a great question. I hope I responded rationally, but my apoplectic-ness may have showed in my complexion and words. In my view, we cannot and should not ignore past research. As Samuel Clemens once wrote, “History doesn’t repeat itself, it only rhymes.” If we don’t know the old stuff, we may miss out on the contemporary rhyming pattern. In our article, 25-years-old now, we also discussed some medication research reporting shenanigans (although we used more professional language. Here’s an excerpt of our discussion about drop-out rates.
Dropout rates. Side effects and adverse events can significantly affect medication study outcomes by causing participants to discontinue medication treatment. For example, in the IMI [imipramine] study with children ( Puig-Antich et al., I987), 4 out of 20 (20%) of the medication group did not complete the study, whereas in the two DMI [desipramine] studies ( Boulos et al., l99 l; Kutcher et al., 1994 ), 6 out of 18 (33%) and 9 out of 30 (30%) medication participants dropped out because of side effects. For each of these studies, participants who dropped out of the treatment groups before completing the treatment protocol were eliminated from data analyses. The elimination of dropout participants from data analyses produced inappropriately inflated treatment-response rates. For example, although Puig-Antich et al. (1987) reported a treatment-response rate of 56% (9 of 16 participants), if all participants are included within the data analyses, the adjusted or intent-to-treat response rate is 45% (9/20). For the three studies that reported the number of medication protocol participants who dropped out of the study, the average reduction in response rate was 16.5%. Overall, intentto-treat response rates ranged from less than 8% to 45% (see Table 2 for intent-to-treat response rates for all reviewed TCA studies).
What’s the value, you might wonder, of looking back 25-years at the methodology and outcomes related to tricyclic antidepressant medication use? You may disagree, but I think the rhyming pattern within antidepressant medication research for youth (and adults) remains. If you’re interested in expanding your historical knowledge about this rhyming, I’ve linked the article here.
Research can be boring; it can be opaque; it can be riddled with stats and numbers. Nevertheless, for me, research remains exciting, both as a source of amazing knowledge, but also as something to read with a critical eye.
John Sommers-Flanagan 