My professional interest in long-term treatment outcomes started when someone I loved dearly began experiencing the anguish of profound depression and anxiety. We did what many families do—encouraging her to “get help.” She subsequently started a treatment designed to provide some immediate relief from the emotional pain.
It worked for a while … until it didn’t. A few good weeks or months, and she was right back in it. Up and down, year after year … with the lows becoming measurably lower over time.
What was going on? I started asking what many families today still wonder: Is it possible for someone grappling with difficult emotional burdens to find deeper, more sustainable healing? If so, what kind of a treatment and healing trajectory would lead there?
That question has motivated almost everything I’ve done in my career in mental health—up to and including this summary review.
Seeing the end from the beginning. Among the most salient of differences between secular society and people of faith is the length of their vision. Whereas religiously-committed communities talk almost incessantly about eternity, life trajectories, and “taking the long-view”—people unmoored from any such formal tie back to the Divine tend to focus most often on the opposite: the immediate moment, what can be known and felt and experienced right now.
The conflict between these emphases can be overstated, of course. It’s a good thing for any of us to be in the present moment and be able to experience it more fully. And I’ve relished opportunities as a mindfulness teacher to help people in a today’s accelerating society learn to draw their attention back from compulsive wanderings into the present or past, and practice resting in an awareness of everything going on right here and right now—in a way that enhances health in a variety of ways.
But there’s another way of “living in the moment” that isn’t so healthy or life giving—and which believers are rightly suspicious of—the kind of “enjoy this moment” and “do what feels right” mentality that almost invariably includes something like,“and stop worrying so much about future consequences.” Whatever thrill that myopic view might provide, few would advance this as a wise way to approach life and its many problems—especially the big ones.
Fear and relief in the age of COVID. Among other things, the pandemic has brought out human fears in a hundred and one flavors. From paralyzing worries about sickness and death, to the many rippling concerns about loneliness, strained relationships, aching losses of jobs and opportunities, and loss of freedoms—there seems to be no end to the fear around us. Across health care as a whole, it remains a rarity to have sustained and rigorous examination of long-term outcomes.
Across health care as a whole, it remains a rarity to have sustained and rigorous examination of long-term outcomes.
So no matter your larger view of the pandemic, it should be understandable to anyone why so many parents today are relieved to hear any news of additional protection for their children that may be available. According to a recent Kaiser Family Foundation survey of parents with children in the 5- to 11-year old age range, approximately one-third of parents want their child to get vaccinated “right away” when an approved vaccination becomes available. Another third of parents understandably want to hear more—especially about details regarding how long the children in this study had been observed and monitored.
In a highly anticipated announcement in late September, a Pfizer news release revealed that its vaccine had been “shown to be safe and effective in young children.” According to its internal findings, the company stated, “These trial results provide a strong foundation for seeking authorization of our vaccine.”
Although the details had not been peer-reviewed, published, or submitted to the FDA yet, news outlets widely celebrated the announcement—even predicting that the FDA may well “authorize a vaccine for younger children in a matter of weeks” on an emergency-use basis, and anticipating that millions of elementary school students may be able to receive shots by Halloween.
Within 24 hours, other news outlets were elaborating on the good news:
Pfizer itself hinted that other positive news for “children 2-5 years of age and children 6 months to 2 years of age [was] expected as soon as the fourth quarter of this year”—so even toddlers may be eligible by Christmas.
So far so good, right?
Safe and effective. A pivotal moment that sparked my own career’s interest in studying long-term pharmacological outcomes was a seminal 1997 study by Graham Emslie and colleagues on the efficacy of fluoxetine (Prozac®) in children. At the time, the trial was heralded as a “landmark study,” with the results widely disseminated in national media as exciting news confirming that “Prozac works for children.”
Over the next year following the release of that single study, America began to see a measurable increase in antidepressant prescriptions for children across the nation, with rates nearly doubling (49%) over the next 5 years. A few years later, another article announced the finding that paroxetine (Paxil®) was “generally well tolerated and effective for major depression in adolescents.” This study likewise became widely cited over the next decade as evidence that Paxil was “safe and effective” for young people—and ultimately leveraged by GlaxoSmithKline to promote the antidepressant for off-label use in teenagers.
Unfortunately, in the years following both announcements, the FDA began getting reports about unanticipated adverse effects for youth from these medications, including tragic child suicides. This prompted urgent reviews of both original studies. In each case, independent researchers learned that outcomes had been selectively reported—with data that contradicted a “safe and effective” conclusion de-emphasized, and not even included in the final analysis. Separate analyses giving equal weight to all scales reached very different conclusions about the safety and efficacy of antidepressant use in children—with one researcher concluding that, contrary to the published conclusions, the study data showed clear “harm” for children. In 2003, Britain’s Medicines and Healthcare products Regulatory Agency (MHRA) concluded that while there was no evidence of paroxetine’s efficacy in children and adolescents, for instance, there was “robust evidence” of a causal link between the drug and suicidal behavior.
What went wrong? And why were initial perceptions of thrilling results so far off from the longer-term realities? Entire books have been written trying to answer that question by respected journalists, researchers, and medical authorities (here and here and here).
But the bottom line is this: In our rush to embrace a potential new solution for threats to childhood health, we failed in the past to give sufficient due diligence to the kind of comprehensive and independent analysis needed to ensure these medical interventions had, in fact, been proven to be “safe and effective” over the long-term. More specifically, in our eagerness to celebrate possible short-term benefits, we neglected to require adequate investigation of the longer-term realities that take more than an 8-week controlled trial to confirm. It’s fair to insist on a heightened level of scrutiny when children are involved, given their unique vulnerabilities.
It’s fair to insist on a heightened level of scrutiny when children are involved, given their unique vulnerabilities.
How long is long enough? When any health or medical intervention is presented in marketing or news reports as “effective,” most Americans understandably assume the intervention has shown some kind of evidence documenting effectiveness over a sustained period of time. But historically speaking, as illustrated above, that has not always been the case.
For decades now, the norm across virtually all areas of healthcare—pharmacologically and therapeutically—has been to evaluate “effectiveness” through rigorous trials that are almost exclusively short-term. I’ve conducted one of these short-term randomized-controlled trials myself, for a mindfulness-based intervention for youth. And in a previous job, I studied outcomes for residential treatment for youth—both short and long-term. Although a handful of such residential treatment studies provide evidence for some enduring effects on participating youth, legitimate critiques have been raised that most such studies “fail to measure outcome after discharge”—with another research team noting that “it is remarkable … “there is very little evidence on long-term outcomes.”
So, that’s what I did. With the encouragement of a wonderful youth non-profit, I called up teenagers and their parents years after residential treatment to find out the actual results of their treatment years later. Many of these kids had graduated with triumphant celebrations that the treatment had “worked” and they were “better.” But as hundreds of conversations quickly revealed, the longer-term picture was quite different.
One thing was painfully clear: short-term effectiveness was simply not the same thing as was long-term, sustainable effectiveness. Evidence of the first is not the same as evidence of the second. And yet, we’re paying remarkably little attention to that very difference—with most people assuming short-term efficacy must mean that everything will continue along fine in the long-term. But that’s not always the case.
Similar concerns have continued to be raised in pharmacological studies as well, with new medications routinely approved on the basis of evidence gathered from studies lasting 8 to 12 weeks. After noting a 7.6-week average trial length for the ADHD drug, Strattera® (atomoxetine HCl), a respected researcher remarked: “One must note that although Strattera was approved, and is marketed and promoted, to treat a ‘chronic’ condition, the only studies reviewed by the FDA were short-term studies.” After acknowledging that more children are being treated at even younger ages, two developmental neuroscientists noted similarly that “surprisingly, despite the obvious need for such information, the long-term effects of therapeutic drug exposure on an immature brain have not been adequately assessed at either the clinical or preclinical stage.”
These kinds of limitations prompted my interest, along with that of Florida State University professor Jeffrey Lacasse, in publishing a peer-reviewed review of the many issues involved, in a 2011 article entitled, “What Does It Mean for an Intervention to ‘Work’?”—with a focus on child and adolescent health interventions. Over the subsequent decade and a half, I’ve continued to explore in a variety of ways what constitutes more lasting, sustainable healing for depression and anxiety.
Improving accountability for long-term studies. Along the way, many of us wondered what could be done to promote and encourage more long-term research as a matter of policy? In 2005, in pursuit of a higher standard of evidence for drugs used over the long term, the U.S. Food and Drug Administration (FDA) implemented a new requirement that drug companies submit longer-term efficacy data as part of the drug approval process. Within a few months, however, the Psychopharmacologic Drugs Advisory Committee (made up of many representatives of the industry) voted 12–0 for a resolution instructing the agency to reverse its decision after 10 pharmaceutical companies submitted testimony that this action would “slow drug development and timely approval of new medications for the treatment of mental illness.”
Decisions such as this have kept the number of longer-term studies for pharmacological intervention of any kind limited. In the absence of more extensive research, it is subsequently the short-term studies that continue to be promoted among the general public as primary evidence of effectiveness. That’s largely what we base our decisions upon. This led Dr. David Cohen to caution that conventional medication studies of “initial, ‘selective’ action” can potentially distract and draw attention away from the more extensive picture of complex, rippling drug effects, including “subsequent cascades of transient and long-lasting neurochemical changes involving other neurotransmitters.”
In the absence of more serious and systematic longer-term research, what can legitimately be said about a health intervention’s true, lasting effects?
In my own view, the only honest answer is …. not much.
We certainly cannot say a great deal about potential adverse effects that may show up over time. For instance, Dr. David Jacobs observed among his own research participants in a psychiatric trial that “it took time for patients to realize” that psychological side effects associated with medication had become problematic—specifically, more time than was allowed in “the six-week time frame of RCTs conducted for FDA approval.” In light of this kind of limitation, Dr. David Cohen and his team pointed out an alarming “inability of conventional clinical trials to provide a true picture of a drug’s … [full] effects.”
Despite all this, when Dr. Lacasse and I published our summary review on outcomes of child and adolescent treatment ten years ago, we were feeling optimistic, writing:
Fortunately, there appears to be some progress in this area as well. Although the practical meaning of “long-term” research naturally varies across settings, there is a growing realization that for a youth treatment or intervention program to be called “effective,” it should be able to demonstrate those effects in more than the immediate beginning or ending of a formal intervention.
Noting a groundswell of interest in long-term research, we cited recommendations at the time that researchers work harder to document any measurable shift observed over at least a 6-to 12-month period after a treatment as a basic standard to guide decisions. Since the misused antidepressant study mentioned earlier had prompted more accountability for drug companies to independently register clinical trials, we were also feeling encouraged that greater transparency might well occur across pharmaceutical studies as a whole. Maybe we were witnessing signs of fundamental improvements in the legitimacy and integrity of medical research?
Not so fast.
Signals from recent opioid and suicide epidemics. Looking back, Americans in 2021 are well aware of at least two major signals that these institutional shifts towards greater transparency with pharmacological outcomes did not materialize as much as we had hoped. Amidst hundreds of thousands of deaths in the opioid epidemic, the country was forced to wake up to the extent to which similar kinds of selective reporting, overstating of positive possibilities, and de-emphasis of adverse effects were continuing to happen in pharmacological studies. A flurry of lawsuits (including from Utah counties) alleged that opioid manufacturers had “trivialized or obscured [opioids’] serious risks,” and convinced doctors and patients “that the compassionate treatment of pain required opioids.”
Despite being aware of reasons for concern, it’s clear now that opioid manufacturers pursued an aggressive marketing campaign with claims that far outpaced what their own highly promoted data had confirmed. As a Utah County Commission resolution put it, despite “knowing of the serious risks and adverse outcomes related to the use of their opioids,” companies “nevertheless set out in the 1990s and 2000s to persuade providers, regulators, and patients that opioids are safe and effective.”
Just one month ago, it was announced that Johnson & Johnson and three large U.S. drug distributors would proceed with a proposed $26 billion settlement—on top of the 10 billion dollar settlement from Purdue Pharma—eclipsing the previous record of health care fraud.
Rather than a sad anomaly, some of these same patterns have been evident elsewhere in healthcare too. In the wake of this heart-breaking opioid fallout, I joined a number of other colleagues in authoring an op-ed warning that “This same minimization of risks has been plainly documented in industry marketing for antidepressants”—pointing to some statistics that many believe illustrate consequences that rival the long-term consequences of opioid deception:
Between the early 1990s and 2008, antidepressant use in teens and adults quadrupled— with rates climbing even further in the past 10 years. In this same period, numbers of suicide and chronic disability for mental illness reached record levels—with 600,000 youth and 4 million adults on disability for mental illness today (compared with 16,200 and 1.25 million in 1987).
Of course, there are many other factors that have influenced spiking suicide rates. But as journalist Bob Whitaker summarized in great depth, the long-term evidence across available psychiatric studies calls for greater scrutiny of concerning effects these medications may elicit at year 5 or 10 or 15 (far beyond what the FDA approved them for). Indeed, compared with the widespread claims that antidepressants are “safe and effective,” every single one of the 22 studies documenting long-term effects of antidepressants that I’ve gathered suggests a far more complex picture—with those on antidepressants long-term statistically more likely to be depressed, in comparison to those who never went on. Compared to early figures demonstrating the strong initial protection provided by COVID-19 vaccination for many, the longer-term picture is much less clear.
Compared to early figures demonstrating the strong initial protection provided by COVID-19 vaccination for many, the longer-term picture is much less clear.
This is not a need limited to mental health or pain management. Across health care as a whole, it remains a rarity to have sustained and rigorous examination of long-term outcomes.
Long-term trajectories of COVID-19 vaccination. I don’t pretend to be a vaccination expert—but I do believe these historic patterns in pharmacological research offer at least a few insights worth considering, especially as we contemplate offering these vaccinations to children on an emergency use basis.
It’s fair to insist on a heightened level of scrutiny when children are involved, given their unique vulnerabilities. In the case of antidepressants, for instance, medications that are more tolerated for adults end up having counterintuitive effects on developing brains (with a doubling of suicidality documented among teens prescribed antidepressants, for example).
Vaccinations, of course, are unique from other medical interventions. And over the last year, COVID-19 vaccinations have been hailed as a modern-day miracle, with the initial differences in rates of infection between the unvaccinated and vaccinated have been dramatic. Soon after the second shot has been received by a plurality of people in a country, there is a period of time in which an almost miraculous reduction in sickness and death among the vaccinated is witnessed, when compared with the unvaccinated.
It’s understandable that the data from this initial period would receive such welcome attention.
Yet if exclusive attention goes to that period of time—if most conclusions are based on it—we might be missing something important. Once again, until we follow the trajectory of outcomes beyond the initial weeks and months following any medical intervention, it’s hard to ascertain the full picture of what’s happening.
As time has gone on, clearly shifts continue to occur in vaccine outcomes for adults. For instance, it’s clear the potency of some immune protection has declined markedly, as was anticipated to some degree. And related to this (and the emerging Delta variant), rates of infection have grown.
Although there is no way to look into the future of the United States, the next best thing might be looking more closely at the experience of other countries that started their vaccination campaigns earlier than we did. The United Kingdom and Israel provide a helpful barometer since they were among the earliest who succeeded in vaccinating the largest percentage of their population. While those numbers won’t be reviewed here in detail, it’s worth noting that the initial happy period immediately after the second dosage has changed over time in measurable ways – with cases and deaths from COVID-19 growing again markedly. Updated figures in Germany, Scotland, and the United States are following suit.
Why exactly this resurgence of cases is occurring, of course, is a question eliciting many different answers. In addition to the Delta variant and declining immunity, most explanations for these spikes focus on the impact of the remaining unvaccinated. In the spirit of exploring all possible factors, it’s important to also not overlook the possibility of what Dr. Susanne Hodgson and colleagues referred to last year as “vaccine-associated enhanced respiratory disease” (VAERD). As I wrote about last year for this magazine, a number of experts early on pointed to this as the single greatest long-term concern —and the question for which good data would be hard to come by until many months had passed. Essentially, this refers to the concern that in some cases, vaccinated individuals (who were less susceptible initially to the virus) may become more susceptible over the long term to the same virus. Various other names that have been used for this phenomenon include “pathogenic priming,” “paradoxical immune enhancement” (PIE), and “antibody-dependent enhancement” (ADE).
God forbid this is happening long-term following our ongoing vaccination efforts. But given the growing incidence of cases, we need to keep watching and stay attentive—especially in the wake of the international data above.
It’s worth asking—if this problem was arising, would our current public discourse be able to detect it? That’s difficult to imagine for some of us, given the widespread censorship of dissenting experts.
At the very least, we might agree that compared to early figures demonstrating the strong initial protection provided by COVID-19 vaccination for many, the longer-term picture is much less clear.
Long-term trajectories of COVID-19 for children. What does this mean for children? Has sufficient youth-specific evidence of safety and effectiveness emerged to feel confident in bringing these injections to younger children?
There seems to be little doubt among journalists and health authorities alike that the vaccinations will shortly be judged as sufficiently safe and effective for children. Just today, the White House has unveiled a plan to get vaccinations out to 5-11 year old’s. And as the Washington Post reported a week ago, the U.S. government has: “purchased enough doses to give two coronavirus shots to all 28 million eligible children, ages 5 to 11”:
Within days of regulators clearing the nation’s first coronavirus vaccine for younger children, federal officials say they will begin pushing out as many as 20 million doses of the Pfizer-BioNTech pediatric vaccine to immunize school-age kids across the United States … If regulators and the CDC give the go-ahead, the shots could start being administered almost immediately.
The same article notes that “many questions remain unanswered only weeks from the possible launch”—for instance: “How do you get clinicians to want to participate in the program? What will the role of pharmacists be? What will the role of schools be? How will we help parents get their questions answered quickly and by whom?”
Notice that one of the critical questions is not whether the vaccinations will be safe and effective for children. That, we have been assured by the companies, is largely answered. Based on this single self-reported trial, we are told the vaccination was “well-tolerated” and with “side effects … generally comparable to those of people between the ages of 16 and 25 years old who received the vaccine.”
I’ve searched for a more in-depth analysis of the results but can find no other information yet (beyond technical elements such as 2268 kids enrolled, with 2/3rds administered 10 μg (micrograms) of vaccine 21 days apart and 1/3rd receiving placebo, and with follow-up planned over the next two years). One thing seems clear—we have not had a whole lot of time elapse to know much about the long-term trajectory for any child taking the COVID-19 vaccine.
Given the foregoing patterns, it seems fair to call for greater patience and deliberation about the questions involved—especially when considering vaccinating children, who have been historically far less prone to this virus. I also suspect that greater respect and consideration for those parents with questions would go a long way towards re-engaging the vaccine-hesitant in the thoughtful, good-faith conversation we need to have (together) about how best to move forward in navigating the continuing pandemic.