Skip to main content

Join the Resistance to ‘Fake News’ in Behavioral Healthcare

August 28, 2019

They may not be coming for your guns, but are they coming for your mental health certification? We are being told that “crazy people” are driving our mass murder rate in America. We are not getting a more refined picture of what constitutes “crazy” just yet, but someone will surely volunteer for such an important job soon.

My sarcastic tone does not reflect a lack of genuine concern. After years of industry leaders arguing that behavioral healthcare is being ignored and underfunded, misguided gun violence solutions would be an unwanted correction. The public is confused about mental illness generally and about the relationship between mental illness and violence specifically.

We should be on guard against a process whereby strong political convictions search for the data to support them. Yet there is a broader concern. We seem to be losing the very basis for evidence-based discussions in this brave new world of fake news.

I am taking fake news as the rubric for many current phenomena. Fake news can be as simple as boldly asserting a view of reality with no support or justification, but the essential quality that I am concerned about is anti-empiricism. As a politician from an earlier generation, Tip O’Neil, once said of political debates, you are entitled to your own opinions, but not your own facts. 

Fake news might infect political debate today, but will it be scientific debate tomorrow? All areas of scientific knowledge can be subjected to the corrosive effects of anti-empiricism. Behavioral healthcare could become a prime target due to long-standing public confusion and stigma. However, every scientific discipline can prepare for attacks on its knowledge base by being scrupulous in rating the certainty of its evidence.

What is known in behavioral healthcare?

This is a serious question and a good one for any scientific discipline to ask periodically. The remainder of this discussion will grapple with the legitimacy of this question rather than bemoaning the nefarious intentions driving many discussions of fake news. The goal is to inoculate the field against periodic distortions from any source about its knowledge base.

Too many people in our field believe we have addressed all concerns about our scientific knowledge by claiming we are an evidence-based field or that we embrace empirically validated clinical models in behavioral healthcare. This might be described as the error of taking your press clippings too seriously. Or to be less cavalier, it is mistaking marketing claims for being rigorous data-driven arguments.

In fact, many of the claims we promote as validated have variable levels of evidence to support them. Let’s start with the question of how much evidence is enough to support a claim as validated. If we accept the FDA as the gold standard because they approve the medications we take, then two randomized controlled clinical studies with positive results are needed to gain approval.

This standard seems like a strong starting point until it is pointed out that the FDA does not withhold its approval despite the existence of numerous studies with negative results. Get ready for a challenge on empirical validation when people file lawsuits alleging results ranging from no benefit to significant harm from approved medications.

What should we expect when challenges are mounted regarding the choice of a particular model of psychotherapy? The claim by many psychotherapists that they feel personally comfortable with one approach over another will not be greeted by hard-nosed empiricists as reassuring. The claim by others that they selected one approach over others based on research will fall apart when Bruce Wampold’s two comprehensive research volumes are uncovered showing that no clinical model is superior to another.

Practices without a body of supporting empirical research will be readily attacked by skeptics. While many people have been helped enormously by 12-step programs, they have no well-designed research studies supporting them. How many treatment programs around the country provide no rigorous studies of their clinical outcomes? It would be a shame to have fake news skeptics expose this deficit.

Comfort in truth telling

This litany of vulnerabilities about the scientific basis for behavioral healthcare should not discourage serious researchers and clinicians. The feared attack on our value proposition as merely fake news can be resisted. But the path to resistance depends more on the humility of the statistician than the hubris of the marketer.

The statistician starts with the acceptance that even well-studied interventions may ultimately prove to have little value (apropos the methodological limitations listed after every research study), while research may one day validate practices that today have only anecdotal support. Don’t let the marketer oversell or undersell the solutions we have today, but rather feel comfort with the limitations of the data and analytics to date.

If a pharmacological agent amasses numerous randomized studies that contradict the studies upon which FDA approval was predicated, then we can proclaim with satisfaction that the scientific imperative for the continued study of well-established solutions is intact. Contradictory findings are part of the expected landscape for investigating clinical outcomes. We should also expect a multi-faceted research palette by promoting multiple methodologies, including everything from randomized controlled trials to retrospective data reviews, along with statistical methods that range from significance testing to meta-analysis.

Science-based or evidence-based

It is commonplace for leaders in behavioral healthcare to claim they take an evidence-based approach or utilize empirically validated clinical techniques. They may go on to recite a list of acronyms like CBT, DBT, ACT, and the like. This is the wrong tactic in preparation for the time when people descend upon the field with a goal of undermining its validity.

We should not get lost in the weeds of defending the data regarding various clinical models. There will be too many weaknesses in the analytics, too many people who received no benefit. Furthermore, we must be prepared for the contrary conclusions of researchers like Wampold that no clinical model has been shown to be superior to any other. A skeptic might then ask why we need so many models.

The answer is that we will keep putting forward clinical models that promise to yield strong clinical results, continuously subjecting those models to rigorous study, and carefully following the data. We should stake our claim on the scientific method, not on the existing state of evidence for any specific model or technique. When skeptics direct their focus on behavioral healthcare with the disparaging bias of finding fake news, our best response is to stress our science-based foundation, not the latest headlines from the field.

Ed Jones, PhD

Ed Jones, PhD, is senior vice president of the Institute for Health and Productivity Management.

Back to Top