Undisclosed financial conflicts of interest in DSM-5-TR: cross sectional analysis

Comparison with other studies

Congruent with our previous studies,78 we found that panel members of the Diagnostic and Statistical Manual of Mental Disorders who received the most remuneration from drug companies were those working in diagnostic areas where drug interventions are often the standard treatment, such as depressive disorders, neurocognitive disorders, and drug induced movement disorders (fig 4). Notably, more than one third of the panel members received “compensation for services other than consulting, including serving as faculty or as a speaker at a venue other than a continuing education program.” (see supplemental figure 1) This category captures what the pharmaceutical industry refers to as key opinion leaders—“physicians who influence their peers’ medical practice, including but not limited to prescribing behavior.”28 Being on a speakers bureau or being a key opinion leader is widely recognized as an egregious financial conflict of interest because the role of the key opinion leader is essentially a marketing one; the talks given are usually presented at educational events sponsored by industry.29 As one author pointed out, key opinion leaders are hired by industry because, as the terms suggests, they can lead (or change) opinions.30 Moreover, key opinion leaders are influential not only because they are often affiliated with prestigious universities, but also because industry provides them with wide ranging and influential platforms and speaking engagements.31

Policy implications

The greatest proportion of compensation (by category of payment) was research funding or research payments (≥70% across all years) (see fig 2). Research funding was excluded from the American Psychiatric Association’s disclosure policy for DSM-5, and some individuals might believe that such compensation does not influence behavior. Nonetheless, evidence is lacking to suggest that simply because money comes in the form of a research grant it does not create an obligation to reciprocate or invoke an implicit bias. Similarly, although some panel members received <$1000 in remuneration, relatively minor gifts do affect physician behavior and prescribing practices; empirical research shows that even small gifts can have a substantial impact on behavior.323334

Additionally, it is important to emphasize that the problem of overdiagnosis is not limited to the inclusion of new disorders in the Diagnostic and Statistical Manual of Mental Disorders. Even seemingly small changes to the manual (eg, to symptomatology of previously included disorders) can have a substantial impact on increasing the number of people who would receive a diagnosis and increasing the number of people prescribed drugs. For example, many researchers and clinicians, including the former chair of the DSM-IV, pointed out that the seemingly small changes to the criteria for attention deficit/hyperactivity disorder (ADHD) in the 2013 edition of the DSM-5 would likely result in a considerable increase in the diagnosis of ADHD and an increase in the prescriptions of stimulants (eg, the DSM-5 made the change that symptoms could appear before age 12 years rather than age 7 years; that symptoms need only to “impact” behavior rather than cause impairment). Interestingly, authors of a recent systematic scoping review of more than 300 studies in children and adolescents found “convincing evidence that ADHD is overdiagnosed in children and adolescents.”35

Finally, panel members also have the authority to eliminate disorders—not just add new ones—and make changes that would help prevent overdiagnosis. In other words, if the developers of the next edition of the Diagnostic and Statistical Manual of Mental Disorders were all free of ties to industry and were a genuinely multidisciplinary group (which would be congruent with the Institute of Medicine’s 2011 recommendation for developing trustworthy clinical guidelines), the manual may be less likely to lead to overdiagnosis and overtreatment.

Strengths and limitations of this study

A major strength of this study is that it provides novel data about the appreciable conflicts of interest in the DSM-5-TR and extends past research on this topic.78 The present study, however, only used information provided by Open Payments, which does not include payments to physicians based outside the US. Although mandatory disclosure of payments for US pharmaceutical and medical device companies that make products covered by Medicare and Medicaid is required, until recently the database did not include payments to non-physicians. (Under the 2021 SUPPORT ACT, Open Payments expanded its database to include non-physician prescribers such as nurse practitioners and physician assistants).13 Of the 168 total panel members (after duplicates were removed), 92 met the inclusion criteria, and thus it is possible that the 76 individuals who did not meet the inclusion criteria were free of commercial ties. Additionally, as other researchers have noted, the amounts listed in the database can change slightly, and although there is a validation process, it can be cumbersome and there “remain ways in which [the data] may be imprecise or inaccurate.”36

Conclusion

To ensure unbiased, evidence based mental health practice, it is important to prohibit industry influence in the Diagnostic and Statistical Manual of Mental Disorders. The extent of conflicts of interest among these panel and task force members reveals how important it is to examine the systemic and institutional practices that allow for these conflicts and that reinforce them as normative.78 Professionalism and the belief in scientific objectivity cannot protect against implicit bias because, as the social psychological literature shows, individuals are unable to assess how conflicts of interest influence decision making.37 Moreover, disclosure may not only make conflicts of interest normative, but may have the iatrogenic effect of worsening bias as empirical research shows.3839

Also, as previously noted, the American Psychiatric Association does not publicly disclose minutes of the meetings of the Diagnostic and Statistical Manual of Mental Disorders, nor does it provide descriptive summaries of proposed changes and reasons for inclusion or exclusion of the proposed changes. As one prominent psychiatric researcher noted, the revision process for the DSM-5 “suffered from lack of an adequate public record of the rationale for changes, thus shortchanging future scholarship . . . DSM-5 was a missed opportunity to increase the conceptual validity of psychiatric diagnosis by aggressively addressing false-positive issues.”40 We believe that the DSM-5-TR also represents this missed opportunity, and we call upon the American Psychiatric Association to ensure greater rigor in, and transparency of, the process for revising the manual. Doing so will help mitigate overdiagnosis and overtreatment.

More than a decade ago, the year before the DSM-5 was published, we made the following recommendations: Individuals who have participated on pharmaceutical companies’ speakers bureaus should be prohibited from being panel members of the Diagnostic and Statistical Manual of Mental Disorders, and there should be a rebuttable presumption of prohibiting conflicts of interest among such panel members. When no independent individuals with the requisite expertise are available, individuals with associations to industry could consult to the panels, but they would not have decision making authority on revisions or inclusion of new disorders.7 Given our present findings, these recommendations are even more pressing 10 years later. As researchers, clinicians, policy makers, and leaders in evidence-based medicine have argued,41 guideline writers should be free of financial relationships with industry, especially those writers who are responsible for such an influential manual on psychiatric taxonomy.