Internet-based mental health interventions have existed nearly as long as the internet itself, but have exploded in popularity since the COVID-19 pandemic. These interventions involve the delivery of self- or therapist-guided evidence based therapies over the internet and can offer increased accessibility to some service users.
During COVID-19, regulations around online treatment of mental health issues were loosened, allowing many new, sometimes opportunistic companies to enter this space. Some of these platforms have exposed users to harm through misleading ads, irresponsible prescribing leading to drug shortages, and questionable practices around privacy and confidentiality giving data brokers greater access to private service user information. Experts have also questioned the quality of therapies delivered by these internet-based platforms.
While conflicts of interest are well documented in medical and mental health research, few studies have examined industry funding of internet-based mental health intervention clinical trials or payments to researchers investigating these platforms. A new meta-analysis published in Psychotherapy Research reveals that internet-based mental healthcare is likely a new stage for industry’s old conflict of interest play.
The current work, led by Thomas Munder and Berit Barthelmes from the University of Zurich, finds that industry sponsorship (InS), researcher financial conflicts of interest (RFCOI), and researcher allegiance (RA) to therapy modalities are common in clinical trials of internet-based interventions (IBI) for depression. Both industry funding and researcher financial conflicts of interest were linked to more positive findings for self-guided online interventions for depression. There was no relationship between positive findings for specific IBI and researcher allegiance in the current work. This relationship was also not significant in therapist-guided online interventions.
The authors write:
“Our study stresses the importance of paying attention to conflicts of interest in research on IBI. Our results suggest that conflicts of interest are frequent and that financial conflicts of interest, in the form of InS and RFCOI, are associated with larger efficacy estimates in trials of self-guided IBI.”
Methodology
The goal of the current work was to examine the prevalence and effects of industry sponsorship, RFCOI, and researcher allegiance on findings in IBI trails for depression. The authors decided the best way to achieve this goal was to collect and analyze data from other studies that had examined IBIs for depression.
To be included in the current research, studies had to compare online based interventions for depression to a group that was either receiving treatment as usual or had been wait-listed for treatment. Additionally, these platforms had to use psychological mechanisms to treat depression and be published in 2000 or later. Studies were excluded if they required their participants to have diagnoses or social problems (like being unemployed) in addition to depression.
InS was only examined in commercial IBIs and marked as present when a trial was funded or co-funded by the company marketing the IBI or when the company provided platform access for free to the researchers investigating the IBI. RFCOI was marked as present when at least one author was an employee of the company marketing the IBI, received payments or other compensation from the interested company, or owned stocks in the company. RA was marked as present when at least one author was involved in developing the IBI or the psychotherapeutic model used by the IBI.
The main outcome measure was severity of depression symptoms at the conclusion of the IBI. This was measured in terms of a standard mean difference (SMD). This measure allowed the researchers to compare differences between studies relying on varied measures of depression. It expresses how far apart the averages from two different groups are, with positive numbers indicating greater symptom reduction.
Seventy-one clinical trials were included in the current meta-analysis. These trials included 94 comparisons of groups receiving an IBI for depression to a control group. In total, these studies examined data from 17,001 service users, 9,462 assigned to an IBI group, and 7,539 assigned to a control group.
Results
Overall, the studies reported a SMD of 0.51, indicating a moderate reduction of depressive symptoms after an IBI. This number was higher for therapist-guided (0.64) compared to self-guided (0.38) IBIs, meaning therapist-guided platforms showed reduced symptoms compared to self-guided platforms. InS, RFCOI, and RA were common in the examined clinical trials.
Thirty-four IBIs (36.17%) were commercial, 24 were non-commercial (25.53%), and commerciality could not be determined in 36 (38.30%). InS was present in 13 commercial IBIs (38.24%) and undetermined in 20 (58.82%). Only one commercial IBI clinical trial was confirmed to not involve industry funding or free access to an industry produced IBI. Eight (23.53%) commercial IBI trials were funded at least in part by industry. Seven (20.59%) received free access to the IBI platform.
RFCOI were present in 29 trials (30.85%), absent in 45 (47.87%), and undetermined in 20 (21.28%). Sixteen of 29 (55.17%) instances of RFCOI involved the first or senior authors. Authors of eight clinical trials (27.59%) received payments for presentations or consultations from the company marketing the IBI they were investigating. Authors of eight clinical trials (27.59%) owned stocks in the company. Five clinical trial authors (17.24%) were employees of the company marketing the investigated IBI, and eight authors had multiple RFCOI (27.59%).
RA was present in 75 (79.79%) investigated IBI clinical trials, absent in 10 (10.64%), and undetermined in 9 (9.57%). Fifty-five of 75 (73.33%) cases of RA involved the first or senior clinical trial author.
The presence of industry InS was associated with a 0.28 SMD increase in the measures of depressive symptoms after participation in an IBI. The presence of RFCOI was linked to a 0.20 SMD increase, “both representing substantial effects in the context of randomized trials.” In other words, industry funded clinical trials, and those where authors were employed, given industry payments, or owned industry stock, showed more favorable results for the IBI under investigation. The authors note that this link was not as clear or as strong for therapist-guided IBIs.
RA was not significantly related to depression score outcomes, but the authors caution that this could be due to their study’s focus on controlled trials rather than comparative studies. Past research has found stronger evidence for RA affecting outcomes in the latter.
Limitations
The included studies used different measures of depression which reduces the certainty of the findings. Inconsistencies in reporting financial conflicts of interest, missing data, and author omission are common in clinical trials. This could mean the findings of the effects of RFCOI on clinical trial outcomes was underestimated in the current work. The study’s design limited the authors ability to detect associations between RA and IBI trial outcomes. A larger sample could have given greater validity to some of the findings, especially regarding InS which was undetermined in a significant number of included trials. The study was not pre-registered. This means the work is more susceptible to bias and there is no way to determine if the authors switched outcome measures. This also limits replicability for future research.
The authors conclude:
“Our results suggest that current meta-analytic estimates of the effects of self-guided IBI are too optimistic, and that the role of self-guided IBI as a stand-alone treatment for depression might need to be reconsidered because it is unclear whether it can bring about clinically relevant change.”
Industry Corruption of Healthcare
Industry corruption of healthcare has been well documented at every level. Industry backed clinical trials produce favorable findings for industry products. Payments directly to physicians and psychiatrists increase prescription rates and lead to greater service user dissatisfaction. Industry employs a system of “ghost management” to determine what research gets funded and published, including ghost-writing research to publish in the names of complicit researchers. They pay researchers and doctors directly to act as “key opinion leaders.” These opinion leaders promote industry interests through biased medical education seminars, contributing industry propaganda to textbooks, and providing industry-promoting commentaries and editorials to academic journals.
****
Munder, T., Barthelmes, B., Weber, L., Flückiger, C., Kerber, A., Krieger, T., Watzke, B., & Wolf, M. (2025). Conflict of interest in trials of internet-based interventions for depression: Systematic review and meta-regression. Psychotherapy Research, 1–11. (Link)
link
