Then vs. Now: Changes in Biomedical Research Over 50 Years

When we began our careers as psychiatrists in the 1970s, the primary locus of psychiatric research (especially psychopharmacology) was in academic-affiliated “research units.” Most medical school psychiatry departments, major state and VA hospitals, and prestigious private institutions dedicated some inpatient beds and clinic space to research. Most research units specialized in specific psychiatric diagnoses: major depression, schizophrenia, bipolar disorder, OCD, borderline personality disorder, anxiety, or eating disorders. At major institutions, there was often an affiliated pre-clinical research program (usually animal research), which added a basic sciences component.

The vast majority of the research was “investigator-initiated.” The primary investigators (PIs) proposed the hypotheses, designed the methodology, managed the paperwork, collected, analyzed, and interpreted the data, and presented and published the findings. Funding usually came through competitive grants from federal and state governments or national and local foundations.

When I (FWP) started clinical research training at the NIMH, there was no formal human subjects or scientific oversight, other than review by my immediate boss. External checks involved the competitive funding process, and later, pre-publication peer-review of the findings. By the time I left 20 years later (1999), there was a highly structured, independent review of the ethical and scientific issues for all human subjects, human tissue, and animal research.

That regulatory structure and intensive scrutiny have only increased by another order or two of magnitude in the quarter century since. Indeed, as I talk with colleagues and former mentees, they tell me that the extraordinary difficulties in obtaining adequate grant funding and the enormous paperwork and administrative burdens of conducting investigator-initiated research have become overwhelming. Many (probably most) traditional academic psychiatric research units have closed. I gather that some of the same difficulties are occurring across other medical specialties.

A New Player in the Biomedical Research World

In 1991, about 80% of pharmaceutical research occurred in academic settings and was conducted by independent investigators. By 2004, the proportions were essentially reversed: independent investigations had fallen to 26% of drug research, with for-profit research organizations conducting the remaining 74%.¹

Most people have never heard of contract research organizations (CROs). CROs are for-profit companies that provide research support and services to drug manufacturers and biomedical technology companies. Initially narrowly focused on specific research tasks that PIs were either not qualified for or not interested in doing, such as biostatistics, patient recruitment, or regulatory compliance, CROs rapidly evolved to cover more and more tasks.

Now a pharmaceutical company can contract with a “full-service” CRO to conduct randomized clinical trials (RCTs) from start to finish. If the trials are successful, the CRO will handle the complicated FDA drug and medical device approval process and even do the commercial marketing. Pharmaceutical companies often trigger aggressive “bidding wars” among CROs when putting out contract proposals². While the contract payer may transfer many or even all of the tasks involved in running a research study to the CRO, in most countries the payer remains legally responsible for the quality and scientific integrity of the CRO’s work.

There are over 1,000 registered CROs around the world, including many in China and India. The top 10 CROs, however, control over 50% of the market, which is projected to reach about $90 billion annually in the next decade. CROs are largely responsible for the enormous jump (over 200-fold) in RCTs conducted in the last two decades, from 2100 in the year 2000 to 454,000 in 2023.

What Has Changed?

Virtually everything. The former academic and the current CRO contractual models of conducting clinical research could hardly be more different. We have gone from a scientific model of research transparency to a commercial model of proprietary information.

In the former, the PIs are highly involved with multiple aspects of the study and responsible for integrating the scientific, clinical, technical, human subjects, and administrative components. Investigators usually are members of the larger society in which they conduct their research. Research subjects are recruited by PIs in the name of the institutions with which they are affiliated. The PIs (and usually their institutions) “own” the data and can analyze and publish it without requiring the funder’s approval. Indeed over the years, data acquired with federal funding has increasingly been made available, after a suitable time period, to the larger scientific community for further analysis. The goal of funders, investigators, and academic institutions is, first and foremost, the advancement of scientific knowledge.

Psychopharmacology Essential Reads

Long-Term Use of Benzodiazepines and Dementia Risk

Does Antipsychotic Medication Work?

In the case of CROs, they may be located in another country and different culture from the research site. Indeed, the labor and tasks of a research project may be divided among multiple CROs located in several countries. The CRO and/or the funder owns the data, and no one else can even see the data without explicit permission, which is not easily obtained. Research subjects are frequently recruited from multiple private practices, but the referring clinicians are usually blinded to their patients’ experimental condition and never see their data or even the overall results, especially when these are negative. In particular, they cannot learn from their patient’s individual response, or even from the study’s overall outcome, how to improve clinical care. Referring clinicians are, however, usually “reimbursed for their time” often to the tune of tens of thousands of dollars for each patient who completes the study. The goal of the funder and CRO is to advance a potential commercial product.

What Does This Mean for the Future of Mental Health Care?

There are a number of important consequences associated with this profound shift in the locus and ethos of mental health and biomedical research, which are beyond the scope of this post (but may be topics for future posts).

Among the more important global consequences is the loss of independent PIs with their greater objectivity, comprehensive understanding of the research, and focus on knowledge over profit. Another often unrecognized consequence is that as a result of the Bayh-Dole Act, which allows academic institutions conducting research with federal funds to commercialize their discoveries, universities have become competitors in this commercial marketplace. University administrations routinely search their scientific portfolios looking for results that can be patented or monetized. Scientists can feel pressured to maximize commercial applications rather than address purely scientific questions. An associated proliferation of non-disclosure research agreements means that academic scientists are less free to share their ideas and preliminary results with colleagues. All of this works against the open advancement of knowledge, and may ultimately be contributing to the accelerating decline of the U.S. as a leader in the biomedical sciences.