To Preprint or Not to Preprint? An Evolving Landscape for the Dissemination of Health Economics and Outcomes Research
Benjamin G. Cohen, PhD, MPH, Monument Analytics, Baltimore, MD, USA; R. Brett McQueen, PhD, Center for Pharmaceutical Outcomes Research, Department of Clinical Pharmacy, University of Colorado Skaggs School of Pharmacy and Pharmaceutical Sciences, Aurora, CO, USA, Monument Analytics, Baltimore, MD, USA; and William V. Padula, PhD, Leonard D. Schaeffer Center for Health Policy & Economics, Department of Pharmaceutical & Health Economics, University of Southern California School of Pharmacy, Los Angeles, CA, USA, Monument Analytics, Baltimore, MD, USA
Editor’s Note: This article takes a closer look at the subject of preprints as presented in the feature article earlier in this issue.
The days of journal subscriptions are all but over. The 21st Century has dawned a renaissance in peer-reviewed journal readership, as researchers contemplate multiple options to publish. One can publish in a traditional journal submission process where readers or institutions pay journal license fees to read or subscribe to the publication. Another option is to publish in open access literature, where the researcher or their sponsor actually pay the journal to make the article openly available to the public. The third, more recent option is to post findings on a “preprint” space so readers can access the content before it has even been peer reviewed. This third option presents a number of noteworthy pros and cons to consider that are particularly important to the health economics and outcomes research (HEOR) community as the demand for timely findings becomes more immediate.
Take the current state of the COVID-19 pandemic. The need to share findings immediately has never been more important, as COVID-19 research findings of the early outbreak in Asia may have benefited billions of susceptible individuals by limiting their exposure to infection in Europe and North America. With few exceptions, the peer review process could not facilitate access to research quickly enough to make such findings accessible as infection spread exponentially. Expedited publishing through preprint media facilitated early access of scientific findings, even if not fully vetted by external peer review.1
"Preprints have, and can, provide value to the field through increased visibility and speed, but one must be cognizant of the challenges associated with preprints when both publishing and reading these articles."
Preprints increase the speed of access to science, allowing stakeholders to interpret data sooner for expedited decision making.2 While preprints eschew traditional assurances of scientific rigor, they do present the HEOR community with a belief that it has embraced through value of information: more data are always better, even if those data encapsulate additional uncertainty.
What the HEOR field has not been scrutinized for in the past, however, is related to a worldwide concern over the spread of misinformation, to which preprints are equally susceptible. An overall increase in information can amplify the noise in the public space and result in the spread of misinformation, especially when left open to interpretation by individuals without technical training or experience who would not normally access peer-reviewed journals with HEOR-specific content, for instance. Peer-reviewed journal articles reliably reach audiences with the appropriate technical expertise to process and interpret the methods and results. Ceding control of the information flow allows for a quicker advance of ideas, yet issues arise when findings are misunderstood.
Along with these concerns, there are a number of other pros and cons of preprint access to HEOR studies that may be more apparent as a result of the COVID-19 pandemic. We provide a brief overview of the following pros and cons of studies in preprint in the Table.
Table. Pros and cons of preprint publication of HEOR findings with respect to the peer-review process.
Model Transparency and Advancement
Preprint articles and open-sourced economic models allow for increased transparency and scrutiny from a wider variety of readers.4 Authors are also able to receive feedback to strengthen the economic model results as they approach the peer review process. Furthermore, printing models and other economic findings only in peer-reviewed literature creates a variability in understanding due to space constraints. Despite the ability to publish appendices and additional information online, it is challenging to include the complete description of a model’s functionality in a journal article. In turn, this limits the interpretation, advancement, and replicability of the model.
Admittedly, publishing through peer review shows comprehension of the theory and application of HEOR methods. HEOR experts conduct highly technical research that requires significant training to understand and appreciate. Increased transparency can allow other HEOR experts to validate and advance available literature, but that does not necessitate preprints.
Publicly available and transparent economic models can be manipulated by HEOR experts to include locally relevant or new data. Users can then generate more individualized and up-to-date results. Advancements in this manner, nevertheless, require the updates to be made by good actors and users who have the skills and abilities to work with the models. Inappropriate use, intentional or not, could lead to spurious, misinterpreted, or misapplied results. Therefore, peer-reviewed journals may learn from health technology assessment agencies (eg, ICER, IVI, NICE, etc) about their experiences with open source and/or model sharing programs.
Looking at Reach and Timeliness
While journals are less likely to publish articles that have already reached a broad audience through preprint servers, preprints can reach a larger audience that may not have journal access—not just laymen, but HEOR experts in low- and middle-income countries with limited institutional subscriptions. However, expanding an article’s reach alone does not expand the pool of individuals who can interpret the results. It is undoubtedly important for HEOR to reach clinicians, health system leadership, and decision makers. Yet, simply increasing the publication’s visibility does not translate into more decision makers accessing a study, such as in preprint form.5 If anything, decision makers would likely embrace the peer-review process so that their own decisions are based on greater certainty. Assurances should be taken to help individuals without HEOR expertise to understand the findings in the appropriate context.
Preprints also provide a timeline advantage for both authors and audiences through quicker access to the public domain. When done appropriately, the expedited timeline and publicly available information allows for greater scientific advancement. The peer-review process, while slower, provides readers with the confidence that the information they are reading is truly science. Without peer review, readers are left to determine the quality of the work on their own. This is unlikely to be a problem for experts in HEOR but is more complicated when applied to the wider readership.
Public Education About Economic Models and the Cost of Accessibility
Lay audiences likely do not have the necessary expertise to understand complicated results produced by HEOR studies (eg, cost-effectiveness analysis or econometric modeling). Allowing for these economic models to reach wider audiences could spur interest in HEOR, but the public currently may not have the ability to understand the nuances of the work.
Preprints are usually accessible to anybody with an internet connection, thereby bypassing the delays and charges associated with the existing peer-review process. While this may seem like a win-win, the journals that act as gatekeepers to high-quality HEOR studies could lose financial support in the long run as preprint popularity increases. Placing peer-reviewed journals in dire financial straits could be a consequence to HEOR that is not worth the risk, considering a dissemination system without checks and balances could flood the HEOR mainstream with misinformation.
Conclusion
The demand for HEOR evidence on a shorter timeline from study conception to reporting results is growing internationally. When acute health consequences such as the COVID-19 pandemic presents a small window of opportunity to share information in order to save lives, preprints offer an alternative to the long haul of peer-review processing. This eagerness to publish results quickly aligns with the attitude of HEOR experts to contribute to improving healthcare’s accessibility and quality at lower costs. That said, more immediate information does not always equal better quality information. HEOR dissemination in preprint comes with a clear tradeoff that it is not yet vetted.
As a result, some preprints simply amplify the noise that would not pass through the peer-review process. The potential for increased misinformation and confusion cannot be understated in the current environment. Preprints have, and can, provide value to the field through increased visibility and speed, but one must be cognizant of the challenges associated with preprints when both publishing and reading these articles. •
References
1. Else H. How a torrent of COVID science changed research publishing—in seven charts. Nature. Published December 17, 2020. https://www.nature.com/articles/d41586-020-03564-y. Accessed March 21, 2021.
2. Schlossberg JA. Correcting COVID-19 vaccine misinformation. UCLA Health. Published December 9, 2020. https://connect.uclahealth.org/2020/12/09/correcting-covid-19-vaccine-misinformation/. Accessed March 21, 2021.
3. Majumder MS, Mandl KD. Early in the epidemic: impact of preprints on global discourse about COVID-19 transmissibility. Lancet Glob Health. 2020;8:(5):e627-e630.
4. Dunlop WCN, Mason N, Kenworthy J, Akehurst RL. Benefits, challenges and potential strategies of open source health economic models. Pharmacoeconomics. 2017;35(1):125-128.
5. Padula WV, McQueen RB, Pronovost PJ. Finding resolution for the responsible transparency of economic models in health and medicine. Med Care. 2017;55(11):915-917.
