Benton Bramwell, ND, and Matt Warnock

Not surprisingly, we commonly think of clinical trials as being conducted in clinics. There are several benefits for the researcher in this traditional approach, including having a central location that handles all subject screening and evaluation with consistency, thus limiting variation. However, there are also reasons to embrace a decentralized approach to clinical trials, where at least some or even all of the activities associated with the trial do not occur at a traditional clinical site. Chief among these is the convenience of the patient, a factor that impacts recruitment and retention, as we explore below. As we contemplate the increasing incidence of decentralized clinical trials (DCTs), a compelling question emerges as to how the shift in control away from institutions and researchers to patients, who will increasingly participate in a location of their choosing and collect much of the data themselves using provided equipment, may impact the amount of data obtained, as well as its cost and quality.

As we broach the terminology associated with this topic, we also wonder if the useful and broad term of DCT might be complemented (or even replaced) by other terms that more specifically describe how the information in a given DCT is collected. A term such as “entirely remote trial (ERT)” might be a more informative descriptor for a trial that conducts all its data collection online and through remote locations. In contrast, “partially remote trial (PRT)” might be helpful to describe a trial that collects some data through remote participation and other data through a clinical site(s). This review primarily explores the potential of decentralized clinical trials in which all trial activities are managed remotely and online.

DCTs and Subject Recruitment and Retention

One significant obstacle in the traditional clinical trial model is the lack of availability of suitable trials in the local area of eligible patients who are willing and able to take part. For example, an estimate from 2019 suggests that about half or more of patients with cancer (55.6%, CI 43.7-67.3%) do not have a clinical trial available to them at the institution where they receive treatment. In addition to a lack of geographically available clinical trials, subjects still must overcome obstacles of transportation to even relatively near trials, sacrifice their time, and deal with the weight of risks associated with study intervention, all when they are often not feeling well.

These and other obstacles make the collection of clinical trial data arduous and explain why so many trials fail to recruit to their target level or even terminate due to lack of recruitment and/or retention. An estimate indicates that in the NLM’s clinical trial repository, 19% of the registered phase 2 and phase 3 clinical trials marked as closed in 2011 were either completed with less than 85% of their target population or terminated early due to recruitment challenges For a longer-term perspective on this issue, we also point to data from 2006-2015, which showed that over that time period 11% of over 13,700 cardiovascular studies were terminated, with the leading cause (41% of the time) being termination from lack of recruitment.

Therefore, it’s not uncommon for inadequate subject recruitment to either limit the data available for analysis or to prevent the data collected from being fully useful. Thus, instances are not uncommon where inadequate subject recruitment either precludes the data ultimately available for analysis or limits the gathering of data to its most useful degree. This situation is problematic both for researchers seeking to answer vital questions and for subjects who participate in studies at various risk levels, only to find their efforts have limited usefulness. These issues in recruitment and retention present an ethical dilemma: Is it fair to ask subjects to accept study risks, conveying the belief that their participation will lead to meaningful results, even when there is a substantial chance that their involvement may not produce fully utilizable data?

For further context on recruitment and retention challenges, it is perhaps useful to step back and consider the related issue of the overall low level of participation from the pool of potentially eligible subjects. One study estimates that only about 5% of eligible subjects choose to participate in clinical trials. Thus, we are making inferences about the effect of an intervention after not examining its impact in 95% of a relevant population where many individual differences in response to a drug occur.

In some cases, perhaps this sampling and other structuring of the research is adequate to lead to reasonably accurate inferences of what to expect in a population. Perhaps in some circumstances it is not, such as when there is any impactful amount of self-selection bias present. Given the amount of contradictory evidence that emerges when there is an effort to replicate clinical trial results, some reservation about the overall strength of the current level of sampling seems warranted. Surely, if we can obtain larger and more informative samples through DCTs than traditional clinical trials, this should be welcome news to the research community.

Relatedly, the sample of subjects recruited for most clinical studies used to generate FDA approval of an intervention may not be representative of national demographics and thus may not provide insight into effects like pharmacogenetic impacts that vary by race. For example, worldwide trials, including US-based plus internationally based trials, used for FDA approval of drugs and biologics between 2015-2019 included Black or African Americans as 7% of study subjects, while the United States 2020 census indicates that Black or African Americans constitute 13.6% of the population. Interestingly, it appears that trials done for the sake of approval specifically within the United States are much better at representing Black or African American subjects than internationally based trials, with Black or African Americans representing 16% of study subjects in the US, perhaps reflecting domestic implementation of FDA guidance encouraging racial diversity in trial recruitment. However, in internationally based trials, only 2% of study subjects are Black or African American.

Conversely, where the rate of participation of Asian subjects is concerned, this population is under-represented in U.S. centered trials (constituting 2% of US trial subjects whereas Asians constitute 6.3% of the US population ). Yet, when we consider the impact of international-based trials on overall participant demographics, Asians represent 11% of total study subjects enrolled in worldwide trials used for FDA approval. Thus, achieving and maintaining racial inclusion at rates comparable to the US population, including for other groups not already mentioned, such as American Indian or Alaska Natives and Pacific Islanders, can be a complex problem, both with studies done for FDA approval within the United States and internationally.

Can decentralized clinical trials significantly help to increase recruitment and retention of subjects representing the diversity of the national population while not being undermined by unacceptable variability and poor precision? The full answer to that is yet to be known, but some interesting initial learnings are appearing in the literature.

A first learning is that subjects seem to like participating in clinical trials online. In an analysis of three remote trials with 706 subjects, researchers found that 97% of subjects reported overall satisfaction with the experience of their trial participation being through Internet visits. Moreover, when researchers offer remote trial subjects a choice of participation via telephone or through interaction via an internet portal, subjects overwhelmingly choose participation through the Internet portal. For example, In the ADAPTABLE trial where this choice was offered, 13,172 subjects opted for participation through the internet portal compared to 1,904 opting for participation via telephone calls through the call center.

A rare silver lining to the COVID pandemic is the stimulation of researchers to explore decentralized clinical trials to collect data. Estimates are that the recruitment rate into clinical trials decreased by about 74%, comparing May 2020 to May 2019. Naturally, researchers began exploring other avenues for getting work done. One of the more interesting studies published during this period is the DeTAP trial, a single-arm observational study of 100 subjects with atrial fibrillation who were recruited, screened, enrolled, and assessed completely remotely. This study demonstrates “pandemic proof recruitment and engagement,” an apt description given its results. Two weeks into the study, the research team introduced recruitment through social media ads. Direct outreach had recruited 6 subjects, but the team found social media recruitment far more effective, attracting 94 subjects in 12 days with several hundred on a waitlist. The 6-month study demonstrated high compliance: 91% of subjects completed their study televisits, 85% finished their surveys, and 90% accurately reported blood pressure and 6-lead ECG data using the provided instruments. The subjects submitted this data through a study-specific app. Additionally, over 80% of the participants expressed willingness to continue in clinical research post-study. However, the demographic makeup of the study—90% white, 3% black, 4% Asian, and primarily urban residents—highlights the need to still intentionally focus recruitment in DCTs to under-represented groups. Possible features of such outreach might be social media ads written with cultural sensitivity to underrepresented racial groups and, where needed, written in the first language of potential subjects.

The DeTAP trial is not alone in finding that subjects strongly prefer participation in research through a decentralized approach. Though smaller in size, a recent study by Sommer found that an option for participation through teleconsultation for screening and telemedicine visits (a decentralized arm) led to enrollment of 18 subjects, whereas recruitment via health clinics into an arm requiring an onsite visit enrolled only 5 subjects. Moreover, the recruitment through the decentralized approach successfully recruited from 10 of the Swiss Cantons (territorial units within Switzerland) and within a rural environment, while the conventional arm recruited from only 2 Cantons. This bodes well for the potential of DCTs with wide enough reach in advertising to recruit diversely not only with regard to geography, but also other demographics, such as economic status.

An additional study of patients with atopic dermatitis from Denmark that required eligible subjects to return a DNA sample prior to enrollment identified 164 potential subjects in 11 days of online recruitment. From the initial group of potential subjects, 65 individuals met all the inclusion criteria and were asked to provide a DNA sample before enrollment, a task completed by 55 subjects. Of these 55 enrolled subjects, 53 completed the study for a retention rate of 96%. This suggests that even with some logistical back and forth with eligible subjects before enrollment, it is still possible to recruit and retain subjects efficiently within the structure of a DCT.

How Does the Use of a Decentralized Approach Impact the Observation Effect?

In addition to overcoming obstacles that lead to poor recruitment and retention, decentralized clinical trials allow the researcher to collect a form of real-world data with potentially broad applicability. As we have explored in previous articles, one of the drawbacks of traditional clinical trials is that they produce data with internal, but not necessarily external validity. That is, the results describe what happened in the group studied, but the ability to translate those results to the real world is limited. Perhaps contributing to this conundrum is a phenomenon common to both quantum physics and clinical trials: the observation effect.

The presence of an observation effect means that results of an experiment are affected by the act of watching. In the case of patients, they behave differently when they are aware that they are participating in a clinical trial. Comment on this issue explains that an observation effect in clinical trial research is specifically called the Hawthorne effect, and that similar to the effect in quantum physics, a greater intensity of observation over time will lead to a greater observation effect.

At this point, we don’t know how much the observation effect impacts data collected in the real world when subjects actively collect data themselves and forward it to researchers, instead of data being collected by researchers at a clinical site. However, there is at least a happy possibility that subjects in a space of their choosing will feel less observed and thus have less propensity to change their behavior. Such a decrease in observation effect could, in turn, offset some increase in variability that may occur when research activities are removed from a more tightly controlled clinical site.

For right now, what seems safe is to simply recognize that the relatively new approach to clinical research represented by DCTs, with data collected in the less-controlled environment of the real world, will collect data with different variability than a more traditional, centralized clinical trial, likely due to a combination of factors. Moreover, if there are some changes in variability of the data collected, but the data obtained has greater applicability and can be obtained more easily, on the whole we may look back in some years and be satisfied with the change in course.

Closing Thoughts

As well as unknowns there are exciting possibilities associated with the promulgation of decentralized clinical trials. While the impacts of subjects collecting data outside of a highly controlled clinical environment are yet to be fully understood, DCTs present an opportunity for data collection that is more convenient, and thus more inviting, to potential subjects. Online recruitment through tools such as social media ads, which might increase participation in trials generally, allows for subjects to be recruited across greater geographical areas and may be tailored to facilitate participation from portions of the population underrepresented in clinical trials. There are indications at this early stage that subject retention in DCTs may also be quite strong and that many subjects are satisfied with their participation in DCTs. Further, DCTs allow for data to be collected within the real-world circumstances of a subject’s life.

In addition to the possible benefits above, we should recognize that fast-moving factors are shaping the world of clinical research. While highlighting the COVID pandemic as the most recent, conspicuous example, it is probably not the last event to make clinical research that relies heavily on onsite clinical visits more difficult. Moreover, although onsite visits are the only responsible or practical method for collecting certain data, other data can be effectively collected by sending devices or surveys directly to subjects and encouraging them to return results. A research paradigm that continues to rely rigidly or nearly exclusively on onsite data collection for information that can be reasonably collected remotely is unnecessarily vulnerable to interruption. Using a decentralized approach wherever possible not only meets the desire of the subject for more convenient participation, it allows research to continue even when onsite research centers must close for a time.

On the whole, despite the unknowns inherent in transitioning to a decentralized approach, it seems that now is a moment of opportunity to embrace and facilitate the advent of decentralized clinical research.

References

1. Unger JM, Vaidya R, Hershman DL, Minasian LM, Fleury ME. Systematic Review and Meta-Analysis of the Magnitude of Structural, Clinical, and Physician and Patient Barriers to Cancer Clinical Trial Participation. J Natl Cancer Inst. 2019 Mar 1;111(3):245-255. doi: 10.1093/jnci/djy221. PMID: 30856272; PMCID: PMC6410951.
2. Carlisle B, Kimmelman J, Ramsay T, MacKinnon N. Unsuccessful trial accrual and human subjects protections: an empirical analysis of recently closed trials. Clin Trials. 2015 Feb;12(1):77-83. doi: 10.1177/1740774514558307. Epub 2014 Dec 4. PMID: 25475878; PMCID: PMC4516407.
3. Baldi I, Lanera C, Berchialla P, Gregori D. Early termination of cardiovascular trials as a consequence of poor accrual: analysis of ClinicalTrials.gov 2006-2015. BMJ Open. 2017 Jun 15;7(6):e013482. doi: 10.1136/bmjopen-2016-013482. PMID: 28619765; PMCID: PMC5577897.
4. Smalley E. Clinical trials go virtual, big pharma dives in. Nat Biotechnol. 2018 Jul 6;36(7):561-562. doi: 10.1038/nbt0718-561. PMID: 29979664. Quotation attributed to Novartis Chief Digital Officer, Bertrand Bodson.
5. Ioannidis JP. Contradicted and initially stronger effects in highly cited clinical research. JAMA. 2005 Jul 13;294(2):218-28. doi: 10.1001/jama.294.2.218. PMID: 16014596.
6. 2015-2019. Drug Trials Snapshots. Summary Report. Five year summary and analysis of clinical trial participation and demographics. U.S. Food and Drug Administration. https://www.fda.gov/media/143592/download?attachment Accessed 08 Jan 2024
7. https://www.census.gov/quickfacts/fact/table/US/POP010220#POP010220 Accessed 08 Jan 2024
8. 2015-2019. Drug Trials Snapshots. Summary Report. Five year summary and analysis of clinical trial participation and demographics. U.S. Food and Drug Administration. https://www.fda.gov/media/143592/download?attachment Accessed 08 Jan 2024
9. Diversity plans to improve enrollment of participants from underrepresented racial and ethnic populations in clinical trials Guidance for Industry. U.S. Department of Health and Human Services. Food and Drug Administration. April 2022. https://www.fda.gov/media/157635/download Accessed 17 Jan 2023
10. 2015-2019. Drug Trials Snapshots. Summary Report. Five year summary and analysis of clinical trial participation and demographics. U.S. Food and Drug Administration. https://www.fda.gov/media/143592/download?attachment Accessed 08 Jan 2024
11. https://www.census.gov/quickfacts/fact/table/US/POP010220#POP010220 Accessed 08 Jan 2024
12. Myers TL, Augustine EF, Baloga E, Daeschler M, Cannon P, Rowbotham H, Chanoff E; 23andMe Research Team; Jensen-Roberts S, Soto J, Holloway RG, Marras C, Tanner CM, Dorsey ER, Schneider RB. Recruitment for Remote Decentralized Studies in Parkinson’s Disease. J Parkinsons Dis. 2022;12(1):371-380. doi: 10.3233/JPD-212935. PMID: 34744053; PMCID: PMC8842745.
13. Jones WS, Mulder H, Wruck LM, et al. ADAPTABLE Team. Comparative Effectiveness of Aspirin Dosing in Cardiovascular Disease. N Engl J Med. 2021 May 27;384(21):1981-1990. doi: 10.1056/NEJMoa2102137. Epub 2021 May 15. PMID: 33999548; PMCID: PMC9908069.
14. Sathian B, Asim M, Banerjee I, Pizarro AB, Roy B, van Teijlingen ER, do Nascimento IJB, Alhamad HK. Impact of COVID-19 on clinical trials and clinical research: A systematic review. Nepal J Epidemiol. 2020 Sep 30;10(3):878-887. doi: 10.3126/nje.v10i3.31622. PMID: 33042591; PMCID: PMC7538012.
15. Sarraju A, Seninger C, Parameswaran V, Petlura C, Bazouzi T, Josan K, Grewal U, Viethen T, Mundl H, Luithle J, Basobas L, Touros A, Senior MJT, De Lombaert K, Mahaffey KW, Turakhia MP, Dash R. Pandemic-proof recruitment and engagement in a fully decentralized trial in atrial fibrillation patients (DeTAP). NPJ Digit Med. 2022 Jun 28;5(1):80. doi: 10.1038/s41746-022-00622-9. PMID: 35764796; PMCID: PMC9240050.
16. Sommer C, Zuccolin D, Arnera V, Schmitz N, Adolfsson P, Colombo N, Gilg R, McDowell B. Building clinical trials around patients: Evaluation and comparison of decentralized and conventional site models in patients with low back pain. Contemp Clin Trials Commun. 2018 Jun 28;11:120-126. doi: 10.1016/j.conctc.2018.06.008. PMID: 30094387; PMCID: PMC6072894.
17. Dal-Ré R. Unperceived similarities between clinical trials and quantum physics. Br J Clin Pharmacol. 2020 Feb;86(2):192-193. doi: 10.1111/bcp.14126. Epub 2020 Jan 14. PMID: 31943305; PMCID: PMC7015734.
18. https://www.fda.gov/media/167696/download Accessed 21 December 2023
19. Darmawan I, Bakker C, Brockman TA, Patten CA, Eder M. The Role of Social Media in Enhancing Clinical Trial Recruitment: Scoping Review. J Med Internet Res. 2020 Oct 26;22(10):e22810. doi: 10.2196/22810. PMID: 33104015; PMCID: PMC7652693.

Published February 24, 2024

About the Authors

Benton Bramwell, ND, is a 2002 graduate of National College of Naturopathic Medicine who practiced primarily in Utah while helping to expand the prescriptive rights of naturopathic physicians in that state. Currently, he owns and operates Bramwell Partners, LLC, providing scientific and regulatory consulting services to both dietary supplement and conventional food companies. He and his wife, Nanette, have six children and two grandchildren; they live in Manti, Utah.

Matt Warnock is an accidental herbalist, who received his MBA and Juris Doctor from BYU, then worked as an attorney, litigator, and business consultant until 2000. He then joined RidgeCrest Herbals, a family business started by his father, and started learning about
herbal medicine, focusing especially on complex herbal formulas. He has two U.S. patents for herbal formulations and methods. He lives near Salt Lake City with his wife, Carol; they are the parents of three children and four grandchildren.