info@biomedres.us
+1 (502) 904-2126
One Westbrook Corporate Center, Suite 300, Westchester, IL 60154, USA
Site Map
ISSN: 2574 -1241
ISSN: 2574 -1241
Received: August 25, 2023; Published: September 05, 2023
*Corresponding author: Jaquellyne Gurgel Penaforte-Saboia, Clinical Research Unit, Walter Cantidio University Hospital, Federal University of Ceará, Fortaleza, Brazil
DOI: 10.26717/BJSTR.2023.52.008289
Randomized clinical trials (RCTs) are the gold-standard method for clinical research. However, major intrinsic challenges to their conduct exist, such as elevated costs, long duration, and difficulty in recruiting. Digital Technology (DT) can provide innovative solutions to overcome these challenges. The aim of this review is to critically evaluate its use in clinical research.
Keywords: Digital Clinical Trials; Digital Technology; Randomized Clinical Trials
Abbreviations: RCTs: Randomized Clinical Trials; DT: Digital Technology; CT: Clinical Trials; DCT: Digital Clinical Trials; FDA: Food and Drug Administration
In times of evidence-based medicine, randomized clinical trials (RCTs) represent one of the highest degrees of scientific evidence, being regarded as the gold standard method of clinical research [1]. The publication of well-designed clinical trials (CT) is capable of rapidly changing current medical practices, contributing greatly to advances in patient care [1,2]. Nevertheless, an important gap in RCT production has been identified. A systematic review of publications between 1995 and 2016 identified only 24 RCTs for Clinical Decision Support for clinical oncology practice [3]. Intrinsic challenges of RCTs that hinder their implementation have been widely noted, including difficulty with adequate patient enrollment, high dropout rates, high costs, and long duration [4]. Researchers have been looking for ways to overcome these challenges, for instance by using methods such as Adaptive Trial Design, Large Simple Trials, and Digital Clinical Trials (DCT) [5,6]. Guo et al argue that traditional CT methods should be reviewed in search of more innovative and agile approaches [7]. The Clinical Trials Transformation Initiative, a public-private partnership with the US Food and Drug Administration (FDA) and more than 60 other organizations, recognizes Digital Technology (DT) as a way to improve the quality and efficiency of CT [8]. Advances in the field of DT have revolutionized the way RCTs are conducted [9]. In the past few years, there has been an exponential growth in the number of CTs published on ClinicalTrials.gov that use the term mHealth, defined as “medical and public health practice supported by mobile devices, such as mobile phones, patient monitoring devices, personal digital assistants, and other wireless devices” [10,11]. Recent data from North America, Europe, and Asia revealed that over 35% of pharmaceutical industries were implementing DT in their CT, while 94% planned to increase its use soon [12].
This review will explore the role of DTs as a means of mitigating some of the difficulties in conducting RCT, focusing on four principal areas:
(i) Enrollment and dropout
(ii) Data collection
(iii) Costs
(iv) Duration
Trial Enrollment and Dropout
Data analysis from the ClinicalTrials.gov registry showed that over 90% of interventional trials enrolled fewer than 100 participants, while only 3–4% enrolled more than 5000 participants [13]. One-third of the trials did not reach the desired sample size [14]. In a review, eighty-one out of 395 RCTs listed on ClinicalTrials.gov were discontinued early, mainly due to problems with recruitment [15]. Researchers have been increasingly evaluating the potential usefulness of complementing recruitment strategies, for example, with DT, in order to include the greatest possible diversity of participants in RCTs [16,17]. There is evidence that social media can be an excellent recruitment method for hard-to-reach populations [18]. Many of the traditional RCTs undesirably restrict the study population to those living in geographic proximity to the study site [19,20]. Only a small percentage of the potential pool of eligible individuals gets invited to participate in clinical trials, limiting not only the number of participants but also the diversity of the sample, which can lead to biased results from an artificially homogeneous population [21-23]. Performing DCTs may allow remote patient participation, enabling the recruitment of otherwise unreachable individuals [24]. In 2016, one of the first applications (APP) developed to conduct a DCT was published. The trial aimed to assess the effects of nutrition labels on food purchases. In approximately 1 year, it was possible to randomize more than 2,000 patients. That showed promise for improving recruitment, delivery of the intervention, and data collection [25]. As an example, Anguera et al. managed to recruit a large number of participants in a short time and with minimal costs in a study of patients with depression that was carried out through mobile devices [26].
While the recruitment of participants can be leveraged in DCTs, participant retention could be a challenge [27]. In 2019, the average patient dropout rate in clinical trials was around 19% [28]. Some older studies have shown that DCTs had a higher dropout rate. This was in part due to less interaction between participants and researchers [29-33]. On the other hand, more recent data showed that modern DCTs may be able to increase participant engagement through more active and secure two-way communication, as well increase patients’ trust in investigators. They can improve patients’ experience and, ultimately, reduce dropout rates [9,24,27,33]. Another typical barrier to the recruitment and permanence of patients in clinical trials is the Hassle Factor of the study. Many studies are investigator-centric, requiring participants to physically attend the clinical site for sample and data collection [9,24,34]. DCTs facilitate decentralization, often allowing the participation of patients without having to leave their location or even their homes, substantially reducing potential inconveniences. Furthermore, DCTs can help mitigate a historical problem associated with randomization in RCTs that can occur when investigators are able to predict who will be randomized to each strategy (e.g., randomization by odd/even dates of birth). Automated randomization processes generated by DT are virtually impossible to be corrupted [2,35]. Accordingly, whether by expanding participation, reducing the hassle factor, or enhancing engagement, DCTs can offer unique features to increase patient recruitment and diminish dropout rates.
Data Collection
In RCTs, data security is extremely relevant, especially regarding the reliability and integrity of the information. DT has allowed a meaningful change in the way data is collected, without interfering with the fundamental principle of ensuring its authenticity, confidentiality, and integrity [6]. While computerized data collection reduces the chances of human error in recording responses [36,37], DT enables computationally intensive encryption for the privacy of participants [27]. DT allows data collection in settings outside a health facility and in real-life conditions, enabling access to new endpoints that have been otherwise impossible to collect in the past [8,24]. Nevertheless, as these technologies enable the continuous collection of new and diverse data, they can bring novel computational and statistical challenges. In addition to prioritizing the collection of critical data for the central scientific questions of CTs, the use of artificial intelligence combined with traditional biostatistical methods can be useful to address the issues associated with interpreting the vast amount of data capable of collection through DT [17,38].
Another major advantage of DTs is the ability to capture data without the presence of a member of the study team, as well as the possibility to monitor adverse events in real time through linked electronic health records [24,27]. On the other hand, when participants themselves enter information into online platforms, there is an increased risk of obtaining inaccurate (such as confusing descriptions of adverse events) or even fictitious data [35,38]. Choosing study protocols that promote the involvement of participants in decisions about data inclusion, as well as the use of digital tools capable of grouping information and analyzing trends to facilitate the identification of fictitious data, are valid strategies to minimize this obstacle [39]. Aside from that, investigators should know how to assess the accuracy and consistency of the DT chosen for data capture. This is essential for ensuring reliable capturing, processing, storage, and transfer of information, providing objective data that accurately represents the outcome [39].
Costs
Healthcare costs have been progressively increasing, leading to a significant economic impact in a world with finite resources [40]. The costs associated with CT are also progressively escalating and are often the greatest barrier to their implementation [41,42]. Data collected over the 2010–2015 period from seven major pharmaceutical companies showed that the median cost for a pharmaceutical industry phase 3 trial was US$21.4 million [43]. Another study reported that these amounts reached up to US$52.9 million [44]. Still, a growing body of evidence has demonstrated the economic benefit of mHealth interventions. A recent systematic review of thirty-nine studies that aimed to evaluate the economic impact of using mHealth tools, showed they were cost-effective and economically beneficial [40]. A study commissioned by the US Department of Health and Human Services showed that the wider use of mobile technologies is one of the most effective means of reducing clinical trial costs for drug development [44]. The US Institute of Medicine has recommended creating digital data collection systems to reduce healthcare costs. This recommendation, along with the Health Information Technology for Economic and Clinical Health (HITECH) Act, increased the adoption of electronic health record systems in the United States from 20.8% in 2004 to 85.9% in 2017 [45,46]. The slow enrollment is a major contributor to increased costs [41,46]. We have already discussed extensively how the use of DT can be useful to improve recruitment strategies. Eliminating ethnic health disparities is one way to significantly reduce overall medical costs [47]. However, in traditional RCTs, minority representation remains inadequate [22,48-50]. Lack of information and understanding about research, and limited access to specialized care centers that serve as referral sources for clinical trials make it difficult to recruit minority populations [51,52]. Jerome et al. by evaluating the use of DT as a strategy to increase patient access to ongoing RCTs, showed that digital media is a cost-effective vehicle to promote awareness of CT [16]. Classically, RCTs tend to have an investigator-centric approach. Study participants often need to travel to academic facilities where investigators and diagnostic technologies are concentrated [27]. Costs with administrative staff (11–29% of the total) and site monitoring (9–14% of the total) were key drivers of direct costs [44,53]. In DCTs, it is possible to capture data and monitor participants remotely, eliminating most patient travel requirements and allowing for the downsizing of research staff [17,28,54].
Duration
More than 70% of the total time required in the development of a new drug (6 to 10 years) is spent in clinical trials [55,56]. Moreover, a low enrollment rate can increase the planned RCT time by almost two times [46]. In contrast, it is known that the development of mHealth tools for clinical trials can be able to accelerate recruitment [57]. Espie et al. used DT in several studies and managed to quickly recruit a large number of participants. Notably, one trial recruited 3755 participants within 24 hours [58]. In addition, several ways in which DT can optimize the recruitment rate have been described above. The need for protocol review also increases CT duration [59]. Data from Tufts CSDD’s 2016 Cost Study reported that CTs with one or more global changes lasted about 18% longer than those without adjustments [60]. Most substantive amendments are implemented while the clinical trial is ongoing, and the delay in implementing these changes is directly associated with increasing the duration of CT with amendments [61,62]. On the other hand, the continuous learning and nearreal- time adaptability of DCTs can significantly reduce this interval of amendment implementation, ultimately reducing the duration of RCTs [17].
A growing body of data has demonstrated that digital technology can be an effective way of conducting the RCTs promoting more effective randomization, reducing the time and costs associated with the study, and improving data quality and security. However, the use of DT in RCTs is still in the early stages of implementation and, as with any new approach, some considerations must be mentioned. For instance, the frequent requirement for complex statistical methods; the possibility of some adjustment in the design and outcomes of the traditional RCT to better adaptation to the DCT; the need for a digital research infrastructure that, in addition to the elevated level of data security, guarantees the validation and usability of the chosen DCT; and the need to implement the interchangeability of several electronic sources to allow portability of DCT data.
Not applicable.
This work was supported by Universidade Federal do Ceará, FUNCAP and CNPq.
The authors have no conflicts of interest to declare.
Abstract
PDF
