Artificial intelligence tools for optimising recruitment and retention in clinical trials: a scoping review protocol

Abstract

Introduction

In recent years, the influence of artificial intelligence technology on clinical trials has been steadily increasing. It has brought about significant improvements in the efficiency and cost reduction of clinical trials. The objective of this scoping review is to systematically map, describe and summarise the current utilisation of artificial intelligence in recruitment and retention process of clinical trials that has been reported in research. Additionally, the review aims to identify benefits and drawbacks, as well as barriers and facilitators associated with the application of artificial intelligence in optimising recruitment and retention in clinical trials. The findings of this review will provide insights and recommendations for future development of artificial intelligence in the context of clinical trials.

Methods and analysis

The review of relevant literature will follow the methodological framework for scoping studies provided by the Joanna Briggs Institute. A comprehensive electronic search will be conducted using the search strategy developed by the authors. Leading medical and computer science databases such as PubMed, Embase, Scopus, IEEE Xplore and Web of Science Core Collection will be searched. The search will encompass analytical observational studies, descriptive observational studies, experimental and quasi-experimental studies published in all languages, without any time limitations, which use artificial intelligence tools in the recruitment and retention process of clinical trials. The review team will screen the identified studies and import them into a dedicated electronic library specifically created for this review. Data extraction will be performed using a data charting table.

Ethics and dissemination

Secondary data will be attained in this scoping review; therefore, no ethical approval is required. The results of the final review will be published in a peer-reviewed journal. It is expected that results will inform future artificial intelligence and clinical trials research.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• The search strategy was developed under the guidance of a librarian with experience in database searches, ensuring the suitability of the search strategies for each database.

• Strict eligibility criteria have been established for inclusion and exclusion for selecting the primary studies for the review.

• This protocol will ensure clarity, transparency and integrity in the research report, therefore, details of the conduct of the study will be presented in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews flow chart.

• While this breadth of coverage is useful for identifying studies overall, it may prioritise quantity over quality, meaning that equal weight is given to studies of varying quality and significance.

Introduction

Artificial intelligence (AI) is an emerging field of technology that focuses on developing theories, methods, technologies and application systems to simulate, extend and expand human intelligence.¹ It encompasses various analytical approaches, such as machine learning (ML), deep learning (DL), natural language processing (NLP) and speech recognition.¹ These approaches enable AI to perform tasks typically associated with intelligent beings.² In recent years, modern Al technology has reached a relatively advanced stage and is being actively applied in various real-life scenarios, including the medical and healthcare domain.³ At present, with the exponential growth of electronic health records (EHRs) data volume and the rapid development of computer technology, the impact of AI technology on clinical trials has seen a consistent and remarkable rise. Clinical trials are studies conducted in humans to evaluate the effectiveness and safety of an intervention,⁴ they play a critical role in the discovery of new treatments and are important practical activity for improving healthcare.⁵ In the traditional clinical trial model, the selection and recruitment mechanism of patient cohorts are critical factors that often contribute to trial failures.⁶ While a large patient pool does not ensure trial success, recruiting unsuitable patients could increase the likelihood of trial failure. These mechanisms often struggle to identify and enrol the most suitable patients in a timely manner. Additionally, they lack the technical infrastructure required to handle complex trials, particularly in the later stages, and face challenges in ensuring reliable and effective compliance control, patient monitoring and clinical key detection systems.⁶

Al has the potential to address these shortcomings in traditional clinical trial design by leveraging branches such as ML and DL. ML, a successful branch of AI, focuses on the developing programmes capable of learning from data.⁷ It can automatically identify meaningful patterns in large datasets, including text, language or images, thereby expediting clinical trials through matching and comprehensive data analysis. The human–computer interface facilitates natural information exchange between computers and humans. These functions can be applied to diverse datasets, such as EHR and medical literature, to improve patient-test matching and recruitment prior to trial initiation, match patients with appropriate clinical trials, and enable automated and continuous patient monitoring during the trial process. For example, a research team in the USA developed an AI-assisted system (Mendel.ai) for cancer clinical trials, which showed superior improvements over conventional practices in aspects like patient prescreening, leading to higher recruitment rates and reduced recruitment time.⁸ Ultimately, this can improve compliance control, yield more reliable and effective endpoint evaluation, save time, and enhance recruitment efficiency.⁶ In general, by leveraging AI in clinical trials, researchers can enhance their ability to identify and manage risks, thereby improving overall trial outcomes. AI also plays a crucial role in analysing biomedical information and enhancing the decision-making process for patient recruitment in clinical trials.⁹ The application of AI technology can greatly improve the efficiency and accuracy of clinical trials by helping clinical trial designers to quickly process similar studies, clinical data, regulatory information and to interpret relevant data. In 2020, an Australian research team introduced an AI-assisted clinical trial matching system specifically for lung cancer patients. The system demonstrated an impressive exclusion accuracy of 95.7% for subjects and a qualification evaluation of 91.6%. This result proves that this automated system is a reliable clinical decision support tool for prescreening extensive patient cohorts and predict patient subpopulations.^{10 11}

While significant progress is being made in the field of AI for clinical trials, it is crucial to acknowledge the responsibility that accompanies the opportunity to transform the clinical trial cycle through AI across various disciplines.⁹ Furthermore, the integration of AI tools in the recruitment and retention process of clinical trials can give rise to potential technical and ethical concerns.¹² In consideration of these factors, we have developed a study protocol for a comprehensive scoping review aiming to provide an in-depth overview, description and summary of the current state of AI technology implementation in the recruitment and retention process within all the kinds of clinical trials as reported in research. This includes examining the predominant application of AI tools in different types of clinical trials, as well as exploring design details such as study objectives. Furthermore, we will analyse the specific types or names of AI technologies used, their functional techniques, technical challenges faced and ethical considerations associated with these tools. Additionally, our objective is to identify both the benefits and drawbacks along with facilitators and barriers related to using AI for optimising recruitment and retention in clinical trials. Ultimately, this review aims to offer valuable insights and recommendations for future advancements in both AI technology and its integration into clinical trial practices.

Materials and methods

A comprehensive scoping review will be conducted, encompassing both peer-reviewed and grey literature, to explore the use of AI tools in optimising recruitment and retention in clinical trials that have been reported in research. The review will follow the methodological framework provided by Joanna Briggs Institute (JBI).¹³ To ensure transparency and adherence to reporting standards, the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension (PRISMA) for Scoping Reviews will serve as a guide for reporting the review¹⁴ (see online supplemental additional file 1). The selection process will be visually represented using a PRISMA flow chart, illustrated in figure 1.

Figure 1.

The PRISMA-ScR flow diagram shown will be used to illustrate the processes involved in conducting this scope review. These processes mainly consist of (1) identification: stated the number of articles identified and their respective sources and provided the number of duplicates found and removed; (2) screening: provided the total number of articles screened and detailed information on the number of full-text articles that have been evaluated as eligible, the number of eligible studies and the number of excluded studies; (3) included: the number of studies provided in this section included following quality assessment, data extraction and thematic analysis, respectively. PRISMA-ScR, Preferred Reporting Items for Systematic Reviews and Meta-analyses extension for Scoping Reviews.

Stage 1: identifying the research question

The primary aim of this review is to provide a comprehensive overview of the current state of AI tool application in recruitment and retention process in clinical trials that has been reported in research. Additionally, our objective is to map out the benefits and drawbacks, facilitators and barriers associated with the development, implementation and evaluation of AI tools in recruitment and retention. By mapping out these factors, we aim to gain a deeper understanding of the potential benefits and challenges of using AI in clinical trial processes.

Based on the objectives of this scoping review, we have formulated the following questions.

1. What are the different types or names of AI tools that are currently available for assisting in optimising the recruitment and retention process in clinical trials?

2. What specific types of clinical trials are AI tools predominantly applied to, and what are the design details (eg, study objective, sample size) of those clinical trials?

3. What are the outcomes, benefits and drawbacks associated with the use of AI tools in assisting optimising recruitment and retention in clinical trials?

4. What are the barriers and facilitators encountered in the utilisation of AI tools to assist in optimising the recruitment and retention process in clinical trials?

Stage 2: identifying relevant studies

Based on the results of our pilot-tested search startegy, we have decided to include peer-reviewed articles that encompass analytical observational study designs including prospective and retrospective cohort studies, case–control studies and analytical cross-sectional studies, descriptive observational study designs including descriptive cross-sectional studies. In addition, studies from experimental and quasi-experimental study designs including randomised controlled trials, non-randomised controlled trials will be included. An electronic search will be conducted in leading medical and computer science databases, including PubMed, Embase, Scopus, IEEE Xplore and Web of Science Core Collection(SCI-EXPANDED, SSCI, A&HCI, ESCI). We developed comprehensive search strategies in the above databases (see online supplemental additional file 2). We will include peer-reviewed publications in all languages, the utilisation of both human and machine translation will be employed as deemed necessary. No time limits will be applied to the literature search. We will conduct additional review by handsearching for other relevant publications cited in the reference section of the included publications. The search terms will include AI, AI, clinical trials, and recruitment and retention. Search terms are different according to different databases. Boolean terms will be used to separate the keywords. The expertise of an experienced librarian will be enlisted to ensure the implementation of a robust search strategy for the review. This will guarantee a comprehensive and thorough identification of relevant studies. To compile all relevant evidence sources, identify, remove duplicate records and screen literature, Covidence will be employed.

Stage 3: study selection

We will establish eligibility criteria to ensure that the selected studies include specific information relevant to our research question. These criteria will be developed based on the JBI approach, which emphasises the formulation of scoping review questions that describe the population, concept and context of the study.¹³

Inclusion criteria

To be included in this scoping review, studies must meet the following criteria:

• The study must be available in full text.

• The study must focus on application of AI technology.

• The study should be observational study and experimental study.

• The study must specifically address the use of AI technology to assist in recruitment and retention in clinical trials.

Population

The scoping review will source all relevant peer-reviewed and grey literature that describes the application of AI in recruitment and retention process in clinical trials. The population sample for the review will be individuals who participate in clinical trials of AI applications, specifically focusing on the recruitment and retention processes, participants of all ages and occupations will be included.

Context

The context of the study focuses on AI-assisted recruitment and retention in clinical trials. The review will encompass studies and grey literature, with no geographical restrictions, to ensure a comprehensive evaluation of the development, implementation and assessment of AI technology in the field. By conducting an extensive search, the review aims to gather a wide range of references to effectively categorise the nature and types of AI technology used for recruitment and retention in clinical trials. This approach will provide a more comprehensive understanding of the topic.

Concept

The concept of AI-assisted recruitment and retention in clinical trials entails the utilisation of various AI techniques to enhance the process of participant recruitment, patient selection, patient retention and patient drop-outs process in clinical trials. The AI mentioned in this review includes all branches of AI, such as ML, DL, NLP, computer vision system, neural networks, etc.

Exclusion criteria

Studies will be excluded if they are:

• Not be available in full text.

• Not include the details of AI technology used.

• Focus on clinical trials conducted on AI, rather than the application of AI to clinical trials.

• Not application of AI in recruitment and retention in clinical trials, but other process, such as AI in clinical trial design.

Stage 4: charting the evidence

Covidence will be used to identify and remove all duplicate articles. The feasibility and efficiency of it will be pilot-tested by two reviewers independently. All publications identified according to these criteria will go through title and abstract screening and then full-text screening, both of which are conducted independently by two authors. In cases where there are disagreements or discrepancies, a consensus will be reached through group discussions involving all three authors.

Stage 5: extracting the data

A data extraction form will be developed to systematically gather relevant information from the included articles. Two reviewers will independently extract the data from the selected studies using a pilot-tested form. In cases where there are discrepancies, a third reviewer will be consulted to reach a consensus. Elements of data extraction will include author(s), year of publication, country, study title, study aim/objective, study population, sample size, technique function, details of study participants and study design. Moreover, we will extract the type or name of AI assistance technology, technical and ethical issues, benefits and drawbacks, barriers and facilitators associated with its application in recruitment and retention process in clinical trials as in table 1. Missing information from the publications will be noted, and authors will be contacted for additional information.

Table 1.

Data extraction table

Reviewer: date of data extraction:
Publication details
Authors
Year
Country
Article title
Study design and details
Study objective
Type or name of the AI technology
Technique function
Sample size
Details of study participants (number/age/sex/study population)
Study design
Outcomes/details of results
Technical issues
Ethical issues
Benefits and drawbacks
Barriers and facilitators
Recommendations

AI, artificial intelligence.

Stage 6 : critical assessment of evidence

While critical appraisal is not mandatory for scoping reviews, it is still useful for reporting the risk of bias in the included studies.¹³ The Newcastle-Ottawa Scale will be used to assess the quality of non-randomised studies,¹⁵ the Consolidated Criteria for Reporting Qualitative Studies will be used to assess the quality of qualitative studies,¹⁶ the Consolidated Standards of Reporting Trials will be used to assess the quality of randomised controlled trials.¹⁷ Two independent reviewers will perform the assessment for each study, and any discrepancies will be resolved through discussion and consensus among the review authors. A critical assessment of the evidence will be presented in tables and figures, and illustrated in the narrative summary.

Stage 7 : collating, summarising and reporting the results

The data collected will be collated and summarised using a basic descriptive approach,¹⁸ employing content analysis described by Pollock et al.¹⁹ The content analysis will categorise the different AI technologies currently employed in assisting the recruitment and retention process in clinical trials based on their bibliographic characteristics, main outcomes, typologies, as well as the technical and ethical issues, benefits and drawbacks, facilitators and barriers that influence their development, implementation and assessment.

Patient and public involvement

Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

Planned start and end dates

The study is planned to start in March 2024 and end by May 2024.

Discussion

The medical industry holds tremendous potential for the application of AI, and one prominent area of AI implementation is in clinical trials. Currently, AI is being used in clinical trials across many countries. The integration of Al in clinical trials offers a promising solution to address the inherent problems associated with traditional clinical trials methodologies and has demonstrated its practicality in clinical trials, it is crucial to ensure that any innovations are subject to rigorous research and development processes in order to maintain their value and reliability. This necessitates the implementation of appropriate regulatory measures to safeguard the integrity and effectiveness of AI applications in clinical trials.

This scoping review aims to comprehensively examine the current state of AI utilisation in recruitment and retention process of clinical trials that has been reported in research. It seeks to identify the advantages and disadvantages of employing of AI in this context and gain a comprehensive understanding of barriers and facilitators that have affected its application. The review will also provide recommendations for future AI developments in clinical trials and offer insights to bridge any existing gaps in the literature. The finding will serve as a valuable resource for stakeholders, informing their decisions and potentially shaping policies and guidelines pertaining to the use of AI in recruitment and retention processes.

However, it is important to acknowledge some limitations of this review protocol. One such limitation is it may prioritise quantity over quality, meaning that equal weight is given to studies of varying quality and significance. Any additional limitations encountered during the study will be addressed in the final full-text scoping review, for example, the results of interventions may be too heterogeneous. Any modifications made to the protocol will be thoroughly documented and incorporated into the final publication of the scoping review results.

Ethics and dissemination

Secondary data will be attained in this scoping review; therefore, no ethical approval is required. Other ethical issues are unexpected. The results of the final study will be published in a peer-reviewed journal. It is expected that results will inform future AI and clinical trials research.

Ethics statements

Patient consent for publication

Not applicable.