Clinical Research Informatics: Contributions from 2022

Summary

Objectives: To summarize key contributions to current research in the field of Clinical Research Informatics (CRI) and to select best papers published in 2022.

Method: A bibliographic search using a combination of Medical Subject Headings (MeSH) descriptors and free-text terms on CRI was performed using PubMed, followed by a double-blind review in order to select a list of candidate best papers to be then peer-reviewed by external reviewers. After peer-review ranking, a consensus meeting between the two section editors and the editorial team was organized to finally conclude on the selected three best papers.

Results: Among the 1,324 papers returned by the search, published in 2022, that were in the scope of the various areas of CRI, the full review process selected four best papers. The first best paper describes the process undertaken in Germany, under the national Medical Informatics Initiative, to define a process and to gain multi-decision-maker acceptance of broad consent for the reuse of health data for research whilst remaining compliant with the European General Data Protection Regulation. The authors of the second-best paper present a federated architecture for the conduct of clinical trial feasibility queries that utilizes HL7 Fast Healthcare Interoperability Resources and an HL7 standard query representation. The third best paper aligns with the overall theme of this Yearbook, the inclusivity of potential participants in clinical trials, with recommendations to ensure greater equity. The fourth proposes a multi-modal modelling approach for large scale phenotyping from electronic health record information. This year's survey paper has also examined equity, along with data bias, and found that the relevant publications in 2022 have focused almost exclusively on the issue of bias in Artificial Intelligence (AI).

Conclusions: The literature relevant to CRI in 2022 has largely been dominated by publications that seek to maximise the reusability of wide scale and representative electronic health record information for research, either as big data for distributed analysis or as a source of information from which to identify suitable patients accurately and equitably for invitation to participate in clinical trials.

Keywords

Observational studies as Topic - real-world data - real-world evidence generation - consent - phenotyping

1 Introduction

Within the 2023 International Medical Informatics Association (IMIA) Yearbook, the goal of the Clinical Research Informatics (CRI) section is to provide an overview of research trends from 2022 publications that demonstrate the progress in multifaceted aspects of medical informatics supporting research and innovation in the healthcare domain. New methods, tools, and CRI systems have been developed in order to enable real-world evidence generation and optimize the life-cycle of clinical trials. Although this year's Yearbook focus should ideally embrace human, animal, plant and planetary considerations, we did not find, in 2022, CRI publications that have such a holistic perspective. The selections made have therefore focused on maximising the inclusivity of people within clinical research.

2 About the Paper Selection

A comprehensive review of articles published in 2022 and addressing a wide range of issues for CRI was conducted. The selection was performed by querying MEDLINE via PubMed (from NCBI, National Center for Biotechnology Information) with a set of predefined MeSH descriptors and free terms: Clinical research informatics, Biomedical research, Nursing research, Clinical research, Medical research, Pharmacovigilance, Patient selection, Phenotyping, Genotype-phenotype associations, Feasibility studies, Eligibility criteria, Feasibility criteria, Cohort selection, Patient recruitment, Clinical trial eligibility screening, Eligibility determination, Patient-trial matching, Protocol feasibility, Real world evidence, Data Collection, Epidemiologic research design, Clinical studies as Topic, Multicenter studies as Topic, and Evaluation studies as Topic. Papers addressing topics of other sections of the Yearbook, such as Translational Bioinformatics, were excluded based on the predefined exclusion of MeSH descriptors such as Genetic research, Gene ontology, Human genome project, Stem cell research, or Molecular epidemiology.

Bibliographic databases were searched twice, initially in November 2022 and then refreshed for late in the year publications in January 2023, considering the electronic publication date. 1,324 papers were found and screened for relevance to CRI, for reporting original research as opposed to opinion or educational papers, and their scientific quality was blindly rated as low, medium, or high by the two section editors based on papers' title and abstract. 55 references classified as high quality contributions to the field by at least one of the two section editors or medium quality by both were considered and classified into the following eleven areas of the CRI domain in order of the number of matching papers (multiple classification choices were permitted): reuse of Electronic Healthcare Records (EHRs), Learning Healthcare System (LHS) data; Big data management, data integration, semantic interoperability and data quality assessment; Data science (data/text mining, Artificial Intelligence (IA), Machine Learning (ML)); Security and data privacy; Feasibility studies, patient recruitment, improved user experiences of CRI systems and Governance (ethical, regulatory, societal, policy issues, stakeholder participation, research networks, team science). A further review by both editors on the basis of full paper screening, again selecting those that scored at least one high or two medium, resulted in nine candidate papers being taken forward to external peer review. In conformance with the IMIA Yearbook process, these nine papers were peer-reviewed by the IMIA Yearbook editors and external reviewers. Each paper was reviewed by at least four reviewers. Four papers were finally selected as best papers ([Table 1]). A content summary of these best papers can be found in the appendix of this synopsis.

Table 1 Selection of best papers for the 2023 IMIA Yearbook of Medical Informatics for the Clinical Research Informatics section. The articles are listed in alphabetical order by the first author's surname.

3 Outlook

The nine candidate best papers for 2022 illustrate recent efforts and trends in different CRI areas: feasibility studies and patient recruitment; interoperability, data harmonisation and data migration; real-world evidence generation and electronic phenotyping; ethical, legal, and social issues including consent and AI ethics; information security and secure record linkage.

3.1 Feasibility Studies, Patient Recruitment

One of the selected best papers, summarized below, evaluates the reuse of Electronic Health Records (EHRs) for clinical trial feasibility queries across multiple hospital sites [[1]]. While this capability is already established through exporting hospital data into interoperable data warehouses like OMOP (Observational medical outcomes partnership) or i2b2, the study aims to replicate it using HL7 FHIR. Validation was performed across four German MII hubs. The paper's significance lies in building tools directly on the increasingly adopted FHIR standard, which has the potential to enhance adoption, reduce costs and complexity, and alleviate the need for specialized skills in alternative models.

3.2 Interoperability, Data Integration and Harmonization

The challenges of database migration and mapping were examined in a publication on re-designing and migrating an existing rare disease registry for it to be used in regulatory context [[2]]. This paper focuses on the remodeling of the Registry of Multiple Osteochondromas (REM) to align with regulatory requirements and recommendations. The study emphasizes the key stages involved in achieving semantic interoperability, data quality, and governance within the registry. To enhance interoperability, ontologies and standards were integrated for proper data collection following FAIR principles. Data quality was improved through the addition of parameters and domains, minimizing human error. Furthermore, a two-level governance structure was established to increase visibility for the scientific community and patients. The remodeled REM registry effectively meets the needs of the scientific community and provides valuable real-world evidence for regulatory purposes.

Another strong paper last year addresses the challenge of integrating data across diverse research environments in collaborative projects [[3]]. While allowing for variation in data collection protocols, achieving interoperability is crucial for successful collaboration. The Clinical Sequence Evidence-Generating Research (CSER) consortium Data Coordinating Center (DCC) shares its experiences in coordinating survey and genomic sequencing data from seven research sites. Fourteen lessons learned and eleven recommendations for future consortia are identified, covering communication and planning, harmonization, informatics, compliance, and analytics. The paper highlights the importance of budgeting for data coordination early in the consortium's development and emphasizes the need for continuous communication and solid data governance approaches.

3.3 Real-world Evidence Generation, Electronic Phenotyping

The routine use of structured electronic healthcare records (EHRs) offers opportunities to address gaps in clinical evidence. However, many publications have underscored the importance of addressing challenges related to verification, validation, data privacy, and the social responsibility of conducting research. To tackle these challenges, the European Society of Cardiology and the BigData@Heart consortium have collaborated with various stakeholders, including patients, clinicians, scientists, regulators, journal editors, and industry experts. They propose the CODE-EHR Minimum Standards Framework [[4]], which provides a practical approach to enhance study design, transparency, and the effective utilization of healthcare data for research purposes, providing a roadmap for future improvements. This paper contributes to the development of guidelines and best practices that promote the responsible and impactful use of big data in healthcare research.

In the same theme, MixEHR-Guided is described in one of the four selected best papers [[5]]. It is an automatic phenotyping model that utilizes a probabilistic joint topic modeling approach to projects a patient's high-dimensional and heterogeneous clinical record onto a low-dimensional latent topic mixture membership over disease phenotypes. The training uses a generative latent topic model inspired by Latent Dirichlet Allocation (LDA), but unlike LDA, the model associates topics with reference phenotypes. This allows the result to be interpretable. The authors apply MixEHR-Guided on MIMIC-III and on PopHR (1.3 M patients from Québec, Canada). 1,500 identifiable phenotypic topics are inferred and exhibit meaningful connections among ICD and ATC codes. The inferred phenotypic topics accurately recovered 9 out of 12 rule-based labels. Based on a qualitative comparison with reference phenotypes, the authors claim that MixEHR-G's phenotype topics are clinically meaningful. MixEHR-G is novel in obtaining simultaneous phenotypic predictions (for ~1,500 phenotypes) and improving the interpretability of phenotypic subjects.

3.4 Ethical, Legal, and Social Issues

The highest scoring best paper in the CRI Section presents a model for obtaining broad consent for secondary use of health care data and human bio-samples that is compliant with Europe's data protection Regulation (the GDPR) and has the endorsement of all key ethical and data protection decision makers in Germany [[6]]. It has up to now been very difficult to gain approval for consent specifications that are relatively broad because the data subject is not deemed to have been well enough informed of the purpose or purposes and organisations processing the data. It is hoped that this example from Germany will inspire other countries to establish a legally acceptable basis for broad consent, which is a critical enabler of big data research ecosystems. Another contrasting ethical paper, also the best paper this year, focuses on the inclusion of diverse populations within clinical trials [[7]]. This paper explains the importance of study populations genuinely representing the patient populations who will utilise the research results (such as an innovative treatment or an algorithm) and highlight some of the main areas of disparity that are found today amongst clinical trial populations. This paper also positively explains how a proactive approach to equity and representativeness can be taken. The third ethical paper that we highlight deals with compensating statistically for biased will inevitably exist within real-world data [[8]]. This focuses on the reuse of existing EHR data within a real-world control arm of the study, which is increasingly attractive in rare disease and paediatric research. To be an accurate control the population included from the real-world data must be comparable, not only from the perspective of representativeness of the individuals but best stages in disease and treatment pathways. The authors describe a statistical approach to estimate the bias of the new external control arm, based on previous observed biases. Since real-world data is increasingly available and clinical trials are increasingly expensive, real-world control arms are gaining favour but must be designed robustly. Finally in this ethics sub-section, the literature survey paper is a scoping review on health equity. Maurud et al. [[9]] conducted a scoping review on ways health equity has been promoted in clinical research informatics with patient implications. They found that the literature in 2022 predominantly focused on bias and equity in the data used for AI development. They interestingly found that although there is a risk of algorithm bias through training with biased data, they also note that AI has at times uncovered bias in existing treatment decision-making. AI can therefore be helpful to reduce bias as well as being vulnerable to bias.

3.5 Information Security

Record linkage-based patient intersection cardinality for rare disease studies using Mainzelliste and secure multi-party computation [[10]]. This paper presents a complementary solution to our best paper on broad consent, to address the challenge of linking patient records from multiple data sources accurately, with the least amount of identifying or potentially identifying data to match the records. The broad consent approach of Zenker et al. [[6]] mentioned above and elaborated below is an approach that can be adopted for pseudonymised data which has a linkage key per patient record. Secure multi-party computation takes the contrasting approach of avoiding the need to incorporate linkage keys within data set, so that each dataset can be processed in an anonymized form, therefore not necessarily requiring consent for this processing. This approach also avoids the need to establish a trusted third-party to perform linkage and then to generate an anonymized data extract for the research users. Most Privacy Preserving Record Linkage solutions utilise a Trusted Third Party (TTP) approach using hashed identifiers to minimise the risk of inappropriate relinkage, but centralised within the TTP the mechanisms that safeguard this process. Secure multi-party computation enables the analysis across multiple data repositories without centralising the data or requiring the centralisation of (hashed) linkage keys. In this paper, the authors extend the Mainzelliste record linkage framework with an error-tolerant, “fuzzy” match algorithm, increasing the linkage quality on noisy and incomplete data.

Die Autoren geben an, dass kein Interessenkonflikt besteht.

Acknowledgement

We would like to acknowledge the support of Adrien Ugon, Martina Hutter, Kate Fultz Hollis, Lina Soualmia, and the whole Yearbook editorial team as well as the reviewers for their contribution to the selection process of the Clinical Research Informatics section for this IMIA Yearbook.

• References

• 1 Gruendner J, Deppenwiese N, Folz M, Köhler T. Kroll B, Prokosch HU, et al. The Architecture of a Feasibility Query Portal for Distributed COVID-19 Fast Healthcare Interoperability Resources (FHIR) Patient Data Repositories: Design and Implementation Study. JMIR Med Inform 2022 May 25;10(5):e36709. doi: 10.2196/36709.
• 2 Mordenti M, Boarini M, D'Alessandro F, Pedrini E, Locatelli M, Sangiorgi L. Remodeling an existing rare disease registry to be used in regulatory context: Lessons learned and recommendations. Front Pharmacol 2022 Sep 23;13:966081. doi: 10.3389/fphar.2022.966081.
• 3 Muenzen KD, Amendola LM, Kauffman TL, Mittendorf KF, Bensen JT, Chen F, et al. Lessons learned and recommendations for data coordination in collaborative research: The CSER consortium experience. HGG Adv 2022 May 20;3(3):100120. doi: 10.1016/j.xhgg.2022.100120.
• 4 Kotecha D, Asselbergs FW, Achenbach S, Anker SD, Atar D, Baigent C, et al. CODE-EHR best practice framework for the use of structured electronic healthcare records in clinical research. BMJ 2022 Aug 29;378:e069048. doi: 10.1136/bmj-2021-069048.
• 5 Ahuja Y, Zou Y, Verm, A, Buckeridge D, Li Y. MixEHR-Guided: A guided multi-modal topic modeling approach for large-scale automatic phenotyping using the electronic health record. J Biomed Inform 2022 Oct;134:104190. doi: 10.1016/j.jbi.2022.104190.
• 6 Zenker S, Strech D, Ihrig, K, Jahns R, Müller G, Schickhardt C, et al. Data protection-compliant broad consent for secondary use of health care data and human biosamples for (bio)medical research: Towards a new German national standard. J Biomed Inform 2022 Jul;131:104096. doi: 10.1016/j.jbi.2022.104096.
• 7 Peters U, Turner B, Alvarez D, Murray M, Sharma A, Mohan S, et al. Considerations for Embedding Inclusive Research Principles in the Design and Execution of Clinical Trials. Ther Innov Regul Sci 2023 Mar;57(2):186-95. doi: 10.1007/s43441-022-00464-3.
• 8 Incerti D, Bretscher MT, Lin R, Harbron C. A meta-analytic framework to adjust for bias in external control studies. Pharm Stat 2023 Jan;22(1):162-80. doi: 10.1002/pst.2266.
• 9 Maurud S, Henni SH, Moen A. Health Equity in Clinical Research Informatics. Yearb Med Inform 2023 Jul 6:138-45. doi: 10.1055/s-0043-1768720.
• 10 Kussel T, Brenner T, Tremper G, Schepers J, Lablans M, Hamacher K. Record linkage based patient intersection cardinality for rare disease studies using Mainzelliste and secure multi-party computation. J Transl Med 2022 Oct 8;20(1):458. doi: 10.1186/s12967-022-03671-6.

Publikationsverlauf

Artikel online veröffentlicht:
26. Dezember 2023