1 Medical Informatics as a Support for Reducing Health Inequalities and Healthcare
Disparities
The Corona Virus Disease 2019 (COVID-19) pandemic has been a particularly powerful
and painful indicator of social inequalities, especially in access to health care.
As the focus of 2021 Yearbook was “Managing Pandemics with Health Informatics – Successes
and Challenges” [[1]], this year’s theme continues with a closer look at health equity in our health
systems. Indeed, social determinants of health have a major impact on mortality and
morbidity from the pandemic, making fragile populations more vulnerable [[2]]. Prior to the COVID-19 health crisis, Ebola had already highlighted the importance
of developing resilient information systems, i.e. capable of guaranteeing the care
and good health of all, even in times of crisis or afterwards, by adapting to the
situation [[3]]. In the meantime, the increase in migration due to war or problems resulting from
climate change also revealed issues that the scientific community must address. In
particular, Chris Lehmann saw an urgent need for action by the medical informatics
researchers to improve international information exchange to facilitate the care of
displaced populations [[4]].
In this frame, the theme of the 31st edition of the International Medical Informatics
Association (IMIA) Yearbook is “Inclusive Digital Health: Addressing Equity, Literacy,
and Bias for Resilient Health Systems”. The Special Section provides a review of significant
papers for 2021 that focus on health equity and informatics tools. Section editors
Brian Dixon and John Holmes found studies that characterized significant challenges
facing biomedical informatics with respect to practices that support equity and literacy
in the design of health information systems [[5]]. Challenges are further addressed in the survey paper where Koehle et al. stress that “it may be easy to look at the current health equity landscape as irreparable,
having been built on hundreds of years of oppression, marginalization, and discrimination.
In this work, we have emphasized collaboration with user and patient groups to define
priorities, ensure accessibility and localization, and consider risks in development
and utilization of digital health tools” [[6]].
2 Highlights of the 31st Edition of the IMIA Yearbook
We are proud to include a chapter on Education by Edward (Ted) H. Shortliffe, who
kindly accepted the invitation of Sabine Koch (IMIA past president) to write a paper
adapted from the presentation he gave at the IMIA François Grémy Award of Excellence
ceremony during Medinfo2021 [[7]]. Be sure to read this chapter and the International Academy of Health Sciences
Informatics (IAHSI) paper by William (Bill) Hersh, the Academy’s President [[8]]. The History of Medical Informatics chapter by Casimir Kulikowski [[9]] provides an analysis of early papers on ethics in clinical decision making relevant
to the equity, fairness, and inclusiveness desired by today’s digital health systems.
In this 31st edition of the IMIA Yearbook, you will also find valuable contributions
from IMIA Working Groups related to the significant topic of inclusive digital health
(to name a few: the need for new generations of electronic health records, the drivers
of inequity in healthcare, the challenges of inclusive health information technology
for vulnerable older adults, open source projects, the comparison of several digital
primary care systems that can promote system equity…), showing the opportunity to
use lessons learned from COVID-19 in the development of digital health, and how they
can contribute to making healthcare systems more resilient, accessible and equitable.
The Research paper by Tiffany Veinot et al. [[10]] also a co-author of a best paper selected in the Consumer Health Informatics section
[[11]], concentrates on nine health informatics studies with a focus on equity in the
U.S. and Canada (disparities related to age, disability or chronic illness, gender/sex,
place of residence, race/ethnicity, sexual orientation, and socioeconomic status)
from which they derive a framework of four considerations for advancing health equity
in informatics: PRAXIS (Participation, Representation, Appropriate methods and interventions,
conteXtualization and structural competence, and Investigation of Systematic differences).
In the Bioinformatics and Translational Informatics (BTI) section’s best papers, co-editors
Mary Lauren Benton and Scott McGrath observed several important trends, including
the use of deep learning approaches to analyze diverse data types, the development
of integrative and web-accessible bioinformatics pipelines, and a continued focus
on the power of individual genome sequencing for precision health [[12]]. The survey paper for BTI by Tang et al. addresses how equity and inclusion should be incorporated in every step of bioinformatics
projects [[13]]. More work needs to be done in computational biology to include specific groups
of patients in algorithms and avoid algorithm bias, and the amount of data available
continues to be a challenge.
The Clinical Research Informatics (CRI) section also named some of the informatics
challenges to handling data. Editors Christel Daniel, Dipak Kalra and Xavier Tannier
wrote that the increasing scale and scope of biomedical data is not only generating
enormous opportunities for improving health outcomes but also raises new challenges
ranging from data acquisition and storage to data analysis and utilization [[14]]. In the CRI survey paper, Kargl et al. addressed the challenges of artificial intelligence (AI) in clinical research and
biobanking. The authors emphasized the need for an ethical-mindfulness and proportionate
approach to AI in biomedical research. They specifically reveal the need for AI ethics
research focused on understanding and resolving practical problems arising from the
use of AI in science and society [[15]].
With a focus on health equity in decision support, Stipelman et al. in their Decision Support (DS) survey for 2021 [[16]] emphasized that Clinical Decision Support (CDS) tools may improve health equity
and outcomes for patients who face disparities. The authors underscored the need for
high-quality analyses of CDS-associated health outcomes, reporting of user-centered
design and implementation strategies used in low-resource settings, and methods to
disseminate CDSs created to improve health equity. DS co-editors Damian Borbolla and
Tiago Colicchio also found a wide variety of research in the field this year as well
as continued interest in CDS to help in cases of COVID-19 in the pandemic [[17]]. This year’s best paper selection highlighted innovation in the CDS evaluation
process, alert fatigue, and a more traditional approach to measuring the efficacy
and efficiency of CDS alerts.
The Human Factors and Organizational Issues (HFOI) section, co-editors Yalini Senathisrajah
and Anthony Solomonides chose four papers that were clearly outstanding and help advance
the field of human factors [[18]]. The papers examine novel and important topics such as the nature of human-machine
interaction behavior and norms, the use of social media-based design for an electronic
health record, and emerging topics such as brain-computer interfaces, thematic development
of electronic health records and usability techniques, and condition-focused patient
facing tools.
The Consumer Health and Informatics (CHI) section also included papers on the coronavirus
but interestingly, section editors Annie Lau and Pascal Staccini discovered that the
COVID-19 topic was retrieved without any use of the term in the query [[19]]. Selected papers for CHI illustrate an area of inclusiveness that should be acknowledged,
such as sexual orientation and methods to reach people in lower income countries.
Also, the topic of disability, deafness, in particular, needs to be acknowledged especially
with the rise in telehealth - how would this community receive care when seeking medical
help relies mainly on the telephone?
Check out the survey paper for CHI that addresses some progress in health care opportunities
for racial and ethnic minorities [[20]]. Valdez and Rogers conclude that future research should utilize community engagement
strategies to design interventions that are attune to multiple racial and ethnic minoritized
populations across geographic regions in addition to numerous intersectional identities
and multiple co-morbidities. In the Keynote paper, co-authored by Valdez, Ancker,
and Veinot [[21]], you will find five “provocations” related to health informatics as a starting
point for building a more conscientious informatics practice. The authors assert that
radical changes in current practice will enhance patient-centered approaches, especially
for marginalized communities.
Read the comprehensive survey paper by Neil Sarkar, who reviewed 256 articles published
between 2018 and 2021 related to Health Information Exchange (HIE) issues [[22]]. Of the seven categories of works he identified, one is dedicated to health disparities
and summarizes ten papers addressing health literacy and equity. Of the three best
papers selected this year from a total of 15 candidates described in the synopsis
by Eta S. Berner and Meryl Bloomrosen, the HIE section co-editors [[23]], two papers are relevant to the issues of this year’s special theme. Nsaghurwe
et al. describe in great detail a five-step approach to HIE development in Tanzania that
may be useful to many other countries developing their national health information
systems [[24]]. Chen et al., showed significant disparities in the ability of hospitals to adopt telehealth depending
on their geographic location [[25]].
The co-editors of the Clinical Information Systems (CIS) section, Werner Hackl and
Alexander Hörbst, did not identify any real novelties in the themes addressed in the
10 candidate articles selected this year compared to recent years [[26]]. The work by Harris et al., which is one of the two best papers selected in this section, conducted a large
study in five different countries in diverse geographic locations to investigate the
potential utility of mobile consultation for health care delivery [[27]]. The authors emphasized that low- and middle-income countries are motivated to
embrace this digital shift, but that this requires addressing the inevitable technical
and privacy issues and a robust health information system that is essential to ensure
continuity of care. Check out the survey paper of the CIS section which focuses on
patient experience and how it is defined and captured in eHealth [[28]]. The authors reviewed 44 papers, and concluded that patient experience is complex
and multifaceted, requiring additional research to better understand this phenomenon.
The survey paper of the Knowledge and Representation Management (KRM) section presents
a selection of 22 papers from 2020 and 2021 focusing on contributions to health inclusivity
and bias [[29]]. The author, Janna Hastings, grouped these works according to their contribution,
namely: (i) semantic resources developed to represent knowledge about health inequalities
and biases, (ii) efforts to enable data sharing across different health information
systems, communities, cultures, and disciplines while ensuring data privacy, and (iii)
solutions that take into account the importance of reducing biases that exist in the
data on which medical AI algorithms are trained in order to achieve inclusiveness.
One of the best papers in the KRM section details the new version of the International
Classification of Diseases, ICD-11 [[30]]. Among the many changes from the previous version is the introduction of a new
chapter entitled “Conditions related to sexual health” that includes gender dysphoria,
which is much less stigmatizing in its description than before. This notion is now
labelled “Gender incongruence” contrary to the way it was done in ICD-10, where the
corresponding concept appeared as a “Gender identity disorder, unspecified” in the
chapter “Mental and behavioural disorders”, thus considering it as a disease [[31]]. The other 14 candidate best papers described by co-editors Licong Cui and Jean
Charlet of the KRM section synopsis did not specifically address issues related to
this year’s theme [[32]], nor did the contributions listed afterward.
The survey paper of the Signals, Sensors, and Imaging Informatics (SSII) section presents
a systematic review of 223 articles related to automated computational segmentation
of the lung and its lobes [[33]]. The authors classified the different works according to the types of methods developed,
namely those based on: (i) thresholding, (ii) region, (iii) shape or pattern, (iv)
traditional machine learning, (v) deep learning, and finally those (iv) guided by
neighborhood anatomy. Twenty public datasets used in these studies are also listed.
SSII co-editors Christian Baumgartner and Thomas Deserno wrote a synopsis featuring
their selection of the best papers of 2021 in which an emerging trend was notably
observed, namely the combination of signal processing and sensor technologies, and
where the development of deep learning approaches for medical image processing remains
substantial [[34]].
The synopsis of the Public Health and Epidemiology Informatics (PHEI) section presents
the 11 best papers they nominated [[35]]. Georgeta Bordea and Gayo Diallo, the co-editors of this section, noted in particular
that some of these works have focused on data that is not considered health data per
se, such as that from mobile apps and social media. Swertz et al., wrote the PHEI survey paper on existing cohort and real-world data catalogues [[36]]. The authors provided recommendations for standardization efforts that should be
undertaken to enable sharing of these catalogues within the scientific community.
The Natural Language Processing (NLP) survey paper from Aramaki et al. focuses on practical clinical applications of NLP methods [[37]]. The authors listed NLP-based systems implemented in ten hospital departments to
illustrate what they call “bedside applications”, while specifying the different types
of NLP tasks addressed. Check out the NLP synopsis by co-editors Natalia Grabar and
Cyril Grouin, who conducted a thorough analysis of the 2021 biomedical NLP literature
[[38]]. They noticed that transformer-based models are extensively studied, that there
is a lot of work on dedicated applications such as COVID-19 and mental health, and
still many papers presenting information extraction methods.
Last but not least, read the well-documented Cancer Informatics (CI) survey paper
from Chaunzwa et al. related to the study and measure of disparities and bias in cancer incidence [[39]]. The authors reviewed clinical informatics studies related to cancer disparities
published from 2018 to 2021, with a focus on domains such as real-world data analysis,
natural language processing, radiomics, genomics, proteomics, metabolomics and metagenomics.
They concluded that care will be needed to ensure that algorithmic bias does not magnify
existing disparities. “class="char-style-override-3">Real-world data yields important
insights into the conduct of clinical trials and registries”, concluded Debra Patt
and Jeremy Warner, co-editors of the CI section, in their review of scientific contributions
published in 2021 on topics related to cancer informatics [[40]]. Indeed, the two selected best papers deal with complementary aspects of real-world
databases and registries for population-level research in cancer [[41], [42]]. Women, the elderly, and racial or ethnic minorities have been relatively excluded
from cancer clinical trials, and this trend appears to be worsening over time, as
studied by Riaz et al. [[43]]. Liu et al. determined that relaxing trial eligibility criteria would result in small changes
in hazard ratios for critical outcomes such as overall survival [[41]]. This has major implications for the future design of cancer clinical trials. Yang
et al. assessed the prevalence of missing data in a very large cancer patient registry,
and determined whether missing data was itself a prognostic factor [[42]]. The authors demonstrated the importance of metadata characteristics in the conduct
and evaluation of real-world data registry studies. They note that “Records with missing
data were more prevalent among Black patients and patients from other racial and ethnic
minority groups, which may reflect long-standing disparities in access to health care
and cancer treatment”.
The 2023 special topic of the Yearbook will be: “Informatics for One Health”. As we
learn to live with COVID-19, we have come to realize that we can live differently.
It is even essential to realize that we must do so at all costs if we want everyone
to continue to live with dignity and equity. Indeed, climate change has a major impact
on our environment and on the lives of animals and humans. It is therefore essential
to consider and ensure the health of all these components, not only our own, as already
underlined in this year’s Yearbook [[44]]. As a recent article by Gray points out, health informatics has a role to play
in addressing climate-related global health issues [[45]]. These observations led us to choose next year’s theme in order to identify and
highlight the contributions of research in medical informatics in the area of one health to encourage our scientific community to take hold of the subject.
3 Changes in the Yearbook Editorial Team
We are pleased to welcome Licong Cui, assistant professor at the University of Texas
Health Science Center (Houston, USA), as a new section editor of the KRM section.
Scott McGrath, program manager for CITRIS Health’s ACTIVATE program at the University
of California Berkeley (California, USA) and Mary Lauren Benton, assistant professor
in the Department of Computer Science at Baylor University (Texas, USA), are the new
editors of the BTI section. Anthony Solomonides, Program Director for Outcomes Research
and Biomedical Informatics at NorthShore University (Illinois, USA), is a new editor
for the HFOI section. For the CRI section, we welcome Xavier Tannier, professor at
Sorbonne Université (Paris, France).
On the departure side, Alexander Hörbst, professor at the Department Medical Technologies
of the Management Center Innsbruck (Austria), who served as an editor of the Clinical
Information Systems section from 2018 to 2022 is leaving the editorial team in 2023.
We thank him warmly for his valuable contribution to the IMIA Yearbook. We also thank
Damian Borbolla, assistant professor at the University of Utah (Utah, USA) and Tiago
Colicchio, assistant professor at the University of Alabama at Birmingham (Alabama,
USA) for their work on DS this year, and we are honored to welcome (back) Christoph
U. Lehmann, professor at University of Texas Southwestern, Dallas (Texas, USA) and
Vignesh Subbian, assistant professor at the University of Arizona, Tucson (Arizona,
USA) as new DS editors next year. Christel Daniel, who has been co-editor of the CRI
section since 2012, is leaving her position as co-editor and we thank her so much
for her valuable work for the past 10 years!
