Lack of regulations and quality management jeopardizes the quality and safety of LDTs, claim experts in clinical laboratory medicine in a commentary to Canadian policymakers
The IHPME members published their comments in the Canadian Medical Association Journal (CMAJ), a peer-reviewed journal owned by Joule Inc., a subsidiary of the Canadian Medical Association. In it, they claim “recent expansion of the molecular diagnostics industry has revealed weaknesses in Canada’s regulatory system for laboratory-developed tests, which are not subject to statutory regulations on medical devices.”
For pathologists and clinical laboratory professionals in both Canada and the United States, these recent actions show the concerns many experts have as they watch the explosive growth in the use of laboratory-developed tests in both countries. In many ways, the swift advances in molecular and genetic diagnostics is outrunning the ability of government regulators to keep pace with use of LDTs in clinical care settings.
In their commentary in CMAJ, the IHPME members also
claim the review and evaluation of LDTs in Canada is inconsistent. Some LDTs they
say, may endure stringent assessments and have endorsements by clinical
guidelines or findings that are published in scientific journals. Other LDTs,
however, may have no analysis at all.
In addition, the IHPME members point out that there is no
national registry kept of LDTs. They theorize that a lack of proper regulation,
controls, and quality management “has potentially jeopardized the delivery of
quality, safe, timely, and appropriate care.”
The researchers calling on Health Canada to address these
issues include:
Fiona A. Miller, PhD, Professor of Health Policy and IHPME Chair in Health Management Strategies;
François Rousseau, PhD, Professor, Department of Molecular Biology, Medical Biochemistry and Pathology, Faculty of Medicine, Laval University, Quebec;
Alberto Gutierrez, PhD, Partner, NDA Partners LLC, former Director, Office of In Vitro Diagnostics and Radiological Health at the FDA’s Center for Devices and Radiological Health (CDRH);
Stuart Hogarth, PhD, Lecturer in Sociology of Science and Technology, University of Cambridge, Cambridge, UK.
During an exclusive presentation offered by The Dark Report (Dark Daily’s sister publication) in 2015, Alberto Gutierrez, PhD (above), who at that time was Director, Office of In Vitro Diagnostics and Radiological Health at the FDA, said, “LDTs are an area that will be difficult to regulate. There is a broad set of tests. Some of the LDTs are very good. Some of them require a lot of expertise from the pathologists and some of them don’t. Regulating LDTs in a way that makes sense and that does not disrupt what’s going on [in clinical laboratories] is going to be difficult.” (Photo copyright: FDA.)
Canadian Scientists Call on Health Canada to Take the
Lead on Regulating LDTs
In the US, the FDA has been making moves to regulate LDTs since 2010, with much opposition from clinical laboratories and In Vitro Diagnostic (IVD) manufacturers. The FDA describes LDTs as internally designed clinical laboratory tests that are developed, manufactured, and used within a single laboratory. They have not undergone government regulatory review, can be simple or complex, and can be utilized to detect a variety of analytes.
Health Canada is the name of a department that falls under
the purview of the Minister of
Health and is part of Canada’s Health
Portfolio. It is responsible for helping Canadians maintain and improve
their health. Other agencies included in the Health Portfolio are:
According to the IHPME paper, however, Health Canada
currently does not have a way to regulate LDTs, and no government agency in
that country is responsible for the oversight of laboratory-developed tests.
Only LDTs that are marketed as test kits are evaluated and reviewed by Health
Canada.
“The current laboratory regulatory system in Canada involves a mixture of public and private entities and operates with oversight from provincial governments, nongovernmental organizations, and professional societies,” the IHPME paper states, adding, “most provinces and territories rely on voluntary standards that are unevenly applied, with little auditing and systematic testing to ensure quality.”
The authors also note that the current lab regulations in
Canada apply only to the operations of the medical laboratories themselves,
encompassing such things as lab environments, personnel, accreditation, and
quality control. They believe the loophole regarding LDTs needs to be addressed,
and they urged Health Canada to “demonstrate leadership” by subjecting these
tests to regulations that are currently applied to medical devices and
pharmaceuticals.
Other Countries Regulate LDTs, though Not Without
Controversy
In support of their call to action, IHPME researchers noted
that Australia, the EU, and the US all have taken steps to regulate LDTs.
The Australian government began oversight of LDTs in 2010 by
subjecting high-risk LDTs to external evaluation and then tracking them in a
public registry.
An EU regulation, which was passed in 2017, will administer
regulatory review of LDTs manufactured on an industrial scale, which targets
commercial laboratories. The law exempts LDTs utilized within individual
hospital laboratories and should be fully implemented by 2022.
Though on its radar since the 1990s, in 2010, the FDA officially
announced its intent to regulate LDTs in the US. The agency released an initial
draft approach for doing so starting in 2014, held a public workshop on the
topic in 2015, and released a
discussion paper in 2017. At this time, however, the FDA is not regulating
LDTs, though the agency remains open to the possibility.
Dark Daily
has reported extensively over the years on the development of LDTs and the
controversy surrounding the FDA’s moves to regulate them.
According to the FDA
website, problems with several high-risk LDTs have been identified,
including:
Claims that are not adequately supported with
evidence;
Lack of appropriate controls which may yield
erroneous results; and
The FDA’s report, titled, “The
Public Health Evidence for FDA Oversight of Laboratory Developed Tests,” reviewed
20 case studies of LDTs for Lyme disease, ovarian cancer, whooping cough,
fibromyalgia, prostate cancer, autism, breast cancer, melanoma, Vitamin D, and
other conditions. The agency concluded that in many instances “patients have
been demonstrably harmed or may have been harmed by tests that did not meet FDA
requirements.”
Klein noted, however, that “The 20 tests described by FDA are mostly a hodgepodge of outlier assays including tests that were never offered, tests for which comparable FDA assays perform poorly, tests for poorly defined disorders with psychologic components, and use of an FDA-approved test off-label.” He continued, “That FDA could find only these dubious examples out of the many thousands of laboratory-developed procedures (LDPs) that benefit patients each day, calls into question the agency’s rationale for expanding its regulatory scope to include LDPs.”
Perhaps this is why the FDA has yet to implement regulations
for LDTs. The controversy continues.
Whether Health Canada will accept the advice of the IHPME
scientists and take steps to regulate laboratory-developed tests in Canada remains
to be seen. As more LDTs are created and manufactured, however, it is probable
that governments will continue to evaluate the administration and oversight of laboratory-developed
tests.
In both Canada and the United States, pathologists, clinical
laboratory managers, and executives at in vitro diagnostic manufacturers
can expect an ongoing tug-of-war between government regulators and the lab
industry over the most appropriate ways to regulate LDTs.
‘Prime editing’ is what researchers are calling the proof-of-concept research that promises improved diagnostics and more effective treatments for patients with genetic defects
Known as Prime Editing, the scientists developed this technique as a more accurate way to edit Deoxyribonucleic acid (DNA). In a paper published in Nature, the authors claim prime editing has the potential to correct up to 89% of disease-causing genetic variations. They also claim prime editing is more powerful, precise, and flexible than CRISPR.
The research paper describes prime editing as a “versatile and precise genome editing method that directly writes new genetic information into a specified DNA site using a catalytically impaired Cas9endonuclease fused to an engineered reverse transcriptase, programmed with a prime editing guide RNA (pegRNA) that both specifies the target site and encodes the desired edit.”
And a Harvard Gazette article states, “Prime editing differs from previous genome-editing systems in that it uses RNA to direct the insertion of new DNA sequences in human cells.”
Assuming further research and clinical studies confirm the
viability of this technology, clinical laboratories would have a new diagnostic
service line that could become a significant proportion of a lab’s specimen
volume and test mix.
In that e-briefing we wrote that Liu “has led a team of scientists in the development of a gene-editing protein delivery system that uses cationic lipids and works on animal and human cells. The new delivery method is as effective as protein delivery via DNA and has significantly higher specificity. If developed, this technology could open the door to routine use of genome analysis, worked up by the clinical laboratory, as one element in therapeutic decision-making.”
Now, Liu has taken that development even further.
“A major aspiration in the molecular life sciences is the ability to precisely make any change to the genome in any location. We think prime editing brings us closer to that goal,” David Liu, PhD (above), Director of the Merkin Institute of Transformative Technologies in Healthcare at the Broad Institute, told The Harvard Gazette. “We’re not aware of another editing technology in mammalian cells that offers this level of versatility and precision with so few byproducts.” (Photo copyright: Broad Institute.)
Cell Division Not Necessary
CRISPR stands for Clustered Regularly Interspaced Short Palindromic Repeats. It is considered the most advanced gene editing technology available. However, it has one drawback not found in Prime Editing—CRISPR relies on a cell’s ability to divide to generate desired alterations in DNA—prime editing does not.
This means prime editing could be used to repair genetic mutations in cells that do not always divide, such as cells in the human nervous system. Another advantage of prime editing is that it does not cut both strands of the DNA double helix. This lowers the risk of making unintended, potentially dangerous changes to a patient’s DNA.
The researchers claim prime editing can eradicate long lengths of disease-causing DNA and insert curative DNA to repair dangerous mutations. These feats, they say, can be accomplished without triggering genome responses introduced by other forms of CRISPR that may be potentially harmful.
“Prime editors are more like word processors capable of
searching for targeted DNA sequences and precisely replacing them with edited
DNA strands,” Liu told NPR.
The scientists involved in the study have used prime editing to perform over 175 edits in human cells. In the test lab, they have succeeded in repairing genetic mutations that cause both Sickle Cell Anemia (SCA) and Tay-Sachs disease, NPR reported.
“Prime editing is really a step—and potentially a significant step—towards this long-term aspiration of the field in which we are trying to be able to make just about any kind of DNA change that anyone wants at just about any site in the human genome,” Liu told News Medical.
Additional Research Required, but Results are Promising
Prime editing is very new and warrants further
investigation. The researchers plan to continue their work on the technology by
performing additional testing and exploring delivery mechanisms that could lead
to human therapeutic applications.
“Prime editing should be tested and optimized in as many cell types as researchers are interested in editing. Our initial study showed prime editing in four human cancer cell lines, as well as in post-mitotic primary mouse cortical neurons,” Liu told STAT. “The efficiency of prime editing varied quite a bit across these cell types, so illuminating the cell-type and cell-state determinants of prime editing outcomes is one focus of our current efforts.”
Although further research and clinical studies are needed to
confirm the viability of prime editing, clinical laboratories could benefit
from this technology. It’s worth watching.
Since Alexa is now programed to be compliant with HIPAA privacy rules, it’s likely similar voice assistance technologies will soon become available in US healthcare as well
Shortages of physicians and other types of caregivers—including
histopathologists
and pathology
laboratory workers—in the United Kingdom (UK) has the UK’s National Health Service (NHS) seeking alternate
ways to get patients needed health and medical information. This has prompted a
partnership with Amazon to use the Alexa virtual assistant to
answer patients healthcare inquiries.
Here in the United States, pathologists and clinical
laboratory executives should take the time to understand this development.
The fact that the NHS is willing to use a device like Alexa to help it maintain
access to services expected by patients in the United Kingdom shows how rapidly
the concept of “virtual clinical care” is moving to become mainstream.
If the NHS can make it work in a health system serving 66-million
people, it can be expected that health insurers, hospitals, and physicians in
the United States will follow that example and deploy similar virtual health
services to their patients.
For these reasons, all clinical laboratories and anatomic
pathology groups will want to develop a strategy as to how their
organizations will interact with virtual health services and how their labs
will want to deploy similar virtual patient information services.
Critical Shortages in Healthcare Services
While virtual assistants have
been answering commonly-asked health questions by mining popular responses on
the Internet for some time, this new agreement allows Alexa to provide
government-endorsed medical advice drawn from the NHS website.
By doing this, the NHS hopes to reduce the burden on
healthcare workers by making it easier for UK patients to access health
information and receive answers to commonly-asked health questions directly from
their homes, GeekWire
reported.
“The public needs to be able to get reliable information
about their health easily and in ways they actually use. By working closely
with Amazon and other tech companies, big and small, we can ensure that the
millions of users looking for health information every day can get simple,
validated advice at the touch of a button or voice command,” Matthew Gould, CEO of NHSX, a division of the NHS that focuses
on digital initiatives, told GeekWire.
The
Verge reported that when the British government officially announced
the partnership in a July press
release, the sample questions that Alexa could answer included:
Alexa, how do I treat a migraine?
Alexa, what are the symptoms of the flu?
Alexa, what are the symptoms of chickenpox?
“We want to empower every patient to take better control of
their healthcare and technology like this is a great example of how people can
access reliable, world-leading NHS advice from the comfort of their home,
reducing the pressure on our hardworking GPs (General Practitioners) and
pharmacists,” said Matt
Hancock, Secretary of State for Health and Social Care, in the press release.
MD
Connect notes that the NHS provides healthcare services free of charge to
more than 66-million individuals residing in the UK. With 1.2 million
employees, the NHS is the largest employer in Europe, according to The
Economist. That article also stated that the biggest problem facing the
NHS is a staff shortage, citing research conducted by three independent
organizations:
Their findings indicate “that NHS hospitals, mental-health
providers, and community services have 100,000 vacancies, and that there are
another 110,000 gaps in adult social care. If things stay on their current
trajectory, the think-tanks predict that there will be 250,000 NHS vacancies in
a decade,” The Economist reported.
UK’s Matt Hancock, Secretary of State for Health and Social Care (above), defends the NHS’ partnership with Amazon Alexa, saying millions already use the smart speaker for medical advice and it’s important the health service uses the “best of modern technology.” Click here to watch the video. (Video and caption copyright: Sky News.)
“This idea is certainly interesting and it has the potential
to help some patients work out what kind of care they need before considering
whether to seek face-to-face medical help, especially for minor ailments that
rarely need a GP appointment, such as coughs and colds that can be safely
treated at home,” Professor
Helen Stokes-Lampard, Chairman at the Royal
College of General Practitioners, and Chair of the Board Of
Directors/Trustees at National
Academy of Social Prescribing, told Sky News.
“However,” she continued, “it is vital that independent
research is done to ensure that the advice given is safe, otherwise it could
prevent people seeking proper medical help and create even more pressure on our
overstretched GP service.”
Amazon has assured consumers that all data obtained by Alexa
through the NHS partnership will be encrypted to ensure privacy and security,
MD Connect notes. Amazon also promised that the personal information will not
be shared or sold to third parties.
Alexa Now HIPAA Compliant in the US
This new agreement with the UK follows the announcement in April
of a new Alexa
Skills Kit that “enables select Covered Entities and their Business
Associates, subject to the US Health
Insurance Portability and Accountability Act of 1996 (HIPAA), to build
Alexa skills that transmit and receive protected
health information (PHI) as part of an invite-only program. Six new Alexa
healthcare skills from industry-leading healthcare providers, payors, pharmacy
benefit managers, and digital health coaching companies are now operating in
our HIPAA-eligible environment.”
Developers of voice assistance technologies can freely use
these Alexa skills, which are “designed to help customers manage a variety of
healthcare needs at home simply using voice—whether it’s booking a medical
appointment, accessing hospital post-discharge instructions, checking on the
status of a prescription delivery, and more,” an Amazon
Developer Alexa blog states.
The blog lists the HIPAA-compliant Alexa skills as:
Express
Scripts: Members can check the status of a home delivery prescription and can
request Alexa notifications when their prescription orders are shipped.
Cigna
Health Today by Cigna (NYSE:CI): Eligible employees with one of Cigna’s
large national accounts can now manage their health improvement goals and
increase opportunities for earning personalized wellness incentives.
Swedish
Health Connect by Providence St.
Joseph Health, a healthcare system with 51 hospitals across seven states
and 829 clinics: Customers can find an urgent care center near them and
schedule a same-day appointment.
Atrium
Health, a healthcare system with more than 40 hospitals and 900 care
locations throughout North and South Carolina and Georgia: Customers in North
and South Carolina can find an urgent care location near them and schedule a
same-day appointment.
Livongo,
a digital health company that creates new and different experiences for people
with chronic conditions: Members can query their last blood sugar reading,
blood sugar measurement trends, and receive insights and Health Nudges that are
personalized to them.
HIPAA Journal notes: “This is not the first time that Alexa skills have been developed, but a stumbling block has been the requirements of HIPAA Privacy Rules, which limit the use of voice technology with protected health information. Now, thanks to HIPAA compliant data transfers, the voice assistant can be used by a select group of healthcare organizations to communicate PHI without violating the HIPAA Privacy Rule.”
Steady increases associated with the costs of medical care
combined with a shortage of healthcare professionals on both continents are
driving trends that motivate government health programs and providers to
experiment with non-traditional ways to interact with patients.
New digital and Artificial
Intelligence (AI) tools like Alexa may continue to emerge as methods for
providing care—including clinical laboratory and pathology advice—to healthcare
consumers.
When patients use telehealth, how do they choose medical laboratories for lab test orders their virtual doctors have authorized?
Doctors On Demand is expanding the nation’s primary care services by launching a virtual care telehealth platform for health insurers and employers. This fits into a growing nationwide trend toward increased use of remote and virtual doctor’s visits. But how should clinical laboratories and anatomic pathology groups prepare for fulfilling virtual doctors’ lab test orders in ways consistent with current scope-of-practice laws?
The rise of virtual care is made possible by innovations in digital
and telecommunication technology. Driven by studies showing more patients are
opting out of conventional primary care visits that take too much time or are
too far away, the healthcare industry is responding by bringing medical
services—including pathology and clinical laboratory—closer to patients through
retail settings and urgent care clinics.
Many pathologists and clinical laboratory managers are unaware of how swiftly patients are becoming comfortable with getting their primary care needs met by other types of caregivers, including virtually. Recently, the Health Care Cost Institute (HCCI) published data showing that visits to primary care physicians declined 18% from 2012 to 2016 among adults under 65 who had employer-sponsored insurance. However, during these same years, visits with nurse practitioners and physician’s assistants increased by 129%!
Another way that providers are making it easier for patients to access healthcare is through the Internet.
Doctor On Demand, a San Francisco-based virtual care provider, is targeting insurers and employers with its Synapse telehealth platform, which integrates into existing health plan networks and enables virtual primary care, according to a news release.
“Through our fully integrated technology platform, we’re putting the patient first and introducing continuity of care not previously available through virtual care solutions,” said Hill Ferguson, CEO of Doctor On Demand in a statement announcing the launch of Synapse on the Humana (NYSE:HUM) health plan network. (Photo copyright: The Business Journals.)
How Synapse Works
Humana is using Synapse in its new On Hand virtual primary
care plan, the news release states. Humana said its members have no copay for
the virtual doctor visits and $5 copays for standard medical laboratory tests
and prescriptions. Synapse’s “smart referrals” function sends referrals to
in-network clinical laboratories, imaging providers, and pharmacies, Healthcare
Dive reported.
“Humana has a deep footprint, and this is a payer looking to create a virtual primary care network as a way to contain cost and thinking about how care is coordinated and delivered,” Josh Berlin, a Principal and Healthcare Co-Practice Leader with advisory firm Citrin Cooperman, told FierceHealthcare.
Changing Primary Care Relationships
Another insurer advancing telehealth is Oscar Health, which offers its own Doctor on Call telehealth platform. The New York City-based health plan reported in a year-end review that 82% of its members had set up a profile that gave them access to a concierge care team and 24/7 telemedicine services, including clinical laboratory test results.
During 2018, Oscar’s concierge teams addressed 1.2 million
questions from 77% of its members, the insurer said.
The graphic above, taken from research conducted by the Health Care Cost Institute (HCCI), shows that while virtual primary care has been expanding, conventional visits to primary care physicians fell 18% from 2012 to 2016 among adults under 65 who had employer-sponsored insurance. Simultaneously, visits with nurse practitioners and physician’s assistants increased by 129%! This indicates a shift in how patients view access to primary care physicians and may explain why telehealth is becoming an attractive option. How will clinical laboratories fit into this new healthcare paradigm? (Photo copyright: HCCI.)
Becker’s Hospital Review reports that telehealth usage by Oscar’s members is five times higher than the average for the healthcare industry.
Will Clinical Laboratories Receive Virtual Referrals?
In a way, it has never been easier for patients to see a
primary care doctor or research symptoms. Additionally, the Internet makes it
possible for patients to self-diagnose, though not always to the benefit of
healthcare providers or the patients.
So, how should clinical laboratories respond to this growing expansion of virtual care doctors? Experts advise lab leaders to reach out to health plans soon and determine their inclusion in virtual healthcare networks. Labs also may benefit by making test scheduling and reporting accessible and convenient to insurance company members and consumers choosing telehealth.
During his keynote presentation at the 24th Annual Executive War College in May, Ted Schwab, a Los Angeles area Healthcare Strategist and Entrepreneur, said, “If you use Google in the United States to check symptoms, you’ll find 350 different electronic applications that will give you medical advice—meaning you’ll get a diagnosis over the Internet. These applications are winding their way somewhere through the regulatory process. (See Schwab’s expanded comments on this trend in, “Strategist Explains Key Trends in Healthcare’s Transformation,” The Dark Report, October 14, 2019.)
Schwab advises that in this “time of change” it’s critical
for labs to take proactive measures. “What we know today is that
providers—including clinical laboratories and pathology groups—who do nothing
will get trampled. However, those providers that do something proactively will
most likely be the winners as healthcare continues to transform.”
With $191 million in startup capital, the genomics startup will draw on existing genetic databases to create personalized medicine therapies for chronic diseases
Why do some people get sick while others do not? That’s what genetic researchers at Maze Therapeutics want to find out. They have developed a new approach to using tools such as CRISPR gene editing to identify and manipulate proteins in genetic code that may be the key to providing personalized protection against specific diseases.
If viable, the results of Maze’s research could mean the development of specific drugs designed to mimic genetic code in a way that is uniquely therapeutic to specific patients. This also would create the need for clinical laboratories to sequence and analyze patients’ DNA to determine whether a patient would be a candidate for any new therapies that come from this line of research.
Based in San Francisco, Maze Therapeutics (Maze) is studying modifier genes—genes that affect the phenotype or physical properties of other genes—and attempting to create drugs that replicate them, reported MIT Technology Review. Maze believes that genetic modifiers could afford a “natural form of protection” against disease.
“If you have a disease-causing gene, and I have the disease-causing gene, why is it that you may be healthy and I may be sick? Are there other genes that come into play that provide a protective effect? Is there a drugging strategy to recover normal phenotype and recover from the illness?” Maze Chief Executive Officer Jason Coloma, PhD, asked in an interview with FierceBiotech.
In 2019, Maze received $191 million in financing from Third Rock Ventures, ARCH Venture Partners, and others, to find ways to translate their findings into personalized medicines, according to a news release. And with the availability of international public genetic databases and CRISPR gene editing, now may be good timing.
“This was the perfect time to get into this space with the tools that were being developed and the amount of data that has been accumulated on the human genetic side,” Charles Homcy, MD, Third Rock Ventures Partner and Maze Scientific Founder, told Forbes, which noted that Maze is tapping existing population-wide genetic databases and large-scale studies, including the United Kingdom’s Biobank and Finland’s Finngen.
To help find genetic modifier drug targets, Maze is accessing CRISPR gene editing capabilities. Jonathan Weissman, PhD, Maze Scientific Founder and Professor of Cellular Molecular Pharmacology at University of California, San Francisco (UCSF), told MIT Technology Review: “You take a cell with a disease-causing gene and then see if you can turn it back to normal. We can do 100,000 experiments at once because each cell is its own experiment.”
“At Maze, we are focused on expanding our understanding of the natural disease protection provided by genetic modifiers through an integrated approach that combines studying natural human genetic variation across the globe and conducting large-scale experiments of gene perturbations,” Charles Homcy, MD (above), Founder and interim CEO of Maze and a partner at Third Rock Ventures, said in a news release. “Through our integrated approach, we believe we will create novel medicines based around those modifiers to treat a number of diseases.” (Photo copyright: Forbes.)
Using CRISPR to Identify the Cause of Disease
One drug research program reportedly progressing at Maze involves developing gene therapy for the neurogenerative disease amyotrophic lateral sclerosis (ALS). The program borrows from previous research conducted by Aaron Gitler, PhD, Professor of Genetics at Stanford University and Maze co-founder, which used CRISPR to find genetic modifiers of ALS. The scientists found that when they removed the protein coding gene TMX2 (Thioredoxin Related Transmembrane Protein 2), the toxicity of proteins building the disease was reduced, reported Chemical and Engineering News.
“We used the CRISPR-Cas9 system to perform genome-wide gene-knockout screens for suppressors and enhancers of C9ORF72 DPR toxicity in human cells,” Gitler and colleagues wrote in Nature Genetics. “Together, our results demonstrate the promise of using CRISPR-Cas9 screens in defining the mechanisms of neurodegenerative diseases.”
“We have the flexibility to think differently. We like to
think of ourselves as part of this new breed of biotech companies,” Coloma told
FierceBiotech.
It’s an exciting time. Clinical laboratories can look
forward to new precision medicine diagnostic tests to detect disease and
monitor the effects of patient therapies. And the research initiatives by Maze
and other genetic companies represent a new approach in the use of genetic code
to create specific drug therapies targeted at specific diseases that work best
for specific patients.
The companion diagnostics that may come from this research would
be a boon to anatomic pathology.
Clinical laboratories working with AI should be aware of ethical challenges being pointed out by industry experts and legal authorities
Experts are voicing concerns that using artificial
intelligence (AI) in healthcare could present ethical challenges that need
to be addressed. They say databases and algorithms may introduce bias into the
diagnostic process, and that AI may not perform as intended, posing a potential
for patient harm.
If true, the issues raised by these experts would have major
implications for how clinical
laboratories and anatomic
pathology groups might use artificial intelligence. For that reason,
medical laboratory executives and pathologists should be aware of possible
drawbacks to the use of AI and machine-learning
algorithms in the diagnostic process.
Is AI Underperforming?
AI’s ability to improve diagnoses, precisely target
therapies, and leverage healthcare data is predicted to be a boon to precision medicine and personalized
healthcare.
For example, Accenture
(NYSE:ACN) says that hospitals will spend $6.6 billion on AI by 2021. This
represents an annual growth rate of 40%, according
to a report from the Dublin, Ireland-based consulting firm, which states,
“when combined, key clinical health AI applications can potentially create $150
billion in annual savings for the United States healthcare economy by 2026.”
But are healthcare providers too quick to adopt AI?
Accenture defines AI as a “constellation of technologies
from machine learning to natural
language processing that allows machines to sense, comprehend, act, and
learn.” However, some experts say AI is not performing as intended, and that it
introduces biases in healthcare worthy of investigation.
Keith Dreyer, DO, PhD, is Chief Data Science Officer at Partners Healthcare and Vice Chairman of Radiology at Massachusetts General Hospital (MGH). At a World Medical Innovation Forum on Artificial Intelligence covered by HealthITAnalytics, he said, “There are currently no measures to indicate that a result is biased or how much it might be biased. We need to explain the dataset these answers came from, how accurate we can expect them to be, where they work, and where they don’t work. When a number comes back, what does it really mean? What’s the difference between a seven and an eight or a two?” (Photo copyright: Healthcare in Europe.)
What Goes in Limits What Comes Out
Could machine learning lead to machine decision-making that
puts patients at risk? Some legal authorities say yes. Especially when computer
algorithms are based on limited data sources and questionable methods, lawyers
warn.
How can AI provide accurate medical insights for people when
the information going into databases is limited in the first place? Ossorio
pointed to lack of diversity in genomic
data. “There are still large groups of people for whom we have almost no
genomic data. This is another way in which the datasets that you might use to
train your algorithms are going to exclude certain groups of people
altogether,” she told HDM.
She also sounded the alarm about making decisions about
women’s health when data driving them are based on studies where women have
been “under-treated compared with men.”
“This leads to poor treatment, and that’s going to be
reflected in essentially all healthcare data that people are using when they
train their algorithms,” Ossorio said during a Machine Learning for Healthcare (MLHC) conference
covered by HDM.
How Bias Happens
Bias can enter healthcare data in three forms: by humans, by
design, and in its usage. That’s according to David Magnus, PhD, Director
of the Stanford Center for
Biomedical Ethics (SCBE) and Senior Author of a paper published in the New England
Journal of Medicine (NEJM) titled, “Implementing Machine
Learning in Health Care—Addressing Ethical Challenges.”
The paper’s authors wrote, “Physician-researchers are
predicting that familiarity with machine-learning tools for analyzing big data
will be a fundamental requirement for the next generation of physicians and
that algorithms might soon rival or replace physicians in fields that involve
close scrutiny of images, such as radiology and anatomical pathology.”
In a news
release, Magnus said, “You can easily imagine that the algorithms being
built into the healthcare system might be reflective of different, conflicting
interests. What if the algorithm is designed around the goal of making money?
What if different treatment decisions about patients are made depending on
insurance status or their ability to pay?”
In addition to the possibility of algorithm bias, the
authors of the NEJM paper have other concerns about AI affecting
healthcare providers:
“Physicians must adequately understand how
algorithms are created, critically assess the source of the data used to create
the statistical models designed to predict outcomes, understand how the models
function and guard against becoming overly dependent on them.
“Data gathered about patient health, diagnostics,
and outcomes become part of the ‘collective knowledge’ of published literature
and information collected by healthcare systems and might be used without
regard for clinical experience and the human aspect of patient care.
“Machine-learning-based clinical guidance may
introduce a third-party ‘actor’ into the physician-patient relationship, challenging
the dynamics of responsibility in the relationship and the expectation of
confidentiality.”
“We need to be cautious about caring for people based on what algorithms are showing us. The one thing people can do that machines can’t do is step aside from our ideas and evaluate them critically,” said Danton Char, MD, Lead Author and Assistant Professor of Anesthesiology, Perioperative, and Pain Medicine at Stanford, in the news release. “I think society has become very breathless in looking for quick answers,” he added. (Photo copyright: Stanford Medicine.)
Acknowledge Healthcare’s Differences
Still, the Stanford researchers acknowledge that AI can
benefit patients. And that healthcare leaders can learn from other industries,
such as car companies, which have test driven AI.
“Artificial intelligence will be pervasive in healthcare in a
few years,” said
Nigam Shah, PhD, co-author of the NEJM paper and Associate Professor of Medicine at Stanford, in the news release. He added that healthcare leaders need to be aware of the “pitfalls” that have happened in other industries and be cognizant of data.
“Be careful about knowing the data from which you learn,” he
warned.
AI’s ultimate role in healthcare diagnostics is not yet fully
known. Nevertheless, it behooves clinical laboratory leaders and anatomic
pathologists who are considering using AI to address issues of quality and
accuracy of the lab data they are generating. And to be aware of potential
biases in the data collection process.
