Case Study Readings
DePuy ASR Metal on Metal Hip Implant
Orthopedic Medical Devices: Ethical Questions, Implant Recalls and Responsibility
hip replacement is a surgical procedure to replace the femoral head and acetabulum with prosthetic implants to improve function, increase mobility, and relieve pain
several recalls of recently poorly functioning implants
Why did the Articular Surface Replacement Implant Fail?
a metal-on-metal total hip replacement consists of a metal socket and a metal femoral head
Johnson and Johnson under a lot of fire due to a failed Articular Surface Replacement
The grove on the inside of the socket limits surface area and increases friction and wear
Limited surface area causes the head to but against the cup’s edge during movement
acetabular cup (socket) and femoral head are too small further increasing fictions
Low tolerance between socket and ball
deformation of the lic ions
causes local inflammation and erosions to the bone resulting in implant loosening, dislocation, cracking and popping sensations, surrounding tissue masses, pseudotumor and areas of frank necrosis
loosening of the components requires revision surgery that is complicated and expensive
various systemic reactions also may occur
Responsibilities of Johnson and Johnson
flaws were never disclosed to surgeons
J&J did not disclose the FDA did not allow one version to be sold in the US
2010: DePuy/J&J found that it had used incorrect of inadequate standards for assessing implant performance before the sale
The company only tested the in-vitro performance of the lab equipment at one angle of implantation, missing a lot of the other angles
Usually implants can be customized to the patient’s body, not true for the DePuy implant
The implant cannot be feasibly applied for every patient
Responsibilities of the Medical Community
Implant design is Multidisciplinary undertaking with input from orthopedic surgeons, bioengineers, and materials scientists
Surgeons must dedicate themselves to the patient first
Federal Regulation of Medical Devices
Safe Medical Devices Act of 1990 requires physicians who utilize devices to report associated serious injuries either to the manufacturer or to the FDA
FDA can impose legal sanctions and fines however it relies on the goodwill and cooperation
Conclusion
Evidence based medicine is increasingly important
More transparency is required
Translational medicine must be more closely observed
High level of evidence must be asked next time
Precaution, governance, and failure of medical implants: THE ASR hip in the UK
Introduction
High revision rates have led to the need to explant and replace the artificial hip due to potential tissue damage
For most, it is a positive change that leads to improvement in quality of life
The challenge has been to develop artificial joins which can withstand higher levels of activity over a greater number of years
Success to failure: an introductory history of the ASR case
ASR Hip and ASR XL was the two major versions that caused failure
Positioning proved to be tricky
high revision rates
the failure of the hips meant more pain, decreased mobility, their ability to work, self-esteem and mental health issues, and their social and family lives
DePuy’s first response was to place the blame on the surgeons
Conjecture among patients on how ASR might’ve unfolded if DePuy withdrew earlier
DePuy has never accepted that there has been a defect with the ASR hip.
MHRA called it out
more after it failed than protecting patient safety
EAG might’ve been able to operate
NJR collects information on joint replacement implants and surgery
made expliciting link between high revision rate and subsequent actions taken by the MHRA to remove poorly performing implants from the market
They claim to identify and remove poor implants quickly and efficiently but not the case for the ASR
surgeons draw on guidance from the British Orthopaedic Association
British Hip Society
Neither body published formal guidance or recommendations prior to device failure and withdrawal
Recognizes that they’re wrong
Surgeons opted to implant alternative hips
Ethical deliberations are very important
People should’ve acted sooner
MHRA should refocus their attention on patients and prioritise their relationship with them over others
Contemporary development and deployment of new tech takes place in a wider social spaces
Values life too cheaply
Lack of clinical trials and transparency in in-vitro testing means most data is from when its in the market
Policy Commerce Alignment
think of precaution as aiding innovation not being scientific
commercial companies need to reduce their contribution to the development of regulatory and testing regimes
post-market surveillance needs to be increased
Key Terms:
Arthoplasty: The surgical repair or replacement of a joing
Acetabulum: the cup shaped anatomical socket in the pelvis where the head of the femur fits
Metal on Metal Bearing: A type of joint prosthesis design where both the femoral head and acetabular ball (socket) are designed from metal alloys
Revision Surgery: A complex surgical procedure to replace, repair, or remove a previously implanted medical device that has failed, worn out, or caused adverse health complications.
In-Vitro Testing: mechanical or biological testing in the lab vs. in a living organism
Substantial Equivalence (510(k) pathway): A fast-track regulatory clearance pathway used by agencies like the FDA that allows a medical device to be cleared without clinical trials if it’s substantially similar to another device in the market.
Precautionary Principle: an ethical and policy making guideline stating that when an activity or technology poses a risk of serious harm to public health or safety, protective action should be taken even if scientific evidence of the harm is not yet fully conclusive.
Beyond Compliance: A voluntary safety and monitoring initiative established in the UK that tracks the introduction and performance of new orthopedic implants in a controlled step wise manner to proactively ensure patient safety.
ImpactPro- AI Bias in Healthcare Algorithms
AI Bias in Healthcare: Using ImpactPro as a Case Study for Healthcare Practitioners’ Duties to Engage in Anti-Bias Measures
The beliefs and prejudices of developers can affect the algorithm
ImpactPro found that New York’s United Health Services failed to recommend black patients to a complex health needs program at the same rate as white patients
healthcare practitioners must engage in a variety of anti-bias measures
implicit bias training
education on bias in medicine
perspective taking
The ImpactPro Study
Main goal of using the algorithm was to better serve patients
also to reduce costs
Healthcare costs was used as a proxy for healthcare needs
No mention was made in regards to race
Black patients generated different healthcare costs than white patients
emergency vs surgical costs
can be explained through larger societal factors systemically
Black patients spend less on their healthcare and their algorithm did not identify Black patients as being sick because they believed that black patients would spend less
Not making decisions in a vacuum, also based on income
Bias in Healthcare
difficult to eradicate bias in individuals and in AI
healthcare providers should have in regard to promote equity when using AI systems
physicians in the West have strong pro-White biases
also true for Native Americans, with a history of mistrust of healthcare professionals (forced sterilization, etc.)
Best Practices for Elimination of Bias
Increase awareness of susceptibility to implicit bias through education
individuating, make a conscious effort to focus on the individual rather than their social category
perspective taking (empathy)
increasing diversity in the population of healthcare providers
Racial Bias in the ImpactPro Healthcare Algorithm
Clinical and Technical Background
AI and ML being used to parse massive electronic health records, diagnostic imaging, and clinical trial enrollments
used to optimize patient care trajectories and reduce medical errors
identify and better manage patients with complex high-risk health needs
major goal is to improve clinical outcomes
secondary was to reduce cost
Core Engineering and Ethical Failures
financial cost was directly equivalent to health severity and failed to account for the socioeconomic and systemic racial disparities in the US healthcare system
Black patients generate lower health expenditures and cost-intensive procedures than white patients
thus it failed as a safety net
Subsequent Actions and Policy Outcomes
2019: landmark study dissected racial bias
created a revised algorithm
replaced healthcare costs with an objective, quantified criterion of health status
became a cornerstone case study in medical Ai ethics
Key Terms:
Machine Learning (ML): A subfield of artificial intelligence focused on building algorithms that can proccess massive historical datasets, identify complex patterns, and make automated predictions or decisions without being explicity programed
Labeling Bias: A systemic error in machine learning where the chosen target outcome that the
model is trained to predict fails to represent the actual real-world concept (e.g., training a model to predict
healthcare spending rather than actual health severity).
Proxy Variable: An easily quantifiable metric used as a substitute for an unobservable, complex, or hard-to- measure concept
Implicit Bias: Unconsciously held attitudes, stereotypes, or negative associations toward specific groups of people that influence an individual behavior, judgment, and decision-making without their conscious awareness.
Individuating: A cognitive anti-bias strategy where a healthcare provider makes a conscious, deliberate effort to focus on a patient specific, personal clinical data rather than relying on social category stereotypes.
Perspective-Taking: An empathetic cognitive strategy where a provider actively imagines a patient feelings and experiences from their point of view to reduce bias and build clinical trust.
Montreal Declaration for the Responsible Development of AI: An international framework of ethical principles and guidelines designed to ensure that artificial intelligence systems promote well-being, equity, autonomy, public responsibility, and human agency.
High-Level Expert Group (HLEG) AI Guidelines: An ethics framework established by the European Commission outlining requirements for Trustworthy AI, emphasizing human oversight, technical robustness, transparency, fairness, and accountability.