By Hugo Caicedo
Traditional medical practice is rooted in advanced knowledge of diseases, their most appropriate treatment, and adequate proficiency in its applied practice. Notably, today, medical treatment does not typically occur until disease symptoms have manifested. While we now have ways to develop therapies that can halt the progression of some symptomatic diseases, symptomatic solutions are not meant to serve as a cure of disease but palliative treatment of late-stage chronic diseases.
The reactive approach in most medical interventions is magnified in that medicine is prone to errors. In November of 1999, the U.S. National Academy of Science, an organization representing the most highly regarded scientists and physician researchers in the U.S., published the report To Err is Human.
The manuscript noted that medical error was a leading cause of patient deaths killing up to 98,000 people in the U.S. every year. One hypothesis that came up was that patient data was being poorly collected, aggregated, and shared among different hospitals and even within the same health system. Health policies such the Health Information Technology for Economic and Clinical Health Act (HITECH) in 2009 and the Affordable Care Act (ACA) in 2010, primarily focused on optimizing clinical and operational effectiveness through the use of health information technology and expansion of government insurance programs, respectively. However, they did not effectively address the issue of medical errors such as poor judgment, mistaken diagnoses, inadequately coordinated care, and incompetent skill that can directly result in patient harm and death.
According to a study published in 2016 by researchers at Johns Hopkins University School of Medicine, medical error is no longer the leading cause of death in the U.S. However, it still ranks up as the third leading cause of death, after heart disease and cancer, causing at least 250,000 deaths every year.
That’s a figure that the authors of the study acknowledge represents an undercount, because they were unable to capture data from deaths in outpatient clinics and nursing homes. In fact, another study reported the number of deaths from medical error to be over 400,000.
Regardless of what the most accurate estimate is, those are big numbers that remind us how inefficient healthcare still is and how patient deaths due to medical error is an under-recognized and under-addressed epidemic in the U.S.
A revealing finding mentioned in the report is that medical error is not included on death certificates or in rankings of cause of death. Every year, the Centers for Disease Control and Prevention (CDC), using death certificates data, compiles the list of the most common causes of death in the U.S. Therefore, in the U.S., there is no pragmatic effort to raise public awareness and inform national research priorities to study medical errors and implement effective safeguards for patients.
The 21st Century Cures Act (which was passed in 2016) encourages healthcare organizations to use real world evidence (RWE) to help support the FDA approval process of new therapies or new indications on existing drugs. However, that RWE does not account for the systematic and comprehensive recording of physician and hospital malpractice. The idea that medical personnel start recording medical errors trustworthily, or without worry for malpractice liability is not realistic.
What is needed instead is proactive, federal-level legislation that provides legal protection to caregivers against lawsuits, the same type as the state-level legal protections that already exists when physicians report morbidity and mortality minutes within their institutions, a.k.a. medical peer review privilege.
Notably, those minutes are not discoverable in a court of law and cannot be used by plaintiff’s counsel in a malpractice action. However, in the context of federal court, a peer review privilege produced by state law does not always apply because the federal rules of evidence only recognize privileges incorporated in the federal common law and privileges produced by federal law.
The Patient Safety and Quality Improvement Act of 2005 tried to address those shortcomings but it has been limited in scope. It is not an error-reporting system per se. It safeguards documents submitted into an organization’s Patient Safety Evaluation System (PSES) followed by submission to a recognized Patient Safety Organization (PSO). If an organization fails to follow these steps, courts will not allow the organization to make a formal request of the medical peer review privilege.
Medical error is an issue that cannot just be addressed with health information technology and artificial intelligence (AI). HITECH has enabled massive development of electronic medical records and advanced predictive analytics to implement clinical decision support systems that better inform medical interventions throughout the full cycle of care. However, as noted before, medical errors are not documented in electronic medical records. The solution is not to replace medical personnel with AI-powered machines.
Despite ethical concerns, and even though there has been unprecedented progress in the fields of AI and machine learning, current artificial neuronal networks applications in healthcare are highly specialized and perform very specific assignments such as pattern recognition and classification tasks. In contrast, the human brain is capable of performing more general and complex tasks with a great deal of self-awareness. Moreover, recent news have shown that one of the most advanced healthcare AI assistants have delivered unsafe and inaccurate recommendations for treating cancer.
To be clear, lowering high healthcare costs while delivering superior value to patients and health systems is not just about changing reimbursement models and engaging in digital transformation, but also about changing the standards on which senior management and policy makers make fundamental decisions that influence the value chain throughout the entire healthcare ecosystem and not just their organization.
This is not the traditional change management, but a cultural management transformation that operates with patient centricity and digital economics in mind as opposed to zero-sum economics.