Home > Work > Reinventing Clinical Decision Support: Data Analytics, Artificial Intelligence, and Diagnostic Reasoning
1 " There was a time when the public had an unquestionable faith in biomedicine and the practitioners who translated it into everyday patient care—and physicians believed that the public's trust was justified based on their educational qualifications and training. But today, many patients believe that individual clinicians must earn their trust, just as a close relative has earned it through shared experience....Gallop polling over the last several decades that demonstrates how much the public's confidence in most US institutions has deteriorated. Confidence in the medical system in particular fell from 80% in 1975 to 37% in 2015. Statistics from the General Social Survey confirm this troubling trend. Baron and Berinsky explain the historical reasons for this shift in attitudes, but the more pressing question is: How can individual clinicians, and the profession as a whole, regain the patients' trust? "
― , Reinventing Clinical Decision Support: Data Analytics, Artificial Intelligence, and Diagnostic Reasoning
2 " Using this technique, Baum et al constructed a forest that contained 1,000 decision trees and looked at 84 co-variates that may have been influencing patients' response or lack of response to the intensive lifestyle modifications program. These variables included a family history of diabetes, muscle cramps in legs and feet, a history of emphysema, kidney disease, amputation, dry skin, loud snoring, marital status, social functioning, hemoglobin A1c, self-reported health, and numerous other characteristics that researchers rarely if ever consider when doing a subgroup analysis. The random forest analysis also allowed the investigators to look at how numerous variables *interact* in multiple combinations to impact clinical outcomes. The Look AHEAD subgroup analyses looked at only 3 possible variables and only one at a time.In the final analysis, Baum et al. discovered that intensive lifestyle modification averted cardiovascular events for two subgroups, patients with HbA1c 6.8% or higher (poorly managed diabetes) and patients with well-controlled diabetes (Hba1c < 6.8%) and good self-reported health. That finding applied to 85% of the entire patient population studied. On the other hand, the remaining 15% who had controlled diabetes but poor self-reported general health responded negatively to the lifestyle modification regimen. The negative and positive responders cancelled each other out in the initial statistical analysis, falsely concluding that lifestyle modification was useless. The Baum et al. re-analysis lends further support to the belief that a one-size-fits-all approach to medicine is inadequate to address all the individualistic responses that patients have to treatment. "
3 " Although these digital tools can improve the diagnostic process and offer clinicians a variety of state-of-the-art treatment options, most are based on a reductionist approach to health and disease. This paradigm takes a divide-and-conquer approach to medicine, "rooted in the assumption that complex problems are solvable by dividing them into smaller, simpler, and thus more tractable units." Although this methodology has led to important insights and practical implications in healthcare, it does have its limitations. Reductionist thinking has led researchers and clinicians to search for one or two primary causes of each disease and design therapies that address those causes.... The limitation of this type of reasoning becomes obvious when one examines the impact of each of these diseases. There are many individuals who are exposed to HIV who do not develop the infection, many patients have blood glucose levels outside the normal range who never develop signs and symptoms of diabetes, and many patients with low thyroxine levels do not develop clinical hypothyroidism. These "anomalies" imply that there are cofactors involved in all these conditions, which when combined with the primary cause or causes bring about the clinical onset. Detecting these contributing factors requires the reductionist approach to be complemented by a systems biology approach, which assumes there are many interacting causes to each disease. "
4 " Since the 19th century, medicine has focused on specific disease states by linking collections of signs and symptoms to single organs.... Systems biology and its offspring, sometimes called Network Medicine, takes a more wholistic approach, looking at all the diverse genetic, metabolic, and environmental factors that contribute to clinical disease. Equally important, it looks at the preclinical manifestations of pathology. The current focus of medicine is much like the focus that an auto mechanic takes to repair a car. The diagnostic process isolates a broken part and repairs or replaces it.... Although this strategy has saved countless lives and reduced pain and suffering, it nevertheless treats the disease and not the patient, with all their unique habits, lifestyle mistakes, environmental exposures, psychosocial interactions, and genetic predispositions. "
5 " The US National Institutes of Health states: "There is a lot of overlap between the terms 'precision medicine' and 'personalized medicine.' According to the National Research Council, 'personalized medicine' is an older term with a meaning similar to 'precision medicine.' However, there was concern that the word 'personalized' could be misinterpreted to imply that treatments and preventions are being developed uniquely for each individual; in precision medicine, the focus is on identifying which approaches will be effective for which patients based on genetic, environmental, and lifestyle factors. "