When dealing with large numbers of cases, clinicians lack real-time comparative evidence and are forced to make decisions despite unknown variables that can dramatically alter outcomes. Such evidence gaps occur every day, especially for patients with multiple conditions, complex medical histories, and diverse ethnic backgrounds. Advances in academic research, including a major project at Stanford University School of Medicine, have led to technological innovations that allow clinicians to generate the evidence they need from research data and anonymized medical records, bridging the evidence gap so doctors can make informed decisions. improve results.
When physicians are asked whether the patient case they are managing has an appropriate care guideline, the answer is usually “no.” As medical professionals can confirm, only about 20% of patients adhere to standard care guidelines. Couple this with the fact that only 20% of existing care guidelines are supported by high-quality evidence, and we arrive at a startling conclusion. only about 4% of patient care situations that a physician must manage have a guideline derived from contingency. controlled clinical trials. In other words, research is almost always lacking.
Such evidence gaps force physicians to make data-free treatment decisions based solely on intuition and experience. This situation is exacerbated when physicians care for patients with multiple illnesses, complex medical histories, or diverse ethnic backgrounds.
At a time when every click can be tracked and medical records are fully electronic, doctors must be able to digitally reference decisions made by other clinics to find out: What has happened to other patients like me?? While the technology exists to perform digital consultations on such data, incorporating the results into routine care requires new solutions. One model is emerging. services staffed with doctors and data scientists who all have longitudinal patient records and can return the required evidence in less than 24 hours.
An academic research project at Stanford University School of Medicine, the Green Button Informatics Consult Service, was developed to deliver technology-enabled clinical consultations. IT departments at some of the country’s most prominent institutions have since followed suit and experimented with similar services. Examples include City of Hope, Columbia University, and Mayo Clinic, which developed the Mayo Clinic Platform to securely organize data so that even the most complex medical queries can be answered as quickly as possible and immediately impact patient care.
The purpose of this article is to explain how these services work and how they can improve patient care and reduce costs.
The evidence gap
Consider this real and personal example (co-author John Halamka’s mother, to be exact): An elderly woman presenting with disturbed mental status, fever, and low serum sodium. He is hospitalized and seen by his primary care physician, who realizes that the patient likely has a urinary tract infection (UTI) and begins treatment with antibiotics and antipyretics. However, a UTI does not fully explain the patient’s low sodium level. Although low sodium levels can be a result of renal sodium clearance, patients with UTIs rarely have low sodium levels, leaving the doctor scrambling for answers.
Unfortunately, there aren’t many answers because no clinical trial has been conducted with 80-year-old females with impaired mental status and abnormally low sodium. But with millions of electronic patient records available, database consultation can allow a doctor to accurately diagnose and treat a patient, rather than just making a best guess.
Even among highly trained and skilled physicians, there are evidence gaps that impair their ability to accurately diagnose and treat certain patients, which is one reason this analysis should be performed regularly. Regular database consultations can answer important clinical questions such as:
- What is the correct diagnosis?
- What diagnostic tests should be prescribed?
- What is the implication of this abnormal lab result or genomic marker?
- What is the typical prognosis for such patients?
- What medications or other treatments should be taken, and in what order, to optimize results?
- Is this procedure worth the risk and/or cost to this patient?
- Can a patient’s life be prolonged or improved with alternative treatments?
Four decades of effort
The idea of consulting medical records to learn what happened to similar patients is not new. It probably began in 1972, when renowned medical researcher Alvan Feinstein published the paper “Assessing Prognosis Using a Speech Mode Computer Program,” in which he described a computer system where doctors, medical data, and technology come together to create real-world evidence to support lung cancer patients. to clinical treatment decisions. Since that landmark idea, several efforts have been made to realize this vision.
One example is the Duke Data Bank for Cardiovascular Disease, which was launched in 1975 and produced reports called: predictions which summarized what happened to similar patients when different treatment choices were made. The hope was that the Duke Databank would become part of clinical practice. However, the cost of obtaining data electronically, the focus on only one medical field (cardiology in this case), and payment limitations limited the effort.
Reports of Duke’s prediction eventually died down, but the concept was not abandoned. Other universities and institutions have worked to overcome logistical barriers and promote the required technologies and supportive policies.
In 2011, pediatricians at Stanford University School of Medicine faced a tough decision about whether to treat a patient with systemic lupus erythematosus (SLE) with an anticoagulant. It wasn’t standard practice, but doctors thought it was the best course of action given the complications. However, there were no adequate studies to confirm that this treatment option is the best risk-based option. To guide them, doctors used the school’s clinical data repository to assess the risk of blood clots. In less than four hours, they were able to review data from the SLE cohort, which included pediatric patients with SLE treated by doctors between October 2004 and July 2009, and make a decision about which anticoagulant to use based on the data. Their success was later published in the New England Journal of Medicine article, Evidence-Based Medicine in the EMR Era.
A workable solution
In 2018, a group of doctors and data scientists at Stanford University School of Medicine piloted the Green Button Informatics Consult Service, which used routinely collected, de-identified data from millions of individuals to provide evidence on demand in situations where sufficient evidence was lacking. . Results can then be immediately analyzed by attending physicians and data scientists to immediately inform patient care. During the pilot test, the service responded to 100 advice requests from 53 users of various professions. The consultations informed individualized patient care, led to changes in institutional practice, and prompted further clinical research, confirming the feasibility of evidence generation to close evidence gaps.
This new opportunity was the catalyst for Atropos Health, which has successfully delivered over 1,800 consultation requests to date. A number of health technology startups are committed to harnessing the power of electronic health record data to improve patient care. Lucem Health, nference, OMNY Health and Atropos Health are just a few examples. Healthcare giants like the Mayo Clinic and Cleveland Clinic are also taking note. Seeing the potential for database consultation at the point of care, Mayo Clinic partnered with Atropos, using the company’s technology to enhance the Mayo Clinic platform.
All of this is made possible in part by federal rules requiring patient records to be moved from paper charts to electronic storage, meaning hundreds of millions of patients’ 10- to 15-year-old medical histories are now routinely accessible. The Mayo Clinic platform, for example, improves care delivery through insights and knowledge derived from de-identified data from 10 million patient records, including lab values, diagnosis codes, vital signs, medications and clinical notes. This data is necessary to increase the effectiveness of database consultations; it gives data analysts and machine learning platforms the basic building blocks needed to turn data into insights.
Expert-in-chain services
Accessing and storing data for analysis makes life easier for data scientists and clinicians looking to improve care delivery, but developing new best practices for care delivery requires building an evidence base. Because data is processed through patient encounters, generating such evidence is time-consuming and costly. Furthermore, finding patterns in data is a manual process for many institutions, adding additional time and labor to the equation. Even at institutions with well-staffed analytical teams, generating such evidence often requires more than 300 hours and $300,000 just to answer clinical questions and develop new care guidelines. This professional approach cannot scale to serve the needs of every patient.
The solution uses new search technology, such as that used by Green Button Informatics Consult Service, and a staff of physicians and data scientists with on-demand access to data sets and expertise in interpreting search results. This approach can achieve the accuracy and speed required for a turnaround time of up to 24 hours needed to close evidence gaps at the bedside. In addition to dramatically reducing the time and labor required to produce this customized clinical information, this approach will almost certainly result in significant cost savings; ongoing research aims to quantify exactly how much.
In the past 40 years, despite huge investments in technology and policy to drive digital innovation in care, doctors still make many clinical decisions without data. Harmonized expert services leveraging now-available technology and data can help change that, fulfilling the early promise of electronic health records to translate real-world clinical insights into improved patient care. They can bring us closer to realizing Feinstein’s vision and close the evidence gap at the point of care.