Jacob Kupietzky is President of HealthCare Transformation, a company dedicated to providing hospitals with experienced interim executives.
One of healthcare’s biggest challenges has always been its lack of precision. If a patient were to ask if a prescribed treatment will be effective, the doctor can only respond with generalities: “Well, in this 2013 study, X percent of patients exhibited Y range of outcomes with Z percent significance.” This is far from a criticism of my profession; each patient brings with them so much variability—be it genetic, behavioral, environmental, demographic or countless other factors—that the best doctors can do is provide only generalized treatment recommendations.
At least, that’s how healthcare used to be—before the advent of personalized medicine.
Personalized Medicine And Big Data
Advancements in technology have created incredible amounts of data in all areas of our lives. The idea behind personalized medicine is that practitioners can take advantage of this patient information to determine “which approaches will be effective for which patients based on genetic, environmental and lifestyle factors,” to borrow a phrase from the National Library of Medicine. When this concept was in its infancy a few years ago, practitioners were limited to such data sources as medical images and patient histories. Today, the variety of sources of patient data is astounding.
• DNA testing. The amount of publicly available genomic data is so vast, it is surpassing that of astronomy.
• Medical imaging at the subcellular level, plus its metadata.
• Diagnostic and treatment information provided in electronic health records.
• Insurance claim and pharmacy refill information.
• Data from home testing kits, such as hormone levels, allergies, food sensitivity and vitamin deficiencies.
• Behavioral and biological data via wearable devices. These days, your Apple watch can do more than tell time—it can count your steps, track your heart rate and even record an echocardiogram.
Personalized Medicine In Practice
So how can all this information inform personalized medicine? By harnessing all of these data points, we can better match the right treatment plans to the right patients. Consider the following examples.
• In a 2017 study in Translational Medicine & Research, the authors discuss how data from wearable monitors, such as out-of-office blood pressure readings and at-home exercise readings, can be used to create treatment plans for preventing hypertension. The same article asserts that by mining omics data, phenotype data, social media, insurance claims information and electronic medical records, practitioners can better create treatment plans to address cardiovascular disease before acute myocardial infarction.
• In this 2018 study published in BMC Medicine, the authors discuss how by integrating molecular profiling with drug-gene interactions, doctors can remove from consideration cancer treatment options that are ineffective or cause adverse events.
• In a 2017 study in Frontiers in Pharmacology, researchers used genetic screening to identify the phenotype of patients with cystic fibrosis. By understanding each patient’s phenotype, doctors can adjust the treatment targeting cystic fibrosis symptoms.
Understanding Costs
While personalized care at first blush sounds like a prohibitively expensive proposition, I believe the idea behind it is actually likely to lower the costs of care. By trading the costs of acquiring and managing data for the costs of unnecessary and imprecise treatment plans, the use of big data can create better outcomes faster. As personalized medicine advances, we can expect to find less trial and error, an increase in preventative procedures and a decrease in corrective treatments.
Potential Challenges
This kind of personalized care is still a developing trend, and as with any paradigm shift, there are several obstacles that need to be addressed.
• Data Management: Storing and processing increasingly large amounts of data can be challenging. The more data we have, the harder it is to make that data FAIR—findable, accessible, interoperable and reusable.
• Data Noise: Sorting through data and understanding the story it is telling you can be challenging, especially when multiple data points (with multiple sample sizes and procedural rigor) are telling conflicting stories. The more data we receive, the more important proper data interpretation becomes.
• Privacy Concerns: Any conversation about data should be accompanied by one about who owns the data (are medical institutions owners of the data, or merely custodians?) and how that data can be used.
• Patient Expectations: With more data becoming available to patients (such as through wearable monitors and at-home testing), patients have greater ability to self-diagnose, without medical training or expertise—and agency to act on those self-diagnoses.
That last challenge truly reinforces the role of the physician and the importance of the doctor-patient relationship. Whether we like it or not, our patients have more access to their own health and biological information than they have ever had before.
Our role isn’t to downplay that information in an effort to increase our own prestige, however. We should put it in the correct context, for sure, but ultimately use that data to create better patient outcomes. With an increase in data comes an increase in expectations, and we must be up to that challenge.
Forbes Business Council is the foremost growth and networking organization for business owners and leaders. Do I qualify?
Read the full article here
