It’s American Heart Month: Here’s How NLP Can Help Shed Light on Heart Disease

69

By David Talby, CTO, John Snow Labs

February marks another American Heart Month: a time to reflect on our cardiovascular health. This is the 57th year the federally designated event has been commemorated, but this year it holds a stronger weight due to the impact of the COVID-19 pandemic on the public’s heart health, including potential harmful effects on the heart and vascular system, according to recent research. To add insult to injury, quarantine has dissuaded many people from going to hospitals for potentially harmful heart-related symptoms, creating poorer outcomes. This paired with more people engaging in unhealthy lifestyle behaviors, such as eating poorly, drinking more alcohol, and limiting physical activity, has only contributed to that.

While it seems hard to find a silver lining here, the coronavirus has forced us to think about outside contributors that may be affecting heart health. While most people understand diet and exercise will impact proclivity for diabetes, which can then lead to more serious illnesses like heart disease, what are all the factors more commonly glazed over? What about employment status or literacy—could these be health contributors, too? As heart disease continues to be the greatest health threat to Americans and is also the leading cause of death worldwide, isn’t it worth at least exploring? 

The short answer: yes. These factors outside of the typical spectrum of care are called social determinants—elements that directly impact a person’s health beyond diseases or drugs, such as access to healthy food, personal safety, housing, employment, literacy, family, employment, and personal freedom. These are often more important than clinical treatment when it comes to managing chronic diseases, like heart disease, and a host of other ailments. Without looking at the full spectrum of a patient’s life, it’s impossible to get to the root cause, thereby preventing the disease, rather than just treating symptoms. But the problem doesn’t exist in whether we should look at social determinants of health, but rather how we look at them. 

Social determinants can often only be read from free-text notes in a healthcare setting—not in structured data. What this means is doctor’s will manually write out details of a patient’s social history, home environment, and similar types of health contributors. Structured data in electronic health records (EHRs) would only consist of lab results, billing codes, and what medications the patient is taking. But if there’s substance abuse, unemployment, homelessness, or illiteracy, those will be in the free-text notes. In order for medical professionals to realistically compile and use this information, they need natural language processing (NLP). NLP is the automated way to connect the tissue between these disparate and siloed data sources to understand how these health events are related. 

In addition to the challenges of linking free-text and structured data, sometimes medical professionals simply don’t know what they’re looking for. What happens to patients with heart disease who take a daily multivitamin and exercise regularly? You may find that their symptoms improve, and that’s great research to have. But what if you’re looking more broadly at any behavior that can improve outcomes of heart disease patients? NLP is the only viable way to correlate all potential variables—sleep, relationships, safety, employment, obesity, etc.—to uncover answers to cases less specific than looking at just multivitamins and exercise. It would be nearly impossible for a doctor or data scientist to read line-by-line and try to connect the dots, even if all the information you needed was in text. But don’t forget, important information also lives in diagnostic imaging reports, social media, and other modalities. You need software to connect the relationship between these things.

Data quality also needs to be a consideration when exploring social determinants of heart disease. While Cardiology is well-known for using data-centric governance models and the American College of Cardiology governs the quality enhancement of 90% of cardiac-related data in the US, data integration is still a problem. In large research projects where information is collected from different entry points and data is available in different formats, it’s common for important information to be missing or inaccurate. Once again, NLP is a good source for researchers working in the cardiology field to mitigate this issue. With existing datasets in this speciality, researchers and data scientists can more easily uncover new findings with increased accuracy. Having curated and standardized data can make researchers’ jobs much easier and save years of headaches.

Even with major advances in healthcare and wellness over the years, cases of heart disease have increased 17.1% over the past decade, resulting in millions of deaths worldwide each year. Despite this, in most cases, heart disease is preventable when people adopt a healthy lifestyle. Intuitively, people understand that maintaining a healthy weight, physical activity, not smoking or drinking excessively, and controlling blood sugar and cholesterol, will impact how healthy they are. But a person who just lost their job and health insurance through their employer is not necessarily going to be prioritizing diet and exercise. Taking social determinants of health into account is often undervalued, but vitally important for patients’ overall health outcomes. Fortunately, technology like NLP has made it easier to start correlating social determinants to heart health, and has the potential to vastly improve prevention and treatment.

LEAVE A REPLY

Please enter your comment!
Please enter your name here

This site uses Akismet to reduce spam. Learn how your comment data is processed.