It is widely believed that social determinants of health (SDOH) may have a greater impact than physical health in determining an individual’s overall well-being, especially in underserved communities. However, understanding how specific SDOH factors affect individuals is extremely difficult because clinicians do not systematically collect SDOH data.
Although SDOH data does exist in patient records, it is too difficult and time-consuming for clinicians to understand because SDOH-related information is often hidden in patient notes. This problem ultimately inhibits their ability to use data to inform decisions about individuals receiving care.
Natural Language Processing (NLP) – a key discipline of artificial intelligence that uses computers to understand written text – can overcome this challenge, and I encourage HHS organizations and hospitals to explore this method of displaying SDOH data, especially as a Technological innovation Over the past few years, the field has matured significantly.
Here are three reasons why health and human services organizations and hospitals should adopt NLP to understand SDOH.
1. Cost savings and more efficient care: For now, clinicians, therapists, and caseworkers spend a lot of time reading typed or handwritten case notes to understand a patient’s condition in order to identify potential treatment options. It’s just a waste of time that could be better used for direct interaction with patients.
The magic of NLP is that it can automatically highlight influential metrics and trends in case or patient notes, quickly revealing SDOH to case staff and clinicians. NLP platforms can alleviate the time spent by health and social services workers combing through difficult-to-manage records by easily highlighting SDOH in cases.
2. Improve results: NLP empowers caseworkers and clinicians with the information they need to make impactful decisions and empowers supervisors to maximize the quality of care provided. This is because NLP provides a deeper understanding of a patient or case.
The Gravity Project is a national public collaboration to create diagnostic codes for SDOH factors with the goal of incorporating these codes into the existing list of medical diagnostic codes. NLP can extract information from unstructured data such as case notes to support and convert it into SDOH-related diagnostic codes. These diagnoses will then trigger interventions to improve outcomes.
3. Risk mitigation: NLP enables organizations to quickly identify patients at the highest risk levels, so that interventions can be targeted to those most in need. I firmly believe that risk can only be truly identified if one understands what is contained in the narrative data. Most risk stratification systems today simply look at claims data to do this. But claims data is incomplete. If care coordinators get the full picture through SDOH, they will have a better tool to identify those at highest risk and where early intervention can prevent more serious health conditions.
Using NLP to understand the SDOH situation is clear. While SDOH is now widely believed to play an important role in an individual’s overall health, we need to make it easier for hospitals and health and human services organizations to consume and understand this data to understand how SDOH affects individual patients. Doing so will only help providers make the best decisions for their patients, resulting in more effective care and improved outcomes.
Photo: vaeenma, Getty Images
