Wellstar Health System

Jennifer Giusti

I recently struck up a conversation with Sophie at my local Farmers Market. When she learned about my professional background, she chatted about her health. I estimated her to be in her 70s. Despite using a cane, she seemed in generally good health, judging by her appearance and armful of fresh veggies. But as we talked, Sophie expressed a disconnect between her perception of her health and that of her doctor: “I’ve tried to explain how sick I feel, but my doctor says my tests look fine,” “The medications he has me on make me so tired,” and “I’m worried I won’t be able to take the trip to see my grandchildren this summer.” As our conversation continued, I saw that Sophie was frustrated and eager for someone to listen to the health goals, that were of most concern to her and her clinicians may be missing vital information predictive of her health outcomes for the next year.

In the medical profession, we traditionally judge an individual’s health based on physical examinations, laboratory results, and radiologic diagnostics. We prescribe treatments based on what scientific evidence tells us about outcomes, like mortality rates. We measure a diabetic’s management in terms of A1C levels. A surgery patient’s recovery is judged on surviving the surgery, and the absence of hospital-acquired infections. It is less common to consider an individual’s health in a broader and holistic context. How do the side effects of the diabetic’s medication impact their energy and ability to participate in daily activities? Was the patient who received a new hip able to walk his daughter down the aisle 6 months after surgery? Taking a person’s overall physical and mental health into account, what is their risk for a negative outcome in the next 12 months?

Today, there is increasing interest in patient reported outcome measures (PROMs), which offer real predictors of quality of life and survival rates using information we can’t always discern from labs or MRIs. As I reflected on Sophie’s frustrations and my interest in healthcare technology, I contemplated how we can enhance and accelerate the use of PROMs by leveraging progressive tools.

First Thing’s First… What are PROs and PROMs?

Patient reported outcomes (PROs) are the “what” – these are reports that come directly from the patient, assessing their own health status, response to treatment, or health behaviors. Patients’ perceptions are vital to disease prevention, diagnosis, treatment, and collaborative health management. PROs can be predictive of healthcare utilization and mortality.

Patient reported outcome measures (PROMs) are the “how” – the tools we use to collect PROs from our patients. There are many PROMs tools, generally formulated as questionnaires using a Likert scale. Various organizations and researchers have taken different approaches to grouping measures into domains. Common elements include overall perception of health/quality of life, physical functioning, symptoms/symptom burden, mental health/emotional distress, pain, fatigue, and the ability to carry out social activities or roles. Here are a few examples of questions you might find on a variety of PROM tools:

 • “Does your health now limit you in climbing one flight of stairs?” (Rated on a 5-point Likert scale from ‘not at all’ to ‘cannot do’)

• “I am satisfied with my ability to do all of the community activities that are really important to me” (Rated on a 5-point scale from ‘not at all’ to ‘very much’)

• “Compared to one year ago, how would you rate your physical health in general now?” (Rated on a 5-point scale from ‘much better’ to ‘much worse’)

This last question is highly correlated with clinical outcomes. Patients who report their health is ‘worse’ or ‘much worse’ than one year ago have a significantly higher mortality rate in the next year.

“ePRO platforms offer integration into the electronic health record (EHR), which takes us a step closer to usable data”

From Pen and Paper to ePROs -

Gathering PROMs can be laborious. I currently have a team of nurses who call patients at incremental points after undergoing a surgical procedure to gather patient reported outcome data. Combining this vital information with rest of the data points we have for a patient is challenging at best and may not happen at all. So, how do we leverage technology to make the gathering, integration, and dissemination of PROM data a routine part of healthcare?

Multiple vendors now market apps supporting ePROs (electronic patient reported outcomes). This can be as simple as digitizing the questionnaires for a patient to complete on their phone or computer. The data can then be received and manipulated by the healthcare provider. ePROs promise to increase patient engagement and response rates. Some also offer the ability to contribute data to larger databases for research and data sharing. While electronic forms may be preferable to paper and pen, it does not go far enough.

ePRO platforms offer integration into the electronic health record (EHR), which takes us a step closer to usable data. However, there are thousands of EHR instances across the country with varied builds and workflows. It will be easy for PROMs to get lost in the plethora of information available in an EHR. Tech companies working in this space will need to be innovative and collaborative in their design, ensuring PROMs are not relegated to a dark corner of the medical record’s ‘media tab’ or added to the ‘noise’ in an already data-rich environment. 

ePROs do not come without a cost. Aside from software licensing fees, organizations that have implemented ePROs report that technical support resource needs are often underestimated. Bias is also a concern, given that segments of the population may not have access to the needed devices. Others may not be proficient with technology or may not trust the security of platform. If left to chance, participation rate differences amongst populations could decrease the prevalence of PROM use for those who may need it most. 

All those cautions aside, I do believe the eventual benefit of ePROs will greatly outweigh the costs. To realize those benefits, clinicians and PROMs experts should partner with machine learning developers. This triad should discern when and how PROMs can be combined with relevant physical findings, traditional diagnostics, and remote patient monitoring data to paint a clear overall picture of a patient’s health status. When we routinely and elegantly integrate PROMs data into everyday clinical workflows, providers will have a powerful tool to identify patients at risk and mitigate poor outcomes.