Eric Makhni, MD, serves as director of quality and informatics of Henry Ford Health System's orthopedic service line and clinical associate professor of orthopedic surgery at Wayne State University School of Medicine, both in Detroit.
After completing a shoulder and sports medicine fellowship at Chicago-based Rush University Medical Center in 2016, Dr. Makhni joined Henry Ford Health System, where he is incorporating patient-reported outcome measures into daily clinical practice across the orthopedic service line to advance clinical care, quality, value-based care and clinical research.
Heavily used in orthopedics, PROMs are validated patient-centric questionnaires that give quantitative outcomes on patient health, and were developed so patients can accurately report how they consider their health.
Here, Dr. Makhni discusses how PROMs are collected, keys to successfully implement them into practice and how they can improve clinical care.
Question: How are PROMs typically collected?
Dr. Eric Makhni: In an ideal world, we would measure PROMs on a routine population health basis, meaning that every patient that comes into the office would get PROMs, which would be reviewed with the patient during the clinical encounter. Unfortunately, there are a couple of things that hinder that. For one, it is difficult to integrate PROMs into a busy ambulatory workflow. It costs money — you have to pay for the collection, whether that's extra personnel, the technology, integrating it into your EMR, or new hardware with tablets and software. The other issue is it causes disruption. When you're already dealing with insurance verification, X-Ray imaging, etc., adding PROMs into the mix makes it very challenging for your workflow.
Q: What are some other challenges in collecting PROMs?
EM: There are challenges from the financial, administrative, workflow and logistical perspectives. The other reason it's difficult is If you are using a lot of PROMs that require knowledge of a diagnosis — within the knee or shoulder, there are dozens of PROMs — then someone has to manually assign a PROM, rather than it being automatically collected. For example, if a patient comes in with a shoulder injury, they don't know what the diagnosis is. It could be a fracture or rotator cuff tear, or any other diagnosis. So, a clinician has to see the patient, make a diagnosis and then assign a PROM.
But by doing that you've disrupted the workflow because now you're adding more work for the provider. What we have done is remove that barrier. We utilize National Institutes of Health patient-reported outcomes measurement information system CAT scores, which are dynamic questionnaires that are domain-specific (pain, function, mental health). These can be automatically assigned to patients according to basic parameters from their appointment information. This allows us to collect PROMs on patients before they are seen by the provider.
Q: What are some best practices for designing a successful PROM system?
EM: There are three conditions that must be met in order to develop a successful and sustainable PROM collection platform: 1) you must utilize existing clinical workflow and personnel, 2) questionnaire assignment must be automated, and 3) the process must be — as Judy Baumhauer, MD, from University Rochester in Brighton, N.Y., likes to say — "invisible to the provider."
Integrating into existing workflow and personnel is important in order to minimize disruption and costs associated with hiring additional personnel. Secondly, it has to be automated. Patients have to get their PROMs before they see the clinician, with an option to fill them out at home ahead of time, or in the clinic before they enter the exam room. You also need to make sure that the PROM survey time takes no more than five minutes, which further limits workflow disruption. This way there is minimal disruption. It's critical for the score to be completed before the patient sees the physician, so they can see the scores to know how they're doing. You need that data ahead of time, and the patient will want to review the data in real-time with the provider.
I really cannot stress enough how important it is to have automated PROM assignment. If a patient is there for a shoulder issue, keep it simple — get a physical function form, a pain form and a mental health form. More and more evidence indicates that PROMIS CAT forms are just as good as diagnosis-specific forms and much quicker to administer; this facilitates automated assignment.
At Henry Ford Health System, we have a very complex clinical workflow, with multiple clinical and support teams that report to different managers. We had to make the platform as easy as possible so that it could seamlessly integrate into the existing workflow. Our solution was to utilize QR code scanning upon appointment check-in at the front desk, which in turn triggered the PROM questionnaires on the tablet. This process required no data entry by the front office support staff, which led to quick and easy adoption. We had nearly 90 percent questionnaire completion in our pilot roll-out, in large part due to this simple process and the efficient nature of the PROMIS CAT forms.
Q: Can solo practitioners and small practices collect PROMs, or is it best suited for large group settings?
EM: Anyone can collect PROMs. You can be in a solo practice using a free collection system or you can be a large multispecialty integrated health system like us. For solo practitioners, it's easier in some ways because you have a little bit more control over the process and the platform. There are a number of free and low-cost platforms that are available for providers who use EMRs that do not have PROM capabilities. Don't forget, collecting scores on paper and pencil is still valuable, even though the advanced analytics and research capabilities may be limited.
Q: How can PROMs help improve clinical care?
EM: There are four ways to incorporate PROMs into clinical practice — clinical care, quality, value-based care and for clinical research. From a clinical perspective, it's best to incorporate PROMs into shared decision-making and postoperative care. You might have a patient with advanced knee arthritis but "normal" function and pain scores; doing a knee replacement surgery on that patient may actually make them worse from a function and pain perspective. So, these scores can be invaluable when incorporated into shared decision-making. In terms of postoperative care, for example, if I have a patient who has undergone arthroscopic rotator cuff repair, I have a good idea of what the expected postop function and pain scores are, based on research we have published. So, if a patient hits that milestone at four months, that's great. They probably don’t have to keep coming in for in-person office visits if they are feeling well, and we will switch to virtual care at that point. Patients love that. On the other hand, if the patient has a sudden decline in their PROM scores, I may want to bring them in to evaluate what could be causing the issues.
Q: What about quality work and value-based care?
EM: The fact that you collect and analyze your outcome scores, you are by definition doing quality work. You're looking at your outcomes and opportunities for improvement. The third way of using PROMs is in value-based delivery. There's a concept called condition-based bundles. Traditional bundled payments start when you have the surgery (the "episode"). But what if that patient didn't need that knee replacement to begin with? They may just need to lose weight, get better pain management, better physical therapy, etc. With condition-based bundles, payers will provide a payment to the physician to treat the patient nonoperatively however they see fit. So, it's going to encourage physicians to do high-value based treatments that improve function and pain, and we measure that through PROMs, which is the cornerstone. You cannot do these programs without PROMs.
Finally, PROMs are the foundation for meaningful clinical research. It goes without saying that you can't do research if you don't have data. While research is not the principal reason to collect PROMs, it is certainly a down-stream benefit. Since we started collecting PROMs at HFHS in a systematic fashion, we have found it to be much easier to conduct high-impact clinical research, as we now have a great amount of data to scrutinize and analyze.