TRANSFORM HF 2023 Collaboration Starter Grant recipients Kimberly Crasta and Ellis Gao – along with their team of collaborators, Joseph Ferenbok, Anna Goldenberg, Juan Duero Posada, Heather Ross, and Dawn Donaldson – want to improve clinic workflow efficiency in outpatient settings.
Kim and Ellis first met when they were placed in the same group project while completing a Master of Health Science in Translational Research as part of the University of Toronto’s Translation Research Program. However, their roles as project partners quickly evolved following some well-timed networking.
While attending TRANSFORM HF’s 2022 Spring Network event, Kim and Ellis learned that the cardiology team at University Health Network (UHN) was hoping to make some improvements in the clinic, such as reduce lengthy wait times for patients and enhance the functionality of the clinic space. With a shared interest in helping to alleviate healthcare resource allocation problems, they saw a unique opportunity to get involved and lend their expertise. The two quickly became close collaborators on a project to develop strategies to facilitate clinic efficiencies and, ultimately, improve care for patients.
We sat down with Kim and Ellis to discuss their TRANSFORM HF-funded project and what’s coming next from this collaboration.
How would you describe the goals of your research in layman’s terms?
Ellis: Our overall goal is to improve clinical efficiency and help smooth out the clinic flow. But to achieve that goal, we first needed to look at what data we could use for analysis of the current clinical landscape. Our initial thinking was to use a large pool of data to assess outpatient clinic flow, which is where Epic – UHN’s electronic health record system – came in because it stores clinic staff’s timed entries for the different stages of a patient’s visit, such as checking in at the front desk. However, we realized the data recorded in Epic wasn’t necessarily reliable for our research needs. Although we knew when a patient arrived, we were unsure about the precision of the other logged interactions – were they real-time records of the different points during a patient’s visit or had they been logged in Epic at a different time than the interaction actually look place?
To effectively perform our analysis, we needed to generate data straight from the clinic. We saw an opportunity to conduct time series tracking, or in other words, record the different points at which stakeholders in the clinic took certain actions. Our team collected data by sitting in clinics and manually tracking the entire flow of 150 patients as they navigated through the various phases of their appointment, such as spending time in the waiting room or going to the lab to get their blood work done.
So, that was our first phase – and right away, we noticed there were some small things that could easily be tweaked to improve flow. It may have only been a couple minutes at a time, but that stacks up over the course of a day – especially when there are sometimes 100 appointments in one day. This data collection – validated by discussions with all the stakeholders in the clinic – was the foundation of figuring out the variables – or stages – that determined the flow of a patient’s visit, such as doing blood work or an ECG, and where adjustments could be made.
Our next goal, which is what we’re working on now, is to standardize these two sources of data – our manual collection and the data from Epic – so that we can create an AI predictive tool and use statistical modeling to analyze trends and improvements in efficiency moving forward.
Our goal is that anybody who is visiting the clinic gets the best care possible – and in the most efficient way possible
Could you tell us about some of the partnerships that you’ve formed through this work?
Kim: While our project started with the Heart Function Clinic at UHN’s Peter Munk Cardiac Centre (PMCC), we have now expanded to the other cardiology clinics within PMCC. This partnership stems from some key collaborators, including Dawn Donaldson who is the clinical manager; Dr. Posada, who has become our Quality Improvement partner; Dr. Ross, who is our Principal Investigator now; and individuals from the University of Toronto’s Computer Science department, who are helping us build out and understand the AI. TRANSFORM HF also helped us get in touch with patient partners to understand their perspectives and get a better sense of how they experience the clinic workflow.
What does the next phase of your project look like?
Kim: Now that we have a large pool of data from the clinics, our first goal is to standardize how stakeholders in the clinic – physicians, administrative staff, nurses – interact with Epic to make sure the patient flow is being tracked effectively. This is important because Epic not only serves as a key tool for communication but also the source of data for when patients need to progress from one location to the next within the clinic, so these metrics are quite vital for understanding what resources need to be allocated where.
During our master’s program, we also developed a questionnaire to help the clinic get a baseline understanding of a patient’s experience. So, as we implement the standardization part of our research, we’ll deploy the questionnaire to monitor changes in patient feedback and experience to determine whether we’re going in the right direction.
Once the standardization phase is completed, we’ll do another round of data analysis based on logged interactions in Epic and see how it has changed. This is where the trend analysis comes in – looking at variables that stand out to us, such as the biggest barriers to wait times or how doctors respond. Looking forward, we aim to use all the data from Epic alongside AI computing power – including comparative analysis and time series prediction – to create a model that does this kind of work in automation with Epic, thereby making it simpler and faster to assess improvements in clinic flow and feed those insights back to the clinic teams.
What do you hope its impact will be on people living with heart failure?
Kim: For people with heart failure, and others as well, we’re really hoping to improve the efficiency of the care that they experience. From what we’ve heard from patients and clinical staff, many people must travel from far away for their appointments and usually need to take a day off work, which not everyone can necessarily afford to do. Our goal is that anybody who is visiting the clinic gets the best care possible – and in the most efficient way possible. If their experience is quicker and smoother, they might not have that secondary impact on how they live their life. Let’s not add additional burden.
Ellis: There are a lot of patients that go through the clinic, and by making it more efficient, there can be a higher capacity for daily flow so that more patients can get the care they need.
TRANSFORM HF’s Collaboration Starter Grants encourage, foster, and support members of our community working collaboratively on research and project proposals that advance our mission and expand our network.
Keep up with our recipients’ journeys by following TRANSFORM HF on Twitter and LinkedIn.
Interested in funding opportunities? Sign up for our newsletter to stay in the know!
