It doesn’t look like performance metrics are going away anytime soon. So why not make them a feature instead of a bug?
Guest Bio: Shawn Dowling MD is the medical director of the Physician Learning Program at the University of Calgary and Clinical Content Lead at the Calgary Zone Emergency Department.
Receiving audit and feedback data typically makes providers uncomfortable and feels punitive. It often seems like an unnecessary intrusion on our autonomy and practice. Why are these metrics important and how can we reframe the process so it doesn’t feel so negative?
- As clinicians, we want to provide the best care possible. Unfortunately, we’re not able to do that without getting a good sense of how we practice in relationship to our colleagues and in relationship to achievable benchmarks of care.
- Providing feedback in a dynamic way (rather than just during performance reviews) is a more fruitful way to help providers identify perceived and unperceived learning needs.
- Using metrics in medicine is still novel. So if the information seems poorly thought out and capricious, this doesn’t mean that we should get rid of them altogether. Instead, it means we need to co-design them with clinicians to make them reflect what we want to get out of the metrics.
Clinicians have a notion that they’re good at self-assessing the quality of care that they provide. There’s a lot of evidence that shows we’re actually not very good at it.
- Feedback is an opportunity to get objective data to show how we practice and to decrease some of the variability in the care being provided.
- Evidence shows that providing metrics is an effective intervention that can be a very impactful tool to get physicians to identify opportunities and implement changes in their practice.
Metrics are often created by administrators and seemingly done more with administrative intent than with any intent at improving patient care. How can we make these more friendly for providers?
- One of the challenges with metrics is that we’re limited to what we can measure, so they tend to focus on providing data that is available to us. In the ED, flow-based metrics are commonly used (ie. time to consults, time to discharge, number of patients seen per hour). The problem is that these numbers are not necessarily meaningful to clinicians.
- The lack of impact of metrics on patient care and the fact that clinicians weren’t involved in developing metrics makes people reluctant to interact with this data.
- Data is meant to be self-reflective, not punitive. It’s meant for you to look at what your practice is like and identify opportunities for improvement. Providers tend to take the results personally, however the results can help identify system-level changes.
- Frontline physicians need to help with designing these metrics and making them reflect what they value as improving patient care.
What’s the best way to present metrics data to individual providers? Hint: It’s not emailing an Excel spreadsheet which shows how many standard deviations you are from the mean.
- Provide the data in a way that’s meaningful to the clinician. Dowling recommends facilitated audit and feedback sessions where the group provides both the data and tools for change.
- In feedback sessions, engage doctors with the metrics by: 1) discussing why audits are important, 2) having one of the clinicians present their own data to the group, highlighting the areas where he/she did well in addition to those points that need improvement, and 3) ending the session by creating system-level and individual action plans and brainstorming ways to put the action plans in place.
The top 3 metrics that are the most helpful to focus on from an emergency department perspective:
- Return visits to the ED – Ideally, this data should be provided in real time, so the provider can review the case while it’s still fresh in his/her memory to determine if something could have been done differently to prevent the “bounceback”.
- Diagnostic imaging (DI) utilization rates – Providers who have a higher DI utilization rate should reflect on what might be the reason for their practice. For many, it’s having a risk averse personality. Once a reason has been identified, you can target a strategy to help change your practice.
- Usage of IV vs oral medications (when an oral option is available and equally effective) – IV medications are more costly, are associated with a longer length of stay, and have potential morbidity from an IV placement.
When giving metrics, you need to paint as full a picture as possible in terms of the care provided. It helps to balance metrics — providing data points that are associated with one another.
- For example, when evaluating CT usage rates or length of stay, also look at the provider’s 72 hour return visits rate. Those who order the fewest advanced imaging studies or who discharge patients the quickest may also have the highest return visit rate.
- The metrics that are currently available do not distinguish good and bad providers. Seeing more patients per hour is not necessarily better, nor is ordering lower than the mean number of CTs.
What is the most sensible way for a group to formulate an action plan for a provider who has a much higher than average length of stay metric?
- First, get the physician to be willing to change. Next, unpackage what contributes to that physician’s flow.
- It can be helpful to pair this provider with someone who’s just a little bit ahead of them in terms of their flow. The goal is to identify small changes that might improve the efficiency of their throughput.
- You need to ask this question: is the solution to help this doctor work faster or should you be changing the system so that excellent care can be provided at the pace this person is most comfortable working?
If there were a mission statement for metrics, what would it be?
- Metrics should be looked at as another tool to help improve the quality of care that we provide.
- Providers should avoid internalizing their data. The data does not define who we are as people or the quality of care we provide.
- The goal is for metrics to not only help with physician-level changes, but also with system-level improvements.