Leslie H. Asks…
I am an administrator for a Cardiology practice in the mid-west, and we have been noticing a high rate in no shows over the past several months. We’ve considered charging patients for not showing up, but I’m not sure if it’s the right way to go. I’ve researched policies to see what other practices do, and some who charge, never really receive a payment so they don’t see much gain.
The MBA Responds…
Thanks for the question, Leslie! The high rate of no shows at your Cardiology practice is analogous to issues in other industries such as a websites abandon rate or a company’s high turnover rate. No matter the industry or the function, these type of business challenges associated with specific metrics can be resolved using quantitative and qualitative analysis techniques in the following manner.
Frame the Problem.As always, the first step in addressing any business challenge is to frame the problem using a well-defined research question. In your case, let’s frame the key question as follows: What are measures our Cardiology practice can implement to reduce our high no show rate? By framing the question in this manner, you can expect solutions that are specific, tactical recommendations to improve the high no show rate (or high turnover, or high abandon rate for other industries). This approach is also powerful in establishing a clear scope and setting boundaries to the problem so you are not “boiling the ocean.” Once you have the key question in hand, the next logical step is to break down the key question into further components (still using the question format). Several examples of these further components could be:
- What are the specific reasons for the high no show rate?
- What diagnostic codes have the highest percentage of no shows?
- Is the high no show rate specific to the Cardiology practice or is the problem persistent across practices?
Develop some initial hypotheses.Armed with our key question and sub questions, your next logical step is to develop a set of initial hypotheses that are likely solutions to resolving your high no show rate. Borrowed from the Scientific Method (learned in middle school!), hypotheses allow you to further structure and focus your problem. The structure and focus enables you/your team to ask the right questions, source the “right” data, and conduct the “right” investigation. Think about every episode of Law & Order you’ve seen, each case leads off with a hypothesis ultimately leading you after 40 minutes to the criminal! An example of a hypothesis is included in your query – charging patients may be a solution to improve the no show rate. Nonetheless, you’ll have to use the next step to validate the hypothesis.
Analyze the Qualitative and Quantitative data/drivers.Next, for each hypothesis, create an action plan that explores both the qualitative and quantitative drivers. What do I mean here? I would first explore the qualitative drivers followed by the quantitative drivers. Qualitative analysis involves conducting interviews, focus groups, and surveys with open-ended questions not only with internal staff but also with patients. Once you have the qualitative data, use it to source the quantitative data needed for confirmation. For example, if you identify in your focus group that patients with a specific diagnosis perceive their in-office waiting time is too long, therefore decide to not show up to an appointment, then secure waiting time data to confirm this notion. If the data confirms the word, then you’ve identified one source of the problem, i.e., fixing in-office waiting time may resolve high no show rate; if the data does not confirm the word, then you’ve identified that you have a perception problem, i.e., an awareness campaign that highlights less than average waiting times may resolve the high no show rate.
Engage in some Predictive Analytics.Almost there, so bear with me! The data from Step 3 provides you insights from historical data which is extremely valuable, but how great would it be to be able to use that data and look forward? Predictive analytics allows you to do that! Using statistics and analysis techniques like logistic regression, you can assign a “score” to each patient that measures their likelihood of being a no show. Pretty cool, right? Your credit score is an example of the “score” I am referring to – your credit score determines how risky you are as a debtor or how likely you are to default on a loan. The patient score will not only let you assess your likelihood of being a no show, but also provides the justification for employing tactics such as overbooking (what airlines do) or performing same day confirmations for those patients with a high score.
Articulate your recommendations.At some point in time, you’re going to have to articulate your approach, data, and solutions to your boss or client so that they can make a decision and take some action. If you have solid data to support each recommendations (and hopefully, a predictive model), it improves that somebody’s ability and confidence in making the decision and taking the action. This leads to fixing the no show problem sooner than later (due to slow decision making or needing additional research). So, in presenting your recommendations, tell the decision-maker a story (starting from Step 1) and make each transition using the logical connections. Finally, always build in a recommendation or next step to measure and monitor your metric (in this case, the no show rate) to observe the trend and re-strategize, as needed.
The aforementioned approach may seem trivial, but given the short-term pressures and the day-to-day fire-fighting it’s easy to take shortcuts and skip steps. Avoid the shortcuts! They only lead to duct tape solutions and don’t address root causes.
Hope this helps, Leslie! As always, please leave your comments/feedback.