It is critical to know how long your patients stay with you for forecasting revenue. We will refer to this metric as your Patient Recall Time.
Patient recall time influences patient lifetime value a lot. It allows you to better budget marketing expenses. Additionally, it can help you predict capacity and future staffing needs.
EHRs generally will not show you how many years a patient will stay with you. And, if you have been in business for less than ten years, you will not have reliable data within your EHR to go on.
Find Max Patient Recall Time
First, you should estimate your max patient recall time. It will help you know when a patient has abandoned your practice.
Your patient recall time is an upper bound of the time in between a patient’s appointments. If most patients see you every 1 year but a bunch also see you every 1.5 years your max patient recall time will include both.
You should not use the mean appointment span. Or an arbitrary period like “1 year”.
If you do, you would miss up to 50% of your patients when you calculate lifespan. These patients are not churned. They are just a little slower to rebook than average.
You can calculate your max Patient Recall Time with or without Excel. Use Excel if you have been in practice for at least three years.
Disclaimer: Technically, you are not finding the “maximum” patient recall time. The true maximum could be an outlier and not representative of your patients.
Without Excel: Estimate max patient recall time
A PERT estimate is best if you are not comfortable with spreadsheets. It is also great if you have been in practice for less than three years.
Guess how long one of your patients or clients typically goes between appointments. Convert this figure to years.
Then, jot down the longest and shortest span between appointments that you have seen. Convert each figure to years. Multiply your first guess by “4” and add it to your shortest and longest guesses. Divide this sum by “6”. You will use this number as your mean patient recall time, Tu. Save it for later.
Now, find your max patient recall time using this formula. You are calculating a sample standard deviation to represent 95% of our patient recall times.
This value Tmax is an upper limit representative of most of your patient recall times. We will refer to it as your max patient recall time from now on.
If a patient does not return within this time frame, they probably will not return.
Note: A PERT estimate is preferable over a three-point estimate because it approximates a Pareto distribution for the appointment spans of your patient population.
With Excel: Normal distribution of max patient recall time
The following steps will help you modify your spreadsheet to calculate patient recall time. Contact meddkit if you would like assistance with Excel for these calculations.
First, export all available encounter data into a flat file like a CSV. If that takes too long, try exporting at least three years from December 31 of last year.
Make sure that it includes the encounter identifier and date. Also include the patient name or, preferably, a unique patient identifier like MRN.
Take care to keep this data in a secure environment. Use a program like Excel or Google Sheets to open the file.
Ensure that your patient identifier is in the first column (A) and the date of the appointment is in the second column (B). Leave your encounter identifiers in the third column (C).
First, sort the patient identifier (or patient name) column from A to Z (ascending). Then, repeat the same sorting logic in the appointment date column. Ensure that you sort the appointment date column second, otherwise the following will not work.
Label column D “DATEDIF”. Enter this formula into the first empty cell and copy it down the column:
This formula will find the number of days between every two sequential dates for a given patient.
Next, label column E “AVERAGE”. Enter this formula into the first empty cell and copy it down the column
Finally, use these two functions to find your average and maximum patient recall times in days, respectively. Insert the first formula into cell F2. Insert the second formula into cell G2.
These functions remove your outliers: the largest and smallest 10% of your sample. Removing outliers helps to avoid data entry errors. It also minimizes the impact of abnormal appointments throwing off your estimates.
The first value is Tu. Your marketing efforts should encourage patients to book another appointment before this time. The second value is Tmax. If a patient does not return within Tmax, they probably will not return.
Find Average Patient Recall Rate
Average patient recall rate is the keystone of patient lifetime value.
Getting your patients to see you twice instead of once per year has the same effect as doubling your gross profit per appointment.
But gross profit per appointment is often much harder to change.
Increase average patient patient recall rate
Increasing average appointments per patient can dramatically improve your bottom line. That is why
smart practitioners work hard for recalls.
Why do many practices fail to maintain a recall program? Because it is a lot of work. It involves setting
up consistent digital communication channels with patients.
If you are busy, you may not have the time for consistent marketing. In these cases a firm like meddkit.com can do it for you.
But, it all starts with knowing how frequently your patients see you today.
Calculate average patient recall rate
Luckily, it is easy to calculate your mean patient recall rate. We will represent it as ru.
In fact, you already have it.
It is the inverse of your mean patient recall time, Tu. You found this when determining your max patient recall time.
If you see each patient every 2.2 years, on average, then you see the same patient every .45 years. Or, if you see each patient every 2 weeks, then your average average annual appointment rate is 26.
Calculate this number every year along with your average patient recall time. Doing so will help you determine if your recall marketing is effective.
With your average patient recall time and rate found, you must now find your patient churn rate.