Using Math to Predict Staffing Needs

Having excess staff is never a good thing. Just enough is the level where most practices are aiming to be. The problem is that many offices work with an erroneous definition of “just enough.”  As a consequence, phones ring off the hook, patients are dissatisfied, overtime is excessive, and basic tasks are neglected.

As an example, consider the challenges in staffing the front desk in a practice that sees 90 patients a day, and a benchmark that says patient registration for that specialty takes an average of 10 minutes. Straight arithmetic says that two full-time people at the front desk are more than enough:

90 patients/day X 10 minutes/patient X 1 hour/60 minutes = 15 hours/day < 2 front desk staff

So why is the waiting room always full of frustrated, waiting patients, while the exam areas are full of frustrated providers waiting to see patients they know are in the office? Here are a few ways to understand why 2 full time people are not enough.

Patient arrivals are not spaced over eight hours.

If the morning appointments are scheduled from 8 a.m. to 11 a.m. and afternoon appointments are scheduled between 1 p.m. and 4 p.m., patients are only being registered about six hours per day, not eight.

90 patients/day X 10 minutes/patient X 1 hour/60 minutes = 15 hours/day

15 hours/day X 1 man-day/6 hours = 2.5 front-desk staff members

Fractional staff should usually be rounded up.

If the workload is steady throughout the day, you can’t add or subtract half a person.  If it happens that the practice schedules really quick appointments in specific windows, it may be able to take advantage of fractional supplemental staffing during those periods.

Patient arrivals occur in waves.

The determining factor in a schedule is how long the appointment is expected to take, and the time required for registration is not usually a part of the calculation.  If the providers expect appointments to require, on average, either 15 minutes or 30 minutes, the logical move is to schedule all appointments on the hour, half-hour and quarter-hour.

The waves are not much of a problem in our example, assuming the practice has staffed the front desk with three people, no registration takes more than 10 minutes per patient, and the patients all arrive on time.  What are the odds? 

Some registrations will take longer than the average.

New patients may have to complete documents upon arrival.  Existing patients may need to update records or discuss a payment plan.  The printer may jam or the computer may need to be rebooted. 

Staffing to the average time required to perform a specific task does not account for ancillary disruptions that are inevitable.

The practice may have picked up that extra half of a staff person by assigning extra duties to the front desk.

That makes sense in terms of fully utilizing staff.  What happens, however, when a patient presents for registration at the same time the phone rings?

Staff members get sick and take vacation.

Scheduling models cover all of the time a practice is open and seeing patients.  No staff members are actually available all of that time.  If we assume each employee will be out of the office 10 days per year, the front desk in our example will be short-staffed for six weeks.

10 days/employee X 3 employees X 1 week/5 days = 6 weeks

Staff members take a leave of absence or quit.

Turnover is the big wild card in determining staff requirements.  The best available method to predict future experience is to look at what has happened in the past.  If turnover at the front desk has historically been 30 percent a year and it takes two months to hire a replacement, the front desk, in our example, can be expected to be short-staffed for almost another eight weeks.

3 employees X 30% X 22 days/month X 2 months X 1 week/5 days = 7.92 weeks

Averages are useful tools, and so are back of the envelope calculations of staffing requirements.  Effective staffing, however, requires active awareness that averages will be exceeded about half of the time, random glitches will occur, and staff members are predictably absent at least part of the time.