By  |  January 21, 2011 | 

This is the third installment on staffing an HMG:

To dig a little deeper, we can now review the daily admission cycles for the group (Figure 2):

Note that the X axis represents hour of the day and the Y axis represents number of admissions over a one year period. At the top, the day is divided into three segments: 0100-0700; 0800-1800; and 1900-2400. The highest volume period for the group is during the evening, however, admission volumes are constant between noon and 2300. Based on this cycle, the group has chosen to add a physician for a swing shift between noon and midnight in addition to a nocturnist between 1900-0700, thus providing coverage during peak admitting hours.

Figure 2


In total, the program staffs 72 work hours per day. The group covers the program 365 days per year, hence 26,280 hours per year must be covered.
If a full time hospitalist is defined as 2000 hours per year (actual hours worked after accounting for paid time off), this program needs 13.1 FTEs, or 26,280 hrs/2000 hrs per FTE.
What happens when the hospital volumes are low? Well, let’s look at volumes by day of the week to see if we can cut back on the number of providers needed on any particular day (Figure 3).

Figure 3

In this figure, we have the number of admissions over a year (Y-axis) plotted against day of the week (X-axis). Based on the drop in admissions over the weekend, the group may decide to cut back staffing, however, note that this is only the new admission volume. Figure 1 demonstrated that total daily census is variable. Based on this data, the group may choose to put its swing shift physician “on call” over the weekend.
How does the program deal with unanticipated volume surges or rapid and sustained growth? This can be managed by opening up more shifts and paying overtime. Alternatively, a cadre of moonlighters could be kept in order to fill in these needs. If the market and culture allows, incrementally adding non-physician providers to absorb the volume may work for some programs.
Figure 4 is an example of how using the threshold volume method can help  predict staffing needs.

Figure 4

This graph represents 9 months of data for a program with 10 FTEs. On the Y-axis is AM census and the blue line is average daily census at 0800. On the right axis is number of days in the month the threshold was exceeded. The red line represents the number of days per month the threshold exceeded 14 patients per provider and the orange line represents the number of days the 16 patient threshold was exceeded.  Based on the slope and fit of the orange trend line, more FTEs are warranted.
Staffing a hospitalist program is a complex endeavor dependent upon multiple variables. This method can help a group ensure appropriate patient ratios, mitigate physician burnout, and plan for growth.

Thanks to John Nelson whose infinite wisdom continues to shape our care delivery systems.

Figure 5

14 Child Threshold?

A family of 14 children and their mother (holding baby) in San Lucas Guatemala. Yours truly and 2 other missionaries (center).

One Comment


Leave A Comment

About the Author:


Related Posts

By  | June 7, 2018 |  0
Everywhere I go these days, one of the top questions on the minds of hospital leaders and hospitalists alike is, “How can we improve hospitalist patient satisfaction scores?” It’s a dilemma. There are people who know way more about this subject than me, but I’m not aware of anyone who has really cracked the nut. […]
By  | March 27, 2018 |  0
There is an orange tree in our backyard planted around the birth of our first child. It thrives 11 years later, providing a plentiful bounty of fruit each winter. Nearby, a mango tree was planted when our second daughter was born. That tree never made it; the roots didn’t take hold, and it was gone […]
By  | February 28, 2018 |  0
“We are playing the same sport, but a different game,” the wise, thoughtful emergency medicine attending physician once told me. “I am playing speed chess – I need to make a move quickly, or I lose – no matter what. My moves have to be right, but they don’t always necessarily need to be the […]