Practice Economics

The To-Don’t List, Part 2


The To-Don’t List, Part 2

A couple of additions to the list that I started last month, in which I mentioned the problems associated with fixed-duration day shifts, a contractual vacation provision, tenure-based salary increases, poor roles for NPs and PAs, and blinded performance reporting. I think most practices would be better off without those things, and this month I’ll add a few more to the list.

I readily admit that there are some relatively rare situations in which the following things might be a good idea. But most hospitalist practices should think about alternatives.

Extra shifts. I think every hospitalist should have, within reason, a chance to work more or less than others in an HM group. And, of course, compensation should match the amount of work. So those who want to work more than the normal, or contractually required, number of shifts should have at it. But I think it is best to avoid categorizing the work into “normal” shifts and “extra” shifts. Essentially, all shifts should be thought of as “normal.”

What is the problem with having an “extra” shift category? It pretty reliably leads to confusion.

This confusion is easiest to illustrate with an example. Consider Dr. Krause, a hospitalist working in a practice with a seven-on/seven-off schedule. However, the first week in July, she works only six days, but she plans to “pay that back” and more when she works a 10-day stretch two months hence. So far, this sounds easy. By the end of September, Dr. Krause will have worked two extra shifts.

But when another hospitalist in Dr. Krause’s group is out sick in August, several hospitalists in the group rearrange their schedules to fill in. In September, Dr. Krause works the two days that she originally was scheduled to be off and trades away three of the consecutive days she was to work in September.

Why should the evening (swing) shift be the same duration as the day shift? Shouldn’t it be however long is necessary?

While it will be clear to Dr. Krause that she will be “even” in the number of shifts worked at the end of September, it probably isn’t clear to anyone else. The person who determines payroll will probably have a really hard time figuring out whether Dr. Krause is to be paid extra for “extra” shifts during any two-week pay period.

The most reliable way to figure out if a doctor worked extra shifts is to add up all worked shifts at the end of the year. But that would mean waiting until the end of the year to compensate the doctor for any extra shifts worked. And most docs would find that really unattractive.

It would be easy enough to just add up the shifts worked every pay period (usually two weeks) and compensate for any above the number expected, but that would then require lowering the salary for any pay period in which the doctor works fewer than the expected number. Although it might not be popular, I see this as the best arrangement. That is, just pay per shift so that there is no need to keep track of whether any particular shift is “normal” or “extra.”

Even if this illustration doesn’t convince you how messy it can be to keep track of extra vs. normal shifts, trust me on this one. It causes lots of problems for lots of physician practices. If your practice is among the few that has a clear-cut system that doesn’t confuse those in payroll, then stick with it.

Shift duration symmetry. Rarely is there a reason to keep every shift the same duration.

Let’s consider a common scenario. A small hospitalist group has a schedule that consists of a 12-hour day shift followed by a 12-hour night shift. As patient volume grows, the day-shift doctor(s) often have to stay after their shift to finish the initial care of new referrals, or the night doctor typically starts their shift with several patients in the ED awaiting admission. So the practice makes a good decision and creates an evening shift, which often is referred to as a “swing shift.” And because all existing shifts are 12 hours, the evening shift will be 12 hours, right?

Not so fast.

Why should the evening shift be the same duration as the day shift? Shouldn’t it be however long is necessary? Practices of no more than about 15 FTEs typically require an evening shift of only about four to six hours. It should start an hour or so before the last day doctor should be finishing work; it should continue until the night doctor has resolved the backlog of patients. As the practice volume grows, it will probably be necessary to lengthen the evening shift until it eventually reaches the same length as other shifts. But there is almost never a real workload or patient-care reason that the shift length needs to be the same duration as other shifts when it is first put into place.

While an evening shift should have a clearly defined start time, it will work best if the end of shift time is left loose and is based on just how busy that night it. For example, it might be reasonable to have the evening doctor accept their last new referral no later than a specified time (10 p.m. is the deadline in my hospitalist group). The swing shift can leave after completing the care of that patient and addressing any other issues that came up during the shift. Some nights, that will mean the evening doctor can leave at 10 p.m.; other nights, it might be 11 p.m. or midnight.

While we’re talking about it, there is no clear reason day and night shifts need to be the same length, either. It is fine to make both 12 hours long, but that isn’t the only reasonable option.

Of course, your compensation formula might influence what can be reasonably done with shift lengths. But if a practice compensates the doctors in a way that requires that all shifts be identical in duration, then the compensation method needs another look. TH

Dr. Nelson has been a practicing hospitalist since 1988 and is co-founder and past president of SHM. He is a principal in Nelson Flores Hospital Medicine Consultants, a national hospitalist practice management consulting firm ( He is course co-director and faculty for SHM’s “Best Practices in Managing a Hospital Medicine Program” course. This column represents his views and is not intended to reflect an official position of SHM.

Next Article:

   Comments ()