Optimizing second shift OR staffing - operating room staffing

AORN Journal,  April, 2003  by Franklin Dexter,  Richard H. Epstein

• E-mail
• Print
• Link

The mean (ie, average) labor cost for second-shift staffing for the day equals the mean of labor costs for each hour during the day. The mean hourly labor cost for a four-week period equals the mean of the labor costs for all second-shift hours during the four-week period. This process of averaging during four-week periods avoids the statistical problem of autocorrelation; specifically, OR workload tends to be correlated between successive hours and weeks. (6) Second-shift workload on one Monday may be correlated to second shift workload on the following Tuesday or Monday; however, the average workload during one four-week period is unlikely to be correlated to the average workload during the next four-week period. That is why it is helpful to calculate labor costs over four-week periods.

For example, if during the four-week period between Feb 3, 2003, and Feb 28, 2003, there were 55 hours for which there was one OR in use for at least part of an hour, 23 hours for which there were two ORs in use for at least part of an hour, and four hours for which there were three ORs in use for at least part of an hour, if one OR had been staffed regularly during the second shift, the mean daily second-shift labor cost for that four-week period would have equaled 10.72 multiplied by the cost of each regularly staffed hour where 10.72 equals

[(1 regularly scheduled OR x 8 hours per day x 20 days) + (23 hours x 1 additional OR in use x 1.75) + (4 hours x 2 additional ORs in use x 1.75)]/20 days.

For each four-week period, the mean hourly labor cost is calculated as if a specific number (eg, zero, one, two, three) of second-shift teams were scheduled to work between 3 and 10:59 PM. For each four-week period, the difference is taken between the mean hourly labor cost of having one rather than zero second-shift teams, two rather than one team, three rather than two teams, and so on.

For instance, the example using Feb 3, 2003, through Feb 28, 2003 previously showed that for this four-week period, with one OR regularly staffed, the mean daily second-shift labor cost would have equaled 10.72 multiplied by the cost of each regularly staffed hour. If the regular staffing was zero ORs, then all ORs would have been covered by perioperative nurses and nurse anesthetists who stayed late on call. The mean daily second-shift labor cost would have equaled 10.08 multiplied by the cost of each regularly staffed hour where 10.08 equals

[(55 hours x 1 additional OR in use x 1.75) + (23 hours x 2 additional ORs in use x 1.75) + (4 hours x 3 additional ORs in use x 1.75)]/20 days.

The difference in mean daily labor cost between having one rather than zero second-shift teams would be 0.64 multiplied by the cost of each regularly staffed hour.

In addition, afternoon OR workload can vary seasonally. For example, workload may be busier in the summer than winter. (7) To check for this type of variability, pair-wise differences were viewed graphically using a nonparametric smoothing line drawn using the LOWESS algorithm. (8) The best second-shift staffing solution has a difference curve that most closely straddles the baseline of a zero difference.