HLTH5693 Module 3 capstone data analysis example

One Cause Recorded Under Two Names: What 1,008 First Cases and Two Weeks of Observation Can Honestly Say About Late Operating Room Starts

Student Name

American College of Education

HLTH5693: Capstone Experience for Healthcare Administration

Module 3 Assignment

Instructor Name

February 5, 2029

The Data and Their Weaknesses

The analysis draws on three sources. The surgical information system provided scheduled start and patient-in-room times for all 1,008 weekday first cases from April through September 2028, with room, service and day of week. Circulating nurses enter a delay reason code for late cases, but codes were present for only 423 of the 595 late cases, 71 percent, and each late case can receive only one code. To see what the codes miss, the capstone observed 80 first cases over two weeks in November 2028, recording when each preoperative step was completed: patient arrival, nursing assessment, anesthesia assessment, the surgeon's update of the history and physical, signed consent, site marking and ready for the operating room. Staffing cost data came from finance.

Each source has a weakness. Missing codes could be missing in a biased way; delays with no obvious owner may be the ones left blank. Single codes hide shared causes. The observation sample is small, from one season, and staff may have behaved differently while observed. Data do not become more reliable by being summarized; the weaknesses must be carried into every conclusion.

How Often, How Late and Where

First cases started on time 41 percent of the time, with a median delay among late cases of 14 minutes. The analysis also examined the spread of start times, as Tiwari et al. (2018) did in their study: the interquartile range of the difference between scheduled and actual entry was 17 minutes, wider than the 13 minutes those authors reported before their improvement program, which suggests the process here is less consistent rather than simply shifted late. Performance varied by day, with Mondays worst at 29 percent on time and Fridays best at 52 percent, and by service, with orthopedics at 32 percent and ophthalmology at 63 percent.

These differences are large enough to be real rather than chance, given 1,008 cases, and they point to where the process is least controlled. Confidence in this finding is high, because it rests on timestamps recorded automatically rather than on judgment.

What Causes the Delays

Among the 423 coded late cases, the codes were: surgeon late, 38 percent; history and physical or consent incomplete, 24 percent; patient late, 11 percent; anesthesia, 9 percent; equipment or room, 8 percent; and other, 10 percent. Taken at face value, surgeons and paperwork are two separate causes, consistent with the pattern Saul et al. (2022) reported at a community hospital, where surgeon practices and preoperative processes led.

The observation changes the interpretation. In 41 of the 80 observed first cases, the patient was not ready for the operating room until after 7:25 a.m. In 29 of those 41, the last step completed was the surgeon's update of the history and physical or the signing of consent, both of which require the surgeon in person, and the surgeon arrived in the preoperative area at a median of 7:21 a.m. Nurses then coded these delays either as surgeon late or as paperwork incomplete, depending on the day. The two leading codes, 62 percent of coded delays together, therefore appear to describe largely one cause: surgeons arriving too close to 7:30 to complete tasks that only they can do. Confidence in this interpretation is moderate, because it rests on 80 cases, but it is consistent across both weeks and all services observed.

Checking the Missing Codes

Before trusting the codes, the analysis compared the 172 late cases without a code with the 423 that had one. The uncoded cases were shorter delays, a median of 8 minutes against 16 for coded cases, and were more common on Fridays and in ophthalmology, the day and service with the best performance. That pattern suggests nurses tend to skip the code when a delay is small, not when its cause is embarrassing, which is reassuring: the missing codes are unlikely to hide a large cause of long delays. It also means the coded distribution describes the longer delays better than the shorter ones.

The analysis also checked whether the observation weeks were typical. The on-time rate during the two observed weeks was 44 percent, close to the six-month rate of 41 percent, and the mix of services was similar. The observed weeks are therefore a reasonable, though small, window on the usual process. The possibility that staff moved faster while observed cannot be ruled out, but if preoperative staff moved faster while being timed, the share of delays left waiting on the surgeon would, if anything, be understated rather than inflated.

What Savings Are Realistic

The analysis applied the method of Dexter and Epstein (2009), counting only rooms whose cases and turnovers regularly exceed eight hours. Only three of the eight rooms, two orthopedic and one general surgery, ran more than eight hours on most days. The loaded cost of the hospital-employed staff in each room, a circulating nurse and a surgical technologist, is about $2.10 per scheduled minute; anesthesia is paid through a fixed contract and does not change with minutes. If a realistic program reduced mean first-case tardiness by 10 minutes in those three rooms, reduced staffed time would be about 33 minutes a day, worth only about $17,000 a year over 250 days.

That figure is far smaller than the $412,000 the hospital spent on perioperative overtime last year, and far smaller than the figure produced by multiplying all late minutes by a full cost of operating room time. The financial case for better starts is modest. The stronger case is strategic, retaining surgeons and their cases, which the data can describe but not value. Confidence in the savings estimate is moderate: the method is established, but the 10-minute improvement is an assumption.

What the Data Support, Suggest and Cannot Say

The data support three conclusions: first cases start late more often than not, with wide variation by day and service; the process is inconsistent, not merely late; and the direct labor savings from improvement are modest. The data suggest, with moderate confidence, that surgeon arrival relative to tasks only surgeons can complete is the main shared cause, and that Mondays and orthopedics are the best places to start. The data cannot say whether late starts cause surgeons to move cases elsewhere, whether they affect patient outcomes or experience, or how much of the overtime they cause, since overtime is not recorded by cause.

These boundaries shape the recommendations in the next module. Recommendations should target the shared surgeon-readiness cause and the variability of the process, be justified mainly by strategy and reliability rather than by savings and include better data collection, especially multiple delay codes per case and a measure of preoperative steps, so that the hospital can test its conclusions over time. The most useful finding is often the one that shows the organization has been counting one problem as two.

References

Dexter, F., & Epstein, R. H. (2009). Typical savings from each minute reduction in tardy first case of the day starts. Anesthesia & Analgesia, 108(4), 1262-1267. https://doi.org/10.1213/ane.0b013e31819775cd

Saul, B., Ketelaar, E., Yaish, A., Wagner, M., Comrie, R., Brannan, G. D., Restini, C., & Balancio, M. (2022). Assessing root causes of first case on-time start (FCOTS) delay in the orthopedic department at a busy level II community teaching hospital. Spartan Medical Research Journal, 7(2), Article 36719. https://doi.org/10.51894/001c.36719

Tiwari, V., Ehrenfeld, J. M., & Sandberg, W. S. (2018). Does a first-case on-time-start initiative achieve its goal by starting the entire process earlier or by tightening the distribution of start times? British Journal of Anaesthesia, 121(5), 1148-1155. https://doi.org/10.1016/j.bja.2018.05.043