An operating room audit to examine for patterns of staff entry/exit: pattern sequencing as a method of traffic reduction

Abstract

Background:

Surgical site infections (SSIs) are estimated at over 16,000 annually and cost hospitals an estimated $1.6 billion per year. Therefore, most operating rooms (ORs) seek methods to reduce the risk of SSI, especially during the intraoperative period. Prior work has established a link between excess traffic through the OR and increased microbial counts, which create a higher risk for SSIs.

Aim/Objectives:

To identify patterns of staff entry into the OR to further reduce the risk of SSIs after total joint arthroplasties.

Methods:

Researchers directly observed 31 total joint arthroplasties, recording every instance the door to the OR suite opened and the personnel, reason for opening and timing during surgical incision. Researchers then utilised the sequential data analysis to search for patterns.

Results:

Despite expected patterns in staff movement during the patterned surgery, researchers found no significant patterns to staff movement during total joint arthroplasty.

Discussion:

This study’s results suggest purposeful education targeted to circulating registered nurses could induce purposeful creation of traffic flow patterns to further decrease traffic and risk of SSI.

Conclusion:

There is no singular pattern to entering and exiting the OR during surgery. Thus, a single-solution approach is not recommended.

Background

A multitude of interventions aim to reduce the risk for surgical site infections (SSIs) through environmental control of microbes in the operating room (OR). Standard methods for protecting the patient from potential microbes from the clinical team include scrubs, facial masks and hair nets. To protect patients from their own microbes, the Centers for Disease Control and Prevention (CDC), in collaboration with the centers for Medicare and Medicaid Services, published the Surgical Care Improvement Project (SCIP), a guideline for standardised quality of care and antimicrobial standards in the OR (The Joint Commission, 2010). The CDC’s Guidelines for Environmental Infection Control in Health-Care Facilities are also in place and recommend the use of positive-pressure ventilation systems and limiting the passage of personnel through the OR doors as much as possible (Sehulster et al., 2004).

Air quality and temperature in the OR forms an essential component in the prevention of SSIs. Recent studies have focused on the role of foot traffic on microbial counts in the OR. One study found a direct correlation between the number of people in the OR and the amount of aerosolised particles in the room, as well as a minor correlation between the number of door swings and the amount of aerosolised particles in the OR (Rezapoor et al., 2018). Another study noted a significant 2.5% increase in colony forming units (CFUs/m³) per person in the OR and a 0.25% increase per door swing (Stauning et al., 2018).

The reason for passage through the OR is a highly studied variable and carries significant potential for interventions aimed at reducing unnecessary traffic. In a study across multiple surgical specialties, researchers found that requests for information constituted 27%–54% of reasons for passage through the OR (Lynch et al., 2009). Another study focusing solely on OR traffic during total joint replacement found that the need for supplies accounted for 23% of all door openings, although 47% of reasons remained unidentifiable (Panahi et al., 2012). Air quality outside of the OR is noted to have higher contaminants than inside the OR (Tang and Wan, 2013). Given that upwards of 45% of door openings are attributable to equipment needs (Loison et al., 2017), the need to investigate disparity in reasons for passage through the OR warrants further investigation.

As tasks in the OR are highly divided by role, the role of persons opening the OR door especially correlates with reason for entry. Those entering and exiting the most frequently likely perform the most commonly needed tasks. Studies across multiple specialties show that circulating nurses are responsible for most (25%–41%) of total door swings (Lynch et al., 2009; Panahi et al., 2012). The Association of periOperative Registered Nurses (AORN) recognises the role nurses play in preventing the unnecessary risk of SSIs and published a guideline through continuing education that established basic principles OR nurses should follow to help prevent SSIs (Spruce, 2014). These guidelines establish protocols for the patient such as diathermy and antibiotic prophylaxis as well as preventative measures for the nurses to reduce environmental microbes, such as facial masks and removal of nail polish. However, despite evidence that movement of nurses in and out of the OR could greatly contribute to an increased risk of infection, the AORN does not address the need for nurses to regulate their traffic through the OR (Spruce, 2014). Further research confirming the role of nurses in high OR traffic could lead to more attention and interventions in this matter. The aim of the present study is to explore the hypothesis that there is an identifiable observable traffic pattern into and out of the OR.

Methods

This prospective observational pilot study with pragmatic sampling observed for patterns of traffic in the operating room at a large metropolitan hospital. This study received institutional review board (IRB) exemption from the university hospital. Participants in the study were the clinical team of the particular surgery pragmatically sampled for observation. Members of the surgical team were made aware on arrival that they were being observed for the purpose of research and that apart from the attending surgeons’ names, no identifying information would be collected throughout the entire process. Staff members cooperation indicated consent to participate unless they chose to withdraw by verbal indication to the researcher. No participants indicated a desire to withdraw.

A dedicated researcher remained in the OR suite and observed 31 OR cases exclusive to primary total joint arthroplasty (TJA) or hemiarthroplasty and recorded data from the opening of the first sterile instrument case to closure of the incision site and dressing application. The research team member recorded data at each opening of the OR doors on a standardised sheet. Standardised variables recorded for staff member passing through included circulating nurse, scrub nurse, facilitator, anesthesia, surgeon, resident, vendor, X-ray and OR attendant. Variables recorded included basic information about type of arthroplasty, date of case, OR suite, attending surgeon and case number, as well as length of case. The researcher also recorded the door used, the role of persons entering/exiting, reason for opening the door and if they entered or exited. Each researcher in the OR received training on data collection from the study’s principal investigator (PI) to maintain consistency across observations.

The researcher recorded the reason for entrance/exit in very specific categories, some of which contained further subcategories. ‘Supply’, ‘instrument’, ‘suture’ and ‘equipment’ were further recorded as pick list or as one-offs. ‘Break’ was further subcategorised as meal, bathroom or other. ‘Communication’ was further recorded per communication surrounding the current case, case to follow or another operating room. Additional options included ‘bringing patient in’, ‘taking bed out’, ‘signing debrief’ and ‘relief’. Finally, an ‘Other’ option was available for circumstances of entry/exit that did not fit into any category. The research team member recorded certain times to assist in determining the length of case, as well as whether passage occurred before incision, during surgery or after closure of the incision but before the dressing application to determine the surgical timepoints of door openings.

Researchers maintained collected data sheets in a locked drawer in a locked office until they were transcribed into a computerised spreadsheet to be analysed using the statistical software SAS. Sequential data analysis was utilised to determine potential patterns in staff traffic.

Results

This research study explored 31 case observations of staff during TJA or hemiarthroplasty, over a six-month period. A total of 2208 data points from 31 cases were analysed and were in the range of 89–300 min. The average length of case lasted 159.9 min, averaging 76 discrete observations per case, or every 2.1 min (28.5/h). Two cases were rejected for use in further analysis due to missing information on timing of incision necessary for staff traffic sequencing. Descriptive statistics on the remaining 29 cases included the frequency of the number of staff entering before and after incision (Figure 1). The frequencies of personnel clustering, determined by the type of staff entering the OR after another staff member, were also determined (Table 1). There was no statistically significant correlation between the length of the case and the number of door openings (P = 0.2403). There was a weak negative correlation for door openings over time with fewer door openings towards the end of the study (r² = 0.01, P < 0.001). After removing an outlier case, length of surgery had a weak negative relation (P = 0.3481). In both scenarios, neither were statistically significant and thus we fail to reject the null hypothesis that the length of the case does not predict the number of people.

Figure 1.

Staff entry rates per surgical timepoint.

Table 1.

Personnel clustering.

	Next person through
First person through	Registered nurse	Scrub	Facilitator	Surgeon	Resident	Anesthesia	Vendor	OR attendant	Imaging	Missing data	Total
Registered nurse	336 (15.48)	35 (1.61)	37 (1.70)	32 (1.47)	25 (1.15)	80 (3.68)	63 (2.90)	0 (0.00)	5 (0.23)	2 (0.09)	622 (28.65)
Scrub	31 (1.43)	88 (4.05)	9 (0.41)	8 (0.37)	25 (1.15)	15 (0.69)	1 (0.05)	3 (0.14)	1 (0.05)	2 (0.10)	195 (8.98)
Facilitator	46 (2.12)	9 (0.41)	76 (3.50)	13 (0.60)	8 (0.37)	24 (1.11)	19 (0.88)	1 (0.05)	3 (0.14)	0 (0.00)	199 (9.17)
Surgeon	32 (1.47)	13 (0.60)	7 (0.32)	39 (1.80)	26 (1.20)	19 (0.88)	35 (1.61)	2 (0.09)	8 (0.37)	1 (0.05)	182 (8.38)
Resident	21 (0.97)	7 (0.32)	7 (0.32)	23 (1.06)	31 (1.43)	21 (0.97)	14 (0.64)	4 (0.18)	3 (0.14)	2 (0.09)	133 (6.13)
Anesthesia	61 (2.81)	20 (0.92)	26 (1.20)	26 (1.20)	18 (0.83)	147 (6.77)	49 (2.26)	8 (0.37)	9 (0.42)	2 (0.10)	366 (16.86)
Vendor	68 (3.13)	10 (0.46)	31 (1.43)	30 (1.38)	14 (0.64)	41 (1.89)	152 (7.00)	8 (0.37)	4 (0.18)	2 (0.10)	362 (16.67)
OR attendant	7 (0.32)	3 (0.14)	0 (0.00)	3 (0.14)	0 (0.00)	7 (0.32)	8 (0.37)	19 (0.88)	1 (0.05)	0 (0.00)	48 (2.21)
Imaging	15 (0.69)	1 (0.05)	2 (0.09)	6 (0.27)	1 (0.05)	3 (0.14)	6 (0.28)	1 (0.05)	8 (0.50)	1 (0.05)	47 (2.16)
Missing data	3 (0.14)	1 (0.05)	0 (0.00)	1 (0.05)	1 (0.05)	1 (0.05)	1 (0.05)	0 (0.00)	0 (0.00)	5 (0.24)	13 (0.63)
Total	621 (28.60)	187 (8.61)	196 (9.03)	187 (8.61)	133 (6.13)	369 (16.72)	363 (16.72)	48 (2.21)	50 (2.30)	17 (0.82)	2171 (100.00)

Values are given as n (%).

Frequency missing = 6.

OR, operating room.

The majority (52.9%) of all traffic occurred during the pre-incision point when the room and patient were being prepared for surgery. Of the traffic, 43.30% occurred after the incision time, when the site was open and most exposed to potential pathogens. Circulating nurses accounted for the largest percentage (28.72%) of all traffic. Vendors and anesthesia team members accounted for another 16.61% and 16.88% of traffic, respectively (Figure 1). Nurses also notably had increased traffic by 6.48% during the open incision period compared to the pre-incision period. Vendors also created a large jump in traffic (25.14%) during the time of incision, although nurses still accounted for the most traffic at every surgical timepoint.

Supplies, instruments, sutures and equipment, which are ideally gathered and brought in together at the start of a case, accounted for 21.64%, 5.45%, 2.18%, and 5.62% of all door swings, respectively, totaling over one-third (34.89%) of all door swings. Bathroom and meal breaks combined accounted for another 9.61% of reasons for entrance and exit. Another 9.08% of door swings were for communication and a further 14.76% were classified as ‘Other’. The remaining reasons for entering and exiting the OR door were typically single passes by a team, such as bringing the patient into the OR and removing the bed, signing the debrief and shift change, and have less opportunity for targeted interventions.

This pilot study of 31 TJAs at a single metropolitan hospital should be repeated using a greater sample size across multiple facilities to allow better generalisation of results.

Discussion

Our results that extend those of prior research find that door openings (door swings) happen on average about once every 2 min (Bashaw and Keister, 2019). Finding that the data support that we reject the null hypothesis of an observable traffic pattern sequence to OR staff entry and exit during TJA informs future efforts at infection prevention. To our knowledge, this is the first staff traffic pattern-sequencing study during arthroplasty. Only weak clusters and associations were found from inter-staff relations as they interact together coming in and out of the OR.

Given the literature suggesting that reducing OR door openings may lower the risk of SSIs, the present study is unique in that it searches for potential traffic patterns by staff type. While organised and time-sectioned events are expected to have predictable patterns of personnel traffic, various analyses identified a fundamental lack of any pattern in staff traffic flow in the OR during TJA. In contrast to the expected finding that persons were most likely to follow others and roles clustered with other roles, there were no identifiable patterns (Table 1). Scrubs, facilitators, anesthesia providers, vendors, OR attendants and X-ray personnel were all most likely to be followed next into the room by a nurse, while nurses themselves most frequently clustered with anesthesia providers. Although statistically insignificant, these associations do highlight the potential for the circulating nurses to influence traffic between many personnel roles.

Another method of analysis examined how often one person left the OR and was immediately the next person to return to identify how many times a single person was responsible for sequential door openings. This information provides a guideline for the timing of personnel and indicates rapidity of movement in and out of the OR suites. The lack of collaboration likely accounted for a marked increase in the number of times the door was opened. Of 2208 observations, only 228 (10.34%) of all entries/exits consisted of two or more people passing together, with 1976 (89.66%) people passing through independently.

The findings raise questions about the individuality of cases. Many surgeons develop a comprehensive list of every required item for each individual surgery (commonly called a ‘pick list’ [PL]). The surgical team uses the PL to gather all the needed items for each case in advance, thereby increasing the efficiency of the room set-up and turnover times. Despite the wide use of PLs, the most common reasons for door swings were to retrieve items not brought into the room during opening. The majority of items brought in as supply, instruments, sutures and equipment was a ‘one-off’, meaning the item retrieved was an incidental not provided on the PL (Table 1). Although this indicates that the PL cart was being appropriately used, the individuality of cases suggests the need for similarly individualised interventions to reduce the risk of infection (Olson, 2019).

Interventions such as checklists are associated with improved outcomes associated with communication (Snavely et al., 2020). However, communication-related traffic was < 10% of all in-and-out movement. Although communication was not the primary reason for entry, opening the OR door and breaching a sterile environment for the purpose of communication can be specifically targeted to reduce traffic. The findings generated by the present study prompted administrative personnel in the OR at this hospital to evaluate new data-driven communication standards aimed at reducing unnecessary traffic through the OR suites.

The inability to find a foot traffic pattern is informative. The results extend findings that a one-size-fits-all solution is unlikely for this population (Panter et al., 2019). Developing multi-focused, data-driven interventions to reduce traffic will likely be a multi-pronged initiative. The role of the nurse in > 25% of door openings may be an essential component. A novel initiative to improve PL-related traffic would likely be most effective with cooperative education targeting both the circulating nurses and the attending surgeon to jointly update the PL for an accurate and comprehensive list to reduce traffic for incidentals. Vendor traffic accounted for the third highest traffic role in the OR, despite their primarily stationary role as advisor to the scrubs and surgeon. Further education aimed at improving vendor preparedness and timing could decrease needless trips to retrieve implants or to step out for patient privacy points.

The present study revealed new knowledge in staff sequence-patterning in the OR. With the knowledge that there is no readily discernable pattern to staff movement during TJA, purposeful education and arrangement could assist staff in developing a pattern of purposeful, minimal movement in and out of the OR. Such education would focus on taking opportunities to minimise traffic, such as asking physicians and residents to leave to scrub in at the same time, or having the circulating nurse prepare a list during opening of all the needed items, and then leaving only once to retrieve them all at once. While this activity would require purposeful and proactive thinking on the part of the surgical team, it has the potential to save lives and money by decreasing the risk for surgical site infections for patients.

Limitations

The Hawthorne Effect is the believed phenomenon that people alter their behavior based on the knowledge that they are being observed (Gottfredson, 2005). In the present study, participants were told that they and their traffic movements through the OR were being observed for the purpose of research. However, as in prior studies (Parikh et al., 2010), the staff’s knowledge of the observations appear to have minimal impact the results, as indicted by a minimal (r² = 0.01) but statistically significant decline in traffic over the 31 cases observed. This may be explained in that OR staff came to recognise that they were being observed and changed their behavior This leads the researchers to believe that the findings reflect as accurately as possible the true circumstances in the OR.

Despite providing multiple options as reasons for exiting or entering the OR and the ability to clarify reasons for leaving with staff, 14.76% of reasons remained accounted as ‘Other’. Although this may be due to observer inconsistency in categorising, or a lack of standardised options, all researchers received individual training from the project manager on proper data collection so the assumed inter-rater reliability leads the researchers to believe that their findings between researchers are accurate.

Conclusion

Surprisingly, there is not a specific pattern to when staff enter or exit the surgical suite during a surgical procedure. Nor is there a pattern as to who enters or exits the surgical suite. Given the inability to identify a singular pattern, it seems intuitive that solution must address multiple causes for entry and exit. Although a one-size-fits-all solution is attractive, it is unlikely that this would significantly alter the number of entry and exits. Future studies should seek to examine multiple interventions to increase the education surrounding the need to reduce door openings.