Using NNAPPS (Nighttime Nurse and Physician Paging System) to Maximize Resident Call Efficiency within 2011 Accreditation Council for Graduate Medical Education (ACGME) Work Hour Restrictions

2012 APDS SPRING MEETING

Using NNAPPS (Nighttime Nurse and Physician Paging System) to Maximize Resident Call Efficiency within 2011 Accreditation Council for Graduate Medical Education (ACGME) Work Hour Restrictions Jason B. Young, MD, Aaron C. Baker, MD, Judie K. Boehmer, RN, Karen M. Briede, RN, Shirley A. Thomas, RN, Cheryl L. Patzer, RN, Christina Pineda, RN, Gina A. Cates, RN, and Joseph M. Galante, MD Department of Surgery, University of California, Davis Medical Center, Sacramento, California OBJECTIVES: To assess if implementing Nighttime Nurse and Physician Paging System (NNAPPS) would improve nurse and physician communication as well as reduce the number of nonurgent pages to residents taking overnight call. DESIGN: NNAPPS was implemented on the busiest General

Surgery and Transplant wards at our University Hospital. We conducted 2 prospective studies that logged pages received by on call surgery residents for 2-month blocks. The logs captured time, source, reason, and action resulting from pages. Independent reviewers determined urgency of the pages. Primary outcome measures were comparison of average nonurgent pages, total pages and total pages per patient during a night shift between the NNAPPS ward and all other wards that care for surgical patients. SETTING: University teaching hospital. PARTICIPANTS: General surgery residents working over-

existed between average nonurgent pages (0.46 vs 2.14), total pages (3.69 vs 6.14) and total pages/patient during a shift (0.38 vs 0.68) when comparing pre- and post-NNAPPS data. CONCLUSIONS: NNAPPS significantly reduced nonurgent pages, total pages and pages per patient during a night shift compared to services with conventional systems. Streamlined paging systems lead to more efficient communication between providers and decrease the nonurgent pages to residents. NNAPPS continued high standards of patient care and improved sleep patterns for residents. (J Surg 69:819-825. © 2012 Association of Program Directors in Surgery. Published by Elsevier Inc. All rights reserved.) KEY WORDS: ACGME, work hour restrictions, resident

work hours, duty hours, paging system, surgery resident COMPETENCIES: Patient Care, Professionalism, Interper-

sonal and Communication Skills, Systems-Based Practice

night call shifts on nine surgical services. RESULTS: In both studies combined, there were a total of 107

night shifts during which 771 pages were received. Total census was 1179 patients. Nurses initiated most pages (67%). Eight percent of pages interrupted patient care, while 40% of pages interrupted resident sleep. Most pages resulted in either a “new order” (39%) or “patient assessment” (22%), while 36% resulted in “no action.” Most pages (56%) were “urgent,” 25% “nonurgent,” and 19% “unable to determine urgency.” Regarding the Transplant ward, significant differences (p ⬍ 0.05)

Correspondence: Inquiries to Joseph M. Galante, MD, Division of Trauma and Emergency Surgery, University of California, Davis Medical Center, 2315 Stockton Blvd, Room 3012, Sacramento, CA 95817; fax: (915)-734-2724; e-mail: joseph.galante@ucdmc. ucdavis.edu

INTRODUCTION In July of 2011, the Accreditation Council for Graduate Medical Education (ACGME) implemented new resident duty hour revisions further limiting the number of hours for house staff.1 This has placed increased demands on balancing patient care duties with receiving adequate education. A clear solution to creating an environment that facilitates appropriate patient care and overnight coverage, while providing residents with adequate experience, education and rest remains difficult to ascertain. In a 2008 statement by the Institute of Medicine, recommendations were made stating that resident shifts longer than 16 hours should include an uninterrupted 5-hour sleep period.2 With regards to the current environment of daunting resident workloads, the likelihood of a 5-hour uninterrupted sleep pe-

Journal of Surgical Education • © 2012 Association of Program Directors in Surgery 1931-7204/$30.00 Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jsurg.2012.08.010

819

riod during a shift lasting more than 16 hours may not be feasible. Instead, the ACGME has recommended that residency programs adopt alertness management and fatigue mitigation strategies to manage the potential negative effects of fatigue on patient care and learning, such as strategic napping or back-up call schedules.1 Paging of residents occurs frequently at night by a variety of hospital staff, including nursing and consulting physicians.3,4 Residents receive at least 1 page per hour during the night and many of these pages are for nonurgent issues and require no action by the residents.3-5 Frequent paging has been shown to be disruptive to patient care, rounding, educational conferences and sleep even though it remains the primary means of communication throughout the hospital.3-7 Multiple pages to residents may produce an environment in which attending to necessary work becomes inefficient. Potentially, each interruption secondary to answering pages may result in an increased chance of being subsequently paged to perform a task that may have been completed if there had been fewer delays from the frequent pages.4 It has been reported that house staff officers may be interrupted every 7-11 minutes while trying to perform and complete an initial history and physical examination, which may be detrimental to establishing a therapeutic relationship with the patient.8 Lurie et al. additionally noted that residents get an average of 2-5 hours of sleep per night while on call and are interrupted every 40-86 minutes during that sleep period.8 We decided to evaluate paging patterns on wards and intensive care units that care for general surgery patients at our University Teaching Hospital. To comply with ACGME regulations, we developed and implemented a new nighttime paging strategy to improve communication between patient care providers. We hypothesized that implementing a new streamlined nighttime paging system would result in improved nurse and physician communication by reducing the number of nonurgent pages to on call residents while maintaining safe and efficient patient care.

METHODS Approval for this study was obtained by the University of California, Davis Institutional Review Board. Resident participation in the study was voluntary and the completion of pager logs was kept anonymous. No additional financial compensation was awarded for the completion of pager logs. University of California, Davis Medical Center (UCDMC) is a 645 bed, acute care, University Teaching Hospital with 11 wards (323 beds) and 7 Intensive Care Units (ICUs) (107 beds) designated solely for surgical or medical/surgical patients. UCDMC is a major referral center for 6 million residents across Northern and Central California and operates inland Northern California’s only Level I Trauma Center for both adult and pediatric emergencies. The primary means of communication between hospital personnel is via an alphanumeric paging system (Unication Alpha Elite; Unication Co., Ltd., Burnaby, BC, Canada, American Messaging Services, LLC, Lewisville, TX). 820

We developed the Nighttime Nurse and Physician Paging System (NNAPPS) over a 2-month period in 2011 by nursing and surgery personnel. NNAPPS consists of screening all pages between 7 p.m. and 7 a.m. by the Charge nurse in which emergency issues are paged out immediately, urgent calls are batched and nonurgent calls are deferred until morning rounds. A logbook is kept on the ward for nonurgent patient care issues, which is reviewed by the resident at 6 a.m., 12 p.m. and 6 p.m. The Charge nurse will periodically review the logbook to ensure that patient care issues are being adequately monitored and addressed. Additionally, a designated resident team member will round between 6 p.m. and 8 p.m. to address any outstanding patient care issues. We then implemented NNAPPS onto the primary and busiest General Surgery ward at our University Teaching Hospital on June 25, 2011, just before the July 2011 ACGME duty hour restrictions. A prospective pilot study was then performed that involved the logging of pages received by nighttime on call first postgraduate year (PGY-1) to fifth postgraduate year (PGY-5) residents within the Department of Surgery working on various surgical services between August and September 2011. To validate our pilot study findings, we conducted a second study by implementing NNAPPS onto our Transplant Surgery ward and compared data compiled over 2-week blocks each of pre- and postNNAPPS implementation between December 2011 and March 2012. The Transplant Surgery service at our University Teaching Hospital currently performs kidney and kidneypancreas transplants as well as performs all vascular access procedures rather than the Vascular Surgery service. Intern surgery residents take in house calls overnight and will subsequently be relieved of work duties after team rounds the following morning. PGY-2 to PGY-5 residents work overnight call shifts from home and will take part in patient care and resident education activities the following day. Logs were organized based on surgical service since the residents may have had surgical patients cared for on multiple wards. The pager log captured service, census, time of page, source of page, reason for page, and what action resulted from the page. The classification of pages as “urgent,” “nonurgent” or “unable to determine urgency” was determined by independent review of 2 chief surgical residents. We defined an “urgent” page as a page that would require adequate attention before the end of the resident’s night shift and not be deferred until the daytime shift, in which case the patient’s health or safety might be adversely affected by such a delay. Examples of “urgent” pages were pages that were mandated per nursing protocol to be sent to the primary team’s physician, such as abnormal vital signs and critical laboratory values, change in a patient’s mental status, low urine output, requests to speak to the patient or patient’s surrogate, requests by the nursing staff to assess the patient at bedside, trauma codes, nontrauma codes and pharmacy pages regarding improper medication orders. We defined a “nonurgent” page as a page that could be appropriately deferred until and adequately addressed during the daytime shift without affecting the patient’s health or safety in an adverse

Journal of Surgical Education • Volume 69/Number 6 • November/December 2012

manner. Examples of “nonurgent” pages were pages for telemetry order renewals, patient restraint order renewals, bowel care requests, electrolyte replacement requests when detailed electrolyte sliding scale orders were active in the patient’s order set and diet order requests when detailed diet orders were active in the patient’s order set. Primary outcome measures were comparison of the average number of nonurgent pages, total pages and total pages per patient during a night shift between the NNAPPS ward and all other wards that care for surgical patients without a specific paging system. A one-way analysis of variance (ANOVA) with Tukey’s honestly significant difference (HSD) post hoc test was used to compare the mean number of pages between services in the initial NNAPPS evaluation and a paired t-test was used to compare pre- and post-NNAPPS implementation mean number of pages on the Transplant Surgery service. A p value ⬍ 0.05 was considered statistically significant. To measure the degree of agreement between the 2 chief residents independently reviewing the urgency of pages, we applied the Cohen’s kappa (␬) coefficient of inter-rater agreement.

RESULTS Paging logs for the initial NNAPPS pilot study were completed by residents on a total of 8 surgical services: General Surgery, Burn, Cardiothoracic (CT), Surgical Intensive Care Unit (SICU) Team 1, SICU Team 2, Emergency General Surgery (EGS), Trauma Ward Team 1, and Trauma Ward Team 2. For the second NNAPPS study, residents on the Transplant Surgery service completed paging logs. All services within the Department of Surgery were asked to participate in the initial study. The Surgical Oncology, Vascular Surgery, and Transplant Surgery services declined to participate. Shortly after the initial study was concluded, we were notified that the Transplant Ward and Transplant Surgery service was considering adopting a novel overnight paging protocol between the nurses and physicians. After further discussion with the Transplant Ward nursing staff and Transplant Surgery service, it was decided they would adopt the NNAPPS model. We therefore decided to conduct an additional study comparing pre-NNAPPS and post-NNAPPS data from the Transplant Surgery service to further validate our results from the initial study. The initial NNAPPS pilot study characteristic data from the services combined is presented in Table 1. Pages were logged over 80 nighttime call shifts totaling 851 resident work hours. Total census was 921 patients. The combined number of pages logged was 637. Independent review of the pages demonstrated that 52.6% were classified as “urgent,” 24.2% as “nonurgent” and 23.2% of the pages as “unable to determine urgency.” Overall agreement between the 2 chief residents classifying the urgency of pages was found to be 97% and Cohen’s ␬ test also revealed significant agreement (␬ ⫽ 0.96). Most pages resulted in either a “new order” (37.4%) or “patient assessment”

TABLE 1. Characteristics of Initial NNAPPS Evaluation (All Services Combined) Characteristic

No. (%)

Services* Night shifts Hours Patients Pages Urgent pages Nonurgent pages Unable to determine urgency of pages Pages resulting in new order Pages resulting in patient assessment Pages resulting in no action Urgent Nonurgent Unable to determine urgency Unable to determine result of pages

8 80 851 921 637 335 (52.6) 154 (24.2) 148 (23.2) 238 (37.4) 158 (24.8) 219 (34.4) 74 (33.8) 117 (53.4) 28 (12.8) 22 (3.4)

NNAPPS ⫽ Nighttime Nurse and Physician Paging System. *Services: General Surgery; Burn; Cardiothoracic; Surgical Intensive Care Unit 1; Surgical Intensive Care Unit 2; Emergency General Surgery; Trauma Ward 1; Trauma Ward 2.

(24.8%), while 34.4% of pages resulted in “no action” taken by the residents and 3.4% of pages had a result that was unable to be determined. Of the pages resulting in “no action,” 33.8% were classified as “urgent,” 53.4% as “nonurgent” and 12.8% as “unable to determine urgency.” Of note, only 5.9% of pages from the service utilizing NNAPPS (General Surgery service) were classified as “nonurgent” compared with the 24.7% of combined pages classified as “nonurgent” from the wards and units not utilizing NNAPPS. Initial NNAPPS pilot study page origination data from the services combined is presented in Table 2. Most pages were initiated by nurses (63.7%), trauma codes (16.5%) and physicians (14.8%). The remainder of pages originated from nontrauma codes, emergency department, hospital bed control, patient calls, medical students, pharmacy, premedical students, radiology and respiratory therapy. Initial NNAPPS pilot study outcomes data from the comparison of services is presented in Table 3. The average number (mean ⫾ SD) of nonurgent pages per night shift received by the service utilizing NNAPPS was 0.05 ⫾ 0.22. Significant differences (p ⬍ 0.05) existed between the average number of nonurgent pages per night shift received by the service utilizing NNAPPS compared with the EGS, Trauma Ward Team 1, and Trauma Ward Team 2 services with patients cohorted on conventional wards or units without a specific paging system. The average number of total pages per night shift received by the service utilizing NNAPPS was 1.45 ⫾ 1.36. Significant differences (p ⬍ 0.05) existed between the average number of total pages per night shift received by the service utilizing NNAPPS compared with the Burn, SICU Team 2, EGS, Trauma Ward Team 1 and Trauma Ward Team 2 services with patients cohorted on conventional wards or units without a specific paging system. The average number of total pages per patient per night shift received by the service utilizing NNAPPS was 0.13 ⫾

Journal of Surgical Education • Volume 69/Number 6 • November/December 2012

821

822

1.10 ⫾ 0.47† 1.26 ⫾ 0.67† 0.90 ⫾ 0.44† 0.48 ⫾ 0.26 0.13 ⫾ 0.13

CT, Cardiothoracic Surgery; SICU, Surgical Intensive Care Unit; EGS, Emergency General Surgery. *Significant difference (p ⬍ 0.05) compared with General Surgery service. †Significant difference (p ⬍ 0.01) compared with General Surgery service. ‡Service Utilizing NNAPPS.

0.80 ⫾ 0.32* 0.79 ⫾ 0.37*

11.20 ⫾ 6.13† 1.45 ⫾ 1.36

0.31 ⫾ 0.14

14.56 ⫾ 5.22† 11.83 ⫾ 2.86† 10.20 ⫾ 4.21† 8.25 ⫾ 3.01* 7.50 ⫾ 3.70

2.20 ⫾ 2.34

Nonurgent pages per night shift Total pages per night shift Total pages per patient per night shift

Outcome

0.05 ⫾ 0.22

6.83 ⫾ 2.92

4.33 ⫾ 1.97* 5.22 ⫾ 3.35† 3.8 ⫾ 1.79* 1.50 ⫾ 1.31 0.50 ⫾ 0.58

SICU 1 (Mean ⴞ SD) TABLE 3. Outcomes of Initial NNAPPS Comparison by Service

0.13. Significant differences (p ⬍ 0.05) existed between the average number of total pages per patient per night shift received by the service utilizing NNAPPS compared with the SICU Team 1, SICU Team 2, EGS, Trauma Ward Team 1, and Trauma Ward Team 2 services with patients cohorted on conventional wards or units without a specific paging system. The second NNAPPS study characteristic data from the Transplant Surgery service is presented in Table 4. Pages were logged over 27 nighttime call shifts totaling 270 resident work hours. Total census was 258 patients. The combined number of pages logged was 134. Independent review of the pages demonstrated that 73.1% were classified as “urgent” and 26.9% as “nonurgent.” Overall agreement between the 2 chief residents classifying the urgency of pages was found to be 93% and Cohen’s ␬ test also revealed significant agreement (␬ ⫽ 0.84). Most pages resulted in either a “new order” (47.8%) or “no action” (41.0%), while 11.2% of pages resulted in “patient assessment.” Of the pages resulting in “no action,” 65.5% were classified as “urgent,” and 34.5% as “nonurgent.” Of note, only 12.5% of pages from the post-NNAPPS group were classified as “nonurgent” compared with 34.9% of pages classified as “nonurgent” from the pre-NNAPPS group. Additionally, 8.2% of combined pages interrupted patient care, while 39.6% of pages interrupted resident sleep. The second NNAPPS study page origination data from the Transplant Surgery service is presented in Table 5. Most pages were initiated by nurses (79.9%), physicians (10.4%) and pharmacists (3.0%). The remainder of pages originated from laboratory technicians, patient calls, premedical students, respiratory therapy, transplant coordinators and unit clerks. The second NNAPPS study outcomes data from the Transplant Surgery service is presented in Table 6. The average number (mean ⫾ SD) of nonurgent pages per night shift received by the post-NNAPPS group was 0.46 ⫾ 0.52. Significant differences (p ⬍ 0.05) existed between the average number of nonurgent pages per night shift received by the post NNAPPS group compared with the pre-NNAPPS group. The average

SICU 2 (Mean ⴞ SD)

*Sum of percentages does not equal 100 secondary to rounding of digits.

0.67 ⫾ 0.89

Trauma Ward 2 (Mean ⴞ SD) EGS (Mean ⴞ SD)

105 (16.5) 1 (0.2) 1 (0.2) 1 (0.2) 1 (0.2) 1 (0.2) 406 (63.7) 11 (1.7) 94 (14.8) 3 (0.5) 2 (0.3) 2 (0.3) 9 (1.4)

CT (Mean ⴞ SD)

Code (trauma) Code (non-trauma) Emergency department consult Hospital bed control Hospital operator (patient call) Medical student Nurse Pharmacist Physician Premedical student Radiology technician Respiratory therapist Unknown

Burn (Mean ⴞ SD)

No. (%)*

General Surgery‡ (Mean ⴞ SD)

Who Paged

Trauma Ward 1 (Mean ⴞ SD)

TABLE 2. Page Origination of Initial NNAPPS Evaluation (All Services Combined)

Journal of Surgical Education • Volume 69/Number 6 • November/December 2012

TABLE 4. Characteristics of Transplant Ward/Transplant Surgery Service NNAPPS Evaluation Characteristic Night shifts Hours Patients Pages Urgent pages Nonurgent Pages Unable to determine urgency of pages Pages resulting in new order Pages resulting in patient assessment Pages resulting in no action Urgent Nonurgent Unable to determine urgency Unable to determine results of pages Patient care interrupted Sleep interrupted

Pre-NNAPPS No. (%)

Post-NNAPPS No. (%)

14 140 128 86 56 (65.1) 30 (34.9) 0 (0)

13 130 130 48 42 (87.5) 6 (12.5) 0 (0)

42 (48.8)

22 (45.8)

10 (11.6)

5 (10.4)

34 (39.5)

21 (43.8)

21 (61.8) 13 (38.2) 0 (0)

15 (71.4) 6 (28.6) 0 (0)

0 (0) 8 (9.3) 37 (43.0)

TABLE 6. Outcomes of Transplant Ward/Transplant Surgery Service NNAPPS Comparison

0 (0) 3 (6.3) 16 (33.3)

number of total pages per night shift received by the postNNAPPS group was 3.69 ⫾ 1.93 and the average number of total pages per patient per night shift received by the postNNAPPS group was 0.38 ⫾ 0.21. Significant differences (p ⬍ 0.05) existed between the average number of total pages and total pages per patient per night shift received by the postNNAPPS group compared with the pre-NNAPPS group. Of note, there were no identifiable patient morbidities or mortalities because of NNAPPS during the data collection period. In addition, since the end of our data collection period, NNAPPS has remained sustainable on both the General Surgery and Transplant wards. Secondary to hospital wide education of NNAPPS, additional systems involving both day and nighttime paging are being integrated onto various wards and units throughout the hospital.

TABLE 5. Page Origination of Transplant Ward/Transplant Surgery Service NNAPPS Evaluation Who Paged

No. (%)*

Microbiology lab Hospital operator (patient call) Nurse Pharmacist Physician Premedical student Respiratory therapist Transplant coordinator Unit clerk

2 (1.5) 2 (1.5) 107 (79.9) 4 (3.0) 14 (10.4) 2 (1.5) 1 (0.7) 1 (0.7) 1 (0.7)

*Sum of percentages does not equal 100 secondary to rounding of digits.

Outcome Nonurgent pages per night shift Total pages per night shift Total pages per patient per night shift

Pre NNAPPS (Mean ⴞ SD)

Post NNAPPS (Mean ⴞ SD)

p Value

2.14 ⫾ 1.96

0.46 ⫾ 0.52

0.02

6.14 ⫾ 2.48

3.69 ⫾ 1.93

0.02

0.68 ⫾ 0.28

0.38 ⫾ 0.21

0.007

DISCUSSION This is the first study to describe the design and implementation of a novel quality improvement nighttime paging system onto a ward providing care primarily for surgery patients at a teaching hospital. We successfully implemented NNAPPS onto a primary and demanding General Surgery ward as our results showed significant reductions in the number of nonurgent pages, total pages and pages per patient during a night shift with the service utilizing NNAPPS when compared with most services with patients cohorted on conventional wards or units without a specific paging system. These results were further validated with our second study design by comparing preNNAPPS and post-NNAPPS implementation data from our Transplant ward which again showed significant differences in the same outcomes as originally analyzed. Twenty-five percent of the pages received in our study were classified as “nonurgent” which is comparable to the number of nonurgent pages reported in previous studies (25%-29%).3,4 Surprisingly, 36% of the pages received in our study resulted in “no action,” which is higher than previously reported numbers of pages resulting in no action (18%-26%).4,5 An explanation for such a large number of pages resulting in “no action” may be secondary to the fact that even though certain pages result in no action by the resident, such pages may still be clinically indicated or in some cases mandated by nursing policy and protocol, such as the reporting of critical laboratory values or vital signs. Additionally, of the pages resulting in no action reported by both Trauma Ward teams, 24% of those were for trauma codes that the ward residents were not required to attend. The fact that roughly one quarter of the pages received by the Trauma Ward residents are for trauma codes that do not specifically concern them raises the question that potentially all paging systems hospital wide should be evaluated from a quality improvement standpoint. The paging patterns directed towards the Trauma Ward residents is one specific area where implementing quality improvement strategies, such as novel paging systems, may lead to significant reduc-

Journal of Surgical Education • Volume 69/Number 6 • November/December 2012

823

tions in the amount of unnecessary pages received by a specific service. The percentage of pages initiated by nursing staff in our study (67%) is similar to that of other studies (41%–50%) in that most pages received by residents were from nursing staff.3,4 These data are consistent with the notion that most patient care communication in a teaching hospital occurs between residents and nursing staff. Additionally, 81% of the total pages originated from either nurses or other physicians, which is consistent with the findings of Morton et al. who used a time-motion model to observe surgery resident activity during overnight call shifts at a University Medical Center and an urban affiliate hospital.9 While 36% of the pages resulted in no action taken by residents, 50% of those pages were classified as nonurgent. These types of pages are a source of significant disruption to patient care, rounding, resident education and sleep. The goal of such paging systems should not be to breed an environment of a reduced number of pages; rather the goal should be to improve the communication between patient care providers regarding patient care and to improve the quality of pages sent to residents, especially those working overnight call shifts. Data from the pre- and post-NNAPPS comparison groups on the Transplant ward demonstrated that 8% of pages interrupted patient care and 40% of pages interrupted resident sleep. These numbers differ from those reported by Harvey, where interns working on medical units in 2 teaching hospitals logged pages during 12-hour night shifts indicated that 19% of pages interrupted patient care and 25% of pages interrupted resident sleep.5 These discrepancies may be explained by the fact that these data come from residents in different specialties caring for a different cohort of patients (surgery residents caring for Transplant Surgery patients vs medicine interns caring for patients on medical units). Regarding amounts of interrupted sleep, a Transplant Surgery team’s schedule can be quite unpredictable, with volume and acuity of patients changing constantly, therefore anytime the chance for a strategic nap arises, the resident is likely to capitalize on this opportunity as rest comes at a premium on these types of services. The surgical residents on the Transplant ward were often interrupted from sleep likely because of logging pages during overnight call shifts as well as trying to utilize as much fatigue mitigation time as possible in between ward and operative duties throughout the night. Additional strategies to reduce the disruptions caused by frequent paging have been previously described by Blum, such as advertising the times of regularly scheduled rounds and conferences to encourage other patient care providers to delay paging during these times, having a designated physician to carry all pagers during these times so that only one physician is interrupted during rounding or conferences, organizing resident teams geographically so that residents may treat patients on only one ward and implementing a system in which the initiator of the page could include an indication of how immediately the page needed to be answered.3 824

Our institution has adopted many of these strategies to assist in reducing the number of interruptions caused by frequent paging, such as a surgery residency program director screening service pagers during educational conferences and attempting to organize resident teams geographically so patients may be cohorted on a single ward or unit. With regards to designating personnel to carry all service pagers during educational conferences, this is a useful strategy during daytime working hours when large scale educational conferences occur, however this strategy is unable to be implemented effectively at night when residents are working overnight call shifts. With regards to cohorting a service’s patient census on one ward or unit, it is not a foolproof method as ward or unit capacity is often exceeded in many tertiary care institutions, therefore the need arises to cohort patients on alternate wards or units. One additional strategy our institution is considering with regards to minimizing frequent paging disruptions is allowing nursing staff to send the specific reason for communication when sending an alphanumeric page to a resident. This strategy would allow the residents to judge the urgency of the pages themselves and ultimately reduce the number of pages requiring a return call during times of patient care, resident education and sleep. There are several limitations to our study. First, the logging of pages was self-reported by the surgical residents, which introduces several potential concerns, such as recall bias and recording of nonurgent or what may have been perceived as unnecessary pages more reliably than urgent or necessary pages. Alternative models of data collection may be employed, such as time-motion studies, in which extensively trained observers accompany and perform the data collection on the subjects being studied as described by both Lurie et al. and Morton et al.8,9 Time-motion studies are not without limitations, since they are expensive, resource intensive and introduce the possibility of a Hawthorne effect. Secondly, incomplete logging of pages by the residents may have affected the reliability of the data. We were unable to track whether or not paging logs were completed in their entirety. Full compliance and reliability of complete paging logs would better be obtained by the aforementioned timemotion data collection model. A potential solution to such a problem would be to ask nursing staff on each of the participating wards to log the pages they were sending out to the residents and then the concordance of data from the 2 groups could be assessed at a later date as previously described by Harvey.5 Even though a portion of our paging logs were subjective and situational, such as the classification of pages as urgent or nonurgent, we also included more objective measures to our paging logs, such as actions taken by residents because of the pages which is similar to the types of objective data collection described previously.5 Thirdly, without a comparison to pre-NNAPPS data, determining the impact of NNAPPS on the General Surgery ward would potentially be problematic, therefore we chose to compare paging pattern data from the General Surgery ward utilizing NNAPPS to paging pattern data from other wards caring for surgical patients without a specific paging system. Different surgical services may have a significantly different

Journal of Surgical Education • Volume 69/Number 6 • November/December 2012

cohort of patients at different acuity levels and different census numbers. We therefore chose an additional surgery ward (Transplant ward) to implement NNAPPS on in an effort to validate our initial pilot study findings by collecting and analyzing both pre-NNAPPS and post-NNAPPS data on the same ward caring for a similar patient cohort. With the same ward serving as its own control by gathering pre-NNAPPS data and comparing it to post-NNAPPS data, the impact of NNAPPS could be more appropriately determined. We believe there was great utility of the initial pilot study implementing NNAPPS on the General Surgery ward, however, in that it did show a significant reduction in the average number of nonurgent pages, total pages and total pages per patient during a night shift between the NNAPPS ward and most other wards or units studied without a specific paging system, albeit there was no comparison to pre-NNAPPS data on the General Surgery ward. Heavy resident workloads and sleep deprivation can amplify fatigue and stress levels with frequent paging, which adds to stress levels, inhibits time for meals, personal hygiene and rest.5 While it is unreasonable to expect hospital personnel to page residents with perfect accuracy, a system to maintain the quality and integrity of pages, such as NNAPPS, offers the opportunity to improve current paging practices.4

The authors have no conflicts of interest to report relative to the preparation or publication of this study.

CONCLUSIONS

5. Harvey R, Jarrett PG, Peltekian KM. Patterns of paging

NNAPPS resulted in a significant reduction of nonurgent pages, pages per night shift and pages per patient during a night shift when compared to services with patients cohorted on conventional wards or units without a specific paging system. A streamlined paging system leads to more efficient communication between patient care providers, decreased nonurgent nighttime pages to residents and improved overall patient care. Residents have more uninterrupted sleep and a reduction in fatigue. Most importantly, NNAPPS achieved all these goals while continuing to maintain high standards in patient care.

ACKNOWLEDGMENTS The authors thank the residents within the Department of Surgery who participated in this study and the nurses on the surgical wards and units for their contributions.

REFERENCES 1. Accreditation Council for Graduate Medical Education.

New-duty hours: ACGME approved standards. Chicago: Common Program Requirements: Accreditation Council for Graduate Medical Education. Effective; 2011. Available at: http://acgme.org/acWebsite/home/ Common_Program_Requirements_07012011.pdf. Accessed March 7, 2012. 2. Institute of Medicine. Resident duty hours: Enhancing sleep,

supervision, and safety; 2008. Available at: http://www. iom.edu/Reports/2008/Resident-Duty-Hours-EnhancingSleep-Supervision-and-Safety.aspx. Accessed March 7, 2012. 3. Blum NJ, Lieu TA. Interrupted care. The effects of paging

on pediatric resident activities. Am J Dis Child. 1992;146: 806-808. 4. Katz MH, Schroeder SA. The sounds of the hospital. Pag-

ing patterns in three teaching hospitals. N Engl J Med. 1988; 319:1585-1589. medical interns during night calls at two teaching hospitals. CMAJ. 1994;151:307-311. 6. Nguyen TC, Battat A, Longhurst C, et al. Alphanumeric

paging in an academic hospital setting. Am J Surg. 2006; 191:561-565. 7. Wong BM, Quan S, Shadowitz S, et al. Implementation

and evaluation of an alphanumeric paging system on a resident inpatient teaching service. J Hosp Med. 2009;4:E34E40. 8. Lurie N, Rank B, Parenti C, et al. How do house officers

spend their nights? A time study of internal medicine house staff on call. N Engl J Med. 1989;320:1673-1677. 9. Morton JM, Baker CC, Farrell TM, et al. What do surgery

residents do on their call nights? Am J Surg. 2004;188:225229.

Journal of Surgical Education • Volume 69/Number 6 • November/December 2012

825