- Email: [email protected]
Paging the Right Resident the First Time on General Internal Medicine: A Quality Improvement Project
The Joint Commission Journal on Quality and Patient Safety 2019; 45:711–716
Paging the Right Resident the First Time on General Internal Medicine: A Quality Improvement Project Joseph Carson, MSc; Stephanie Gottheil, MD, FRCPC; Bailey Dyck, MD, PhD, FRCPC; Tim Rice, RN
Background: Interprofessional hospital communication is vital for high-quality patient care. However, staff have reported that nursing pages are often sent to the wrong residents, leading to service delays, interruptions, and safety risks. The aim of this quality improvement project was to reduce day shift pages to general internal medicine (GIM) teams by 25% over 10 months by helping nurses page the right residents the first time. Methods: This study was conducted at a Canadian tertiary academic hospital and involved three GIM teams on seven inpatient wards. The Model for Improvement was used to explore root causes and redesign how nurses and switchboard operators contacted residents. Multiple change ideas were tested: posting daily resident assignments on digital monitors, redirecting switchboard pages to internal medicine residents, and forwarding pagers in learning sessions. The primary outcome was the average number of pages/team/week to GIM residents. Evaluation was conducted with statistical process control charts and qualitative feedback. Results: A total of 19,925 pages were reviewed from 226 resident shifts over 39 weeks. Average pages/team/week (Monday to Friday, 08:00 to 17:00) decreased by 38.3% (133 to 82) postimplementation. More nurses reported often or always knowing which residents were assigned to patients, increasing from 0% to 38.1%. Fewer residents reported often or always receiving pages about another resident’s patient, decreasing from 50.0% to 26.7%. Conclusion: Quality improvement methods were used to streamline the paging process on GIM wards, resulting in fewer pages and improved communication efficiency.
C
ommunication between interprofessional health care providers is vital to provide safe, effective, and timely care to hospital inpatients. Inefficient communication has been linked to delays in patient care, medical errors, and staff frustration.1,2 This issue can be magnified on academic general internal medicine (GIM) wards, where multidisciplinary students with various levels of experience complete short-term training rotations. Various modes of communication have been used to address this problem, including overhead paging, numeric and alphanumeric pagers, online messaging, and smartphones. Of these, personal pagers remain a key means of communication for the majority of GIM teams. However, paging is often inefficient and can lead to interprofessional frustration for recipients and senders.2 In a retrospective study of four GIM teams across two tertiary academic centers, 14% of pages were found to be sent to the wrong physician; of these pages, 36% were sent to an off-duty provider, 22% were sent during the evening, and 21% were sent during protected academic time.3 Even more concerning, 15% of pages sent to the wrong physician were emergent in nature, and 32% were urgent.3 Although several programs have piloted changes away from pager-based systems, the majority of hospital-based clinicians continue to use pagers for the majority of pa1553-7250/$-see front matter © 2019 The Joint Commission. Published by Elsevier Inc. All rights reserved. https://doi.org/10.1016/j.jcjq.2019.08.001
tient care communication.4,5 One study attempted to improve communication by introducing a team pager held by the senior resident, who would receive all incoming pages.6 Although this resulted in a reduction of pages sent to the wrong physician from 14% preimplementation to 3% postimplementation, the assigned residents experienced a dramatic increase in pages, from 55 to 83 calls per week. Furthermore, this system does not improve efficiency because it requires the senior resident to then contact other team members to address the issue.6 To date, no dedicated interventions to optimize the ability to contact the correct physician for an assigned patient have been reported. At one academic medical center, a nurse-physician task force implemented a clarification process by posting information sheets on each inpatient floor that indicated the correct hospitalist to contact for that floor.7 Although this intervention did not address a diverse GIM team with multiple physician and trainee members, the number of misdirected pages halved after the intervention and nurses reported an appreciation for the posted information. This highlights the importance of interventions to help nurses and other allied health professionals contact the right physician the first time. To determine local problem severity, we surveyed GIM residents and nurses about their experiences with interprofessional communication. Many staff reported a quality gap in the current paging system. Nurses did not know which residents were assigned patients on a daily basis, and this
712
Joseph Carson, MSc, et al.
confusion contributed to safety risks, workflow interruptions, and discharge delays. Although we could not feasibly measure unsuccessful attempts by nurses to page the right resident, we hypothesized that a streamlined communication process would lead to fewer overall pages. We aimed to reduce day shift pages to GIM residents by 25% in 10 months, by May 31, 2018. METHODS
In this quality improvement project, we analyzed the baseline quality gap and tested solutions. The current state was explored by mapping the pager communication process, surveying staff, and analyzing call logs. Change ideas were developed by reviewing literature, surveying staff, and identifying positive deviance at an affiliated hospital. We used a driver diagram and priority matrix to determine which ideas to implement. Changes were tested with incremental PlanDo-Study-Act (PDSA) cycles, based on the Model for Improvement.8 Our project team included a quality improvement consultant, an internal medicine fellow, an internal medicine chief resident, a clinical nurse, and a senior hospital administrator. The project involved multidisciplinary staff from inpatient medicine services, telecommunications, decision support, and the resident education program. Local Context
Three GIM physician teams at one tertiary academic center were included. Each team consisted of one attending physician, one or two senior residents, three or four junior residents, and several medical students. Junior residents were either from the internal medicine program or non–
Paging the Right Resident the First Time
internal medicine residents from other programs, such as family medicine, emergency medicine, or surgery. Residents rotated on a monthly basis, but the number of residents per team was relatively consistent. Each GIM team was assigned a roster of patients admitted across several floors of the hospital. Every morning, GIM teams would meet and assign the care of each patient to a junior resident or a medical student. Patient assignments were done verbally during team meetings and would occasionally be written down on printed patient lists by team members. Patient assignments were not recorded in the electronic health record and were not shared outside of physician teams. Some GIM team members would be absent from the morning assignment if they were away, post-call, or on vacation. Each team member was expected to carry and attend to a personal pager during daytime hours (Monday to Friday, 08:00 to 17:00). Residents were expected to disable or forward their pagers when they were absent for any reason. In the current state, nurses used multiple methods to identify and page residents (Figure 1). Nurses did not carry pagers themselves or have any handheld communication devices, but they would typically use desk phones located in the nursing station to page residents when needed. Switchboard operators selected residents based on the alphabetized order of their training programs (for example, emergency medicine, then family medicine, then internal medicine), or they would select at random. In a survey of 26 nurses, 0% (0/26) reported often or always knowing which residents were assigned to their patients, 11.5% (3/26) often or always believed the process to page residents was effi-
Figure 1: The current state map was created with nurses and residents to identify communication pathways. SMR, senior medical resident.
Volume 45, No. 10, October 2019
cient, and 11.5% (3/26) reported often or always paging the right resident on the first attempt. In a survey of 24 residents, 50.0% (12/24) reported often or always being paged about another resident’s patient, and 29.2% (7/24) were often or always paged on evenings and weekends when off duty (Table 1). Qualitative feedback was also solicited from nurses and residents to explore their experiences with interprofessional communication. Both nurses and residents commented on inefficiencies in paging, interruptions in patient care, and concerns about patient safety. One nurse commented: “It’s extremely difficult to get a hold of the correct resident normally, but during an emergency it is worse because we can’t afford to be away from the patient’s bedside. The paging system feels like it is all trial and error.” Another nurse made the following suggestion: “I would like the medical teams to list their patient assignments daily so there isn’t so much guess work.” This feedback was used to help in the design of quality improvement interventions. Interventions
Change Idea 1: Displaying Physician Assignments. We discovered that an affiliate hospital posted pager numbers on large digital monitors (showboards) located in each ward’s nursing station. Showboards listed patient names, assigned nurses, and discharge planning tasks using Cerner Capacity Management software (Cerner Corporation, North Kansas City, Missouri). We used PDSA cycles at our hospital to confirm that showboard data could be entered centrally and remotely, and pager numbers would appear in the correct ward locations. Initially, a patient care facilitator (PCF) recorded daily assignments at the physician team meeting and entered the information onto the digital showboard, mimicking the affiliate hospital. However, this process did not align with our scheduling and rounding practices. In the next PDSA cycles, senior residents were asked to hand a list of physician assignments to any ward clerk for data entry onto the digital showboard. Residents did not have access to the software that was used for digital showboard data entry and so could not enter assignments themselves. Senior residents were reminded by PCFs at the beginning of new resident rotations,
713
and they often assigned this task to a junior resident or medical student. Even though resident assignments began to appear regularly on showboards, many nurses continued to ask the switchboard to contact physician teams. We learned that some nurses were not aware of the new process, while others found inaccurate pager numbers. To raise awareness, clinical educators conducted brief educational sessions with nurses, signs were posted, and ward clerks reminded nurses to look at the showboards. Nurses suggested stapling a cover page over resident lists, which were revised to indicate that senior residents should be paged only for urgent issues. We also ensured that ward clerks deleted numbers at the end of each day shift to prevent old information from appearing the next morning. Change Idea 2: Redirecting Switchboard Calls. In the new paging process, the list of daily resident assignments was usually posted by 11:00 after team rounds. However, some nurses reported not knowing who to page beforehand and called the switchboard. We interviewed switchboard operators and learned that they preferentially selected non– internal medicine residents, based on the alphabetical order of their home training program. Listings were revised to prioritize internal medicine residents, who generally had more expertise to handle questions about any GIM patient. Change Idea 3: Forwarding Pagers During Academic Half Days. Residents attended a mandatory threehour teaching session every week (academic half day). During this time, nurses reported difficulty in contacting residents, and pager logs confirmed a high number of calls. Residents reported a lack of clarity from attending physicians about whether they were expected to respond to calls during academic half days. We worked with the Department of Medicine to implement a call forwarding policy, encouraging residents to forward their pagers to another physician. We also promoted the new policy with brief education sessions and awareness e-mails from the chief resident. Measures and Data Collection
Three outcome measures were developed to assess the success of our project: (1) the average number of day shift
714
Joseph Carson, MSc, et al.
pages/team/week from GIM wards and the switchboard to residents; (2) the percentage of nurses who knew which resident was assigned to their patients; and (3) the percentage of residents who were paged about another resident’s patient. The analysis was limited to day shift pages, as there was only one resident covering the team after 17:00, and so the most responsible resident was easily identified. Pager log data were collected by the hospital telecommunications department. GIM nurses completed anonymous online surveys at baseline and after showboard implementation. Residents completed anonymous online surveys at baseline and monthly at the end of their GIM rotations. Survey results were analyzed using Pearson’s chi-square test with SPSS 25 (IBM Corp., Armonk, New York). Average pager call frequency was analyzed continuously over time using QI Macros 2018 (KnowWare International, Inc., Denver) statistical process control (SPC) charts. Process measures were used to determine if change ideas were being implemented as expected. These included (1) the number of days/week that resident assignments were on the showboards; (2) the percentage of pages directed to non–internal medicine residents; (3) the percentage of pages from the switchboard to residents; and (4) the percentage of residents who forwarded their pagers during academic half days. Data were obtained from telecom logs and manual audits and analyzed using SPSS Pearson’s chi-square test and QI Macros SPC charts. Our balance measure was the percentage of residents who reported being paged while off duty, obtained in our monthly surveys. We hypothesized that nurses might use expired daily assignment numbers to contact residents if the process was not implemented correctly.
Paging the Right Resident the First Time
Ethical Considerations
This quality improvement project was exempt from Research Ethics Board review, in accordance with the Government of Canada Tri-Council Policy Statement on Research Ethics.9 RESULTS
We reviewed 19,925 pager call logs from July 2017 to June 2018 (39 weeks). Pages were included from 226 continuous resident day shifts (Monday to Friday, 08:00 to 17:00, excluding holidays)—161 shifts (71.2%) from internal medicine residents and 65 shifts (28.8%) from non– internal medicine residents. Call logs from medical students and attending physicians were excluded, as we believed that residents received most pages from GIM nurses. The primary outcome measure, average pages/team/ week, decreased by 38.3% (from 133 to 82) and demonstrated special cause variation after implementation (Figure 2). The percentage of nurses who reported knowing often or always which resident was assigned to their patients increased from 0% to 38.1%, while the percentage of residents reportedly paged often or always about another resident’s patient decreased from 50.0% to 26.7% (Table 1). Process measures showed that staff updated digital showboards more than 80% of the time (high fidelity). As well, the percentage of non–internal medicine residents receiving 25 or more pages per month pages per month decreased significantly from 82.0 to 20.0 (p < 0.0001). The percentage of pages from the switchboard decreased from 36.9 to 13.6, demonstrating special cause variation (Figure 3 ). However, even after the call forwarding policy was implemented, no
Figure 2: Statistical process control (SPC) I-chart showing the average number of pages from wards/switchboard to team residents. UCL, upper control limit; CL, center line (recalculated when changes corresponded with special-cause variation); LCL, lower control limit; PCF,patient care facilitator.
Volume 45, No. 10, October 2019
715
Figure 3: Statistical process control (SPC) P-chart showing the percentage of pages from the switchboard to team residents. UCL, upper control limit; CL, center line (recalculated when changes corresponded with special-cause variation); LCL, lower control limit.
residents forwarded their pagers during academic half days. The balance measure showed a decrease in residents reporting off-duty pages from 29.2% to 6.7% (Table 1).
DISCUSSION
Interprofessional communication on GIM wards lacked efficiency. Because nurses were uncertain whom to contact, we posted daily patient assignments to direct nurses to the most appropriate resident. We also asked switchboard operators to prioritize internal medicine residents, who were more likely to have the required knowledge to answer clinical questions. We believed this combination of changes and new nursing habits led to a sustained decrease in pages and an increase in staff-reported communication efficiency. We used multiple PDSA cycles to establish the process for posting daily assignments. Staff adherence was initially inconsistent, particularly at the beginning of new resident rotations, requiring daily follow-up and motivation. Pager volumes actually increased during the initial implementation period, which likely reflected the difficulty in creating new habits for both residents and nurses. Measuring daily adherence was therefore helpful to identify and address performance gaps in a timely manner. Seeking feedback from residents, nurses, and ward clerks was also important to shape implementation strategies. For example, nurses told us that, out of habit, they would still page randomly from a resident list even when daily assignments were posted. They suggested taping a cover page over the resident list to nudge them toward the digital showboard. We also heard that daily postings were often delayed until late morning
due to physician rounding or ward clerks not seeing the list delivered to an unattended ward desk. These challenges led us to revisit current-state mapping with switchboard staff to direct morning pages appropriately. Although we implemented a new pager forwarding policy for academic half days, we believe it was not followed due to lack of encouragement from attending physicians and resident concern that pager forwarding may be perceived poorly by attending physicians. Limitations
There were several limitations to this study. Although the number of pages was measured, the urgency of those pages and any adverse events occurring from incorrect paging were not assessed. We were also unable to determine how often pages were directed to the correct resident. As well, the accuracy of self-reported communication inefficiency by nurses and residents could not be confirmed. More frequent staff surveys and more survey responses may have improved the perception of paging practices and decreased recall bias. It is also possible that other factors contributed to the reduction in pager volumes, such as seasonal variation in patient and staff census and residents rotating each month.
CONCLUSION
This quality improvement project aimed to reduce unnecessary pages by helping nurses page the right resident the first time. The root cause analysis found that daily resident assignments were not shared with nonphysician team members. non–internal medicine residents received more
716
Joseph Carson, MSc, et al.
pages than internal medicine residents. and residents were paged during protected teaching sessions. A number of changes were implemented: posting daily patient assignments on showboards, redirecting switchboard pages to internal medicine residents, and creating a call forwarding policy for academic half day sessions. Results demonstrated a sustained, significant decrease in total pages and an improvement in communication efficiency on GIM. Based on the success of this project, similar changes were subsequently implemented at an affiliated hospital. Acknowledgments. The authors wish to acknowledge contributions from the following staff: Donna Fan and Ashley Hogan for current state mapping; Chris Dotzert and Robin Francis for data analysis; Kim McIntyre, Marnie Bensette, Lara McKinley, and Marika Wilton for switchboard changes; Jamie Gregor for physician leadership; and Victoria Hospital medicine residents, nurses, communication clerks, and patient care facilitators for improving our pager process. Conflicts of Interest. All authors report no conflicts of interest.
Joseph Carson, MSc, is Quality Improvement Consultant, Department of Medicine, Schulich School of Medicine & Dentistry, Western University, London, Ontario, Canada. Stephanie Gottheil, MD, FRCPC, is Resident Physician, Department of Medicine, Schulich School of Medicine & Dentistry, Western University, and London Health Sciences Centre (LHSC). Bailey Dyck, MD, PhD, FRCPC, is Resident Physician, Department of Medicine, Schulich School of Medicine & Dentistry, Western University, and LHSC. Tim Rice, RN, is Director, Medicine and Family Medicine, LHSC. Please address correspondence to Joseph Carson, [email protected].
REFERENCES 1. Kummerow Broman K, et al. Characterizing the clamor: an in-depth analysis of inpatient paging communication. Acad Med. 2016;91:1015–1021.
Paging the Right Resident the First Time 2. Wu RC, et al. The intended and unintended consequences of communication systems on general internal medicine inpatient care delivery: a prospective observational case study of five teaching hospitals. J Am Med Inform Assoc. 2013;20:766–777. 3. Wong BM, et al. Frequency and clinical importance of pages sent to the wrong physician. Arch Intern Med. 2009 Jun 8;169:1072–1073. 4. O’Leary KJ, et al. Hospital-based clinicians’ use of technology for patient care-related communication: a national survey. J Hosp Med. 2017;12:530–535. 5. Quan SD, et al. It’s not about pager replacement: an in-depth look at the interprofessional nature of communication in healthcare. J Hosp Med. 2013;8:137–143. 6. Wong BM, et al. Getting the message: a quality improvement initiative to reduce pages sent to the wrong physician. BMJ Qual Saf. 2012;21:855–862. 7. Rosenthal L. Enhancing communication between night shift RNs and hospitalists: an opportunity for performance improvement. J Nurs Adm. 2013;43:59–61. 8. Langley GL, et al. The Improvement Guide: A Practical Approach to Enhancing Organizational Performance. 2nd ed. San Francisco: Jossey-Bass, 2009. 9. Government of Canada, Interagency Advisory Panel on Research Ethics. Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans. Activities Not Requiring REB Review: Article 2.5. (Updated: Feb 15, 2018.) Accessed Aug 12, 2019. http://www.pre.ethics.gc.ca/eng/policy-politique/ initiatives/tcps2- eptc2/chapter2- chapitre2/#ch2_en_a2.5.
Paging the Right Resident the First Time on General Internal Medicine: A Quality Improvement Project Joseph Carson, MSc; Stephanie Gottheil, MD, FRCPC; Bailey Dyck, MD, PhD, FRCPC; Tim Rice, RN
Background: Interprofessional hospital communication is vital for high-quality patient care. However, staff have reported that nursing pages are often sent to the wrong residents, leading to service delays, interruptions, and safety risks. The aim of this quality improvement project was to reduce day shift pages to general internal medicine (GIM) teams by 25% over 10 months by helping nurses page the right residents the first time. Methods: This study was conducted at a Canadian tertiary academic hospital and involved three GIM teams on seven inpatient wards. The Model for Improvement was used to explore root causes and redesign how nurses and switchboard operators contacted residents. Multiple change ideas were tested: posting daily resident assignments on digital monitors, redirecting switchboard pages to internal medicine residents, and forwarding pagers in learning sessions. The primary outcome was the average number of pages/team/week to GIM residents. Evaluation was conducted with statistical process control charts and qualitative feedback. Results: A total of 19,925 pages were reviewed from 226 resident shifts over 39 weeks. Average pages/team/week (Monday to Friday, 08:00 to 17:00) decreased by 38.3% (133 to 82) postimplementation. More nurses reported often or always knowing which residents were assigned to patients, increasing from 0% to 38.1%. Fewer residents reported often or always receiving pages about another resident’s patient, decreasing from 50.0% to 26.7%. Conclusion: Quality improvement methods were used to streamline the paging process on GIM wards, resulting in fewer pages and improved communication efficiency.
C
ommunication between interprofessional health care providers is vital to provide safe, effective, and timely care to hospital inpatients. Inefficient communication has been linked to delays in patient care, medical errors, and staff frustration.1,2 This issue can be magnified on academic general internal medicine (GIM) wards, where multidisciplinary students with various levels of experience complete short-term training rotations. Various modes of communication have been used to address this problem, including overhead paging, numeric and alphanumeric pagers, online messaging, and smartphones. Of these, personal pagers remain a key means of communication for the majority of GIM teams. However, paging is often inefficient and can lead to interprofessional frustration for recipients and senders.2 In a retrospective study of four GIM teams across two tertiary academic centers, 14% of pages were found to be sent to the wrong physician; of these pages, 36% were sent to an off-duty provider, 22% were sent during the evening, and 21% were sent during protected academic time.3 Even more concerning, 15% of pages sent to the wrong physician were emergent in nature, and 32% were urgent.3 Although several programs have piloted changes away from pager-based systems, the majority of hospital-based clinicians continue to use pagers for the majority of pa1553-7250/$-see front matter © 2019 The Joint Commission. Published by Elsevier Inc. All rights reserved. https://doi.org/10.1016/j.jcjq.2019.08.001
tient care communication.4,5 One study attempted to improve communication by introducing a team pager held by the senior resident, who would receive all incoming pages.6 Although this resulted in a reduction of pages sent to the wrong physician from 14% preimplementation to 3% postimplementation, the assigned residents experienced a dramatic increase in pages, from 55 to 83 calls per week. Furthermore, this system does not improve efficiency because it requires the senior resident to then contact other team members to address the issue.6 To date, no dedicated interventions to optimize the ability to contact the correct physician for an assigned patient have been reported. At one academic medical center, a nurse-physician task force implemented a clarification process by posting information sheets on each inpatient floor that indicated the correct hospitalist to contact for that floor.7 Although this intervention did not address a diverse GIM team with multiple physician and trainee members, the number of misdirected pages halved after the intervention and nurses reported an appreciation for the posted information. This highlights the importance of interventions to help nurses and other allied health professionals contact the right physician the first time. To determine local problem severity, we surveyed GIM residents and nurses about their experiences with interprofessional communication. Many staff reported a quality gap in the current paging system. Nurses did not know which residents were assigned patients on a daily basis, and this
712
Joseph Carson, MSc, et al.
confusion contributed to safety risks, workflow interruptions, and discharge delays. Although we could not feasibly measure unsuccessful attempts by nurses to page the right resident, we hypothesized that a streamlined communication process would lead to fewer overall pages. We aimed to reduce day shift pages to GIM residents by 25% in 10 months, by May 31, 2018. METHODS
In this quality improvement project, we analyzed the baseline quality gap and tested solutions. The current state was explored by mapping the pager communication process, surveying staff, and analyzing call logs. Change ideas were developed by reviewing literature, surveying staff, and identifying positive deviance at an affiliated hospital. We used a driver diagram and priority matrix to determine which ideas to implement. Changes were tested with incremental PlanDo-Study-Act (PDSA) cycles, based on the Model for Improvement.8 Our project team included a quality improvement consultant, an internal medicine fellow, an internal medicine chief resident, a clinical nurse, and a senior hospital administrator. The project involved multidisciplinary staff from inpatient medicine services, telecommunications, decision support, and the resident education program. Local Context
Three GIM physician teams at one tertiary academic center were included. Each team consisted of one attending physician, one or two senior residents, three or four junior residents, and several medical students. Junior residents were either from the internal medicine program or non–
Paging the Right Resident the First Time
internal medicine residents from other programs, such as family medicine, emergency medicine, or surgery. Residents rotated on a monthly basis, but the number of residents per team was relatively consistent. Each GIM team was assigned a roster of patients admitted across several floors of the hospital. Every morning, GIM teams would meet and assign the care of each patient to a junior resident or a medical student. Patient assignments were done verbally during team meetings and would occasionally be written down on printed patient lists by team members. Patient assignments were not recorded in the electronic health record and were not shared outside of physician teams. Some GIM team members would be absent from the morning assignment if they were away, post-call, or on vacation. Each team member was expected to carry and attend to a personal pager during daytime hours (Monday to Friday, 08:00 to 17:00). Residents were expected to disable or forward their pagers when they were absent for any reason. In the current state, nurses used multiple methods to identify and page residents (Figure 1). Nurses did not carry pagers themselves or have any handheld communication devices, but they would typically use desk phones located in the nursing station to page residents when needed. Switchboard operators selected residents based on the alphabetized order of their training programs (for example, emergency medicine, then family medicine, then internal medicine), or they would select at random. In a survey of 26 nurses, 0% (0/26) reported often or always knowing which residents were assigned to their patients, 11.5% (3/26) often or always believed the process to page residents was effi-
Figure 1: The current state map was created with nurses and residents to identify communication pathways. SMR, senior medical resident.
Volume 45, No. 10, October 2019
cient, and 11.5% (3/26) reported often or always paging the right resident on the first attempt. In a survey of 24 residents, 50.0% (12/24) reported often or always being paged about another resident’s patient, and 29.2% (7/24) were often or always paged on evenings and weekends when off duty (Table 1). Qualitative feedback was also solicited from nurses and residents to explore their experiences with interprofessional communication. Both nurses and residents commented on inefficiencies in paging, interruptions in patient care, and concerns about patient safety. One nurse commented: “It’s extremely difficult to get a hold of the correct resident normally, but during an emergency it is worse because we can’t afford to be away from the patient’s bedside. The paging system feels like it is all trial and error.” Another nurse made the following suggestion: “I would like the medical teams to list their patient assignments daily so there isn’t so much guess work.” This feedback was used to help in the design of quality improvement interventions. Interventions
Change Idea 1: Displaying Physician Assignments. We discovered that an affiliate hospital posted pager numbers on large digital monitors (showboards) located in each ward’s nursing station. Showboards listed patient names, assigned nurses, and discharge planning tasks using Cerner Capacity Management software (Cerner Corporation, North Kansas City, Missouri). We used PDSA cycles at our hospital to confirm that showboard data could be entered centrally and remotely, and pager numbers would appear in the correct ward locations. Initially, a patient care facilitator (PCF) recorded daily assignments at the physician team meeting and entered the information onto the digital showboard, mimicking the affiliate hospital. However, this process did not align with our scheduling and rounding practices. In the next PDSA cycles, senior residents were asked to hand a list of physician assignments to any ward clerk for data entry onto the digital showboard. Residents did not have access to the software that was used for digital showboard data entry and so could not enter assignments themselves. Senior residents were reminded by PCFs at the beginning of new resident rotations,
713
and they often assigned this task to a junior resident or medical student. Even though resident assignments began to appear regularly on showboards, many nurses continued to ask the switchboard to contact physician teams. We learned that some nurses were not aware of the new process, while others found inaccurate pager numbers. To raise awareness, clinical educators conducted brief educational sessions with nurses, signs were posted, and ward clerks reminded nurses to look at the showboards. Nurses suggested stapling a cover page over resident lists, which were revised to indicate that senior residents should be paged only for urgent issues. We also ensured that ward clerks deleted numbers at the end of each day shift to prevent old information from appearing the next morning. Change Idea 2: Redirecting Switchboard Calls. In the new paging process, the list of daily resident assignments was usually posted by 11:00 after team rounds. However, some nurses reported not knowing who to page beforehand and called the switchboard. We interviewed switchboard operators and learned that they preferentially selected non– internal medicine residents, based on the alphabetical order of their home training program. Listings were revised to prioritize internal medicine residents, who generally had more expertise to handle questions about any GIM patient. Change Idea 3: Forwarding Pagers During Academic Half Days. Residents attended a mandatory threehour teaching session every week (academic half day). During this time, nurses reported difficulty in contacting residents, and pager logs confirmed a high number of calls. Residents reported a lack of clarity from attending physicians about whether they were expected to respond to calls during academic half days. We worked with the Department of Medicine to implement a call forwarding policy, encouraging residents to forward their pagers to another physician. We also promoted the new policy with brief education sessions and awareness e-mails from the chief resident. Measures and Data Collection
Three outcome measures were developed to assess the success of our project: (1) the average number of day shift
714
Joseph Carson, MSc, et al.
pages/team/week from GIM wards and the switchboard to residents; (2) the percentage of nurses who knew which resident was assigned to their patients; and (3) the percentage of residents who were paged about another resident’s patient. The analysis was limited to day shift pages, as there was only one resident covering the team after 17:00, and so the most responsible resident was easily identified. Pager log data were collected by the hospital telecommunications department. GIM nurses completed anonymous online surveys at baseline and after showboard implementation. Residents completed anonymous online surveys at baseline and monthly at the end of their GIM rotations. Survey results were analyzed using Pearson’s chi-square test with SPSS 25 (IBM Corp., Armonk, New York). Average pager call frequency was analyzed continuously over time using QI Macros 2018 (KnowWare International, Inc., Denver) statistical process control (SPC) charts. Process measures were used to determine if change ideas were being implemented as expected. These included (1) the number of days/week that resident assignments were on the showboards; (2) the percentage of pages directed to non–internal medicine residents; (3) the percentage of pages from the switchboard to residents; and (4) the percentage of residents who forwarded their pagers during academic half days. Data were obtained from telecom logs and manual audits and analyzed using SPSS Pearson’s chi-square test and QI Macros SPC charts. Our balance measure was the percentage of residents who reported being paged while off duty, obtained in our monthly surveys. We hypothesized that nurses might use expired daily assignment numbers to contact residents if the process was not implemented correctly.
Paging the Right Resident the First Time
Ethical Considerations
This quality improvement project was exempt from Research Ethics Board review, in accordance with the Government of Canada Tri-Council Policy Statement on Research Ethics.9 RESULTS
We reviewed 19,925 pager call logs from July 2017 to June 2018 (39 weeks). Pages were included from 226 continuous resident day shifts (Monday to Friday, 08:00 to 17:00, excluding holidays)—161 shifts (71.2%) from internal medicine residents and 65 shifts (28.8%) from non– internal medicine residents. Call logs from medical students and attending physicians were excluded, as we believed that residents received most pages from GIM nurses. The primary outcome measure, average pages/team/ week, decreased by 38.3% (from 133 to 82) and demonstrated special cause variation after implementation (Figure 2). The percentage of nurses who reported knowing often or always which resident was assigned to their patients increased from 0% to 38.1%, while the percentage of residents reportedly paged often or always about another resident’s patient decreased from 50.0% to 26.7% (Table 1). Process measures showed that staff updated digital showboards more than 80% of the time (high fidelity). As well, the percentage of non–internal medicine residents receiving 25 or more pages per month pages per month decreased significantly from 82.0 to 20.0 (p < 0.0001). The percentage of pages from the switchboard decreased from 36.9 to 13.6, demonstrating special cause variation (Figure 3 ). However, even after the call forwarding policy was implemented, no
Figure 2: Statistical process control (SPC) I-chart showing the average number of pages from wards/switchboard to team residents. UCL, upper control limit; CL, center line (recalculated when changes corresponded with special-cause variation); LCL, lower control limit; PCF,patient care facilitator.
Volume 45, No. 10, October 2019
715
Figure 3: Statistical process control (SPC) P-chart showing the percentage of pages from the switchboard to team residents. UCL, upper control limit; CL, center line (recalculated when changes corresponded with special-cause variation); LCL, lower control limit.
residents forwarded their pagers during academic half days. The balance measure showed a decrease in residents reporting off-duty pages from 29.2% to 6.7% (Table 1).
DISCUSSION
Interprofessional communication on GIM wards lacked efficiency. Because nurses were uncertain whom to contact, we posted daily patient assignments to direct nurses to the most appropriate resident. We also asked switchboard operators to prioritize internal medicine residents, who were more likely to have the required knowledge to answer clinical questions. We believed this combination of changes and new nursing habits led to a sustained decrease in pages and an increase in staff-reported communication efficiency. We used multiple PDSA cycles to establish the process for posting daily assignments. Staff adherence was initially inconsistent, particularly at the beginning of new resident rotations, requiring daily follow-up and motivation. Pager volumes actually increased during the initial implementation period, which likely reflected the difficulty in creating new habits for both residents and nurses. Measuring daily adherence was therefore helpful to identify and address performance gaps in a timely manner. Seeking feedback from residents, nurses, and ward clerks was also important to shape implementation strategies. For example, nurses told us that, out of habit, they would still page randomly from a resident list even when daily assignments were posted. They suggested taping a cover page over the resident list to nudge them toward the digital showboard. We also heard that daily postings were often delayed until late morning
due to physician rounding or ward clerks not seeing the list delivered to an unattended ward desk. These challenges led us to revisit current-state mapping with switchboard staff to direct morning pages appropriately. Although we implemented a new pager forwarding policy for academic half days, we believe it was not followed due to lack of encouragement from attending physicians and resident concern that pager forwarding may be perceived poorly by attending physicians. Limitations
There were several limitations to this study. Although the number of pages was measured, the urgency of those pages and any adverse events occurring from incorrect paging were not assessed. We were also unable to determine how often pages were directed to the correct resident. As well, the accuracy of self-reported communication inefficiency by nurses and residents could not be confirmed. More frequent staff surveys and more survey responses may have improved the perception of paging practices and decreased recall bias. It is also possible that other factors contributed to the reduction in pager volumes, such as seasonal variation in patient and staff census and residents rotating each month.
CONCLUSION
This quality improvement project aimed to reduce unnecessary pages by helping nurses page the right resident the first time. The root cause analysis found that daily resident assignments were not shared with nonphysician team members. non–internal medicine residents received more
716
Joseph Carson, MSc, et al.
pages than internal medicine residents. and residents were paged during protected teaching sessions. A number of changes were implemented: posting daily patient assignments on showboards, redirecting switchboard pages to internal medicine residents, and creating a call forwarding policy for academic half day sessions. Results demonstrated a sustained, significant decrease in total pages and an improvement in communication efficiency on GIM. Based on the success of this project, similar changes were subsequently implemented at an affiliated hospital. Acknowledgments. The authors wish to acknowledge contributions from the following staff: Donna Fan and Ashley Hogan for current state mapping; Chris Dotzert and Robin Francis for data analysis; Kim McIntyre, Marnie Bensette, Lara McKinley, and Marika Wilton for switchboard changes; Jamie Gregor for physician leadership; and Victoria Hospital medicine residents, nurses, communication clerks, and patient care facilitators for improving our pager process. Conflicts of Interest. All authors report no conflicts of interest.
Joseph Carson, MSc, is Quality Improvement Consultant, Department of Medicine, Schulich School of Medicine & Dentistry, Western University, London, Ontario, Canada. Stephanie Gottheil, MD, FRCPC, is Resident Physician, Department of Medicine, Schulich School of Medicine & Dentistry, Western University, and London Health Sciences Centre (LHSC). Bailey Dyck, MD, PhD, FRCPC, is Resident Physician, Department of Medicine, Schulich School of Medicine & Dentistry, Western University, and LHSC. Tim Rice, RN, is Director, Medicine and Family Medicine, LHSC. Please address correspondence to Joseph Carson, [email protected].
REFERENCES 1. Kummerow Broman K, et al. Characterizing the clamor: an in-depth analysis of inpatient paging communication. Acad Med. 2016;91:1015–1021.
Paging the Right Resident the First Time 2. Wu RC, et al. The intended and unintended consequences of communication systems on general internal medicine inpatient care delivery: a prospective observational case study of five teaching hospitals. J Am Med Inform Assoc. 2013;20:766–777. 3. Wong BM, et al. Frequency and clinical importance of pages sent to the wrong physician. Arch Intern Med. 2009 Jun 8;169:1072–1073. 4. O’Leary KJ, et al. Hospital-based clinicians’ use of technology for patient care-related communication: a national survey. J Hosp Med. 2017;12:530–535. 5. Quan SD, et al. It’s not about pager replacement: an in-depth look at the interprofessional nature of communication in healthcare. J Hosp Med. 2013;8:137–143. 6. Wong BM, et al. Getting the message: a quality improvement initiative to reduce pages sent to the wrong physician. BMJ Qual Saf. 2012;21:855–862. 7. Rosenthal L. Enhancing communication between night shift RNs and hospitalists: an opportunity for performance improvement. J Nurs Adm. 2013;43:59–61. 8. Langley GL, et al. The Improvement Guide: A Practical Approach to Enhancing Organizational Performance. 2nd ed. San Francisco: Jossey-Bass, 2009. 9. Government of Canada, Interagency Advisory Panel on Research Ethics. Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans. Activities Not Requiring REB Review: Article 2.5. (Updated: Feb 15, 2018.) Accessed Aug 12, 2019. http://www.pre.ethics.gc.ca/eng/policy-politique/ initiatives/tcps2- eptc2/chapter2- chapitre2/#ch2_en_a2.5.