ORIGINAL REPORTS
Orthopedic Resident Work-Shift Analysis: Are We Making the Best Use of Resident Work Hours? Kamran S. Hamid, MD,* Benedict U. Nwachukwu, MD,† Eugene Hsu, MD,‡ Colston A. Edgerton, BS,* David R. Hobson, MS,* and Jason E. Lang, MD* *
Department of Orthopaedic Surgery, Wake Forest Baptist Medical Center, Winston-Salem, North Carolina; Hospital for Special Surgery, New York, New York; and ‡Department of Anesthesiology & Critical Care Medicine, The Johns Hopkins Hospital, Baltimore, Maryland †
BACKGROUND: Surgery programs have been tasked to
meet rising demands in patient surgical care while simultaneously providing adequate resident training in the midst of increasing resident work-hour restrictions. The purpose of this study was to quantify orthopedic surgery resident workflow and identify areas needing improved resident efficiency. We hypothesize that residents spend a disproportionate amount of time involved in activities that do not relate directly to patient care or maximize resident education. METHODS: We observed 4 orthopedic surgery residents on
the orthopedic consult service at a major tertiary care center for 72 consecutive hours (6 consecutive shifts). We collected minute-by-minute data using predefined work-task criteria: direct new patient contact, direct existing patient contact, communications with other providers, documentation/ administrative time, transit time, and basic human needs. A seventh category comprised remaining less-productive work was termed as standby.
directly interact with existing patients (p ¼ 0.006) or attend to basic human needs (p ¼ 0.003). CONCLUSIONS: Orthopedic surgery residents spend a
large proportion of their time performing documentation/ administrative-type work and their workday can be operationally optimized to minimize nonvalue-adding tasks. Formal workflow analysis may aid program directors in systematic process improvements to better align resident skills with tasks. LEVEL OF EVIDENCE: III ( J Surg 71:216-221. Published
by Elsevier Inc. on behalf of the Association of Program Directors in Surgery) KEYWORDS: orthopedic, resident, workflow, value, analysis, surgery, documentation COMPETENCIES: Patient Care, Medical Knowledge, Prac-
tice-Based Learning and Improvement, Systems-Based Practice
RESULTS: In a 720-minute shift, residents spent on an
average: 191 minutes (26.5%) performing documentation/ administrative duties, 167.0 minutes (23.2%) in direct contact with new patient consults, 129.6 minutes (17.1%) in communication with other providers regarding patients, 116.2 (16.1%) minutes in standby, 63.7 minutes (8.8%) in transit, 32.6 minutes (4.5%) with existing patients, and 20 minutes (2.7%) attending to basic human needs. Residents performed an additional 130 minutes of administrative work off duty. Secondary analysis revealed residents were more likely to perform administrative work rather than
Correspondence: Inquiries to Kamran S. Hamid, MD, MPH, Department of Orthopaedic Surgery, Wake Forest Baptist Medical Center, Medical Center Boulevard, Winston-Salem, NC 27157-1070; fax: þ1-336-716-6286; e-mail: khamid@ wakehealth.edu, [email protected]
216
INTRODUCTION In 2003, the Accreditation Council for Graduate Medical Education (ACGME) limited resident work hours to no more than 80 hours per week.1 The changes were implemented largely to reduce resident fatigue, thereby creating an environment for safe patient care and optimal educational performance.2 Despite these changes, resident fatigue has continued to be an issue.3,4 In 2010, the ACGME published results from an investigative task force established to perform a systematic review of its policies. The task force evaluated reports from the medical community, public, and an Institute of Medicine report calling for further workhour restrictions.5 The 2010 ACGME report led to a
Journal of Surgical Education Published by Elsevier Inc. on behalf of the 1931-7204/$30.00 Association of Program Directors in Surgery http://dx.doi.org/10.1016/j.jsurg.2013.07.006
revised policy and further residency work-hour restrictions that became effective July 1, 2011.6 Despite these revisions, some legislators and academic organizations have continued to call for further restrictions in work hours to bring the United States more in alignment with countries such as the United Kingdom, France, and New Zealand where residents work between 37 and 70 hours per week.7,8 The orthopedic community faces the challenge of providing adequate resident training in the face of decreasing resident work hours, growth in the demand for surgery, greater specialization within the field, and increased complexity of patient care.9 To meet these challenges, orthopedic surgery residents must engage in clinical duties and selfdirected learning with increased efficiency. As understood through business practices, efficient care involves optimizing time, human resources, and productivity while minimizing waste.10 This concept applied to orthopedic residency training implies high-quality care delivered efficiently while maximizing the opportunities for education. Given the mounting need for improved efficiency in orthopedic surgical training, we conducted an orthopedic resident workflow analysis to understand the components of a typical resident workday and to identify inefficient practices in the postgraduate surgical setting. To date, there is no reported formal work-shift analysis of orthopedic residents. Our hypothesis was that orthopedic surgery residents spend a disproportionate amount of time involved in activities that do not relate directly to patient care or maximize resident education.
MATERIALS AND METHODS We selected residents engaged in work on an orthopedic resident consultation service at a major tertiary care center as our subjects. Junior residents from 11 outside medical centers were polled at an orthopedic conference before undertaking the study to assess for generalizability of our potential results. Based on the residents’ responses, we found that this consult service highly approximates the time constraints placed on orthopedic residents at each of the other large multidisciplinary institutions. As did the other centers, our site had a fully integrated electronic medical record which the residents were proficient in. Under an Internal Review Board-approved protocol, we conducted a prospective observational evaluation of residents on this service. We observed 4 Postgraduate Year 3 (PGY-3) residents during alternating 12-hour shifts for 72 consecutive hours (6 consecutive shifts) using 2 trained observers per shift. PGY-3 residents were selected as subjects as they are the most senior residents taking primary call and theoretically will have the least variation in practice as a group with resultant increased precision of results. Each observer was pretrained in shadowing the resident without disrupting resident workflow and independently recorded
minute-by-minute data on resident activities during each 12-hour shift. Study data were collected on standardized log sheets. In recording resident activity, observers were required to assign each recorded activity and its duration to 1 of 6 predefined resident activity categories (defined later). We used Student t tests to determine whether there were any significant differences in the activities that residents were more likely to perform during their call shift. Consult Service The orthopedic surgery consult service at our institution comprises alternating 12-hour shifts among 10 junior orthopedic residents with supervision from chief residents and attending surgeons. Orthopedic surgery residents are consulted by emergency department and inpatient physicians seeking expert musculoskeletal evaluation of patients. These patients are seen and assessed by the orthopedic surgery resident who subsequently discusses patient care with a chief resident and supervising staff physician to arrive at a disposition. Measures A priori, we defined 6 major categories comprising a resident’s workflow, as follows: (1) direct patient contact time with new consults, (2) direct patient contact time with existing patients, (3) time spent communicating with other providers regarding patients, (4) documentation/administrative time, (5) transit time, and (6) basic human needs. Secondarily we defined a seventh category for resident duties not fitting within these “necessary” categories, we termed this seventh category as standby. Owing to the lack of a formalized didactic structure on this consult service during the weekend, we did not have a separate category for education, which is obviously a necessary part of every resident’s daily workflow. On most consult services, learning is often built into the workup and diagnosis of the consult patient. Direct patient contact time was defined as time spent directly in the presence of patients—this time could be spent either with a newly consulted patient or with an existing patient. Communications regarding patients was defined as communication with other physicians or health care providers regarding patient care. This category included all forms of verbal communication such as telephone calls and direct communications with nurses and ancillary staff. Documentation/administrative time was defined as time spent performing duties including history and physical examination documentation, any form of electronic or paper patient note writing, preparing discharge paperwork, updating of patient lists, prescription writing, reviewing documentation from other providers, or physician order entry. Transit time was defined as time spent in commute and encompassed time spent walking between specific destinations and duties. Basic human needs were defined as time spent eating,
Journal of Surgical Education Volume 71/Number 2 March/April 2014
217
sleeping, and stepping away to use the restroom. Standby time was calculated post hoc as time not dedicated to these 6 major categories (i.e., 720 min minus the sum of time spent on the 6 other categories). In this study, we measured all clinical activities in terms of their value, which is defined as the activity’s contribution toward a patient’s health outcomes per resident time spent (a surrogate for health care dollars). Activities that contribute directly to improving patient outcomes, such as direct patient contact and communication with other providers, are value positive because the physician is using this time to gather data to make a diagnosis, performing procedures, or communicating and jointly making decisions with other members of the patient’s care team, which would directly affect the patient’s clinical outcome. Documentation/ administrative tasks serve a dual purpose of billing and communication; although much of documentation is gratuitous, it is still often essential for communication thus it has been deemed value-neutral overall. Transit time and basic human needs will be considered value positive because they cannot be replaced at this time, though technological gains are quickly supplanting the need for transit. Standby time is a value-negative activity because neither does it improve the patient’s outcome nor is it an activity without which the patient’s care would be compromised. Based on work suggesting that work-hour restrictions have caused an increase in postshift documentation/administrative-type work,11 residents were also instructed to email their respective data abstractors a self-reported approximation of the amount of time spent performing documentation/administrative tasks after each shift was over. Residents were asked to report amount of time spent updating electronic patient lists and writing notes for patients seen during that particular shift. Subjects We recruited and obtained consent from all 4 orthopedic surgery residents designated to be on call for the consult service during our randomly designated 72-hour study period.
Subjects were blinded as to the purpose of the study and were encouraged to pursue their daily activities in a normal manner as possible. Residents were also instructed not to interact with the observers. All 4 residents were PGY-3 orthopedic surgery residents from a 25-resident orthopedic surgery residency program in a southeastern metropolitan area. Statistical Analysis Descriptive data were calculated by averaging times collected by each pair of independent data abstractors, with calculated times independently verified by 2 authors. Differences within categories for each shift were deemed to be adequate if less than 15 minutes (2% of the shift). We determined clinically relevant workflow category comparisons a priori and used Student t tests to determine statistical significance between compared categories. All analyses were conducted using SAS version 9.2 software (SAS Institute Inc., Cary, NC).
RESULTS In a 12-hour shift (720 min), orthopedic residents included as part of this study spent on average: 191 minutes (26.5%) performing documentation/administrative duties, 167.0 minutes (23.2%) in direct contact with new patient consults, 129.6 minutes (17.1%) in communication regarding patients, 63.7 minutes (8.8%) in transit, 32.6 minutes (4.5%) with existing patients, 20 minutes (2.7%) attending to basic human needs, and 116.2 (16.1%) minutes in standby (Table 1). Of note, an aggregate of zero minutes was spent on sleep by all residents during the entirety of the study and time allocated to basic human needs ranged from a minimum of 1 minute over 12 hours to a maximum of 41 minutes. Residents performed an additional 130 minutes of administrative work after each 12-hour shift: of this, an average of 25 minutes was spent updating patient lists and 105 minutes were spent on postcall note writing (Fig. 1). Our secondary analysis revealed that for each call shift, there was no statistically significant difference in the time
TABLE 1. Resident Time Allocation During 12-hour Orthopedic Consultation Shifts (Minutes) Direct Patient Contact Time‐ New Consults Shift 1 Shift 2 Shift 3 Shift 4 Shift 5 Shift 6 Averages per 12-h shift
159.5 193.0 170.0 168.0 126.5 185.0 167.0
Direct Patient Contact TimeExisting Patients 12.0 15.0 87.5 16.0 21.5 43.5 32.6
Communication Documentation/ Basic Regarding Administrative Transit Human Patients Time Time Needs* Standby 119.0 169.0 167.0 34.5 180 108.0 129.6
80.5 146.5 149.0 377.0 135.5 257.5 191.0
45.0 71.0 62.5 57.0 88 58.5 63.7
22.5 15.5 1.0 18.0 41 22.0 20.0
281.5 110.0 83.0 49.5 127.5 45.5 116.2
*A total of zero minutes were spent sleeping by the call residents over the study period. 218
Journal of Surgical Education Volume 71/Number 2 March/April 2014
FIGURE 1. Average time spent by call residents on activities per 12-hour orthopedic consultation service shift (minutes).
shift residents spent doing administrative work vs time spent in direct contact with new patients or time spent communicating with other health care personnel. However, residents were more likely to perform administrative work rather than directly interact with existing patients (p ¼ 0.006) or attend to basic human needs (p ¼ 0.003). Residents were also more likely to spend time in transit (p ¼ 0.003) or in standby (p ¼ 0.02) than attend to essential needs (Table 2). During our 72-hour study period, there were 57 new consultations, an average of 9.5 orthopedic consults per 12hour shift. In the 52 weeks leading up to our study, there were on an average 45.6 consults for similar 72-hour periods (range: 27-73; standard deviation: 11.3) (Fig. 2).
DISCUSSION In this study, we characterize the workflow for orthopedic surgery residents on a busy consult service at a major tertiary
care center. We sought to characterize typical time allocation for various resident activities and identify areas for process improvement. Although residents were engaged in a diverse array of duties for significant portions of their shifts, documentation/administrative duties represented the largest component of their workday. Further, residents continued to engage in these documentation and administrative duties beyond official shift hours. Additionally, residents spent a significant proportion of their time in standby—nonvaluecreating activity. A prior study by Dassinger et al.12 investigating surgical resident workflow followed 1 senior surgical resident on an inpatient pediatric surgical unit for a total of 18.9 hours. Based on their activity classification criteria, the authors found that this resident spent 58% of her time engaged in patient care, 15% in education, 12% in transit, 12% in communication, and 1.5% in “pure waste.” The authors encourage increased workflow efficiency to decrease time spent by residents performing nonvalue-adding activity,
TABLE 2. Clinically Relevant Time Allocation Category Comparisons Documentation/administrative time 191.0 min Documentation/administrative time 191.0 min Documentation/administrative time 191.0 min Documentation/administrative time 191.0 min Transit time 63.7 min Standby 116.2 min Standby 116.2 min Standby 116.2 min
Direct contact time with existing patients 32.6 min Direct contact time with new patients 167.0 min Communication regarding patients 129.6 min Basic human needs 20.0 min Basic human needs 20.0 min Basic human needs 20.0 min Direct contact time with new patients 167.0 min Direct contact time with existing patients 32.6 min
Journal of Surgical Education Volume 71/Number 2 March/April 2014
p ¼ 0.006 p ¼ 0.6 p ¼ 0.2 p ¼ 0.003 p ¼ 0.003 p ¼ 0.02 p ¼ 0.2 p ¼ 0.05
219
FIGURE 2. Emergency department and inpatient orthopedic consultations per 72-hour weekend in the 52 weekends leading into the study period. (Red bar denotes study period.)
namely phone communication, social work/care coordination, transit, and waste. Although the study offers many novel perspectives, the work has several weaknesses including poor generalizability. In most surgical residencies, senior residents tend to engage in more value-positive activity (i.e., most spend their time in the operating room or in didactics and are less frequently involved in the documentation or administrative role often apportioned to junior residents). Thus the study does not fully capture the possibilities for suboptimal resident workflow. Our study represents the first attempt in the literature to quantify orthopedic residents’ workflow. A major finding of our study is that residents spend a large proportion of time performing documentation/administrative tasks. In a strict sense, documentation/administrative tasks are essential to patient care for communicating between teams, patient workup, education—a portion of these tasks can be considered value-positive activity. Additionally, learning appropriate documentation during residency is an important skill to gain for medicolegal and coding/billing purposes once out in practice. However, the majority of this work is often a duplication of efforts and is considered to produce little value and is not aligned with the 6 core competencies of medical education espoused by the ACGME.13 Attending physicians are not immune to the effects of burdensome administrative work resulting in decreased time with patients and ancillary staff.14 For these reasons, administrative-type work typically falls to the most junior members of orthopedic rotations. Several authors have noted that since the 80-hour workweek was introduced, administrative-type workload, and senior resident case volume has stayed the same, whereas junior resident caseload has dramatically decreased.15,16 This trend suggests that junior residents are disproportionately performing administrative-type work at the expense of educational experiences. Given the ACGMEs expectations for resident education and the resources invested on educating orthopedic residents, duplication of effort is now an unacceptably poor return on investment. 220
We did not explicitly track the specific components of standby time; however, qualitative data from our abstractors suggest that this nonvalue-creating time included activities such as time spent waiting to communicate via pager system, time spent attempting to locate health care personnel associated with patients’ care after arriving at the right patient location, time spent waiting for study results to arrive or trying to locate final impressions/results for studies, and time spent waiting to log on to work stations and other miscellaneous activity. One solution to reduce standby time is for individual institutions to adopt an operations management perspective to resident workflow. Such an approach would involve conducting even more granular workflow analyses to identify 1 or 2 major nonvalue-creating operation processes during a resident’s workday and then implementing new procedures to address the identified workflow bottlenecks. For example, at our institution, we have become aware that resident use of pagers creates significant workflow disruptions. Residents often receive pages that communicate very little information and without knowing whether the page is regarding a mundane or life-threatening issue residents will spend precious time locating a phone or workstation or both to track down and respond to the page sender. We have thus begun transitioning to advanced technological platforms utilizing smart phones that best allow residents to coordinate care with the greatest operational efficiency. Based on the results of this investigation, several process innovations have been implemented at our institution to reduce value-neutral and value-negative resident activity. Our study has limitations, including limited external validity. However, we believe that for this particular type of study our single center approach is optimal. Multicentered workflow data would likely obscure trends and would not allow program directors to suitably identify and isolate internal processes in need of improvement. As with all studies in which subjects are being directly observed, this study is subject to information bias as result of the Hawthorne effect.
Journal of Surgical Education Volume 71/Number 2 March/April 2014
This study highlights the importance for orthopedic residency programs and residency directors to develop awareness of the components of residents’ workdays. In particular, residencies should look to minimize standby time, nonvalue-adding activity, and in scenarios where documentation/administrative-type tasks comprise a sizeable portion of a resident’s workflow, programs should reassess whether such tasks are appropriate to their residents’ skill level, training, and educational needs.
CONCLUSIONS Orthopedic surgery residents spend a large proportion of their time performing documentation/administrativetype work, and their workday can be operationally optimized to minimize nonvalue-adding tasks. Formal workflow analysis may aid program directors in systematic process improvements to better align resident skills with tasks.
ACKNOWLEDGMENTS Special thanks to Brent Witten, MD and Shasta Henderson, MD for their contributions to this manuscript.
2008/Resident-Duty-Hours-Enhancing-Sleep-Supervisio n-and-Safety.aspx.〉; 2008 Accessed July 27, 2011. 6. The ACGME 2011 Duty Hour Standards: Enhancing
Quality of Care, Supervision and Resident Professional Development. Available at: 〈http://www.acgme-2010s tandards.org/pdf/monographs/jgme-monograph.pdf〉; 2011 Accessed July 27, 2011. 7. Blum AB, Raiszadeh F, Shea S, et al. US public
opinion regarding proposed limits on resident physician work hours. BMC Med. 2010;8:1-12. 8. Czeisler CA. Wake-promoting therapeutic medications
not an appropriate alternative to implementation of safer work schedules for resident physicians. Mayo Clin Proc. 2010;85:302-303. 9. Robbins L, Bostrom M, Craig E, Sculco TP. Proposals
for change in orthopaedic education: recommendations from an Orthopaedic Residency Directors’ Peer Forum. J Bone Joint Surg Am. 2010;92:245-249. 10. Stratton TD, Rudy DW, Sauer MJ, Perman JA,
Jennings CD. Lessons from industry: one school’s transformation toward “lean” curricular governance. Acad Med. 2007;82:331-340. 11. Mourad M, Vidyarthi AR, Hollander H, Ranji SR.
REFERENCES 1. Common Program Requirements 2003. Available at:
〈http://www.acgme.org/acWebsite/dutyHours/dh_du tyHoursCommonPR.pdf〉; Accessed July 27, 2011.
2. Gaba DM, Howard SK. Patient safety: fatigue among
Shifting indirect patient care duties to after hours in the era of work hours restrictions. Acad Med. 2011;86:586-590. 12. Dassinger MS 3rd, Eubanks JW 3rd, Langham MR Jr.
Full work analysis of resident work hours. J Surg Res. 2008;147:178-181.
clinicians and the safety of patients. N Engl J Med. 2002;347:1249-1255.
13. ACGME Core Competencies. Available at: 〈http://www.
3. Pelligrini VD Jr, Peabody T, Dinges DF, Moody J,
14. Park SY, Lee SY, Chen Y. The effects of EMR
Fabri PF. Symposium resident work-hour guidelines. A sentence or an opportunity for orthopaedic education? J Bone Joint Surg Am. 2005;87:2576-2586.
deployment on doctors’ work practices: a qualitative study in the emergency department of a teaching hospital. Int J Med Inform. 2012;81:204-217.
4. Kamath AF, Baldwin K, Meade LK, Powell AC,
15. Carlin AM, Gasevic E, Shepard AD. Effect of the 80-
Mehta S. The increased financial burden of further proposed orthopaedic resident work-hour reductions. J Bone Joint Surg Am. 2011;93:e31(1–8).
5. Resident Duty Hours: Enhancing Sleep, Supervision,
and Safety. Available at: 〈http://iom.edu/Reports/
acgme.org/acwebsite/RRC_280/280_corecomp.asp〉; 2011.
hour work week on resident operative experience in general surgery. Am J Surg. 2007;193:326. 16. Spencer AU, Teitelbaum DH. Impact of work hour
restrictions on residents’ operative volume on a subspecialty surgical service. J Am Coll Surg. 2005;200:670.
Journal of Surgical Education Volume 71/Number 2 March/April 2014
221
Orthopedic Resident Work-Shift Analysis: Are We Making the Best Use of Resident Work Hours? Kamran S. Hamid, MD,* Benedict U. Nwachukwu, MD,† Eugene Hsu, MD,‡ Colston A. Edgerton, BS,* David R. Hobson, MS,* and Jason E. Lang, MD* *
Department of Orthopaedic Surgery, Wake Forest Baptist Medical Center, Winston-Salem, North Carolina; Hospital for Special Surgery, New York, New York; and ‡Department of Anesthesiology & Critical Care Medicine, The Johns Hopkins Hospital, Baltimore, Maryland †
BACKGROUND: Surgery programs have been tasked to
meet rising demands in patient surgical care while simultaneously providing adequate resident training in the midst of increasing resident work-hour restrictions. The purpose of this study was to quantify orthopedic surgery resident workflow and identify areas needing improved resident efficiency. We hypothesize that residents spend a disproportionate amount of time involved in activities that do not relate directly to patient care or maximize resident education. METHODS: We observed 4 orthopedic surgery residents on
the orthopedic consult service at a major tertiary care center for 72 consecutive hours (6 consecutive shifts). We collected minute-by-minute data using predefined work-task criteria: direct new patient contact, direct existing patient contact, communications with other providers, documentation/ administrative time, transit time, and basic human needs. A seventh category comprised remaining less-productive work was termed as standby.
directly interact with existing patients (p ¼ 0.006) or attend to basic human needs (p ¼ 0.003). CONCLUSIONS: Orthopedic surgery residents spend a
large proportion of their time performing documentation/ administrative-type work and their workday can be operationally optimized to minimize nonvalue-adding tasks. Formal workflow analysis may aid program directors in systematic process improvements to better align resident skills with tasks. LEVEL OF EVIDENCE: III ( J Surg 71:216-221. Published
by Elsevier Inc. on behalf of the Association of Program Directors in Surgery) KEYWORDS: orthopedic, resident, workflow, value, analysis, surgery, documentation COMPETENCIES: Patient Care, Medical Knowledge, Prac-
tice-Based Learning and Improvement, Systems-Based Practice
RESULTS: In a 720-minute shift, residents spent on an
average: 191 minutes (26.5%) performing documentation/ administrative duties, 167.0 minutes (23.2%) in direct contact with new patient consults, 129.6 minutes (17.1%) in communication with other providers regarding patients, 116.2 (16.1%) minutes in standby, 63.7 minutes (8.8%) in transit, 32.6 minutes (4.5%) with existing patients, and 20 minutes (2.7%) attending to basic human needs. Residents performed an additional 130 minutes of administrative work off duty. Secondary analysis revealed residents were more likely to perform administrative work rather than
Correspondence: Inquiries to Kamran S. Hamid, MD, MPH, Department of Orthopaedic Surgery, Wake Forest Baptist Medical Center, Medical Center Boulevard, Winston-Salem, NC 27157-1070; fax: þ1-336-716-6286; e-mail: khamid@ wakehealth.edu, [email protected]
216
INTRODUCTION In 2003, the Accreditation Council for Graduate Medical Education (ACGME) limited resident work hours to no more than 80 hours per week.1 The changes were implemented largely to reduce resident fatigue, thereby creating an environment for safe patient care and optimal educational performance.2 Despite these changes, resident fatigue has continued to be an issue.3,4 In 2010, the ACGME published results from an investigative task force established to perform a systematic review of its policies. The task force evaluated reports from the medical community, public, and an Institute of Medicine report calling for further workhour restrictions.5 The 2010 ACGME report led to a
Journal of Surgical Education Published by Elsevier Inc. on behalf of the 1931-7204/$30.00 Association of Program Directors in Surgery http://dx.doi.org/10.1016/j.jsurg.2013.07.006
revised policy and further residency work-hour restrictions that became effective July 1, 2011.6 Despite these revisions, some legislators and academic organizations have continued to call for further restrictions in work hours to bring the United States more in alignment with countries such as the United Kingdom, France, and New Zealand where residents work between 37 and 70 hours per week.7,8 The orthopedic community faces the challenge of providing adequate resident training in the face of decreasing resident work hours, growth in the demand for surgery, greater specialization within the field, and increased complexity of patient care.9 To meet these challenges, orthopedic surgery residents must engage in clinical duties and selfdirected learning with increased efficiency. As understood through business practices, efficient care involves optimizing time, human resources, and productivity while minimizing waste.10 This concept applied to orthopedic residency training implies high-quality care delivered efficiently while maximizing the opportunities for education. Given the mounting need for improved efficiency in orthopedic surgical training, we conducted an orthopedic resident workflow analysis to understand the components of a typical resident workday and to identify inefficient practices in the postgraduate surgical setting. To date, there is no reported formal work-shift analysis of orthopedic residents. Our hypothesis was that orthopedic surgery residents spend a disproportionate amount of time involved in activities that do not relate directly to patient care or maximize resident education.
MATERIALS AND METHODS We selected residents engaged in work on an orthopedic resident consultation service at a major tertiary care center as our subjects. Junior residents from 11 outside medical centers were polled at an orthopedic conference before undertaking the study to assess for generalizability of our potential results. Based on the residents’ responses, we found that this consult service highly approximates the time constraints placed on orthopedic residents at each of the other large multidisciplinary institutions. As did the other centers, our site had a fully integrated electronic medical record which the residents were proficient in. Under an Internal Review Board-approved protocol, we conducted a prospective observational evaluation of residents on this service. We observed 4 Postgraduate Year 3 (PGY-3) residents during alternating 12-hour shifts for 72 consecutive hours (6 consecutive shifts) using 2 trained observers per shift. PGY-3 residents were selected as subjects as they are the most senior residents taking primary call and theoretically will have the least variation in practice as a group with resultant increased precision of results. Each observer was pretrained in shadowing the resident without disrupting resident workflow and independently recorded
minute-by-minute data on resident activities during each 12-hour shift. Study data were collected on standardized log sheets. In recording resident activity, observers were required to assign each recorded activity and its duration to 1 of 6 predefined resident activity categories (defined later). We used Student t tests to determine whether there were any significant differences in the activities that residents were more likely to perform during their call shift. Consult Service The orthopedic surgery consult service at our institution comprises alternating 12-hour shifts among 10 junior orthopedic residents with supervision from chief residents and attending surgeons. Orthopedic surgery residents are consulted by emergency department and inpatient physicians seeking expert musculoskeletal evaluation of patients. These patients are seen and assessed by the orthopedic surgery resident who subsequently discusses patient care with a chief resident and supervising staff physician to arrive at a disposition. Measures A priori, we defined 6 major categories comprising a resident’s workflow, as follows: (1) direct patient contact time with new consults, (2) direct patient contact time with existing patients, (3) time spent communicating with other providers regarding patients, (4) documentation/administrative time, (5) transit time, and (6) basic human needs. Secondarily we defined a seventh category for resident duties not fitting within these “necessary” categories, we termed this seventh category as standby. Owing to the lack of a formalized didactic structure on this consult service during the weekend, we did not have a separate category for education, which is obviously a necessary part of every resident’s daily workflow. On most consult services, learning is often built into the workup and diagnosis of the consult patient. Direct patient contact time was defined as time spent directly in the presence of patients—this time could be spent either with a newly consulted patient or with an existing patient. Communications regarding patients was defined as communication with other physicians or health care providers regarding patient care. This category included all forms of verbal communication such as telephone calls and direct communications with nurses and ancillary staff. Documentation/administrative time was defined as time spent performing duties including history and physical examination documentation, any form of electronic or paper patient note writing, preparing discharge paperwork, updating of patient lists, prescription writing, reviewing documentation from other providers, or physician order entry. Transit time was defined as time spent in commute and encompassed time spent walking between specific destinations and duties. Basic human needs were defined as time spent eating,
Journal of Surgical Education Volume 71/Number 2 March/April 2014
217
sleeping, and stepping away to use the restroom. Standby time was calculated post hoc as time not dedicated to these 6 major categories (i.e., 720 min minus the sum of time spent on the 6 other categories). In this study, we measured all clinical activities in terms of their value, which is defined as the activity’s contribution toward a patient’s health outcomes per resident time spent (a surrogate for health care dollars). Activities that contribute directly to improving patient outcomes, such as direct patient contact and communication with other providers, are value positive because the physician is using this time to gather data to make a diagnosis, performing procedures, or communicating and jointly making decisions with other members of the patient’s care team, which would directly affect the patient’s clinical outcome. Documentation/ administrative tasks serve a dual purpose of billing and communication; although much of documentation is gratuitous, it is still often essential for communication thus it has been deemed value-neutral overall. Transit time and basic human needs will be considered value positive because they cannot be replaced at this time, though technological gains are quickly supplanting the need for transit. Standby time is a value-negative activity because neither does it improve the patient’s outcome nor is it an activity without which the patient’s care would be compromised. Based on work suggesting that work-hour restrictions have caused an increase in postshift documentation/administrative-type work,11 residents were also instructed to email their respective data abstractors a self-reported approximation of the amount of time spent performing documentation/administrative tasks after each shift was over. Residents were asked to report amount of time spent updating electronic patient lists and writing notes for patients seen during that particular shift. Subjects We recruited and obtained consent from all 4 orthopedic surgery residents designated to be on call for the consult service during our randomly designated 72-hour study period.
Subjects were blinded as to the purpose of the study and were encouraged to pursue their daily activities in a normal manner as possible. Residents were also instructed not to interact with the observers. All 4 residents were PGY-3 orthopedic surgery residents from a 25-resident orthopedic surgery residency program in a southeastern metropolitan area. Statistical Analysis Descriptive data were calculated by averaging times collected by each pair of independent data abstractors, with calculated times independently verified by 2 authors. Differences within categories for each shift were deemed to be adequate if less than 15 minutes (2% of the shift). We determined clinically relevant workflow category comparisons a priori and used Student t tests to determine statistical significance between compared categories. All analyses were conducted using SAS version 9.2 software (SAS Institute Inc., Cary, NC).
RESULTS In a 12-hour shift (720 min), orthopedic residents included as part of this study spent on average: 191 minutes (26.5%) performing documentation/administrative duties, 167.0 minutes (23.2%) in direct contact with new patient consults, 129.6 minutes (17.1%) in communication regarding patients, 63.7 minutes (8.8%) in transit, 32.6 minutes (4.5%) with existing patients, 20 minutes (2.7%) attending to basic human needs, and 116.2 (16.1%) minutes in standby (Table 1). Of note, an aggregate of zero minutes was spent on sleep by all residents during the entirety of the study and time allocated to basic human needs ranged from a minimum of 1 minute over 12 hours to a maximum of 41 minutes. Residents performed an additional 130 minutes of administrative work after each 12-hour shift: of this, an average of 25 minutes was spent updating patient lists and 105 minutes were spent on postcall note writing (Fig. 1). Our secondary analysis revealed that for each call shift, there was no statistically significant difference in the time
TABLE 1. Resident Time Allocation During 12-hour Orthopedic Consultation Shifts (Minutes) Direct Patient Contact Time‐ New Consults Shift 1 Shift 2 Shift 3 Shift 4 Shift 5 Shift 6 Averages per 12-h shift
159.5 193.0 170.0 168.0 126.5 185.0 167.0
Direct Patient Contact TimeExisting Patients 12.0 15.0 87.5 16.0 21.5 43.5 32.6
Communication Documentation/ Basic Regarding Administrative Transit Human Patients Time Time Needs* Standby 119.0 169.0 167.0 34.5 180 108.0 129.6
80.5 146.5 149.0 377.0 135.5 257.5 191.0
45.0 71.0 62.5 57.0 88 58.5 63.7
22.5 15.5 1.0 18.0 41 22.0 20.0
281.5 110.0 83.0 49.5 127.5 45.5 116.2
*A total of zero minutes were spent sleeping by the call residents over the study period. 218
Journal of Surgical Education Volume 71/Number 2 March/April 2014
FIGURE 1. Average time spent by call residents on activities per 12-hour orthopedic consultation service shift (minutes).
shift residents spent doing administrative work vs time spent in direct contact with new patients or time spent communicating with other health care personnel. However, residents were more likely to perform administrative work rather than directly interact with existing patients (p ¼ 0.006) or attend to basic human needs (p ¼ 0.003). Residents were also more likely to spend time in transit (p ¼ 0.003) or in standby (p ¼ 0.02) than attend to essential needs (Table 2). During our 72-hour study period, there were 57 new consultations, an average of 9.5 orthopedic consults per 12hour shift. In the 52 weeks leading up to our study, there were on an average 45.6 consults for similar 72-hour periods (range: 27-73; standard deviation: 11.3) (Fig. 2).
DISCUSSION In this study, we characterize the workflow for orthopedic surgery residents on a busy consult service at a major tertiary
care center. We sought to characterize typical time allocation for various resident activities and identify areas for process improvement. Although residents were engaged in a diverse array of duties for significant portions of their shifts, documentation/administrative duties represented the largest component of their workday. Further, residents continued to engage in these documentation and administrative duties beyond official shift hours. Additionally, residents spent a significant proportion of their time in standby—nonvaluecreating activity. A prior study by Dassinger et al.12 investigating surgical resident workflow followed 1 senior surgical resident on an inpatient pediatric surgical unit for a total of 18.9 hours. Based on their activity classification criteria, the authors found that this resident spent 58% of her time engaged in patient care, 15% in education, 12% in transit, 12% in communication, and 1.5% in “pure waste.” The authors encourage increased workflow efficiency to decrease time spent by residents performing nonvalue-adding activity,
TABLE 2. Clinically Relevant Time Allocation Category Comparisons Documentation/administrative time 191.0 min Documentation/administrative time 191.0 min Documentation/administrative time 191.0 min Documentation/administrative time 191.0 min Transit time 63.7 min Standby 116.2 min Standby 116.2 min Standby 116.2 min
Direct contact time with existing patients 32.6 min Direct contact time with new patients 167.0 min Communication regarding patients 129.6 min Basic human needs 20.0 min Basic human needs 20.0 min Basic human needs 20.0 min Direct contact time with new patients 167.0 min Direct contact time with existing patients 32.6 min
Journal of Surgical Education Volume 71/Number 2 March/April 2014
p ¼ 0.006 p ¼ 0.6 p ¼ 0.2 p ¼ 0.003 p ¼ 0.003 p ¼ 0.02 p ¼ 0.2 p ¼ 0.05
219
FIGURE 2. Emergency department and inpatient orthopedic consultations per 72-hour weekend in the 52 weekends leading into the study period. (Red bar denotes study period.)
namely phone communication, social work/care coordination, transit, and waste. Although the study offers many novel perspectives, the work has several weaknesses including poor generalizability. In most surgical residencies, senior residents tend to engage in more value-positive activity (i.e., most spend their time in the operating room or in didactics and are less frequently involved in the documentation or administrative role often apportioned to junior residents). Thus the study does not fully capture the possibilities for suboptimal resident workflow. Our study represents the first attempt in the literature to quantify orthopedic residents’ workflow. A major finding of our study is that residents spend a large proportion of time performing documentation/administrative tasks. In a strict sense, documentation/administrative tasks are essential to patient care for communicating between teams, patient workup, education—a portion of these tasks can be considered value-positive activity. Additionally, learning appropriate documentation during residency is an important skill to gain for medicolegal and coding/billing purposes once out in practice. However, the majority of this work is often a duplication of efforts and is considered to produce little value and is not aligned with the 6 core competencies of medical education espoused by the ACGME.13 Attending physicians are not immune to the effects of burdensome administrative work resulting in decreased time with patients and ancillary staff.14 For these reasons, administrative-type work typically falls to the most junior members of orthopedic rotations. Several authors have noted that since the 80-hour workweek was introduced, administrative-type workload, and senior resident case volume has stayed the same, whereas junior resident caseload has dramatically decreased.15,16 This trend suggests that junior residents are disproportionately performing administrative-type work at the expense of educational experiences. Given the ACGMEs expectations for resident education and the resources invested on educating orthopedic residents, duplication of effort is now an unacceptably poor return on investment. 220
We did not explicitly track the specific components of standby time; however, qualitative data from our abstractors suggest that this nonvalue-creating time included activities such as time spent waiting to communicate via pager system, time spent attempting to locate health care personnel associated with patients’ care after arriving at the right patient location, time spent waiting for study results to arrive or trying to locate final impressions/results for studies, and time spent waiting to log on to work stations and other miscellaneous activity. One solution to reduce standby time is for individual institutions to adopt an operations management perspective to resident workflow. Such an approach would involve conducting even more granular workflow analyses to identify 1 or 2 major nonvalue-creating operation processes during a resident’s workday and then implementing new procedures to address the identified workflow bottlenecks. For example, at our institution, we have become aware that resident use of pagers creates significant workflow disruptions. Residents often receive pages that communicate very little information and without knowing whether the page is regarding a mundane or life-threatening issue residents will spend precious time locating a phone or workstation or both to track down and respond to the page sender. We have thus begun transitioning to advanced technological platforms utilizing smart phones that best allow residents to coordinate care with the greatest operational efficiency. Based on the results of this investigation, several process innovations have been implemented at our institution to reduce value-neutral and value-negative resident activity. Our study has limitations, including limited external validity. However, we believe that for this particular type of study our single center approach is optimal. Multicentered workflow data would likely obscure trends and would not allow program directors to suitably identify and isolate internal processes in need of improvement. As with all studies in which subjects are being directly observed, this study is subject to information bias as result of the Hawthorne effect.
Journal of Surgical Education Volume 71/Number 2 March/April 2014
This study highlights the importance for orthopedic residency programs and residency directors to develop awareness of the components of residents’ workdays. In particular, residencies should look to minimize standby time, nonvalue-adding activity, and in scenarios where documentation/administrative-type tasks comprise a sizeable portion of a resident’s workflow, programs should reassess whether such tasks are appropriate to their residents’ skill level, training, and educational needs.
CONCLUSIONS Orthopedic surgery residents spend a large proportion of their time performing documentation/administrativetype work, and their workday can be operationally optimized to minimize nonvalue-adding tasks. Formal workflow analysis may aid program directors in systematic process improvements to better align resident skills with tasks.
ACKNOWLEDGMENTS Special thanks to Brent Witten, MD and Shasta Henderson, MD for their contributions to this manuscript.
2008/Resident-Duty-Hours-Enhancing-Sleep-Supervisio n-and-Safety.aspx.〉; 2008 Accessed July 27, 2011. 6. The ACGME 2011 Duty Hour Standards: Enhancing
Quality of Care, Supervision and Resident Professional Development. Available at: 〈http://www.acgme-2010s tandards.org/pdf/monographs/jgme-monograph.pdf〉; 2011 Accessed July 27, 2011. 7. Blum AB, Raiszadeh F, Shea S, et al. US public
opinion regarding proposed limits on resident physician work hours. BMC Med. 2010;8:1-12. 8. Czeisler CA. Wake-promoting therapeutic medications
not an appropriate alternative to implementation of safer work schedules for resident physicians. Mayo Clin Proc. 2010;85:302-303. 9. Robbins L, Bostrom M, Craig E, Sculco TP. Proposals
for change in orthopaedic education: recommendations from an Orthopaedic Residency Directors’ Peer Forum. J Bone Joint Surg Am. 2010;92:245-249. 10. Stratton TD, Rudy DW, Sauer MJ, Perman JA,
Jennings CD. Lessons from industry: one school’s transformation toward “lean” curricular governance. Acad Med. 2007;82:331-340. 11. Mourad M, Vidyarthi AR, Hollander H, Ranji SR.
REFERENCES 1. Common Program Requirements 2003. Available at:
〈http://www.acgme.org/acWebsite/dutyHours/dh_du tyHoursCommonPR.pdf〉; Accessed July 27, 2011.
2. Gaba DM, Howard SK. Patient safety: fatigue among
Shifting indirect patient care duties to after hours in the era of work hours restrictions. Acad Med. 2011;86:586-590. 12. Dassinger MS 3rd, Eubanks JW 3rd, Langham MR Jr.
Full work analysis of resident work hours. J Surg Res. 2008;147:178-181.
clinicians and the safety of patients. N Engl J Med. 2002;347:1249-1255.
13. ACGME Core Competencies. Available at: 〈http://www.
3. Pelligrini VD Jr, Peabody T, Dinges DF, Moody J,
14. Park SY, Lee SY, Chen Y. The effects of EMR
Fabri PF. Symposium resident work-hour guidelines. A sentence or an opportunity for orthopaedic education? J Bone Joint Surg Am. 2005;87:2576-2586.
deployment on doctors’ work practices: a qualitative study in the emergency department of a teaching hospital. Int J Med Inform. 2012;81:204-217.
4. Kamath AF, Baldwin K, Meade LK, Powell AC,
15. Carlin AM, Gasevic E, Shepard AD. Effect of the 80-
Mehta S. The increased financial burden of further proposed orthopaedic resident work-hour reductions. J Bone Joint Surg Am. 2011;93:e31(1–8).
5. Resident Duty Hours: Enhancing Sleep, Supervision,
and Safety. Available at: 〈http://iom.edu/Reports/
acgme.org/acwebsite/RRC_280/280_corecomp.asp〉; 2011.
hour work week on resident operative experience in general surgery. Am J Surg. 2007;193:326. 16. Spencer AU, Teitelbaum DH. Impact of work hour
restrictions on residents’ operative volume on a subspecialty surgical service. J Am Coll Surg. 2005;200:670.
Journal of Surgical Education Volume 71/Number 2 March/April 2014
221
