Moore, Shelly Smith MS, AdEd; Nettleman, Mary D. MD, MS; Beyer, Sandy MPA, MD; Chalasani, Kalpana MD; Fairbanks, Rollin J. MD, MS; Goyal, Munish MD; Carter, Miquel
Ms. Smith Moore is assistant professor and Dr. Nettleman is professor and chair, Division of General Medicine, Department of Internal Medicine; Dr. Beyer, Dr. Chalasani, Dr. Fairbanks, and Dr. Goyal were medical students, all at the Virginia Commonwealth University School of Medicine, when this article was written. Mr. Carter is a student intern, Virginia Commonwealth University, Richmond.
Correspondence and requests for reprints should be addressed to Dr. Nettleman, Associate Professor and Chair, Division of General Medicine, Virginia Commonwealth University, P.O. Box 980102, Richmond, VA 23298-0102.
This study was sponsored in part by the Generalist Physician Initiative of Virginia and the Division of General Internal Medicine, Department of Internal Medicine Virginia Commonwealth University.
The authors are indebted to the house officers who made the study possible, and to the residency directors of the two programs.
Purpose: Night call is a significant part of residents' education, but little information about their night-call activities is available. This study recorded residents' activities during night-call rotations on internal medicine and pediatrics wards.
Method: In June and July 1997, on-call pediatrics and internal medicine residents at an urban academic medical center were accompanied by trained observers on the general wards between the hours of 7 PM and 7 AM. The types and duration of activities were recorded.
Results: Residents were observed for 106 nights. Internal medicine and pediatrics residents spent their time similarly. They spent 5.3 hours and 5.7 hours per night, respectively, on “basic” activities such as eating, resting, chatting, and sleeping, and an average of 2.6 hours and 2.2 hours, respectively, on chart review and documentation. In both programs, discussing the case with team members averaged 1.5 hours per night and use of the computer averaged slightly more than half an hour. Internal medicine residents spent approximately 1.5 hours on patients' history and physical examinations while pediatrics residents spent 1.3 hours. With each new patient, internal medicine residents spent an average of 19.7 minutes and pediatrics residents spent 16.5 minutes. The only significant difference between the two groups of residents was that the pediatrics residents spent more time per night on procedures than did the internal medicine residents (37 minutes versus 14 minutes, p < 0.01).
Conclusions: Residents from both programs spent a surprising amount of time each night on chart review and documentation. In fact, they spent more time with charts than with patients. Whether this activity truly contributes to residents' education or improved patients' outcomes is not clear.
More than a decade ago, Lurie and colleagues published an in-depth analysis of the activities of internal medicine housestaff during night call.1 At that time, the American College of Physicians was struggling with the issue of residents' workloads,2 and Lurie's data served as an important framework for discussion and decision making. Now, 11 years later, the entire face of medicine has changed because of the rise of managed care and increasing regulations of both patient care and resident education.
Although some focused studies have been performed,3–10 no current and comprehensive study has evaluated how house officers spend their nights. To document the nature and extent of their activities, we performed a time—motion study where on-call residents' activities were directly observed for 106 nights. Because no comprehensive study has addressed the activities of pediatrics housestaff, both internal medicine and pediatrics residents were included in the study.
This time—motion study was conducted between the hours of 7 PM and 7 AM in June and July 1997 at a large, nonprofit, urban academic medical center with more than 700 beds. The hospital provides primary care for the metropolitan Richmond area and is a major tertiary referral center for all of Virginia and surrounding states. The principal thirdparty payers for inpatients are Medicare and Medicaid, accounting for 44% of inpatient charges during the study period. At the time of the study, a minority of inpatients (13%) were members of a managed care organization and 22% of all inpatients had no third-party payer.
The participants in this study were second- and third-year residents staffing the general internal medicine and pediatrics inpatient wards. Interns were not included because new interns arrive in July and faculty were concerned that observers could complicate this stressful time for these inexperienced housestaff. The study was confined to the general wards. On-call residents did not follow patients in intensive care units. In both programs, a single resident was responsible for night-call duties, which include supervising interns and students, performing history and physical examinations on new patients, and managing medical problems that might arise in existing patients. Residents entered the hospital in the morning preceding night call and left the afternoon following night call. On the day following night call, the resident took no new admissions.
Four medical students acted as observers for this study. The relevant literature, research purpose, study design, and observation and documentation techniques were reviewed with each observer. All observers had prior clinical training, including experience with hospitalized patients and research. Each observer was monitored during the first encounter to ensure that activities were being categorized appropriately. The observers were available for six of every seven days. The day off was rotated so that each weekday was equally represented.
The study was limited to evaluating the time residents spent on each activity; there was no attempt to measure the quality or effectiveness of their work. The observers recorded residents' time spent in activities in five categories: telephone communication; patients' history and physical examinations; procedures; case review and documentation, which included discussing the case with the team, using a computer, and chart review and documentation; and basic or non—patient-related activities, such as sleeping and eating. Observations were recorded in a log that had been patterned after studies from the literature1–8 and revised with input from faculty, house-staff, and investigators. Minor adjustments were made to the data-collection instrument during a one-night pilot study. The observers were instructed to follow the residents everywhere with two exceptions: the restroom and the on-call bedroom when the resident was asleep. The observer was stationed near the call room and was available by beeper. If housestaff left the on-call bedroom, they were asked to notify the observer to go with them. Because the observers did not directly watch housestaff in the on-call bedroom, it was not possible to identify whether the resident was sleeping, resting, snacking, or reading. Therefore, the “basic” category was further divided into eating, resting, sleeping, and waiting as well as reading, walking around, and informal conversation. The observers were instructed not to interrupt the activities of the housestaff and to hold questions until natural breaks in activities, such as when traveling to different locations within the hospital.
The purpose of the study was explained to each resident. It was emphasized that the study, was not intended to evaluate individuals' performances or to justify bed allocations. Residents' names did not appear on any forms or datacollection instruments. Verbal consent was obtained from all the residents in the study. Only one pediatrics resident refused to participate.
The mean time and standard deviation were computed for each activity. The nonparametric Mann—Whitney U test was used to compare the numbers of minutes per night the internal medicine and pediatrics residents spent on each activity. For internal medicine housestaff only, the chi-square test was used to compare the amount of time spent on each activity in our study with the results of the 1988 study by Lurie et al. of internal medicine housestaff in a university hospital.1 No similar reference study was available to compare activities for general pediatrics housestaff. All p values were two-tailed.
Comparison of Internal Medicine and Pediatrics Residents' Activities
Sixteen housestaff (eight internal medicine, eight pediatrics) were observed for 106 nights, 54 for internal medicine and 52 for pediatrics, during the months of June and July 1997. In rank order, the residents in internal medicine and pediatrics appeared to spend their time in similar activities (see Table 1). The only statistically significant difference in the amounts of time spent on activities between internal medicine and pediatrics residents was that the pediatricians spent more time doing procedures. A small amount of all residents' time was spent on “other” activities, including searching for charts, bed management, and x-ray review.
Case review and documentation was the second most time-consuming activity for residents. This category was composed of discussion of the case with the intern and other team members, using the computer to enter orders and review laboratory results, and chart review and documentation (see Table 1). Importantly, chart review and documentation accounted for the majority of time spent within this category, more than 2.5 hours required per night for each internal medicine resident and 2.1 hours per night for each pediatrics resident. Residents in the two programs spent similar amounts of time reviewing cases: a combined mean of 55 minutes reviewing cases with an intern, seven minutes reviewing cases with an attending, and 25 minutes reviewing cases with medical students or other team members.
Residents' history and physical examination activities ranked third for both programs (see Table 1). Internal medicine residents spent a mean of 71 minutes per night with new patients and 17 minutes with existing patients. Pediatrics residents spent a mean of 51 minutes per night with new patients and 25 minutes with existing patients. During the study period, the internal medicine residents admitted an average of 3.6 patients between 7 PM and 7 AM and spent an average of 20 minutes with each patient. The pediatrics residents admitted an average of 3.1 patients per night. Residents did not admit more than ten patients in any 24-hour period. Pediatrics residents spent an average of 16.5 minutes with each new patient, while internal medicine residents spent 19.7 minutes.
Overall, residents spent 37 fewer minutes on “down-time” activities, such as eating, resting, and sleeping, in July than they did in June. Also, in July residents spent 30 more minutes per night on case review and documentation than did residents in June.
Comparison of 1988 and 1997 Data for Internal Medicine Housestaff
We compared the activities of the internal medicine residents in our study with those from a 1988 study by Lurie et al.1 of internal medicine residents at a university hospital. Because the two studies were performed at different institutions and different times, the comparison is useful to establish a framework, but it cannot be relied upon for defining precise trends. The most striking difference between the two studies was in the amounts of time spent in case review and documentation. Our participants spent 38% of their time on this activity, while the residents in the 1988 study spent only 14% of their time in this activity (p < 0.01).
Internal medicine residents admitted an average of 3.6 new patients per night in 1997 compared with three patients per night in the 1988 study. The average amount of time spent per new patient's history and physical examination was 19.7 minutes in our study, compared with 15 minutes in the 1988 study.
The internal medicine residents in our study spent less time on basic activities such as eating or sleeping than did their counterparts in the 1988 study (p < 0.01). In addition, our residents spent less time on “other” activities. Importantly, the 1988 study found that much of this “other” time was spent searching for charts or transporting patients and specimens. In our study, the majority of “other” time was spent reviewing radiographs and in bed management.
Both internal medicine and pediatrics residents spent a major portion of their night call on case review and documentation. In some respects, housestaff duties may reflect the overall trend in medicine towards increased documentation and paperwork and the increased documentation itself may lead to larger charts, contributing to the long review time.
There is no national standard against which to compare our data. In fact, ours is the only large-scale study of pediatrics residents' night call. The last major large-scale observational study of internal medicine residents' night call was reported over a decade ago.1 Like that study, we found that the predominant activity for internal medicine housestaff consisted of “basic” functions (sleeping, resting, chatting, eating, and reading). Interesting differences between the results of the two studies were also found. Compared with the 1988 study, the internal medicine residents in our study spent more time on case review and documentation. There was a significant decrease in “other” work in our study; specifically, the residents in our study spent very little time transporting patients and blood work or looking for charts.1,4 Because the two studies were carried out in different institutions at different times, the cause(s) of these differences cannot be confirmed. It is not known whether these differences represent trends in the field, differences between the two hospitals, a trend toward increased patient acuity that requires longer case review and documentation, or differences resulting from study design.
Previous studies have used self-report instruments or work-sampling methods.3–5,7 Our study involved more labor-intensive direct observation to maximize accuracy and reduce bias.3,7 Nonetheless, our study has some limitations. Although two programs were studied, both were located at a single institution. Other institutions might have fewer admissions per resident or use fully electronic medical records. However, most hospitals have not yet abandoned paper charts. All laboratory results in this study were available on computer and all orders were entered into the computer. On average, computer work took more than half an hour per night. In institutions that still use paper for these activities, the proportion of time residents spend with charts would be expected to be larger. On-call residents took admissions during the day and continued into the night. Approximately half of admissions occurred during the night. Therefore, some of the case review and documentation might have been work left over from the daytime admissions. Similarly, for patients admitted in the early morning hours, case review and documentation might be put off to the next day.
In summary, chart review and documentation were an enormous part of residents' activities during night call. Whether such paperwork truly contributes to better patient care or education is a matter open to serious debate and one that should be researched. A careful assessment of duplicate documentation by residents, interns, students, attending physicians, and ancillary personnel might help to reduce this burden. Technologic efforts to streamline documentation should also be explored. Certainly, program directors should be involved in any hospital-based project that requires additional resident documentation. Finally, program directors should consider the impact on residents' education when regulators increase documentation requirements.
1. Lurie N, Rank B, Parenti C, Woolley T, Snoke W. How do house officers spend their nights? A time study of internal medicine house staff on call. N Engl J Med. 1989;320:1673–7.
2. American College of Physicians. Working conditions and supervision for residents in internal medicine programs: recommendations. Ann Intern Med. 1989;110:657–63.
3. Guarisco S, Oddone E, Simel D. Time analysis of a general medicine service: results from a random work sampling study. J Gen Intern Med. 1994;9:272–7.
4. Hayward RSA, Rockwood K, Sheehan GJ, Bass EB. A phenomenology of scut. Ann Intern Med. 1991;115:372–6.
5. Harvey R, Jarrett PG, Peltekian KM. Patterns of paging medical interns during night calls at two teaching hospitals. Can Med Assoc J. 1994;151:307–11.
6. Buff DD, Shabti R. The night float system of resident on call: what do nurses think? J Gen Intern Med. 1995;10:400–2.
7. Wood VC, Markert RJ, McGlynn TJ. Internal medicine residents' perceptions of the balance between service and education in their night-call activities. Acad Med. 1993;68:640–2.
8. Rosenberg M, McNulty D. Beyond night float? The impact of call structure on internal medicine residents. J Gen Intern Med. 1995;10:95–8.
9. Gottlieb DJ, Peterson CA, Parenti CM, Lofgren RP. Effects of a night float system on house staff neuropsychologic function. J Gen Intern Med. 1993;8:146–8.
10. Yedidia MJ, Schwartz MD, Hirschkorn C, Lipkin M. Learners as teachers: the conflicting roles of medical residents. J Gen Intern Med. 1995;10:615–23.