Impact of reforms and work environment on resident time allocation in a Swiss internal medicine division: a time motion study with a before-and-after comparison

DOI: https://doi.org/https://doi.org/10.57187/5190

Introduction

Background

_d1uhyfglc7ryWorkplace organisation refers to how an institution defines employee schedules, interactions, meetings and responsibilities. While this structure is shaped at the institutional level, the actual content and intensity of work have also evolved substantially. Over the last decades, the job of general internal medicine residents has changed due to escalating polymorbidity and complexity of inpatients, growing volumes of clinical data and increased economic pressure on healthcare systems [1, 2].

In this evolving setting, previous studies have shown that workplace organisation is critical to ensuring high-quality care, sustaining healthcare delivery and fostering long-term professional commitment among physicians [3, 4, 5]. It is the framework for residents to achieve high quality and efficiency of provided care, added-value medical education and work satisfaction [6, 7].  

In this context, key concerns are clinical workflow and workload experienced by residents which include administrative burden, work continuity and time spent with patients. Indeed, use of information technologies has increased in healthcare and the role of the electronic medical record (EMR) is crucial. However, negative effects have also been described: increased time spent by physicians on administrative tasks and note writing, and reduced communication with patients [8, 9]. Fletcher et al. observed that the largest proportion of residents’ time (40%) was spent on clinical computer work [10]. EMR systems still fail to meet physicians’ expectations around harnessing, synthesising and presenting available data [2]. Moreover, on a more human level, teamwork characterises hospitals, but multiplicity of partners can lead to many interruptions, upsetting quality and work continuity. Westbrook et al. [11] showed that doctors multitasked 20% of the time and were interrupted every 21 minutes. Finally, increased time spent with patients improves patient satisfaction, patient education as well as health promotion activities, and reduces inappropriate prescribing and malpractice claims [12, 13]. Block et al. showed that residents spent a minority (12%) of their time in direct patient care [4, 14].

Conceptual framework

We adopted the Job Demands–Resources (JD-R) framework to interpret our findings. This model considers wellbeing and efficiency as the result of a balance between job demands (e.g. workload, task-switching) and job resources (e.g. tools, support, scheduling). In our context, reforms sought to reduce demands and increase resources, allowing us to link specific interventions to observed changes in residents’ activities [15].

Local context: a teaching hospital

In 2015, like many teaching hospitals, we sought to improve workplace organisation and logistical aspects of the clinical learning environment in response to growing time pressure on residents. We convened a multidisciplinary working group to identify priority areas for reform, using findings from a local time and motion study and a literature review. This group, composed of physicians, administrative staff and educators, worked iteratively over several months to analyse baseline data, benchmark practices and design targeted interventions. The process and methodological details of this working group are described in the appendix. These efforts led to the implementation of organisational changes with high potential for improving workplace efficiency and resident workflow.

Study objectives

Quantitative evidence on the effects of organisational reforms on resident activities is lacking. We aimed to assess how resident time allocation changed after organisational reforms with a primary focus on time spent on administrative tasks. As these took place in an evolving environment, we also had to assess how residents adapted to an increasing workload.

Methods

Study design

We assessed the impact of organisational and structural interventions by means of a before-and-after comparison. The “before” observation took place between May and July 2015, our baseline study. Its method and results were published by Wenger et al. [16]. Briefly, activities indirectly related to patients predominated, about half the workday was spent using a computer and residents switched from one task to another up to 15 times per hour [16, 17]. At the time, none of the reforms described below had been implemented. The “after” observation took place between May and July 2018 using the same method regarding training of observers, definition of activities and data collection.

Setting and participants

We conducted our study at Lausanne University Hospital, a 1500-bed tertiary care centre in Switzerland. Its division of internal medicine, in addition to its clinical care and research missions, hosts residents for one to two years, usually during the second part of their postgraduate training. Accredited as a Type A university training centre by the Swiss national authority for postgraduate medical education, it constitutes a mandatory rotation in the postgraduate curriculum for internal medicine. Training combines supervised patient care with structured educational activities, including daily bedside rounds, teaching sessions, multidisciplinary meetings and protected academic time.

The division of internal medicine is divided into eight wards. In both before and after periods, each ward was staffed with 1 senior physician, 1 chief resident and 2 to 4 residents. Each resident was responsible for 6 to 10 beds. Resident working hours are legally limited to 50 hours per week, including on-calls, and are usually scheduled between 42 and 50 hours.

There were day, evening and night shifts. Typical day shifts consisted in a daily patient round, supervision, training and new patient admissions. Medical staffing was reduced to 1 chief resident and 2 residents during evening and night shifts. Evening shifts – from 16:00 to 23:00 – mostly involved late patient admissions and emergencies. Night shifts were not evaluated.

Changes in the setting between the before and after periods occurred within a broader, system-wide trend towards shorter hospital stays and higher patient turnover. In 2015, the division of internal medicine employed 43 residents and operated an average of 196 beds (range: 191–228), admitting 6200 patients that year. By 2018, the division had 40 residents and operated 168 beds (162–178) with 5500 admissions. In line with legislative changes, official shift start and end times were slightly adjusted between 2015 and 2018, as detailed in figure 1.

Figure 1

Division work plan in 2015 and 2018. Time blocks are shown for a standard weekday daytime shift, excluding public holidays, and indicating scheduled periods. “Miscellaneous” includes tasks not predefined in the schedule. “Meetings & boards” in 2015 related to patient orientation meetings and multidisciplinary boards, which did not occur at fixed times and involved only the residents responsible for the patients discussed; therefore, they were displayed as separate scheduled blocks and integrated into the morning report in 2018. Compared to 2015, the 2018 schedule reflects key organisational changes: morning rounds were delayed to allow for extended preparation time, and teaching sessions were moved to the early afternoon to prioritise uninterrupted clinical work in the morning.

All residents employed by the division of internal medicine during the study period were eligible for inclusion. Residents were excluded if they could not be observed on the day of observation (ill or displaced elsewhere) or were not working weekday dayshifts on the observed inpatient wards (research position, intermediate care, night or weekend shift). Owing to routine resident turnover, none of the residents observed in 2015 were among those observed in 2018. Observed shifts were randomly selected and stratified by weekday to ensure representativeness. Two shifts per resident were recorded whenever possible.

Ethics approval and consent to participate

The Human Research Ethics Committee of Canton de Vaud certified that the study was exempt from human subject ethics review. All residents were informed of the study and provided written consent. No patient identifier or health information was recorded.

Consent for publication was part of the consent to participate.

Data collection procedures

To quantify activities, observers followed residents using a dedicated tablet application allowing real-time recording. The observers could document 22 mutually exclusive activities grouped into five categories (appendix table 1). They could select contexts: presence of a patient, presence of a colleague, use of a phone and use of a computer. After selecting an activity or a context, the application automatically recorded the starting time of the selection.

Observers were 22 paid undergraduate medical students (6 in 2015 and 16 in 2018). Training consisted in an e-learning, a teaching session, a 1-hour video of residents engaging in typical medical activities, an 8-hour dry run observing residents and recording, as well as a final session to resolve remaining issues and to ensure reproducibility. Two observers sequentially covered day shifts with a handoff after the first six hours. Observers had continuous access to activity definitions provided in appendix table 1 and to the research team, ensuring clarity if needed.

To characterise participants, we collected sex, age, country of graduation from medical school, year of postgraduate training and distance between home and hospital. To assess the workload of the clinical division, we collected overall mean length of stay (LOS), number of admissions per week and number of patients cared for, for each observed shift.

Organisational interventions

The working group ultimately designed nine interventions, all of which are described in detail in the appendix. The three main axes of reform were:

Firstly, we structurally strengthened administrative support by reallocating resources to hire dedicated medical secretaries, integrated into the medical teams. Before the reforms, one secretary served two wards (≈48 beds) from a separate office, mainly producing discharge summaries. After the reforms, one secretary per ward (≈24 beds) worked within the medical team with additional tasks, doubling total administrative support from 6.3 to 12.6 full-time equivalent (FTE). In parallel, we collaborated with the IT department to streamline the EMR interface by introducing structured templates and auto-filled fields, and continuing to improve the efficiency and consistency of discharge summary production [18, 19].

Secondly, we increased the frequency of case management rounds from twice per week to once daily by transforming the morning handoff into a longer interprofessional decision meeting including discussion about discharges and potential barriers [20–24].

Thirdly, we moved non-clinical activities such as the daily morning postgraduate training sessions (45 minutes long) to the afternoon, increased preparation time by postponing medical rounds by one hour and reinforced the importance of structured medical rounds [25–27].

From a JD-R perspective, our 2018 reforms sought to reduce demands – through the delegation of discharge summaries and other administrative tasks, and the restructuring of interprofessional rounds – while increasing resources via an enhanced EMR interface, improved scheduling and additional administrative support.

Outcomes

The primary outcome was chosen to reflect the administrative burden of residents and was defined as the total time spent on tasks considered delegable, i.e. activities that do not require core medical judgement and can be handled by non-physician staff. This included four activities defined in appendix table 1: patient administrative tasks (e.g. booking appointments), non-patient administrative tasks (e.g. professional e-mails), discharge summary writing and retrieving information from records (e.g. EMR, archives or by contacting providers). In contrast, tasks such as “writing in the medical record” were excluded from the primary outcome due to their reflective nature and contribution to clinical reasoning and coordination.

As secondary outcomes, we assessed day shifts for task-switching rate, defined as the number of times per hour that a resident switched from one activity to another, as described by Méan et al. [17]. We also measured mismatch rate, defined as an activity observed but not scheduled for the corresponding timeframe (e.g. a resident observed handling inpatient admissions at 13:23 during a scheduled postgraduate training session from 13:00 to 13:45). All possible activities were cross-referenced with the timetable prior to the 2015 analysis, independently reviewed by the research team and finalised by consensus. In 2018, two team members reviewed and validated the table before repeating the analysis. Lastly, we compared the effective duration of observed shifts.

Bias

We identified and mitigated many biases related to the before-and-after design. 1) Observation bias was reduced by instructing observers to avoid interacting with residents except to clarify an activity or context. We recruited medical students as observers, as they were more likely to understand and accurately code the activities performed. Observers had no hierarchical reporting line other than the research team. 2) Confounding bias from external workload factors could have arisen from variations in emergency room occupancy, bed availability in critical care, rehabilitation centres and nursing homes, as well as from weekday-related fluctuations. We mitigated these last biases by recording and adjusting the number of admissions and discharges per day, mean LOS, number of patients cared for, number of patient-shift-equivalents cared for and total patients in the division. 3) Seasonal confounding due to variations in occupancy rates and disease patterns was minimised by conducting observations during the same period of the year in both study phases.

Statistical analysis

The sample size calculation was based on other studies and on the results of the 2015 study [10, 11, 14, 16, 28]. We calculated the sample size necessary to detect a 20-minute reduction in the primary outcome (time dedicated to administrative tasks). The following information was used: average time spent on administrative tasks of 92 minutes per dayshift, with a standard error of 36 minutes; one-sided test, significance of 0.05 and power of 0.8. This led to a sample size of 64 in 2015 and 70 in 2018, corresponding to two observed shifts per resident.

In this paper, we present the day shift analysis only. The percentage of time devoted to a specific activity was calculated by dividing the time for that activity by the total shift duration. Statistical analyses were performed using Stata version 16 (Stata Corp, College Station, TX, USA). Residents’ characteristics are presented as either a mean (standard deviation [SD]) or median [interquartile range (IQR)] for continuous data, or as a count (percentage) for categorical data. Comparisons between the 2015 and 2018 observation periods were performed using mixed-effects linear regression models to account for repeated observations per resident. Study period (2015 vs 2018) was included as a fixed effect. Models included a random intercept for each resident, representing its vertical shift from the general mean. No random slopes were specified as there was no resident who participated in both surveys. Mixed models were adjusted for individual-level covariates (sex and postgraduate training year) and for division-level workload indicators (mean length of stay, weekly admissions and number of patients cared for per shift), as pre-specified in the study protocol. Results are expressed as means and their corresponding 95% confidence intervals of the time dedicated to each activity. Statistical significance was set at two-sided p <0.05.

The study was conducted in accordance with the registered protocol, and no protocol deviations occurred.

Results

Participants, shifts and division workload

Altogether, the activities of 75 residents were collected. Resident characteristics are presented in table 1. We recorded 66 shifts in 2015 and 76 shifts in 2018, summing 1478 hours of observation. When assessing division workload, mean admissions rose from 96.4 to 146.3 per week (+51.7%) between the observation periods, while mean length of stay decreased from 15.5 to 8.5 days (–45.2%). During daytime shifts, each resident was responsible for an average of 8.6 patients in 2015 and 8.5 in 2018 (table 2). Table 3 shows the distribution of activities during day shifts.

Outcome data and main results

Primary outcome

In the before-and-after comparison, mean time spent on administrative tasks during day shifts increased to 2018, from 92 to 139 minutes (95% CIs: 76–107 and 125–152) per day, p <0.001, table 3, figure 2). Of the four administrative tasks, “looking for information” increased the most: from 38 to 59 minutes per day (95% CIs: 31–46 and 52–65, p <0.001). Figure 3 presents the distribution of administrative tasks during the day. In 2018, these took place mainly between 08:30 and 10:00 and were scattered throughout the afternoon.

Figure 2

Mean time spent in five activity categories during weekday day shifts in 2015 and 2018. Mean time spent per shift on five activity categories (clinical, administrative, academic, personal, transit), expressed in minutes with percentage of total shift time in brackets. Data are unadjusted; see table 3 for adjusted estimates. Asterisks indicate highly significant changes in duration (p <0.001). Clinical: clinical tasks; admin: administrative tasks; academic: training and academic activities; pers: personal time; transit: transit from a task to another. Percentages are calculated over total shift duration.

Figure 3

Temporal distribution of administrative tasks during weekday day shifts in 2015 and 2018. The figure illustrates how administrative activities were distributed across the working day and how this distribution changed between study periods. This representation complements the quantitative results by highlighting peak administrative activity and potential targets for workflow interventions. Each line represents one shift, ordered from shorter to longer. Each dot corresponds to a 15-minute interval. The colour scale indicates the percentage of time spent on administrative tasks during each interval.

Adjustments made for division workload did not modify primary outcome results (table 3).

Secondary outcomes

_4bok3kb790q6Between 2015 and 2018, mean task-switching rate increased significantly from 15 to 20 switches per hour (95% CIs: 14–17 and 19–21, p <0.001). Switches occurred more frequently during clinical tasks (+25%, p <0.001) but less during personal time, dedicated to residents’ needs (–42%, p = 0.001). Task-switching rate was lower during the newly introduced morning meeting (08:00–08:30), medical rounds (10:00–11:00), lunchtime and teaching rounds (12:30–13:30), but higher during admissions and office work (after 14:00, appendix figure 1).

The mean mismatch rate decreased overall from 38.8% (4.4h/11.5h; 95% CI: 36.1–41.4%) to 31.7% (3.3h/10.2h; 95% CI: 29.4–34.0%, p <0.001). Clinical tasks produced less mismatch (–8.5%, appendix table 2). Personal time created more mismatch in 2018 (+14.8%, p <0.001). Looking at the weekly mismatch distribution, hotspots remained just before lunchtime, at noon and during the nurse desk round at 16:00 (appendix figure 2). Less restrictive periods at the beginning of the day freed residents to resolve most urgent issues before ward wounds.

Mean effective day shift duration decreased from 11h38m to 10h45m (95% CIs: 11h15m–12h01m and 10h30m–10h59m, p <0.001). Simultaneously, personal time increased from 32 to 63 minutes (95% CIs: 22–44 vs 53–72, p <0.001).

Adjustments made for division workload did not modify secondary outcome results.

Contexts

On average, residents were less exposed to patients in 2018 (113 to 92 minutes per day shift, 95% CIs: 101–126 and 80–103, p = 0.011). When expressed as a percentage of the total shift time, the difference was not statistically significant (table 4). Patients were seen mainly during medical rounds and between 14:00 and 17:00, with a similar temporal distribution between 2015 and 2018.

The mean time spent using a computer decreased significantly from 327 to 290 minutes per day shift (95% CIs: 306–348 and 272–308, p = 0.009, table 4). This reduction was particularly seen during patient rounds: percentage of time spent with a computer decreased from 51% (61/118 minutes in 2015) to 37% (35/93 minutes in 2018) (95% CIs: 45–56% and 32–42%, p <0.001), while percentage of time spent with patients remained identical.

Discussion

Key results

Totalling almost 1500 hours of observation, our study shows that administrative time (+49 min per day) and task-switching (+5 per hour) increased between 2015 and 2018. Conversely, the mean dayshift duration significantly decreased (–53 minutes) while personal time increased (+29 min). These changes occurred in the context of a substantial reduction of mean LOS (–45%) and a rise in weekly admissions (+52%). The central question is whether these results reflect the impact of our organisational reforms or are primarily driven by larger system-level pressures.

Interpretation of results in light of reforms

The increases in administrative time and task-switching were unexpected, thus implying complex underlying mechanisms of reforms, and requiring challenging interpretation. Of note, Schuurman et al. [29] found a similar percentage of administrative tasks during a shift. We suggest four effects of reforms that could explain an increase instead of a decrease of administrative time:

1) Delegation to medical secretaries aimed to free residents from non-medical-added-value tasks. However, the quality and completeness of administrative outputs (e.g. discharge summaries) possibly increased thus not reducing residents’ involvement in these processes.

2) Rescheduling rounds and teaching sessions may have allowed better medical round preparation but also created space that was filled with administrative work, possibly improving quality without decreasing quantity.

3) Promoting longer uninterrupted work periods was intended to improve efficiency. Yet task-switching increased (from 15 to 20 per hour, +33%), possibly reflecting the need to juggle more activities in a shorter shift within a busier division.

4) Improvement of the EMR by involving senior physicians in IT development aimed to streamline information retrieval. Nevertheless, time spent searching for information increased by 57% and became the main pre-round activity, possibly due to the rising volume of medical data and faster patient turnover [30].

Interpreted through the JD-R model, our reforms partially rebalanced demands and resources. Added resources reduced some administrative and cognitive load, but new or persistent demands (e.g. patient complexity) may have offset benefits. JD-R highlights that gains in wellbeing and efficiency require resources sufficient to counterbalance evolving demands.

Weight of context and systemic pressures

Between 2015 and 2018, the division faced a halving of mean length of stay and an increase in weekly admissions, consistent with a system-wide trend towards shorter hospitalisations and higher patient turnover. This structural shift inevitably increased the number of admissions and discharges per resident, each requiring substantial clinical and administrative work. However, the available data do not allow us to verify this hypothesis. These workload indicators (mean LOS, weekly admissions, patients per shift) were included as adjustment variables in mixed models. Adjustments did not materially change the results, suggesting that the impact of systemic pressures remained pervasive. Possible explanations include variability in how administrative tasks are distributed among residents, the contribution of administrative work unrelated to patient turnover, partial compensation from reforms, collinearity between workload variables and limited power to detect marginal effects of turnover.

This finding highlights a key message of our study: organisational reforms alone, particularly those limited to time reallocation, may be insufficient to counteract the impact of large-scale systemic changes, even if this observation may partly reflect the impact of our organisational reforms, particularly those aimed at streamlining discharge processes.

Other observed effects beyond administrative work

Despite the lack of measurable reduction in administrative time, our reforms coincided with several other changes that may reflect improvements in workflow and working conditions:

• Shorter shifts and reduced mismatch rate suggest better alignment between scheduled and actual activities, even in the face of increased admissions and higher turnover. This may indicate improved planning, smoother handovers and a more predictable daily structure, which are important for both efficiency and resident satisfaction.
• Increased personal time within shifts points towards a healthier work–rest balance, with residents more frequently able to take breaks without sacrificing clinical duties. Such pauses may help maintain concentration, reduce fatigue and promote resilience over the course of the day.
• A stable proportion of time with patients (as a percentage of total shift time) suggests that direct clinical contact was preserved despite a reduction in absolute minutes. This implies that, even under increased workload pressures, residents maintained their focus on patient-facing activities, which is critical for care quality and training. Patients continued to be seen mainly during morning rounds and mid-afternoon in both study periods. We hypothesise that planning medical rounds later in the morning may have improved preparation and the quality of patient contact, but this would require confirmation in a dedicated study.

In addition, time spent using computers decreased in absolute terms despite ongoing healthcare digitalisation, matching other studies such as Wieler et al. [31]. This finding could reflect multiple factors: improved usability of digital tools, better digital proficiency among residents, increased efficiency in data entry and retrieval, and a broader distribution of EMR-related tasks to other members of the care team.

Importance for future management

Our results challenge the idea that targeted organisational changes within a residency programme can, by themselves, substantially reduce administrative workload in a high-pressure health system. Even carefully designed reforms may have a limited measurable effect when systemic trends – such as increased patient turnover and reduced length of stay – exert a dominant influence on daily work patterns.

For residency programme directors and hospital managers, this means that structural changes to time allocation should not be implemented in isolation. They are more likely to yield tangible benefits when combined with broader interventions addressing the complexity of clinical workflows, the efficiency of interprofessional communication and the optimisation of administrative processes across the institution. Such measures may include streamlined discharge procedures, intelligent task allocation across the care team, integration of digital tools with improved usability and proactive workload monitoring.

In practice, aligning residency reforms with hospital-wide strategies is crucial to prevent local initiatives from being overshadowed by external pressures. This requires sustained collaboration between educational leaders, clinical managers and hospital administration, supported by robust monitoring of both quantitative (e.g. time-use metrics, workload indicators) and qualitative (e.g. resident satisfaction, perceived efficiency) outcomes. Without such coordinated efforts, the potential benefits of localised reforms risk remaining imperceptible in quantitative analyses, even when they may improve qualitative aspects of work.

Finally, in light of growing concerns about physician burn-out, exhaustion and attrition, our results highlight the importance of monitoring quantitative indicators and their sometimes unexpected evolution, while also adopting a multifaceted approach towards improving physicians’ working conditions and wellbeing, thereby aiming at enhancing recruitment and ensuring the future supply of medical professionals, in the inpatient sector and beyond.

Broader perspective

Measuring resident’s activities with a stopwatch relies on the scientific management put forward by Frederick Taylor [32]. However, time alone is an imperfect proxy for quality, as people are not machines. Residents need incentives other than wages to compensate for overtime. In a companion qualitative study, we interviewed our residents through focus groups and found that time management remains their main challenge [33]. Time constraints mean constant efficiency-seeking due to more admissions and pressure to reduce costs and lengths of stay.

Reforms to improve work conditions and adapt to the evolution of medicine will always be needed, but time management strategies alone are insufficient. Physicians also need a clear sense of purpose and professional identity to remain engaged. The rapid evolution of hospital medicine – marked by higher patient turnover, shorter lengths of stay and expanding administrative requirements – has reshaped the resident’s role. Preserving meaning in daily clinical work, fostering professional identity and maintaining strong engagement with the human aspects of care are essential. In high-pressure environments, these elements may require dedicated reflective practice, mentorship and collaborative models of care, integrated with structural reforms.

Strengths and limitations

Our study presents 1500 hours of observation, making it one of the biggest datasets in this field. This extensive amount of highly contextualised data allows precise quantitative responses. We collected two snapshots three years apart, at the same time of year and using exactly the same method, thus avoiding seasonality and methodological biases. Demographic characteristics were similar in both groups. Another strength is the use of a dedicated electronic tool to record activities in real time, providing granular, accurate and reproducible measurements.

Our study has several limitations. First, measurement-related limitations: despite standardised observer training in both study periods (same tutorial, video material and procedure), coding bias cannot be excluded. For instance, distinguishing between “writing in the medical record” and “looking for information” in the EMR may have been challenging. We did not assess the extent of implementation of each intervention, limiting interpretation of their actual impact. In addition, all interventions were evaluated together, preventing us from attributing effects to individual components.

Second, limitations affecting internal validity: in a before-and-after design, numerous confounding factors – such as patient complexity, medical student involvement or nurse staffing – could have influenced results. We estimated that the division’s workload increased substantially due to shorter length of stay and more weekly admissions; although adjustments for these variables did not alter our results, we cannot exclude an overwhelming effect of workload compared to reforms.

Third, limitations affecting external validity: we observed a limited number of residents in a single division of a university hospital, and our findings reflect only the activity of mainly third-year internal medicine residents during daytime shifts. Our results may not be generalisable to surgical disciplines, outpatient settings or more senior physicians. Night shifts were not observed, and changes in nocturnal activity could have affected administrative tasks during the following day.

Finally, limitations inherent to the time and motion design: this method does not capture quality of care, quality of work or resident wellbeing.

Conclusion

Organisational reforms in residency programmes can improve certain aspects of workflow, but our findings show that they may not be sufficient to offset the impact of broader systemic pressures such as increased patient turnover and reduced length of stay. Addressing administrative burden and task fragmentation in such contexts requires coordinated, institution-wide strategies that go beyond time reallocation, integrating workflow optimisation with initiatives that sustain professional meaning, identity and engagement among residents, aiming at improving physicians’ working conditions and wellbeing, and ensuring the future supply of medical professionals.

Data sharing statement

The study protocol is available from Dr Garnier (antoine.garnier[at]h-fr.ch). The statistical analysis code is available from Prof. Marques-Vidal (pedromanuel.marques-vidal[at]chuv.ch). The study dataset consists of individual-level, time-stamped observational and administrative data collected within a single Swiss hospital division. All data used for analysis were deidentified prior to analysis. However, due to the granular temporal structure of the data and the institutional context, there remains a residual risk of reidentification. For this reason, the full raw dataset cannot be deposited in an open public repository. Deidentified aggregated data (e.g. at shift or day level), data dictionaries and related documentation can be made available upon reasonable request to Dr Garnier. Data access is restricted to academic, non-commercial research purposes and is conditional on institutional approval and a signed data transfer agreement. Data will be made available for secondary analyses consistent with the objectives of the present study, without a predefined end date. The source code for the MEDAY application is available “as is” at https://github.com/agarnier00/MEDAY.

Acknowledgments

The authors sincerely thank SGAIM-Foundation and its members for providing funding support, all the student-observers, the observer team leader Ribal Bou Mjahed and Prof. Claudio Sartori.

The authors confirm contribution to the paper as follows: Study conception and design: AG, JC, PM-V, FB, MM, DG, MM, OL, PV, GW, VK. Data collection: AG, VK, FB, JC. Analysis and interpretation of results: AG, VK, PM-V, FB, JC. Draft manuscript preparation: AG, VK. All authors reviewed the results and approved the final version of the manuscript.

References

1. Alkureishi MA, Lee WW, Lyons M, Press VG, Imam S, Nkansah-Amankra A, et al. Impact of Electronic Medical Record Use on the Patient-Doctor Relationship and Communication: A Systematic Review. J Gen Intern Med. 2016 May;31(5):548–60. doi: https://doi.org/10.1007/s11606-015-3582-1 

2. Ammenwerth E, Spötl HP. The time needed for clinical documentation versus direct patient care. A work-sampling analysis of physicians’ activities. Methods Inf Med. 2009;48(1):84–91. doi: https://doi.org/10.3414/ME0569 

3. Bes I, Shoman Y, Al-Gobari M, Rousson V, Guseva Canu I. Organizational interventions and occupational burnout: a meta-analysis with focus on exhaustion. Int Arch Occup Environ Health. 2023 Nov;96(9):1211–23. doi: https://doi.org/10.1007/s00420-023-02009-z 

4. Block L, Habicht R, Wu AW, Desai SV, Wang K, Silva KN, et al. In the wake of the 2003 and 2011 duty hours regulations, how do internal medicine interns spend their time? J Gen Intern Med. 2013 Aug;28(8):1042–7. doi: https://doi.org/10.1007/s11606-013-2376-6 

5. Céline B, Matteo M, Michael S, Friedrich S, Vanessa K, David G, et al. Running against the clock: a qualitative study of internal medicine residents’ work experience. Swiss Med Wkly. 2022 Aug;152(3334):w30216. doi: https://doi.org/10.4414/SMW.2022.w30216 

6. Castioni J, Hagenbuch A, Tâche J, Cappai M, Jovanovic M, Sartori C. [Delegation of medico-administrative tasks : what do medical interns and secretaries think?]. Rev Med Suisse. 2017 Nov;13(584):2048–51. doi: https://doi.org/10.53738/REVMED.2017.13.584.2048 

7. Curley C, McEachern JE, Speroff T. A firm trial of interdisciplinary rounds on the inpatient medical wards: an intervention designed using continuous quality improvement. Med Care. 1998 Aug;36(8 Suppl):AS4–12. doi: https://doi.org/10.1097/00005650-199808001-00002 

8. Demerouti E, Bakker AB, Nachreiner F, Schaufeli WB. The job demands-resources model of burnout. J Appl Psychol. 2001 Jun;86(3):499–512. doi: https://doi.org/10.1037/0021-9010.86.3.499 

9.        Dugdale DC, Epstein R, Pantilat SZ: Time and the patient-physician relationship. J Gen Intern Med 1999, 14 Suppl 1(Suppl 1):S34-40. 

10. Erickson SM, Rockwern B, Koltov M, McLean RM, Medical P; Medical Practice and Quality Committee of the American College of Physicians. Putting Patients First by Reducing Administrative Tasks in Health Care: A Position Paper of the American College of Physicians. Ann Intern Med. 2017 May;166(9):659–61. doi: https://doi.org/10.7326/M16-2697 

11. Fletcher KE, Visotcky AM, Slagle JM, Tarima S, Weinger MB, Schapira MM. The composition of intern work while on call. J Gen Intern Med. 2012 Nov;27(11):1432–7. doi: https://doi.org/10.1007/s11606-012-2120-7 

12. Gachoud D, Monti M, Waeber G, Bonvin R, Vollenweider P, Waeber G, et al. [Conducting ward rounds: a balance between care and teaching]. Rev Med Suisse. 2013 Oct;9(404):2013–6. doi: https://doi.org/10.53738/REVMED.2013.9.404.2013 

13. Garnier A, Castioni J, Kraege V, Méan M, Vollenweider P, Waeber G. Open complaints and compliments about electronic medical records: internists’ top five. Prim Hosp Care. 2019;19(4):121–5. 

14. Geary S, Cale DD, Quinn B, Winchell J. Daily rapid rounds: decreasing length of stay and improving professional practice. J Nurs Adm. 2009 Jun;39(6):293–8. doi: https://doi.org/10.1097/NNA.0b013e3181a72ab8 

15.      Gironda Cuéllar SI, Ando V, Colleaux AH, Claivaz V, Gachoud D, Monti M: Implementation of a checklist to structure interprofessional daily ward rounds to improve adherence to standards of care. Swiss Med Wkly 2024, 154(Suppl 276). 

16. Judge TA, Thoresen CJ, Bono JE, Patton GK. The job satisfaction-job performance relationship: a qualitative and quantitative review. Psychol Bull. 2001 May;127(3):376–407. doi: https://doi.org/10.1037/0033-2909.127.3.376 

17. Kahn JS, Witteles RM, Mahaffey KW, Desai SA, Ozdalga E, Heidenreich PA. A 15-year review of the Stanford Internal Medicine Residency Program: predictors of resident satisfaction and dissatisfaction. Adv Med Educ Pract. 2017 Aug;8:559–66. doi: https://doi.org/10.2147/AMEP.S138467 

18. Méan M, Garnier A, Wenger N, Castioni J, Waeber G, Marques-Vidal P. Computer usage and task-switching during resident’s working day: disruptive or not? PLoS One. 2017 Feb;12(2):e0172878. doi: https://doi.org/10.1371/journal.pone.0172878 

19. Nardi R, Scanelli G, Corrao S, Iori I, Mathieu G, Cataldi Amatrian R. Co-morbidity does not reflect complexity in internal medicine patients. Eur J Intern Med. 2007 Sep;18(5):359–68. doi: https://doi.org/10.1016/j.ejim.2007.05.002 

20. Nørgaard K, Ringsted C, Dolmans D. Validation of a checklist to assess ward round performance in internal medicine. Med Educ. 2004 Jul;38(7):700–7. doi: https://doi.org/10.1111/j.1365-2929.2004.01840.x 

21. O’Mahony S, Mazur E, Charney P, Wang Y, Fine J. Use of multidisciplinary rounds to simultaneously improve quality outcomes, enhance resident education, and shorten length of stay. J Gen Intern Med. 2007 Aug;22(8):1073–9. doi: https://doi.org/10.1007/s11606-007-0225-1 

22. Poissant L, Pereira J, Tamblyn R, Kawasumi Y. The impact of electronic health records on time efficiency of physicians and nurses: a systematic review. J Am Med Inform Assoc. 2005;12(5):505–16. doi: https://doi.org/10.1197/jamia.M1700 

23. Prystajecky M, Lee T, Abonyi S, Perry R, Ward H. A case study of healthcare providers’ goals during interprofessional rounds. J Interprof Care. 2017 Jul;31(4):463–9. doi: https://doi.org/10.1080/13561820.2017.1306497 

24. Schuurman AR, Bos SA, de Wit K, de Graaf R, Wiersinga WJ. [A day in the life of a medical resident on the ward]. Ned Tijdschr Geneeskd. 2018;161:D2480. 

25. Taylor FW. The principles of scientific management. New York: Routledge; 2004. doi: https://doi.org/10.4324/9780203498569 

26. Wee KZ, Lai AY. Work Engagement and Patient Quality of Care: A Meta-Analysis and Systematic Review. Med Care Res Rev. 2022 Jun;79(3):345–58. doi: https://doi.org/10.1177/10775587211030388 

27. Wenger N, Méan M, Castioni J, Marques-Vidal P, Waeber G, Garnier A. Allocation of Internal Medicine Resident Time in a Swiss Hospital: A Time and Motion Study of Day and Evening Shifts. Ann Intern Med. 2017 Apr;166(8):579–86. doi: https://doi.org/10.7326/M16-2238 

28. Westbrook JI, Ampt A, Kearney L, Rob MI. All in a day’s work: an observational study to quantify how and with whom doctors on hospital wards spend their time. Med J Aust. 2008 May;188(9):506–9. doi: https://doi.org/10.5694/j.1326-5377.2008.tb01762.x 

29. Wieler J, Lehman E, Khalid M, Hennrikus E, Ross S. A day in the life of an internal medicine resident – a time study: what is changed from first to third year? Adv Med Educ Pract. 2020 Mar;11:253–8. doi: https://doi.org/10.2147/AMEP.S247974 

30. Wild D, Nawaz H, Chan W, Katz DL. Effects of interdisciplinary rounds on length of stay in a telemetry unit. J Public Health Manag Pract. 2004;10(1):63–9. doi: https://doi.org/10.1097/00124784-200401000-00011 

31. Wilson A, McDonald P, Hayes L, Cooney J. Health promotion in the general practice consultation: a minute makes a difference. BMJ. 1992 Jan;304(6821):227–30. doi: https://doi.org/10.1136/bmj.304.6821.227 

32. Yinusa A, Faezipour M. Optimizing Healthcare Delivery: A Model for Staffing, Patient Assignment, and Resource Allocation. Appl Syst Innov. 2023;6(5):78. doi: https://doi.org/10.3390/asi6050078 

33. Zulman DM, Shah NH, Verghese A. Evolutionary Pressures on the Electronic Health Record: caring for Complexity. JAMA. 2016 Sep;316(9):923–4. doi: https://doi.org/10.1001/jama.2016.9538 

Appendix

The appendix is available in the pdf version of the article at https://doi.org/10.57187/5190.