Apprenticeships Ease the Transition to Independent Call

Radiologic Education

Apprenticeships Ease the Transition to Independent Call: An Evaluation of Anxiety and Confidence Among Junior Radiology Residents Andrew T. Trout, MD, Page I. Wang, MD, Richard H. Cohan, MD, Janet E. Bailey, MD, Shokoufeh Khalatbari, MS, Jamie D. Myles, PhD, N. Reed Dunnick, MD Rationale and Objectives: The aims of this study were to quantify resident anxiety when beginning independent call and to assess whether an apprenticeship experience (buddy call) can lessen anxiety and improve confidence. Materials and Methods: A prospective cohort comparison of two groups of radiology residents beginning independent call, one of which was provided with a buddy call experience, was performed. Anxiety and confidence were assessed using the Endler Multidimensional Anxiety Scales–State (EMAS-S), with total score, autonomic emotional, and cognitive worry components, and a five-point, Likert-type scale, respectively. Both groups were asked about the perceived value of a buddy call experience. Results: EMAS-S scores improved significantly over 5 days of call in both groups (control, n = 10, P = .0005; buddy call, n = 9, P = .0001), and image interpretation confidence correspondingly increased (control, P = .0004; buddy call, P = .003). Compared to the control group, autonomic emotional scores were significantly lower in the buddy call group on the first day of independent call (P = .040), and cognitive worry and total EMAS-S scores were significantly lower on day 5 (both P values = .03). Buddy call was independently associated with improved autonomic emotional and film interpretation confidence scores (both P values = .02). All members of the buddy call group indicated that the experience was very helpful in preparing for call. Conclusions: Beginning independent call is associated with high anxiety, and buddy call reduces that anxiety, beyond the effect of time alone. Residents who participated in buddy call found it helpful in preparing for independent call. These findings support the use of buddy call and tiered call structures as means to introduce junior residents to independent call. Key Words: Call; residents; anxiety; confidence. ªAUR, 2011

R

esidency is stressful. The impact of the stresses faced by residents has been well documented, with several studies showing high levels of depression and anxiety among residents and worsening mood over the course of residency (1–3). Resident stress has also been demonstrated to have negative physiologic effects and to have an adverse impact on job performance (4–7). One of the most stressful components of residency is overnight call. Trainees are expected to operate increasingly independently, and the limits of their knowledge and organizational skills are tested. Despite the high stress associated with overnight call, the call experience is a necessary and valuable part of residency

Acad Radiol 2011; 18:1186–1194 From the Department of Radiology, University of Michigan Health System, 1500 East Medical Center Drive, Ann Arbor, MI 48109 (A.T.T., P.I.W., R.H.C., J.E.B., N.R.D.); and the Michigan Institute for Clinical and Health Research, University of Michigan, Ann Arbor, Michigan (S.K., J.D.M.). Received February 18, 2011; accepted April 9, 2011. Statistical services were supported by grant UL1RR024986 from the Michigan Institute for Clinical and Health Research (Ann Arbor, MI). Address correspondence to: A.T.T. e-mail: [email protected] ªAUR, 2011 doi:10.1016/j.acra.2011.04.015

1186

training, as it allows residents to make independent decisions and to develop confidence in their skills. Anecdotally, senior residents report that the stresses of call diminish with time. Despite this natural acclimatization, some residency programs can also provide support for junior residents at the beginning of their call experience to reduce the initial stress. Clinical residencies may support junior residents by scheduling them for call with senior residents, knowing that the availability of a more experienced colleague for consultation can provide security for junior trainees. Similarly, some radiology residencies have adopted a model in which junior residents work alongside senior residents before they begin independent call (a ‘‘buddy call’’ system). Although a preliminary apprenticeship may ease the transition to independent call, it is resource intensive, because several residents are assigned to perform a task that might be accomplished by a single trainee. To our knowledge, there are no published data that demonstrate whether such apprenticeship experiences reduce resident stress during subsequent independent call. At our institution, all first-year radiology residents previously rotated through two 5-day (Monday to Friday) afternoon/evening call rotations (12 PM to 9 PM) in their first year of residency. On these rotations, they worked alone under

Academic Radiology, Vol 18, No 9, September 2011

indirect faculty supervision. First-year residents were required to provide preliminary interpretations of radiographs obtained in the emergency department, as well as occasional inpatient radiographs, and they were also required to perform fluoroscopic procedures. Given the Accreditation Council for Graduate Medical Education’s limitations on first-year residents’ taking unsupervised call, these rotations allowed first-year residents to have faculty-supervised exposure to on-call cases prior to beginning true independent night call in their second year. At our institution, second-year residents provide nighttime coverage for plain radiographs and fluoroscopic procedures from 9 PM to 8 AM, while third-year and fourth-year residents provide nighttime coverage for ultrasound and cross-sectional imaging studies. Although our program previously prepared first-year residents for their initial afternoon call by rotating them through a prescribed series of daytime rotations, including cardiothoracic, gastrointestinal, genitourinary, pediatric, and musculoskeletal imaging, these residents did not have any preliminary exposure to the on-call experience prior to their first day of the independent afternoon/evening call rotation. Anecdotally, the initial call experience was described as high-stress, as residents were required to transition from the full faculty support available during the day to providing semi-independent preliminary radiographic interpretations. With knowledge of the buddy call systems in place at other residencies, as well as of the graded responsibility in place in many clinical residencies, we set out to explore the potential value of such an arrangement in our program. The purpose of our study was to quantify the stresses experienced by residents beginning independent call and to determine the independent effects of time and a buddy call experience on this stress. Specifically, our goals were to measure the degree of stress perceived by radiology residents beginning call as it has historically occurred at our institution, to measure changes in stress over the first few days of this call, and to assess whether a buddy call experience lessens the degree of stress experienced by residents beginning independent call. MATERIALS AND METHODS Institutional review board approval was obtained for this prospective cohort study, and written informed consent was obtained from participating residents. First-year radiology residents in consecutive class years were invited to participate in this project. Residents who entered our program on July 1, 2008, served as the control group, beginning call as it had traditionally occurred at our institution (no buddy call). Residents who entered our program 1 year later, on July 1, 2009, served as the experimental (buddy call) group and were provided with a buddy call experience that involved two 4.5-hour rotations (4:30 PM to 9 PM) on consecutive days within the month prior to beginning independent call. The length of the buddy call experience was selected to be a minimal additional time commitment and to sum to the equivalent of a single inde-

APPRENTICESHIPS ALLEVIATE CALL ANXIETY

pendent afternoon/evening call shift (12 PM to 9 PM; 9 hours) to facilitate data analysis. During the buddy call rotations, first-year residents worked in parallel with experienced residents with indirect in-house faculty supervision. First-year residents were expected to participate in all clinical activities (interpretation of radiographs, conveying findings to referring clinicians, answering clinical questions, and reviewing cases with the staff radiologist during a subsequent sign-out session). The senior residents were instructed in advance that this was to be an orientation experience for the first-year residents and that they should seek to prepare them for what might be experienced during independent call. For the purposes of data analysis, the stressful nature of the call experience was subdivided into two components: anxiety and confidence. Anxiety was assessed using the Endler Multidimensional Anxiety Scales–State (EMAS-S) inventory (Western Psychological Services, Los Angeles, CA). This is a 20-item questionnaire that asks respondents to indicate on a five-point, Likert-type scale (ranging from ‘‘not at all’’ to ‘‘very much’’) their current experience relative to each item. The 20 items in the instrument are subdivided into 10 assessing the cognitive worry (CW) (self-evaluation) domain and 10 assessing the autonomic emotional (AE) (physiologic) domain of state anxiety. Examples of CW and AE elements include ‘‘feel uncertain’’ and ‘‘mouth feels dry,’’ respectively. The instrument generates scores for both the CW and AE components, as well as a total state anxiety score. Normative data are available for US adults, allowing for subsequent categorization of raw scores into population-based percentiles. Confidence in interpretive and procedural skills (ability to interpret plain radiographs or perform various fluoroscopic procedures, including lumbar puncture, joint aspiration, and gastrointestinal studies) was also assessed with a five-point, Likert-type scale ranging from ‘‘very unsure’’ to ‘‘very confident’’ (Appendix A). To assess the impact of stress on resident performance, self-reported interpretive accuracy was also recorded. Each resident was asked to report the total number of exams interpreted each shift and to tally all interpretive errors, subdividing errors into ‘‘minor changes in interpretation’’ (those that were not reported to the ordering physician but which may have been documented in the final radiology report) and ‘‘major changes in interpretation’’ (those that required notification of the referring physician because of a possible change in patient management) (Appendix B). Residents identified these errors in interpretation during their subsequent signout with the faculty radiologist. The anticipated subjective value of a buddy call experience for the control group and the actual subjective value of the buddy call group were assessed on the final day of the participants’ first independent call week by asking the following questions: for the control group, ‘‘In terms of preparing for call and decreasing anxiety around beginning call; would it be helpful to work two additional half shifts with an experienced resident prior to working alone?’’ and for the 1187

TROUT ET AL

experimental group, ‘‘In terms of preparing for call and decreasing anxiety around beginning call; was it helpful to work two buddy call shifts with an experienced resident prior to working alone?’’ A follow up question asked each group to estimate the degree of helpfulness of a buddy call experience (‘‘somewhat helpful,’’ ‘‘very helpful,’’ or ‘‘extremely helpful’’; Appendix A). All participating residents were assigned identification numbers to preserve anonymity, and numbered survey instruments were administered at the beginning of their respective shifts by one of the authors (A.T.T., P.I.W., R.H.C.), who left the room while the residents completed the instruments.

Academic Radiology, Vol 18, No 9, September 2011

RESULTS Demographics

Of the 22 residents invited to participate, one of 11 (9.1%) from the control group (male) and two of 11 (18%) from the experimental group (both male) declined. The remaining members of the control group thus consisted of 10 residents, two women (20%) and eight men, with a mean age of 30.1
0.9 years. The remaining members of the buddy call group consisted of nine residents, three women (33.3%) and six men, with a mean age of 29.2
0.8 years. Control Group

Control Group

Anxiety and confidence were assessed on days 1, 2, and 5 of the initial 5-day afternoon/evening call rotation (12 PM to 9 PM). The additional questions concerning the perceived value of a buddy call system were asked on day 5. Accuracy and the total number of studies interpreted were assessed on days 1 and 2 of the independent call rotation. Buddy Call Group

Anxiety and confidence were assessed at the beginning of the first buddy call shift as well as on days 1, 2, and 5 of the resident’s subsequent independent call rotation. The additional questions concerning the actual value of a buddy call system were asked on day 5 of the independent evening call rotation. Accuracy and the total number of studies interpreted were assessed over the course of both buddy call shifts and on days 1 and 2 of the evening call rotation. Statistical Analysis

Total EMAS-S score, CW and AE component scores, and skills confidence (plain-film interpretation, fluoroscopic procedures) were analyzed as outcome measures. Continuous variables were summarized using means and standard deviations and categorical variables as counts and percentages. Wilcoxon’s rank-sum test was used to compare the various scores (outcome measures) between the two cohorts (control and buddy call) for each time point. Repeated-measures analysis was performed to test the time effect on each outcome measure per cohort, and multivariate repeated-measures analysis was used to test the effect of time and buddy call together on each outcome measure. Compound symmetry was used as covariance structure in the above repeatedmeasures analyses. Residual plots were examined to test the fit of the models. Post hoc analyses were performed using Scheffe ’s test in the comparison of scores among different days. To compare the scores between the two cohorts on their first exposure to call, a two-sample t test was used. For all statistical analyses, P values < .05 were considered significant. Analysis was performed using SAS version 9.2 (SAS Institute Inc, Cary, NC). 1188

Mean total anxiety scores (EMAS-S) declined from 48.5 on day 1 of the 5-day evening call rotation to 29.3 on day 5 (Table 1). Relative to normative data for US adults, these scores represent anxiety at the 91st percentile on day 1 declining to the 65th percentile by day 5 (Table 1). This decrease in scores over the 5 days of the rotation was statistically significant (P < .001). The change in scores between days 1 and 2 trended toward significance (P = .06). There was no significant difference in scores between days 2 and 5. For the component CW and AE scores, the decreases in scores between days 1 and 5 were also statistically significant for each (P = .0001 and P = .002, respectively). CW scores also decreased significantly between days 2 and 5 (P = .02). Changes in AE scores between days 1 and 2 trended toward, but did not reach, statistical significance (P = .07). The other changes were not significant. Control group confidence in the interpretation of plain radiographs increased significantly from a mean of 2.5 on day 1 to 3.5 on day 5 (P = .0004; Table 2). Neither of the incremental changes (between days 1 and 2 or days 2 and 5) was statistically significant. There were increases in confidence in performing fluoroscopically guided lumbar punctures, joint aspirations, and gastrointestinal studies, but none of these reached statistical significance. Residents in the control group self-reported high accuracy in their interpretation of on-call examinations; however, the small number of errors precludes statistical analysis of these data due to limited power. Minor interpretive errors occurred at a mean rate of 7.6% on day 1 and 8.8% on day 2. The reported mean rate of major errors in interpretation was 1.1% on day 1 and 1.8% on day 2. Buddy Call Group

The mean total anxiety score during the buddy call shift was 34.9. This increased slightly to 36.3 on the first day of independent call, a change that was not statistically significantly different. Total scores decreased to 23.7 by day 5 of independent evening call (Table 1), which represents a statistically significant improvement from both the buddy call shift (P < .01) and the first independent shift (P = .001). Intervening changes in scores were not statistically significant. Total

Academic Radiology, Vol 18, No 9, September 2011

APPRENTICESHIPS ALLEVIATE CALL ANXIETY

TABLE 1. Anxiety Scores for the Control (No Buddy Call) and Buddy Call Groups as Assessed by the Endler Multidimensional Anxiety Scales–State Anxiety Domain Control group (n = 10)

Call Day Buddy Call

Day 1

Day 2

Day 5

23.6 17.0–30.2 88

17.4 12.8–22.0 69

13.2 10.2–16.2 52

24.9 19.4–30.4 91

21.1 16.3–25.9 82

16.1 13.4–18.8 71

48.5 36.8–60.2 91

38.5 29.4–47.6 79

29.3 24.4–34.2 65

15.7 10.7–20.6 62

15.6 11.2–19.9 69

13.0 10.3–15.7 54

10.7 10.0–11.3 32

19.2 14.0–24.5 76

20.8 14.9–26.7 79

16.2 12.2–20.3 66

13.0 8.9–17.1 43

34.9 25.1–44.7 73

36.3 26.5–46.2 77

29.2 23.2–35.2 62

23.7 19.1–28.2 39

Autonomic emotional Mean score 95% CI Percentile Cognitive worry Mean score 95% CI Percentile Total score Mean score 95% CI Percentile Buddy call group (n = 9) Autonomic emotional Mean score 95% CI Percentile Cognitive worry Mean score 95% CI Percentile Total score Mean score 95% CI Percentile CI, confidence interval.

EMAS-S anxiety scores for the buddy shift, day 1, and day 5 ere at the 73rd percentile, the 77th percentile, and the 39th percentile of US adults, respectively (Table 1). Changes in component CW and AE scores were also statistically significant over the 5-day evening call rotation (Table 1). Mean CW scores changed from 19.2 during the buddy call shift to 20.8 on day 1 and 13.0 on day 5 of the independent evening call shifts (P = .01 for buddy shift vs day 5, P = .001 for day 1 vs day 5). Mean AE scores changed from 15.7 during the buddy call shift to 15.6 on day 1 and 10.7 on day 5 of the independent evening call shifts (P = .03 for buddy call shift vs both day 1 and day 5). Confidence in plain-film interpretive skill increased from a mean of 2.4 during the buddy call shift to 3.2 on day 1 of independent evening call, with a further increase to 4.1 by day 5 (Table 2). The differences between film interpretation confidence during the buddy shift and days 1, 2, and 5 of the independent call week were all statistically significant (P = .01, P < .0001, and P < .0001, respectively). The change between days 1 and 5 of the independent evening call week was also statistically significant (P = .003). Confidence in performing fluoroscopically guided joint aspirations also increased significantly between the buddy call shift and days 2 and 5 of the independent rotation (P = .03 and P = .002, respectively) but did not change significantly between days 1

and 5 of the independent evening call week. Over the course of the independent call week, there were also absolute increases in confidence in performing lumbar punctures, joint aspirations, and gastrointestinal studies under fluoroscopy, but none of these changes reached statistical significance (Table 2). Self-reported interpretation accuracy by residents in the buddy call group was quite good; however, as with the control group, small numbers did not provide sufficient power to perform statistical analysis. The mean rate of minor interpretive errors was 7.3% during the buddy call shift and 9.0% and 7.3% on days 1 and 2 respectively. The self-reported rates of major interpretative errors were 1.2% during the buddy call shift and 0.6% and 0.4% during days 1 and 2 of independent evening call. Differences Between Groups

On day 1 of independent evening call, AE scores were significantly lower in the buddy call group than in the control group (P = .04), but CW and total EMAS-S scores did not differ significantly (Table 3, Figs 1–3). By day 2, there was no significant difference between the groups for any of the EMAS-S scores. Scores diverged again on day 5, with both CWand total EMAS-S scores significantly lower in the buddy call group (P = .03 for both). AE scores did not differ between 1189

TROUT ET AL

Academic Radiology, Vol 18, No 9, September 2011

TABLE 2. Confidence in Interpretive and Procedural Skills Confidence Control group (n = 10)

TABLE 3. Statistical Significance of Comparison of Control and Buddy Call Group Scores at Each Time Point

Call Day Buddy Call Day 1

Day 2

Statistical Significance (P)

Day 5 Measure

Plain radiographic interpretation Mean score 95% CI Fluoroscopic LP Mean score 95% CI Fluoroscopic joint aspiration Mean score 95% CI Fluoroscopic GI study Mean score 95% CI

2.5 3.0 3.5 2.0–3.0 2.4–3.6 3.0–4.0 2.4 2.5 2.5 1.4–3.4 1.6–3.4 1.5–3.5

2.8 3.0 2.9 1.9–3.7 2.2–3.8 2.2–3.6 2.9 3.5 3.5 2.2–3.6 3.0–4.0 2.9–4.1

Cognitive worry Autonomic emotional Total EMAS-S score Film confidence LP confidence Joint confidence GI confidence

Day 1 .2672 .0402y .1110 .0543 .8340 1.000 .2616

Day 2 .1693 .1346 .1094 .0848 .8359 .6141 .7618

Day 5 y

.0258 .1202 .0262y .0880 .8004 .1429 .6528

First Shift* .1121 .0469y .0632 .8679 .6680 .4193 .4277

EMAS-S, endler multidimensional anxiety scales–state; GI, gastrointestinal; LP, lumbar puncture. *First shift for each group: day 1 for the control group, buddy call shift for the experimental group. y Statistically significant difference.

Buddy call group (n = 9) Plain radiographic interpretation Mean score 95% CI Fluoroscopic LP Mean score 95% CI Fluoroscopic joint aspiration Mean score 95% CI Fluoroscopic GI study Mean score 95% CI

2.4 1.9–3.0

3.2 2.7–3.7

3.7 3.3–4.0

4.1 3.7–4.6

2.1 1.0–3.2

2.4 1.4–3.4

2.7 1.7–3.6

2.8 1.7–3.9

2.3 1.4–3.3

2.9 2.3–3.5

3.3 2.6–4.1

3.7 3.0–4.3

3.3 2.3–4.3

3.6 2.8–4.3

3.7 2.9–4.4

3.7 2.8–4.5

CI, confidence interval; GI, gastrointestinal; LP, lumbar puncture.

groups at this time point. There was no significant difference in confidence scores between groups at any time point (day 1, 2, or 5 of independent call; Table 3). When we compared both groups’ first exposure of any kind to call (the buddy call shift relative to day 1 of the control group’s independent evening call rotation), AE scores were significantly lower in the buddy call group (P = .047), and total scores trended toward being significantly lower (P = .06). CW and confidence scores were not significantly different between the groups (Table 3).

day 5, while plain-film interpretation confidence significantly increased between all three time points. The effect of buddy call on total EMAS-S scores trended toward, but did not reach, statistical significance (Table 4).

Time and Buddy Call Effects

Subjective Value of a Buddy Call Experience

In the multivariate analysis, time on call had a significant association with CW and AE scores, total EMAS-S scores, and plain-film interpretation and joint aspiration confidence, with anxiety decreasing and confidence increasing between days 1 and 5 (Table 4). Buddy call was associated with significantly lower AE scores and higher plain-film interpretation confidence (Table 4). Because of the effect of buddy call, AE scores were significantly lower between day 1 versus day 2 and day 1 versus

All members of the control group indicated that they thought a buddy call experience would have been useful in terms of preparing for independent call (even though they had had no such experience). The mean predicted value of such an experience assigned on a scale of 1 (‘‘somewhat helpful’’) to 3 (‘‘extremely helpful’’) was 1.7 (95% confidence interval, 1.2–2.2). After both buddy call and independent call, all members of the buddy call group responded that buddy call had been

1190

Figure 1. Mean cognitive worry scores for the control (no buddy call) and buddy call groups at each measured time point. Scores are raw scores derived from the Endler Multidimensional Anxiety Scales–State and are out of a maximum score of 50. Error bars define the standard error of the mean. *Statistically significant difference.

Academic Radiology, Vol 18, No 9, September 2011

Figure 2. Mean autonomic emotional scores for the control (no buddy call) and buddy call groups at each measured time point. Scores are raw scores derived from the Endler Multidimensional Anxiety Scales–State instrument and are out of a maximum score of 50. Error bars define the standard error of the mean. *Statistically significant difference.

Figure 3. Mean total Endler Multidimensional Anxiety Scales–State scores for the control (no buddy call) and buddy call groups at each measured time point. These scores represent the sum of the cognitive worry and autonomic emotional component scores and are out of a maximum score of 100. Error bars define the standard error of the mean. *Statistically significant difference.

helpful in preparing them for independent call. The mean value, on the same scale, that they assigned to the buddy call experience was 2.1 (95% confidence interval, 1.3–2.9).

DISCUSSION The on-call experience is a necessary part of residency training, as it allows residents to begin to make independent decisions and to develop confidence in their skills. Not

APPRENTICESHIPS ALLEVIATE CALL ANXIETY

surprisingly, initial on-call experiences produce high levels of stress in residents, who are often abruptly confronted with new responsibilities. Such stress can manifest itself in many ways, including physiologic responses (increased heart rate, blood pressure, fatigue) and emotional responses (anxiety, mood alterations) (6–10). These undesired responses can have an adverse impact on performance; Feddock et al (4) reported an 11% rate of self-identified omissions in patient care by oncall internal medicine interns, most of which were associated with high pressure (5). Our study has shown that radiology residents who take independent call without prior call exposure experience high levels of initial anxiety, with EMAS-S scores at the 91st percentile for age-matched controls. This finding is concordant with the observations of prior studies, which have demonstrated high levels of stress among on-call residents training in various disciplines. For example, using survey instruments (the State-Trait Anxiety Inventory and the Profile of Mood States) Berkoff and Rusin (10) showed that among first-year and second-year pediatric residents, those on call had higher anxiety levels than did a control group that was not on call, and those on call developed significantly worsening moods over the 24-hour call period. Mehler and Anderson (6) showed corresponding physiologic changes in internal medicine residents, who demonstrated significant increases in blood pressure and pulse rate while on call, and Tendulkar et al (7) showed that on-call surgery residents had significantly increased mean and maximum heart rates, as well as increased white blood cell counts. One might argue that no dedicated investment needs to be made to reduce the stress associated with call, as initial stress is inevitable and will simply resolve itself over time. Data to support this assertion are available in the literature, with Tendulkar et al (7) showing that although surgical residents at all levels of training manifested physiologic signs of stress (increased heart rates) while on call, there was an inverse relationship between the maximum heart rate of an individual and his or her number of years in residency, suggesting acclimatization over time. Similarly, Gordon et al (1) found that tension-anxiety scores in internal medicine interns decreased significantly between the beginning of internship and the end of the first month and remained lower throughout the year. Our data also support this phenomenon, as we noted significant decreases in resident anxiety in both groups over the course of the first week of call, with a corresponding significant increase in confidence. Other findings have suggested that reduction in stress and improvement in performance over time spent on call do not always occur. In a study by Gordon et al (1), although anxiety decreased over time among interns, anger-hostility and fatigue scores increased significantly during the year, suggesting that although anxiety resolved, emotional stressors persisted. Schneider and Phillips (2) saw a similar persistence of emotional stressors and documented relatively constant rates of anxiety and/or depression in 39 residents followed through their intern year. The investigators reflected that there was 1191

TROUT ET AL

Academic Radiology, Vol 18, No 9, September 2011

TABLE 4. Multivariate Analysis of Time and Buddy Call Effects on Anxiety and Confidence Scores Statistical Significance Measure Cognitive worry Autonomic emotional Total EMAS-S score Film confidence Joint confidence

Buddy Call Effect

Time Effect

Day 1 to 2

Day 2 to 5

Day 1 to 5

.14 .02* .057 .02* .37

<.0001* <.0001* <.0001* <.0001* .01*

.001* .02* .003* .007* .09

.001* .09 .01* .007* .70

<.0001* <.0001* <.0001* <.0001* .02*

EMAS-S, endler multidimensional anxiety scales–state. *Statistically significant difference.

surprisingly little adaptation to the stress of the job and suggested that the stresses of residency perhaps overwhelm the coping and support mechanisms of the residents. Given the conflicting data about the effect of time alone on the stresses experienced by residents, an argument can be made for seeking ways to mitigate the negative effects of this stress, such as practicing on-call tasks prior to taking call and the implementation of support systems (such as the use of a preliminary buddy call system or available attending physician coverage). Furthermore, the potential deleterious effects of even transient initial stress during call should be recognized. Several approaches to addressing on-call stresses have been described. Some investigators have shown that practicing the skills that will be needed in the on-call environment can improve confidence. Two studies evaluating the effect of a surgery skills course (‘‘surgery boot camp’’) on medical student confidence with surgical techniques showed improved confidence scores for various measures after the skills course (11,12). Towbin et al (13) also showed an improved sense of preparedness and decreased nervousness in radiology residents who were given access to a picture archiving and communication system simulator and sample cases prior to taking independent call. Another approach has been to provide tiered support systems on call. This has been accomplished in some clinical disciplines that structure their call in a manner so that junior residents are provided with on-call support from senior residents or faculty members. In the field of radiology, Joffe et al (14) looked at the effect of introducing overnight teleradiology coverage by faculty members and found that both preshift anxiety and during-shift stress were reduced among residents. To our knowledge, however, there are few other published data on the effects of differing call structures on resident anxiety and confidence. In our study, we have demonstrated that a buddy call experience for junior radiology residents has a time-independent, significant effect of reducing anxiety and increasing confidence among residents beginning call. Residents who were provided with a buddy call experience had significantly lower autonomic emotional (physiologic) anxiety scores during both their initial exposure to call (a buddy shift) and during their first independent call shift than did members of the control group. Importantly, 1192

this effect appears durable, with members of the buddy call group continuing to report significantly lower anxiety on the last day of their independent call rotation. This improvement related to buddy call likely represents a combination of both recognition by the resident of their own clinical skills as well as an effect of being oriented to the work flow and work environment. In addition to the objective value of a buddy call experience, there is support for the practice in the subjective data we collected as well. All members of the control group predicted that buddy call would be useful in terms of preparing for independent call, and the entire buddy call group felt that the experience was helpful, a belief that may justify the implementation of a buddy call system in and of itself. Although our study and others have demonstrated the value of buddy call experiences, such systems must be implemented in a manner that preserves the opportunity for junior residents to develop independent decision-making skills. This is particularly important as opportunities to develop such skills are increasingly threatened by limitations on unsupervised practice by junior radiology residents and a movement toward 24-hour faculty physician coverage. Balancing the importance of independent resident experience, patient safety, and the needs of the clinical services will be a continuing issue for radiology and other residencies. This study had several limitations. First, despite our relatively large radiology residency program, our sample size was limited by the number of residents enrolling each year in our program. These small numbers limit statistical power and may explain why some comparisons did not reach statistical significance. Second, the study design may have introduced bias. Cohorts were selected on the basis of a convenience sample, without measurement of the baseline anxiety levels of the two groups. Differences in baseline anxiety between our control and buddy call groups might have accounted for some of the findings of this study. It is worth noting, however, that the similarities in scores between the two groups at several time points, particularly midweek of independent call, suggests that significant baseline differences are unlikely. Third, given the lack of similar studies in the literature, the duration of the buddy call experience was determined by the study team on the basis of convenience for the participants

Academic Radiology, Vol 18, No 9, September 2011

rather than on evidence suggesting a minimum effective duration. It may be the case that the amount of time spent in apprenticeship was insufficient to result in significant effects. Fourth, data concerning the interpretive accuracy of our residents was self-reported, and the residents may not have been uniform or entirely accurate in their self-assessments. Finally, the results of this study may be unique to our institution. Although we believe that the conclusions can be generalized, the data are derived from our residents, who could be in some way different from residents in other radiology residencies and in other clinical disciplines. In summary, our prospective study confirms that there is a high degree of stress associated with beginning independent call for junior radiology residents and demonstrates the value of an apprenticeship or buddy call experience prior to taking independent call for reducing that stress. Although anxiety decreases and confidence increases fairly rapidly over time, buddy call has an additional and time-independent effect on these changes. Furthermore, buddy call and graded responsibility on call are valued by residents, who feel that such experiences are helpful in preparing them for call. We believe that our findings strongly support the use of a buddy call system as a means to introduce junior residents to independent call. As a result of this study, we have now formally incorporated such an experience into the radiology residency curriculum at our institution.

APPRENTICESHIPS ALLEVIATE CALL ANXIETY

REFERENCES 1. Gordon GH, Hubbell FA, Wyle FA, et al. Stress during internship: a prospective study of mood states. J Gen Intern Med 1986; 1:228–231. 2. Schneider SE, Phillips WM. Depression and anxiety in medical, surgical, and pediatric interns. Psychol Rep 1993; 72:1145–1146. 3. Zare SM, Galanko J, Behrns KE, et al. Psychological well-being of surgery residents before the 80-hour work week: a multiinstitutional study. J Am Coll Surg 2004; 198:633–640. 4. Feddock CA, Hoellein AR, Wilson JF, et al. Do pressure and fatigue influence resident job performance? Med Teach 2007; 29:495–497. 5. Goetz T, Preckel F, Zeidner M, et al. Big fish in big ponds: a multilevel analysis of test anxiety and achievement in special gifted classes. Anxiety Stress Coping 2008; 21:185–198. 6. Mehler PS, Anderson RJ. Mechanism of pressor response in medical house officers on call. Ann Intern Med 1987; 106:560–561. 7. Tendulkar AP, Victorino GP, Chong TJ, et al. Quantification of surgical resident stress ‘‘on call.’’ J Am Coll Surg 2005; 201:560–564. 8. Endler NS, Edwards JM, Vitelli R. Endler multidimensional anxiety scales manual. Los Angeles, CA: Western Psychological Services, 1991. 9. Orton DI, Gruzelier JH. Adverse changes in mood and cognitive performance of house officers after night duty. BMJ 1989; 298:21–23. 10. Berkoff K, Rusin W. Pediatric house staff’s psychological response to call duty. J Dev Behav Pediatr 1991; 12:6–10. 11. Esterl RM Jr, Henzi DL, Cohn SM. Senior medical student ‘‘boot camp’’: can result in increased self-confidence before starting surgery internships. Curr Surg 2006; 63:264–268. 12. Peyre SE, Peyre CG, Sullivan ME, et al. A surgical skills elective can improve student confidence prior to internship. J Surg Res 2006; 133: 11–15. 13. Towbin AJ, Paterson B, Chang PJ. A computer-based radiology simulator as a learning tool to help prepare first-year residents for being on call. Acad Radiol 2007; 14:1271–1283. 14. Joffe SA, Burak JS, Rackson M, et al. The effect of international teleradiology attending radiologist coverage on radiology residents’ perceptions of night call. J Am Coll Radiol 2006; 3:872–878.

1193

TROUT ET AL

APPENDIX A Additional Survey, Day 5 (Experimental Group)

How confident do you feel right now in your ability to accurately interpret any plain film you may encounter on call? 1 2 3 4 5 Very Unsure Somewhat Neutral Somewhat Very Confident Unsure Confident

Academic Radiology, Vol 18, No 9, September 2011

In terms of preparing for call and decreasing anxiety around beginning call was it helpful to work two ‘‘buddy call’’ shifts with an experienced resident prior to working alone? , Yes , No If yes, how helpful was this?

1 Somewhat Helpful

2 Very Helpful

3 Extremely Helpful

How confident do you feel right now in your ability to perform a lumbar puncture while on call? APPENDIX B 1 2 3 4 5 Very Unsure Somewhat Neutral Somewhat Very Confident Unsure Confident

How confident do you feel right now in your ability to perform a joint aspiration while on call? 1 2 3 4 5 Very Unsure Somewhat Neutral Somewhat Very Confident Unsure Confident

How confident do you feel right now in your ability to perform any requested GI study while on call? 1 2 3 4 5 Very Unsure Somewhat Neutral Somewhat Very Confident Unsure Confident

1194

Resident Accuracy Assessment

Please score the accuracy of your interpretation as you staff cases for the evening. These scores will remain confidential, will not be used in any way to grade performance and will not be used outside of this research project. DATE: __________ TOTAL EXAMS: __________ SIGNIFICANT CHANGES IN INTERPRETATION Please tally changes in the interpretation which required notifying the managing clinician due to a possible change in patient management. MINOR CHANGES IN INTERPRETATION Please tally all other changes that were made but were not significant enough to change patient management or require reporting to the referring physician.