Pierce C, Keniston A, Stickrath C. Frequency of Attending Physician–Led Discussion of Test-Ordering Principles During Teaching Rounds. JAMA Intern Med. 2016;176(2):261-262. doi:10.1001/jamainternmed.2015.6979
High-value care is essential for patients and sustainable health care. In 2012, the American Board of Internal Medicine launched its Choosing Wisely campaign to help physicians more thoughtfully consider diagnostic testing. The American College of Physicians published 5 questions physicians should ask before ordering tests.1 Because educators are encouraged to teach these principles to trainees, we sought to quantify how frequently attending physicians lead their discussion.2- 4
We obtained approval for the study from the University of Colorado multiple institutional review board. Each enrolled attending physician and all observed team members provided written consent to participation. Observers received compensation for their time. No study participant received compensation. We trained fourth-year medical students as nonparticipant observers of medicine teaching rounds. A single observer shadowed a rounding episode led by each enrolled attending physician. Observers used standardized definitions and a stopwatch to annotate specific attending physician teaching behaviors in real time. The observers transcribed these notes to an investigator-developed instrument tracking in binary whether, for each patient, attending physician–led discussion mapped to the following 5 American College of Physicians test-ordering principles5:
whether a diagnostic test other than a complete blood cell count or basic metabolic, hepatic, or coagulation panel was previously performed;
whether diagnostic test results would affect care;
whether a test result represented—or a study under consideration might produce—a false-positive result;
whether the patient would experience short-term harm if a test were not ordered; and
whether the team considered patient preferences toward a diagnostic study.
We recorded the duration of every encounter, team time at the bedside, duration of teaching by the attending physician, and whether a patient was new or already known to the attending physician. The observer standardization process consisted of having observers annotate events of a 20-minute videotaped encounter and classify events iteratively until findings were consistent across all observers.
We observed 17 different rounding episodes between December 6, 2013, and December 18, 2014, at Denver Health and the University of Colorado Anschutz, consisting of of 168 patient encounters and 17 rounding days (11 at a county hospital and 6 at a university hospital) and involving 16 unique attending physicians. Data analysis was conducted from January 6, 2015, to October 9, 2015, at Denver Health. Only 35 (20.8%) of the encounters involved attending physician–led discussion of any test-ordering principle. The Table reports that short-term harm was discussed least frequently (2.4%), and the tests’ effects on care were discussed most frequently (13.7%).
For new patients compared with those who were known, we noted a nonsignificant finding regarding frequent discussion about tests’ effects on care (P = .09). We found no correlation with time spent at the bedside. Total attending physician teaching time was positively correlated with discussion of any test-ordering principle (P < .001).
To our knowledge, this is the first publication documenting the frequency of discussion of American College of Physicians test-ordering principles by attending physicians on rounds. In our view, discussion of these principles offers potential benefits, including improved test-ordering appropriateness and more patient-centered care. Because ward rounds continue to serve as a key educational forum for trainees, our observation that only 1 in 5 encounters involves attending physician–led discussion of these principles is discouraging.
Limitations of our study include the low occurrence rate and small sample size, thereby limiting detection of between-group differences. Our observer standardization process could have been more robust. Because annotation occurred in real time and episodes were not recorded, our study may be underreporting actual event frequencies. We did not track whether individual principles were discussed more than once per encounter or whether residents led discussion about test appropriateness. The single-institution nature and singular discipline limit generalizability.
Our secondary analysis conflicts with findings of a previous study6 showing that outpatient physicians who spend more face time with patients order fewer radiographic studies. However, we tracked discussion rather than test ordering.
The suggestion that more frequent discussion occurs for new compared with known patients could arise from more frequent ordering of diagnostic tests during the initial 24 hours of hospitalization. Medical educators could consider standardizing discussion for all testing of newly admitted patients to optimize care and educational value. Larger studies are needed to explore further potential links between on-rounds teaching, trainee acquisition of high-value care principles, and patient outcomes.
Corresponding Author: Cason Pierce, MD, MA, Denver Health Medical Center, 660 Bannock St, MC4000, Denver, CO 80204 (firstname.lastname@example.org).
Published Online: December 21, 2015. doi:10.1001/jamainternmed.2015.6979.
Author Contributions: Dr Pierce had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Pierce, Stickrath.
Acquisition, analysis, or interpretation of data: Pierce, Stickrath.
Drafting of the manuscript: Pierce, Keniston.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Keniston.
Obtained funding: Stickrath.
Administrative, technical, or material support: Pierce, Stickrath.
Study supervision: Pierce, Stickrath.
Conflict of Interest Disclosures: None reported.
Funding/Support: Funding for conducting observations was provided by a grant through the Academy of Medical Educators at the University of Colorado Anschutz.
Role of the Funder/Sponsor: The University of Colorado Anschutz had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Previous Presentation: This article was presented in part as a poster at the Society for General Internal Medicine Meeting; April 22, 2015; Toronto, Ontario, Canada.
Additional Contributions: Daniel Ozzello, MD (University of Colorado Anschutz), served as a research assistant. He observed rounding encounters and collected primary data. He received monetary compensation for the time he spent observing rounding sessions.