Description of a Multi-faceted COVID-19 Pandemic Physician Workforce Plan at a Multi-site Academic Health System.

BACKGROUND: The evolving COVID-19 pandemic has and continues to present a threat to health system capacity. Rapidly expanding an existing acute care physician workforce is critical to pandemic response planning in large urban academic health systems. INTERVENTION: The Medical Emergency-Pandemic Operations Command (MEOC)-a multi-specialty team of physicians, operational leaders, and support staff within an academic Department of Medicine in Calgary, Canada-partnered with its provincial health system to rapidly develop a comprehensive, scalable pandemic physician workforce plan for non-ventilated inpatients with COVID-19 across multiple hospitals. The MEOC Pandemic Plan comprised seven components, each with unique structure and processes. METHODS: In this manuscript, we describe MEOC's Pandemic Plan that was designed and implemented from March to May 2020 and re-escalated in October 2020. We report on the plan's structure and process, early implementation outcomes, and unforeseen challenges. Data sources included MEOC documents, health system, public health, and physician engagement implementation data. KEY RESULTS: From March 5 to October 26, 2020, 427 patients were admitted to COVID-19 units in Calgary hospitals. In the initial implementation period (March-May 2020), MEOC communications reached over 2500 physicians, leading to 1446 physicians volunteering to provide care on COVID-19 units. Of these, 234 physicians signed up for hospital shifts, and 227 physicians received in-person personal protective equipment simulation training. Ninety-three physicians were deployed on COVID-19 units at four large acute care hospitals. The resurgence of cases in September 2020 has prompted re-escalation including re-activation of COVID-19 units. CONCLUSIONS: MEOC leveraged an academic health system partnership to rapidly design, implement, and refine a comprehensive, scalable COVID-19 acute care physician workforce plan whose components are readily applicable across jurisdictions or healthcare crises. This description may guide other institutions responding to COVID-19 and future health emergencies.

The cirrhosis care Alberta (CCAB) protocol: implementing an evidence-based best practice order set for the management of liver cirrhosis - a hybrid type I effectiveness-implementation trial.

BACKGROUND: Liver cirrhosis is a leading cause of morbidity, premature mortality and acute care utilization in patients with digestive disease. In the province of Alberta, hospital readmission rates for patients with cirrhosis are estimated at 44% at 90 days. For hospitalized patients, multiple care gaps exist, the most notable stemming from i) the lack of a structured approach to best practice care for cirrhosis complications, ii) the lack of a structured approach to broader health needs and iii) suboptimal preparation for transition of care into the community. Cirrhosis Care Alberta (CCAB) is a 4-year multi-component pragmatic trial which aims to address these gaps. The proposed intervention is initiated at the time of hospitalization through implementation of a clinical information system embedded electronic order set for delivering evidence-based best practices under real-world conditions. The overarching objective of the CCAB trial is to demonstrate effectiveness and implementation feasibility for use of the order set in routine patient care within eight hospital sites in Alberta. METHODS: A mixed methods hybrid type I effectiveness-implementation design will be used to evaluate the effectiveness of the order set intervention. The primary outcome is a reduction in 90-day cumulative length of stay. Implementation outcomes such as reach, adoption, fidelity and maintenance will also be evaluated alongside other patient and service outcomes such as readmission rates, quality of care and cost-effectiveness. This theory-based trial will be guided by Normalization Process Theory, Consolidated Framework on Implementation Research (CFIR) and the Reach-Effectiveness-Adoption-Implementation-Maintenance (RE-AIM) Framework. DISCUSSION: The CCAB project is unique in its breadth, both in the comprehensiveness of the multi-component order set and also for the breadth of its roll-out. Lessons learned will ultimately inform the feasibility and effectiveness of this approach in "real-world" conditions as well as adoption and adaptation of these best practices within the rest of Alberta, other provinces in Canada, and beyond. TRIAL REGISTRATION: ClinicalTrials.gov: NCT04149223, November 4, 2019.

Correction to: Internal Medicine Point of Care Ultrasound: Indicators It's Here to Stay.

This editorial, "Internal Medicine Point of Care Ultrasound: Indicators It's Here to Stay" (DOI: 10.1007/s11606-019-05268-0), was intended to accompany "Education Indicators for Internal Medicine Point-of-Care Ultrasound: a Consensus Report from the Canadian Internal Medicine Ultrasound (CIMUS) Group".

The attributes of an effective teacher differ between the classroom and the clinical setting.

Most training programs use learners' subjective ratings of their teachers as the primary measure of teaching effectiveness. In a recent study we found that preclinical medical students' ratings of classroom teachers were associated with perceived charisma and physical attractiveness of the teacher, but not intellect. Here we explored whether the relationship between these variables and teaching effectiveness ratings holds in the clinical setting. We asked 27 Internal Medicine residents to rate teaching effectiveness of ten teachers with whom they had worked on a clinical rotation, in addition to rating each teacher's clinical skills, physical attractiveness, and charisma. We used linear regression to study the association between these explanatory variables and teaching effectiveness ratings. We found no association between rating of physical attractiveness and teaching effectiveness. Clinical skill and charisma were independently associated with rating of teaching effectiveness (regression coefficients [95 % confidence interval] 0.73 [0.60, 0.85], p < 0.001 and 0.12 [0.01, 0.23], p = 0.03, respectively). The variables associated with effectiveness of classroom and clinical teachers differ, suggesting context specificity in teaching effectiveness ratings. Context specificity may be explained by differences in the exposure that learners have to teachers in the classroom versus clinical setting-so that raters in the clinical setting may base ratings upon observed behaviours rather than stereotype data. Alternatively, since subjective ratings of teaching effectiveness inevitably incorporate learners' context-specific needs, the attributes that make a teacher effective in one context may not meet the needs of learners in a different context.

Diagnosing technical competence in six bedside procedures: comparing checklists and a global rating scale in the assessment of resident performance.

PURPOSE: To compare procedure-specific checklists and a global rating scale in assessing technical competence. METHOD: Two trained raters used procedure-specific checklists and a global rating scale to independently evaluate 218 video-recorded performances of six bedside procedures of varying complexity for technical competence. The procedures were completed by 47 residents participating in a formative simulation-based objective structured clinical examination at the University of Calgary in 2011. Pass/fail (competent/not competent) decisions were based on an overall global assessment item on the global rating scale. Raters provided written comments on performances they deemed not competent. Checklist minimum passing levels were set using traditional standard-setting methods. RESULTS: For each procedure, the global rating scale demonstrated higher internal reliability and lower interrater reliability than the checklist. However, interrater reliability was almost perfect for decisions on competence using the overall global assessment (Kappa range: 0.84-1.00). Clinically significant procedural errors were most often cited as reasons for ratings of not competent. Using checklist scores to diagnose competence demonstrated acceptable discrimination: The area under the curve ranged from 0.84 (95% CI 0.72-0.97) to 0.93 (95% CI 0.82-1.00). Checklist minimum passing levels demonstrated high sensitivity but low specificity for diagnosing competence. CONCLUSIONS: Assessment using a global rating scale may be superior to assessment using a checklist for evaluation of technical competence. Traditional standard-setting methods may establish checklist cut scores with too-low specificity: High checklist scores did not rule out incompetence. The role of clinically significant errors in determining procedural competence should be further evaluated.

Think global, act local: medical students contextualize global health education.

BACKGROUND: There is considerable heterogeneity in the extent to which global health education is emphasized in undergraduate medical curricula. Here, we performed an exploratory analysis to test the hypothesis that exposure to global health education may influence the attitudes of medical students toward the treatment of local vulnerable patient populations. METHODS: All pre-clerkship students at an urban Canadian university were invited to attend a voluntary global health education session on challenges in treating human immunodeficiency virus (HIV) in the developing world. Those who attended as well as those who did not completed pre- and post-session surveys measuring willingness to treat patients with HIV and related attitudes. A repeated measure analysis of variance (ANOVA) was performed to assess the effect of the intervention on attitudes toward locally affected populations. RESULTS: A total of 201 (81.4%) and 143 (58.3%) students completed the pre- and post-session surveys, respectively. Students who scored their willingness to treat patients with HIV within highest 10% of the scale on the pre-session survey were excluded from the analysis to account for a ceiling effect. On repeated measure ANOVA, willingness to treat local patients with HIV increased significantly following the session (P < 0.01). Students intending to attend the session also reported a greater propensity to treat patients with HIV than those who did not (P = 0.03). DISCUSSION: In this exploratory study, we find that following exposure to a global health lecture on the challenges of HIV in the developing world, students possessed more favorable attitudes toward the treatment of marginalized local patient populations, a finding that may be exploited in undergraduate and continuing medical education.

Estimation of post-test probabilities by residents: Bayesian reasoning versus heuristics?

Although the process of diagnosing invariably begins with a heuristic, we encourage our learners to support their diagnoses by analytical cognitive processes, such as Bayesian reasoning, in an attempt to mitigate the effects of heuristics on diagnosing. There are, however, limited data on the use ± impact of Bayesian reasoning on the accuracy of disease probability estimates. In this study our objective was to explore whether Internal Medicine residents use a Bayesian process to estimate disease probabilities by comparing their disease probability estimates to literature-derived Bayesian post-test probabilities. We gave 35 Internal Medicine residents four clinical vignettes in the form of a referral letter and asked them to estimate the post-test probability of the target condition in each case. We then compared these to literature-derived probabilities. For each vignette the estimated probability was significantly different from the literature-derived probability. For the two cases with low literature-derived probability our participants significantly overestimated the probability of these target conditions being the correct diagnosis, whereas for the two cases with high literature-derived probability the estimated probability was significantly lower than the calculated value. Our results suggest that residents generate inaccurate post-test probability estimates. Possible explanations for this include ineffective application of Bayesian reasoning, attribute substitution whereby a complex cognitive task is replaced by an easier one (e.g., a heuristic), or systematic rater bias, such as central tendency bias. Further studies are needed to identify the reasons for inaccuracy of disease probability estimates and to explore ways of improving accuracy.

In-group bias in residency selection.

BACKGROUND: More than half of all Canadian medical graduates match to residency programs within the same university as their medical school. Here we describe two studies designed to explore whether there is partiality for internal applicants in the resident selection process. METHODS: We first performed an observational study in which we compared the ratings of 14 'internal' and 89 'external' applicants to the University of Calgary Internal Medicine Training Program by resident and faculty raters. Following this we then asked residents to rate anonymous application packages in which we manipulated applicants' affiliation to our training program. RESULTS: In our first study, we found that residents rated internal applicants significantly higher for both application packages (mean (SD)) rating for internal versus external applicants (4.86 (0.36) vs. 4.36 (0.57), d = 1.05, p = 0.002) and interviews (4.93 (0.27) vs. 4.36 (0.7), d = 1.07, p = 0.003). There was no difference in the faculty ratings of internal and external applicants. In our second study, we found that residents rated applicants with an affiliation to our program - either attending the local medical school or having completed an elective - higher than applicants with no affiliation to our program. CONCLUSIONS: Our finding support in-group bias during resident selection, possibly due to the interdependent relationship between residents and students. Considering the career implications of residency matching, we feel that further studies are needed to identify and mitigate sources of bias in the residency application process.

The effect of a biofeedback-based stress management tool on physician stress: a randomized controlled clinical trial.

BACKGROUND: Physicians often experience work-related stress that may lead to personal harm and impaired professional performance. Biofeedback has been used to manage stress in various populations. OBJECTIVE: To determine whether a biofeedback-based stress management tool, consisting of rhythmic breathing, actively self-generated positive emotions and a portable biofeedback device, reduces physician stress. DESIGN: Randomized controlled trial measuring efficacy of a stress-reduction intervention over 28 days, with a 28-day open-label trial extension to assess effectiveness. SETTING: Urban tertiary care hospital. PARTICIPANTS: Forty staff physicians (23 men and 17 women) from various medical practices (1 from primary care, 30 from a medical specialty and 9 from a surgical specialty) were recruited by means of electronic mail, regular mail and posters placed in the physicians' lounge and throughout the hospital. INTERVENTION: Physicians in the intervention group were instructed to use a biofeedback-based stress management tool three times daily. Participants in both the control and intervention groups received twice-weekly support visits from the research team over 28 days, with the intervention group also receiving re-inforcement in the use of the stress management tool during these support visits. During the 28-day extension period, both the control and the intervention groups received the intervention, but without intensive support from the research team. MAIN OUTCOME MEASURE: Stress was measured with a scale developed to capture short-term changes in global perceptions of stress for physicians (maximum score 200). RESULTS: During the randomized controlled trial (days 0 to 28), the mean stress score declined significantly for the intervention group (change -14.7, standard deviation [SD] 23.8; p = 0.013) but not for the control group (change -2.2, SD 8.4; p = 0.30). The difference in mean score change between the groups was 12.5 (p = 0.048). The lower mean stress scores in the intervention group were maintained during the trial extension to day 56. The mean stress score for the control group changed significantly during the 28-day extension period (change -8.5, SD 7.6; p < 0.001). CONCLUSION: A biofeedback-based stress management tool may be a simple and effective stress-reduction strategy for physicians.

Glycemic control and use of the insulin sliding scale in hospitalized patients with diabetes.

This study assessed the extent of use of the subcutaneous insulin sliding scale (ISS) with hospitalized medical patients and examined the association between ISS use and glucose control. Despite some concerns about efficacy and suitability, the ISS is often used in the hospital studied. Researchers reviewed records of patients with a secondary diagnosis of diabetes who were admitted to the medical teaching unit of a tertiary care hospital. On day 1, 45.2% of patients were on an ISS alone. Patients on an ISS were more likely to experience hyperglycemia compared with patients on scheduled regimens, but they also had fewer hypoglycemic episodes.