‘Choosing Wisely’ culture among Brazilian cardiologists

‘Choosing Wisely’ culture among Brazilian cardiologists Abstract Objective (i) To describe how aligned the ‘Choosing Wisely’ concept is with the medical culture among Brazilian cardiologists and (ii) to identify predictors for physicians’ preference for avoiding wasteful care. Design Cross-sectional study. Setting Brazilian Society of Cardiology. Participants Cardiologists who agree to fill a web questionary. Intervention A task force of 12 Brazilian cardiologists prepared a list of 13 ‘do not do’ recommendations, which were made available on the Brazilian Society of Cardiology website for affiliates to assign a supported score of 1 to 10 to each recommendation. Main Outcome Measurement Score average for supporting recommendations. Results Of 14 579 Brazilian cardiologists, 621 (4.3%) answered the questionnaire. The top recommendation was ‘do not perform routine percutaneous coronary intervention in asymptomatic individuals’ (mean score = 8.0 ± 2.9) while the one with the lowest support was ‘do not use an intra-aortic balloon pump in infarction with cardiogenic shock’ (5.8 ± 3.2). None of the 13 recommendations presented a mean grade >9 (strong support); 7 recommendations averaged 7–8 (moderate support) followed by 6 recommendations with an average of 5–7 (modest support). Multivariate analysis independently identified predictors of the score attributed to the top recommendation; being an interventionist and time since graduation were both negatively associated with support. Conclusions (i) The support of Brazilian cardiologists for the ‘Choosing Wisely’ concept is modest to moderate, and (ii) older generations and enthusiasm towards the procedure one performs may be factors against the ‘Choosing Wisely’ philosophy. patient-centred care, shared decision-making, appropriateness, under-use, over-use, patient safety, decision analysis Introduction In 2012, the ‘Choosing Wisely’ campaign was launched by the American Board of Internal Medicine (ABIM) Foundation to promote debate about overuse of tests, treatments, and procedures in the United States [1]. The initiative was triggered by the growing perception of overuse in the United States, which can be defined as ‘the provision of medical services for which the potential for harm exceeds the potential for benefit’ [2]. Thus, Choosing Wisely is focused on preventing harm by avoiding unnecessary tests and treatments. In the campaign, the ABIM Foundation invited medical societies to provide 5–10 ‘do not do’ recommendations regarding common overused procedures. It has been a successful initiative that spread to Canada [3] and several other countries [4], leading to the foundation of Choosing Wisely International. In this context, countries share experiences and discuss strategies for campaign implementation in different parts of the globe. In 2015, Brazil was represented for the first time in a Choosing Wisely international meeting, which was held in London. The Brazilian Society of Cardiology had a seat in the meeting, when campaign methodologies and Principals of Choosing Wisely were discussed: physician-led, patient-focused, evidence-based, multi-professional, and transparent [4]. Then, the Brazilian Society of Cardiology launched the first Choosing Wisely campaign in Brazil under the auspices of Choosing Wisely International. Although overuse has mainly been associated with high-income countries [5], it has been recognized as a growing problem in low-medium income countries. Despite the vast majority of Brazilians being covered by a health public system of relatively low resources, overuse is prevalent. A typical example is the Brazilian leadership in C-section labour, which is performed in 45% of cases, the majority with no specific indication [6]. Cardiology is a medical specialty with a myriad of tests and procedures for which utility varies according to the situation, being a potential source of overuse when indication is detached from an appropriate clinical context. Therefore, our group understood that the Choosing Wisely campaign in the Brazilian Society of Cardiology was an opportunity to explore cardiologists’ culture regarding the issue of avoiding low-value care. The present study aims to (i) describe the method utilized to establish the five recommendations of the Brazilian Society of Cardiology, (ii) describe how strong (or weak) the Choosing Wisely culture is among cardiologists in Brazil and (iii) identify predictors of physicians’ preference for recognizing and avoiding wasteful care. Methods The Brazilian Society of Cardiology has 14 463 members distributed throughout 27 states, predominantly São Paulo, Rio de Janeiro, Bahia, Minas Gerais and Rio Grande do Sul, which embrace 60% of the members. Initially, 12 experts were chosen from among clinical cardiologists, imaging specialists, electrophysiologists, and interventionists to constitute the Choosing Wisely Task Force. No regional criteria were utilized in defining this panel, which was selected based on knowledge and interest in evidence-based medicine, clinical epidemiology, over-diagnosis and overtreatment. The panel brainstormed and defines 13 ‘do not do’ recommendations. The recommendations had to be related to commonly recommended procedures with one of the following criteria: (i) treatments not proved to be beneficial; (ii) treatments proved not to be beneficial; (iii) diagnosis of questionable beneficial effect and potential harm, leaning towards over-diagnosis; and (iv) serial tests of no utility. The 13 recommendations and their rational are depicted in Table 1. Table 1. Recommendation, justifications and abbreviations Recommendation  Justification  Abbreviation  Mean score  Tests overuse   Do not order calcium score to follow the evolution of subclinical atherosclerosis.  Increase of calcium score over time has no prognostic role and this test offers substantial radiation [7].  Follow-up calcium score  7.9 ± 2.9   Do not order routine echocardiogram because of valvular dysfunction of mild degree.  Low yield of testing in detecting relevant changes [8].  Echo in mild valvular dysfunction  7.3 ± 3.0   Do not order non-invasive tests for obstructive coronary artery disease as a routine in pre-operatory evaluation of non-cardiac surgery in asymptomatic individuals with normal functional capacity.  The diagnosis of obstructive CAD and subsequent revascularization does not reduce cardiovascular risk related to surgery (over-diagnosis and potential overtreatment) [9].  Follow-up testing for previous revascularization  7.6 ± 3.0   Do not order non-invasive tests as screening of obstructive coronary artery disease (CAD) in asymptomatic individuals with normal ventricular function.  The diagnosis of stable coronary disease and subsequent revascularization does not improve outcomes (over-diagnosis and potential overtreatment).  CAD Screening  7.1 ± 3.1   Do not order non-invasive tests for obstructive coronary artery disease as a routine in asymptomatic individuals with a history of revascularization procedure.  The diagnosis of stable coronary disease and subsequent revascularization does not improve outcomes (over-diagnosis and potential overtreatment) [10].  Follow-up testing for previous revascularization  6.6 ± 3.0   Do not order carotid ultrasound for syncope investigation in the absence of neurological signs.  Syncope not related to neurologic event is not caused by carotid obstruction (over-diagnosis) [11].  Carotid US for syncope  5.9 ± 3.3  Invasive treatment overuse   Do not perform percutaneous coronary revascularization (PCI) in asymptomatic individuals.  PCI in stable CAD does not improve prognosis, only improves symptoms. In an asymptomatic patient, obviously, there is no symptom to be improved [10].  PCI in asymptomatic  8.0 ± 2.9   Do not perform PCI in a persistently occluded coronary artery after a late myocardial infarction.  PCI of persistently occluded coronary artery does not improve prognosis after myocardial infarction [12].  PCI in occluded artery  7.6 ± 2.9   Avoid routine pulmonary artery catheterization (PAC) in patients with acute heart failure.  PAC does not improve prognosis in heart failure [13].  PAC in acute HF  6.9 ± 3.0   Do not close persistent foramen oval (PFO) to prevent recurrence of cryptogenic stroke.  PFO closure does not reduce stroke recurrence [14].  PFO closure  5.9 ± 3.1   Do not use a routine intra-aortic balloon pump (IOBP) to treat cardiogenic shock due to myocardial infarction.  IOBP does not reduce mortality in patients with shock due to myocardial infarction [15].  IOBP in shock  5.8 ± 3.2  Medical treatment overuse   Do not prescribe beta-blockers before non-cardiac surgery as a means to prevent cardiovascular events related to the procedure.  Beta-blocker does not prevent cardiovascular events related to surgery [16].  Pre-op beta-blocker  7.0 ± 3.0   Do not use aspirin to prevent embolic stroke in patients with atrial fibrillation.  Aspirin does not prevent embolic stroke [17].  Aspirin for embolic stroke  6.9 ± 3.3  Recommendation  Justification  Abbreviation  Mean score  Tests overuse   Do not order calcium score to follow the evolution of subclinical atherosclerosis.  Increase of calcium score over time has no prognostic role and this test offers substantial radiation [7].  Follow-up calcium score  7.9 ± 2.9   Do not order routine echocardiogram because of valvular dysfunction of mild degree.  Low yield of testing in detecting relevant changes [8].  Echo in mild valvular dysfunction  7.3 ± 3.0   Do not order non-invasive tests for obstructive coronary artery disease as a routine in pre-operatory evaluation of non-cardiac surgery in asymptomatic individuals with normal functional capacity.  The diagnosis of obstructive CAD and subsequent revascularization does not reduce cardiovascular risk related to surgery (over-diagnosis and potential overtreatment) [9].  Follow-up testing for previous revascularization  7.6 ± 3.0   Do not order non-invasive tests as screening of obstructive coronary artery disease (CAD) in asymptomatic individuals with normal ventricular function.  The diagnosis of stable coronary disease and subsequent revascularization does not improve outcomes (over-diagnosis and potential overtreatment).  CAD Screening  7.1 ± 3.1   Do not order non-invasive tests for obstructive coronary artery disease as a routine in asymptomatic individuals with a history of revascularization procedure.  The diagnosis of stable coronary disease and subsequent revascularization does not improve outcomes (over-diagnosis and potential overtreatment) [10].  Follow-up testing for previous revascularization  6.6 ± 3.0   Do not order carotid ultrasound for syncope investigation in the absence of neurological signs.  Syncope not related to neurologic event is not caused by carotid obstruction (over-diagnosis) [11].  Carotid US for syncope  5.9 ± 3.3  Invasive treatment overuse   Do not perform percutaneous coronary revascularization (PCI) in asymptomatic individuals.  PCI in stable CAD does not improve prognosis, only improves symptoms. In an asymptomatic patient, obviously, there is no symptom to be improved [10].  PCI in asymptomatic  8.0 ± 2.9   Do not perform PCI in a persistently occluded coronary artery after a late myocardial infarction.  PCI of persistently occluded coronary artery does not improve prognosis after myocardial infarction [12].  PCI in occluded artery  7.6 ± 2.9   Avoid routine pulmonary artery catheterization (PAC) in patients with acute heart failure.  PAC does not improve prognosis in heart failure [13].  PAC in acute HF  6.9 ± 3.0   Do not close persistent foramen oval (PFO) to prevent recurrence of cryptogenic stroke.  PFO closure does not reduce stroke recurrence [14].  PFO closure  5.9 ± 3.1   Do not use a routine intra-aortic balloon pump (IOBP) to treat cardiogenic shock due to myocardial infarction.  IOBP does not reduce mortality in patients with shock due to myocardial infarction [15].  IOBP in shock  5.8 ± 3.2  Medical treatment overuse   Do not prescribe beta-blockers before non-cardiac surgery as a means to prevent cardiovascular events related to the procedure.  Beta-blocker does not prevent cardiovascular events related to surgery [16].  Pre-op beta-blocker  7.0 ± 3.0   Do not use aspirin to prevent embolic stroke in patients with atrial fibrillation.  Aspirin does not prevent embolic stroke [17].  Aspirin for embolic stroke  6.9 ± 3.3  CAD, Coronary Artery Disease; PCI, Percutaneous coronary revascularization; PAC, Pulmonary artery catheterization; PFO, Persistent foramen oval; IOBP, Intra-aortic balloon pump. Table 1. Recommendation, justifications and abbreviations Recommendation  Justification  Abbreviation  Mean score  Tests overuse   Do not order calcium score to follow the evolution of subclinical atherosclerosis.  Increase of calcium score over time has no prognostic role and this test offers substantial radiation [7].  Follow-up calcium score  7.9 ± 2.9   Do not order routine echocardiogram because of valvular dysfunction of mild degree.  Low yield of testing in detecting relevant changes [8].  Echo in mild valvular dysfunction  7.3 ± 3.0   Do not order non-invasive tests for obstructive coronary artery disease as a routine in pre-operatory evaluation of non-cardiac surgery in asymptomatic individuals with normal functional capacity.  The diagnosis of obstructive CAD and subsequent revascularization does not reduce cardiovascular risk related to surgery (over-diagnosis and potential overtreatment) [9].  Follow-up testing for previous revascularization  7.6 ± 3.0   Do not order non-invasive tests as screening of obstructive coronary artery disease (CAD) in asymptomatic individuals with normal ventricular function.  The diagnosis of stable coronary disease and subsequent revascularization does not improve outcomes (over-diagnosis and potential overtreatment).  CAD Screening  7.1 ± 3.1   Do not order non-invasive tests for obstructive coronary artery disease as a routine in asymptomatic individuals with a history of revascularization procedure.  The diagnosis of stable coronary disease and subsequent revascularization does not improve outcomes (over-diagnosis and potential overtreatment) [10].  Follow-up testing for previous revascularization  6.6 ± 3.0   Do not order carotid ultrasound for syncope investigation in the absence of neurological signs.  Syncope not related to neurologic event is not caused by carotid obstruction (over-diagnosis) [11].  Carotid US for syncope  5.9 ± 3.3  Invasive treatment overuse   Do not perform percutaneous coronary revascularization (PCI) in asymptomatic individuals.  PCI in stable CAD does not improve prognosis, only improves symptoms. In an asymptomatic patient, obviously, there is no symptom to be improved [10].  PCI in asymptomatic  8.0 ± 2.9   Do not perform PCI in a persistently occluded coronary artery after a late myocardial infarction.  PCI of persistently occluded coronary artery does not improve prognosis after myocardial infarction [12].  PCI in occluded artery  7.6 ± 2.9   Avoid routine pulmonary artery catheterization (PAC) in patients with acute heart failure.  PAC does not improve prognosis in heart failure [13].  PAC in acute HF  6.9 ± 3.0   Do not close persistent foramen oval (PFO) to prevent recurrence of cryptogenic stroke.  PFO closure does not reduce stroke recurrence [14].  PFO closure  5.9 ± 3.1   Do not use a routine intra-aortic balloon pump (IOBP) to treat cardiogenic shock due to myocardial infarction.  IOBP does not reduce mortality in patients with shock due to myocardial infarction [15].  IOBP in shock  5.8 ± 3.2  Medical treatment overuse   Do not prescribe beta-blockers before non-cardiac surgery as a means to prevent cardiovascular events related to the procedure.  Beta-blocker does not prevent cardiovascular events related to surgery [16].  Pre-op beta-blocker  7.0 ± 3.0   Do not use aspirin to prevent embolic stroke in patients with atrial fibrillation.  Aspirin does not prevent embolic stroke [17].  Aspirin for embolic stroke  6.9 ± 3.3  Recommendation  Justification  Abbreviation  Mean score  Tests overuse   Do not order calcium score to follow the evolution of subclinical atherosclerosis.  Increase of calcium score over time has no prognostic role and this test offers substantial radiation [7].  Follow-up calcium score  7.9 ± 2.9   Do not order routine echocardiogram because of valvular dysfunction of mild degree.  Low yield of testing in detecting relevant changes [8].  Echo in mild valvular dysfunction  7.3 ± 3.0   Do not order non-invasive tests for obstructive coronary artery disease as a routine in pre-operatory evaluation of non-cardiac surgery in asymptomatic individuals with normal functional capacity.  The diagnosis of obstructive CAD and subsequent revascularization does not reduce cardiovascular risk related to surgery (over-diagnosis and potential overtreatment) [9].  Follow-up testing for previous revascularization  7.6 ± 3.0   Do not order non-invasive tests as screening of obstructive coronary artery disease (CAD) in asymptomatic individuals with normal ventricular function.  The diagnosis of stable coronary disease and subsequent revascularization does not improve outcomes (over-diagnosis and potential overtreatment).  CAD Screening  7.1 ± 3.1   Do not order non-invasive tests for obstructive coronary artery disease as a routine in asymptomatic individuals with a history of revascularization procedure.  The diagnosis of stable coronary disease and subsequent revascularization does not improve outcomes (over-diagnosis and potential overtreatment) [10].  Follow-up testing for previous revascularization  6.6 ± 3.0   Do not order carotid ultrasound for syncope investigation in the absence of neurological signs.  Syncope not related to neurologic event is not caused by carotid obstruction (over-diagnosis) [11].  Carotid US for syncope  5.9 ± 3.3  Invasive treatment overuse   Do not perform percutaneous coronary revascularization (PCI) in asymptomatic individuals.  PCI in stable CAD does not improve prognosis, only improves symptoms. In an asymptomatic patient, obviously, there is no symptom to be improved [10].  PCI in asymptomatic  8.0 ± 2.9   Do not perform PCI in a persistently occluded coronary artery after a late myocardial infarction.  PCI of persistently occluded coronary artery does not improve prognosis after myocardial infarction [12].  PCI in occluded artery  7.6 ± 2.9   Avoid routine pulmonary artery catheterization (PAC) in patients with acute heart failure.  PAC does not improve prognosis in heart failure [13].  PAC in acute HF  6.9 ± 3.0   Do not close persistent foramen oval (PFO) to prevent recurrence of cryptogenic stroke.  PFO closure does not reduce stroke recurrence [14].  PFO closure  5.9 ± 3.1   Do not use a routine intra-aortic balloon pump (IOBP) to treat cardiogenic shock due to myocardial infarction.  IOBP does not reduce mortality in patients with shock due to myocardial infarction [15].  IOBP in shock  5.8 ± 3.2  Medical treatment overuse   Do not prescribe beta-blockers before non-cardiac surgery as a means to prevent cardiovascular events related to the procedure.  Beta-blocker does not prevent cardiovascular events related to surgery [16].  Pre-op beta-blocker  7.0 ± 3.0   Do not use aspirin to prevent embolic stroke in patients with atrial fibrillation.  Aspirin does not prevent embolic stroke [17].  Aspirin for embolic stroke  6.9 ± 3.3  CAD, Coronary Artery Disease; PCI, Percutaneous coronary revascularization; PAC, Pulmonary artery catheterization; PFO, Persistent foramen oval; IOBP, Intra-aortic balloon pump. An electronic form describing the 13 recommendations was made available on the Society’s website for all members to grade the importance of each item from 1 to 10, in ascending order. Ten would mean total agreement with the ‘do not do’ recommendation and 1 total disagreement. Over three weeks, Society members received email invitations with a link to the electronic form. The form could only be accessed with the personal login and password for each member in the Brazilian Society of Cardiology. Members could vote only once. Cardiologists were instructed to grade each recommendation independently of the other recommendations with no hierarchical relationship among them. This method was supposed to provide a non-biased evaluation of each item. To represent the original and independent opinion of each member, no evidence-based or logical explanation was provided for the recommendation. Monetary cost was not underlined as a major issue in grading the recommendation. The emphasis of the campaign was to prevent patient harm. Data analysis Means and standard deviations of grades were calculated for each recommendation. The top five means were selected as the Choosing Wisely Recommendations of the Brazilian Society of cardiology. The coefficient of variability of scores was used to express heterogeneity of opinion among cardiologists according to specific characteristics. Average scores were compared between groups defined by high or low levels of evidence, diagnostic or non-diagnostic procedure, and invasive or non-invasive procedure by student’s t-test. The score attributed to the best-evaluated recommendation was used as a proxy for the Choosing Wisely concept to identify predictors of physicians’ support. Univariate associations between this score and different variables were assessed by ANOVA for categorical predictors or simple linear regression for numeric predictors. Variables associated with the score (P < 0.15) entered the multivariate analysis of covariance to identify independent predictors. Final statistical significance was defined by a P value <0.05. SPSS Statistical Software was used for statistical analysis. Results Sample characteristics Among 14 579 Brazilian cardiologists, 621 answered the electronic form, comprising 4.3% of the population. This sample had a mean age of 48 ± 12 years old and was younger, and this variable was less dispersed than the entire population’s age of 57 ± 34 years old. Men constituted 80% of the sample, similar to 71% in the population. Geographic distribution of cardiologists in the sample was similar to the population. The majority of cardiologists were clinical cardiologists (59%) followed by imaging specialists (27%), interventional cardiologists (8%), and electrophysiologists (6%). Around one-third of the sample was university professors or faculty in teaching hospitals (Table 2). Table 2. Sample characteristics with some comparison with population characteristics Characteristics  Sample  Population  Size  621  14.579  Age (years)  48 ± 12  57 ± 34  Male gender  498 (80%)  10.325 (71%)  Geographic region in Brazil   South  434 (70%)  8.710 (60%)   Central  38 (6.1%)  1.104 (7.6%)   North  149 (24%)  4.765 (33%)  Residence in state capital  400 (64%)    University professor  171 (28%)    Faculty in teaching hospital  219 (35%)    Maximum degree   Medical doctor only  76 (12%)     Medical doctor with residence  331 (53%)     Master’s degree  111 (18%)     PhD  103 (17%)    Interventional cardiologist  48 (7.7%)    Characteristics  Sample  Population  Size  621  14.579  Age (years)  48 ± 12  57 ± 34  Male gender  498 (80%)  10.325 (71%)  Geographic region in Brazil   South  434 (70%)  8.710 (60%)   Central  38 (6.1%)  1.104 (7.6%)   North  149 (24%)  4.765 (33%)  Residence in state capital  400 (64%)    University professor  171 (28%)    Faculty in teaching hospital  219 (35%)    Maximum degree   Medical doctor only  76 (12%)     Medical doctor with residence  331 (53%)     Master’s degree  111 (18%)     PhD  103 (17%)    Interventional cardiologist  48 (7.7%)    Table 2. Sample characteristics with some comparison with population characteristics Characteristics  Sample  Population  Size  621  14.579  Age (years)  48 ± 12  57 ± 34  Male gender  498 (80%)  10.325 (71%)  Geographic region in Brazil   South  434 (70%)  8.710 (60%)   Central  38 (6.1%)  1.104 (7.6%)   North  149 (24%)  4.765 (33%)  Residence in state capital  400 (64%)    University professor  171 (28%)    Faculty in teaching hospital  219 (35%)    Maximum degree   Medical doctor only  76 (12%)     Medical doctor with residence  331 (53%)     Master’s degree  111 (18%)     PhD  103 (17%)    Interventional cardiologist  48 (7.7%)    Characteristics  Sample  Population  Size  621  14.579  Age (years)  48 ± 12  57 ± 34  Male gender  498 (80%)  10.325 (71%)  Geographic region in Brazil   South  434 (70%)  8.710 (60%)   Central  38 (6.1%)  1.104 (7.6%)   North  149 (24%)  4.765 (33%)  Residence in state capital  400 (64%)    University professor  171 (28%)    Faculty in teaching hospital  219 (35%)    Maximum degree   Medical doctor only  76 (12%)     Medical doctor with residence  331 (53%)     Master’s degree  111 (18%)     PhD  103 (17%)    Interventional cardiologist  48 (7.7%)    Preference-based scores for each recommendations The mean scores for each recommendation varied from a minimum of 5.79 ± 3.17 (median = 6.0; interquartile range = 3–9) for ‘IOBP in shock’ to a maximum of 9.98 ± 2.92 (median = 10; 7–10) for ‘PCI in asymptomatic’. Seven recommendations had a mean score between 7 and 8, three recommendations were between 6 and 7 and three others had a mean between 5 and 6. No recommendation had a mean score <5. The average of the means of all recommendations was 6.95 ± 0.75 (median = 7; 6.2–7.6) (Fig. 1). The most frequent score was 10, leading to a non-normal distribution with negative skewness. The proportion of individuals giving grades ≤ 5 varied from 20% for the recommendation of highest grade to 48% for the recommendation of the lowest grade. Figure 1 View largeDownload slide Mean score for each recommendation according to Brazilian cardiologists. Figure 1 View largeDownload slide Mean score for each recommendation according to Brazilian cardiologists. The list of top 5 chosen recommendations was as follows: (1) Do not perform percutaneous coronary revascularization (PCI) in asymptomatic individuals, (2) Do not order calcium score to follow the evolution of subclinical atherosclerosis, (3) Do not order non-invasive tests for obstructive coronary artery disease as a routine in pre-operatory evaluation of non-cardiac surgery in asymptomatic individuals with normal functional capacity, (4) Do not perform PCI in a persistently occluded coronary artery after a late myocardial infarction and (5) Do not order routine echocardiogram because of valvar dysfunction of mild degree. Variability of scores There was a reasonable amount of variation among cardiologists for all recommendations with coefficients varying from 37% to 55%. The highest coefficient of variability was related to the recommendation of lowest grade (IOBP in shock) while the lowest variability took place in the recommendation of highest grade (PCI in asymptomatic). There was a strong negative correlation between average grade and coefficient of variability (r = −0.96; P < 0.001) (Figure 2). Figure 2 View largeDownload slide Correlation between grades and coefficient variation of grades, showing the higher the grade, the less variation of opinion exists (r = −0.96; P < 0.001). Figure 2 View largeDownload slide Correlation between grades and coefficient variation of grades, showing the higher the grade, the less variation of opinion exists (r = −0.96; P < 0.001). Influence of level of evidence on scores Among the 13 recommendations, 7 had direct evidence indicating lack of beneficial effect according to randomized clinical trials (CAD screening, PCI in asymptomatic, PFO closure, IOBP in shock, pre-op beta-blocker, PCI in occluded artery, and PAC in acute HF), the so-called evidence-based recommendations. The remaining 6 recommendations (pre-operative CAD screening, follow-up testing for previous revascularization, echo in mild valvular dysfunction, follow-up calcium score, carotid US for syncope, and aspirin for embolic stroke) were based on small trials, indirect evidence, or lack of plausible beneficial effect. The average of scores for evidence-based recommendations was 6.9 ± 0.80, similar to the average of 7.0 ± 0.74 for the recommendations not based on randomized trials (P = 0.78). Recommendations regarding diagnostic tests (CAD Screening, pre-operative CAD screening, follow-up testing for previous revascularization, follow-up calcium score, echo in mild valvular dysfunction, and carotid US for syncope) scored similarly to recommendations regarding treatments (7.1 ± 0.73 vs. 6.9 ± 0.88, P = 0.68). The same was observed for recommendations related to invasive procedures (PCI in asymptomatic, PCI in occluded artery, PFO closure, IOBP in shock, and PAC in acute HF) versus non-invasive procedures (6.8 ± 0.97 vs. 7.0 ± 0.63, P = 0.67). Predictors of preference-based scores Table 3 depicts univariate and multivariate analyses to evaluate predictors of the scores attributed by each cardiologist to the recommendation of highest score (PCI in asymptomatic). Table 3. Predictors of physician’s score for PCI in asymptomatic recommendation Dependent variable: grade PCI  Univariate  Multivariate    Coefficient®  P value  Coefficient®  P value  Independent Variables          Age (years)  −0.036  <0.001      Male gender  −0.38  0.20      Residence in country side  −0.27  0.27      University professor  0.24  0.36      Faculty in teaching hospital  0.30  0.22      Interventional cardiologist  −0.71  0.11  −0.83  0.05  Years of medical degree  −0.037  <0.001  −0.038  <0.001  Maximum degree   Medical doctor only  −0.80  0.07  −0.50  0.15   Medical doctor with residence  −0.35  0.29       Master’s degree  −0.14  0.73       PhD  Reference        Geographic region in Brazil   North  −0.068  0.81       Central  −0.337  0.50       South  Reference        Dependent variable: grade PCI  Univariate  Multivariate    Coefficient®  P value  Coefficient®  P value  Independent Variables          Age (years)  −0.036  <0.001      Male gender  −0.38  0.20      Residence in country side  −0.27  0.27      University professor  0.24  0.36      Faculty in teaching hospital  0.30  0.22      Interventional cardiologist  −0.71  0.11  −0.83  0.05  Years of medical degree  −0.037  <0.001  −0.038  <0.001  Maximum degree   Medical doctor only  −0.80  0.07  −0.50  0.15   Medical doctor with residence  −0.35  0.29       Master’s degree  −0.14  0.73       PhD  Reference        Geographic region in Brazil   North  −0.068  0.81       Central  −0.337  0.50       South  Reference        Table 3. Predictors of physician’s score for PCI in asymptomatic recommendation Dependent variable: grade PCI  Univariate  Multivariate    Coefficient®  P value  Coefficient®  P value  Independent Variables          Age (years)  −0.036  <0.001      Male gender  −0.38  0.20      Residence in country side  −0.27  0.27      University professor  0.24  0.36      Faculty in teaching hospital  0.30  0.22      Interventional cardiologist  −0.71  0.11  −0.83  0.05  Years of medical degree  −0.037  <0.001  −0.038  <0.001  Maximum degree   Medical doctor only  −0.80  0.07  −0.50  0.15   Medical doctor with residence  −0.35  0.29       Master’s degree  −0.14  0.73       PhD  Reference        Geographic region in Brazil   North  −0.068  0.81       Central  −0.337  0.50       South  Reference        Dependent variable: grade PCI  Univariate  Multivariate    Coefficient®  P value  Coefficient®  P value  Independent Variables          Age (years)  −0.036  <0.001      Male gender  −0.38  0.20      Residence in country side  −0.27  0.27      University professor  0.24  0.36      Faculty in teaching hospital  0.30  0.22      Interventional cardiologist  −0.71  0.11  −0.83  0.05  Years of medical degree  −0.037  <0.001  −0.038  <0.001  Maximum degree   Medical doctor only  −0.80  0.07  −0.50  0.15   Medical doctor with residence  −0.35  0.29       Master’s degree  −0.14  0.73       PhD  Reference        Geographic region in Brazil   North  −0.068  0.81       Central  −0.337  0.50       South  Reference        Univariate analysis identified ‘age’ and ‘years of graduation’ as negatively associated with scores, meaning that older physicians have a lower level of agreement with recommendations chosen by the panel. In addition, ‘intervention cardiologists and physicians with no post-graduation training’ (residence, master or doctor of sciences) had statistical trends towards lower scores. Other variables such as ‘gender, geographic place of living’, and ‘being a faculty or professor’ did not influence scores. Age was not entered in multivariate analysis to avoid collinearity with ‘years of graduation’. In this analysis, years of graduation (P < 0.001) and intervention cardiologist (P = 0.05) remained as independent predictors of lower preference-based scores. Despite being independent predictors, the influence of years of graduation (Coefficient ®=−0.038) and intervention cardiologist (Coefficient ®=−0.83) on the final score was weak. The R2 of the model was only 0.029. Discussion In this article, we described the first Brazilian experience of applying the Choosing Wisely method to create a 5-topic list of ‘do not do’ recommendations. It was an initiative by the Brazilian Society of Cardiology, one of the largest and most powerful medical societies in the country.Subjectively, it was a well-received initiative, commemorated by many cardiologists who felt that this type of discussion is essential for our progress. Since this campaign followed the cardinal principals of Choosing Wisely [4], this can be interpreted as evidence of its effectiveness for these principals. First, it was physician-led with no participation of health care providers or government; second, it was patient-focused in the sense that prevention of harm was the major issue, as opposed to cost reduction; third, all recommendations proposed by the task force were driven by evidence of harm or lack of evidence as a justification of ‘do not do’; fourth, it was multi-professional as the task force had representation of the main subspecialties in cardiology. Objectively, an overall score of 7 could be interpreted as reasonable support from cardiologists. However, one should look deeper at the data and make a more comprehensive interpretation before evaluating how cardiologists think regarding the ‘less is more’ paradigm. Firstly, it is a sample of individuals who were persuaded to spend 15 min of their time to answer a thoughtful questionnaire on medical decisions. It is only natural that this sample represents physicians more in line with the Choosing Wisely concept. Considering this, an average score of 7 may be interpreted as modest support for the recommendations, especially when 6 recommendations had a score below 7 and the maximal score was only 8. Secondly, there was a high variability of scores among individuals, even for the highest-ranked recommendations. This variability was not influenced by geographical region, indicating that it was not influenced by availability of resources. If universal principals of evidence-based decision-making were uniform in the minds of physicians, a more homogeneous result would probably take place. High variability has been associated with a physician-centred approach rather than a patient-centred or evidence-based decision [18]. Two of our data analyses support this last statement. A multivariate model was built to predict the score each physician gave to the highest-ranked recommendation. The model found that interventional cardiologists are less prone to support a recommendation contrary to his or her practice even if the procedure has been extensively tested and does not provide beneficial effect according to high level of evidence (PCI in the stable and asymptomatic patient) [19]. It may be one example of the so-called ego bias: when someone views his or her own performance as better than that shown by statistics or by someone else. In addition, we found older individuals less prone to support this Choosing Wisely recommendation, which seems to be a non-scientific bias [20]. This bias is probably supported by the traditional cultural stance that more is better, the so-called do-something bias [21]. Despite identifying these two independent predictors of individual preferences against the Choosing Wisely paradigm, we should recognize that the model was not able to explain the majority of score’s variability (R2 = 0.029). This may indicate that preferences are determined by a highly complex and variable set of individualities not captured by our analysis, including values and traits of personality. Thus, this difficulty in predicting cardiologists’ preferences suggests (once again) a lack of scientific thinking. Furthermore, there was no association between level of evidence and recommendation’s score, suggesting that evidence is not a major drive in physician’s preference. Choosing Wisely is not a regulatory process. Its primary intention focused on cultural change, which ultimately might result in behavioural change [22]. Future research should test how much this campaign is able to change culture and behaviour in different environments, specialties, and countries. Although data regarding the cultural impact of the Choosing Wisely campaign remain scarce [23], our subjective perception during the process was that medical and patient communities tend to embrace the concept. It is worth contextualizing how the Choosing Wisely campaign is evolving in Brazil. The first initiative in the Brazilian Society of Cardiology was received with great enthusiasm and became a landmark example for other societies. Following cardiology, family medicine, hospitalists and mastology have completed their processes. Others have initiated the campaign, such as intensive care and gastrointestinal specialty. Moreover, some hospitals have implemented the campaign to reduce overuse, and the Medical and Public Health School of Bahia has launched an educational and research project about Choosing Wisely. In Brazil, the use of social networks has helped spreading the discussion to the general public. We must recognize some limitations of this study: it is a retrospective analysis with no a priori hypothesis; unpredictable sources of variability might have influence physicians answers like time constraints and desire to be recognized as a rational physician [24]; answers to a questionnaire do not necessarily reflect the actual medical decision physicians would make in real-world situations; and, most importantly, we did not use a probabilistic sample of cardiologists. Our sample was made of those interested in contributing with the voting, which may have overestimated preferences for the Choosing Wisely paradigm. Therefore, this should be a considered an exploratory study and should be taken as a hypothesis generator. In conclusion, in this first effort to create a Choosing Wisely list in Brazil, we observed that cardiologists’ support for the concept is modest and highly variable. In an exploratory multivariable analysis, we found that specialty conflicts of interest and the traditional medical thinking of older generations might be factors against changing the culture towards avoiding low-value choices. Funding This work was not supported by any kind of funding. References 1 Choosing Wisely [Available from: http://www.choosingwisely.org/. 2 Chassin MR, Galvin RW, the National Roundtable on Health Care Q. The urgent need to improve health care quality: Institute of medicine national roundtable on health care quality. JAMA  1998; 280: 1000– 5. Google Scholar CrossRef Search ADS PubMed  3 Canada CW. https://choosingwiselycanada.org Acessed [18 Feb 2017] 4 Levinson W, Kallewaard M, Bhatia RS et al.  . ‘Choosing Wisely’: a growing international campaign. BMJ Qual Saf  2015; 24: 167– 74. Google Scholar CrossRef Search ADS PubMed  5 Corallo AN, Croxford R, Goodman DC et al.  . A systematic review of medical practice variation in OECD countries. Health Policy  2014; 114: 5– 14. Google Scholar CrossRef Search ADS PubMed  6 Barros AJD, Santos IS, Matijasevich A et al.  . Patterns of deliveries in a Brazilian birth cohort: almost universal cesarean sections for the better-off. Rev saude publ  2011; 45: 635– 43. Google Scholar CrossRef Search ADS   7 McEvoy JW, Blaha MJ, DeFilippis AP et al.  . Coronary artery calcium progression: an important clinical measurement? J Am College Cardiol  2010; 56: 1613. Google Scholar CrossRef Search ADS   8 Lindekleiv H, Løchen M, Mathiesen EB et al.  . Echocardiographic screening of the general population and long-term survival: a randomized clinical study. JAMA Intern Med  2013; 173: 1592– 8. Google Scholar CrossRef Search ADS PubMed  9 McFalls EO, Ward HB, Moritz TE et al.  . Coronary-artery revascularization before elective major vascular surgery. N Engl J Med  2004; 351: 2795– 804. Google Scholar CrossRef Search ADS PubMed  10 Stergiopoulos K, Boden WE, Hartigan P et al.  . Percutaneous coronary intervention outcomes in patients with stable obstructive coronary artery disease and myocardial ischemia: a collaborative meta-analysis of contemporary randomized clinical trials. JAMA Intern Med  2014; 174: 232– 40. Google Scholar CrossRef Search ADS PubMed  11 Kadian-Dodov D, Papolos A, Olin JW. Diagnostic utility of carotid artery duplex ultrasonography in the evaluation of syncope: a good test ordered for the wrong reason. Eur Heart J Cardiovasc Imaging  2015; 16: 621– 5. Google Scholar PubMed  12 Hochman JS, Lamas GA, Buller CE et al.  . Coronary intervention for persistent occlusion after myocardial infarction. N Engl J Med  2006; 355: 2395– 407. Google Scholar CrossRef Search ADS PubMed  13 The EI, Coordinators* ES. Evaluation study of congestive heart failure and pulmonary artery catheterization effectiveness: the escape trial. JAMA  2005; 294: 1625– 33. CrossRef Search ADS PubMed  14 Udell JA, Opotowsky AR, Khairy P et al.  . Patent foramen ovale closure vs medical therapy for stroke prevention: meta-analysis of randomized trials and review of heterogeneity in meta-analyses. Can J Cardiol  2014; 30: 1216– 24. Google Scholar CrossRef Search ADS PubMed  15 Thiele H, Zeymer U, Neumann F-J et al.  . Intraaortic balloon support for myocardial infarction with cardiogenic shock. N Engl J Med  2012; 367: 1287– 96. Google Scholar CrossRef Search ADS PubMed  16 Bangalore S, Wetterslev J, Pranesh S et al.  . Perioperative beta blockers in patients having non-cardiac surgery: a meta-analysis. The Lancet  9654; 372: 1962– 76. Google Scholar CrossRef Search ADS   17 Sato H, Ishikawa K, Kitabatake A et al.  . Low-dose aspirin for prevention of stroke in low-risk patients with atrial fibrillation. Stroke  2006; 37: 447. Google Scholar CrossRef Search ADS PubMed  18 Malhotra A, Maughan D, Ansell J et al.  . Choosing Wisely in the UK: the Academy of Medical Royal Colleges’ initiative to reduce the harms of too much medicine. Br Med J  2015; 350: h2308. Google Scholar CrossRef Search ADS   19 Boden WE, O’Rourke RA, Teo KK et al.  . Optimal medical therapy with or without pci for stable coronary disease. N Engl J Med  2007; 356: 1503– 16. Google Scholar CrossRef Search ADS PubMed  20 Dawson NV, Arkes HR. Systematic errors in medical decision making. J Gen Intern Med  1987; 2: 183– 7. Google Scholar CrossRef Search ADS PubMed  21 Thomas KB. The consultation and the therapeutic illusion. Br Med J  1978; 1: 1327– 8. Google Scholar CrossRef Search ADS PubMed  22 Wolfson D, Santa J, Slass L. Engaging physicians and consumers in conversations about treatment overuse and waste: a short history of the Choosing Wisely campaign. Acad Med  2014; 89: 990– 5. Google Scholar CrossRef Search ADS PubMed  23 Bhatia RS, Levinson W, Shortt S et al.  . Measuring the effect of Choosing Wisely: an integrated framework to assess campaign impact on low-value care. BMJ Qual Saf  2015; 24: 523– 31. Google Scholar CrossRef Search ADS PubMed  24 Parsons HM. What happened at Hawthorne? Science  1974; 183: 922. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press in association with the International Society for Quality in Health Care. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png International Journal for Quality in Health Care Oxford University Press

Loading next page...
 
/lp/ou_press/choosing-wisely-culture-among-brazilian-cardiologists-VawYxMqI8m
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press in association with the International Society for Quality in Health Care. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
1353-4505
eISSN
1464-3677
D.O.I.
10.1093/intqhc/mzy028
Publisher site
See Article on Publisher Site

Abstract

Abstract Objective (i) To describe how aligned the ‘Choosing Wisely’ concept is with the medical culture among Brazilian cardiologists and (ii) to identify predictors for physicians’ preference for avoiding wasteful care. Design Cross-sectional study. Setting Brazilian Society of Cardiology. Participants Cardiologists who agree to fill a web questionary. Intervention A task force of 12 Brazilian cardiologists prepared a list of 13 ‘do not do’ recommendations, which were made available on the Brazilian Society of Cardiology website for affiliates to assign a supported score of 1 to 10 to each recommendation. Main Outcome Measurement Score average for supporting recommendations. Results Of 14 579 Brazilian cardiologists, 621 (4.3%) answered the questionnaire. The top recommendation was ‘do not perform routine percutaneous coronary intervention in asymptomatic individuals’ (mean score = 8.0 ± 2.9) while the one with the lowest support was ‘do not use an intra-aortic balloon pump in infarction with cardiogenic shock’ (5.8 ± 3.2). None of the 13 recommendations presented a mean grade >9 (strong support); 7 recommendations averaged 7–8 (moderate support) followed by 6 recommendations with an average of 5–7 (modest support). Multivariate analysis independently identified predictors of the score attributed to the top recommendation; being an interventionist and time since graduation were both negatively associated with support. Conclusions (i) The support of Brazilian cardiologists for the ‘Choosing Wisely’ concept is modest to moderate, and (ii) older generations and enthusiasm towards the procedure one performs may be factors against the ‘Choosing Wisely’ philosophy. patient-centred care, shared decision-making, appropriateness, under-use, over-use, patient safety, decision analysis Introduction In 2012, the ‘Choosing Wisely’ campaign was launched by the American Board of Internal Medicine (ABIM) Foundation to promote debate about overuse of tests, treatments, and procedures in the United States [1]. The initiative was triggered by the growing perception of overuse in the United States, which can be defined as ‘the provision of medical services for which the potential for harm exceeds the potential for benefit’ [2]. Thus, Choosing Wisely is focused on preventing harm by avoiding unnecessary tests and treatments. In the campaign, the ABIM Foundation invited medical societies to provide 5–10 ‘do not do’ recommendations regarding common overused procedures. It has been a successful initiative that spread to Canada [3] and several other countries [4], leading to the foundation of Choosing Wisely International. In this context, countries share experiences and discuss strategies for campaign implementation in different parts of the globe. In 2015, Brazil was represented for the first time in a Choosing Wisely international meeting, which was held in London. The Brazilian Society of Cardiology had a seat in the meeting, when campaign methodologies and Principals of Choosing Wisely were discussed: physician-led, patient-focused, evidence-based, multi-professional, and transparent [4]. Then, the Brazilian Society of Cardiology launched the first Choosing Wisely campaign in Brazil under the auspices of Choosing Wisely International. Although overuse has mainly been associated with high-income countries [5], it has been recognized as a growing problem in low-medium income countries. Despite the vast majority of Brazilians being covered by a health public system of relatively low resources, overuse is prevalent. A typical example is the Brazilian leadership in C-section labour, which is performed in 45% of cases, the majority with no specific indication [6]. Cardiology is a medical specialty with a myriad of tests and procedures for which utility varies according to the situation, being a potential source of overuse when indication is detached from an appropriate clinical context. Therefore, our group understood that the Choosing Wisely campaign in the Brazilian Society of Cardiology was an opportunity to explore cardiologists’ culture regarding the issue of avoiding low-value care. The present study aims to (i) describe the method utilized to establish the five recommendations of the Brazilian Society of Cardiology, (ii) describe how strong (or weak) the Choosing Wisely culture is among cardiologists in Brazil and (iii) identify predictors of physicians’ preference for recognizing and avoiding wasteful care. Methods The Brazilian Society of Cardiology has 14 463 members distributed throughout 27 states, predominantly São Paulo, Rio de Janeiro, Bahia, Minas Gerais and Rio Grande do Sul, which embrace 60% of the members. Initially, 12 experts were chosen from among clinical cardiologists, imaging specialists, electrophysiologists, and interventionists to constitute the Choosing Wisely Task Force. No regional criteria were utilized in defining this panel, which was selected based on knowledge and interest in evidence-based medicine, clinical epidemiology, over-diagnosis and overtreatment. The panel brainstormed and defines 13 ‘do not do’ recommendations. The recommendations had to be related to commonly recommended procedures with one of the following criteria: (i) treatments not proved to be beneficial; (ii) treatments proved not to be beneficial; (iii) diagnosis of questionable beneficial effect and potential harm, leaning towards over-diagnosis; and (iv) serial tests of no utility. The 13 recommendations and their rational are depicted in Table 1. Table 1. Recommendation, justifications and abbreviations Recommendation  Justification  Abbreviation  Mean score  Tests overuse   Do not order calcium score to follow the evolution of subclinical atherosclerosis.  Increase of calcium score over time has no prognostic role and this test offers substantial radiation [7].  Follow-up calcium score  7.9 ± 2.9   Do not order routine echocardiogram because of valvular dysfunction of mild degree.  Low yield of testing in detecting relevant changes [8].  Echo in mild valvular dysfunction  7.3 ± 3.0   Do not order non-invasive tests for obstructive coronary artery disease as a routine in pre-operatory evaluation of non-cardiac surgery in asymptomatic individuals with normal functional capacity.  The diagnosis of obstructive CAD and subsequent revascularization does not reduce cardiovascular risk related to surgery (over-diagnosis and potential overtreatment) [9].  Follow-up testing for previous revascularization  7.6 ± 3.0   Do not order non-invasive tests as screening of obstructive coronary artery disease (CAD) in asymptomatic individuals with normal ventricular function.  The diagnosis of stable coronary disease and subsequent revascularization does not improve outcomes (over-diagnosis and potential overtreatment).  CAD Screening  7.1 ± 3.1   Do not order non-invasive tests for obstructive coronary artery disease as a routine in asymptomatic individuals with a history of revascularization procedure.  The diagnosis of stable coronary disease and subsequent revascularization does not improve outcomes (over-diagnosis and potential overtreatment) [10].  Follow-up testing for previous revascularization  6.6 ± 3.0   Do not order carotid ultrasound for syncope investigation in the absence of neurological signs.  Syncope not related to neurologic event is not caused by carotid obstruction (over-diagnosis) [11].  Carotid US for syncope  5.9 ± 3.3  Invasive treatment overuse   Do not perform percutaneous coronary revascularization (PCI) in asymptomatic individuals.  PCI in stable CAD does not improve prognosis, only improves symptoms. In an asymptomatic patient, obviously, there is no symptom to be improved [10].  PCI in asymptomatic  8.0 ± 2.9   Do not perform PCI in a persistently occluded coronary artery after a late myocardial infarction.  PCI of persistently occluded coronary artery does not improve prognosis after myocardial infarction [12].  PCI in occluded artery  7.6 ± 2.9   Avoid routine pulmonary artery catheterization (PAC) in patients with acute heart failure.  PAC does not improve prognosis in heart failure [13].  PAC in acute HF  6.9 ± 3.0   Do not close persistent foramen oval (PFO) to prevent recurrence of cryptogenic stroke.  PFO closure does not reduce stroke recurrence [14].  PFO closure  5.9 ± 3.1   Do not use a routine intra-aortic balloon pump (IOBP) to treat cardiogenic shock due to myocardial infarction.  IOBP does not reduce mortality in patients with shock due to myocardial infarction [15].  IOBP in shock  5.8 ± 3.2  Medical treatment overuse   Do not prescribe beta-blockers before non-cardiac surgery as a means to prevent cardiovascular events related to the procedure.  Beta-blocker does not prevent cardiovascular events related to surgery [16].  Pre-op beta-blocker  7.0 ± 3.0   Do not use aspirin to prevent embolic stroke in patients with atrial fibrillation.  Aspirin does not prevent embolic stroke [17].  Aspirin for embolic stroke  6.9 ± 3.3  Recommendation  Justification  Abbreviation  Mean score  Tests overuse   Do not order calcium score to follow the evolution of subclinical atherosclerosis.  Increase of calcium score over time has no prognostic role and this test offers substantial radiation [7].  Follow-up calcium score  7.9 ± 2.9   Do not order routine echocardiogram because of valvular dysfunction of mild degree.  Low yield of testing in detecting relevant changes [8].  Echo in mild valvular dysfunction  7.3 ± 3.0   Do not order non-invasive tests for obstructive coronary artery disease as a routine in pre-operatory evaluation of non-cardiac surgery in asymptomatic individuals with normal functional capacity.  The diagnosis of obstructive CAD and subsequent revascularization does not reduce cardiovascular risk related to surgery (over-diagnosis and potential overtreatment) [9].  Follow-up testing for previous revascularization  7.6 ± 3.0   Do not order non-invasive tests as screening of obstructive coronary artery disease (CAD) in asymptomatic individuals with normal ventricular function.  The diagnosis of stable coronary disease and subsequent revascularization does not improve outcomes (over-diagnosis and potential overtreatment).  CAD Screening  7.1 ± 3.1   Do not order non-invasive tests for obstructive coronary artery disease as a routine in asymptomatic individuals with a history of revascularization procedure.  The diagnosis of stable coronary disease and subsequent revascularization does not improve outcomes (over-diagnosis and potential overtreatment) [10].  Follow-up testing for previous revascularization  6.6 ± 3.0   Do not order carotid ultrasound for syncope investigation in the absence of neurological signs.  Syncope not related to neurologic event is not caused by carotid obstruction (over-diagnosis) [11].  Carotid US for syncope  5.9 ± 3.3  Invasive treatment overuse   Do not perform percutaneous coronary revascularization (PCI) in asymptomatic individuals.  PCI in stable CAD does not improve prognosis, only improves symptoms. In an asymptomatic patient, obviously, there is no symptom to be improved [10].  PCI in asymptomatic  8.0 ± 2.9   Do not perform PCI in a persistently occluded coronary artery after a late myocardial infarction.  PCI of persistently occluded coronary artery does not improve prognosis after myocardial infarction [12].  PCI in occluded artery  7.6 ± 2.9   Avoid routine pulmonary artery catheterization (PAC) in patients with acute heart failure.  PAC does not improve prognosis in heart failure [13].  PAC in acute HF  6.9 ± 3.0   Do not close persistent foramen oval (PFO) to prevent recurrence of cryptogenic stroke.  PFO closure does not reduce stroke recurrence [14].  PFO closure  5.9 ± 3.1   Do not use a routine intra-aortic balloon pump (IOBP) to treat cardiogenic shock due to myocardial infarction.  IOBP does not reduce mortality in patients with shock due to myocardial infarction [15].  IOBP in shock  5.8 ± 3.2  Medical treatment overuse   Do not prescribe beta-blockers before non-cardiac surgery as a means to prevent cardiovascular events related to the procedure.  Beta-blocker does not prevent cardiovascular events related to surgery [16].  Pre-op beta-blocker  7.0 ± 3.0   Do not use aspirin to prevent embolic stroke in patients with atrial fibrillation.  Aspirin does not prevent embolic stroke [17].  Aspirin for embolic stroke  6.9 ± 3.3  CAD, Coronary Artery Disease; PCI, Percutaneous coronary revascularization; PAC, Pulmonary artery catheterization; PFO, Persistent foramen oval; IOBP, Intra-aortic balloon pump. Table 1. Recommendation, justifications and abbreviations Recommendation  Justification  Abbreviation  Mean score  Tests overuse   Do not order calcium score to follow the evolution of subclinical atherosclerosis.  Increase of calcium score over time has no prognostic role and this test offers substantial radiation [7].  Follow-up calcium score  7.9 ± 2.9   Do not order routine echocardiogram because of valvular dysfunction of mild degree.  Low yield of testing in detecting relevant changes [8].  Echo in mild valvular dysfunction  7.3 ± 3.0   Do not order non-invasive tests for obstructive coronary artery disease as a routine in pre-operatory evaluation of non-cardiac surgery in asymptomatic individuals with normal functional capacity.  The diagnosis of obstructive CAD and subsequent revascularization does not reduce cardiovascular risk related to surgery (over-diagnosis and potential overtreatment) [9].  Follow-up testing for previous revascularization  7.6 ± 3.0   Do not order non-invasive tests as screening of obstructive coronary artery disease (CAD) in asymptomatic individuals with normal ventricular function.  The diagnosis of stable coronary disease and subsequent revascularization does not improve outcomes (over-diagnosis and potential overtreatment).  CAD Screening  7.1 ± 3.1   Do not order non-invasive tests for obstructive coronary artery disease as a routine in asymptomatic individuals with a history of revascularization procedure.  The diagnosis of stable coronary disease and subsequent revascularization does not improve outcomes (over-diagnosis and potential overtreatment) [10].  Follow-up testing for previous revascularization  6.6 ± 3.0   Do not order carotid ultrasound for syncope investigation in the absence of neurological signs.  Syncope not related to neurologic event is not caused by carotid obstruction (over-diagnosis) [11].  Carotid US for syncope  5.9 ± 3.3  Invasive treatment overuse   Do not perform percutaneous coronary revascularization (PCI) in asymptomatic individuals.  PCI in stable CAD does not improve prognosis, only improves symptoms. In an asymptomatic patient, obviously, there is no symptom to be improved [10].  PCI in asymptomatic  8.0 ± 2.9   Do not perform PCI in a persistently occluded coronary artery after a late myocardial infarction.  PCI of persistently occluded coronary artery does not improve prognosis after myocardial infarction [12].  PCI in occluded artery  7.6 ± 2.9   Avoid routine pulmonary artery catheterization (PAC) in patients with acute heart failure.  PAC does not improve prognosis in heart failure [13].  PAC in acute HF  6.9 ± 3.0   Do not close persistent foramen oval (PFO) to prevent recurrence of cryptogenic stroke.  PFO closure does not reduce stroke recurrence [14].  PFO closure  5.9 ± 3.1   Do not use a routine intra-aortic balloon pump (IOBP) to treat cardiogenic shock due to myocardial infarction.  IOBP does not reduce mortality in patients with shock due to myocardial infarction [15].  IOBP in shock  5.8 ± 3.2  Medical treatment overuse   Do not prescribe beta-blockers before non-cardiac surgery as a means to prevent cardiovascular events related to the procedure.  Beta-blocker does not prevent cardiovascular events related to surgery [16].  Pre-op beta-blocker  7.0 ± 3.0   Do not use aspirin to prevent embolic stroke in patients with atrial fibrillation.  Aspirin does not prevent embolic stroke [17].  Aspirin for embolic stroke  6.9 ± 3.3  Recommendation  Justification  Abbreviation  Mean score  Tests overuse   Do not order calcium score to follow the evolution of subclinical atherosclerosis.  Increase of calcium score over time has no prognostic role and this test offers substantial radiation [7].  Follow-up calcium score  7.9 ± 2.9   Do not order routine echocardiogram because of valvular dysfunction of mild degree.  Low yield of testing in detecting relevant changes [8].  Echo in mild valvular dysfunction  7.3 ± 3.0   Do not order non-invasive tests for obstructive coronary artery disease as a routine in pre-operatory evaluation of non-cardiac surgery in asymptomatic individuals with normal functional capacity.  The diagnosis of obstructive CAD and subsequent revascularization does not reduce cardiovascular risk related to surgery (over-diagnosis and potential overtreatment) [9].  Follow-up testing for previous revascularization  7.6 ± 3.0   Do not order non-invasive tests as screening of obstructive coronary artery disease (CAD) in asymptomatic individuals with normal ventricular function.  The diagnosis of stable coronary disease and subsequent revascularization does not improve outcomes (over-diagnosis and potential overtreatment).  CAD Screening  7.1 ± 3.1   Do not order non-invasive tests for obstructive coronary artery disease as a routine in asymptomatic individuals with a history of revascularization procedure.  The diagnosis of stable coronary disease and subsequent revascularization does not improve outcomes (over-diagnosis and potential overtreatment) [10].  Follow-up testing for previous revascularization  6.6 ± 3.0   Do not order carotid ultrasound for syncope investigation in the absence of neurological signs.  Syncope not related to neurologic event is not caused by carotid obstruction (over-diagnosis) [11].  Carotid US for syncope  5.9 ± 3.3  Invasive treatment overuse   Do not perform percutaneous coronary revascularization (PCI) in asymptomatic individuals.  PCI in stable CAD does not improve prognosis, only improves symptoms. In an asymptomatic patient, obviously, there is no symptom to be improved [10].  PCI in asymptomatic  8.0 ± 2.9   Do not perform PCI in a persistently occluded coronary artery after a late myocardial infarction.  PCI of persistently occluded coronary artery does not improve prognosis after myocardial infarction [12].  PCI in occluded artery  7.6 ± 2.9   Avoid routine pulmonary artery catheterization (PAC) in patients with acute heart failure.  PAC does not improve prognosis in heart failure [13].  PAC in acute HF  6.9 ± 3.0   Do not close persistent foramen oval (PFO) to prevent recurrence of cryptogenic stroke.  PFO closure does not reduce stroke recurrence [14].  PFO closure  5.9 ± 3.1   Do not use a routine intra-aortic balloon pump (IOBP) to treat cardiogenic shock due to myocardial infarction.  IOBP does not reduce mortality in patients with shock due to myocardial infarction [15].  IOBP in shock  5.8 ± 3.2  Medical treatment overuse   Do not prescribe beta-blockers before non-cardiac surgery as a means to prevent cardiovascular events related to the procedure.  Beta-blocker does not prevent cardiovascular events related to surgery [16].  Pre-op beta-blocker  7.0 ± 3.0   Do not use aspirin to prevent embolic stroke in patients with atrial fibrillation.  Aspirin does not prevent embolic stroke [17].  Aspirin for embolic stroke  6.9 ± 3.3  CAD, Coronary Artery Disease; PCI, Percutaneous coronary revascularization; PAC, Pulmonary artery catheterization; PFO, Persistent foramen oval; IOBP, Intra-aortic balloon pump. An electronic form describing the 13 recommendations was made available on the Society’s website for all members to grade the importance of each item from 1 to 10, in ascending order. Ten would mean total agreement with the ‘do not do’ recommendation and 1 total disagreement. Over three weeks, Society members received email invitations with a link to the electronic form. The form could only be accessed with the personal login and password for each member in the Brazilian Society of Cardiology. Members could vote only once. Cardiologists were instructed to grade each recommendation independently of the other recommendations with no hierarchical relationship among them. This method was supposed to provide a non-biased evaluation of each item. To represent the original and independent opinion of each member, no evidence-based or logical explanation was provided for the recommendation. Monetary cost was not underlined as a major issue in grading the recommendation. The emphasis of the campaign was to prevent patient harm. Data analysis Means and standard deviations of grades were calculated for each recommendation. The top five means were selected as the Choosing Wisely Recommendations of the Brazilian Society of cardiology. The coefficient of variability of scores was used to express heterogeneity of opinion among cardiologists according to specific characteristics. Average scores were compared between groups defined by high or low levels of evidence, diagnostic or non-diagnostic procedure, and invasive or non-invasive procedure by student’s t-test. The score attributed to the best-evaluated recommendation was used as a proxy for the Choosing Wisely concept to identify predictors of physicians’ support. Univariate associations between this score and different variables were assessed by ANOVA for categorical predictors or simple linear regression for numeric predictors. Variables associated with the score (P < 0.15) entered the multivariate analysis of covariance to identify independent predictors. Final statistical significance was defined by a P value <0.05. SPSS Statistical Software was used for statistical analysis. Results Sample characteristics Among 14 579 Brazilian cardiologists, 621 answered the electronic form, comprising 4.3% of the population. This sample had a mean age of 48 ± 12 years old and was younger, and this variable was less dispersed than the entire population’s age of 57 ± 34 years old. Men constituted 80% of the sample, similar to 71% in the population. Geographic distribution of cardiologists in the sample was similar to the population. The majority of cardiologists were clinical cardiologists (59%) followed by imaging specialists (27%), interventional cardiologists (8%), and electrophysiologists (6%). Around one-third of the sample was university professors or faculty in teaching hospitals (Table 2). Table 2. Sample characteristics with some comparison with population characteristics Characteristics  Sample  Population  Size  621  14.579  Age (years)  48 ± 12  57 ± 34  Male gender  498 (80%)  10.325 (71%)  Geographic region in Brazil   South  434 (70%)  8.710 (60%)   Central  38 (6.1%)  1.104 (7.6%)   North  149 (24%)  4.765 (33%)  Residence in state capital  400 (64%)    University professor  171 (28%)    Faculty in teaching hospital  219 (35%)    Maximum degree   Medical doctor only  76 (12%)     Medical doctor with residence  331 (53%)     Master’s degree  111 (18%)     PhD  103 (17%)    Interventional cardiologist  48 (7.7%)    Characteristics  Sample  Population  Size  621  14.579  Age (years)  48 ± 12  57 ± 34  Male gender  498 (80%)  10.325 (71%)  Geographic region in Brazil   South  434 (70%)  8.710 (60%)   Central  38 (6.1%)  1.104 (7.6%)   North  149 (24%)  4.765 (33%)  Residence in state capital  400 (64%)    University professor  171 (28%)    Faculty in teaching hospital  219 (35%)    Maximum degree   Medical doctor only  76 (12%)     Medical doctor with residence  331 (53%)     Master’s degree  111 (18%)     PhD  103 (17%)    Interventional cardiologist  48 (7.7%)    Table 2. Sample characteristics with some comparison with population characteristics Characteristics  Sample  Population  Size  621  14.579  Age (years)  48 ± 12  57 ± 34  Male gender  498 (80%)  10.325 (71%)  Geographic region in Brazil   South  434 (70%)  8.710 (60%)   Central  38 (6.1%)  1.104 (7.6%)   North  149 (24%)  4.765 (33%)  Residence in state capital  400 (64%)    University professor  171 (28%)    Faculty in teaching hospital  219 (35%)    Maximum degree   Medical doctor only  76 (12%)     Medical doctor with residence  331 (53%)     Master’s degree  111 (18%)     PhD  103 (17%)    Interventional cardiologist  48 (7.7%)    Characteristics  Sample  Population  Size  621  14.579  Age (years)  48 ± 12  57 ± 34  Male gender  498 (80%)  10.325 (71%)  Geographic region in Brazil   South  434 (70%)  8.710 (60%)   Central  38 (6.1%)  1.104 (7.6%)   North  149 (24%)  4.765 (33%)  Residence in state capital  400 (64%)    University professor  171 (28%)    Faculty in teaching hospital  219 (35%)    Maximum degree   Medical doctor only  76 (12%)     Medical doctor with residence  331 (53%)     Master’s degree  111 (18%)     PhD  103 (17%)    Interventional cardiologist  48 (7.7%)    Preference-based scores for each recommendations The mean scores for each recommendation varied from a minimum of 5.79 ± 3.17 (median = 6.0; interquartile range = 3–9) for ‘IOBP in shock’ to a maximum of 9.98 ± 2.92 (median = 10; 7–10) for ‘PCI in asymptomatic’. Seven recommendations had a mean score between 7 and 8, three recommendations were between 6 and 7 and three others had a mean between 5 and 6. No recommendation had a mean score <5. The average of the means of all recommendations was 6.95 ± 0.75 (median = 7; 6.2–7.6) (Fig. 1). The most frequent score was 10, leading to a non-normal distribution with negative skewness. The proportion of individuals giving grades ≤ 5 varied from 20% for the recommendation of highest grade to 48% for the recommendation of the lowest grade. Figure 1 View largeDownload slide Mean score for each recommendation according to Brazilian cardiologists. Figure 1 View largeDownload slide Mean score for each recommendation according to Brazilian cardiologists. The list of top 5 chosen recommendations was as follows: (1) Do not perform percutaneous coronary revascularization (PCI) in asymptomatic individuals, (2) Do not order calcium score to follow the evolution of subclinical atherosclerosis, (3) Do not order non-invasive tests for obstructive coronary artery disease as a routine in pre-operatory evaluation of non-cardiac surgery in asymptomatic individuals with normal functional capacity, (4) Do not perform PCI in a persistently occluded coronary artery after a late myocardial infarction and (5) Do not order routine echocardiogram because of valvar dysfunction of mild degree. Variability of scores There was a reasonable amount of variation among cardiologists for all recommendations with coefficients varying from 37% to 55%. The highest coefficient of variability was related to the recommendation of lowest grade (IOBP in shock) while the lowest variability took place in the recommendation of highest grade (PCI in asymptomatic). There was a strong negative correlation between average grade and coefficient of variability (r = −0.96; P < 0.001) (Figure 2). Figure 2 View largeDownload slide Correlation between grades and coefficient variation of grades, showing the higher the grade, the less variation of opinion exists (r = −0.96; P < 0.001). Figure 2 View largeDownload slide Correlation between grades and coefficient variation of grades, showing the higher the grade, the less variation of opinion exists (r = −0.96; P < 0.001). Influence of level of evidence on scores Among the 13 recommendations, 7 had direct evidence indicating lack of beneficial effect according to randomized clinical trials (CAD screening, PCI in asymptomatic, PFO closure, IOBP in shock, pre-op beta-blocker, PCI in occluded artery, and PAC in acute HF), the so-called evidence-based recommendations. The remaining 6 recommendations (pre-operative CAD screening, follow-up testing for previous revascularization, echo in mild valvular dysfunction, follow-up calcium score, carotid US for syncope, and aspirin for embolic stroke) were based on small trials, indirect evidence, or lack of plausible beneficial effect. The average of scores for evidence-based recommendations was 6.9 ± 0.80, similar to the average of 7.0 ± 0.74 for the recommendations not based on randomized trials (P = 0.78). Recommendations regarding diagnostic tests (CAD Screening, pre-operative CAD screening, follow-up testing for previous revascularization, follow-up calcium score, echo in mild valvular dysfunction, and carotid US for syncope) scored similarly to recommendations regarding treatments (7.1 ± 0.73 vs. 6.9 ± 0.88, P = 0.68). The same was observed for recommendations related to invasive procedures (PCI in asymptomatic, PCI in occluded artery, PFO closure, IOBP in shock, and PAC in acute HF) versus non-invasive procedures (6.8 ± 0.97 vs. 7.0 ± 0.63, P = 0.67). Predictors of preference-based scores Table 3 depicts univariate and multivariate analyses to evaluate predictors of the scores attributed by each cardiologist to the recommendation of highest score (PCI in asymptomatic). Table 3. Predictors of physician’s score for PCI in asymptomatic recommendation Dependent variable: grade PCI  Univariate  Multivariate    Coefficient®  P value  Coefficient®  P value  Independent Variables          Age (years)  −0.036  <0.001      Male gender  −0.38  0.20      Residence in country side  −0.27  0.27      University professor  0.24  0.36      Faculty in teaching hospital  0.30  0.22      Interventional cardiologist  −0.71  0.11  −0.83  0.05  Years of medical degree  −0.037  <0.001  −0.038  <0.001  Maximum degree   Medical doctor only  −0.80  0.07  −0.50  0.15   Medical doctor with residence  −0.35  0.29       Master’s degree  −0.14  0.73       PhD  Reference        Geographic region in Brazil   North  −0.068  0.81       Central  −0.337  0.50       South  Reference        Dependent variable: grade PCI  Univariate  Multivariate    Coefficient®  P value  Coefficient®  P value  Independent Variables          Age (years)  −0.036  <0.001      Male gender  −0.38  0.20      Residence in country side  −0.27  0.27      University professor  0.24  0.36      Faculty in teaching hospital  0.30  0.22      Interventional cardiologist  −0.71  0.11  −0.83  0.05  Years of medical degree  −0.037  <0.001  −0.038  <0.001  Maximum degree   Medical doctor only  −0.80  0.07  −0.50  0.15   Medical doctor with residence  −0.35  0.29       Master’s degree  −0.14  0.73       PhD  Reference        Geographic region in Brazil   North  −0.068  0.81       Central  −0.337  0.50       South  Reference        Table 3. Predictors of physician’s score for PCI in asymptomatic recommendation Dependent variable: grade PCI  Univariate  Multivariate    Coefficient®  P value  Coefficient®  P value  Independent Variables          Age (years)  −0.036  <0.001      Male gender  −0.38  0.20      Residence in country side  −0.27  0.27      University professor  0.24  0.36      Faculty in teaching hospital  0.30  0.22      Interventional cardiologist  −0.71  0.11  −0.83  0.05  Years of medical degree  −0.037  <0.001  −0.038  <0.001  Maximum degree   Medical doctor only  −0.80  0.07  −0.50  0.15   Medical doctor with residence  −0.35  0.29       Master’s degree  −0.14  0.73       PhD  Reference        Geographic region in Brazil   North  −0.068  0.81       Central  −0.337  0.50       South  Reference        Dependent variable: grade PCI  Univariate  Multivariate    Coefficient®  P value  Coefficient®  P value  Independent Variables          Age (years)  −0.036  <0.001      Male gender  −0.38  0.20      Residence in country side  −0.27  0.27      University professor  0.24  0.36      Faculty in teaching hospital  0.30  0.22      Interventional cardiologist  −0.71  0.11  −0.83  0.05  Years of medical degree  −0.037  <0.001  −0.038  <0.001  Maximum degree   Medical doctor only  −0.80  0.07  −0.50  0.15   Medical doctor with residence  −0.35  0.29       Master’s degree  −0.14  0.73       PhD  Reference        Geographic region in Brazil   North  −0.068  0.81       Central  −0.337  0.50       South  Reference        Univariate analysis identified ‘age’ and ‘years of graduation’ as negatively associated with scores, meaning that older physicians have a lower level of agreement with recommendations chosen by the panel. In addition, ‘intervention cardiologists and physicians with no post-graduation training’ (residence, master or doctor of sciences) had statistical trends towards lower scores. Other variables such as ‘gender, geographic place of living’, and ‘being a faculty or professor’ did not influence scores. Age was not entered in multivariate analysis to avoid collinearity with ‘years of graduation’. In this analysis, years of graduation (P < 0.001) and intervention cardiologist (P = 0.05) remained as independent predictors of lower preference-based scores. Despite being independent predictors, the influence of years of graduation (Coefficient ®=−0.038) and intervention cardiologist (Coefficient ®=−0.83) on the final score was weak. The R2 of the model was only 0.029. Discussion In this article, we described the first Brazilian experience of applying the Choosing Wisely method to create a 5-topic list of ‘do not do’ recommendations. It was an initiative by the Brazilian Society of Cardiology, one of the largest and most powerful medical societies in the country.Subjectively, it was a well-received initiative, commemorated by many cardiologists who felt that this type of discussion is essential for our progress. Since this campaign followed the cardinal principals of Choosing Wisely [4], this can be interpreted as evidence of its effectiveness for these principals. First, it was physician-led with no participation of health care providers or government; second, it was patient-focused in the sense that prevention of harm was the major issue, as opposed to cost reduction; third, all recommendations proposed by the task force were driven by evidence of harm or lack of evidence as a justification of ‘do not do’; fourth, it was multi-professional as the task force had representation of the main subspecialties in cardiology. Objectively, an overall score of 7 could be interpreted as reasonable support from cardiologists. However, one should look deeper at the data and make a more comprehensive interpretation before evaluating how cardiologists think regarding the ‘less is more’ paradigm. Firstly, it is a sample of individuals who were persuaded to spend 15 min of their time to answer a thoughtful questionnaire on medical decisions. It is only natural that this sample represents physicians more in line with the Choosing Wisely concept. Considering this, an average score of 7 may be interpreted as modest support for the recommendations, especially when 6 recommendations had a score below 7 and the maximal score was only 8. Secondly, there was a high variability of scores among individuals, even for the highest-ranked recommendations. This variability was not influenced by geographical region, indicating that it was not influenced by availability of resources. If universal principals of evidence-based decision-making were uniform in the minds of physicians, a more homogeneous result would probably take place. High variability has been associated with a physician-centred approach rather than a patient-centred or evidence-based decision [18]. Two of our data analyses support this last statement. A multivariate model was built to predict the score each physician gave to the highest-ranked recommendation. The model found that interventional cardiologists are less prone to support a recommendation contrary to his or her practice even if the procedure has been extensively tested and does not provide beneficial effect according to high level of evidence (PCI in the stable and asymptomatic patient) [19]. It may be one example of the so-called ego bias: when someone views his or her own performance as better than that shown by statistics or by someone else. In addition, we found older individuals less prone to support this Choosing Wisely recommendation, which seems to be a non-scientific bias [20]. This bias is probably supported by the traditional cultural stance that more is better, the so-called do-something bias [21]. Despite identifying these two independent predictors of individual preferences against the Choosing Wisely paradigm, we should recognize that the model was not able to explain the majority of score’s variability (R2 = 0.029). This may indicate that preferences are determined by a highly complex and variable set of individualities not captured by our analysis, including values and traits of personality. Thus, this difficulty in predicting cardiologists’ preferences suggests (once again) a lack of scientific thinking. Furthermore, there was no association between level of evidence and recommendation’s score, suggesting that evidence is not a major drive in physician’s preference. Choosing Wisely is not a regulatory process. Its primary intention focused on cultural change, which ultimately might result in behavioural change [22]. Future research should test how much this campaign is able to change culture and behaviour in different environments, specialties, and countries. Although data regarding the cultural impact of the Choosing Wisely campaign remain scarce [23], our subjective perception during the process was that medical and patient communities tend to embrace the concept. It is worth contextualizing how the Choosing Wisely campaign is evolving in Brazil. The first initiative in the Brazilian Society of Cardiology was received with great enthusiasm and became a landmark example for other societies. Following cardiology, family medicine, hospitalists and mastology have completed their processes. Others have initiated the campaign, such as intensive care and gastrointestinal specialty. Moreover, some hospitals have implemented the campaign to reduce overuse, and the Medical and Public Health School of Bahia has launched an educational and research project about Choosing Wisely. In Brazil, the use of social networks has helped spreading the discussion to the general public. We must recognize some limitations of this study: it is a retrospective analysis with no a priori hypothesis; unpredictable sources of variability might have influence physicians answers like time constraints and desire to be recognized as a rational physician [24]; answers to a questionnaire do not necessarily reflect the actual medical decision physicians would make in real-world situations; and, most importantly, we did not use a probabilistic sample of cardiologists. Our sample was made of those interested in contributing with the voting, which may have overestimated preferences for the Choosing Wisely paradigm. Therefore, this should be a considered an exploratory study and should be taken as a hypothesis generator. In conclusion, in this first effort to create a Choosing Wisely list in Brazil, we observed that cardiologists’ support for the concept is modest and highly variable. In an exploratory multivariable analysis, we found that specialty conflicts of interest and the traditional medical thinking of older generations might be factors against changing the culture towards avoiding low-value choices. Funding This work was not supported by any kind of funding. References 1 Choosing Wisely [Available from: http://www.choosingwisely.org/. 2 Chassin MR, Galvin RW, the National Roundtable on Health Care Q. The urgent need to improve health care quality: Institute of medicine national roundtable on health care quality. JAMA  1998; 280: 1000– 5. Google Scholar CrossRef Search ADS PubMed  3 Canada CW. https://choosingwiselycanada.org Acessed [18 Feb 2017] 4 Levinson W, Kallewaard M, Bhatia RS et al.  . ‘Choosing Wisely’: a growing international campaign. BMJ Qual Saf  2015; 24: 167– 74. Google Scholar CrossRef Search ADS PubMed  5 Corallo AN, Croxford R, Goodman DC et al.  . A systematic review of medical practice variation in OECD countries. Health Policy  2014; 114: 5– 14. Google Scholar CrossRef Search ADS PubMed  6 Barros AJD, Santos IS, Matijasevich A et al.  . Patterns of deliveries in a Brazilian birth cohort: almost universal cesarean sections for the better-off. Rev saude publ  2011; 45: 635– 43. Google Scholar CrossRef Search ADS   7 McEvoy JW, Blaha MJ, DeFilippis AP et al.  . Coronary artery calcium progression: an important clinical measurement? J Am College Cardiol  2010; 56: 1613. Google Scholar CrossRef Search ADS   8 Lindekleiv H, Løchen M, Mathiesen EB et al.  . Echocardiographic screening of the general population and long-term survival: a randomized clinical study. JAMA Intern Med  2013; 173: 1592– 8. Google Scholar CrossRef Search ADS PubMed  9 McFalls EO, Ward HB, Moritz TE et al.  . Coronary-artery revascularization before elective major vascular surgery. N Engl J Med  2004; 351: 2795– 804. Google Scholar CrossRef Search ADS PubMed  10 Stergiopoulos K, Boden WE, Hartigan P et al.  . Percutaneous coronary intervention outcomes in patients with stable obstructive coronary artery disease and myocardial ischemia: a collaborative meta-analysis of contemporary randomized clinical trials. JAMA Intern Med  2014; 174: 232– 40. Google Scholar CrossRef Search ADS PubMed  11 Kadian-Dodov D, Papolos A, Olin JW. Diagnostic utility of carotid artery duplex ultrasonography in the evaluation of syncope: a good test ordered for the wrong reason. Eur Heart J Cardiovasc Imaging  2015; 16: 621– 5. Google Scholar PubMed  12 Hochman JS, Lamas GA, Buller CE et al.  . Coronary intervention for persistent occlusion after myocardial infarction. N Engl J Med  2006; 355: 2395– 407. Google Scholar CrossRef Search ADS PubMed  13 The EI, Coordinators* ES. Evaluation study of congestive heart failure and pulmonary artery catheterization effectiveness: the escape trial. JAMA  2005; 294: 1625– 33. CrossRef Search ADS PubMed  14 Udell JA, Opotowsky AR, Khairy P et al.  . Patent foramen ovale closure vs medical therapy for stroke prevention: meta-analysis of randomized trials and review of heterogeneity in meta-analyses. Can J Cardiol  2014; 30: 1216– 24. Google Scholar CrossRef Search ADS PubMed  15 Thiele H, Zeymer U, Neumann F-J et al.  . Intraaortic balloon support for myocardial infarction with cardiogenic shock. N Engl J Med  2012; 367: 1287– 96. Google Scholar CrossRef Search ADS PubMed  16 Bangalore S, Wetterslev J, Pranesh S et al.  . Perioperative beta blockers in patients having non-cardiac surgery: a meta-analysis. The Lancet  9654; 372: 1962– 76. Google Scholar CrossRef Search ADS   17 Sato H, Ishikawa K, Kitabatake A et al.  . Low-dose aspirin for prevention of stroke in low-risk patients with atrial fibrillation. Stroke  2006; 37: 447. Google Scholar CrossRef Search ADS PubMed  18 Malhotra A, Maughan D, Ansell J et al.  . Choosing Wisely in the UK: the Academy of Medical Royal Colleges’ initiative to reduce the harms of too much medicine. Br Med J  2015; 350: h2308. Google Scholar CrossRef Search ADS   19 Boden WE, O’Rourke RA, Teo KK et al.  . Optimal medical therapy with or without pci for stable coronary disease. N Engl J Med  2007; 356: 1503– 16. Google Scholar CrossRef Search ADS PubMed  20 Dawson NV, Arkes HR. Systematic errors in medical decision making. J Gen Intern Med  1987; 2: 183– 7. Google Scholar CrossRef Search ADS PubMed  21 Thomas KB. The consultation and the therapeutic illusion. Br Med J  1978; 1: 1327– 8. Google Scholar CrossRef Search ADS PubMed  22 Wolfson D, Santa J, Slass L. Engaging physicians and consumers in conversations about treatment overuse and waste: a short history of the Choosing Wisely campaign. Acad Med  2014; 89: 990– 5. Google Scholar CrossRef Search ADS PubMed  23 Bhatia RS, Levinson W, Shortt S et al.  . Measuring the effect of Choosing Wisely: an integrated framework to assess campaign impact on low-value care. BMJ Qual Saf  2015; 24: 523– 31. Google Scholar CrossRef Search ADS PubMed  24 Parsons HM. What happened at Hawthorne? Science  1974; 183: 922. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press in association with the International Society for Quality in Health Care. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com

Journal

International Journal for Quality in Health CareOxford University Press

Published: Mar 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off