You are here

  • Home
  • Archive
  • Volume 12, Issue 4
  • Evidence for overuse of cardiovascular healthcare services in high-income countries: protocol for a systematic review and meta-analysis

Article Text

Article menu
Cardiovascular medicine
Protocol
Evidence for overuse of cardiovascular healthcare services in high-income countries: protocol for a systematic review and meta-analysis
  1. Derek J Roberts1,2,3,
  2. Emma E Sypes4,
  3. Sudhir K Nagpal1,
  4. Daniel Niven5,
  5. Mamas Mamas6,
  6. Daniel I McIsaac7,
  7. Carl van Walraven2,3,8,
  8. Risa Shorr9,
  9. Ian D Graham2,3,
  10. Henry Thomas Stelfox5,
  11. Jeremy Grimshaw2,3
  1. 1Division of Vascular and Endovascular Surgery, Department of Surgery, The Ottawa Hospital and University of Ottawa, Ottawa, Ontario, Canada
  2. 2School of Epidemiology and Public Health, Faculty of Medicine, University of Ottawa, Ottawa, Ontario, Canada
  3. 3Clinical Epidemiology Program, The Ottawa Hospital Research Institute, Ottawa, Ontario, Canada
  4. 4Faculty of Medicine, University of Ottawa, Ottawa, Ontario, Canada
  5. 5Department of Critical Care Medicine, University of Calgary, Calgary, Alberta, Canada
  6. 6Keele University, Keele, UK
  7. 7Anesthesiology and Pain Medicine, The Ottawa Hospital, Ottawa, Ontario, Canada
  8. 8Department of Medicine, The Ottawa Hospital and University of Ottawa, Ottawa, Ontario, Canada
  9. 9Learning Services, The Ottawa Hospital, Ottawa, Ontario, Canada
  1. Correspondence to Dr Derek J Roberts; Derek.Roberts01{at}gmail.com

Abstract

Introduction Overuse of cardiovascular healthcare services, defined as the provision of low-value (ineffective, harmful, cost-ineffective) tests, medications and procedures, may be common and associated with increased patient harm and health system inefficiencies and costs. We seek to systematically review the evidence for overuse of different cardiovascular healthcare services in high-income countries.

Methods and analysis We will search MEDLINE, EMBASE and Evidence-Based Medicine Reviews from 2010 onwards. Two investigators will independently review titles and abstracts and full-text studies. We will include published English-language studies conducted in high-income countries that enrolled adults (mean/median age ≥18 years) and reported the incidence or prevalence of overuse of cardiovascular tests, medications or procedures; adjusted risk factors for overuse; or adjusted associations between overuse and outcomes (reported estimates of morbidity, mortality, costs or lengths of hospital stay). Acceptable methods of defining low-value care will include literature review and multidisciplinary iterative panel processes, healthcare services with reproducible evidence of a lack of benefit or harm, or clinical practice guideline or Choosing Wisely recommendations. Two investigators will independently extract data and evaluate study risk of bias in duplicate. We will calculate summary estimates of the incidence and prevalence of overuse of different cardiovascular healthcare services across studies unstratified and stratified by country; method of defining low-value care; the percentage of included females, different races, and those with low and high socioeconomic status or cardiovascular risk; and study risks of bias using random-effects models. We will also calculate pooled estimates of adjusted risk factors for overuse and adjusted associations between overuse and outcomes overall and stratified by country using random-effects models. We will use the Grading of Recommendations, Assessment, Development and Evaluation to determine certainty in estimates.

Ethics and dissemination No ethics approval is required for this study as it deals with published data. Results will be presented at meetings and published in a peer-reviewed journal.

PROSPERO registration number CRD42021257490.

  • Cardiac Epidemiology
  • Cardiology
  • Vascular medicine
  • Vascular surgery
  • Cardiovascular imaging
  • Cardiothoracic surgery
http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

https://doi.org/10.1136/bmjopen-2021-053920

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Strengths and limitations of this study

    • We seek to systematically review the evidence for overuse of different cardiovascular healthcare services (tests, medications or procedures) in high-income countries from 2010 onwards.

    • Strengths of the study include our detailed prespecification of study methods in a study protocol; our comprehensive search strategy and inclusion of studies from all high-income countries (as defined by the World Bank) rather than just those located in North America; and our planned meta-analysis to create summary estimates of incidence and prevalence and risk factors for overuse of different cardiovascular healthcare services in high-income countries.

    • Potential limitations of the study include the exclusion of non-English language and grey literature studies; further, our study may not provide information on whether newer cardiovascular healthcare services are overused in high-income countries.

    Introduction

    Low-value care has been defined as ‘a healthcare service (procedure, test or medication) in which evidence suggests it confers no or very little benefit for patients, or risk of harm exceeds probable benefit, or more broadly, the added costs of the intervention do not provide proportional added benefits.’1 Overuse (ie, provision of low-value care) is a well-recognised problem in high-income countries with substantial consequences for healthcare systems.2–4 In North America, low-value care accounts for an estimated 30% or more of healthcare.5 6 Overuse exposes patients to unnecessary harms and consumes critical human and financial resources needed to provide high-value healthcare services. It also threatens the financial sustainability of healthcare systems,7 8 and potentially contributes to climate change through increased healthcare waste and greenhouse emissions.9–11

    Low-value cardiovascular healthcare services likely drive overuse in healthcare.12 In a systematic review examining overuse in the USA between 1978 and 2010, most used low-value services were cardiovascular.13 Estimated incidences of cardiovascular overuse varied from low (eg, coronary angiography for certain low-risk patients) to high (eg, carotid endarterectomy for certain high-risk patients), and many low-value services were persistently overused across the review period.13 Cardiovascular service overuse has been highlighted by the Choosing Wisely campaign since its inception.14 Among the first ‘do not do’ recommendations submitted by 25 American societies, 21% targeted cardiovascular testing.15 These recommendations were provided not only by cardiovascular societies, but also other non-cardiovascular societies, indicating that overuse of cardiovascular services may be pervasive across healthcare in high-income countries.15

    Although many of the cardiovascular services examined in the overuse literature are valuable when used appropriately, they are not when used for certain indications. For example, percutaneous coronary intervention (PCI), carotid endarterectomy and endovascular abdominal aortic aneurysm repair (EVAR) are all efficacious when appropriately indicated. However, PCI may be overused in stable coronary artery disease16 17 and carotid endarterectomy and EVAR may be overused in asymptomatic, high-risk patients with limited life expectancy.18–20 Cardiovascular diagnostic testing similarly provides useful and sometimes necessary clinical information, but its overuse can lead to a cascade of additional healthcare.21 For example, one study found that patients who received a low-value ECG were five times more likely to have further testing or appointments than patients who did not.22 The resulting cascade of healthcare involved specialist consultations, additional testing (eg, transthoracic echocardiograms and nuclear stress tests) and cardiac catheterisations.22

    Overuse of cardiovascular tests, procedures or medications may be costly and harmful to patients.23 Although the downstream effects of overuse are incompletely understood, frameworks suggest that they include direct medical costs (eg, prolonged hospitalisation), non-medical costs (eg, travel costs for unforeseen medical appointments) and indirect costs (eg, lost productivity).23 Overuse of cardiovascular procedures may lead to increased morbidity and mortality and unanticipated hospital admissions with an average length of stay of 2 days.18 In one study, 1 in 7 low-value EVARs, 1 in 13 low-value carotid endarterectomies, and 1 in 12 low-value renal artery angioplasties were associated with hospital-acquired complications, most commonly healthcare-associated infections.18

    Overuse is estimated to cost between $75.7 and $101.2 billion dollars annually in the USA,8 and cardiovascular testing in asymptomatic or low-risk patients may represent some of the highest cost low-value services.24 For certain overused cardiovascular services, such as low-value ECGs, the cost of the test is low, but the high rates of overuse contribute to substantial costs.24 Other cardiovascular services are more expensive, but have lower documented rates of overuse. For example, overuse of carotid artery disease screening (a low-cost test) in asymptomatic patients and PCI (a high-cost test) in those with stable coronary artery disease is estimated to cost $274 and $212 million dollars per year in the USA, respectively.19 Importantly, these estimates likely underestimate the financial burden of overuse on healthcare systems as they do not take into consideration the costs of the downstream effects of overuse.21 25

    Both indirect and direct methods have been used to measure overuse.26 Indirect measurement examines geographical variations in the use of healthcare services independent of differences in patient or population characteristics.2 For example, the finding of a ninefold variation in rates of elective PCI across countries suggests potential PCI overuse.27 The problem with indirect measurement, however, is that divergent rates of use do not necessarily represent inappropriate utilisation in the high-use areas.28 In comparison, direct measurement uses an evidence-informed definition of a low-value service to detect overuse.2

    This low-value care definition may be sourced from Choosing Wisely recommendations, clinical practice guidelines or outputs from expert consensus processes.2 Using this definition, researchers study overuse in administrative databases, insurance claims data or national registries.29 30 Using a Delphi approach, a 17-member expert panel judged that 11.6% of PCIs for non-acute indications in the US National Cardiovascular Data Registry were low value.17 Direct measurements are considered to be the most reliable indicator of overuse and can be used to guide de-implementation research and interventions.2

    Objectives

    The primary objective of this systematic review is to summarise reported direct estimates of the incidence and prevalence of overuse of different cardiovascular healthcare services (procedures, tests or medications) in adults who received care in high-income countries (as defined by the World Bank31) from 2010 to present. We will also determine whether overuse of these services varies by country; method of defining low-value care; the percentage of included females, different races, and those with low and high socioeconomic status or cardiovascular risk; and study risks of bias. The secondary objectives are to identify and summarise risk factors for overuse of different cardiovascular healthcare services and determine whether their overuse is associated with increased morbidity, mortality, costs or lengths of hospital stay. These data will be used to identify those low-value cardiovascular services that de-implementation interventions may have the greatest potential to improve quality of care, patient outcomes and healthcare spending.

    Methods and analysis

    Protocol, reporting and registration

    We prespecified our methods following recommendations for conducting systematic reviews and meta-analyses of incidence, prevalence and prognostic factor studies.32–34 The protocol is reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocols (PRISMA-P) statement (see online supplemental file 1 for the completed PRISMA-P checklist). It was registered on PROSPERO, the international prospective register of systematic reviews (registration number: CRD42021257490).35 36 The sponsor, the University of Ottawa, had no role in protocol development. The start date for the study was 1 July 2021 and the planned end date is 1 June 2022.

    Clinical questions

    We formulated study clinical questions using suggested frameworks for posing clinical questions for systematic reviews of incidence, prevalence and prognostic factor studies.32 33 37

    Primary clinical question

    • In adults (mean/median age ≥18 years) who received healthcare in a high-income country, what is the cumulative incidence, incidence rate, and point or period prevalence of overuse of different cardiovascular healthcare services overall and stratified by country; method of defining low-value care; the percentage of included females, different races, and those with low and high socioeconomic status or cardiovascular risk; and study risks of bias (as outlined below)?

    Secondary clinical questions

    1. In adults (mean/median age ≥18 years) who received healthcare in a high-income country, which factors increase the adjusted odds of overuse of different cardiovascular healthcare services?

    2. In adults (mean/median age ≥18 years) who received healthcare in a high-income country, is overuse of a certain cardiovascular healthcare service associated with an increased adjusted mortality, morbidity, cost and length of hospital stay?

    Definitions

    A high-income country will be defined according to the World Bank as a country with a gross national income per capita >US$12 696 in the 2022 fiscal year, calculated using the World Bank Atlas Method (see table 1 for a list of these countries).31 We will define overuse as a direct measurement of the provision of low-value care. Low-value care will be defined as a healthcare service (procedure, test or medication) ‘in which evidence suggests it confers no or very little benefit for a particular patient population, or risk of harm exceeds probable benefit, or more broadly, the added costs of the intervention do not provide proportional added benefits.’1 Acceptable direct methods of defining low-value care will include: (1) recommendations from literature review and multidisciplinary iterative panel processes (eg, the RAND/University of California at Los Angeles Appropriateness Rating Method38), (2) healthcare services with reproducible evidence of a lack of benefit or harm (with supporting literature supplied by study authors), or (3) clinical practice guideline or Choosing Wisely recommendations from the study’s country or region of origin.2 13 Cardiovascular healthcare services will be defined as cardiac or extracranial peripheral vascular (ie, aortic, extracranial carotid or other peripheral arterial) tests, medications or procedures. Tests may include laboratory tests, ECGs or diagnostic imaging tests, while procedures may include interventional (eg, PCI) or surgical (eg, carotid endarterectomy) procedures.

    View this table:
    Table 1

    High-income countries as defined by the World Bank31

    Information sources

    We will search MEDLINE; EMBASE; and the databases contained within Evidence-Based Medicine Reviews (ACP Journal Club; the Cochrane Central Register of Controlled Trials, Database of Systematic Reviews, and Methodology Register Database; Database of Abstracts of Reviews of Effects; Health Technology Assessment Database; and National Health Service Economic Evaluation Database) from 1 January 2010 without restrictions. We will begin our search in 2010 because recommendations regarding low-value care may have changed considerably over a time period beyond a decade in length. To identify additional citations, we will use the PubMed ‘related articles’ feature, and manually search bibliographies of included studies and relevant systematic and narrative review articles identified during the search. As we anticipate that the systematic review will identify a large number of peer-reviewed published studies, we will not search or include grey literature as this may challenge the feasibility of completing the evidence synthesis.

    Search strategy

    With the assistance of an information-scientist/medical librarian (RS), we created the initial MEDLINE and EMBASE search strategies. We first extracted Medical Subject Heading (MeSH) terms used to index studies known to meet inclusion criteria for our systematic review as well as those used in a previous systematic review on overuse of healthcare services in the USA.13 We then used those terms to search for additional relevant studies in PubMed and extracted MeSH terms those studies were indexed under.13 This process was repeated until we did not find any additional relevant MeSH indexing terms.13 Finally, we searched for Emtree terms that were similar to the above MeSH terms in EMBASE and created a list of non-MeSH/non-Emtree keywords for cardiovascular healthcare services and overuse. Using a combination of these MeSH/Emtree terms and keywords, we then constructed search filters covering the search themes cardiovascular healthcare services and overuse. After the MEDLINE search strategy was created, we submitted it to another information-scientist/medical librarian to peer review it using the Peer-Review of Electronic Search Strategies (PRESS) guideline39 (see box 1 for our PRESS’d MEDLINE search strategy).

    Box 1

    PRESS’d MEDLINE search strategy

    1. exp Health Services Misuse/

    2. ((overuse* or unnecessary or ineffective or overtreat* or overdiagnos* or overutilis* or overutiliz* or low-value or or wasteful* or appropriateness or inappropriate) adj10 (test* or imaging or procedure* or drug* or medication* or therapy or service* or intervention*)).tw,kf.

    3. (overuse* or unnecessary or ineffective or overtreat* or overdiagnos* or overutilis* or overutiliz* or low-value or wasteful* or appropriateness or inappropriate).tw,kf. and (“in data review” or in process or publisher or “pubmed not medline”).st.

    4. choos* wisely.mp.

    5. or/1–4

    6. exp Cardiovascular Surgical Procedures/

    7. exp heart diseases/di, dg, su, th or exp vascular disease/di, dg, su, th

    8. ((cardiovascular or cardiac or vascular or heart) adj2 disease*).ti.

    9. exp Cardiac Imaging Techniques/ or exp Diagnostic Techniques, Cardiovascular/

    10. (angiogra* or echocardiogra* or electrocardiogra* or myocardial perfusion imaging or duplex ultrasound or lower extremity arterial duplex ultrasound or carotid duplex ultrasound).kf,tw.

    11. (percutaneous coronary intervention* or cabg or coronary arter* bypass or renal arter* angioplast* or carotid endarterectom* or endovascular abdominal aortic aneurysm repair* or peripheral arter* bypass or coronary revasculari?ation*).tw,kf.

    12. ((cardiac or cardiovascular) adj2 (test* or imaging or procedure* or drug* or medication* or therapy or service* or intervention*)).tw,kf.

    13. exp Troponin/bl

    14. (troponin adj2 test*).tw,kf.

    15. Defibrillators, Implantable/

    16. or/6–15

    17. 5 and 16

    18. (overuse* or unnecessary or ineffective or overtreat* or overdiagnos* or overutilis* or overutiliz* or low-value or wasteful* or appropriateness or inappropriate).tw,kf. adj10 (angiogra* or echocardiogra* or electrocardiogramECG* or myocardial perfusion imag* or duplex ultrasound or lower extremity arterial duplex ultrasound or carotid duplex ultrasound).kf,tw.

    19. (overuse* or unnecessary or ineffective or overtreat* or overdiagnos* or overutilis* or overutiliz* or low-value or wasteful* or appropriateness or inappropriate).tw,kf. adj10 (percutaneous coronary intervention* or cabg or coronary arter* bypass or renal arter* angioplast* or carotid endarterectom* or endovascular abdominal aortic aneurysm repair* or peripheral arter* bypass or coronary revasculari?ation* or implant* cardioverter* or implant* defibrillator*).kf,tw.

    20. (overuse* or unnecessary or ineffective or overtreat* or overdiagnos* or overutilis* or overutiliz* or low-value or wasteful* or appropriateness or inappropriate).tw,kf. adj10 (cardiac test* or cardiovascular test* or cardiac imaging or cardiovascular procedure* or cardiovascular drug* or cardiovascular medication*).kf,tw.

    21. or/17–20

    22. exp animals/ not humans/

    23. 21 not 22

    24. limit 23 to yr=“2010 -Current”

    25. limit 24 to english language

    PRESS, Peer-Review of Electronic Search Strategies.

    Data management and selection process

    The titles and abstracts of citations identified during the search will be imported into Covidence (https://www.covidence.org/). Thereafter, two investigators will use Covidence to remove duplicates and independently review the titles and abstracts of articles identified by the search and select any article deemed potentially relevant by either investigator for full-text review. Finally, two investigators will review the full-text of all potentially relevant citations and select studies for inclusion in the systematic review. Disagreements regarding study inclusion will be resolved via consensus or arbitration by a third investigator. Chance-corrected agreement between investigators regarding full-text article inclusion will be calculated using a kappa statistic.40

    Eligibility criteria and outcomes

    We will use the following study inclusion criteria33 37:

    • The study included adults (mean/median age ≥18 years) who received healthcare in a high-income country (as defined by the World Bank31).

    • The study reported one or more of the following outcomes (or these outcomes could be calculated from the data provided):

      • Cumulative incidence, incidence rate or point or period prevalence of overuse of a cardiovascular healthcare service or services.41

      • Odds ratios (ORs), risk ratios (RRs) or hazard ratios (HRs) (and surrounding standard errors (SEs) or 95% confidence intervals (CIs)) adjusted for the presence of other risk factors or confounding variables and relating one or more potential risk factor of interest to overuse of a cardiovascular healthcare service or services.

      • ORs, RRs, HRs or other measures (and surrounding SEs or 95% CIs) describing differences in mortality, morbidity, costs or lengths of hospital stay associated with overuse of a certain cardiovascular healthcare service or services instead of no overuse and adjusted for the presence of other risk factors or confounding factors.

    • The study design was observational (ie, cohort, case–control or cross-sectional41 42) and used an acceptable method of defining low-value care as outlined above.13 We will also include studies of interventions to decrease overuse if sufficient data on overuse were reported for the control group.13

    Disagreements between investigators as to whether a healthcare service is cardiovascular or not will be resolved via consensus or arbitration by a third investigator. We will exclude studies: (1) published in non-English languages; (2) published only as an abstract; (3) published before the year 2010 or those that only enrolled patients before 2010; (4) that reported only unadjusted risk factors for overuse or unadjusted associations between overuse and outcomes; or (5) that measured inefficient care (eg, use of a brand name instead of equivalent generic drug).13

    Data items and collection process

    Once we have established consensus on which studies should be included in the systematic review, two investigators will independently extract data in duplicate using a predesigned electronic data extraction spreadsheet. This spreadsheet will be piloted on a representative sample of five included studies. We will extract the following data from included studies: (1) design, data source and setting of the study (country, region and rural vs urban13); (2) patient recruitment period; (3) the direct method of defining overuse of different cardiovascular healthcare services (eg, clinical practice guideline) and the year it was published (where relevant); (4) sample size; (5) included patient characteristics (for descriptive purposes), including percentages of patient sex, race and socioeconomic status and patients with coronary, cerebrovascular and peripheral artery disease; pulmonary disease; diabetes; chronic kidney disease; cancer; and a smoking history; (6) reported estimates of the included patients’ risk of cardiovascular events or their estimated life expectancy (as reported by the authors); (7) reported cumulative incidences, incidence rates and point or period prevalences of overuse of cardiovascular healthcare services; (8) reported adjusted risk factors for overuse of cardiovascular healthcare services (and their surrounding 95% CIs); (9) reported adjusted associations between overuse of cardiovascular healthcare services and estimates of morbidity, mortality, costs and lengths of hospital stay (and their surrounding 95% CIs or standard deviations (SDs)); and (10) which other prognostic or confounding factors were adjusted for when evaluating associations between potential risk factors and overuse or between overuse and morbidity, mortality, costs, and lengths of hospital stay. Three investigators will independently extract data when they are only presented visually (eg, within a bar graph) and then their results will be averaged.13 If more than one method of defining low-value care or overuse was used, we will extract each of the estimates and report and analyse them separately.13

    Risk of bias assessment

    Two investigators will independently evaluate the risk of bias of studies reporting incidence and prevalence estimates using the Joanna Briggs Institute’s critical appraisal checklist of studies reporting prevalence data.32 The Joanna Briggs checklist includes questions about whether the sample frame was appropriate to address the target population, participants were sampled in an appropriate way, sample size was adequate, study subjects and setting were described in detail, the data analysis was conducted with sufficient coverage of the identified sample, valid methods were used for the identification of the condition, the condition was measured in a standard and reliable way and statistical analyses were appropriate.32 Those studies that reported risk factors for overuse or associations between overuse and outcomes will also be independently evaluated by two investigators using the Quality in Prognosis Studies tool.43 44 This tool includes questions regarding study participation and attrition; potential risk factor and outcome description and measurement; confounding measurement and account; and methods and reporting of statistical analyses.43 44 For those studies that used administrative data, we will also examine whether the study authors considered the accuracy (sensitivity and specificity) of the codes used to define variables. Disagreements regarding risk of bias assessments will be resolved by consensus.

    Qualitative data synthesis

    We will perform a narrative synthesis of the included studies and their reported data before considering meta-analyses.45 We will first tabulate characteristics of the included studies, including their design; data source; setting; recruitment period; included patients; studied low-value cardiovascular healthcare service or services; direct method of defining overuse of each service; reported incidences, incidence rates, and point or period prevalences of overuse of different services; and reported risk factors for overuse of different services or associations between overuse and estimates of morbidity, mortality, costs and lengths of stay. This tabulation will allow us to cluster the low-value cardiovascular healthcare services by test, medication or procedure and identify potentially duplicate data. It will also allow us to determine if any definitions of low-value care may have changed over the 10-year study period and to identify which studies provided similar enough information be included in meta-analyses.

    Quantitative data synthesis and statistical analyses

    Where it was not reported, we will calculate the cumulative incidence, incidence rate and point or period prevalence of overuse of different cardiovascular healthcare services. Cumulative incidence will be calculated using the following formula:

    Embedded Image

    where the total population at risk will be defined as the number of adults who have not been exposed to the low-value cardiovascular healthcare service. Incidence rate will be determined using the formula:

    Embedded Image

    Point or period prevalence will be determined using the formula:

    Embedded Image

    The SE and 95% CI of these proportions will then be determined using the Clopper-Pearson exact binomial method.

    Where we identify more than one study that provided estimates of overuse of the same low-value healthcare practice, incidence or prevalence estimates for overuse of that practice will be pooled using DerSimonian and Laird random-effects models.46 As suggested by Barendregt et al, we will transform these proportional estimates using a double arcsine transformation prior to meta-analyses.32 47 The data will then be back-transformed to incidence and prevalence estimates after meta-analysis.32

    We will use the OR (for dichotomous outcomes) or standardised mean difference (for continuous outcomes) as the summary measures of choice for pooled risk factor and outcome analyses. Similar adjusted risk factor estimates and outcome associations will be pooled using DerSimonian and Laird random-effects models.46 Where the OR was not reported, we will pool log-transformed RRs or HRs instead. When adjusted estimates were calculated from the same data source across several studies, we will include the estimate derived from the largest study. As a sensitivity analyses, we will also recalculate the estimate using that derived from the other, smaller studies as studies may have variably adjusted their estimates for potentially confounding factors.

    We will inspect forest plots, calculate I2 inconsistency statistics and conduct tests of homogeneity (p<0.10 considered significant given the low power of these tests) to assess for between-study heterogeneity in the above estimates.48–50 We will consider I2 statistics >25%, >50% and >75% to represent low, moderate and high degrees of heterogeneity, respectively.49 In the presence of at least low between-study heterogeneity in our pooled estimates of incidence and prevalence, we will conduct subgroup meta-analyses and meta-regression using DerSimonian and Laird random-effects models. We will use the following predictor variables in these stratified meta-analyses and meta-regressions: (1) country; (2) direct method of defining low-value care; (3) whether the study was population-based versus not; (4) the percentage of included females, different races, and those with low and high socioeconomic status or low and high cardiovascular risk (as defined by the authors); and (5) whether there was a high or lower risk of bias relating to measuring overuse. Where different countries disagree regarding the definition of low-value care, we will allow the local setting to make the decision about what is low value and only report summary estimates of low-value care per country rather than across all studies. We will also report stratified analyses within country-level estimates rather than across all studies.

    We will evaluate for evidence of small study effects potentially due to publication bias by visually inspecting funnel plots of incidence and prevalence of overuse and using Begg’s and Egger’s tests (p<0.05 considered significant).51 We will use the study sample size instead of the inverse of the SE on the y-axis as this may perform more favourably in these analyses.32 52 Statistical analyses will be performed by a PhD-trained meta-analyst using Stata MP V.13.1 (Stata Corporation, College Station, Texas, USA).

    Certainty in the cumulative evidence

    We will use the Grading of Recommendations, Assessment, Development and Evaluation for assessment of evidence about prognostic factors to determine the certainty in the estimates of association between the reported risk factors and overuse or between overuse and estimates of morbidity, mortality and costs.53 To do this, we will first assess the risk of bias, imprecision, inconsistency, indirectness and publication bias associated with the evidence for the reported risk factors.54–58 The overall certainty in these estimates will then be adjudicated as high (further evidence is high) (‘further research is very unlikely to change our confidence in the estimate’), moderate (‘further research is likely to have an important impact on our confidence in the estimate and may change the estimate’), low (‘further research is very likely to have an important impact on our confidence in the estimate and is likely to change the estimate’) or very low (‘very uncertain about the estimate’).36

    Patient and public involvement

    Patients and the public were not involved in the creation of the protocol. Patients will be involved in subsequent studies examining barriers and facilitators to reducing overuse of cardiovascular healthcare.

    Ethics and dissemination

    No ethics approval is required for this study as it deals with published data. Results of the systematic review will be presented at cardiovascular and other relevant meetings and published in a peer-reviewed journal. Potential limitations of the study include the exclusion of non-English language and grey literature studies. Further, our study may not provide information on whether newer cardiovascular healthcare services are overused in high-income countries. Outputs of the study will include summary estimates of the incidence and prevalence of overuse of different cardiovascular healthcare services in high-income countries, risk factors for their overuse and pooled estimates of the burden of overuse of different cardiovascular healthcare services on patients and healthcare systems. Through the involvement of several international stakeholders in cardiovascular medicine and surgery and de-implementation science, our results will hopefully be of immediate interest and use to guide discussion regarding prioritisation of services.

    De-implementation of healthcare services should begin with the identification and prioritisation of low-value services.59 60 Perhaps the most well-known example of this is the Choosing Wisely campaign, where lists of low-value services are published by individual societies that are disseminated to healthcare providers (eg, in academic journals) and the public (eg, public-facing campaigns).61 62 However, evidence suggests that simply identifying a low-value service has little impact on overuse overall.63–66 Prioritisation of specific low-value services by systematically identifying factors such as rates of overuse and the associated economic burden and risk of harm to patients and healthcare systems (as in this study) may help focus de-implementation efforts.

    Systematic review methodology may be used to synthesise evidence on cardiovascular healthcare service overuse that could be used to help guide prioritisation decisions. A systematic review of overuse in the USA across all areas of medicine was published in 2012.13 This study included a narrative description of the published evidence for overuse up until 2010. Subsequent narrative updates have been published annually from 2016 to 2019.4 67–69 These reviews provide valuable narrative syntheses of the rapidly emerging evidence on medical overuse. However, because they examined all studies of medical overuse, they did not provide detailed information on each of the included studies and their findings. They also included studies published across variable time periods with different and important risks of bias. Finally, they did not provide data from other high-income countries or summary estimates of the frequency of overuse across the included studies. Therefore, these evidence syntheses are likely not able to help with prioritisation decisions regarding overuse of cardiovascular healthcare services.

    The outputs of this systematic review will be used to inform a research programme that aims to reduce overuse of cardiovascular healthcare services. Subsequent steps will include cohort studies designed to determine whether the priority low-value cardiovascular healthcare services identified in the systematic review are those that are also overused in Canada. This will be followed by an interview-based study to determine barriers and facilitators to overuse and de-implementation of low-value cardiovascular Canadian healthcare services. We will use this information to design de-implementation randomised controlled trials that aim to reduce overuse of priority low-value cardiovascular healthcare services across different Canadian regions. These studies may serve as a template to inform similar programmes of work in other countries.

    Ethics statements

    Patient consent for publication

    References

    1. ↵
      1. Elshaug AG,
      2. Rosenthal MB,
      3. Lavis JN, et al
      . Levers for addressing medical underuse and overuse: achieving high-value health care. Lancet 2017;390:191–202.doi:10.1016/S0140-6736(16)32586-7pmid:http://www.ncbi.nlm.nih.gov/pubmed/28077228
      OpenUrlCrossRefPubMed
    2. ↵
      1. Brownlee S,
      2. Chalkidou K,
      3. Doust J, et al
      . Evidence for overuse of medical services around the world. The Lancet 2017;390:156–68.doi:10.1016/S0140-6736(16)32585-5pmid:http://www.ncbi.nlm.nih.gov/pubmed/28077234
      OpenUrlPubMed
    3. ↵
      1. Elshaug AG,
      2. Watt AM,
      3. Mundy L, et al
      . Over 150 potentially low-value health care practices: an Australian study. Med J Aust 2012;197:556–60.doi:10.5694/mja12.11083pmid:http://www.ncbi.nlm.nih.gov/pubmed/23163685
      OpenUrlCrossRefPubMedWeb of Science
    4. ↵
      1. Morgan DJ,
      2. Dhruva SS,
      3. Coon ER, et al
      . 2018 update on medical overuse. JAMA Intern Med 2019;179:240–6.doi:10.1001/jamainternmed.2018.5748
      OpenUrl
    5. ↵
      1. Information CIfH
      . Unnecessary care in Canada, 2018. Available: http://www.cihi.cacopyright@cihi.caISBN978-1-77109-569-3 [Accessed October 3, 2018].
    6. ↵
      1. Io M
      . The healthcare imperative: lowering costs and improving outcomes: workshop series summary. Washington, DC: The National Academies Press, 2010.
    7. ↵
      1. Berwick DM
      . Avoiding overuse-the next quality frontier. Lancet 2017;390:102–4.doi:10.1016/S0140-6736(16)32570-3pmid:http://www.ncbi.nlm.nih.gov/pubmed/28077229
      OpenUrlCrossRefPubMed
    8. ↵
      1. Shrank WH,
      2. Rogstad TL,
      3. Parekh N
      . Waste in the US health care system: estimated costs and potential for savings. JAMA 2019;322:1501–9.doi:10.1001/jama.2019.13978pmid:http://www.ncbi.nlm.nih.gov/pubmed/31589283
      OpenUrlCrossRefPubMed
    9. ↵
      1. Eckelman MJ,
      2. Sherman JD,
      3. MacNeill AJ
      . Life cycle environmental emissions and health damages from the Canadian healthcare system: an economic-environmental-epidemiological analysis. PLoS Med 2018;15:e1002623.doi:10.1371/journal.pmed.1002623pmid:http://www.ncbi.nlm.nih.gov/pubmed/30063712
      OpenUrlPubMed
    10. ↵
      1. Mercer C
      . How health care contributes to climate change. CMAJ 2019;191:E403–4.doi:10.1503/cmaj.109-5722pmid:http://www.ncbi.nlm.nih.gov/pubmed/30962202
      OpenUrlFREE Full Text
    11. ↵
      1. Watts N,
      2. Amann M,
      3. Arnell N, et al
      . The 2019 report of the Lancet countdown on health and climate change: ensuring that the health of a child born today is not defined by a changing climate. Lancet 2019;394:1836–78.doi:10.1016/S0140-6736(19)32596-6pmid:http://www.ncbi.nlm.nih.gov/pubmed/31733928
      OpenUrlCrossRefPubMed
    12. ↵
      1. Huang X,
      2. Rosenthal MB
      . Overuse of cardiovascular services: evidence, causes, and opportunities for reform. Circulation 2015;132:205–14.doi:10.1161/CIRCULATIONAHA.114.012668pmid:http://www.ncbi.nlm.nih.gov/pubmed/26195488
      OpenUrlFREE Full Text
    13. ↵
      1. Korenstein D,
      2. Falk R,
      3. Howell EA, et al
      . Overuse of health care services in the United States: an understudied problem. Arch Intern Med 2012;172:171–8.doi:10.1001/archinternmed.2011.772pmid:http://www.ncbi.nlm.nih.gov/pubmed/22271125
      OpenUrlCrossRefPubMedWeb of Science
    14. ↵
      1. Beller GA
      . Tests that may be overused or misused in cardiology: the choosing wisely campaign. J Nucl Cardiol 2012;19:401–3.doi:10.1007/s12350-012-9569-ypmid:http://www.ncbi.nlm.nih.gov/pubmed/22547398
      OpenUrlCrossRefPubMed
    15. ↵
      1. Morden NE,
      2. Colla CH,
      3. Sequist TD, et al
      . Choosing wisely--the politics and economics of labeling low-value services. N Engl J Med 2014;370:589–92.doi:10.1056/NEJMp1314965pmid:http://www.ncbi.nlm.nih.gov/pubmed/24450859
      OpenUrlCrossRefPubMedWeb of Science
    16. ↵
      1. Ko DT,
      2. Guo H,
      3. Wijeysundera HC, et al
      . Assessing the association of appropriateness of coronary revascularization and clinical outcomes for patients with stable coronary artery disease. J Am Coll Cardiol 2012;60:1876–84.doi:10.1016/j.jacc.2012.06.056pmid:http://www.ncbi.nlm.nih.gov/pubmed/23062534
      OpenUrlFREE Full Text
    17. ↵
      1. Chan PS,
      2. Patel MR,
      3. Klein LW, et al
      . Appropriateness of percutaneous coronary intervention. JAMA 2011;306:53–61.doi:10.1001/jama.2011.916
      OpenUrlCrossRefPubMedWeb of Science
    18. ↵
      1. Badgery-Parker T,
      2. Pearson S-A,
      3. Dunn S, et al
      . Measuring hospital-acquired complications associated with low-value care. JAMA Intern Med 2019;179:499–505.doi:10.1001/jamainternmed.2018.7464pmid:http://www.ncbi.nlm.nih.gov/pubmed/30801628
      OpenUrlPubMed
    19. ↵
      1. Schwartz AL,
      2. Landon BE,
      3. Elshaug AG, et al
      . Measuring low-value care in medicare. JAMA Intern Med 2014;174:1067–76.doi:10.1001/jamainternmed.2014.1541pmid:http://www.ncbi.nlm.nih.gov/pubmed/24819824
      OpenUrlPubMed
    20. ↵
      1. Goldstein LB,
      2. Bushnell CD,
      3. Adams RJ, et al
      . Guidelines for the primary prevention of stroke: a guideline for healthcare professionals from the American heart association/American stroke association. Stroke 2011;42:517–84.doi:10.1161/STR.0b013e3181fcb238pmid:http://www.ncbi.nlm.nih.gov/pubmed/21127304
      OpenUrlAbstract/FREE Full Text
    21. ↵
      1. Ganguli I,
      2. Lupo C,
      3. Mainor AJ, et al
      . Prevalence and cost of care cascades after low-value preoperative electrocardiogram for cataract surgery in fee-for-service Medicare beneficiaries. JAMA Intern Med 2019;179:1211–9.doi:10.1001/jamainternmed.2019.1739pmid:http://www.ncbi.nlm.nih.gov/pubmed/31158270
      OpenUrlPubMed
    22. ↵
      1. Bhatia RS,
      2. Bouck Z,
      3. Ivers NM, et al
      . Electrocardiograms in low-risk patients undergoing an annual health examination. JAMA Intern Med 2017;177:1326–33.doi:10.1001/jamainternmed.2017.2649pmid:http://www.ncbi.nlm.nih.gov/pubmed/28692719
      OpenUrlPubMed
    23. ↵
      1. Korenstein D,
      2. Chimonas S,
      3. Barrow B, et al
      . Development of a conceptual map of negative consequences for patients of overuse of medical tests and treatments. JAMA Intern Med 2018;178:1401–7.doi:10.1001/jamainternmed.2018.3573pmid:http://www.ncbi.nlm.nih.gov/pubmed/30105371
      OpenUrlPubMed
    24. ↵
      1. Mafi JN,
      2. Russell K,
      3. Bortz BA, et al
      . Low-cost, high-volume health services contribute the most to unnecessary health spending. Health Aff 2017;36:1701–4.doi:10.1377/hlthaff.2017.0385pmid:http://www.ncbi.nlm.nih.gov/pubmed/28971913
      OpenUrlAbstract/FREE Full Text
    25. ↵
      1. Mandrola J,
      2. Morgan DJ
      . The important but rarely studied cascade of care. JAMA Netw Open 2019;2:e1913315.doi:10.1001/jamanetworkopen.2019.13315pmid:http://www.ncbi.nlm.nih.gov/pubmed/31617920
      OpenUrlPubMed
    26. ↵
      1. Miller G,
      2. Rhyan C,
      3. Beaudin-Seiler B, et al
      . A framework for measuring low-value care. Value Health 2018;21:375–9.doi:10.1016/j.jval.2017.10.017pmid:http://www.ncbi.nlm.nih.gov/pubmed/29680091
      OpenUrlPubMed
    27. ↵
      1. Corallo AN,
      2. Croxford R,
      3. Goodman DC, et al
      . A systematic review of medical practice variation in OECD countries. Health Policy 2014;114:5–14.doi:10.1016/j.healthpol.2013.08.002pmid:http://www.ncbi.nlm.nih.gov/pubmed/24054709
      OpenUrlCrossRefPubMedWeb of Science
    28. ↵
      1. van Walraven CV,
      2. Paterson JM,
      3. Kapral M, et al
      . Appropriateness of primary total hip and knee replacements in regions of Ontario with high and low utilization rates. CMAJ 1996;155:697–706.pmid:http://www.ncbi.nlm.nih.gov/pubmed/8823215
      OpenUrlAbstract
    29. ↵
      1. Chalmers K,
      2. Badgery-Parker T,
      3. Pearson S-A, et al
      . Developing indicators for measuring low-value care: mapping choosing wisely recommendations to hospital data. BMC Res Notes 2018;11:163.doi:10.1186/s13104-018-3270-4pmid:http://www.ncbi.nlm.nih.gov/pubmed/29506573
      OpenUrlPubMed
    30. ↵
      1. Bhatia RS,
      2. Levinson W,
      3. 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.doi:10.1136/bmjqs-2015-004070pmid:http://www.ncbi.nlm.nih.gov/pubmed/26092165
      OpenUrlAbstract/FREE Full Text
    31. ↵
      1. The World Bank
      . World bank country and lending groups: country classification, 2022. Available: https://datahelpdesk.worldbank.org/knowledgebase/articles/906519 [Accessed December 15, 2021].
    32. ↵
      1. Munn Z,
      2. Moola S,
      3. Lisy K, et al
      . Methodological guidance for systematic reviews of observational epidemiological studies reporting prevalence and cumulative incidence data. Int J Evid Based Healthc 2015;13:147–53.doi:10.1097/XEB.0000000000000054pmid:http://www.ncbi.nlm.nih.gov/pubmed/26317388
      OpenUrlCrossRefPubMed
    33. ↵
      1. Riley RD,
      2. Moons KGM,
      3. Snell KIE, et al
      . A guide to systematic review and meta-analysis of prognostic factor studies. BMJ 2019;364:k4597.doi:10.1136/bmj.k4597pmid:http://www.ncbi.nlm.nih.gov/pubmed/30700442
      OpenUrlFREE Full Text
    34. ↵
      1. Mueller M,
      2. D'Addario M,
      3. Egger M, et al
      . Methods to systematically review and meta-analyse observational studies: a systematic scoping review of recommendations. BMC Med Res Methodol 2018;18:44.doi:10.1186/s12874-018-0495-9pmid:http://www.ncbi.nlm.nih.gov/pubmed/29783954
      OpenUrlCrossRefPubMed
    35. ↵
      1. Moher D,
      2. Shamseer L,
      3. Clarke M, et al
      . Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Syst Rev 2015;4:1.doi:10.1186/2046-4053-4-1pmid:http://www.ncbi.nlm.nih.gov/pubmed/25554246
      OpenUrlCrossRefPubMed
    36. ↵
      1. Shamseer L,
      2. Moher D,
      3. Clarke M, et al
      . Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015: elaboration and explanation. BMJ 2015;350:g7647.doi:10.1136/bmj.g7647pmid:http://www.ncbi.nlm.nih.gov/pubmed/25555855
      OpenUrlCrossRefPubMed
    37. ↵
      1. Moons KGM,
      2. de Groot JAH,
      3. Bouwmeester W, et al
      . Critical appraisal and data extraction for systematic reviews of prediction modelling studies: the charms checklist. PLoS Med 2014;11:e1001744.doi:10.1371/journal.pmed.1001744pmid:http://www.ncbi.nlm.nih.gov/pubmed/25314315
      OpenUrlCrossRefPubMed
    38. ↵
      The RAND/UCLA appropriateness method user’s manual. Available: http://www.rand.org/pubs/monograph_reports/MR1269.html [Accessed 1 Jan 2014].
    39. ↵
      1. McGowan J,
      2. Sampson M,
      3. Salzwedel DM, et al
      . PRESS peer review of electronic search strategies: 2015 guideline statement. J Clin Epidemiol 2016;75:40–6.doi:10.1016/j.jclinepi.2016.01.021pmid:http://www.ncbi.nlm.nih.gov/pubmed/27005575
      OpenUrlCrossRefPubMed
    40. ↵
      1. Landis JR,
      2. Koch GG
      . The measurement of observer agreement for categorical data. Biometrics 1977;33:159–74.doi:10.2307/2529310pmid:http://www.ncbi.nlm.nih.gov/pubmed/843571
      OpenUrlCrossRefPubMedWeb of Science
    41. ↵
      A dictionary of epidemiology. 5th Edition. Oxford, UK: Oxford University Press, 2008.
    42. ↵
      1. Dekkers OM,
      2. Egger M,
      3. Altman DG, et al
      . Distinguishing case series from cohort studies. Ann Intern Med 2012;156:37–40.doi:10.7326/0003-4819-156-1-201201030-00006pmid:http://www.ncbi.nlm.nih.gov/pubmed/22213493
      OpenUrlCrossRefPubMed
    43. ↵
      1. Hayden JA,
      2. Côté P,
      3. Bombardier C
      . Evaluation of the quality of prognosis studies in systematic reviews. Ann Intern Med 2006;144:427–37.doi:10.7326/0003-4819-144-6-200603210-00010pmid:http://www.ncbi.nlm.nih.gov/pubmed/16549855
      OpenUrlCrossRefPubMedWeb of Science
    44. ↵
      1. Hayden JA,
      2. van der Windt DA,
      3. Cartwright JL, et al
      . Assessing bias in studies of prognostic factors. Ann Intern Med 2013;158:280–6.doi:10.7326/0003-4819-158-4-201302190-00009pmid:http://www.ncbi.nlm.nih.gov/pubmed/23420236
      OpenUrlCrossRefPubMedWeb of Science
    45. ↵
      1. Popay J,
      2. Roberts H,
      3. Sowden A
      . Guidance on the conduct of narrative synthesis in systematic reviews. Lancaster, PA: ESRC Methods Programme, 2006.
    46. ↵
      1. DerSimonian R,
      2. Laird N
      . Meta-Analysis in clinical trials. Control Clin Trials 1986;7:177–88.doi:10.1016/0197-2456(86)90046-2pmid:http://www.ncbi.nlm.nih.gov/pubmed/3802833
      OpenUrlCrossRefPubMedWeb of Science
    47. ↵
      1. Barendregt JJ,
      2. Doi SA,
      3. Lee YY, et al
      . Meta-analysis of prevalence. J Epidemiol Community Health 2013;67:974–8.doi:10.1136/jech-2013-203104pmid:http://www.ncbi.nlm.nih.gov/pubmed/23963506
      OpenUrlAbstract/FREE Full Text
    48. ↵
      1. Higgins JPT,
      2. Thompson SG
      . Quantifying heterogeneity in a meta-analysis. Stat Med 2002;21:1539–58.doi:10.1002/sim.1186pmid:http://www.ncbi.nlm.nih.gov/pubmed/12111919
      OpenUrlCrossRefPubMedWeb of Science
    49. ↵
      1. Higgins JPT,
      2. Thompson SG,
      3. Deeks JJ, et al
      . Measuring inconsistency in meta-analyses. BMJ 2003;327:557–60.doi:10.1136/bmj.327.7414.557pmid:http://www.ncbi.nlm.nih.gov/pubmed/12958120
      OpenUrlFREE Full Text
    50. ↵
      1. Systematic Reviews in Health Care
      . Meta-analysis in context. London, U.K.: BMJ Publishing Group, 2001.
    51. ↵
      1. Begg CB,
      2. Mazumdar M
      . Operating characteristics of a RANK correlation test for publication bias. Biometrics 1994;50:1088–101.doi:10.2307/2533446pmid:http://www.ncbi.nlm.nih.gov/pubmed/7786990
      OpenUrlCrossRefPubMedWeb of Science
    52. ↵
      1. Hunter JP,
      2. Saratzis A,
      3. Sutton AJ, et al
      . In meta-analyses of proportion studies, funnel plots were found to be an inaccurate method of assessing publication bias. J Clin Epidemiol 2014;67:897–903.doi:10.1016/j.jclinepi.2014.03.003pmid:http://www.ncbi.nlm.nih.gov/pubmed/24794697
      OpenUrlCrossRefPubMed
    53. ↵
      1. Foroutan F,
      2. Guyatt G,
      3. Zuk V, et al
      . Grade guidelines 28: use of grade for the assessment of evidence about prognostic factors: rating certainty in identification of groups of patients with different absolute risks. J Clin Epidemiol 2020;121:62–70.doi:10.1016/j.jclinepi.2019.12.023pmid:http://www.ncbi.nlm.nih.gov/pubmed/31982539
      OpenUrlCrossRefPubMed
    54. ↵
      1. Guyatt GH,
      2. Oxman AD,
      3. Vist G, et al
      . GRADE guidelines: 4. Rating the quality of evidence--study limitations (risk of bias). J Clin Epidemiol 2011;64:407–15.doi:10.1016/j.jclinepi.2010.07.017pmid:http://www.ncbi.nlm.nih.gov/pubmed/21247734
      OpenUrlCrossRefPubMedWeb of Science
    55. ↵
      1. Guyatt GH,
      2. Oxman AD,
      3. Montori V, et al
      . GRADE guidelines: 5. Rating the quality of evidence--publication bias. J Clin Epidemiol 2011;64:1277–82.doi:10.1016/j.jclinepi.2011.01.011pmid:http://www.ncbi.nlm.nih.gov/pubmed/21802904
      OpenUrlCrossRefPubMed
    56. ↵
      1. Guyatt GH,
      2. Oxman AD,
      3. Kunz R, et al
      . GRADE guidelines 6. Rating the quality of evidence--imprecision. J Clin Epidemiol 2011;64:1283–93.doi:10.1016/j.jclinepi.2011.01.012pmid:http://www.ncbi.nlm.nih.gov/pubmed/21839614
      OpenUrlCrossRefPubMed
    57. ↵
      1. Guyatt GH,
      2. Oxman AD,
      3. Kunz R, et al
      . GRADE guidelines: 7. Rating the quality of evidence--inconsistency. J Clin Epidemiol 2011;64:1294–302.doi:10.1016/j.jclinepi.2011.03.017pmid:http://www.ncbi.nlm.nih.gov/pubmed/21803546
      OpenUrlCrossRefPubMed
    58. ↵
      1. Guyatt GH,
      2. Oxman AD,
      3. Kunz R, et al
      . GRADE guidelines: 8. Rating the quality of evidence--indirectness. J Clin Epidemiol 2011;64:1303–10.doi:10.1016/j.jclinepi.2011.04.014pmid:http://www.ncbi.nlm.nih.gov/pubmed/21802903
      OpenUrlCrossRefPubMed
    59. ↵
      1. Niven DJ,
      2. Mrklas KJ,
      3. Holodinsky JK, et al
      . Towards understanding the de-adoption of low-value clinical practices: a scoping review. BMC Med 2015;13:255.doi:10.1186/s12916-015-0488-zpmid:http://www.ncbi.nlm.nih.gov/pubmed/26444862
      OpenUrlCrossRefPubMed
    60. ↵
      1. Grimshaw JM,
      2. Patey AM,
      3. Kirkham KR, et al
      . De-implementing wisely: developing the evidence base to reduce low-value care. BMJ Qual Saf 2020;29:409–17.doi:10.1136/bmjqs-2019-010060pmid:http://www.ncbi.nlm.nih.gov/pubmed/32029572
      OpenUrlAbstract/FREE Full Text
    61. ↵
      1. Levinson W,
      2. Kallewaard M,
      3. Bhatia RS, et al
      . 'Choosing wisely': a growing international campaign. BMJ Qual Saf 2015;24:167–74.doi:10.1136/bmjqs-2014-003821pmid:http://www.ncbi.nlm.nih.gov/pubmed/25552584
      OpenUrlAbstract/FREE Full Text
    62. ↵
      1. Born KB,
      2. Coulter A,
      3. Han A, et al
      . Engaging patients and the public in choosing wisely. BMJ Qual Saf 2017;26:687–91.doi:10.1136/bmjqs-2017-006595pmid:http://www.ncbi.nlm.nih.gov/pubmed/28600453
      OpenUrlFREE Full Text
    63. ↵
      1. Rosenberg A,
      2. Agiro A,
      3. Gottlieb M, et al
      . Early trends among seven recommendations from the choosing wisely campaign. JAMA Intern Med 2015;175:1913–20.doi:10.1001/jamainternmed.2015.5441pmid:http://www.ncbi.nlm.nih.gov/pubmed/26457643
      OpenUrlPubMed
    64. ↵
      1. Siddiqi AZ,
      2. Grigat D,
      3. Vatanpour S, et al
      . Transfusions in patients with iron deficiency anemia following release of choosing wisely guidelines. CJEM 2021;23:475–9.doi:10.1007/s43678-021-00082-2pmid:http://www.ncbi.nlm.nih.gov/pubmed/33721287
      OpenUrlPubMed
    65. ↵
      1. Niven DJ,
      2. Rubenfeld GD,
      3. Kramer AA, et al
      . Effect of published scientific evidence on glycemic control in adult intensive care units. JAMA Intern Med 2015;175:801–9.doi:10.1001/jamainternmed.2015.0157pmid:http://www.ncbi.nlm.nih.gov/pubmed/25775163
      OpenUrlPubMed
    66. ↵
      1. Anderson TS,
      2. Leonard S,
      3. Zhang AJ, et al
      . Trends in low-value carotid imaging in the Veterans health administration from 2007 to 2016. JAMA Netw Open 2020;3:e2015250.doi:10.1001/jamanetworkopen.2020.15250pmid:http://www.ncbi.nlm.nih.gov/pubmed/32886120
      OpenUrlPubMed
    67. ↵
      1. Morgan DJ,
      2. Dhruva SS,
      3. Coon ER, et al
      . 2017 update on medical overuse: a systematic review. JAMA Intern Med 2018;178:110–5.doi:10.1001/jamainternmed.2017.4361pmid:http://www.ncbi.nlm.nih.gov/pubmed/28973402
      OpenUrlPubMed
    68. ↵
      1. Morgan DJ,
      2. Dhruva SS,
      3. Coon ER, et al
      . 2019 update on medical overuse. JAMA Intern Med 2019;179:1568.doi:10.1001/jamainternmed.2019.3842
    69. ↵
      1. Morgan DJ,
      2. Dhruva SS,
      3. Wright SM, et al
      . 2016 update on medical overuse: a systematic review. JAMA Intern Med 2016;176:1687–92.doi:10.1001/jamainternmed.2016.5381pmid:http://www.ncbi.nlm.nih.gov/pubmed/27654002
      OpenUrlPubMed

    Supplementary materials

    • Supplementary Data

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Footnotes

    • Twitter @DerekRoberts01, @mmamas1973

    • HTS and JG contributed equally.

    • Contributors DJR conceived the study idea. All authors (DJR, EES, SKN, DN, MM, DIM, CvW, RS, IDG, HTS, and JG) contributed to the design of the study. DJR and EES drafted the study protocol, which was revised for important intellectual content by SKN, DN, MM, DIM, CvW, RS, IDG, HTS and JG. All authors (DJR, EES, SKN, DN, MM, DIM, CvW, RS, IDG, HTS and JG) provided final approval of the manuscript and agree to be accountable for the accuracy and integrity of the work.

    • Funding This work received a grant from the University of Ottawa Faculty of Medicine’s Summer Studentship Program (awarded to DJR and EES).

    • Disclaimer This funder had no role in protocol development.

    • Competing interests None declared.

    • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.