Development of Choosing Wisely Recommendations for an Inpatient

Letters

Figure. Trends of Radioactive Iodine (RAI) Therapy Misuse in Papillary Thyroid Microcarcinomas Over Time, by Facility Type

Academic/Research Hospital Programs Comprehensive Community Hospital Programs Community Hospital Programs RAI P < .001 P < .001 P = .26 100 100 100 No Yes

80 80 80

71.6 72.5 71.2 72.2 75.0 74.3 74.8 75.0 72.3 74.6 77.6 76.6 77.6 60 77.5 78.4 60 82.2 60 78.5 77.3 78.3 80.2 81.5 83.6 86.4 81.4

40 40 40 RAI Misuse, % RAI Misuse, % RAI Misuse, %

20 20 20 25.0 25.7 25.2 25.0 28.4 27.5 28.8 27.7 27.8 22.5 25.4 22.4 22.7 23.4 21.6 21.5 21.7 22.4 18.6 19.8 18.5 16.4 13.6 17.8 0 0 0 2004 20052006 2007 2008 2009 2010 2011 2004 20052006 2007 2008 2009 2010 2011 2004 20052006 2007 2008 2009 2010 2011 Year of Diagnosis Year of Diagnosis Year of Diagnosis

Misuse of RAI therapy in papillary thyroid microcarcinomas in academic/research programs, comprehensive community programs, and community programs.

In conclusion, our study demonstrates that nearly one- Commission on Cancer have not verified and are not responsible for the analytic fourth of patients with thyroid cancer may receive unnec- or statistical methods used, or the conclusions drawn from these data by the investigators. esary RAI treatment. Ongoing efforts should be undertaken to Additional Contributions: Mohamed Abdelgadir Adam, MD, Section of educate health care professionals in the appropriate use of RAI Endocrine Surgery, Department of Surgery, Duke University Medical Center, therapy to optimize patient care. Durham, North Carolina, and Linda Youngwirth, MD, Duke Clinical Research Institute, and Section of Endocrine Surgery, Department of Surgery, Duke Paolo Goffredo, MD University Medical Center, Durham, North Carolina, assisted with data acquisition and statistical work. They were not compensated for their Samantha M. Thomas, MB contributions. Michaela A. Dinan, PhD 1. Cooper DS, Doherty GM, Haugen BR, et al. Revised American Thyroid Jennifer M. Perkins, MD Association management guidelines for patients with thyroid nodules and Sanziana A. Roman, MD differentiated thyroid cancer. Thyroid. 2009;19(11):1167-1214. Julie A. Sosa, MD 2. Smallridge RC, Ain KB, Asa SL, et al. American Thyroid Association guidelines for management of patients with anaplastic thyroid cancer. Thyroid. 2012;22 (11):1104-1139. Author Affiliations: Duke Clinical Research Institute, Durham, North Carolina (Goffredo, Dinan, Sosa); Department of Biostatistics, Duke University, Durham, 3. Kloos RT, Eng C, Evans DB, et al. Medullary thyroid cancer: management North Carolina (Thomas); Department of Medicine, Duke University, Durham, guidelines of the American Thyroid Association. Thyroid. 2009;19(6):565-612. North Carolina (Perkins); Section of Endocrine Surgery, Department of Surgery, 4. Lee SL. Complications of radioactive iodine treatment of thyroid carcinoma. Duke University Medical Center, Durham, North Carolina (Roman, Sosa). J Natl Compr Canc Netw. 2010;8(11):1277-1287. Corresponding Author: Sanziana A. Roman, MD, Section of Endocrine Surgery, 5. National Cancer Data Base. 2014. https://www.facs.org/quality%20programs Department of Surgery, Duke University School of Medicine, DUMC #2945, /cancer/ncdb/puf. Accessed May 1, 2014. Durham, NC 27710 ([email protected]). 6. National Cancer Institute. Surveillance, Epidemiology, and End Results (SEER) Published Online: February 16, 2015. doi:10.1001/jamainternmed.2014.8020. Program. http://seer.cancer.gov/data/. Published 2014. Accessed June 1, 2014. Author Contributions: Dr Sosa had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Creating a List of Low-Value Health Care Study concept and design: Goffredo, Dinan, Roman, Sosa. Activities in Swiss Primary Care Acquisition, analysis, or interpretation of data: Goffredo, Thomas, Dinan, In 2010, the idea emerged of creating lists of low-value health Perkins. Drafting of the manuscript: Goffredo, Dinan, Perkins, Roman. care activities as a way to confront rising medical costs and en- Critical revision of the manuscript for important intellectual content: All authors. courage cost-conscious care. The Good Stewardship Working Statistical analysis: Goffredo, Thomas, Dinan, Roman. Group1 and Brody2 pioneered the idea of “top 5” lists, leading Obtained funding: Goffredo. Administrative, technical, or material support: Dinan, Roman, Sosa. to the Choosing Wisely 3 Study supervision: Perkins, Roman, Sosa. Related article page 642 and campaign. Building on this Conflict of Interest Disclosures: None reported. Invited Commentary page 644 momentum, there has been Funding/Support: Dr Goffredo is supported by the Fondazione Italiana per la widespread interest in pro- Ricerca sul Cancro (Italian Foundation for Cancer Research). posing additional lists.4 In 2012, the Swiss Society of General Role of the Funder/Sponsor: The funding source had no role in the design and Internal Medicine committed to creating a list for Swiss am- conduct of the study; collection, management, analysis, and interpretation of bulatory internal medicine. the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. Disclaimer: Some of the data used in the study are derived from a deidentified Methods | A review of publications was performed using the National Cancer Data Base file. The American College of Surgeons and the search terms low value, disinvestment, less is more, and avoidable

640 JAMA Internal Medicine April 2015 Volume 175, Number 4 (Reprinted) jamainternalmedicine.com

Downloaded From: https://jamanetwork.com/ on 09/25/2021 Letters

Figure. Flowchart of Recommendations Through the Delphi Process

50 Under consideration Chosen for next round 15 pass to Eliminated round 3 40 Final selection

recommendations 17 are 30 10 reranked 14 Agreement Agreement Frequency Committee 5 ranking ranking ranking selection international

20 38 13 Recommendations, No. Recommendations, 6

10 5 Swiss A total of 50 recommendations were

novel 7 ranked during the Delphi process, 12 0 recommendations including 38 existing international recommendations and 12 novel From Review Round 1 Round 2 Round 3 Final recommendations from the experts. of Publications From the top 10 recommendations of Results the Delphi process, an expert panel chose the final top 5 list.

Table. Top 10 Recommendations Based on Frequency Scorea

Frequency Score Agreement Score Rank Recommendation (32-96)b (0-10)c 1 Do not obtain imaging studies in patients with nonspecific low 94 9.56 back pain 2 Do not prescribe antibiotics for uncomplicated URTIs 92 9.40 3 Do not perform the PSA test to screen for prostate cancer 90 9.59 without a discussion of the risks and benefits 4 Do not perform laboratory testing in patients with a clinical 87 9.03 diagnosis of an uncomplicated URTI 5 Do not continue pharmacological treatment of GERD with 82 9.50 long-term acid suppression therapy without titrating to the lowest effective dose Abbreviations: DEXA, dual-energy x-ray absorptiometry; GERD, 6 Do not routinely prescribe antibiotics for acute 81 9.50 mild-to-moderate sinusitis gastroesophageal reflux disease; PSA, prostate-specific antigen; 7 Do not use antimicrobials to treat bacteriuria in 80 9.16 URTI, upper respiratory tract immunocompetent older adults infection. 8 Do not routinely obtain radiographic imaging for patients who 78 9.91 meet diagnostic criteria for uncomplicated acute rhinosinusitis a Boldface indicates items retained 9 Do not obtain preoperative chest radiography in the absence of a 77 9.26 for top 5. clinical suspicion b Frequency scores are from round 3. 10 Do not use DEXA screening for osteoporosis in women younger 72 9.16 c Agreement scores are from rounds 1 than 65 or men younger than 70 and 2.

care. Three sets of lists were identified (1103 recommenda- formed based on Swiss Society of General Internal Medicine tions)3,5,6 as of March 5, 2013. Two physicians excluded recom- members who expressed a specific interest in this subject. mendations that were not relevant to ambulatory internal medi- In round 1, experts gave their level of agreement with the cine (eg, specialized medicine, pediatrics), leading to an initial international recommendations using a 10-point Likert scale. list of 38 international recommendations. Experts could also propose additional recommendations. Af- An online Delphi process was then applied, using succes- ter a review of publications to ensure their validity based on sive electronic survey instruments placed on the Survey Mon- available evidence, 12 of 21 novel recommendations were re- key website (www.surveymonkey.com). All committee mem- tained. bers of the Swiss Society of General Internal Medicine and the In round 2, recommendations with intermediate scores in Swiss Society of Family Medicine, along with professors from round 1 (average scores, 7-9) were reranked based on experts’ the divisions of General Internal Medicine and Family Medi- level of agreement, along with the 12 novel recommendations. cine at the 5 Swiss university medical schools, were invited to For round 3, recommendations with scores greater than 9 participate as experts. A 7-member advisory committee was were graded based on a 3-point Likert scale in 3 areas: fre-

jamainternalmedicine.com (Reprinted) JAMA Internal Medicine April 2015 Volume 175, Number 4 641

Downloaded From: https://jamanetwork.com/ on 09/25/2021 Letters

quency, costs, and patient harm. Frequency was defined as how Study concept and design: Selby, Gaspoz, Rodondi, Neuner-Jehle, Perrier, often the average general practitioner is faced with the deci- Cornuz. Acquisition, analysis, or interpretation of data: Selby, Rodondi, Neuner-Jehle, sion to perform the test or prescribe the treatment. Costs were Perrier, Zeller, Cornuz. direct costs and not those of unanticipated adverse effects or Drafting of the manuscript: Selby, Zeller, Cornuz. complications. Harms were defined as potential harms from Critical revision of the manuscript for important intellectual content: All authors. the test or treatment, including those that could be expected Statistical analysis: Selby. Obtained funding: Gaspoz, Cornuz. from the recommendation. For reasons of implementation, the Administrative, technical, or material support: Selby, Perrier, Cornuz. final list was limited to 5 of the 10 most frequent recommen- Study supervision: Gaspoz, Rodondi, Perrier, Cornuz. dations. Conflict of Interest Disclosures: None reported. Funding/Support: This study was supported by a grant from the Swiss Society Results | Of the 59 experts contacted, 35 agreed to participate of General Internal Medicine and the Department of Ambulatory Care and (59%; mean [SD] age, 51 [6.3] years; 27 men [77%]). A flow- Community Medicine, Lausanne University. chart of the recommendations is shown in the Figure. Through Role of the Funder/Sponsor: The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of rounds 1 and 2, a total of 50 items were ranked based on an the data; preparation, review, or approval of the manuscript; and decision to agreement scale of 1 to 10, including the 12 novel recommen- submit the manuscript for publication. dations. The mean (SD) agreement score was 8.52 (0.80) of 10. Additional Contributions: Christoph Meier, MD, Department of Medicine, Of the 18 recommendations reviewed in round 3, the top 10, Triemli Hospital, provided guidance throughout this project, 35 expert ranked by perceived frequency, are seen in the Table. The fi- participants volunteered their time, and Joana Le Boudec, MD, Department of Ambulatory Care and Community Medicine, University of Lausanne, assisted nal top 5 list was made by consensus of the advisory commit- with drafting the first survey instrument and reviewing international tee, who believed there would be too much overlap if there recommendations. None were financially compensated. were 2 recommendations for respiratory tract infections. 1. Good Stewardship Working Group. The “top 5” lists in primary care: meeting the responsibility of professionalism. Arch Intern Med. 2011;171(15):1385-1390. Discussion | Our study illustrates a method to allow medical so- 2. Brody H. Medicine’s ethical responsibility for health care reform: the top five cieties to create their own national lists based on existing in- list. N Engl J Med. 2010;362(4):283-285. ternational work.3,5,6 Our high agreement scores suggest that 3. American Board of Internal Medicine. Choosing Wisely website. http://www .choosingwisely.org/. Accessed January 7, 2015. there is enough consensus to allow for the adaptation of such 4. Hurley R. Can doctors reduce harmful medical overuse worldwide? lists in other countries. [published online July 3, 2014]. BMJ. doi:10.1136/bmj.g4289. The Good Stewardship Working Group1 used a small com- 5. National Institute for Health and Care Excellence. NICE 'do not do' mittee for the generation and initial selection of recommen- recommendations. National Institute for Health and Care Excellence website. dations and a larger group of 255 health care professionals for https://www.nice.org.uk/proxy/?sourceurl=http://www.nice.org.uk validation; we started from an initial list of international rec- /usingguidance/donotdorecommendations/index.jsp.Accessed January 7, 2015. ommendations and used a panel of 35 experts for selection. 6. Elshaug AG, Watt AM, Mundy L, Willis CD. Over 150 potentially low-value health care practices: an Australian study. Med J Aust. 2012;197(10):556-560. We are currently conducting an implementation study among Swiss general practitioners.

Kevin Selby, MD LESS IS MORE Jean-Michel Gaspoz, MD, MPH Development of Choosing Wisely Recommendations Nicolas Rodondi, MD, MAS for an Inpatient Internal Medicine Service Stefan Neuner-Jehle, MD, MPH As part of the American Board of Internal Medicine Founda- Arnaud Perrier, MD tion’s Choosing Wisely campaign,1 more than 60 specialty so- Andreas Zeller, MD cieties have published lists of 5 tests, procedures, or treat- Jacques Cornuz, MD, MPH ments that physicians and patients should question. Recognizing the opportunity Author Affiliations: Department of Ambulatory Care and Community for the provision of higher- Medicine, University of Lausanne, Lausanne, Switzerland (Selby, Cornuz); Related article page 640 and Division of Primary Care Medicine, Department of Community Medicine, Invited Commentary page 644 value care in our own set- Primary Care, and Emergency Medicine, Geneva University Hospitals, Geneva, ting, we developed a Choos- Switzerland (Gaspoz); Department of General Internal Medicine, Bern ing Wisely list for the inpatient General Internal Medicine University Hospital, Bern, Switzerland (Rodondi); Institute of General Practice and Health Services Research, University of Zurich, Zurich, Switzerland service at Mount Sinai Hospital (MSH), one of the affiliated (Neuner-Jehle); Department of Medicine, University of Geneva, Geneva, teaching hospitals at University of Toronto. Switzerland (Perrier); Institute of Primary Health Care, University of Basel, Basel, Switzerland (Zeller). Methods | All attending physicians who care for patients on the Corresponding Author: Kevin Selby, MD, Department of Ambulatory Care and General Internal Medicine service at MSH were invited to rec- Community Medicine, University of Lausanne, Policlinique médicale ommend 2 commonly ordered tests or procedures for inclu- universitaire, Rue de Bugnon 44, 1011 Lausanne, Switzerland (kevin.selby @hospvd.ch). sion on our Choosing Wisely list and to participate in an itera- Published Online: February 23, 2015. doi:10.1001/jamainternmed.2014.8111. tive voting process to identify the 5 items that are least likely to be of net benefit to patients. Once solicitation of initial sug- Author Contributions: Drs Selby and Cornuz had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy gestions was complete, the participants were asked to select of the data analysis. exactly 5 items for the final Choosing Wisely list via an anony-

642 JAMA Internal Medicine April 2015 Volume 175, Number 4 (Reprinted) jamainternalmedicine.com

Downloaded From: https://jamanetwork.com/ on 09/25/2021 Letters

Table 1. List of Choosing Wisely Suggestions and Voting Results

Votes Itema Round 1 Round 2 Troponin without suspected acute coronary syndrome 4 8 CT head in delirious patients without reason to suspect abnormality (head trauma, anticoagulation, abnormal 67 neuro exam) Daily complete blood count, electrolytes, creatinine×3dwithout clear indication 5 7 2-D echocardiogram for patients with recent echo and clinical exam consistent with prior result 5 6 Urinalysis or urine culture in the absence of symptoms of urinary tract infection 3 6 Daily calcium, magnesium, phosphate in absence of abnormalities 4 5 Urinary catheters in the absence of indications (obstruction, critical illness, palliative care) 4 4 Magnetic resonance imaging of the brain in patients with clinical diagnosis of stroke 3 4 Folate 34 New benzodiazepine prescriptions for sleep 3 4 Carotid dopplers in patients with stroke who would not be surgical candidates 3 4 Erythrocyte sedimentation rate for anything except temporal arteritis or monitoring response to therapy in 33 connective tissue disease CT chest for uncomplicated pneumonia 33 Nil per osb for stroke patients until speech language pathology assessment when bedside swallowing test would 42 suffice Colace (docusate sodium; Purdue Products LP) 4 2 Ultrasound of the kidney for uncomplicated pyelonephritis 3 2 Repeat thyroid-stimulating hormone 31 Chest x-ray to follow up heart failure 2NA Nonferritin iron studies for anemia 2NA Repeat albumin 2NA Gamma-glutamyl transferase 1NA Sputum/blood cultures for low-risk community-acquired pneumonia (no recent hospitalization/COPD/diabetes) 1 NA Abdominal x-ray to assess for constipation 1NA Chest x-ray to follow up pneumonia before 4-6 wk 1 NA Rheumatoid factor 1NA Urine eosinophils 1NA Lactate 0NA Hepatitis serologies without considering pattern of liver enzyme elevation 0 NA Telemetry for stroke patients (as opposed to Holter monitors) 0 NA Venereal Disease Research Laboratory test 0NA Autoimmune serologies beyond screening tests (antinuclear antibody, RF, complements) 0 NA Total 75 75

Abbreviations: COPD, chronic obstructive pulmonary disease; CT, computed tomography; exam, examination; NA, items eliminated for having zero votes in the first round could not be voted on in the second; RF, rheumatoid factor. a Original wording of suggestions, as provided by the participating physicians, was preserved. b Indicates that nothing can be eaten or drunk.

mous online survey. The physicians were not asked to rank items received 2 or fewer votes (<15% of participants) in the their selections. The results of each round of voting were dis- first round of voting and were eliminated. Consensus regard- tributed to the participants, and the least popular items were ing the top 5 items (those with the most votes) for the final list eliminated from consideration before subsequent rounds. The was achieved in the second round of voting. The top 5 recom- selection process was repeated until 5 items had more votes mendations were each endorsed by a minimum of 40% of the than the remainder of the suggestions, comprising our final participants, with the top recommendation being endorsed by Choosing Wisely list. Institutional review board approval was more than 50% of participants. The final Choosing Wisely list neither obtained nor waived because no patient information for the inpatient internal medicine service at MSH is shown in was used. Table 2.

Results | Fifteen attending physicians agreed to participate, and Discussion | To our knowledge, this is the first instance of an in- a total of 31 initial suggestions were received. The initial list dividual hospital division developing its own Choosing Wisely and results of voting are published in Table 1. Fourteen (45%) list. While the selected items may not be generalizable be-

jamainternalmedicine.com (Reprinted) JAMA Internal Medicine April 2015 Volume 175, Number 4 643

Downloaded From: https://jamanetwork.com/ on 09/25/2021 Letters

Table 2. Final Choosing Wisely Recommendations for the Inpatient Internal Medicine Service at Mount Sinai Hospital

Ranking Recommendation 1 Do not order troponins in the absence of angina, new or worsened heart failure, EKG changes consistent with an acute coronary syndrome, or sudden death with ventricular fibrillation 2 Do not order a) daily complete blood counts (in the absence of active bleeding, symptomatic anemia, febrile neutropenia, or primary hematologic disorder) or b) daily electrolyte panels and/or creatinine levels (in the absence of existing abnormalities such as hypo/hypernatremia or hypo/hyperkalemia, renal failure, active diuresis or intravenous fluid replacement) 3 Do not order computed tomography of the brain in delirious patients in the absence of abnormal neurological findings, a history of falls, or head trauma 4 Do not order urinalysis or urine culture for patients in the absence of symptoms of urinary tract infection or, in patients unable to provide a history, the systemic inflammatory response syndrome 5 Do not order echocardiograms in patients with known heart failure, a recent echo report (eg, within one year if patient has been stable over that time), and an identified precipitant Abbreviation: EKG, electrocardiogram.

yond our scope of practice, the exercise of identifying areas Invited Commentary of improvement can be applied in any setting. A potential weak- Challenges in Choosing Wisely’s International Future: ness of our study is the reliance on expert opinion, although Support, Evidence, and Burnout substantial evidence supports each of the items we gener- In the short time since its April 2012 launch by the American ated. A further possible weakness is the potential for random Board of Internal Medicine Foundation, the Choosing Wisely variation in voting patterns to determine the final outcome ow- campaign has affected more than 60 US specialty societies. ing to our small sample size. However, the general consensus Now the campaign is becoming an international phenom- among participants was that the number of high-quality sug- enon, as evidenced by Selby gestions on the initial list was in excess of 5, making it un- 1 2 Related articles pages 640 et al and Gupta and Detsky likely that our final list is invalid. Finally, in a small depart- and 642 in this issue. These publica- ment like ours, we were concerned about the potential for a tions should be considered in small number of individuals having an undue influence over the context of other national efforts, the most prominent being the final outcome, which we mitigated through the anony- Choosing Wisely Canada,3 which identifies itself as being mod- mous online voting process. eled after its American counterpart and having “spread to Aus- Our list is unique to the inpatient internal medicine ser- tralia, Germany, Italy, Japan, Netherlands, Switzerland and vice, and perhaps to our local environment at MSH. As the com- elsewhere.”This rapid expansion is a heartening sign that there plexity of medicine increases, physicians will face difficult is increasing international sentiment against wasteful medi- choices regarding how best to investigate and treat patients’ cal practices. However, actually decreasing wasteful and harm- conditions. We must also be mindful of the value our care pro- ful health care will require both patient and physician com- vides in a resource-constrained world. Initiatives such as this mitment as well as objective evidence of effectiveness. If either are one way to address these problems in the future. is found wanting, the results will be underwhelming. Both Choosing Wisely and its global counterparts seek to Shlok Gupta, MD reduce low-value health care by generating “top 5” lists of Allan S. Detsky, MD, PhD, FRCPC practices that should be questioned by patients and health care professionals; as such, effecting change from these sug- Author Affiliations: Department of Medicine, Mount Sinai Hospital, Toronto, gested lists will require constituents’ support of the lists. Ontario, Canada (Gupta, Detsky); Institute of Health Policy, Management and Therefore, it is fitting that the process should be designed to Evaluation, University of Toronto, Toronto, Ontario, Canada (Detsky). optimize support. The first such top 5 list, generated by the Corresponding Author: Allan S. Detsky, MD, PhD, FRCPC, Mount Sinai Hospital, 600 University Ave, Room 429, Toronto, ON M5G 1X5, Canada (adetsky American Academy of Family Physicians and published in @mtsinai.on.ca). Archives of Internal Medicine in 2011,4 was developed using a Published Online: February 23, 2015. doi:10.1001/jamainternmed.2014.8119. modified nominal group process. Three years later, the Swiss Author Contributions: Drs Gupta and Detsky had full access to all the data in Society of General Internal Medicine used a process of litera- the study and take responsibility for the integrity of the data and the accuracy ture review supplemented by expert opinion, followed by an of the data analysis. electronic Delphi process in multiple rounds to identify the Study concept and design: All authors. 2 Acquisition, analysis, or interpretation of data: All authors. top 5 recommendations. Gupta and Detsky chose to solicit Drafting of the manuscript: All authors. recommendations from physicians, followed by a voting Critical revision of the manuscript for important intellectual content: All authors. process to select the top recommendations. The above pro- Statistical analysis: Gupta. cesses are quasi-systematic methods that focus on physician Administrative, technical, or material support: Detsky. participation to increase support; however, they rely on sub- Conflict of Interest Disclosures: None reported. jective perceptions of feasibility and benefit to cost ratio 1. American Board of Internal Medicine Foundation. Choosing wisely: an initiative of the ABIM foundation. http://www.choosingwisely.org. Accessed rather than high-quality evidence. An alternative would be January 6, 2014. to develop a systematic process to identify the practices that

644 JAMA Internal Medicine April 2015 Volume 175, Number 4 (Reprinted) jamainternalmedicine.com

Downloaded From: https://jamanetwork.com/ on 09/25/2021 Letters

and cost reductions. Doing so might be accomplished with clearly the will to change exists among physicians and pa- traditional methods of evidence-based recommendations, tients. Systematic, repeated, deliberate effort is required, and such as systematic review and health technology assess- tools such as dashboards, performance reporting, financial in- ment. Alternatively, one might leverage already-generated centives, benchmarking, and repeated feedback loops may be lists of practices that are known to have evidence of harm useful. We believe all top 5 lists should be accompanied by an and are high in cost. Several such lists already exist, with implementation plan and should be evaluated and continu- groups having completed systematic reviews as well as ously monitored to assess their effect on low-value health care sophisticated cost modeling to project potential cost utilization. reduction.4-6 Choosing Wisely and Choosing Wisely Canada aspired to Fiona Clement, PhD initiate conversations about eliminating low-value health Blake Charlton, MD care,3,7 and they have succeeded in realizing their goal. We are now ready and eager for the next steps—to realize a measur- Author Affiliations: Department of Community Health Sciences, O’Brien Institute of Public Health, University of Calgary, Calgary, Alberta, Canada able decrease in low-value health care utilization. Currently, (Clement); Department of Internal Medicine, University of California, San we are not aware of any evidence that the top 5 lists, in the Francisco (Charlton). United States or abroad, have reduced low-value medical prac- Corresponding Author: Fiona Clement, PhD, Department of Community tices. The American Board of Internal Medicine Foundation Health Sciences, O’Brien Institute of Public Health, University of Calgary, 3289 studied the effect of their campaign using a telephone survey Hospital Dr NW, Third Floor, Teaching, Research and Wellness Bldg, Calgary, AB T2N 4Z6, Canada ([email protected]). of 600 US physicians and found that 21% had heard of the Published Online: February 23, 2015. doi:10.1001/jamainternmed.2014.8122. Choosing Wisely campaign and that, among this subgroup, 62% Conflict of Interest Disclosures: None reported. reported having reduced unnecessary testing in the past year.8 These results suggest, at best, a modest effect; but because the 1. Selby K, Gaspoz J-M, Rodondi N, et al. Creating a list of low-value health care activities in Swiss primary care [published online February 23, 2015]. JAMA results are self-reported, they cannot be used to estimate the Intern Med. doi:10.1001/jamainternmed.2014.8111. net effect on cost-effective care. Building on the success of the 2. Gupta S, Detsky AS. Development of Choosing Wisely recommendations for Choosing Wisely campaign will require demonstration of a re- an inpatient internal medicine service [published online February 23, 2015]. duction in wasteful practices. As global enthusiasm for top 5 JAMA Intern Med. doi:10.1001/jamainternmed.2014.8119. lists mounts, so will the desire to use the lists to shape policy 3. Canadian Medical Association. What is CWC? Choosing Wisely Canada website. and practices. Evidence of the effectiveness of top 5 lists must www.choosingwiselycanada.org/about/what-is-cwc. Accessed November 11, 2014. grow, not just the number of lists—otherwise, physicians may 4. Aguilar IBZ, Casher D, Choi RY, et al; Good Stewardship Working Group. The “top 5” lists in primary care: meeting the responsibility of professionalism. Arch question the value of the campaigns. The resulting skepti- Intern Med. 2011;171(15):1385-1390. cism and cynicism are likely to lead to decreased support. 5. Elshaug AGWA, Watt AM, Mundy L, Willis CD. Over 150 potentially low-value Even more dangerous to the movement than the present health care practices: an Australian study. Med J Aust. 2012;197(10):556-560. 9 lack of evidence would be top 5 list “burnout.”Survey studies 6. National Institute for Health and Care Excellence. NICE ‘do not do’ have shown that physicians who were faced with multiple recommendations. National Institute for Health and Care Excellence website. guidelines on a single topic become less certain of how to pro- www.nice.org.uk/proxy/?sourceurl=http://www.nice.org.uk/usingguidance /donotdorecommendations/index.jsp. Accessed November 11, 2014. ceed. Professional organizations around the world already pub- 7. American Board of Internal Medicine Foundation. About. Choosing Wisely lish guidelines that outline the best evidence-based prac- website. www.choosingwisely.org/about-us/. Accessed November 11, 2014. tices. While top 5 lists are not guidelines, it is likely that the 8. Perry Undem Research/Communication. Unnecessary tests and procedures finding of guideline burnout is generalizable to top 5 lists. This in the health care system: what physicians say about the problem, the causes, potential seems more likely now that individual institutions and the solutions: results from a national survey of physicians. www are adopting their own top 5 lists. For example, Gupta and .choosingwisely.org/wp-content/uploads/2014/04/042814_Final-Choosing -Wisely-Survey-Report.pdf. Accessed November 11, 2014. Detsky2 describe the creation of a top 5 list for the general in- 9. Gagliardi ARBM, Brouwers MC, Palda VA, Lemieux-Charles L, Grimshaw JM. ternal medicine inpatient service at Mount Sinai Hospital in How can we improve guideline use? a conceptual framework of Toronto, Ontario, Canada, that might be considered in addi- implementability. Implement Sci. 2011;6(26):26. tion to the top 5 lists from the 21 Canadian specialty societies that partnered with Choosing Wisely Canada. It is not hard to imagine a near future in which every service at an individual Stress Testing Before Low-Risk Surgery: So Many institution generates a top 5 list to be considered in the con- Recommendations, So Little Overuse text of the national specialty society top 5 lists, all without spe- The Choosing Wisely (CW) campaign, which commenced in cific evidence of effectiveness. Particularly, if some of these 2011, focuses on reducing medical services that are of ques- lists are discordant, they will be unlikely to change behavior. tionable value or may be harmful.1 In 1996 and 2002, guide- Internationally, health care costs are increasing without a lines from the American College of Cardiology and the Ameri- commensurate improvement in health outcomes. Therefore, can Heart Association implied

we strongly believe in the global spirit behind the Choosing Editor's Note page 647 that routine stress testing be- Wisely campaigns and movement; that is why we advocate that fore low-risk surgeries should the priority in 2015 should be the thoughtful implementation be avoided; this was codified in the 2007 guidelines2 because and rigorous evaluation of existing top 5 lists. Changing be- the risk of cardiac complications from these surgeries is very havior is more complex and challenging than writing a list, but low. Consequently, 7 specialty societies for the CW campaign

jamainternalmedicine.com (Reprinted) JAMA Internal Medicine April 2015 Volume 175, Number 4 645

Downloaded From: https://jamanetwork.com/ on 09/25/2021