Physician-Reported Misuse of Thyroid Ultrasonography

In the US, there is an ongoing national dialogue about avoiding unnecessary medical tests, with the Choosing Wisely campaign recommending against ordering thyroid ultrasonographic examinations for abnormal thyroid function test results.^1,2 However, despite this national dialogue, little is known about physician-reported use of thyroid ultrasonography, a known driver of thyroid cancer incidence.^3,4

Between October 2018 and August 2019, we surveyed surgeons, endocrinologists, and primary care physicians who were identified as being involved in thyroid cancer care by patients with differentiated thyroid cancer who were diagnosed between 2014 and 2015 and affiliated with Surveillance, Epidemiology, and End Results (SEER) registries of Georgia and Los Angeles, California.⁴ This study was approved by institutional review boards of participating institutions. Survey completion was considered implied consent for participation.

Physicians were asked to select clinical scenarios in which they would schedule a thyroid or neck ultrasonography examination (eMethods in the Supplement). For the analysis, we grouped response options into categories (clinically supported, clinically unclear, and not clinically supported), based on clinical practice guidelines and review by thyroid experts.^2,5,6 Since data are conflicting or limited, the clinically unclear reasons included a history of head or neck radiation and a family history of thyroid cancer. Physicians were also asked to select the informational sources most influential in their decisions for treating patients with thyroid nodules and thyroid cancer (clinical guidelines, meetings, experts, training, colleagues, journal articles, and employer guidelines). We generated descriptive statistics for all categorical variables, and we report nonweighted frequencies and weighted percentages. Statistical analyses (performed with R version 3.5.2 [R Foundation for Statistical Computing] and Stata version 15.1 [StataCorp]) incorporated nonresponse weights. All P values less than .05 were considered statistically significant.

A total of 610 physicians responded, yielding a 65% response rate. The cohort consisted of primary care physicians (n = 162), endocrinologists (n = 176), otolaryngologists (n = 130), and general surgeons (n = 134). Of the respondents, 405 (66.0%) were men, 392 (67.3%) self-identified as White, 347 (59.6%) worked in private practice, and 432 (69.7%) reported that they had read the 2015 American Thyroid Association (ATA) management guidelines for adult patients with thyroid nodules and differentiated thyroid cancer and/or the 2017 National Comprehensive Cancer Network (NCCN) Clinical Practice Guidelines in Oncology–Thyroid Carcinoma. Two hundred eighty-eight physicians (46.9%) reported that they had read the 2015 ATA guidelines alone, 33 (5.3%) had read the 2017 NCCN guidelines alone, and 111 (17.6%) had read both the ATA and NCCN guidelines. These physician respondents had treated 74% of the previously surveyed 2014-2015 SEER patient cohort.⁴

Most of the physicians reported use of ultrasonography for clinically supported reasons: palpable nodules (599 [98.2%]), large goiter (560 [91.8%]), nodule seen on another imaging test (538 [88.1%]), and new-onset hoarseness or compressive symptoms (398 [65.8%]). However, a substantial number endorsed use for clinically unsupported reasons: patient request (202 [32.7%]), abnormal thyroid function test results (169 [28.0%]), and positive thyroid antibody test result (136 [22.5%]) (Figure 1). Three physicians (0.5%) endorsed use of ultrasonography for fatigue, and no physician endorsed use for routine well-patient visits.

Multivariable logistic regression analysis identified factors associated with thyroid ultrasonography misuse (Figure 2). Endocrinologists (odds ratio [OR], 2.46 [95% CI, 1.22-4.97], otolaryngologists (OR, 2.87 [95% CI, 1.49-5.54]), and general surgeons (OR, 2.20 [95% CI, 1.19-4.06]) were more likely to schedule ultrasonography examinations in response to a patient request compared with primary care physicians. Physicians in private practice were more likely to schedule ultrasonography examinations for abnormal thyroid function test results (OR, 2.45 [95% CI, 1.28-4.69]) and positive thyroid antibody test results (OR, 2.61 [95% CI, 1.28-5.32]) compared with those in academic medical centers. Physicians who managed 10 or fewer patients with thyroid nodules per year, compared with more than 50 such patients, were less likely to schedule ultrasonography examinations for positive thyroid antibody test results (OR, 0.41 [95% CI, 0.18-0.93]).

Most surveyed physicians (428 [69.3%]) reported recently published clinical guidelines as most influential in their decisions for treating patients with thyroid nodules and thyroid cancer.

To our knowledge, this is the first survey study to evaluate use of thyroid ultrasonography by a diverse cohort of physicians involved in the care of a population-based cohort of patients with thyroid cancer. A substantial number of physicians endorsed use of ultrasonography for reasons not supported by clinical guidelines and in conflict with the Choosing Wisely recommendations.^2,5,6

Our study has limitations in that direct links between physician-reported practice patterns and unnecessary thyroid cancer diagnoses were not possible, and physicians were not asked if they personally performed thyroid ultrasonography examinations or specific details about patients’ ultrasonography requests. Despite limitations, this study highlights the need for focused physician education on clinically supported and unsupported indications for use of thyroid ultrasonography, with potential roles for future clinical practice guidelines, patient decision-making aids, and clinical decision-making support tools.

Accepted for Publication: May 17, 2020.

Corresponding Author: Megan R. Haymart, MD, Metabolism, Endocrinology, and Diabetes, University of Michigan, 2800 Plymouth Rd, North Campus Research Complex, Bldg 16, Room 408E, Ann Arbor, MI 48109 (meganhay@med.umich.edu).

Published Online: August 12, 2020. doi:10.1001/jamasurg.2020.2507

Correction: This article was corrected on October 21, 2020, to fix numbers placed out of order on the x-axes in Figure 1 and incorrect word placement in Figure 2, with “General surgery site,” corrected to “General surgery,” “Los Angeles County practice” corrected to “Los Angeles County site,” and “Setting” corrected to “Practice setting.”

Author Contributions: Dr Haymart had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Haymart.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Chen, Haymart.

Critical revision of the manuscript for important intellectual content: Reyes-Gastelum, Radhakrishnan, Hamilton, Ward.

Statistical analysis: Reyes-Gastelum.

Obtained funding: Haymart.

Administrative, technical, or material support: Hamilton, Ward, Haymart.

Supervision: Hamilton, Haymart.

Conflict of Interest Disclosures: Drs Hamilton and Ward reported a subcontract from the University of Michigan during the conduct of this study. No other disclosures were reported.

Funding/Support: This study is supported by the National Cancer Institute (grant R01 CA201198 [Dr Haymart]). Dr Haymart also receives funding from the Agency for Healthcare Research and Quality (grant R01 HS024512). Dr Chen receives support from the National Institutes of Diabetes and Digestive and Kidney Diseases (grant T32DK07245). The collection of cancer incidence data used in this study was supported by the California Department of Public Health pursuant to California Health and Safety Code (section 103885); the US Centers for Disease Control and Prevention’s National Program of Cancer Registries (cooperative agreement 5NU58DP003862-04/DP003862), the National Cancer Institute’s Surveillance, Epidemiology, and End Results Program (contract HHSN261201000035C, awarded to the University of Southern California). The collection of cancer incidence data in Georgia was supported by the National Cancer Institute (contract HHSN261201800003I; task order HHSN26100001) and the US Centers for Disease Control and Prevention (cooperative agreement 5NU58DP003875-04).

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Disclaimer: The ideas and opinions expressed herein are those of the authors, and endorsement by the State of California and State of Georgia Departments of Public Health, the National Cancer Institute, and the US Centers for Disease Control and Prevention or their contractors and subcontractors is not intended nor should be inferred.

Additional Contributions: Brittany L. Gay, BA, University of Michigan, assisted with the figures. She was compensated for this contribution. We acknowledge with gratitude the physicians who responded to our survey.