[Skip to Content]
[Skip to Content Landing]
Editorial
July 2017

Cancer Screening, Overdiagnosis, and Regulatory Capture

Author Affiliations
  • 1Dartmouth Institute of Health Policy and Clinical Practice, Geisel School of Medicine, Hanover, New Hampshire
JAMA Intern Med. 2017;177(7):915-916. doi:10.1001/jamainternmed.2017.1198

The US Preventive Services Task Force (USPSTF) has given thyroid cancer screening a D grade, recommending against screening.1 They conclude with moderate certainty that screening for thyroid cancer in asymptomatic persons results in harms that outweigh the benefits. Given the historical context—both in pathology and medical care epidemiology—it should have been an easy decision.

At the end of World War II, there was an alarming publication by a group of Chicago surgeons reporting that 17% of nontoxic thyroid nodules harbored thyroid cancer. The surgeons concluded that this finding makes “surgical therapy quite urgent in this seemingly harmless lesion.”2(p883) A few months later, New York surgeons reported that apparently benign thyroid nodules and innocent-appearing breast lumps had a similar prevalence of cancer.3 They concluded that it is as justifiable to perform a thyroidectomy for a thyroid nodule as a biopsy for a breast mass.

In 1947, a Boston pathologist responded to these high prevalence estimates by raising questions about the criteria for thyroid cancer and how they might differ among pathologists. He argued that “the ultimate criterion for the diagnosis of cancer is that the disease naturally results in the death of the patient harboring it.”4(p221) And he went on to emphasize that thyroid cancer was an extremely rare cause of death.

The extent of the high prevalence/rare death mismatch was made even more evident in 1985. Finnish pathologists reported on their work systematically sectioning the thyroid gland in 101 consecutive autopsies.5 They found that 36% contained papillary thyroid cancer, the majority of which were very small. Given the number of small cancers they found and the number that they reasoned they missed, the researchers deduced that virtually everybody would have some evidence of thyroid cancer if examined carefully enough.6 They concluded that small papillary thyroid cancers are “so common in Finland, possibly also in other countries, that they can be regarded as a normal finding.”5(p538) Theoretically, at least, there was a bottomless reservoir of undetected thyroid cancer.

But there was no epidemiologic evidence that medical interventions were beginning to tap this reservoir until the mid 1990s. SEER data7,8 show that thyroid cancer incidence had been relatively stable in the United States—but has since shot up 3-fold (Figure). All of the increase is due to papillary thyroid cancer. And despite the dramatic increase in incidence, mortality from thyroid cancer has remained stable—a combination that is pathognomonic for overdiagnosis.

Place holder to copy figure label and caption
Figure.
Thyroid Cancer Incidence and Mortality in the United States
Graphic Jump Location
Thyroid Cancer Incidence and Mortality in the United States

While it was clear what was happening, it was less clear why it was happening.9 Some of the increase undoubtedly reflected incidental detection—serendipitous detection of thyroid cancer in the course of evaluating some other problem. The increased use of chest and neck computed tomography was one culprit; the increased use of carotid ultrasonography was another. But some of the detection was apparently purposeful: the result of a systematic physical examination of the neck followed by ultrasonographic evaluation of detected nodules. That sounds like screening.

Then, about a decade ago, public service announcements began to appear encouraging people to have their physicians “check your neck.” The Light of Life Foundation campaign featured actual testimonials of patients describing their positive health behaviors on the day before they were diagnosed with thyroid cancer. The ads used compelling language: “Thyroid cancer doesn’t care how healthy you are. It can happen to anyone. Including you. That’s why it is the fastest growing cancer in the US. Ask your doctor to check your neck. It could save your life.”10 The campaign’s title—and its main slogan—was “Confidence Kills.”

That’s a great public health message: if you feel good, you are about to die.

Patients with newly diagnosed thyroid cancer typically have been treated aggressively. As of 2013 in the United States, over 80% underwent total thyroidectomy and subsequently required lifelong thyroid replacement therapy. Nearly half received some form of radiation.7 Perhaps not surprisingly, among the corporate sponsors of the Confidence Kills campaign are corporations such as Sanofi Genzyme11—which sells an adjunctive pharmacological agent to accompany thyroid radiation therapy (Thyrogen).

But the poster child for how far wrong thyroid cancer screening can go is not the United States. It is South Korea.12 With the new millennium, many Korean providers chose to offer ultrasonographic screening to complement a government program of screening for other cancers. Within a decade, thyroid cancer incidence had increased 15-fold. Thyroid cancer became the most common cancer in Korea—more common than breast, colon and lung cancer.12

The epidemic was fueled by the misleading positive feedback that often accompanies cancer screening: a sense of urgency driven by rising incidence, a more favorable stage and size distribution, and a dramatic increase in 5-year survival.13 The positive feedback loop in thyroid cancer screening was further reinforced because the overdiagnosis harm of screening was construed as a benefit in media stories about “survivors”—individuals with screen-detected cancer who felt they owed their life to the screening test.14

This phenomenon shows no signs of reversing. In fact, a website promoting Korean medical tourism now advertises Korea as the place be treated for thyroid cancer—touting “the highest thyroid cancer survival rate in the world.”15 But just as in the United States, thyroid cancer mortality in Korea remains stable.

How would I summarize the effect of thyroid cancer screening? Massive overdiagnosis with no change in mortality.

The Task Force recommendation against thyroid cancer screening might have been an easy decision scientifically but not politically. Recently, the Task Force has been coming under attack.16 Last November, the House Energy and Commerce Health Subcommittee considered the USPSTF Transparency and Accountability Act in response to “growing concerns over a number of USPSTF recommendations”—particularly recent recommendations regarding breast and prostate cancer screening.17

Currently, the Task Force is an independent volunteer panel of 16 members who are nationally recognized experts in prevention, evidence-based medicine, and primary care. The proposed legislation would expand its composition to include specialty care clinicians as well as “relevant stakeholders from the medical products manufacturing community.”18 Would it really be helpful to include manufacturers of neck ultrasonographic equipment and thyroid replacement drugs in the evaluation of thyroid cancer screening?

Here is where the economists start worrying about regulatory capture.19 The few parties with a high-stakes interest in policy decisions can be expected to focus their resources in obtaining the recommendations they prefer, while the rest of us are less energized because we have only a tiny individual stake in the outcome. You might not think of the Task Force as a regulator, but that does not mean that it cannot be captured. And mass screening certainly meets the prerequisite of having financial benefit concentrated among the few, while the costs and harms are spread across the many.

There is a substantial amount of money involved in this decision; mass screening involves large swaths of the population—and with subsequent testing and additional treatments, it can be a gift that keeps on giving. The issue goes well beyond breast, prostate, and thyroid cancer screening. The biotech industry is devoting tremendous efforts to developing new diagnostic tools: genetic tests, biomarkers, immunosignatures, nanocytology, and liquid biopsies. While these may prove useful to a few high-risk patients, the most sought-after market will be mass screening, and the Task Force might be considered as a “supreme court” for screening: adjudicating competing claims based on a careful review of the scientific evidence. The few do not want this kind of scrutiny. But for the rest of us, it is of upmost importance that the Task Force remain independent of professional and financial interests. Let us not create a new swamp.

Back to top
Article Information

Corresponding Author: H. Gilbert Welch, MD, MPH, Dartmouth Institute of Health Policy and Clinical Practice, Geisel School of Medicine, 1 Medical Center Dr, Lebanon, NH 03766 (h.gilbert.welch@dartmouth.edu).

Published Online: May 9, 2017. doi:10.1001/jamainternmed.2017.1198

Conflict of Interest Disclosures: None reported.

Additional Contributions: I am indebted to Robert G. Hansen, PhD, for alerting me to the concept of regulatory capture.

References
References
1.
US Preventive Services Task Force.  Screening for thyroid cancer  [published May 9, 2017].  JAMA. doi:10.1001/jama.2017.4011Google Scholar
2.
Cole  WH, Slaughter  DP, Rossiter  LJ.  Potential dangers of nontoxic nodular goiter.  JAMA. 1945;127(14):883-888. doi:10.1001/jama.1945.02860140001001Google ScholarCrossref
3.
Hinton  JW, Lord  JW.  Is surgery indicated in all cases of nodular goiter, toxic and nontoxic?  JAMA. 1945;129(9):605-606. doi:10.1001/jama.1945.02860430021006Google ScholarCrossref
4.
Vanderlaan  WP.  The occurrence of carcinoma of the thyroid gland in autopsy material.  N Engl J Med. 1947;237(7):221-222.PubMedGoogle ScholarCrossref
5.
Harach  HR, Franssila  KO, Wasenius  VM.  Occult papillary carcinoma of the thyroid: a “normal” finding in Finland.  Cancer. 1985;56(3):531-538.PubMedGoogle ScholarCrossref
6.
Welch  HG.  Should I Be Tested for Cancer? Oakland, CA: University of California Press; 2006:79-82.
7.
SEER*Stat Database. Incidence—1973-2013. https://seer.cancer.gov/data/seerstat/nov2015/. Accessed April 7, 2017.
8.
SEER*Stat Database. Mortality—1969-2013. https://seer.cancer.gov/data/citation.html. Accessed April 7, 2017.
9.
Davies  L, Ouellette  M, Hunter  M, Welch  HG.  The increasing incidence of small thyroid cancers.  Laryngoscope. 2010;120(12):2446-2451.PubMedGoogle ScholarCrossref
10.
Light of Life Foundation. Confidence Kills. http://lightoflifefoundation.org/advocacy/campaigns/confidence-kills/. Accessed April 20, 2017.
11.
Light of Life Foundation. Sponsors and partners. http://lightoflifefoundation.org/about-us/sponsors-and-partners/. Accessed April 20, 2017.
12.
Ahn  HS, Kim  HJ, Welch  HG.  Korea’s thyroid-cancer “epidemic”.  N Engl J Med. 2014;371(19):1765-1767.PubMedGoogle ScholarCrossref
13.
Welch  HG, Schwartz  LM, Woloshin  S.  Are increasing 5-year survival rates evidence of success against cancer?  JAMA. 2000;283(22):2975-2978.PubMedGoogle ScholarCrossref
14.
Raffle  AE, Gray  JAM.  Screening: Evidence and Practice. Oxford: Oxford University Press; 2007.
15.
Koreakoreakorea. Thyroid cancer. http://www.koreakoreakorea.com/package/thyroid-cancer/. Accessed April 5, 2017.
16.
Allen  M. HHS Nominee Tom Price targeted panel that urged fewer cancer screenings. https://www.propublica.org/article/hhs-nominee-tom-price-targeted-panel-that-urged-fewer-cancer-screenings. Accessed April 5, 2017.
17.
Ault  A. GOP begins new scrutiny of US Preventive Services Task Force. http://www.medscape.com/viewarticle/872599. Accessed April 5, 2017.
18.
114th Congress (2015-2016). H.R.1151—USPSTF Transparency and Accountability Act of 2015. https://www.congress.gov/bill/114th-congress/house-bill/1151?q=. Accessed April 5, 2017.
19.
Wikipedia. Regulatory capture. https://en.wikipedia.org/wiki/Regulatory_capture. Accessed April 5, 2017.
×