Optimizing TSH Testing: Minimizing Overdiagnosis and Unnecessary Interventions

Abstract

In this chapter, the routine application of thyroid function tests for hypothyroidism detection is evaluated, with a focus on the problems of overdiagnosis and unnecessary treatment, especially in subclinical scenarios. It critically assesses the lack of solid evidence behind common interventions, like the widespread prescription of levothyroxine without definitive clinical need. Highlighting evidence-based guidelines for hypothyroidism screening, the chapter argues against indiscriminate thyroid testing. It advocates for a thoughtful approach to thyroid disorder management, urging cautious use of TSH tests to reduce needless medical actions and improve patient outcomes.

1. Introduction

The journey from clinical trial discoveries to their practical application in healthcare is remarkably slow, often taking up to 17 years. This slow process reveals a significant healthcare paradox: the hesitancy to adopt proven interventions while outdated or less effective practices continue, particularly in thyroid health management [1, 2]. This situation points to a significant gap between knowledge generation and its application, underscoring a critical area for improvement across all medical areas, including endocrinology [3]. The frequent use of Thyroid Stimulating Hormone (TSH) tests has come under scrutiny for contributing to the overdiagnosis and unnecessary treatment of conditions like subclinical hypothyroidism, which remains a contentious area in medical practice [4, 5, 6]. It advocates for a critical reassessment of TSH testing’s role, aiming to align it more closely with the latest evidence and clinical guidelines without necessarily proposing new protocols but emphasizing the importance of evidence-based practice [7, 8, 9].

2. Challenges in hypothyroidism screening

In healthcare, screening is intended to detect diseases in apparently healthy individuals with risk factors, aiming to decrease morbidity and mortality through early intervention. In contrast, early diagnosis seeks to promptly identify diseases in individuals already exhibiting symptoms. The foundation for modern screening practices was established in 1968 by Wilson and Jungner in a seminal WHO publication, where they outlined key criteria [10]. Applying these principles to hypothyroidism screening reveals several challenges: limitations in the diagnostic accuracy of the TSH test, a weak association between symptoms and hypothyroidism diagnosis, and the inconsistent relationship between minor TSH elevations and adverse outcomes. Additionally, treatment with levothyroxine often lacks significant efficacy, leading to recommendations against treating most patients with subclinical hypothyroidism [5, 11]. These complexities are detailed in Table 1, which applies Wilson and Jungner’s screening criteria to hypothyroidism, revealing the unique challenges in this field.

Principle	Assessment for hypothyroidism
1. The condition should be a significant health problem	Manifest hypothyroidism is significant due to its potential complications. However, the clinical relevance of subclinical hypothyroidism is debatable since its treatment has not shown to improve significant clinical outcomes in controlled trials.
2. There should be an accepted treatment for patients with recognized disease	Treatment with levothyroxine is common but does not necessarily improve clinical outcomes in subclinical hypothyroidism.
3. Facilities for diagnosis and treatment should be available	Broad access to TSH testing can promote its overuse, increasing the risk of unnecessary treatments that negatively impact patients and the healthcare system.
4. There should be a recognizable latent or early symptomatic stage	Early identification of hypothyroidism is complicated because general symptoms, such as fatigue and cold sensitivity, are not exclusive to this condition, making diagnosis based solely on clinical signs challenging.
5. There should be a suitable test or examination	TSH variability without proper age adjustments can lead to erroneous diagnoses and an increase in overdiagnosis, particularly in older populations.
6. The test should be acceptable to the population	The TSH test is a minimally invasive and low-risk procedure.
7. The natural history of the condition should be adequately understood	Uncertainty regarding the relationship between TSH levels and nonspecific symptoms, along with unclear therapeutic benefits, poses a challenge to understanding the natural course of subclinical hypothyroidism.
8. There should be an agreed policy on whom to treat as patients	There is widespread consensus against universal screening tests in the general population.
9. The cost of case finding should be economically balanced in relation to possible expenditure on medical care as a whole	Broad screening with TSH tests can lead to substantial expenses and the unjustified classification of patients, necessitating regular consultations, levothyroxine treatment, additional expenses, and more medical examinations.

Table 1.

Assessment of Wilson and Jungner’s principles for population screening of hypothyroidism.

2.1 Limitations in TSH and symptom use for hypothyroidism diagnosis

TSH is critical for screening and early diagnosis of hypothyroidism. However, interpreting its levels is complex due to variations caused by circadian rhythms, individual variability, and the effects of aging [12, 13, 14]. A longitudinal study observed individuals without thyroid conditions for 5 years, revealing that 62% of TSH levels initially ranging from 5.5 mIU/L to 10 mIU/L naturally returned to normal in later evaluations without any medical intervention [15]. This variability underscores the importance of confirming abnormal TSH levels with a subsequent measurement before initiating any treatment [4]. Adhering to the guidelines from the National Health and Nutrition Examination Survey (NHANES III), which recommend a TSH upper threshold of 7.5 mIU/L for individuals older than 80, is crucial. Ignoring the necessity for age-based threshold adjustments could inadvertently increase the likelihood of overdiagnosis and consequent unnecessary treatments, along with their potential adverse effects [3, 16].

Moreover, TSH diagnostic accuracy is influenced by pretest probability. It reliably indicates thyroid dysfunction in high-suspicion cases, such as patients with a history of thyroid surgery or using medications affecting the thyroid. However, its precision diminishes when solely assessing symptoms like fatigue, common in hypothyroidism [17, 18, 19]. Symptoms alone lack specificity, as population studies show no significant increase in mild thyroid dysfunction symptomatology compared to normal function [8]. Additionally, up to 25% of individuals without thyroid disorders may exhibit symptoms erroneously interpreted as hypothyroidism [20].

2.2 Adverse outcomes of subclinical hypothyroidism and effects of levothyroxine treatment

Observational studies suggest that TSH levels above 10 mIU/L are associated with an increased risk of cardiac events, such as heart failure and coronary artery disease [6]. However, for nonpregnant adults with subclinical hypothyroidism and TSH levels at or below 10 mIU/L, there is no consistent evidence linking these slight TSH elevations to an increased risk of mortality, cardiovascular events, cognitive impairment, frailty, decreased bone mineral density, worsened quality of life, or an increased occurrence of symptoms [4, 6, 21].

Moreover, substantial evidence supporting the effectiveness of levothyroxine treatment in these patients is lacking [22]. Results from various clinical trials, including the TRUST trial that enrolled 737 older patients with mild TSH elevations in a randomized, double-blind, placebo-controlled study, revealed no significant improvements in symptoms such as fatigue, quality of life, cognitive function, or cardiometabolic parameters after normalizing TSH levels with 12 months of levothyroxine use [23].

Despite these findings, the use of levothyroxine has significantly increased in many countries, with about 7% of the U.S. population now receiving this hormone therapy [24, 25]. Often, the indications for initiating thyroid hormone therapy include inappropriate criteria, such as patients with normal TSH levels, management of mild subclinical hypothyroidism, suppression of thyroid nodule growth, or addressing nonspecific symptoms like depression, weight loss, and fatigue [3, 26]. A study by Brito and colleagues examining levothyroxine use between 2008 and 2018 in 58,706 commercially insured adult patients and Medicare beneficiaries found that 30.5% began levothyroxine therapy despite having normal hormone levels, 61% for subclinical hypothyroidism, and 8.4% for overt hypothyroidism. Among those treated for subclinical hypothyroidism, the majority (57.9%) had TSH levels below 10 mIU/L, with primary care physicians initiating most prescriptions (47.5%) [26].

As previously discussed, the treatment for subclinical hypothyroidism, particularly with levothyroxine, does not clearly demonstrate benefits and is associated with significant adverse effects. The trend toward lowering the threshold for treating mild subclinical conditions has led to an increase in iatrogenic medical complications. Specifically, levothyroxine treatment resulting in suppressed TSH levels is linked to a heightened risk of arrhythmias and bone fractures, with hazard ratios of 1.60 (95% CI: 1.10–2.33) and 2.02 ([95% CI: 1.55–2.62]), respectively, compared to patients maintaining normal TSH levels [27]. Furthermore, research shows that 40–50% of patients over 65 on levothyroxine have TSH levels lower than 0.45 mIU/L, indicating widespread overtreatment [28]. This not only increases the burden of medication and its associated costs but also necessitates regular doctor visits, frequent blood tests, and lifestyle adjustments for taking medication on an empty stomach [29]. Importantly, once started, about 90% of patients with subclinical hypothyroidism are likely to require lifelong treatment, emphasizing the need for careful evaluation before initiating levothyroxine therapy [30].

2.3 Current guidelines for hypothyroidism screening

The perspectives of medical organizations on universal screening for hypothyroidism vary considerably. While the American Thyroid Association advocates for assessments every 5 years to identify thyroid dysfunction in asymptomatic individuals aged over 35, potentially impacting around 3.5 billion adults worldwide [28, 31], other medical organizations question this approach for nonpregnant adults [17, 32]. These guidelines, as outlined in Table 2, provide insights into the recommendations set forth by various medical bodies regarding hypothyroidism screening practices.

Organization	Current guidelines
USPSTF (United States Preventive Services Task Force) [17]	Insufficient evidence to assess universal screening.
ATA (American Thyroid Association), AACE (American Association of Clinical Endocrinology) [27]	Identifying high-risk individuals and evaluating thyroid function every 5 years in asymptomatic individuals over 35.
Association for Clinical Biochemistry (ACB), the British Thyroid Association (BTA), and the British Thyroid Foundation (BTF) [30]	Do not endorse routine universal screening. Instead, they recommend identifying cases in women with nonspecific symptoms suggestive of an underlying thyroid disorder
Canadian Task Force on Preventive Health Care [29]	Do not perform screening tests for thyroid dysfunction in asymptomatic nonpregnant adults

Table 2.

Guidelines for hypothyroidism screening by medical organizations.

The United States Preventive Services Task Force (USPSTF), in its 2015 review, determined insufficient evidence to support mass thyroid screening in asymptomatic adults, citing a lack of conclusive data on its effect in preventing significant adverse outcomes such as cardiovascular diseases or improving cognitive function. Concerns about overdiagnosis and overtreatment were also emphasized [17].

Conversely, health authorities in the United Kingdom, including the Association for Clinical Biochemistry, British Thyroid Association, and British Thyroid Foundation, do not advocate for screening in the general healthy population. Instead, they promote an early diagnostic approach in high-risk individuals with indicative symptoms [33]. Despite these guidelines, up to 30% of individuals over 65 undergo thyroid evaluations annually [15, 34, 35]. An analysis of 2936 subjects in this age range, with normal or slightly altered thyroid results, revealed that less than 0.5% progressed to overt thyroid dysfunction within 5 years, raising questions about the necessity of screening in this age group [34]. This situation underscores the discrepancy between current medical practice and clinical consensus, highlighting the importance of reassessing screening practices to avoid unnecessary medical interventions [3].

3. Challenges and strategies for de-implementation

3.1 Barriers to de-implementation

De-implementation emerges as a critical strategy for reducing low-value care, such as the excessive use of TSH testing [36]. Its primary objectives are to minimize patient harm, optimize resource utilization, and enhance overall population health [37]. However, it encounters significant hurdles, including resistance and conflicting motivations from healthcare providers and patients, as well as external influences stemming from economic, political, cultural, and historical factors [38]. To tackle these barriers, a series of comprehensive approaches is outlined in Table 3. These strategies aim to navigate and overcome the obstacles to de-implementation, ultimately promoting the judicious use of the TSH test.

Category	Strategy/insight	Implementation challenge	Example application
De-implementation challenges	Overcoming resistance from healthcare providers and patients; navigating economic, political, and cultural influences.	Resistance to change due to established practices and external pressures.	Implementing a multifaceted educational campaign to address misconceptions and emphasize the evidence-based approach to TSH testing.
Education and training	Updating healthcare professionals on current guidelines; emphasizing high-value care in medical training	Knowledge gaps; inertia in updating practices according to new evidence.	Hosting workshops and seminars to disseminate the latest research and guidelines on when TSH testing and treatment are warranted.
EHR-based tools	Utilizing CPOE and CDSS to guide clinical decisions and reduce inappropriate interventions.	Ensuring EHR tools are user-friendly and effectively integrated into clinical workflows.	Integrating alerts into EHRs that prompt for clinical justification when ordering TSH tests, especially for borderline cases.
Audit and feedback	Regular performance evaluations against standards to encourage adherence to evidence-based practices.	Establishing meaningful benchmarks and ensuring feedback is constructive and actionable.	Providing quarterly reports to clinicians comparing their TSH testing patterns with best practice guidelines, highlighting areas for improvement.
Patient engagement	Involving patients in care decisions, particularly regarding the necessity and implications of TSH testing.	Balancing educational efforts to ensure patients have realistic expectations without undermining trust.	Developing patient education materials that detail the potential risks of overtreatment and the importance of evidence-based testing.
Institutional policy adjustments	Revising policies to limit TSH testing to clinically justified scenarios.	Aligning institutional policies with current evidence while managing stakeholder interests.	Instituting a review process for TSH test orders to ensure compliance with updated clinical guidelines.
Assessing sustainability	Evaluating long-term adherence to de-implementation strategies and maintaining practice change.	Addressing persistent or resurfacing inappropriate testing practices over time	Establishing an annual review process to assess the impact of de-implementation strategies on TSH testing rates and identify areas for ongoing improvement.
Cost-effectiveness and broader analysis	Considering the long-term financial impact of reducing unnecessary TSH tests.	Demonstrating the value of de-implementation beyond immediate cost savings.	Conducting a cost-benefit analysis to showcase the long-term savings from avoiding unnecessary treatments and complications.
Research and innovation	Encouraging research on TSH testing impacts and developing innovative approaches to reduce unnecessary testing.	Securing funding and interest for research focused on diagnostic stewardship.	Pilot studies assessing the impact of targeted educational programs on TSH testing rates.

Table 3.

Enhanced strategies for optimizing TSH testing.

Norton and Chambers have underscored the multifaceted nature of the de-implementation process, where diverse factors exert influence across different levels [39]. Patient preferences significantly impact prescribing practices, posing challenges for physicians striving to adhere to evidence-based care [9]. Cognitive dissonance arises when physicians recognize that certain interventions, such as thyroid screenings or treatments for subclinical hypothyroidism, may not benefit specific patients. However, addressing patient concerns, potential litigation, and the importance of maintaining the doctor-patient relationship complicate decision-making [40]. This internal conflict, compounded by external pressures such as malpractice concerns, fear of missing a diagnosis, and the need to meet patient expectations, is further influenced by habituation, extensive training, and years of practice [41].

Given these complexities, sociological theories, like Rogers’ theory of the diffusion of innovations, suggest that the majority of physicians, around 84%, are often slow to adopt changes in their clinical practice [42]. Considering this delay in adopting new guidelines is crucial for developing effective communication strategies and ongoing training programs. These efforts should directly target perceived barriers and aim to expedite the adoption of evidence-based practices [43].

3.2 Strategies for de-implementation

The field of de-implementation is advancing with the emergence of more effective evidence-based strategies [44]. Addressing the identified challenges requires a comprehensive approach that involves the practices of healthcare professionals, incorporates the perspectives and needs of patients, and updates institutional policies [44, 45].

Several de-implementation strategies have been explored in the literature, which include support for clinical decision-making, education of both physicians and patients, and the implementation of multicomponent interventions, as reported in a systematic review that identified nine strategies for reducing low-value care. Additionally, other mentioned strategies include cost-sharing, which shifts the financial burden of low-value care onto patients; provider performance reports, providing patients with information about their healthcare provider’s use of low-value care services; pay-for-performance, offering financial incentives to physicians; insurer restrictions that limit reimbursement for low-value care; risk-sharing; and provider feedback, also identified as effective interventions [46].

The majority of evidence suggests that multicomponent strategies are effective and could be applied in de-implementation programs targeting the indiscriminate use of TSH.

3.2.1 Education and training for healthcare professionals

The factors influencing healthcare professionals are closely tied to their beliefs, knowledge, and skills [47]. It is crucial to acknowledge that clinicians often encounter challenges in staying updated with scientific literature and may be influenced by habits acquired during their medical training, which typically prioritizes diagnosis over “do nothing” [48, 49, 50]. Therefore, resident training presents a critical opportunity to introduce the concept of high-value care, as behaviors learned during this period tend to persist throughout one’s professional career [49].

Although some medical schools have integrated healthcare economics content into their curricula, many physicians still have significant knowledge gaps in these areas. Training programs frequently overlook resource management and cost-conscious practices, essential elements for promoting more efficient, evidence-based healthcare [50].

Recent reviews have indicated that the majority of de-implementation strategies focus on educational approaches aligned with the “Training and Education of Stakeholders” category of the ERIC (Expert Recommendations for Implementing Change) framework, emphasizing direct training and widespread dissemination of educational materials as key strategies [45]. To discourage the indiscriminate ordering of TSH tests, various educational strategies can be implemented, such as focused training sessions on updated guidelines, distribution of educational materials, and conducting periodic assessments. These initiatives should be complemented with specific, measurable, achievable, relevant, and time-bound (SMART) objectives [47]. For instance, evaluating the impact on reducing TSH tests by 15% among primary care physicians over the next 6 months would provide tangible and measurable results within a defined timeframe.

However, it is essential to recognize that education alone may not suffice to change deeply entrenched practices [1]. Research indicates that relying solely on education for de-implementation often yields limited effectiveness, emphasizing the need for a multifaceted, multi-level approach in current strategies, which has shown to yield more promising results [51]. Moreover, the implementation process tends to accelerate significantly when clear policies and effective communication channels are established among research, Pharmacy Benefit Management (PBM) Services, and clinical practice [1]. Therefore, it is critical to seamlessly integrate educational strategies into a comprehensive framework to ensure their effectiveness.

3.2.2 Use of EHR-based tools

The utilization of Electronic Health Record (EHR)-based tools, including Computerized Physician Order Entry (CPOE) and Clinical Decision Support Systems (CDSS), emerges as a powerful strategy to influence medical practice [45]. These tools rely on theories of individual and organizational learning, facilitating the implementation of effective strategies to replace inappropriate interventions with evidence-based ones [36]. However, it is paramount for these tools to remain intuitive and straightforward, incorporating contextual alerts that guide physicians in real-time and allowing for customization to meet patient needs. This ensures that the process does not deteriorate into a low-value management endeavor [52].

To illustrate, there are various approaches to reducing the number of requested TSH tests. For instance, implementing an alert within the CPOE system would trigger when a physician requests a TSH test without a clear indication or adherence to established clinical guidelines. This alert would prompt the physician to provide a specific justification for the TSH test order before proceeding. Additionally, regularly reviewing predefined order lists in the EHR system, which include the TSH test, proves effective [36, 44].

Clinical Decision Support Systems (CDSS) represent advanced tools for influencing behavior. They deliver real-time alerts on best practices, reminders, and evidence-based recommendations. For example, integrating alerts or reminders into CDSS prompts clinicians to reassess their decisions to request TSH tests without indication and opt for more appropriate interventions aligned with current best practices and guidelines, such as conducting TSH tests in populations truly at risk and benefiting from levothyroxine use [44]. In summary, effectively leveraging these EHR-based tools offers a potent avenue for enhancing clinical decision-making and optimizing patient care while also contributing to the de-implementation of indiscriminate TSH testing [36, 44].

3.2.3 Audit and feedback

Audit and feedback are essential tools in assessing professional performance, as they are evidence-based and designed to compare individual practice with established professional standards and objectives [53]. This evaluation process, known as “auditing” performance, is communicated both individually and collectively to motivate compliance with professional norms and standards, such as following recommendations to limit hypothyroidism screening or overuse of levothyroxine in various contexts [44, 53, 54].

Additionally, accessible dashboards can be utilized in this audit, allowing clinicians to compare their performance with others and pre-established benchmarks. These dashboards not only facilitate comparison but also provide an opportunity for clinicians to self-audit their performance and reflect on their own practices [44]. This, in turn, promotes continuous improvement in the quality of medical care provided [53].

3.2.4 Strengthening patient engagement in de-implementation processes

Engaging patients in the process of removing unnecessary medical practices through educational efforts and shared decision-making is key in medical settings, particularly when patients hold specific beliefs or expectations about certain treatments [44]. In scenarios like subclinical hypothyroidism, providing patients with thorough information about the risks, benefits, and alternatives to levothyroxine treatment helps to realign their expectations with evidence-based standards [45, 55]. Evidence from a systematic review and meta-analysis confirms that actively including patients and ensuring they have access to relevant information are effective strategies for phasing out unnecessary medical interventions [56]. These methods foster meaningful interactions between patients and healthcare providers, creating a culture of shared decision-making that effectively reduces the use of treatments that are not essential [56].

3.2.5 Institutional policy adjustment

To address the issue of excessive TSH screening and the overuse of LT4, it is crucial to implement changes in institutional policies. This could involve restricting access to these practices in cases where they are not supported by medical evidence. By establishing specific guidelines that limit the use of TSH testing and the prescription of LT4 to clinical scenarios where they are clearly indicated, we can promote a more evidence-based and patient-centered approach to medical care [36, 44].

4. Assessing long-term sustainability of de-implementation strategies

The evaluation of de-implementation and its long-term sustainability demands a comprehensive approach. While physicians and patients may acknowledge the risks associated with indiscriminate TSH testing, various psychological, cultural, and social factors may perpetuate its request, necessitating attention to overcome this barrier [36].

A thorough examination of cost-effectiveness is essential [9]. While the upfront expense of a TSH test is clear, it is imperative to consider broader implications for long-term savings. By forecasting outcomes for patients with undiagnosed subclinical hypothyroidism, potential cost savings become apparent. These savings stem from avoiding expenses associated with levothyroxine treatment, follow-up appointments, and complications arising from overdosing, such as hospitalizations for cardiac arrhythmias or costs related to bone fractures resulting from chronic levothyroxine overdose.

Furthermore, de-implementation represents an ongoing and dynamic process. Even if successful initially, it is crucial to assess its penetration, indicating the extent to which the practice discontinues across all levels of the healthcare system, and its sustainability, reflecting the ability to maintain the interruption of the practice over time [9].

An integral aspect of de-implementation science involves considering potential unintended consequences [36]. For instance, discontinuing a previously recommended intervention, such as including TSH testing in routine check-ups, may lead to patients losing confidence in the medical system. Additionally, clinicians may compensate for de-implementation by increasing the use of other medical interventions, such as conducting additional blood tests like free T4 or anti-TPO alongside TSH, or requesting thyroid ultrasounds in response to patient requests.

5. Conclusion

The significance of optimizing thyroid function tests to prevent unnecessary diagnoses and treatments, especially in cases of subclinical hypothyroidism, has been emphasized in this chapter. It is evident that adopting an individualized and risk-based approach to patient selection for thyroid function testing is crucial, given the variability of TSH levels and their limited correlation with symptoms. Despite the challenges in reducing the overuse of TSH testing, specific strategies, such as healthcare professional education, leveraging electronic health record-based tools, and enhancing patient engagement in the de-implementation process, have been identified as essential steps toward successfully de-implementing low-value medical practices.

Looking ahead, it is imperative to address the limitations present in current literature, including inconsistencies in terminology. Moreover, exploring variations in de-implementation strategies based on taxonomy, along with investigating combined de-implementation approaches and conducting further qualitative research to understand the preferences of physicians, payers, and patients, is crucial avenues for future exploration. By pursuing these paths, tailored interventions can be developed to effectively discourage the use of low-value medical practices across diverse geographical and economic contexts, thereby enhancing healthcare quality and promoting the efficient utilization of healthcare resources. Advancements in these proposed future directions hold the potential to optimize TSH testing, minimizing overdiagnosis and unnecessary interventions.

Acknowledgments

I express my heartfelt gratitude to my family members for their unwavering support and encouragement throughout the completion of this work.

Conflict of interest

The author declares no conflict of interest. This work is not funded.