What is the right dose of modern medicine?

A recent set of papers published in The Lancet makes for sobering reading in relation to both the over-use and under-use of modern medical care^1–5. The authors provide evidence for both over-use and under-use in high-income countries and low- and middle-income countries, alike. Under-use results in preventable loss of life, disability and suffering³. Over-use results in, at the very least, a staggering waste of resources² but can also result in net harm. Estimation of the extent of the over- and under-use of medical care is challenging and the causes are complex, although some of the drivers have been identified as inequalities in information, wealth and power⁵.

How do these concepts of over- and under-use of medical care apply specifically to the health of women? There are plenty of examples where the under-use of modern medicine remains an issue such as the fact that, in India, cervical cancer remains the commonest cause of cancer death in women aged 30–69 years⁶ when there are effective methods of primary (vaccination) and secondary prevention (screening)⁷. Regrettably, over-use of medicine also plagues the care of women in the same setting, as evidenced by high rates of hysterectomy in the private sector in the absence of clinical indications⁸.

One of the most challenging forms of over-use arises from over-diagnosis, and the over-treatment that follows. Over-diagnosis is the identification of a condition which, in the absence of screening, would not have become clinically evident in the person’s lifetime⁹. If you are in the over-diagnosed group and you are treated, treatment is all harm. Depending on the magnitude of the over-diagnosis and the amount of harm caused by treatment, the impact of screening for a condition may be net harm¹⁰. The greatest problem, up to now, for women in terms of over-diagnosis has been mammographic screening for breast cancer¹¹. However, a new threat for women has emerged in the over-diagnosis of thyroid cancer.

There have been considerable increases in the incidence of thyroid cancer in the last three decades, in both men and women, but, in some countries, the increase has been greater in women than in men¹². Despite the recent changes in incidence, mortality from the disease remains low¹³. Historically, the incidence of thyroid cancer has been about three times higher in women than in men¹⁴, although this ratio has changed in some countries recently¹². Thyroid cancer is the sixth most common cancer in women in the United States¹⁵ and the seventh most common in Australia¹⁶ but, if current trends continue, it is likely to move up the incidence tables¹⁷. Most of the increase in incidence has occurred in small papillary carcinomas¹³ where the outlook in terms of prognosis is excellent¹⁸.

There appears to be no adequate explanation for the upward trend in incidence in terms of changes in known risk factors such as childhood exposure to ionizing radiation, trends in diabetes, obesity or reproductive patterns or the availability of nutritional iodine¹⁹. The most likely explanation is a change in the way the condition is diagnosed in the context of what is now understood to be a large reservoir of cases of subclinical disease¹³. In terms of the reservoir of disease, a study of 101 consecutive autopsies in Finland found that small papillary cancers were found in one-third of cases and that tumors were commonly multi-focal²⁰.

Before relatively recent changes in the use of imaging techniques, thyroid cancer presented clinically as a nodule which was either visible or palpable in the neck. Now, it is more likely that a thyroid nodule will be identified incidentally following imaging for another indication (chest CT or MRI of the neck) than present clinically²¹. Another contributing factor in terms of imaging is the ease with which ultrasound can be applied to the neck²¹.

So how large a problem is the over-diagnosis of thyroid cancer? The answer depends on the country, but modeling from countries with cancer registries, which allow the creation of age-specific trends in thyroid cancer starting before the introduction of ultrasound of the neck, suggest that it is of the order of 90% for South Korea, 70–80% for the United States, Italy, France and Australia, and 50% in countries such as England, Scotland, Denmark, Finland, Norway and Sweden^12,22. What distinguishes South Korea is the participation of adults in a cancer screening program which includes thyroid ultrasonography, and the people most likely to participate in this program are women aged 50–59 years (26%)²². The ratio of women to men in terms of thyroid cancer incidence in Korea has now reached 5.3¹².

Thyroid cancer is a heterogeneous condition and conventional management has included total or partial thyroidectomy, with or without node dissection, followed, where indicated, by radiation therapy or TSH suppression²³. However, change is afoot with an argument that conventional management of small follicular variant papillary thyroid cancers might violate the dictum of 'first do no harm’²⁴. There has been a proposal to rename small encapsulated follicular variant papillary thyroid cancers as non-invasive follicular thyroid neoplasms with papillary-like nuclear features (NIFTP) which will allow a 'restrained risk-adapted' approach to managment²⁴, which will likely include the option of active surveillance¹⁸ rather than immediate surgery.

Even for those people who have had primary treatment for a differentiated thyroid cancer, extensive use of various imaging modalities for follow-up may be contributing to treatment for recurrence without clear evidence of improved disease-specific survival²⁵.

Drivers for overuse of medicine are complex and include powerful financial factors. However, within the individual doctor–patient interaction, patient factors include beliefs about the value of medical tests and clinician factors such as technological enthusiasm¹. A recent systematic review has shown that doctors tend to under-estimate harm from cancer screening²⁶. Possible reasons for this include a ‘preoccupation with pathophysiologic mechanisms’, rather than ‘evidence-based effectiveness’, or it may just be a case of not keeping up with the evidence. The authors of the review also acknowledge the ‘therapeutic illusion’ (unjustified enthusiasm for treatment on the part of both doctors and patients)²⁶. However, other contributing factors for the doctors include medicolegal concerns, anticipated regret and commission bias²⁶. The Lancet series suggests that, in order to get the dose of modern medicine right, we need the public to be educated, engaged and empowered¹ but that doctors have an obligation to lead the way⁵. We will need to do things which are uncomfortable such as the re-naming of conditions to avoid the word cancer, the studies of watchful waiting in appropriate patient groups and, perhaps most importantly, resist the temptation to apply an ultrasound machine to the neck of a woman who is asymptomatic.

Conflict of interest

The author reports no conflict of interest. The author alone is responsible for the content and writing of this paper.

Source of funding

Nil.