Preventing overdiagnosis
How to stop harming healthy people
The article Preventing overdiagnosis: how to stop harming the healthy is published in the British Medical Journal (May 2012).
Translation (with small abbreviations) Evgenia Ryabtseva
Recently, there has been more and more evidence suggesting that medicine harms healthy people due to an increasingly early diagnosis and an increasing expansion of the concepts of various diseases. In connection with the announcement of an international conference dedicated to improving the understanding of the problem of overdiagnosis, Ray Moynihan, Jenny Doust and David Henry from Bond University (Robina, Queensland, Australia) and the Institute of Clinical Evidence-Based Sciences (Toronto, Canada) analyzed the causes of this phenomenon and possible methods of solving the problem.
The vaunted ability of medicine to help the sick has recently been increasingly compromised by its tendency to harm healthy people. Scientific publications are increasingly appearing that cause public alarm due to the fact that too many people are victims of overdoses (1), overtreatment (2) and overdiagnosis (3).
Screening programs detect early stages of cancer that would never lead to symptoms or death (4), sensitive diagnostic methods identify "anomalies" so minor that they would remain benign throughout a person's life (5). At the same time, the expansion of definitions of diseases means that even people with minor risks they acquire medical "labels" for life and receive treatment for the rest of their lives that does not benefit most of them (3, 6). According to statistics, more than $200 billion is wasted annually in the United States (7), while the cumulative burden of overdiagnosis poses a serious threat to human health.
In a narrow sense, overdiagnosis is the diagnosis of a disease in a person who does not have symptoms, which ultimately would not lead to the appearance of symptoms or to his premature death (3). In a broader sense, the term overdiagnosis applies to related problems of excessive medicalization and associated hypertreatment, distortion of diagnoses, displacement of the limits of the norm and erroneous detection of diseases. All this contributes to the fact that people who are in the low-risk group or have minor problems are treated as patients (8).
The adverse consequences of overdiagnosis include the negative impact of unnecessarily assigning people "labels", the harm caused by excessive diagnostic and therapeutic procedures, as well as the loss of financial and other resources that could be spent on the treatment or prevention of real diseases. The difficulty lies in determining the nature and scope of the problem in a broad sense, identifying its nature and causes, as well as in developing a set of reactions at different levels, ranging from clinical to cultural.
At the clinical level, the key goal is to identify clearer distinctions between benign abnormalities and anomalies that may cause harm to health in the future. To raise awareness among both the public and professionals, more reliable information is needed about the risk of overdiagnosis, especially regarding screening. If we look deeper, the accumulating evidence in favor of the fact that harm is inflicted on healthy people calls into question the belief in favor of early detection of cancer and updating (towards expansion) of disease criteria.
The International Scientific Conference "Prevention of Overdiagnosis", which aims to deepen understanding and awareness of the problem of overdiagnosis and its prevention, will take place from September 10 to 12, 2013 in the USA. The Dartmouth Institute of Health Policy and Clinical Practice, the British Medical Journal, the leading US consumer organization Consumer Reports and Bond University participate in its organization. Experts believe that the holding of this conference is very timely.
Hyperdiagnosis has many driving factors, including the best intentions, but the key point is the achievements of technology. The literature describes several broad and interrelated mechanisms of overdiagnosis: screening of people without symptoms, the use of highly sensitive tests for people with symptoms, the facts of accidental overdiagnosis – "incidentaloma" (from the English "incidentaloma"), as well as the expansion of the concept of "disease". These mechanisms are not mutually exclusive, and a stricter classification of various forms of overdiagnosis will be developed during the discussion at the 2013 conference.
Overdiagnosis as a result of screeningThis mechanism is involved in cases when, during a screening program, people without symptoms are diagnosed with a form of the disease that would never have led to symptoms or premature death.
Sometimes such forms of diseases are called pseudo-diseases. Contrary to the common view that cancer is dangerous in any case and ultimately fatal, some forms of malignant tumors may regress, not progress or grow so slowly that they will not have time to harm a person's health before his death from other causes (5). Randomized clinical trials and others will be discussed below studies in which populations that have been and have not been screened were compared. According to their results, a significant part of the malignant tumors identified in the framework of popular screening programs are pseudoassociations (4, 12). The results of the study of autopsy material indicate the existence in the general population of a large reservoir of subclinical diseases, including tumors of the prostate, mammary and thyroid glands, which in most cases do not cause harm to health [12]. Screening of the heart condition of people who do not have symptoms or are in a low-risk group can also lead to overdiagnosis of atherosclerosis of the coronary arteries and subsequent unnecessary interventions (13). Apparently, the nature and scale of overdiagnosis, as well as the number of pseudo-diseases detected as a result of screening, is limited today, but it continues to increase and, according to experts, "concern about overdiagnosis is justified" (14).
Increasing the sensitivity of testsPeople who turn to doctors with complaints of symptoms may also become victims of overdiagnosis, as the improvement of technologies and diagnostic methods makes it possible to detect less severe forms of diseases and disorders.
It becomes obvious that a significant part of the detected early "deviations" will not progress. This raises difficult questions regarding the identification of cases in which diagnostic markers and therapeutic approaches should be used, which have traditionally been used for much more severe forms of the disease.
IncidentalomsDuring diagnostic scanning of the abdominal cavity, pelvic organs, chest, head and neck, the proportion of patients in whom any "accidental findings" are detected during the examination for other reasons can reach 40% (15).
Some of these findings are tumors, and most of these "accidental" ("accidental tumors") are benign. Early accidental detection of a malignant tumor benefits only a very small number of patients, while the remaining majority suffers from anxiety and side effects of further examinations and treatment of "abnormalities" that would never harm them. As some authors have shown, the rapidly increasing incidence of certain types of cancer against the background of relatively stable mortality rates is a phenomenon indicating the prevalence of overdiagnosis due to screening or the detection of an incident – see Figure (12).
Detection and mortality rates from five types of cancer in the United States, 1975-2005
A – thyroid cancer, B – melanoma, C – kidney cancer, D – prostate, E – breastUnnecessarily lengthy definitions of diseases
Another mechanism of overdiagnosis is to expand the boundaries of the definition of diseases and reduce the thresholds for the use of therapy to the limits at which awarding a person with a medical diagnosis and subsequent treatment can do more harm than good to a person.
Changing diagnostic criteria for many conditions increases the number of people diagnosed with diseases (16), resulting in the presence of at least one chronic disease in practically all elderly people (17). This pattern extends to asymptomatic conditions fraught with the risk of developing negative manifestations, such as osteoporosis, in which treatment of patients at low risk of fractures may do more harm than good (18), as well as behavioral abnormalities, such as female sexual dysfunction, in which common problems are falsely classified as disorders (19).
Such changes in diagnostic criteria are made by commissions of health experts who have financial ties with companies that directly profit from increasing the number of patients (20). In the book Overdiagnosed – Overdiagnosed (3) published in 2011, Welch and co-authors argue that many people who are prescribed long-term therapy as a result of diagnosis, aimed at reducing slightly elevated cholesterol levels or eliminating minor manifestations of osteoporosis, in reality are victims of overdiagnosis, so the symptoms to prevent which treatment is prescribed, they have and without treatment, they would never have developed.
A similar form of overdiagnosis is observed in cases where the diagnosis is made outside the already expanded diagnostic criteria. This may be due to the fact that the inadequate standards indicated by the manufacturer exaggerate the frequency of occurrence of deviations (21), in cases when diagnostic methods determine random or normal fluctuations in biomarker levels as true deviations (22), or when important indicators are not taken into account during diagnosis (3).
Examples of overdiagnosisThe growing amount of information about overdiagnosis indicates that on a different scale this problem can occur when detecting a variety of diseases, including those for which there is a problem of underdiagnosis at the same time.
For a number of conditions, the evidence for the existence of overdiagnosis remains inconclusive, while for the rest they are becoming more and more compelling.
Breast cancerThe most convincing evidence of overdiagnosis was obtained by studying the results of screening conducted to detect breast cancer, but its estimated scale varies widely.
In 2007, the Lancet Oncology journal published a systematic review (24), the authors of which found that the proportion of cases of overdiagnosis of invasive breast cancer in women over 50 ranged from 1.7% to 54%. According to the estimates of Australian scientists (25), this figure is at least 30%, while a Norwegian study (26) determined it in the range of 15-25%. A systematic review published in 2009 in the BMJ (4) concluded that up to a third of breast cancer cases detected during screening can be regarded as cases of overdiagnosis. At the same time, even if there is convincing evidence obtained during population studies, it is currently impossible to distinguish between cancerous tumors that will harm the body in the future and those that will remain harmless.
Thyroid cancerWhile the risk of detecting "abnormalities" of the thyroid gland is quite high, the probability that it will harm a person's health is small (3, 27).
Analysis of the growing amount of evidence in favor of the fact that many of the newly detected cases of thyroid cancer are small, low-aggressive tumors that do not require therapy (28), which in itself is associated with the risk of nerve damage and the need for subsequent prolonged treatment (3).
Diabetes of pregnant womenThe revision of the criteria for the diagnosis of gestational diabetes made in 2010 recommended a significant reduction in the diagnostic threshold, which more than doubled (to almost 18%) increased the number of pregnant women with such a diagnosis (29).
Proponents of such changes claim that total screening according to the new definition will reduce the frequency of problems, including "the size of children is too large for their "gestational age" (29). At the same time, critics are calling for an urgent discussion of this issue before the larger implementation of the new definition, since many women may be subjected to overdiagnosis and hypertreatment. They note that with mild deviations, the test results are poorly reproducible, the evidence of the benefits of therapy for newly diagnosed pregnant women is very weak, and the benefits for their health are moderate at best (30, 31).
Chronic renal failureToday, more than 10% of the adult population of the United States lives with a diagnosis of chronic kidney failure in one form or another (32).
The current definition, introduced into practice as part of the new clinical guidelines (33), states that the estimated glomerular filtration rate, which remains below 60 ml/min/1.73 m2 for 3 or more months, should be considered abnormal. Critics claim that this innovation automatically creates the likelihood of overdiagnosis, especially among the elderly (34).
According to an article published last year by Winearls and Glassock, the new classification system is a kind of "fishing trawler" that "catches significantly more innocent objects than it should" (23). They estimate that up to a third of people over the age of 65 may meet the new criteria, although less than 1 in 1,000 a year will develop end-stage renal failure. The authors also point to the existence of serious problems with the reliability and reproducibility of the results of the assessment of glomerular filtration rate and express concern that many elderly people are diagnosed on the basis of a single potentially unreliable laboratory analysis. In another publication, they also claim that "most of the patients diagnosed with chronic renal failure do not have symptoms of detectable kidney disease," they also note that some organizations are trying to abandon the controversial new definition and raise the threshold values of diagnostic indicators, which significantly reduces the frequency of detection of the disease (35). In response to criticism, proponents of innovations defend the new definition as "clear, simple and useful" (36).
AsthmaDespite the fact that asthma can be a serious disease and there is a risk of its underdiagnosis and insufficient treatment, a number of studies indicate that in this case there may be a significant frequency of overdiagnosis.
One large study conducted in 2008 showed that almost 30% of people who were diagnosed with asthma did not have this disease, and 66% of them did not need medication for asthma during the 6-month follow-up period (37). The authors concluded that "a significant proportion of people... an overdiagnosis of asthma can be carried out and medications for its treatment can be prescribed unnecessarily." In the same year, a Dutch study showed that out of 1,100 patients using aerosols of corticosteroid drugs, 30% can use these drugs without obvious indications (38).
Pulmonary embolismDoctors consider pulmonary embolism to be a diagnosis that is "unacceptable to miss", as it can have disastrous consequences.
Historically, this disease was diagnosed only in cases when the blockade was so strong that it caused a heart attack of a part of the lung or hemodynamic instability. Treatment of such patients with anticoagulants or thrombolytic drugs was considered mandatory. Today, lung angiography using a CT scanner allows detecting small blood clots, and doctors still do not agree whether treatment is always necessary in such cases (39). The analysis of trends observed before and after the introduction of this diagnostic method into clinical practice by Weiner and co-authors showed that the almost doubled number of detected cases is "a reflection of the epidemic of diagnostic examinations that gave rise to overdiagnosis." At the same time, most of the additionally detected cases are "clinically insignificant" and "would not have become fatal even without diagnosis and treatment" (40). Currently, an observational study is being conducted to study the safety of non-treatment of people who have very small blood clots (41).
Hyperactivity and attention deficit disorderMuch has been written about the expansion of diagnostic definitions of mental illness and the potential danger of overtreatment (42).
The debate on this issue has intensified due to the emergence of statements according to which modern approaches to diagnosis can contribute to the spread of overdiagnosis of diseases such as bipolar disorder, autism and hyperactivity and attention deficit disorder (43, 44). One of the reasons for concern is the possible overdiagnosis in children who do not have the right to vote on the fairness of a diagnosis that can break from life. This is most relevant in relation to hyperactivity and attention deficit disorder (45). A recent study conducted in Canada, which included almost a million children, showed that boys born in December (usually the youngest in the class) are 30% more likely to be diagnosed with this and 40% more likely to be prescribed appropriate medications than boys born in January. Based on this, the authors state that their findings "raise concerns about the potential harm caused by overdiagnosis and excessive prescribing of treatment" (46).
Forces driving overdiagnosisThe factors driving overdiagnosis lie deep within the medical culture and the wider public, which makes the scale of the problems associated with any attempts to combat these mechanisms not obvious.
The main factor is the development of technology. In 1998, Black wrote that the possibility of detecting minor deviations definitely leads to an increase in the frequency of occurrence of any disease. This, in turn, leads to an overestimation of the benefits of therapy, since milder forms of diseases are treated, and improvements in health status are mistakenly recognized as a success of treatment. This creates a "false feedback loop" that triggers a vicious "cycle of intensification of diagnosis and treatment, which, in the end, can do more harm than good" (5).
Industries profiting from the expansion of diagnostic and treatment markets have a great impact on medical professionals and the wider public through the formation of monetary ties with professionals and patients and the financing of advertising, the target group of which are direct consumers; research funds; campaigns to increase awareness of diseases and medical education (8). What is the most It is important that the members of the commissions that formulate definitions of diseases and threshold therapeutic indicators often have financial ties with companies that are interested in increasing profits by expanding the market (20). Similarly, doctors and their associations may be interested in increasing the patient population within their specialization, and the involvement of doctors in diagnostic and therapeutic technologies, in which they may have a commercial interest, may also lead to unnecessary diagnostics.
The desire to avoid lawsuits and the psychology of regret is another driving force behind overdiagnosis, as doctors can be punished for ignoring early symptoms of the disease, which, as a rule, does not happen with overdiagnosis. Indicators of the quality of work equating to the treatment of a larger number of patients can also stimulate overdiagnosis in order to receive remuneration.
An intuitive belief in early detection of diseases, fueled by a deep trust in medical technology, is at the center of the problem of overdiagnosis. We are gradually beginning to regard simply entering the "risk" group for the development of the disease in the future as a disease itself. Starting with the treatment of high blood pressure, which became "fashionable" in the middle of the 20th century, an increasing proportion of the healthy population is being diagnosed and treated for an increasing list of asymptomatic conditions, justified only by the estimated risk of their occurrence in the future. Despite the fact that this approach has reduced suffering and prolonged the lives of many patients, for people who have become victims of overdiagnosis, it has unnecessarily turned life experience into a tangled network of chronic diseases. The cultural norm of "the more, the better" is confirmed by recent evidence that patient satisfaction depends on increased access to diagnosis and treatment, even though more active medical intervention may be associated with causing more harm (49, 50).
ConclusionAnxiety about overdiagnosis does not negate the understanding that many people suffer from a lack of the medical care they need.
On the other hand, the resources spent on unnecessary medical procedures can be much more usefully used for the treatment and prevention of real diseases. The problem lies in distinguishing between one and the other, as well as in obtaining and distributing information that will help people make more informed decisions in cases where diagnosis and treatment can cause more harm than good.
The list of references to the article is given in a separate file.
Portal "Eternal youth" http://vechnayamolodost.ru
13.07.2012