Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

FormalPara Case

Sadie was a 45-year-old woman who presented with three days of right-sided weakness and numbness. She had a history of systemic lupus erythematosus and had a prior episode of visual loss due to inflammation of her optic nerve (optic neuritis) several years prior. An MRI on admission showed a large lesion in the left frontal lobe. A biopsy was performed showing that this was a demyelinating lesion , and it was felt to be either due to multiple sclerosis (MS) or systemic lupus erythematosus (Fig. 18.1).

She was treated with intravenous steroids , which reduced the inflammation, though she never fully recovered. Two years later, she presented with the acute onset of bilateral leg weakness such that she could not walk. An MRI at that time revealed inflammation throughout most of her cervical spinal cord. NMO antibodies, which had not been sent during her first admission, were positive. This confirmed the diagnosis of neuromyleitis optica (NMO) , an inflammatory disease that presents with pathology of the optic nerves/chiasm, spinal cord, brainstem, or, less commonly, the brain.

FormalPara What Dr. Yesky Was Thinking

In 2009, when I met Sadie, she did not have NMO, at least by the criteria at the time.

In 2006, the following formal diagnostic criteria were as follows:

  • Optic neuritis.

  • Acute myelitis.

  • At least two of three supportive criteria:

  • Continuous spinal cord MRI lesion extending greater than three spinal cord segments in length.

  • Brain MRI not meeting the criteria for multiple sclerosis.

  • NMO-IgG antibody detected in the blood.

As new discoveries about NMO accumulated, these criteria were revised in 2015 to include brainstem syndromes , lesions of the thalamus that may present with narcolepsy , and symptomatic brain lesions . As such, many patients who might have been explicitly excluded from the 2006 diagnostic criteria for NMO, especially if their brain MRI met the diagnostic criteria for MS, are now known to have NMO.

At the time, it did not occur to me to send the blood test that would have made the diagnosis. By obeying the 2006 criteria, as if they had been delivered from stone tablets on Mount Sinai, I missed a chance to initiate immunosupressive treatment in Sadie. When the revised criteria were released the next year, it was obvious she had NMO all along. I suppose the mistake I could make now would be to assume that these 2015 criteria are set in stone and no further developments will occur.

FormalPara Discussion

Nosology is a branch of medicine that deals with classification of diseases, disorders, syndromes, and injuries. The most widely used classification system is the International Statistical Classification of Diseases and Related Health Problems created by the World Health Organization. It contains 22 chapters and nearly 15,000 codes for diseases, abnormal signs, symptoms, social circumstances, and external causes of injury or diseases (Fig. 18.2).

Categorizing diseases this way is extremely important as it allows clinicians to communicate meaningfully to each other across time and space about the patients they treat. Additionally, little medical research would be possible without formal definitions of diseases. Certainly, many patients will have a disease that fits squarely within a well-established diagnostic category. However, the formal boundaries of many diseases can be rather arbitrary. It is often difficult to separate one disease from another, and sometimes there is great controversy as to whether a person even has a disease or not. Occasionally, a completely new disease emerges. Medicine is not a static field, and clinicians who allow themselves to be hemmed in by formal diagnostic criteria may miss a new disease or a new manifestation of an old one. Additionally, diagnostic criteria may be overly broad, encompassing multiple different diseases. The nosology trap occurs when clinicians uncritically rely on formal diagnostic criteria to separate the sick from the healthy, to make a diagnosis, and to determine treatments.

Consider MS, the main disease I treat. At present, MS is classified primarily by the time course over which patients develop symptoms. The majority of patients present with relapsing-remitting disease . These patients develop the rather abrupt onset of neurological symptoms, which last for several days to weeks, and then gradually abate. In order to make a diagnosis of MS, a patient must have evidence, either based on their history or MRI, of central nervous system inflammation in at least two different places at different times. Patients with a single episode of inflammation cannot be said to have MS, but are instead said that have clinically isolated syndrome. Other patients have progressive disease, where symptoms slowly progress over the course of months to years without any abrupt changes (Fig. 18.3).

Fig. 18.1
figure 1

FLAIR and post-contrast T1WI demonstrate a large white matter lesion in the left frontal lobe with peripheral enhancement (red arrow)

Full size image

Yet, this disease classification system is clearly incomplete. It says nothing about which symptoms a given patient experiences. Some of my patients are in wheelchairs, but have no cognitive or visual impairment. Other patients can walk fairly well, but have visual impairment or cognitive limitations. It also says nothing about the severity of impairment caused by the disease. I know several patients in their twenties who need full-time care due to their disability. At the other extreme, there are many patients in their sixties and seventies who have mild impairment that does not prevent them from leading a full and productive life. It’s not entirely clear what is the best way to classify MS.

Even the diagnostic criteria for this seemingly well-established disease were modified in 2010, and further tweaked in 2017. Based on changes made that year, many patients who had clinically isolated syndrome prior to 2010 could now be diagnosed with MS. The existing categories of MS were further changed in 2014, and several new subcategories were added. I am sure that the current classification scheme will not be the final one. Similarly, the International League Against Epilepsy revamped its basic seizure classification scheme in 2017. Terms that had been used for 35 years were eliminated and replaced with a new vocabulary [1]. These examples show that even the definition of diseases that have been described since antiquity are still subject to change today.

The implications of these changes are not just academic. All of the treatments for MS are approved based on the classification of the disease into relapsing and progressive forms. Additionally, because of changes in the definition of diseases, it is often impossible to compare studies conducted in different times. A study of patients with MS today can enroll a different patient population from a trial conducted only several years prior, making it problematic to compare results from one study to the next.

Many times, it is not clear where one disease stops and another begins. In the classification of medical conditions, there is a tension between the “lumpers” and “splitters ,” which are opposing camps that emerge when diseases have to be placed within rigorously specified categories. Lumpers combine things into broad categories, while splitters divide things into smaller categories.

There are many times when the “splitters” have won, and diseases that were once considered to be a single condition were recognized as different diseases. For example, NMO was once thought to be a variant of MS. Prior to discover of the antibody responsible for NMO, it was called opticospinal MS due to its propensity to attack the optic nerves and spinal cord. We now know that NMO is a completely separate disease from MS, and while the symptoms often overlap, some of the treatments for MS may actually worsen NMO. A clinician who treated NMO patients with medications for MS may have inadvertently done significant harm. Even today, it’s still not clear whether MS is a single disease with a wide variety of manifestations or several different diseases that we currently lump together because we don’t know any better. Time will tell.

At other times, a single disease has been split into multiple different types and subtypes based on advances in medical knowledge. For example, Charcot–Marie–Tooth disease is a relatively common peripheral nervous system disorder first described in 1886. It affects approximately one in 2,500 people in the US. It is characterized by the slow progression of weakness, gait disturbances, and numbness. It has evolved from its original designation as one disease and is now known to be caused by several different genetic mutations that lead to a broadly similar clinical presentation. Though they are still all called Charcot–Marie–Tooth disease, they have different symptoms, inheritance patterns, and prognoses. So, is Charcot–Marie–Tooth really one disease? Or it is multiple different genetic diseases that are artificially lumped together as one condition for the sake of convenience?

At other times, the lumpers have won, and multiple different diseases have been consolidated into a single entity. For example, several neurodegenerative diseases that existed when I was in medical school, such as Shy-Drager syndrome , striatonigral degeneration , and olivopontocerebellar atrophy , have now been merged into an entity called multiple system atrophy . Medical students today won’t learn about these diseases, as they have effectively “disappeared” into a single category. Currently, multiple system atrophy is itself divided into two types depending if the patient has cerebellar dysfunction or symptoms that resemble Parkinson’s disease . Perhaps the splitters will again tackle multiple system atrophy, and future generations of medical students will learn about diseases that have yet to be named.

The classification of many cancers is also being radically altered by examining their underlying molecular aberrations. The most common type of primary brain tumor, known as a glioma , was divided into four categories for decades, based primarily on its appearance under a microscope. Today, gliomas are defined largely by the presence or absence of certain genetic markers, which have enormous prognostic value. In fact, it is often not clear which conditions should even earn the terrifying designation “cancer [2].” Several conditions, previously classified as cancer and often treated as such, have recently been changed, so that the word “cancer” was eliminated from their description. In 2013, experts suggested renaming a breast condition called ductal carcinoma in situ, eliminating the word “carcinoma” so that potentially harmful treatments were avoided and patients were not needlessly frightened. Similarly, in 2016, an international team of doctors decided that a thyroid tumor (encapsulated follicular variant of papillary thyroid carcinoma), had been classified and treated as cancer, is not actually a cancer as it has a benign prognosis even without treatment [3].

Not only are old diseases reclassified, but new diseases are occasionally still identified. Most obviously, this includes new infectious diseases. For example, severe acute respiratory syndrome (SARS) is a viral infection that emerged in China in 2002 and spread to nearly 40 countries, causing 8,098 cases of the disease and 774 deaths. Additionally, new antibody-mediated and genetic diseases are being discovered as well. These diseases have likely existed for decades or centuries, but are just being recognized now as a distinct disease entity. Anti-NMDA receptor encephalitis , a disease that is gaining increasing recognition today, was only formally characterized in 2007 [4]. It presents with psychosis, cognitive disturbances, and seizures, often in young women with benign ovarian tumors. Undoubtedly, many patients were treated for years as having a primary psychotic disorder, such as schizophrenia , without the neurological nature of their disease being recognized.

Along these lines, many diseases that were once thought to be “psychiatric” are now realized to have a physiological basis. As late as the 1970s, many dystonias, which are characterized by abnormal movements and postures due to sustained, involuntary muscle contractions, were thought to be psychiatric in nature, when in fact they have a genetic basis [5]. Certain dystonias caused by muscle overuse in writers, musicians, and athletes were referred to as “occupational neurosis [6].” Today, almost 30 different genetic mutations have been identified that lead to a wide spectrum of dystonias. As the classification for this disease changed from psychiatric to neurologic, the treatment changed from psychotherapy to medications.

Large sums of money can be at stake if something is labeled a disease or not, and where the boundaries of a certain disease are set. Widening the diagnostic boundaries of diseases is known as disease-mongering .

Science writer Lynn Payer defined disease-mongering as having the following characteristics [7]:

  • Stating that normal human experiences are abnormal and in need of treatment.Advertising a common symptom as a serious disease.

  • Claiming to recognize suffering, which is not present.

  • Defining a disease such that a large number of people have it.

  • Defining a disease’s cause as an ambiguous hormonal imbalance or deficiency.

  • Associating a disease with a public relations spin campaign.

  • Directing the framing of public discussion of a disease.

  • Intentionally misusing statistics to exaggerate treatment benefits and setting a dubious clinical endpoint in research.

  • Advertising a treatment as without side effect.

A 2002 editorial in the British Medical Journal (BMJ) titled Selling Sickness: The Pharmaceutical Industry and Disease Mongering warned of disease-mongering, saying:

Inappropriate medicalisation carries the dangers of the unnecessary labelling, poor treatment decisions, iatrogenic illness, and economic waste, as well as the opportunity costs that result when resources are diverted away from treating or preventing more serious disease. At a deeper level it may help to feed unhealthy obsessions with health, obscure or mystify sociological or political explanations for health problems, and focus undue attention on pharmacological, individualised, or privatised solution [8].

Disease-mongering is a common tactic of both pharmaceutical companies and practitioners of complementary and alternative medicine (CAM) who profit when previously healthy people are considered ill. Conditions such as adrenal fatigue, chronic Lyme disease, electromagnetic hypersensitivity, leaky gut syndrome, multiple chemical sensitivity, and non-celiac gluten sensitivity are unrecognized conditions that CAM practitioners promote right before they try to sell you the cure. Once a condition is labeled a disease, the genie is out of the bottle, and it can be hard to put back in.

Similarly, the cut-off for bona fide diseases can be lowered, creating conditions such as “pre-dementia ” and “pre-diabetes .” With MS, the criteria have been relaxed to make it possible to diagnosis the disease in people who have had a single neurological event combined with MRI data. It is not clear that this is an improvement for patients, but it forces insurance companies to pay for medications in people who might have previously been ineligible for them. In the 1990s, the definition of osteoporosis was redefined such that nearly 25% of post-menopausal women were considered to have the “disease.” Soon after this, Merck launched the osteoporosis drug Fosamax, which became a blockbuster, selling billions of dollars. In 2008, the US National Osteoporosis Foundation again altered their guidelines such that 72% of white women older than 65% and 93% of those older than 75 were suggested to take drugs for osteoporosis [9]. Similarly, in 2017 guidelines from the American Heart Association and the American College of Cardiology redefined hypertension as a blood pressure greater than 130/80 mmHg, down from the previous value of 140/90 mmHg [10]. Almost half of adults in the US will meet criteria for hypertension based on this revision, up from about one-third. While the guidelines stressed that vast majority of the new cases could be managed by lifestyle interventions, invariably this will lead to the use of more medications (though by some estimates the numbers will be rather low) [11].

When expert panels meet to define diseases, they almost never advocate for making the criteria for the disease more stringent. Raymond Moynihan and colleagues reviewed 16 publications representing 14 diseases where expert panels suggested changing diagnostic criteria. The disease definition widened for 10 diseases, narrowed for only one, and had an unclear impact in five diseases [12]. Additionally, expert panels rarely consider the potential harms of over-diagnosis. Even more concerning is that many panels have been found to have members with undeclared conflicts of interest with pharmaceutical companies [12].

In response to disease-mongering , an international group of multidisciplinary experts created a checklist of eight questions to consider when modifying the definition of a disease [13].

  1. 1.

    Definition: What are the differences between the previous and new definition?

  2. 2.

    Number of people affected: How will the new disease definition change the incidence and prevalence of the disease?

  3. 3.

    Trigger: What is the trigger for considering the modification of the disease definition?

  4. 4.

    Prognostic ability: How well does the new definition of disease predict clinically important outcomes compared with the previous definition?

  5. 5.

    Disease definition precision and accuracy: What is the repeatability, reproducibility, and accuracy of the new disease definition?

  6. 6.

    Benefit: What is the incremental benefit for patients classified by the new disease definition versus the previous definition?

  7. 7.

    Harm: What is the incremental harm for patients classified by the new definition versus the previous definition?

  8. 8.

    New benefit and harms: What is the net benefit and harm for patients classified by the new definition versus the previous definition? (Fig. 18.4)

Fig. 18. 2
figure 2

Chapters in International Statistical Classification of Diseases and Related Health Problems. (World Health Organization. Classifications. ICD-11. 2018. Retrieved from http://www.who.int/classifications/icd/en/.)

Full size image
Fig. 18. 3
figure 3

The main subtypes of multiple sclerosis. (Diagnosis of multiple sclerosis. [Image file]. In: Wikipedia. (n.d.). Retrieved from https://en.wikipedia.org/wiki/Diagnosis_of_multiple_sclerosis#/media/File:Ms_progression_types.svg.)

Full size image
Fig. 18. 4
figure 4

Checklist of items to consider when modifying a disease definition

Full size image

It is often difficult to determine if a patient has a disease at all. Family medicine doctor Ian McWhinney wrote, defining the word “disease” can be as difficult as defining “beauty, truth, and love [14].” A survey of 3,280 doctors, nurses, and laypeople in Finland by Kari Tikkinen and colleagues showed it is difficult to determine what does and does not constitute disease and that professional clinicians can have a different opinion from the general public [15]. There was broad agreement amongst responders that conditions such as malaria and lung cancer are diseases. There was also broad agreement that conditions such as aging, wrinkles, and homosexuality are not diseases. However, there was disagreement about whether conditions such as gambling addiction, infertility, and drug addiction are diseases.

In an attempt to demonstrate the “slipperiness of the notion of disease,” The BMJ conducted a survey in 2002 asking their readers to identify “non-diseases,” which they defined as “a human process or problem that some have defined as a medical condition but where people may have better outcomes if the problem or process was not defined in that way [16].” The suggestions, compiled in an international classification of non-disease, included several conditions that most clinicians would today consider diseases, such as alcohol dependency, anorexia, attention deficit disorder, and tics. It also included many conditions that few would consider diseases today, including loneliness, boredom, flat feet, colic, aging, and false memory syndrome. While these conditions can cause undoubtedly immense suffering, it is not clear that people suffering from them are necessarily ill. Perhaps future generations will disagree.

Clinicians struggle to know what to do with people who are suffering but not necessarily ill. People with conditions that are generally not regarded as diseases often chafe at the idea that they are healthy. Strong communities form, often on social media and under the influence of passionate, charismatic leaders. They often rail against a seemingly callous and dogmatically rigid medical system that fails to legitimatize their suffering. People with Morgellons disease, for example, believe their skin has been infested with yet to be identified parasites. According to The Morgellons Research Foundation, “The symptoms include itching, biting and crawling sensations, ‘filaments’ or fibers which emerge from the skin, skin lesions which range from minor to disfiguring, joint pain, debilitating fatigue, changes in cognition, memory loss, mood disturbance and serious neurological manifestations [17].” Under close analysis, the fibers are often found to be cotton and other clothing items. A $600,000 study conducted by the CDC failed to elucidate the etiology, and most clinicians view the disorder as a delusional parasitosis. Not surprisingly, Morgellons sufferers did not accept this conclusion. Jan Smith, a leading patient advocate, said in an interview, “I’m pretty sure they’ll say we’re all delusional. There’s so much more to this than a medical condition. There’s something being hidden [18].”

Clifton K. Meador, a professor of medicine, argued in 1965 that patients with such conditions should be labeled as having a specific non-disease [19]. He wrote:

Patients are frequently seen on referral with a specific disease diagnosis, and yet investigation fails to substantiate the referral diagnosis; in fact it may not reveal any disease. What, then, does the patient have? He must have something. The argument will be presented that he or she has a particular nondisease. This is certainly more reasonable than the common error of continuing to label such patients with nonexistent diseases.

However, a small study by KB Thomas in 1987 showed that this might be a mistake, at least in communicating with patients. He studied 200 patients for whom no definitive diagnosis could be made. Some of the patients had a positive consult, where they received a diagnosis and told they would improve. Others had a negative consult, where no diagnosis was made and no prognosis for improvement was rendered. Two weeks later, 64% of the patients who received a “positive” consultation improved, while only 39% who received a “negative” consultation improved [20].

Even if no disease diagnosis can be made, clinicians should remember that patients are still suffering. As Mark L. Eberhard, director of the division of parasitic diseases and malaria at the CDC, said about patients with Morgellons, “These people are definitely suffering from something. It has impacted their lives greatly [18].”

Political and cultural norms can further complicate determinations of what is and is not considered a disease. Culture-bound syndrome is a well-recognized set of psychiatric and somatic symptoms that are recognized as a disease within a specific society or culture. Dozens such syndromes exist. Koro, for example, is seen in parts of Africa and Asia. It is a delusional disorder in which people believe their genitals are retracting and will disappear. Running amok occurs in Indonesia and Malaysia, and is defined as “an episode of sudden mass assault against people or objects usually by a single individual following a period of brooding.”

Western cultures have their own culture-bound syndromes. A condition known as neurasthenia was described by a New York neurologist George Beard in 1869. It was characterized by headaches, palpitations, hypertension, neuralgia, fatigue, anxiety, and depression. He believed it was caused by busy society women, overworked businessmen, and the general challenges of living in an increasingly industrial world. It was also called Americanitis, as it was common in the US. It was thought to be a condition that could only afflict people with active, intelligent minds and racist explanations were given as to why only white people could get the “disease.” National parks were established in part to let people with neurasthenia rest [21]. While neurasthenia essentially died as a diagnosis 100 years ago, its echoes are seen with conditions such as chronic fatigue syndrome and chronic Lyme disease.

More recently, homosexuality was classified as a disease by the American Psychiatric Association (APA) in its official Diagnostic and Statistical Manual of Mental Disorders until 1973. Even today, some fundamentalist Christian groups continue to advocate for conversion therapy as a means to “cure” homosexuality. While the notion of homosexuality as a disease likely seems abhorrent to most people, it is possible that there currently stigmatized behaviors that future generations might view as a sign of mental illness. For example, psychiatrist Alvin Poussaint argued that extreme racism should be considered a delusional disorder and that it is amenable to treatment. He wrote, “It is time for the American Psychiatric Association to designate extreme racism as a mental health problem by recognizing it as a delusional psychotic symptom [22].”

Other critics argue that psychiatric illnesses in particular have a significant cultural component and that medicalizing them using Western values carries potentially grave consequences for non-Western cultures. Ethan Waters wrote in The New York Times that:

For more than a generation now, we in the West have aggressively spread our modern knowledge of mental illness around the world. We have done this in the name of science, believing that our approaches reveal the biological basis of psychic suffering and dispel prescientific myths and harmful stigma. There is now good evidence to suggest that in the process of teaching the rest of the world to think like us, we’ve been exporting our Western “symptom repertoire” as well. That is, we’ve been changing not only the treatments but also the expression of mental illness in other cultures. Indeed, a handful of mental-health disorders — depression, post-traumatic stress disorder and anorexia among them — now appear to be spreading across cultures with the speed of contagious diseases. These symptom clusters are becoming the lingua franca of human suffering, replacing indigenous forms of mental illness.

Finally, defining someone as “diseased” has important implications for how patients view themselves, how they are viewed by their loved ones, and by society at large. Receiving a medical diagnosis may have significant practical implications for a person’s life, determining whether someone is exempt from work or entitled to benefits. For the most part, sick people are thought of as needing care, sympathy, and treatment.

However, this is not always the case, and at times defining someone as “diseased” can be used as a weapon to stigmatize and discriminate against them. As the example of homosexuality shows, psychiatric diagnoses in particular have all too often been used as weapons to denigrate certain classes of people and stigmatize individuals. The most famous example of this occurred in a 1964 magazine article titled, 1189 Psychiatrists Say Goldwater is Psychologically Unfit to Be President! In it, psychiatrists said that the Republican candidate for president, Barry Goldwater, was mentally unfit for the job. He was called “unpredictable,” “emotionally unstable,” and “a dangerous lunatic! [23]” Barry Goldwater successfully sued for libel, and the APA established an ethical principle, known informally as the “Goldwater rule,” that prohibited psychiatrists from giving their medical opinion about someone unless they have personally evaluated them and consent has been obtained. (There have been calls to end this rule with the election of Donald Trump.)

Even more dramatically, psychiatric diagnoses have been used at times to subject people to unethical medical procedures such as forced sterilization. Most famously, Carrie Buck was forcibly sterilized in 1927 at the age of 19 for being “feeble-minded” and “promiscuous.” Her case made it to the Supreme Court, where famed jurist Oliver Wendell Holmes defended her sterilization by famously arguing that “Three generations of imbeciles are enough.” The ruling legitimized Virginia’s sterilization procedures, which were only repealed in 1974. Her sister Doris was also sterilized without her knowledge when she was hospitalized for appendicitis. Carrie Buck’s one child, conceived at the age of 17 as the result of a rape, was considered to be above-average intelligence, and there is no evidence Ms. Buck herself was particularly impaired. She was described as an avid reader until her death in 1983.

FormalPara Conclusion

The principle of linguistic relativity (the Sapir–Whorf hypothesis) posits that the structure of a language affects the cognition and thoughts of those who speak it. While this is a controversial notion, it is clear that, to some degree in medicine, the classification schemes used to categorize diseases affect how clinicians think about them. Given that medical diagnoses are often in flux, clinicians must be careful not to be hemmed in by formal diagnostic criteria for diseases. Obviously, this is not to say that formal diagnostic criteria are meaningless or without use. Clinicians need a framework to conceptualize diseases and to communicate with each other. These criteria are also necessary for medical research. In a clinical study of any disease, it is important that only persons who meet the criteria for that disease be enrolled in the study.

However, by rigidly assuming that medical diagnostic categories are fixed entities without room for change, clinicians may fail to recognize heretofore undescribed variations of a disease or even an entirely new disease. Alternatively, they may respond to the name of the disease- certainly encapsulated follicular variant of papillary thyroid carcinoma and ductal carcinoma in situ sound scary- rather than to the actual threat the “disease” poses. Additionally, some diseases are culturally bound. A condition that is considered a disease today might be considered a variant of normal in the future, and vice versa. Finally, clinicians who sit on expert panels should be very cautious before expanding the boundary of disease, carefully considering the risks and benefits of the proposed change.