Is Long Covid a functional disorder?

In 1978, I submitted my first paper to the British Medical Journal. It concerned what we now refer to as functional disorders, the phenomenon of illness with no identified disease causing it. I have been interested since. A recent editorial by an Oxford colleague, Anton Pick, discussing whether patients with Long Covid had a functional disorder, has stimulated me to write this blog post on functional disorders, the biopsychosocial model of illness, our propensity to categorise and draw binary distinctions, and stigmatisation of any condition thought to have a psychological cause. These are all weighty issues, and the post can only introduce topics. You will have to consider them further; the considerations include ethical and legal matters such as the nature of illness and society’s responsibility for supporting patients, philosophical concerns about choice, free will, and the nature of experiences as part of an illness, and linguistic issues around the meaning of words.

Table of Contents


“But that patient does not have real pain.”, “Are you saying that my pain is imaginary?”, “You just think my problems are all in my mind!”  Statements or questions like these are frequent and reveal a continuing culture of devaluing any illness or experience not demonstrably due to tissue damage or disease. Pain is the starkest example for illustrating the problem, but the discussion applies to all symptoms (patient-reported experiences) arising within an illness.

When I hear such statements, I immediately point out that pain is an experience generated within the brain, like the colour blue, the feeling of warmth, fatigue, and pins and needles (paraesthesia). Sometimes I continue that a person’s experience is, arguably, the only reality they can be sure about. This is the position of Empiricism put forward by many philosophers, including John Locke, Immanuel Kant, Rene Descartes, and David Hume. I will often add that pain is our experience when our well-being is threatened, emphasising that not all threats are physical and that depression and other psychological threats are all accompanied by pain.

However, these philosophical arguments rarely satisfy patients, their relatives, and healthcare professionals. Philosophy does not explain how grief can cause pain, why some people are more likely than others to have distressing experiences after trauma, or what aspect of Covid-19 infection leads to prolonged generalised symptoms. People want explanations consistent with their own beliefs and models of illness.

Philosophical and clinical discussions are also bedevilled by the words used, and I will start by distinguishing maladies, illness, sickness, disease (pathology), and disability.

What is in a word?

Words are labels that summarise a complex concept into a single item. We sit on many objects of different shapes, sizes, costs etc., but we label them all chairs. Some chairs have another label applied sometimes, such as a weapon used to hit someone, firewood when burned, or a table used as a table. When an object can be seen and felt, it is usually easy to match a word to the thing, though the context can influence the word used, as the use of an object may determine the word used.

Intangible concepts such as justice, the colour blue, headache, and illness are also labelled using words. Unfortunately, knowing precisely what is meant is more challenging because there is no object or visible use to anchor the meaning. Misunderstandings and different understandings of words bedevil discussions about any part of healthcare.

I have discussed the words associated with ill health recently, and I will restate the meanings of the words I use in the post; they are similar to the intentions suggested by Bjørn Hofmann. The Venn diagram below also shows the distinctions between the central concepts.

Malady, Illness, Sickness, Condition, Disease


The malady is an all-encompassing, non-specific word with no associated implications, which Bjørn Hofmann summarises thus, “Malady will be used as a generic term covering a wide range of terms, such as disease, illness, sickness, but also injury, lesion, defect, disorder, deformity, disability, impairment, infirmity, etc.” 


Sickness refers to a socially-determined state, a public role with associated expectations.  To quote Bjørn Hofmann, “Sickness determines whether a person is entitled to treatment and economic rights, exemption from social duties, such as work (sick leave), but also whether a person is legally accountable for his or her actions.”


Illness refers to the person’s experiences and self-determined state leading to the person taking on a role and associated expectations. Still, the role and expectations may not be validated by society. To quote Bjørn Hofmann, “Illness, on the other hand, has emotions and experience, such as anxiety, fear, pain, and suffering, as its basic phenomena.”

Condition (or disorder).

Many people are sick (i.e. there is a social agreement that they are ill) but do not have an identifiable disease.  For example, people may have alcoholism or depression, headache or back pain, or undiagnosed muscle wasting (sarcopenia). In many cases, risk factors and associated illnesses may be known, and mechanisms that account for symptoms may be known, but no single, unique cause has been identified. These sicknesses are most likely multifactorial. Some of these conditions may be labelled functional.

I will use condition (synonym, disorder) as a non-specific, syndromic label which can be used to identify a group of people with sickness who have a similar syndrome (set of symptoms or signs) and prognosis. Some conditions will be associated with a disease, but in others, there may be no known single cause, or the aetiology may be multifactorial.

In other words, all people who are sick will have a condition, but only some will have a disease.


I will use disease to refer to an externally verifiable specific cause for the sickness. It is “the cause of the sickness, where the cause is definitely and uniquely linked to the sickness and is characterised by a distinct structural and/or physiological abnormality within the body.”

Long Covid, a single cause?

One human characteristic is to seek the cause for an observation, exemplified by the phrase Root Cause Analysis. What caused that accident? Who caused the failure to protect that child? What is the reason for the current healthcare NHS crisis? People expect simple answers. For example, faulty brakes caused the accident; the social worker should have noticed the child’s injuries; the NHS crisis arises from poor management and organisation. These answers imply a simple, single solution, which means we can avoid undertaking a fuller investigation and committing to a more complex and challenging solution.

These answers and solutions ignore the multifactorial nature influencing events, even those with straightforward causation. If you google for the cause of malaria, you will find that “Malaria is caused by a single-celled parasite of the genus plasmodium.”  But immediately, a second factor is mentioned, “The parasite is transmitted to humans most commonly through mosquito bites.” It is also well-known that poverty, malnutrition, lack of protective netting etc., are all associated with an increased risk of death of a child from malaria. They are additional causes in that reversal of these factors reduces the risk of death from malaria.

This is obvious for all conditions where no disease is known; most of the supporting charities state they are supporting research into the underlying cause, with substantial implications that:

  1. Their situation has a unique, specific single cause without which the condition would not occur
  2. Finding the cause will lead to an effective treatment.

The hypothesis that every condition has a single cause is untrue. Consider multiple sclerosis, where the evidence for a genetic contribution is strong. Only one in eight identical twins will develop multiple sclerosis. Consider carcinoma of the lung, where 85% of smokers do not get carcinoma of the lung. (I am not advocating that smoking is safe!) Many people who have had a stroke ask, quite reasonably, why they had a stroke and are not satisfied with the explanation that it was due to hypertension, pointing out that (a) they have had raised blood pressure for years with no stroke and (b) their siblings with high blood pressure never had a stroke.

Our success over the last 200 years in identifying causes for sickness and going on to find cures for the sickness has led us to overlook the many other factors associated with and contributing towards most conditions. For example, socio-economic factors are strongly related to most conditions, as highlighted in the Marmot review in 2010.

Therefore, the suggestion by Anton Pick that “the symptoms patients experience are, in fact, caused by more than one mechanism and that ‘Long Covid’ is too big an umbrella for the range of presentations seen.” is sound.

Diagnosis should be non-binary.

Many take one further step from the assumption that one cause exists for a condition and assume that if a person has a condition, that condition accounts for all the problems experienced. More specifically, people who have acquired the label of functional illness are considered not to have any other condition.

This flies in the face of everyday experience, where having two or more verified diseases is now expected. Having a functional disorder does not protect you from another condition. Most people think that you can have, for example, multiple sclerosis or a functional disorder, but you cannot have both.

This leaves a problem. Many people with symptoms and disabilities undoubtedly have a disease that could account for a proportion of their symptoms, but they have many problems that cannot be attributed to the known disease. For example, a patient with multiple sclerosis suddenly lost her eyesight after attending a talk on how to help people who were blind. This person had had many other episodes of extreme losses, which all improved dramatically after a few days or weeks. While any single episode she had was possible with the condition, their onset, recovery, and rapid and complete resolution, and the frequency and initial severity of the episodes were not. These would have been labelled functional in any other person without neurological disease. She did have multiple sclerosis and a functional disorder.

A binary approach to a malady necessitates a single label. A label has advantages, such as summarising the problems, their prognosis, and sometimes their treatment. It often satisfies the patient by giving a simple explanation. It is usually required by bureaucracy, for example, to access sickness benefits.

Labels have many disadvantages. They give an impression of certainty and thus stop people from thinking and being curious, they are often not reviewed critically, and many labels are associated with a judgemental attitude and lead to stigma.

Mood and other psychological and social factors often influence the manifestations of disease and the experiences of a patient. We must become comfortable with saying that problems A, B, and C can be attributed primarily to factors X, problems D, E, and F are 50% related to X and 50% to Y, and problems G, H, and I have some other cause. In diagnosing a patient’s problems, we should move away from binary labels to a formulation.

Long Covid and stigma.

In a meta-ethnographic synthesis of the experiences of stigma amongst people with a functional neurological disorder, Ciarán Foley and colleagues found that people diagnosed with a functional neurological disorder experienced stigma as delegitimisation of their experience (“what you experience is not real”) and through social exclusion. Maria Annadale and colleagues, in a scoping review, found stigma was prevalent in people with functional (non-epileptic) seizures.

It is probable that people with Long Covid, especially people who were not sufficiently ill to be admitted to a hospital, will experience stigmatisation. This stigmatisation is part of a societal attitude towards all mental health disorders, attributing them less attention, resource, and sympathy. Indeed, many people with functional disorders probably resist treatments aimed at psychological factors or influencing psychological processes precisely because of the cultural devaluation of psychological illness.

Long Covid, uncertainty and the biopsychosocial model

Anton Pick says, “we need courage to sit with the uncertainty that surrounds Long Covid.” Yes, we must live with the uncertainty we face in all rehabilitation. Accepting and managing complexity and certainty is one of the seven rehabilitation capabilities I have suggested. Long Covid is undoubtedly complex, and we only have two years to research it. We cannot know the prognosis after two years, and there are few randomised controlled trials of rehabilitation interventions.

Throughout his editorial Anton Pick stresses the multifactorial nature of all conditions and that these relevant influences extend well beyond the limited purview of the biomedical model. For example, he says, “the relationship between proposed biomedical mechanisms and disease outcomes is not straightforward.”

The uncertainty arises not simply from the lack of research evidence and will remain for years to come. Uncertainty is an integral part of all healthcare practice, especially in rehabilitation and other specialities whose approach is based on the biopsychosocial model of illness.

The way forward.

People with Long Covid should receive the same approach to rehabilitation as any other person, avoiding preconceptions or assumptions about ultimate causes or mechanisms that are not supported by firm evidence. As Anton Pick stresses, we must maintain our curiosity and question all statements presented as facts. His article implies, in places, that there is a single condition that will be diagnosed by validated and dedicated tests but, more generally, he agrees that the label probably encompasses several or many discrete conditions.

The vital first steps are for any patient with persisting disabling problems to be seen by a multi-professional team that includes all professions relating to their problems for an assessment and expert formulation, identifying causal links where this can be done but accepting that often these will not be known. The team must avoid the influence of labels.

The team then must decide on the best initial interventions. This requires expert and wise judgement and attention to the patient’s beliefs, expectations, and goals. A team will likely make a more thoughtful decision that an individual, however expert. Team members and groups must avoid giving undue weight to their own beliefs. Intervention should be considered using available evidence acquired from patients with similar problems. Without any contrary evidence, the team should consider the range of known effective interventions in rehabilitation.


Patients with symptoms and problems attributed to Long Covid should be considered as all other people presenting to healthcare, especially rehabilitation. They should be assessed by a team who should avoid any assumptions about the causes and nature of their condition or conditions. As with any other diagnostic label applied to any other person, the label on presentation, Long Covid, should be accepted as a starting point but evaluated critically. At all points, one must realise it is a label with no independent verification; this does not mean it is invalid. However, it should always be considered the best available label but subject to revision. As professionals and as a society, we must be more focused on solving patients’ health problems and less concentrated on speculation about causes and binary labels. Formal research programmes are essential to investigate causes and causal mechanisms that will eventually improve rehabilitation.

Scroll to Top

Subscribe to Blog

Enter your email address to receive an email each time a new blog post is published. 
Then press the black ‘Subscribe’ button.