A Case Definition of a New Disease: A Review of the US Working Definition (USG) and 2024 NASEM Definition for Long COVID

Abstract

SARS-CoV-2 infection can be followed by prolonged symptoms, signs and sequelae, collectively known under the term Long COVID. Hundreds of millions are estimated to suffer from Long COVID. Long COVID, therefore, is a public health crisis that deserves the utmost urgency from all relevant stakeholders, from policymakers to advocacy groups, researchers and healthcare providers. The development of effective definitions and guidelines for Long COVID is crucial to support patients and carers. In this review, I address the following two case definitions of Long COVID developed in the US as a case study for a broader discussion on the sequelae of SARS-CoV-2 infection: the U.S. Government (USG) working definition for Long COVID and the NASEM definition published in 2024. In the first part of this review, I provide a critical appraisal of the USG in light of research, pathophysiology and lived experience, building upon my intervention as a patient expert on a National Academies of Sciences, Engineering, and Medicine (NASEM) panel for defining Long COVID, which examined the USG. In the second part, I raise some pressing concerns to address when approaching Long COVID as a disease entity and as a concept, which I originally submitted to NASEM. In the third part, I offer a critical appraisal of the NASEM definition, the most recent benchmark for Long COVID in the US. The review highlights the importance of broad, expansive and inclusive definitions for Long COVID, accounting for the disease’s heterogeneous, fluctuating and multi-system manifestations. Clinical case definitions for Long COVID must retain their focus on the broader spectrum and scope of the disease entity, while incorporating feedback from people with lived experience, advocates and patient-researchers.

1. Introduction

Coronavirus disease 2019 (COVID-19) is caused by the virus “severe acute respiratory syndrome coronavirus 2” (SARS-CoV-2). Since the first case was officially reported in late 2019, in Wuhan, China, the global population has been massively infected with SARS-CoV-2, with reinfections becoming increasingly common, especially after the emergence of the variant of concern (VOC) Omicron in late 2021 [1]. According to official data only, around 7 million people have died from COVID-19 worldwide as of June 2025 [2]. However, data based on metrics, such as excess mortality, have suggested the pandemic’s death toll to be at least 20–28 million by 2022–2023 [3,4]. These data alone make the COVID-19 pandemic one of deadliest in recorded history [5].

Many more people, however, have experienced a wide range of prolonged symptoms, signs and sequelae following SARS-CoV-2 infection, which are collectively known under the patient-coined term Long COVID [5,6,7]. Long COVID can affect all organs and body systems [8,9,10,11,12,13]. Multiple mechanisms have been proposed to explain the dozens of symptoms and multi-system pathology of Long COVID, which can result in severe disability and fatal outcomes [

Table 1. Proposed and putative mechanisms of Long COVID (bibliography).

Proposed and Putative Mechanisms of Long COVID
Viral persistence
Immune dysfunction Autoimmunity Vascular pathology including vasculitis, microvascular dysfunction and endothelial dysfunction Coagulation abnormalities Macro- and micro-thrombi Gut and oral dysbiosis Lung perfusion defects Tissue damage Fibrosis Apoptosis (cell death) Inflammation Reactivation of latent pathogens like herpesviruses DNA modifications   The list is not exhaustive. More than one mechanisms can be present in a patient

]. The number of people who suffer or have suffered from Long COVID has been estimated to range in the hundreds of millions [5,8,14,15]. This is in line with concerns over severe morbidity from the pandemic put forward already in 2020 [6]. The World Health Organization (WHO) itself expects, as of April 2023, that “hundreds of millions of people will necessitate longer-term care” following SARS-CoV-2 infection [16].

Long COVID, therefore, is a public health crisis that deserves the utmost urgency from all relevant stakeholders, from policymakers to advocacy groups, researchers and healthcare providers. The development of effective definitions and guidelines for Long COVID is crucial to support patients and carers. Patient perspectives are important for the development of efficacious case definitions of diseases, as patients provide different kinds of expertise (e.g., lived experience and patient-led research) that can complement clinical experience from physicians and healthcare professionals [6,7,17]. This issue has been brought to the fore by Long COVID itself, a disease entity identified, named and defined by those affected by the condition during the early pandemic [5,6,7,18,19,20].

In this review, I address two case definitions for Long COVID as a case study for a broader discussion of critical concerns over the sequelae of SARS-CoV-2 infection, as follows: (i) the U.S. Government (USG) working definition for Long COVID (hereafter, USG or USG definition; US) and (ii) the definition put forward by the US National Academies of Sciences, Engineering, and Medicine (NASEM) in 2024 (hereafter, NASEM definition) [21,22]. While the NASEM definition is proposed as the current benchmark in the US [22,23], the USG remains relevant to the time of writing as it can still be used in literature and research on Long COVID; it also reflects the process of knowledge building for defining a new disease entity such as Long COVID: the steps taken to define Long COVID, and any challenges encountered, deserve discussion and preservation in the biomedical literature both to address current approaches to post-infectious diseases and for future reference in the event of a novel pandemic with sequelae of infection. The NASEM definition itself could be updated in the future.

In the first part of the review, I will go through the USG in light of research, pathophysiology, and lived experience, building upon my intervention at NASEM as a patient expert (Section 3.1 and Section 3.2). In the second part, I will raise some pressing concerns to address when approaching Long COVID as a disease entity and as a concept, which I originally submitted to NASEM while assessing the USG (Section 3.3). Lastly, I will provide a critical review of the NASEM definition as published in 2024 [22]. Awareness around the issues raised in this paper can support the creation of effective definitions and guidelines for Long COVID. The merits of an expansive and inclusive definition, especially for clinical purposes, will be addressed. Other topics of heightened interest discussed in this review include Long COVID as a new, heterogeneous disease entity affecting people of all ages, genders and prior health statuses; access to care and diagnostics; Long COVID as a novel concept useful to address the prolonged disease process triggered and potentially driven by SARS-CoV-2 infection; Long COVID in children; Long COVID patient-driven advocacy; and concerns about potential longer-term sequelae of SARS-CoV-2 infection, such as cancer and neurodegenerative disease, which could become apparent only in years or decades to come.

2. Methods, Scope and Rationale

A critical review of the US working definition for Long COVID (USG) (Section 3.1 and Section 3.2), a critical discussion of pressing concerns to address when approaching Long COVID as a concept and disease entity (Section 3.3), and a critical review of the NASEM definition (Section 3.4 and Section 3.5) are provided. The USG and NASEM definitions are examined in view of my lived experience as a person living with Long COVID and a participant in Long COVID advocacy–research since 2020 [6,7]. This paper is a revised, expanded and written version, with a bibliography, of my oral presentation for the “patient perspective panel” of the NASEM Committee Meeting 3 on “Examining the Working Definition for Long COVID”, to which I participated virtually from Italy on 12 May 2023 [24]. The meeting was part of a “series of activities” a committee from NASEM was asked to “conduct. to examine the [then] current U.S. Government (USG) working definition for Long COVID and related technical terms” [24]. I submitted a shorter version of this review as a white paper via email to the committee’s attention on 26 May 2023. I submitted an expanded version of the white paper to the Open Science Framework in December 2023 for public fruition [25]. The critical review of the NASEM definition was conducted in 2025.

My aims in contributing to NASEM were the following: (i) to urge action on the Long COVID crisis; (ii) to highlight the severity, scope and scale of Long COVID, which must remain paramount in any future definition; (iii) to invite recognition of patient-driven advocacy–research in Long COVID and support the incorporation of patient perspectives into case definitions. The present review is updated to include knowledge available as of Summer 2025, drawing from literature on COVID-19 and Long COVID included in PubMed and LitCovid. Nonetheless, with the exception of the NASEM definition’s appraisal (Section 3.4 and Section 3.5), the review maintains, when feasible, the basic structure and arguments of the white paper delivered to NASEM in 2023: this is to preserve, in the peer-reviewed literature, the process of knowledge building that contributed to the NASEM definition published in 2024. The aim is to increase transparency into decision making, by documenting the scientific and lived experience offered to policymakers on Long COVID, As such, this work is also offered to the attention of health historians and scholars of the COVID-19 pandemic.

The review largely stems from tasks I was asked to perform by NASEM (e.g., providing my feedback on the USG) and is not intended as a meta-analysis or systemic review on Long COVID. The pressing issues raised in Section 3.3 build upon the white paper I submitted to the NASEM committee in 2023 and remain relevant to the time of writing, including in view of any future revision of the current NASEM definition. The bibliography is indicative and not exhaustive. More information on the activities carried out by NASEM on Long COVID can be found in relevant NASEM-related literature [21,22,24].

3. Results and Outcomes

3.1. US Working Definition of Long COVID (USG)

The USG reads as follows:

“Long COVID is broadly defined as signs, symptoms, and conditions that continue or develop after initial COVID-19 or SARS-CoV-2 infection. The signs, symptoms, and conditions are present four weeks or more after the initial phase of infection. Long COVID may be multisystemic; and may present with a relapsing–remitting pattern and progression or worsening over time, with the possibility of severe and life-threatening events even months or years after infection. Long COVID is not one condition. Long COVID represents many potentially overlapping entities, likely with different biological causes and different sets of risk factors and outcomes.”

Overall, I think the USG definition is a good and inclusive definition as a starting point. It is broad in scope and attentive to the heterogeneous manifestations of Long COVID. It is simple and concise. My comments in Section 3.2 build upon the definition as of May 2023, which was examined at NASEM. I do not suggest an alternative definition myself for this review, nor did I at NASEM. The NASEM definition is reviewed in Section 3.4 and Section 3.5.

3.2. Observations on the US Working Definition of Long COVID (USG)

Here, I provide a critical review of the USG, sentence by sentence.

“Long COVID is broadly defined as signs, symptoms, and conditions that continue or develop after initial COVID-19 or SARS-CoV-2 infection.” (USG definition first sentence)

This is appropriate, as the scientific literature shows a vast range of symptoms, signs and conditions associated with SARS-CoV-2 infection and Long COVID [6,9,10,11,12,13].

It is important to highlight that a positive PCR (polymerase chain reaction), RAT (rapid antigen test) or antibody (serology) test should not be necessary for diagnosis. Clinical diagnosis of COVID-19 and Long COVID based on symptoms, signs, epidemiology (e.g., exposure history, time and place) and other relevant tests, such as scans, should be enough. In this regard, tests of relevance for diagnosis include, but are not limited to, the presence of a positive CT and/or SPECT-CT scan, suggestive of COVID-19-related lung pathology: patients, especially from the early pandemic, might have these tests in the absence of, or in addition to, a positive PCR [26,27,28]. Many infected with SARS-CoV-2 do not have access to tests such as a PCR, RAT or antibody test. In particular, many infected in the first wave in 2020 were denied or failed to receive adequate testing [6,29,30]. Moreover, the risk of false negatives for PCR, RAT or serology is relevant under certain circumstances, including timing of administration, type of testing and individual immune response to SARS-CoV-2 infection [31,32,33,34].

As of June 2025, novel diagnostic challenges must be acknowledged due, but not limited to the following: (i) poor surveillance of SARS-CoV-2 circulation, which makes individual cases and outbreaks potentially difficult to capture; (ii) the clinical presentation of COVID-19 and Long COVID after the emergence of SARS-CoV-2 Omicron and more recent variants; these remain comparatively less studied than earlier variants from 2020–2021, including as regard their acute disease presentation and sequelae; (iii) the clinical presentation of COVID-19 and Long COVID in an increasingly complex and variegated immune landscape, where people are likely to have received multiple COVID-19 vaccine doses and been infected with SARS-CoV-2 more than once.

“The signs, symptoms, and conditions are present four weeks or more after the initial phase of infection.” (second sentence of the USG definition)

I broadly agree with the four-week threshold as a starting point [6], although I acknowledge there are different perspectives on the timeline. An example is the WHO’s clinical case definition of post-COVID-19 condition, a term which is often used as a synonym for Long COVID (similarly to other terms, such as PASC or post-acute sequelae of SARS-CoV-2 infection, and post-COVID conditions). The WHO defines post-COVID-19 condition as the continuation or development of new symptoms 3 months after the initial SARS-CoV-2 infection, with these symptoms lasting for at least 2 months, with no other explanation [35].

In previous work, co-authors and I have noted that “the point at which COVID-19 moves out of its acute phase remains unclear, and may vary in different patients” [6]; schematic and oversimplified expectations about the natural history of COVID-19 and Long COVID, therefore, should be avoided (Section 3.3.3 and Section 3.3.4). In addition, we underlined the need to not miss early thresholds for intervention [6]. An early timeframe for eligibility for a Long COVID diagnosis might be helpful in this regard. Nonetheless, patients should be able to access care at any time over their disease course (e.g., six weeks from infection, even if formal diagnosis is possible only after three months).

There can be delayed onset of symptoms, signs and conditions after the acute phase of COVID-19 (the duration of which remains of contingent definition) and down the disease line, namely, far from infection. New symptoms, signs and conditions have been reported to appear in time, even after asymptomatic, subclinical, pauci-symptomatic or so-called “mild” onset of COVID-19 [6,10,36,37,38,39]. Long COVID can be the outcome of different levels of disease severity in the acute phase. This can range from asymptomatic/pauci-symptomatic to life-threatening COVID-19 that required hospitalization [6,10,36,37,40,41] Long COVID can affect patients of all ages, genders, ethnicities and prior health statuses [6,10,18,19,37,42,43,44].

“Long COVID may be multisystemic; and may present with a relapsing–remitting pattern and progression or worsening over time, with the possibility of severe and life-threatening events even months or years after infection.” (third sentence of the USG definition)

These are strong points that should be maintained [45]. This must include the mention of severe and life-threatening events far from infection. The existence of such events and life-threatening conditions (e.g., stroke and thromboembolism) is proven by lived experience [19] and case reports [38,39,46,47,48,49], as well as data on mortality and increased risk for death in Long COVID and large patient cohorts assessed following SARS-CoV-2 infection [9,50,51,52,53,54]. Awareness around this point is fundamental to save lives. While it is possible—but not assured—that the highest risk of death and severe events in Long COVID is closer to infection in some [55], it must be considered that people are being reinfected with SARS-CoV-2. In addition, the longer-term effects of infection remain to be fully elucidated, as further discussed in Section 3.3.8. Some conditions–complications associated with Long COVID, such as diabetes and cardiovascular disease, can be fatal or disabling far from onset, especially if no treatment is provided [56].

I suggested to add “[with the possibility] of lifelong disability” to the USG. It is already known that some outcomes of SARS-CoV-2 infection do not have a cure in view of current medical knowledge or are irreversible, such as, in extreme cases, amputations and transplants [57,58]. There is no certainty as of 2025 that a cure for Long COVID will be found, leading to full disease resolution. In addition, there is no certainty that any such cure will be deployed globally and made available to all those in need. People should not be denied access to care and support services in view of erroneous expectations about potential recovery. Several studies on Long COVID have already shown persistence of pathology, low rates of disease resolution in many sampled patient cohorts, and ongoing impairment years from infection [40,51,52,55,59].

“Long COVID is not one condition.” (fourth sentence of the USG definition)

This point deserves further clarification. It must be noted, however, that Long COVID can manifest with many associated conditions and complications affecting all organs and body systems [6,8,9,10,11,12,13,60,61,62], including but not limited to the cardiovascular system [48,52,54,63], lungs [37,64,65,66], heart [67,68], eyes [69], pancreas [70,71], brain [72,73,74], liver [75], immune system [76,77], gastrointestinal system [78], kidneys [79], and skin [80]. Multiple organs can be affected at the same time [12], The presence of conditions such as diabetes in Long COVID has been highlighted [81]. The extent of pathology and disease severity can vary from individual to individual, as well as over time [19,39,40,41]. The kinds of organs and body systems affected—or more severely affected—have also been shown to vary from person to person and across the disease course [10,19,37,39,40,41]. For example, Qurban and Mullen described the case of a 35-year-old male cancer survivor who (i) developed severe COVID-19 in December 2021; (ii) was hospitalized in the intensive care unit (ICU) with pulmonary emboli, labile blood glucose levels and superadded Klebsiella pneumonia infection; (iii) was readmitted in January 2022, testing positive for SARS-CoV-2 again, and treated with oxygen; (iv) was found with pancreatitis on a routine FDG-PET scan weeks later, a condition which was interpreted as COVID-19-related in view of anamnesis and presentation; (v) was PCR-negative and had apparently recovered from pancreatitis by six months; but (vi) remained with diabetes and in need of insulin at the latest documented follow-up (long-term pulmonary assessment was not available in the case report) [71].

“Long COVID represents many potentially overlapping entities, likely with different biological causes and different sets of risk factors and outcomes.” (fifth sentence of the USG definition)

This is a valid point and is noted in the literature [6,10,22]. More clarity is needed about what “entities” mean in the context of case definitions such as the USG.

3.3. Long COVID: Pressing Issues to Address

In the second part of this review, I provide further comments, based on research, pathophysiology and lived experience, which can help relevant stakeholders to address pressing issues around Long COVID as a disease entity and as a concept. Awareness of these issues can contribute to confirming or ameliorating the US definitions addressed and developed at NASEM (USG and NASEM definitions), as well as any definition and guidelines for Long COVID more in general. Stakeholders who can benefit from the following discussion include but are not limited to policymakers, users of the USG and NASEM definitions, healthcare professionals and health bodies, researchers, patients, carers, advocacy groups and the media.

3.3.1. Tests and Diagnostics: Recognize Issues Around Access and Quality of the Tests Available

It must be recognized that not everyone has access to the best diagnostic investigations. Barriers may include costs, poor access to healthcare facilities and unwillingness of providers to prescribe certain tests [44]. In addition, some damage to organs and tissues can be identified only, or primarily, via procedures such as autopsies (which were critical to early discoveries in pathophysiology [82,83]), biopsies and/or through sophisticated diagnostics, which are often available only in research settings [64]. It is also possible that some biological damage cannot be appropriately identified with existing tools. Damage to specific organs and body systems, therefore, can be present, even when patients do not have evidence of it on paper. Lack of access to appropriate testing should not preclude people with symptoms, impairments and physical limitations from receiving adequate support, such as disability benefits; accommodations; sick leave; additional, more sophisticated clinical investigations; and even treatment when needed and possible.

3.3.2. Concerns over Excessively Narrow Definitions

Concerns must be raised over the possible adoption of excessively narrow definitions of Long COVID. Care must be taken to avoid the following:

• “Hasty” diagnoses, namely, diagnoses made too fast, especially in a time of crisis like a pandemic, in the context of emergency medicine, and when facing imperfect scientific knowledge about an emerging disease entity [6].
• Overly restrictive, specific or narrow clinical case definitions, for example, based on a few symptoms only [84]. Overly narrow definitions could exclude people from a diagnosis of Long COVID and any potential diagnostics or treatments.
• Overly “firm”, inflexible classifications in sub-types (such as phenotypes and endotypes), which could constrain any treatment options for patients and lead to imprecise clinical case definitions, when (i) knowledge about Long COVID remains incomplete; (ii) each patient can face their own specific clinical challenges, which might not necessarily fit into predetermined classifications; (iii) the course of illness can be fluid, relapsing-remitting and progressive (e.g., the clinical presentation of patients might change in time) [19,45]; and (iv) not everyone with Long COVID has access to adequate investigations to prove they have a specific manifestation of the disease entity (Section 3.3.1).

While billions of people are likely to have been infected with SARS-CoV-2, only a minority would seem to have been offered in-depth clinical and biomedical investigations for Long COVID. Subclinical and pauci-symptomatic pathology, in particular, might remain undiagnosed. For example, Heiss and colleagues used low-field-strength MRI to show persistent pulmonary dysfunction in both children and adolescents who appeared to have recovered from COVID-19 and in those classified as having Long COVID [65]. Knowledge about disease presentations and pathophysiology across the patient population as a whole remains, therefore, incomplete. While thousands of studies on Long COVID pathophysiology exist (LitCovid), the range of biological damage from SARS-CoV-2 infection remains to be fully elucidated as of 2025. An inclusive and expansive definition (broad) would protect, under such a definition, patients that could otherwise be dismissed as not matching overly restrictive diagnostic criteria.

Long COVID, however, does not mean “vague symptoms” or an “ill-defined syndrome” as it has erroneously been described in some papers and press articles [85]. Rather, it is an intentionally inclusive and “capacious” [6] term, attentive to the heterogeneous manifestations and developments in time of the biological pathology triggered and potentially driven by SARS-CoV-2 infection.

Critically, therefore, an inclusive and expansive case definition should not preclude support, treatment and specific sub-diagnoses under Long COVID when the evidence for such conditions is found in a patient (e.g., thromboembolic disease [9,52], heart failure [52], diabetes [81,86], dysautonomia [11], pericarditis [67], immune dysfunction [76,77] and kidney disease [79]). As noted in Section 3.2, many conditions and complications have already been linked to Long COVID in the literature, including with fatal outcomes.

Similarly, a broad and expansive definition, especially if adopted for use in a clinical context, should not preclude the use of specific research or clinical endpoints in relevant settings.

3.3.3. Heterogeneous Manifestations and Personalized Approaches

Personalized and interdisciplinary approaches are crucial, given the broad and heterogeneous manifestations of COVID-19 itself [6,57,83]. Heterogeneity across the COVID-19 patient population appears to be reflected in the “long” tail of the disease (Long COVID). Not all patients will necessarily present with the same symptoms, signs and sequelae in Long COVID (Section 3.2).

Consideration must be paid to different disease pathways, within a fluctuating disease process [10,45] and clinical presentations that can be multi-phasic, progressive and multifaceted [6,7,18,19,45]. The natural history and development of COVID-19 and Long COVID can vary across the patient population. Lang and colleagues, for example, reported a critical case of COVID-19 in a 44-year-old female, leading to lung transplant approximately two months post infection [57], while Ghenu and co-authors described prolonged COVID-19 lung pathology in a vaccinated 62-year-old male, with a fatal outcome around 2.5 months after positive PCR [49]. In contrast, Buonsenso and colleagues addressed the case of a female adolescent with milder acute presentation but evidence of pulmonary pathology almost one year into Long COVID [37]. Guidelines and definitions, therefore, “need to reflect the lived experience” [87] and be attentive to disease heterogeneity. The ongoing emergence of new SARS-CoV-2 variants, as well as reinfections, might also change the clinical presentation of patients with each reinfection, or after infection with a different variant. This topic deserves further attention (Section 3.3.9).

3.3.4. Hospitalization and Disease Severity

Hospitalization alone should not be used as an inflexible indicator of COVID-19 and Long-COVID severity. Many factors—including gender, class, race, ethnicity and bed availability in COVID-19 hotspots—can play, or have played, a significant role in access to care, rather than severity alone [5,6,7,19,20]. There are patients who were not hospitalized in acute COVID-19 and yet died from reported COVID-19-related complications beyond four weeks from onset [38,48]. On the other hand, people with a milder acute COVID-19 presentation, as well as an apparent resolution of symptoms after the initial infection, are reported to have been hospitalized in the acute phase; for example, Ghosh and colleagues presented the case of a 56-year-old male with pericarditis as the sole (identified) manifestation of acute COVID-19, who was discharged on day five and appeared to have been recovered on follow up after three weeks [88]. Other patients have been hospitalized or seen as outpatients in Long COVID but not in acute COVID-19 [37]. Some patients have remained hospitalized for weeks to months with different outcomes, ranging from fatal events [39,49] to discharge [57,89].

Many people with disease classified as “mild” at onset have developed Long COVID [6,7,10,11,37]. In previous work, co-authors and I have cautioned against the use of “mild” as a way to strictly characterize COVID-19 and Long COVID without acknowledging biases in the value attached to this term since the early pandemic and, more general, in medicine [6,7]. In early 2020, the presence of the clinical entity known as COVID-19 pneumonia, together with hospitalization, was often considered evidence of disease severity [6,7]. Yet, for example, Bandirali and colleagues described patients from an early Italian hotspot with radiological evidence of COVID-19 pneumonia on X-ray, who were minimally symptomatic and seen as outpatients in their clinic [90]. Symptoms in the acute phase, therefore, are not necessarily a clear indication of the underlying pathology and its severity. At any rate, symptoms of acute and Long COVID affecting an individual’s level of functioning and quality of life should not be called “mild” [6].

3.3.5. Early Treatment

Early treatment is fundamental to avoid deterioration and fatal events [6]. Access to early treatment and appropriate care can be difficult to obtain, however, for many people living with Long COVID [44,91,92], as already reported in 2020 [7,29]. The provision of early diagnosis and any possible treatment should be an area of utmost urgency. This warrants caution in delaying diagnosis and adopting a narrow case definition of Long COVID, whereby patients might be denied access to adequate care and support as a consequence (i.e., because they are not diagnosed).

3.3.6. Long COVID as a Framework to Approach Novelty and Associated Conditions

Long COVID, as a concept, offers a unifying framework to understand and reflect upon pathology triggered by SARS-CoV-2 infection [6]. First, it allows for addressing features which are, or could be, specific to SARS-CoV-2 infection [93], not least viral persistence of SARS-CoV-2 [94,95]. Other COVID-19-specific features could yet be identified.

SARS-CoV-2 infection could trigger, or drive directly, if the virus persists, a change in the whole biological landscape of the human body (e.g., metabolic profiles, in the cardiovascular system, in the coagulation cascade and in how people might respond to medication, medical procedures and infectious diseases other than COVID-19) [96,97]. This issue must be addressed urgently. Research, clinical trials, disease monitoring and disease management need to acknowledge the unknown. Relevant stakeholders, such as health bodies and healthcare professionals, must be ready to consider and integrate new, emerging evidence on Long COVID fast and effectively. Avoiding a narrow case definition could afford flexibility in this regard: a broad definition would not need to be updated each time a new discovery is made, while such new knowledge could be still integrated rapidly in, for example, research studies.

Recognition of any COVID-19-specific damage must not preclude access to treatment, where such treatment is appropriate (e.g., for diabetes post-COVID-19). Some conditions associated with Long COVID, such as cardiovascular disease, are particularly worrisome as they can be fatal [38,48,52].

Another important issue, which remains comparatively less addressed in research and clinical practice, is whether diseases, such as diabetes or rheumatoid arthritis, when diagnosed after SARS-CoV-2 infection, present with the same features of such diseases as they existed pre-pandemic. Namely, if one is diagnosed after COVID-19 with a disease meeting the clinical case definition of, for instance, rheumatoid arthritis or diabetes, it must be considered whether such conditions might present specific features induced by the trigger (SARS-CoV-2 infection) or influenced by the patient having had COVID-19—for example, a more significant cardiovascular component than one would have expected pre-infection or additional COVID-19-related comorbidities, such as the sequelae of COVID-19 lung pathology [19]. For example, Baimukhamedov and colleagues described the case of a 67-year-old male who developed rheumatoid arthritis after severe acute COVID-19, while displaying residual lung pathology weeks after infection. The patient had returned to work months post-COVID-19, but a longer follow up was not available [98]. Rubino and colleagues, on the other hand, discussed whether the type of diabetes triggered by COVID-19 could be a “a new type of diabetes”, as well as its “natural history” and “management” [99].

This issue has implications for the diagnosis and treatment of many conditions after SARS-CoV-2 infection. For example, it is worth addressing the extent to which medications normally prescribed for such conditions would remain equally effective and safe after SARS-CoV-2 infection.

3.3.7. Differential Diagnosis and Diagnosis of Exclusion

Differential diagnoses for the identification of conditions that are not Long COVID remains crucial (e.g., lung cancer when the patient presents with respiratory symptoms). Yet, Long COVID should not be a diagnosis of exclusion. This could minimize suffering, disenfranchise patients from care and make diagnosis more difficult. Ramifications and complications of Long COVID, such as metabolic, neurological, immunological, cardiovascular, pulmonary and renal conditions, should be part of clinical diagnosis and research endpoints, with due consideration to issues raised in this review (e.g., Section 3.3.6).

3.3.8. Long-Term Sequelae from Infections Including Cancer and Neurodegenerative Conditions; SARS; Existing Knowledge

The risk of longest-term sequelae from SARS-CoV-2 infection, such as, but not limited to, neurodegenerative pathology or cancer, remains to be elucidated. This risk should be addressed in view of emerging evidence [100,101,102,103]. Other infections, including viral infections, have been linked to conditions such as cancer and neurodegenerative diseases [104,105]. This calls for an application of the precautionary principle, with dedicated research and epidemiological surveillance on such prospective sequelae. Chronic inflammation, immunosuppression and DNA modifications are implicated in cancer susceptibility and documented following SARS-CoV-2 infection [100,101]. Chia and colleagues, in addition, reported that influenza viruses and SARS-CoV-2 awakened metastatic breast cancer cells in the lungs in an animal model [102]. Concerns over neurodegenerative disease following SARS-CoV-2 infection have been raised in view of research on pathophysiology, including in animal models [106], while cohorts assessed after COVID-19 have shown evidence of cognitive decline [107], brain pathology [72,73,74,107] and higher rates of neurodegenerative diseases as opposed to controls [22,103]. Further epidemiological surveillance, therefore, is strongly recommended.

Infections, including viral infections, are also linked to a wide array of sequelae and specific conditions emerging or becoming apparent years or decades after infection. Examples include but are not limited to multiple sclerosis and the widely discussed link with the Epstein–Barr virus, AIDS from HIV infection and post-polio syndrome from poliovirus infection [108,109,110,111]. Not rarely, some such infections might start with an asymptomatic or pauci-symptomatic acute phase, followed by later sequelae. This evidence cautions against considering SARS-CoV-2 infection a short-term risk even in apparently “mild” acute cases. Adequate investigations into SARS-CoV-2’s longer-term effects are warranted.

COVID-19 and Long COVID are some of the most studied disease entities in medical history. As of 23 May 2023, a search I carried out in the repository for medical papers PubMed using *COVID rendered over 356,000 results, with over 460,000 as of 10 June 2025. There are several thousand publications on Long COVID alone (LitCovid). This research constitutes a significant body of knowledge to build upon. Novel COVID-19 research is a massive addition to existing biomedical knowledge predating the pandemic and involving conditions related to infections, such as myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) and dysautonomia [10], as well as viral persistence and sequelae in relation to pathogens other than SARS-CoV-2, such as the Dengue or Ebola viruses [112,113,114]. There is evidence of prolonged illness from SARS-CoV infection–SARS (long SARS–post SARS) [5,115].

In view of all this, it should not be said Long COVID is “mysterious” or a “medical mystery”, as it has been described in some venues [116], nor it should be assumed all Long COVID patients will recover with time (some SARS survivors did not [5,115]). There is no doubt viruses, including SARS-CoV-2, are significant contributors to the disease burden in humans. The risks from SARS-CoV-2 infection, therefore, should be communicated appropriately to the general public. Clinicians and other relevant stakeholders should be promptly informed (e.g., by health authorities) on both established and emerging biomedical data. The dissemination of appropriate information is critical for building or adopting effective guidelines for patient care.

3.3.9. Omicron and Reinfections

More research is needed on the variant of concern (VOC) Omicron, the more recent BA.2.86, with its descendant JN.1, any further VOC and highly mutated variant emerging in the future, and the effects of multiple infections with SARS-CoV-2 over time. This is with particular regard to pathophysiology and clinical manifestations. Some existing research has elucidated variant-specific differences in the clinical presentation of COVID-19 [117]. A case definition of Long COVID must be flexible enough to allow for integration of emerging knowledge around these issues. Reinfections have been reported to increase the risk for Long COVID and impact negatively on the health of people already living with Long COVID [118].

3.3.10. Long COVID in Children and COVID-19 in Pregnancy

Attention to pediatric Long COVID is critical. The existence of Long COVID in children and young people was already reported in 2020 [7,18,19,36,42]. Research on pathophysiology and clinical manifestations is now growing fast, although pediatric patients remain underrepresented in Long-COVID research. As for Long COVID in adults, studies include but are not limited to detailed case reports [37], case series [74], analyses of large patient cohorts following SARS-CoV-2 infection as opposed to controls [86] and systemic reviews [36,119,120]. Research has addressed issues such as but not limited to the long-term clinical characteristics of children with Long COVID [121]; increased risk for conditions such as diabetes and autoimmune diseases following SARS-CoV-2 infection [86,122]; evidence of prolonged inflammation [37]; and altered metabolic phenotypes [96], as well as the impact of pediatric Long COVID on organs such as the liver [75], lung [37,65,66], brain [72,74] and heart [36]. However, more research is needed to fully elucidate pathophysiology and potential treatments—as for Long COVID in adults. Long COVID in children is an urgent topic also in the context of repeated infections. The impact of SARS-CoV-2 infection on pregnancy, the fetus and embryos is another urgent area to address, given the growing evidence of harm in this regard [123,124].

The response to the COVID-19 and Long COVID crisis will have a deep impact on the future of children who are being infected with SARS-CoV-2, not rarely multiple times. Children may also face specific challenges to care and diagnosis, such as dependence on caregivers, including when being preverbal, or poor social and medical recognition of Long COVID in pediatric patients [125].

3.3.11. Advocacy and Patient-Driven Research

Long COVID is the patient-coined and a patient-preferred term for the long-term health effects of SARS-CoV-2 infection [7,45]. I invited the NASEM committee to acknowledge the immense amount of collective, grassroots, international often unpaid and unrecognized labor, the Long-COVID community has mobilized to achieve recognition, including for children [5,6,7,18,19,20,42,45]. These contributions to knowledge must be recognized in future clinical and research efforts, as well as in policymaking. People living with Long COVID must be afforded equitable access to decision making for the creation of effective guidelines and definitions.

3.4. 2024 NASEM Long COVID Definition

The NASEM definition reads as follows:

“Long COVID (LC) is an infection-associated chronic condition (IACC) that occurs after SARS-CoV-2 infection and is present for at least 3 months as a continuous, relapsing and remitting, or progressive disease state that affects one or more organ systems.

LC manifests in multiple ways. A complete enumeration of possible signs, symptoms, and diagnosable conditions of LC would have hundreds of entries. Any organ system can be involved, and LC patients can present with single or multiple symptoms, such as shortness of breath, cough, persistent fatigue, post-exertional malaise, difficulty concentrating, memory changes, recurring headache, lightheadedness, fast heart rate, sleep disturbance, problems with taste or smell, bloating, constipation, and diarrhea. single or multiple diagnosable conditions, such as interstitial lung disease and hypoxemia, cardiovascular disease and arrhythmias, cognitive impairment, mood disorders, anxiety, migraine, stroke, blood clots, chronic kidney disease, postural orthostatic tachycardia syndrome (POTS) and other forms of dysautonomia, myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS), mast cell activation syndrome (MCAS), fibromyalgia, connective tissue diseases, hyperlipidemia, diabetes, and autoimmune disorders such as lupus, rheumatoid arthritis, and Sjogren’s syndrome. Important Features of LC:

LC can follow asymptomatic, mild, or severe SARS-CoV-2 infection. Previous infections may have been recognized or unrecognized.

LC can be continuous from the time of acute SARS-CoV-2 infection or can be delayed in onset for weeks or months following what had appeared to be full recovery from acute infection.

LC can affect children and adults, regardless of health, disability, or socioeconomic status, age, sex, gender, sexual orientation, race, ethnicity, or geographic location.

LC can exacerbate pre-existing health conditions or present as new conditions.

LC can range from mild to severe. It can resolve over a period of months or can persist for months or years.

LC can be diagnosed on clinical grounds. No biomarker currently available demonstrates conclusively the presence of LC.

LC can impair individuals’ ability to work, attend school, take care of family, and care for themselves. It can have a profound emotional and physical impact on patients and their families and caregivers.”

Overall, the NASEM definition is attentive to the heterogeneous manifestations of Long COVID. It is also attentive to equity considerations, such as recognition of people’s profound suffering and Long COVID in children. It is less concise than the USG, but it conveys more information. A number of points need additional elaboration, including about any potential drawbacks (e.g., use of “disease state” instead of disease or disease entity; mention of “anxiety”). My comments in Section 3.5 build upon the definition as of 2024–2025, which emerged from the NASEM’s activities [24]. I do not propose an alternative definition myself for this review. The full scope of the NASEM definition, including the accompanying volume by NASEM and Fineberg et al. [21], would deserve further attention in a separate article.

3.5. Observations on the 2024 NASEM Long COVID Definition

Here, I provide a critical review of the NASEM definition, sentence by sentence.

“Long COVID (LC) is an infection-associated chronic condition (IACC) that occurs after SARS-CoV-2 infection and is present for at least 3 months as a continuous, relapsing and remitting, or progressive disease state that affects one or more organ systems.” (NASEM definition’s first sentence)

Further discussion is necessary to better elucidate the terms “IACC” and “disease state” (instead of disease or disease entity). However, rejection of the term “syndrome” [22] is an important step, given some problems with this term in medical usage [6,42], where it can convey the idea of “an amorphous ailment” with “no physical basis” [22]. Formal recognition of disease persistence in Long COVID (“chronic”) is appropriate and important. Also appropriate is an acknowledgement of the “continuous”, “relapsing and remitting” or “progressive” nature of Long COVID [45], which has already been mentioned in the USG (Section 3.1).

Failure to mention life-threatening outcomes directly in the definition is a limitation, especially because this point was already present in the USG (Section 3.1 and Section 3.2), and Long COVID is documented to be fatal in some cases [50] (which NASEM-related literature, indeed, acknowledges [22]). The three-month timeline deserves further scrutiny to address any potential drawbacks, including the risks of limiting diagnosis, research and access to care (in view of what falls within or outside the proposed timeline). Issues around timeline are further addressed in Section 3.2 and below. The presence of pathology in one or more organs in Long COVID is documented in the biomedical literature [8,9,10,11,12,13] and is a strong point of the NASEM definition.

“[Long COVID] manifests in multiple ways. A complete enumeration of possible signs, symptoms, and diagnosable conditions of [Long COVID] would have hundreds of entries. Any organ system can be involved, and [Long COVID] patients can present with single or multiple symptoms, such as shortness of breath, cough, persistent fatigue, post-exertional malaise, difficulty concentrating, memory changes, recurring headache, lightheadedness, fast heart rate, sleep disturbance, problems with taste or smell, bloating, constipation, and diarrhea.” (NASEM definition’s second sentence)

Attention to the heterogeneous manifestations of Long COVID is a strong point. Recognition that “any organ system can be affected” is also a strong point, which is documented in the biomedical literature [8,9,10,11,12,13]. The symptoms and signs highlighted by NASEM are all mentioned in the literature [10,11,13]. It is appropriate for the definition to underline that not all signs, symptoms and conditions can be directly mentioned in the definition itself (given the broad range of sequelae in Long COVID). Mentioning only some symptoms, however, could make those symptoms appear more important than others. Nonetheless, the definition clarifies the symptoms mentioned are mere examples (“such as”).

“[Long COVID patients can present with] single or multiple diagnosable conditions, such as interstitial lung disease and hypoxemia, cardiovascular disease and arrhythmias, cognitive impairment, mood disorders, anxiety, migraine, stroke, blood clots, chronic kidney disease, postural orthostatic tachycardia syndrome (POTS) and other forms of dysautonomia, myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS), mast cell activation syndrome (MCAS), fibromyalgia, connective tissue diseases, hyperlipidemia, diabetes, and autoimmune disorders such as lupus, rheumatoid arthritis, and Sjogren’s syndrome.” (NASEM definition’s second sentence, continued)

The focus on the heterogeneous manifestations of Long COVID is appropriate. However, mentioning only some conditions directly in the definition could make those conditions appear more important than others, which would be a limitation. Nonetheless, the definition clarifies the conditions mentioned are only examples (“such as”). More discussion is needed on whether “diagnosable conditions” in Long COVID, such as rheumatoid arthritis or diabetes, would manifest as they did before the pandemic or whether they might present novel or unusual features due to the trigger (SARS-CoV-2 infection). This point is addressed more thoroughly in Section 3.3.6.

The emphasis on “anxiety” and “mood disorders” must be cautioned against, as it could support views of Long COVID as a psychological disorder, which Long COVID is not. While “anxiety” and “mood disorders” might be diagnosed in Long COVID, (i) the risk of misdiagnosis remains high, especially for certain groups like women; (ii) conditions such as “mood disorders” might be simply an effect of long-term illness, as it can happen with, for example, cancer; and (iii) symptoms of biological pathology, such as brain inflammation or brain hypometabolism [72,73,74], might be erroneously diagnosed as psychological conditions [6].

The NASEM definition also lists some “important features” of Long COVID, of which I offer a critical review as follows:

“[Long COVID] can follow asymptomatic, mild, or severe SARS-CoV-2 infection. Previous infections may have been recognized or unrecognized.” (NASEM definition’s first important feature)

This is a strong point, which is recognized in the literature (Section 3.1 and Section 3.2).

“[Long COVID] can be continuous from the time of acute SARS-CoV-2 infection or can be delayed in onset for weeks or months following what had appeared to be full recovery from acute infection.” (NASEM definition’s second important feature)

This is an important point, which acknowledges delayed onset from infection (Section 3.2).

“[Long COVID] can affect children and adults, regardless of health, disability, or socioeconomic status, age, sex, gender, sexual orientation, race, ethnicity, or geographic location.” (NASEM definition’s third important feature)

Recognition that everyone can potentially develop Long COVID is a strength of the definition, which follows the literature (Section 3.2). Recognition of Long COVID in children is crucial (Section 3.3.10).

“[Long COVID] can exacerbate pre-existing health conditions or present as new conditions.” (NASEM definition’s fourth important feature)

This is another pertinent observation, which acknowledges the wide-ranging implications of SARS-CoV-2 infection on human health (Section 3.2 and Section 3.3). Further discussion is needed on whether “pre-existing health conditions” might develop new features (e.g., at the cellular level) following SARS-CoV-2 infection. As suggested in Section 3.3.6, novelty might have implications for disease management, treatment and care.

“[Long COVID] can range from mild to severe. It can resolve over a period of months or can persist for months or years.” (NASEM definition’s fifth important feature)

Recognition of diverse degrees in disease severity is important. However, an additional focus on subclinical and pauci-symptomatic manifestations would have been welcomed (Section 3.3.2). There is also an apparent lack of emphasis in the NASEM definition on very severe manifestations and fatal outcomes, which are documented in the literature [50] (but see also [22]). There should be recognition that Long COVID can potentially last more than “years” and be permanent, although the definition acknowledges Long COVID as “chronic”. The possibility of lifelong disability deserves further attention (Section 3.2). Decades-long pathology is documented in SARS survivors and after other infections (Section 3.3.8). The potentially ambiguous value attached in medicine to “mild”, which might minimize people’s suffering, needs further attention [6] (Section 3.3.4).

“[Long COVID] can be diagnosed on clinical grounds. No biomarker currently available demonstrates conclusively the presence of [Long COVID].” (NASEM definition’s sixth important feature)

It is important to recognize that Long COVID can be diagnosed clinically. However, it must be more extensively addressed as to whether and to what extent biomarkers exist to help diagnose Long COVID.

“[Long COVID] can impair individuals’ ability to work, attend school, take care of family, and care for themselves. It can have a profound emotional and physical impact on patients and their families and caregivers.” (NASEM definition’s seventh important feature)

Recognition of the profound impact of Long COVID on patients and carers is a strong point of the definition.

In addition, NASEM-related literature [22] includes the following:

Recognizes and adopts officially the patient-coined term Long COVID (Section 3.3.11).

Recognizes the contributions of Long COVID advocacy, although with some imperfections (e.g., the Long COVID movement included professional researchers and physicians [5,6,7], not only “citizen scientists” [22]).

Acknowledges that the NASEM committee “determined at the onset” to engage with patients to foster “interdisciplinary dialogue”, “transparency” and “acceptibility of the definition” [22], which is a positive step (Section 3.3.11).

Recognizes that many infections can produce long-term symptoms and sequelae [5] (Section 3.3.8).

Acknowledges that antibody tests might not be reliable in picking up all SARS-CoV-2 infections (Section 3.2).

Recognizes Long COVID is not a diagnosis of exclusion (Section 3.3.7).

Highlights equity considerations, such as recognizing Long COVID in different patient groups and its severity (e.g., [22], Box 2).

Recognizes that “the 3-month disease threshold should not be taken as a reason to ignore symptoms at an earlier stage”, as addressed in Section 3.2. However, unintended consequences might arise around this point, for example, if only the definition and not the accompanying literature is considered by, for instance, healthcare professionals.

Recognizes Long COVID can be fatal, although this point is not included in the NASEM definition itself (Section 3.2)

4. Discussion and Conclusions

The COVID-19 pandemic is one of the most severe public health challenges the world has faced from an emerging infectious disease since the 1918–1920 influenza pandemic (the so-called “Spanish Flu”) and the HIV/AIDS epidemic [5]. In addition to an estimated death toll of at least 20 million people, as of 2022–2023, the pandemic is leaving behind a severe toll of morbidity in the form of Long COVID [5]. This will affect the health, economic standing and general wellbeing of populations for decades and generations. It is, therefore, imperative to move forward with urgency as regards the development of efficacious treatments, definitions and diagnostics for Long COVID.

In this review, I provided a critical appraisal of the following two case definitions for Long COVID developed in the US: the US Government working definition of Long COVID, or USG, and the more recent NASEM definition. My critical review of the USG builds upon the case definition as presented and discussed at the US National Academies of Sciences, Engineering, and Medicine (NASEM) in May 2023, when I was invited to join (virtually) a panel as a patient expert. Work by NASEM led to the current NASEM definition, which was published in 2024 [22]. In the review, I also highlighted some pressing issues that relevant stakeholders will want to consider when addressing Long COVID as a disease entity and as a concept. Reflection on these points can contribute to building effective case definitions and guidelines for Long COVID within and beyond NASEM and the US.

In particular, I underlined the importance of having an inclusive and expansive (broad) definition of Long COVID, especially for clinical purposes. An inclusive and expansive definition would maximize access to care, treatments, trials and support for people living with Long COVID, while minimizing the number of people potentially left behind (i.e., excluded from diagnosis and care in view of more restrictive, narrower definitions). The use of a broad definition is supported by the heterogeneous manifestations and fluctuating disease processes in Long COVID. Wide in scope, an inclusive and expansive definition will help to cover the broad range of biological pathology triggered and potentially driven by SARS-CoV-2 infection, which is being elucidated by scientific research. The use of an expansive definition should not preclude recognition and treatment of conditions that fall under Long COVID, for which diagnosis and treatment could be possible (e.g., diabetes and cardiovascular disease). It should, also, not preclude the use of narrower definitions for specific purposes, such as designing clinical trials or administering any potential therapeutics. A broad definition would allow for incorporating fast and effectively any emerging biomedical evidence while accommodating one’s individual disease presentation. The heterogeneous manifestations of Long COVID call for personalized and multidisciplinary approaches to care.

The review also highlighted the following: (1) It is imperative to promote adequate diagnostics and pathways to care, which are not available to all those in need. (2) We must be cautious about “hasty” and excessively narrow definitions for Long COVID. (3) We must reflect on the definition of “severity” in relation to COVID-19: hospitalization alone should not be taken as evidence of severity, while terms such as “mild” might misrepresent people’s suffering. (4) Long COVID is a novel concept useful for addressing the prolonged disease process triggered and potentially driven by SARS-CoV-2 infection. (5) Long COVID is a multi-system disease entity, incorporating multiple sequelae of infection. Long COVID should not be a diagnosis of exclusion, although people with Long COVID might meet the clinical case definition of one or more conditions already described before the pandemic, such as diabetes or cardiovascular diseases. (6) There is significant existing knowledge on COVID-19, Long COVID and other prolonged diseases triggered by infections, including from SARS-CoV infection (Long SARS–post SARS). This evidence highlights the urgency of the Long COVID crisis, as it shows many people living with Long COVID might not recover spontaneously over time. Existing and emerging knowledge should be used promptly and effectively to accelerate research, trials and access to any treatment. (7) There are concerns over potential longer-term sequelae of SARS-CoV-2 infection, which could become apparent in the years and decades to come, such as, but not limited to, cancer or neurodegenerative disease. (8) There is a need for more research on Omicron, new emerging variants and reinfections. (9) There is a need for more research on and recognition of Long COVID in children. (10) It is urgent to address the impact of COVID-19 and Long COVID on pregnancy and the fetus. (11) It is important to acknowledge the ground-breaking impact of Long-COVID collective, international, grassroots advocacy and patient-driven research, which led to recognition of Long COVID. Patient-driven expertise must be incorporated into knowledge building and policymaking.

Given the extent of the COVID-19 pandemic, there is no time to waste. People living with Long COVID deserve support, research, treatment and full recognition. They should not be excluded from diagnosis and access to care. It is imperative to move forward with urgency with the development of efficacious treatments and advanced diagnostics for Long COVID. The recognition of patient rights is also of the utmost importance, including access to disability services, social security and prevention of further infections.

Clinical case definitions for Long COVID must retain their focus on the broader spectrum and scope of the disease entity while incorporating feedback from people with lived experience, advocates and patient-researchers.