Long COVID is a complex, multi-system illness with the potential for a substantial impact on society, from increased health care costs to economic and productivity losses. Symptoms may persist for weeks or months following acute SARS-CoV-2 infection, come and go over time, or manifest as new onset chronic conditions, such as heart disease, diabetes, kidney disease and neurological conditions.

Long COVID is an umbrella term used to describe both ongoing symptoms in the medium-term (4–12 weeks) and longer-term sequelae beyond 12 weeks known as post-COVID syndrome (National Institute for Health and Care Excellence) or post COVID-19 condition (World Health Organization).

This review analyses the available Australian and international literature to understand the impact and scale of long COVID, including:

  • incidence and prevalence of long COVID in Australia and internationally
  • whether SARS-CoV-2 variants and vaccination modify the risk of developing long COVID
  • demographic, clinical and social determinants of long COVID
  • outcomes and impact of long COVID on patients, such as burden of disease, health service use, quality of life and patient experience
  • data deficiencies and research gaps around long COVID.

Many studies from the early phases of the pandemic were conducted before clear definitions were developed and produced wide variation in results. In addition, there has been no consensus on a core set of health outcomes to be measured and reported for long COVID which has also translated into inconsistent findings.

Prevalence of long COVID

As most cases of COVID-19 have occurred in Australia during 2022 studies of the occurrence of long COVID have only recently gathered momentum. From the limited data available, current prevalence estimates of long COVID (defined as >12 weeks) in Australia range from 5% to 10% of COVID-19 cases.

Wide variation in estimates has been reported from international data, ranging from 9% to 81% in a global systematic review. Sources of heterogeneity include methodological differences between studies including definitions of long COVID and follow-up time, geographic region, demographic and clinical profile of study participants and acute
COVID-19 disease severity.

Studies using stricter case definitions for long COVID have produced more modest estimates. The prevalence of post COVID-19 condition (>12 weeks) ranged from 8% to 17% in studies from the UK. The global prevalence of post COVID-19 condition was estimated to be 6.2% of symptomatic COVID-19 infections when only symptoms of fatigue, cognitive problems or shortness of breath were counted.

Many studies lack non-COVID-19 comparison groups that are needed to establish whether the prevalence can be attributed to COVID-19, which is particularly important for studies that rely on self-report of a diverse range of signs and symptoms that are not unique to long COVID.

Regardless of the precise definition of long COVID used by individual studies, most studies find a relationship with severity of acute COVID-19. Prevalence is highest in patients who were admitted to an intensive care unit (ICU) for COVID-19, followed by hospitalised patients, and lowest in non-hospitalised patients.

There is growing evidence that the risk of long COVID has been lower during the Omicron wave compared with earlier SARS-CoV-2 variants. However, because many people were vaccinated when the Omicron variant emerged, observed differences in risk of long COVID in relation to different SARS-CoV-2 variants could be due to vaccination. A meta-analysis of 18 studies found that the risk of long COVID was 32% lower (relative risk [RR] 0.68, 95% confidence interval [CI] 0.53–0.87) based on studies using a >4-week definition and 25% lower (RR 0.75, 95% CI 0.64–0.88) for other definitions combined for people double vaccinated against SARS-CoV-2 compared to unvaccinated people.

Determinants of long COVID

There is growing evidence that severity of acute disease, age, female sex and comorbidities are the most common risk factors for the development of long COVID:

  • Severity of illness during the acute COVID-19 infection has been identified by numerous studies as an important predictor of long COVID. This includes the number of symptoms, length of hospital stay and ICU admission.
  • Long COVID has an inverted U-shaped relationship with age and is most common in middle-aged adults.
  • Studies have consistently shown that females experience a higher prevalence of self-reported long COVID than males, a finding that is independent of demographic and clinical characteristics.
  • Poorer underlying health is also related to an increased risk of long COVID. Pre-existing chronic conditions and their associated risk factors, such as obesity and smoking, increase the risk of developing long COVID.

People from lower socioeconomic groups, certain occupations and ethnic backgrounds may also be at a higher risk of developing long COVID. However, there is a lack of robust research focusing on social determinants and long COVID. Understanding the burden of long COVID in specific population groups is important to target prevention and develop treatment and care programs.

Very few protective factors other than SARS-CoV-2 vaccination have been identified. There is emerging but preliminary evidence that management of acute COVID-19 infection with antiviral medication and physical activity may reduce the risk of long COVID.

Long COVID and chronic conditions

Some people develop a range of multi-organ symptoms that may arise as a direct complication during the acute COVID-19 illness or develop over the longer term leading to new-onset chronic conditions.

Studies of large health databases, predominantly from the US, have identified an increased risk of a range of chronic outcomes, including cardiovascular disease, metabolic disorders, and mental and neurological complaints up to 12 months following infection. Imaging and laboratory studies have demonstrated persistent structural damage to the heart which may result in increased hospitalisation for cardiovascular events such as heart attacks.

One of the most common neurological complaints is ‘brain fog’ characterised as difficulties with cognitive function, attention and memory. Some symptoms of long COVID, particularly persistent fatigue and post-exertional malaise, overlap with myalgic encephalomyelitis (ME), also called chronic fatigue syndrome (CFS). ME/CFS has also been associated with previous viral infections and the underlying pathophysiology between these sets of symptoms may be similar for the 2 conditions. Continued research into long COVID may provide further understanding of ME/CFS. Likewise, established research on ME/CFS may point to clues worth investigating in long COVID.

Impact of long COVID

The review examines 4 dimensions of the impact of long COVID: the population health impact through the burden of disease and mortality, impacts on the health system, quality of life and social impacts, and the patient experience with long COVID.

  • In Australia, since the start of the pandemic and up to 30 September 2022, there had been 10,279 deaths due to COVID-19 of which 123 (1.2%) were classified as being due to post COVID-19 condition. Long COVID contributed to 10% of the total burden of disease from COVID-19 in Australia in the first few months of 2022.
  • Several studies have reported increased post-acute COVID-19 health care utilisation and costs, including rehospitalisation, emergency department visits, outpatient visits and primary care attendances.
  • A significant proportion of people with long COVID report limitations on their daily activities and a reduced quality of life. In the COVID-19 Impact Monitoring Survey, 22% of respondents with symptoms lasting for 3 months or more reported their ability to carry out day-to-day activities had reduced substantially compared to before COVID-19. The impact of persisting symptoms can impact on workforce participation, including delays in return to work, and ongoing residual difficulties that impact the ability to perform the same duties or limit working hours.
  • The term ‘long COVID’ emerged as social terminology to describe patient’s experiences of the long-term health effects of SARS-CoV-2 infection. The use of online support groups and social media has been a key tool for patient advocacy, demonstrating the evolution of social attitudes and experiences of long COVID. In the early phase of the pandemic, long COVID sufferers expressed a lack of belief and recognition of their illness by health care professionals and struggled to access medical care. Over time, the sentiments have become more positive, reflecting increased knowledge, acceptance and awareness of long COVID and health system responses to the condition.

Data deficiencies and future research

One of the main limitations of long COVID research is the inconsistency in the definition of long COVID used. Specially, there is difficulty in effectively defining the condition’s symptoms and time course for research and clinical purposes. Two definitions have been developed by the National Institute for Health and Care Excellence (NICE) and the World Health Organization (WHO) that define parameters around the timing and duration of symptoms, but both remain broad in relation to symptoms. An international consensus study has produced a core outcome set for adults with post COVID-19 condition consisting of 12 outcomes in the domains of clinical, life impact and survival to improve harmonisation and comparability across studies.

In September 2020 WHO activated an International Classification of Disease, 10th Revision (ICD-10) code for post COVID-19 condition. Analysis of the use of the code in US health care records has shown that the uptake varies widely and currently underestimates the frequency of long COVID. As use of the code becomes more consistent, health care records will provide large and rich sources of data to understand the impact of long COVID, such as patterns of health service use among long COVID patients. However, health records may be impacted by behavioural differences in seeking care, the need for care depending on the severity of long COVID symptoms, disparities in the availability of care, obtaining a diagnosis of long COVID and having that diagnosis recorded in the patient record. These issues may lead to lack of representation in health records of some population groups.

Most of the evidence presented in this review has been from research conducted overseas. Some opportunities for research to monitor the impact of long COVID in Australia include:

  • The national COVID-19 linked data set is a project being conducted by the AIHW to link COVID-19 infection notifications from states and territories to national administrative health data sets including deaths, hospitals, aged care, immunisation, Medicare Benefits Schedule (MBS) and Pharmaceutical Benefits Scheme (PBS) data. This data asset will allow investigation of health outcomes post-COVID-19 infection, including the occurrence, risk factors and impact of long COVID. This information will be valuable for understanding health service demands arising from long COVID and could be useful in designing targeted long COVID models of care.
  • Long COVID clinics have been established across Australia to provide specialised care to people having long-term symptoms after COVID-19 infection. These clinics provide an opportunity to collect data on long COVID progression in affected individuals.
  • Large-scale national surveys, like those established in the UK and USA, provide rapid and relevant long COVID information including the tracking of the prevalence over time. This information is important for planning prevention and health care demand.
  • Several Australian longitudinal studies have included questions on COVID that will allow for analysis of post-COVID-19 outcomes. This includes the 45 and Up Study conducted by the Sax Institute.

Long COVID is a new condition and therefore the evidence so far is limited by a relatively short follow-up time since infection, particularly in Australia where most of the acute burden of COVID-19 has occurred in 2022 to date. Research and monitoring of long COVID is required to understand its impact in the Australian population and to corroborate findings with the evidence from other countries.