The COVID-19 pandemic has been present in the world for over four years, causing a significant burden on individuals known as “long Covid.” Long Covid poses significant challenges for those who have been infected with COVID-19 since it disrupts their everyday life in many ways. Furthermore, the government has faced difficulties in allocating resources, both in terms of funding and healthcare professionals, to adequately support these patients. The definition of long Covid was ambiguous in terms of patient monitoring, prompting various non-profit organizations, such as the World Health Organization (WHO), the National Institute for Health and Care Excellence (NICE), the Scottish Intercollegiate Guidelines Network (SIGN), and the Royal College of General Practitioners (RCGP), to attempt to define it through guideline development. ¹ Hence, this term was succinctly defined as “signs and symptoms that develop during or following an infection consistent with COVID-19, continue for more than 12 weeks, and are not explained by an alternative diagnosis. The condition usually presents with clusters of symptoms, often overlapping, that can fluctuate and change over time and can affect any system in the body. Long Covid includes both ongoing symptomatic COVID-19 (from 4 to 12 weeks) and post-COVID-19 syndrome (12 weeks or more)”.

More than 65 million people worldwide have been afflicted with long Covid for a period of three years since the COVID-19 pandemic started. ² The symptoms endured in a chronic manner for a significant period, coinciding with various systemic manifestations such as cognitive decline, fatigue, lethargy, dyspnea, and reduced functional capacity. These manifestations persisted for several months. Individual symptoms were analyzed in systematic reviews and meta-analyses, ³ with the aggregate prevalence of these symptoms being subject to considerable variation across studies and temporal variation in symptom patterns. One of the symptoms that is frequently reported is fatigue. It has been detected in patients ranging from 21.59 to 39.45%. Cardiopulmonary complications, including symptoms such as cough, sore throat, chest pain, and palpitations, are reported in approximately 12.21% to 18.41% of cases. Approximately 11.65-16.58% of the population reported neurological issues, which included headaches, cognitive impairment, memory deficits, difficulties with attention, dizziness, loss of scent and taste, and hearing problems. Anxiety, post-traumatic stress disorder (PTSD), sleep disturbances, and depression were reported by approximately 16.81% to 26.00% of the population. Thirteen symptoms were documented in the southern region of Thailand according to a cross-sectional study. The study identified a prevalence of long Covid in approximately 79% of all cases that were infected with the virus. The most frequently reported symptoms were lethargy (72.9%), cough (66.0%), and muscle pain (54.1%). ⁴

The characteristics of COVID-19 patients have exhibited dynamic shifts across different waves of the pandemic. In the initial wave, older adults, especially those aged 60 and above, were notably vulnerable to severe outcomes, with a higher prevalence of comorbidities such as hypertension and cardiovascular diseases. ⁵ Age demographics shifted during subsequent outbreaks, as infections increased among younger individuals. The emergence of variants, such as the B.1.1.7 lineage, introduced uncertainties in transmission dynamics. ⁶ The Omicron variant, unlike previous waves, had a significant impact on real-world vaccine implementation. Most individuals in this cohort have received a minimum of two doses of the COVID vaccination. Thus, it is possible that everyone has already attained herd immunity or population immunity. Patients who have previously been vaccinated experience a decrease in mortality rates, and the severity of cases is significantly diminished. ^{7– 9} Nevertheless, different SARS-CoV-2 virus variants may reduce the efficacy of vaccines. ¹⁰

Our previous study examined the vaccine effectiveness (VE) status in Thailand during the Omicron variant dominant epidemic. In Thailand, we are administering a mixture of various vaccine types and brands, including CoronaVac, ChAdOx1, and n-CoV-19-BNT162b2. The administration of 4 doses of CoronaVac-CoronaVac-ChAdOx1 nCoV-19-BNT162b2 demonstrated the highest vaccine effectiveness (VE). This vaccine regimen was able to prevent infection by 65% and reduce the occurrence of moderate-to-critical diseases by 82%. ¹¹ Hence, reducing the severity of acute illness resulted in a decrease in the occurrence of long Covid, which served as a potential predictor. However, in COVID-19, different mutational variants change properties as they confer immune escape and influence vaccine efficacy. Different vaccine durations revealed different ranges of protection before infection.

While effective vaccination may decrease the severity of acute illness, long Covid is influenced by various factors including sex (specifically females), smoking, weight, and comorbidities. ^{12– 19} These variables also contributed to an increased risk of long Covid. As follows, Omicron might be a difference in long Covid prevalence. So, the aims of this study were to find the overall prevalence, systematic symptoms prevalence, and prognostic factors of long Covid during the Omicron variant dominant epidemic at 3 and 6 months in Thailand.

This study investigates the point prevalence of long Covid in adult patients (aged 18 years and older) during the Omicron variant-dominant period of the COVID-19 pandemic. We further analyze the symptom-based prevalence and predictors associated with long Covid at 3- and 6-months post-infection, utilizing data from patients at Thammasat University Hospital (TUH) and Thammasat Field Hospital (TFH) in Thailand. Our findings reveal significant point prevalence rates of long Covid, with 52.59% of patients affected at 3 months and 28.57% at 6 months. The data underscore that the prevalence of long Covid was notably higher at 3 months, nearly double that observed at 6 months, reflecting the evolving clinical trajectory over time.

The point prevalence of prior research on extended COVID exhibited significant variation, encompassing a range of symptoms. Systematic reviews and meta-analysis (SR/MA) gathered publications on extended COVID from January 1, 2020 to November 2, 2021. They reported a combined prevalence of 13.6% (ranging from 1.2% to 68.0%) according to health records. ²⁰ LUO et al. conducted a study in Hong Kong during the Omicron variant dominant epidemic to investigate the frequency of long Covid. They identified a total of forty-three symptoms via telephone follow-up and determined that 55.0% of patients experienced at least one persistent long Covid symptom for a duration of at least 6 months but no longer than 12 months. ²¹ The authors noted that there was a significant variation and likelihood of bias in the study due to the absence of consistent criteria, as well as issues related to recall, nonresponse, and differences in follow-up. ²²

Many studies reported individual symptoms such as fatigue and weakness, memory impairment, insomnia, cough, and headache. ^{15, 23– 27} Systematic reviews and meta-analysis by Natarajan et al. categorized phenotype of symptoms. To compare their reviews and our study on classification, their study showed only five groupings. Unlike our study, overall mental disorders (i.e., sleep disturbance, post-traumatic stress disorder (PTSD), anxiety, depression, anxiety/depression) had the highest pooled prevalence with 21.3%. Overall cardiopulmonary problems (i.e., chest pain, sore throat, dyspnea, palpitations, cough), overall general problem (i.e., mobility dysfunction, fatigue, general pain, joint pain, muscle pain, hairfall, skin rash, weight loss, fever), overall neurological problems (i.e., headache, cognitive impairment, memory deficit, attention/concentration deficit, dizziness, smell loss, taste loss, hearing problem) showed pooled prevalence of 14.9%, 14.4% and 13.9%, respectively. Overall gestational problems (i.e., poor appetite, diarrhea, nausea, vomiting, abdominal pain) had the lowest pooled prevalence at 6.2%. However, most of this SR/MA were reported high of heterogeneity ( I ² ) more than 90% in each single criterion.

When we analyzed prognostic factors for long Covid by multivariable logistic regression, severe-to-critical illness during the acute phase was a strong predictor at 3 months. A similar trend was observed at 6 months; however, the association did not reach statistical significance at 6 months. Female sex and healthcare worker occupation were significant predictors at both time points, while obesity was significant only at 3 months. Many publications reported severe & critical acute illness as precipitating long Covid because the damage occurred to the host varied depending on disease severity. ^{15, 17} However, studies have shown that some asymptomatic COVID patients who were not hospitalized still developed long Covid, ^{28– 31} but our study had a different result. Healthcare workers who have recovered from a COVID-19 infection may nevertheless come into frequent contact with COVID patients and hard work, ^{32– 34} and some studies reported that healthcare workers were more frequently reinfected with COVID. ³⁵ Therefore, this component exhibited a second-order level of strength in our study. Female was a strong associated factor with long Covid at 3 and 6 months, suggesting that fluctuations in hormonal regulation in females might influence a robust immune response against viral infection. ^{36– 40} Almost all studies suggested that females were more likely to develop long Covid than males. ^{15– 19, 41} As for obesity, every 1 kg/m ² of BMI increase in patients was shown to elevate the risk of long Covid. Also, obesity triggered severity during acute illness, which led to more potential of having long Covid. ⁴² Obesity was found in Vimercati, L et al. and Subramanian, A et al. which studied long Covid. They found that overweight (BMI > 25 kg/m ²) people was more likely to develop long Covid. ^{19, 42} The potential interaction between age and BMI also merits attention. These two factors are physiologically linked, as aging is associated with notable changes in body composition, such as increased visceral fat, loss of muscle mass, and alterations in metabolic regulation. Such age-related shifts can influence immune function and inflammatory pathways, potentially shaping how obesity contributes to long COVID risk. Additionally, older adults may develop sarcopenic obesity, which could influence symptom persistence in a manner distinct from that of younger individuals with elevated BMI. In our analysis, however, the age–obesity interaction did not reach statistical significance. Higher education level, particularly postgraduate attainment, was associated with a reduced risk of long COVID at 3 months. This finding may reflect differences in health literacy, access to healthcare resources, adherence to medical advice, and the ability to engage in effective self-management during recovery. Individuals with higher educational attainment may also have greater capacity to access reliable health information and implement preventive strategies, which could contribute to improved recovery trajectories. However, this interpretation should be approached cautiously. It is possible that educational level serves as a proxy for socioeconomic status, occupational flexibility, or baseline health differences rather than exerting a direct biological effect. Furthermore, individuals with higher education may differ in symptom perception or reporting behavior, which could influence observed associations. An important finding in our study was the association between gastrointestinal symptoms during the acute phase and long COVID at 6 months. Although the overall prevalence of gastrointestinal symptoms at follow-up was low, diarrhea during the acute infection was independently associated with persistent symptoms. This finding suggests that early gastrointestinal involvement may indicate underlying systemic inflammatory responses that contribute to prolonged symptom persistence.

This study did not report some predictors that were found in other studies, such as vaccine dose-related long Covid cause of type of vaccine, older adults, comorbidities including chronic obstructive pulmonary disease, diabetes mellitus, chronic kidney disease, cancer, and symptoms at acute illness. ^{17– 19, 42– 46} The data showed a significant p-value, but the 95% confidence interval had no impact on the effect. However, the study did not explore treatment during acute illness such as treatment satisfaction, early or late treatment, and traditional Thai medicinal usage.

Our study highlights a large sample size of long Covid evidence particularly in the dominant wave of Omicron variant dominant epidemic in Thailand by telephone survey. As the Omicron variant dominant epidemic was announced by WHO from November 26 ^th, 2021, our data seemed to contain less contamination from other variants. ^{9, 47– 51} Furthermore, we used RT-PCR technique which had high specificity to COVID. Consequently, false positives were extremely rare in this research. However, there were three limitations to this study. The non-standardization of the definition and criteria for long Covid remained unclear. Therefore, the evidence was difficult to compare and represent true prevalence. Since the data was gathered through telephone interviews, physical and laboratory examinations were not provided for the investigation of long Covid. Inter-rater reliability might affect the consistency of results. Lastly, test-negative COVID infection and population were not considered for the control group in this study, ⁸ leading to increased noise for the true prevalence of long Covid studies.

This study was approved by the Human Research Ethics Committee of Thammasat University (Medicine), ethical approval number MTU-EC-PE-1-280/65 dated 12 January 2023. Verbal consent has been obtained from all participants via telephone that was approved by the ethics committee.