Health Outcomes and Reinfection Among COVID-19 Survivors Four Years After Wuhan Hospital Dischar

Objectives To evaluate health outcomes and identify risk factors for reinfection and persistent symptoms among COVID-19 survivors 4 years after hospital discharge.

Outcome measures Self-reported symptom questionnaire, Chronic Obstructive Pulmonary Disease Assessment Test, Hospital Anxiety and Depression Scale and Checklist Individual Strength (CIS) fatigue subscale. Long covid was defined according to WHO criteria.

Results Median age was 58 years and 50.2% were male. Reinfection during December 2022–April 2023 occurred in 36.1%; 21 developed pneumonia and 14 required hospitalisation. At least 12 months after reinfection, 12.1% reported sequelae compared with 46.9% after the initial infection. At 4 years, 16.7% reported long covid symptoms, commonly fatigue, chest tightness, cough and dyspnoea. In multivariable analysis, risk factors for abnormal fatigue (CIS ≥27) included age (OR 1.020, 95% CI 1.007 to 1.034; p=0.003), reinfection (OR 2.393, 95% CI 1.708 to 3.352; p<0.001), severe disease (OR 1.553, 95% CI 1.088 to 2.218; p=0.015) and tumour (OR 3.420, 95% CI 1.177 to 9.936; p=0.024).

Conclusions At 4 years post discharge, symptom burden was lower than at earlier follow-up time points for most survivors. Reinfection and older age were associated with persistent symptoms.

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: https://creativecommons.org/licenses/by-nc/4.0/ .

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center's RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Longitudinal follow-up of a fixed cohort with standardised telephone interviews at 1, 2, 3 and 4 years after discharge.

Attrition over follow-up and differences between participants and non-participants may introduce selection bias.

Telephone-based self-reported outcomes without in-person assessment may lead to misclassification, and the hospitalised cohort may limit generalisability to non-hospitalised populations.

As of 26 May 2024, the global COVID-19 pandemic has resulted in approximately 775 million infections and over 7 million deaths, according to data from the WHO. 1 Although public health policies, vaccination programmes and antiviral therapies have helped reduce mortality rates from SARS-CoV-2 infection and many patients have recovered, 2 increasing evidence suggests that numerous patients continue to experience a variety of long-term symptoms after recovery. 3 These symptoms include cough, fatigue, shortness of breath, memory decline, anxiety and muscle pain, commonly referred to as ‘long COVID’. The WHO defines long covid as the presence of symptoms that appear 3 months after the onset of COVID-19 and persist for at least 2 months, which cannot be explained by an alternative diagnosis. 4 These persistent symptoms not only affect patients’ physical health but also have significant negative impacts on their psychological well-being and quality of life, further limiting their social participation and work capacity. 5

Previous studies have shown that while the symptom burden in COVID-19 survivors gradually decreases within 3 years after hospital discharge, notable health impacts can still be observed in the third year. 6 Patients with severe illness during hospitalisation, particularly those admitted to the intensive care unit (ICU), have a higher risk of persistent symptoms. 7 8 The work by Ziyad et al also indicates an increased risk of a range of neurological, gastrointestinal and endocrine system sequelae following the acute phase of COVID-19. 9–11 Furthermore, due to changes in China’s COVID-19 prevention policies at the end of 2022, the country experienced two major Omicron variant waves in 2023, leading to a substantial number of reinfections. 12 13 Studies have shown that survivors with long covid have a higher proportion of reinfection compared with those without long covid. 14 15

Therefore, long-term follow-up of COVID-19 patients is crucial. The high hospitalisation rate among patients infected with the original strain of COVID-19 provides a robust study population for investigating long covid. Our team has previously conducted follow-up studies on long covid at 1, 2 and 3 years after discharge. 6–8 This study aims to elucidate the 4-year health outcomes of COVID-19 survivors and assess the impact of reinfection and long covid on this patient population.

This is a longitudinal cohort study targeting COVID-19 survivors discharged from Huoshenshan Hospital and Taikang Tongji Hospital in Wuhan between 12 February 2020 and 10 April 2020. All adult patients with laboratory-confirmed COVID-19 were screened for eligibility. Exclusion criteria included: (1) those who declined to participate, (2) those who could not be contacted and (3) those who died before follow-up. Follow-ups were conducted at 1 year, 2 years and 3 years after hospital discharge. The fourth-year follow-up was conducted from 8 March 2024 to 15 April 2024, and data analysis was performed from 20 April 2024 to 1 May 2024.

All patients were contacted in the order of their discharge dates documented in their medical records and were interviewed via telephone by trained physicians. The questionnaires used included: self-reported symptom questionnaire, Checklist Individual Strength (CIS) fatigue subscale, Chronic Obstructive Pulmonary Disease (COPD) Assessment Test (CAT), Activity of Daily Living Scale and Hospital Anxiety and Depression Scale (HADS; including the HADS-Anxiety (HADS-A) and the HADS-Depression (HADS-D)). The CAT was initially designed to assess symptom burden in patients with COPD but is also applicable for evaluating the symptom burden in COVID-19 survivors. 16 A CAT score ≥10 was used to indicate a medium-to-high symptom impact, based on the threshold recommended by the Global Initiative for Chronic Obstructive Lung Disease (GOLD) guidelines. 17 HADS was used to measure mood symptoms of anxiety and depression. 18 Each subscale consists of seven questions with a 4-point Likert scale (0–3). Scores of at least 8 indicate the presence of symptoms of anxiety or depression. CIS was used to assess fatigue levels over the past 2 weeks. The scale comprises eight items scored on a 7-point Likert scale, with a total score ranging from 8 to 56, and a cut-off value of ≥27 is used to indicate abnormal fatigue. 19 Reinfection was ascertained for the prespecified observation window (December 2022 to April 2023). The criteria for identifying COVID-19 reinfection in our study were based on telephone follow-ups with patients. During these follow-ups, patients were first asked whether they believed they had been reinfected with COVID-19 during this period. If the response was affirmative, we further inquired whether they had undergone COVID-19 testing, either via nucleic acid testing (PCR-based) or antigen testing (colloidal gold-based). Only patients who reported undergoing testing and receiving a positive result were classified as having experienced COVID-19 reinfection. The self-reported symptom questionnaire included symptoms such as sweating, chest tightness, myalgia, palpitations, cough, chest pain, dizziness, expectoration, dyspnoea, headache, oedema, taste change, smell reduction and sore throat. 20 All the questionnaires we used are shown in online supplemental eTables 1-4 .

Clinical data for patients during hospitalisation were retrieved from electronic medical records, including demographic characteristics (self-reported race, age, sex and cigarette smoking) and clinical characteristics (self-reported comorbidities, symptoms and chest images). Long covid was defined by WHO. 4 All data were double entered and validated using EpiData software V.3.1 (EpiData Association).

Continuous variables were presented as median (Q1–Q3) and compared using the Mann-Whitney U test. Categorical variables were presented as numbers and percentages and compared using the Pearson χ² test or Fisher’s exact test, as appropriate. For repeated binary outcomes measured in the same individuals across the 2-year, 3-year and 4-year follow-ups, overall comparisons across time points were performed using Cochran’s Q test. When the overall test was statistically significant, pairwise comparisons between time points were conducted using McNemar’s tests with Holm adjustment for multiple comparisons. To identify risk factors associated with symptom burden and reinfection, univariable logistic regression analyses were initially performed to screen for potential risk factors. Age and sex were included a priori in the multivariable models because of their clinical relevance and were retained regardless of statistical significance. Additional candidate variables with univariable p<0.10 were then entered, and backward elimination was used to retain variables with p<0.05. All tests were two-sided, and p<0.05 was considered statistically significant. Data were analysed using SPSS statistical software V.26.0 (IBM SPSS Statistics) and R statistical software V.4.1.1 (R Project for Statistical Computing).

No a priori sample size calculation was performed because this was a longitudinal follow-up of a fixed cohort of patients discharged after COVID-19 hospitalisation; we attempted to include all eligible survivors who could be contacted for the 4-year interview.

Among the 3988 patients discharged alive after the index COVID-19 hospitalisation, 1076 (26.9%) completed the 4-year follow-up interview and were included in the final analysis ( figure 1 ). As shown in figure 1 , follow-up participation decreased over time (2433 at 1 year, 1864 at 2 years, 1594 at 3 years and 1076 at 4 years), primarily due to refusal or inability to be contacted, and a small number of participants died after discharge before a scheduled follow-up interview ( figure 1 ). The median age of included participants was 58 years (Q1–Q3 47.0–66.0 years), and 541 (50.2%) were male. The median duration of hospital stay was 14 days (Q1–Q3 9–20 days). During hospitalisation, 795 (73.9%) received oxygen therapy, 7 (0.7%) required mechanical ventilation and 17 (1.6%) were admitted to the ICU ( table 1 ). Additionally, 388 (36.1%) experienced reinfection, among whom 21 developed pneumonia and 14 required hospitalisation ( table 2 ). In addition, 27 participants reported suspected reinfection without PCR/antigen confirmation and were not classified as reinfected in the primary analysis. Overall, 938 (87.2%) participants reported having received COVID-19 vaccination by the time of the 4-year follow-up interview; among reinfected participants (n=388), 334 (86.1%) reported COVID-19 vaccination.