A Multi-Center Retrospective Cohort Analysis of COVID-19 Vaccine Effectiveness in Patients with Mild or Asymptomatic Infection During the Omicron Outbreak in Guangzhou

Introduction

Since SARS-CoV-2 emerged in December 2019, causing COVID-19, the virus has spread worldwide, infecting over 700 million people and resulting in more than 6 million deaths by February 2023 [1]. In response, vaccines have been rapidly developed and deployed, with nearly 70% of the global population receiving at least one dose and over 13 billion doses administered overall [2]. Evidence consistently shows that vaccination significantly reduces the risk of severe illness and mortality [3–6]. Despite this, vaccines are less effective at preventing mild or asymptomatic infections and do not entirely halt virus transmission [7]. Consequently, unvaccinated populations remain key drivers of viral spread, particularly in regions with high vaccination coverage, highlighting the need for strategies that protect vulnerable groups and reduce overall infection rates [8].

The continuing evolution of SARS-CoV-2, including variants with higher transmissibility and partial immune evasion [9–12], has complicated pandemic control efforts. Variants such as B.1.351 and P.1 have raised concerns about the reduced effectiveness of vaccines and other interventions. Studies suggest that BNT162b2 may offer weaker protection against these strains [12]. Additionally, immunity tends to wane over time, emphasizing the potential benefits of heterologous booster strategies, as recommended by the World Health Organization (WHO) [13]. Vaccine effectiveness is further influenced by co-infections, concurrent medications [14], and demographic factors like age, sex, and socioeconomic status [14]. Vaccination may also affect isolation dynamics [15, 16]. Therefore, population-based studies are critical for accurately assessing vaccine effectiveness and informing strategies to overcome vaccine hesitancy and guide public health decisions [17].

Asymptomatic infections, which contribute substantially to SARS-CoV-2 transmission [17], have become an essential focus of study. Vaccines have demonstrated efficacy in reducing asymptomatic cases [18], suggesting that vaccination could shorten both isolation periods and viral shedding in these patients, thereby limiting the onward transmission of the virus. To examine this hypothesis, we conducted a multi-center study evaluating the effect of COVID-19 vaccination on isolation duration and viral shedding, aiming to inform optimal vaccine strategies to curb community spread.

Materials and Methods

Patient selection

Adults aged 18 years or older with first-time COVID-19 infection were enrolled from Fangcang isolation centers in Pazhou and Yongning between November and December 2022. Only patients with asymptomatic or mild disease were included; those with severe or critical illness were excluded. Individuals with underlying health conditions or other specific circumstances were also omitted from the study (Table E1). This investigation followed the strengthening of the reporting of observational studies in epidemiology (STROBE) guidelines (Online Supplement 2) [19] and received approval from the Ethics Committee of The First Affiliated Hospital of Guangzhou Medical University (approval number ES-2023-116-01).

Study design

We collected comprehensive demographic data, including age, sex, marital status, ethnicity, occupation, and province of residence. Admission information (such as hospital stay duration, date when the health code changed to yellow, dates of nucleic acid tests (NATs), and laboratory indicators including NAT outcomes and cycle threshold (Ct) values) was also recorded. Vaccination details, including the type of vaccine and the administration date, were documented.

Patient discharge from the Fangcang isolation centers followed these criteria: (1) body temperature remained normal for at least three consecutive days; (2) notable improvement in respiratory symptoms; (3) clear resolution of acute infiltrative lesions on pulmonary imaging; and (4) completion of seven days of centralized medical observation, with nasal and pharyngeal swabs collected for NAT on days 6 and 7 (minimum 24-hour interval between samples). Patients were eligible for discharge if the Ct values for both the nucleocapsid (N) gene and ORF1ab gene in both NATs were ≥ 35 (via fluorescence quantitative PCR with a detection threshold of 40) or if tests returned negative (Ct < 35). Patients who did not meet these criteria remained in isolation until all requirements were satisfied.

The discharge rate was defined as the proportion of patients meeting the criteria for release from isolation, while the isolation rate was calculated as 1 minus the discharge rate. Viral shedding duration was defined as the interval from the first positive NAT to the first day of continuous negative results. Full vaccination referred to completion of the primary vaccine series, whereas a booster indicated any additional doses received after achieving full vaccination [16]. The negative rate was defined as the proportion of patients achieving continuous negative test results, with the positive rate calculated as 1 minus the negative rate.

Statistical analysis

Continuous variables following a normal distribution were reported as mean ± standard deviation (SD), and those not normally distributed were expressed as median with interquartile range (IQR). Categorical variables were presented as counts and percentages. Group differences were assessed using analysis of variance, Kruskal–Wallis test, Chi-square test, or Fisher’s exact test, depending on data characteristics.

The effects of vaccination on isolation and viral shedding duration were examined using multivariable Cox regression models. Differences in negative and discharge rates among vaccine types and dosage groups were evaluated through Kaplan–Meier curves and multivariable Cox regression analyses. For missing Ct values, imputation was performed using the median Ct value of the respective positive or negative group from the same day. To address sample size disparities across vaccine regimens, propensity score matching was applied in sensitivity analyses using nearest-neighbor matching with a 1:1 ratio and a caliper of 0.2 SD of the propensity score probit. A P-value < 0.05 was considered statistically significant. All analyses were conducted using R software (version 4.1.2, R Project for Statistical Computing, Vienna, Austria).

Results

Patient recruitment and baseline characteristics

A total of 6,560 COVID-19 patients were included, comprising 3,584 from Pazhou and 2,976 from Yongning. Patient baseline characteristics are summarized in Table 1. Among these, 69 individuals received heterologous vaccination combining inactivated and recombinant protein vaccines. Overall, 47.9% of participants were female, and the median age was 39 years. The majority (80.5%) were married. At admission, the median Ct values for the nucleocapsid (N) gene and ORF1ab gene were 31.6 and 29.0, respectively.

¹Includes only vaccinated cases (n = 6173).

In this study population, most participants (90.6%) had received inactivated COVID-19 vaccines, whereas 3.06% were administered recombinant SARS-CoV-2 spike protein subunit vaccines, and 0.85% received adenovirus-based vaccines. Among the 6,173 vaccinated individuals, nearly three-quarters (71.9%) had obtained a booster dose, with 67.3% having received their most recent vaccine within the year preceding infection.

Isolation Duration and Discharge Outcomes in Vaccinated Versus Unvaccinated Patients

Analysis of isolation trends revealed that 50% of vaccinated patients remained in isolation until day 9, while an equivalent proportion of unvaccinated patients were discharged by day 10 (P < .001; Figure 1(A)). Table 2) summarizes the baseline characteristics for both groups. After adjusting for potential confounders including sex, age, marital status, and geographic location, vaccination was associated with a significantly faster discharge, with vaccinated individuals having a 21.1% higher likelihood of being released from isolation within 14 days compared to unvaccinated individuals (hazard ratio [HR] = 1.211; 95% confidence interval [CI]: 1.084–1.351; P < .001) (Figure 1(B)).

Figure 1. Impact of vaccination on 14-day isolation outcomes: (A) Isolation trends among vaccinated versus unvaccinated patients. The red line indicates the proportion of vaccinated individuals remaining in isolation, while the blue line shows the corresponding rate for unvaccinated individuals. (B) The likelihood of discharge from the isolation center within 14 days for vaccinated and unvaccinated groups is also shown.

Viral Clearance and Negative Test Trends in Vaccinated Versus Unvaccinated Patients

Analysis of viral shedding revealed that half of the vaccinated participants had converted to negative COVID-19 test results by approximately day 7.5. In contrast, unvaccinated individuals reached the same milestone slightly later, around day 8 (P < .001; Figure 2(A)). When adjusting for confounding factors such as age, sex, marital status, and geographic region, vaccination was associated with a faster rate of viral clearance, with vaccinated patients exhibiting a 23.9% higher probability of achieving negative tests within 14 days compared to those unvaccinated (hazard ratio [HR] = 1.239; 95% confidence interval [CI]: 1.113–1.378; P < .001) (Figure 2(B)).

Figure 2. Effect of vaccination on 14-Day positivity outcomes: (A) Trends in positivity rates during isolation among patients receiving different vaccine types and doses. The red line depicts the proportion of vaccinated individuals remaining positive. In contrast, the blue line shows the corresponding trend for unvaccinated patients. (B) Likelihood of achieving negative test results within 14 days across vaccine types and dosage groups.

Comparison of Isolation Duration and Discharge Between Heterologous and Inactivated Vaccine Recipients

Among all vaccinated participants, half were discharged by approximately day 8, whereas unvaccinated individuals reached the same milestone around day 9. Importantly, patients who received heterologous vaccination displayed a higher probability of discharge within 14 days compared with those receiving only inactivated vaccines or other regimens (P = 0.029; Figure 3(A)). Baseline characteristics for patients across different vaccination strategies are summarized in Table 3.

¹Includes only vaccinated cases (n = 6173).

To explore whether the type of vaccine influenced isolation outcomes—specifically the likelihood of discharge from the isolation center within 14 days—a detailed analysis was performed, adjusting for potential confounders including sex, age, marital status, province of residence, and number of vaccine doses. The findings indicated no statistically significant difference in 14-day isolation rates between patients receiving heterologous vaccination and those who had only inactivated vaccines (hazard ratio [HR] = 1.226; 95% confidence interval [CI]: 0.963–1.562; P = 0.099) (Figure 3(B)). These results suggest that, within the 14-day observation window, both vaccination strategies yielded comparable outcomes in terms of isolation.

Comparison of Viral Shedding Duration and Positive Test Rates Between Heterologous and Inactivated Vaccine Recipients

Between days 6 and 8, approximately 50% of patients tested positive for COVID-19. When comparing viral shedding durations, patients who received either inactivated vaccines alone or heterologous vaccination exhibited shorter viral shedding periods compared with other vaccination groups (P = 0.015). Notably, individuals who received heterologous vaccination cleared the virus more quickly than those vaccinated solely with inactivated vaccines (P = 0.011) (Figure 4(A)).