The COVID-19 pandemic highlighted the critical need to understand the effectiveness of vaccines in preventing both infection and transmission of SARS-CoV-2. Vaccine effectiveness against infection (VE-infection) refers to the reduction in susceptibility to infection, while vaccine effectiveness against infectiousness (VE-infectiousness) describes the reduction in the ability to transmit the virus after infection. Determining VE-infectiousness is challenging, but household studies provide a valuable approach. Policy decisions regarding vaccination prioritization depend on understanding both VE-infection and VE-infectiousness. If vaccines primarily prevent infection, then vaccinating only vulnerable populations might suffice. However, if vaccines also reduce transmission, then wider vaccination strategies become more beneficial. Previous research has demonstrated that COVID-19 vaccines significantly reduce the risk of SARS-CoV-2 infection, although this effect varies by variant and wanes over time. The emergence of highly transmissible variants like Delta and Omicron further complicates the picture. This study investigates the VE-infection and VE-infectiousness of COVID-19 vaccines against Delta and Omicron variants in the Netherlands during the latter half of 2021 and the first half of 2022, aiming to provide data relevant for informing public health strategies.

Extensive research has explored the effectiveness of COVID-19 vaccines against infection. Studies consistently show a substantial reduction in SARS-CoV-2 infection risk with vaccination, although the level of protection differs across variants, and vaccine efficacy decreases over time. The Delta variant, with its increased transmissibility, posed a significant challenge, and the subsequent emergence of Omicron, even more transmissible, further emphasized the dynamic nature of the virus. Studies have investigated the infectiousness of these variants, revealing that Omicron's infectiousness surpasses that of Delta, regardless of vaccination status. This study builds on this existing knowledge by specifically focusing on household transmission, which provides a unique setting to evaluate both VE-infection and VE-infectiousness simultaneously.

This study utilizes data from the VASCO (Vaccine Study COVID-19) prospective cohort study, encompassing approximately 45,000 Dutch adults aged 18-85. Participants were enrolled between May 3, 2021, and December 15, 2021, and followed for five years. Monthly questionnaires tracked vaccination status and self-reported positive SARS-CoV-2 tests. Starting in August 2021, participants who reported a positive test were asked to complete an additional questionnaire about their household members one month later, gathering information on their age, COVID-19 testing, test results, and vaccination status. Households were included if the VASCO participant was the first to test positive (index case), excluding households with co-primary cases or those where members did not provide consent. Household members who didn't undergo SARS-CoV-2 testing within the specified window (2 weeks before to 4 weeks after the index case's positive test) were excluded. Vaccination status was determined based on self-reported data and, where available, the national vaccination register. Individuals were categorized as unvaccinated, having received a primary series, or a booster. Generalized estimating equation (GEE) logistic regression models were used to estimate VE-infection and VE-infectiousness, adjusting for age, vaccination status, calendar week, and household size. Sensitivity analyses were conducted to assess the impact of time since vaccination, prior infection, and age on the estimates. The study period covered both the Delta and Omicron dominant periods in the Netherlands, allowing for analysis of vaccine effectiveness against both variants.

A total of 3399 questionnaires concerning 4105 household members were included in the analysis. During the Delta-dominant period, the adjusted VE-infection for primary vaccination was 47% (95% CI, −27% to 78%), while the adjusted VE-infectiousness was 70% (95% CI, 28% to 87%) for primary series and 93% (95% CI, 56% to 99%) for booster vaccination. However, the sample size for booster vaccination was limited in the Delta period. The Omicron period showed a different picture. VE-infection was negative for both primary vaccination (−36%, 95% CI, −87% to 1%) and the first booster (−28%, 95% CI, −77% to 7%). VE-infectiousness remained positive, at 45% (95% CI, −14% to 74%) for the primary series and 64% (95% CI, 31% to 82%) for the first booster. Sensitivity analyses revealed that including index cases with prior infection resulted in lower VE-infectiousness estimates. Additionally, VE-infection estimates were lower after three months since vaccination compared to the first three months. The study observed a similar secondary attack rate (SAR) during both Delta and Omicron periods (around 40%), indicating that the proportion of household members testing positive was not significantly altered.

This study demonstrates the effectiveness of COVID-19 vaccination against SARS-CoV-2 Delta and Omicron infectiousness within household settings, particularly for index cases without prior infection. The findings regarding VE-infectiousness for Delta align with some previous research, while the results for Omicron demonstrate a lower but still positive effect compared to the negative VE-infection for Omicron. The strikingly low and negative VE-infection estimates for Omicron warrant careful consideration. Several factors could contribute to this, including a lack of adjustment for prior infections in household members, the potential for increased transmission risk in the close-contact household environment, the high proportion of symptomatic infections in the index cases, and waning vaccine immunity over time. Sensitivity analyses highlight that these factors are crucial to understand the findings. The study's strength lies in its relatively large sample size and its ability to assess vaccine effectiveness during the later Omicron waves due to free self-testing availability. The findings support vaccination, especially among individuals in close contact with vulnerable populations, to reduce both infection risk and onward transmission.

This study provides valuable insights into the effectiveness of COVID-19 vaccination against SARS-CoV-2 Delta and Omicron infection and transmission within households. While vaccination showed effectiveness against infectiousness of both variants, the results for VE-infection for Omicron were considerably lower than expected. This underscores the importance of booster doses to maintain protection and to consider the various factors influencing VE estimates. Further research should focus on understanding the dynamics of waning vaccine efficacy, transmission rates in households, the impact of prior infections, and testing behavior to refine our understanding of vaccine effectiveness in the face of emerging variants.

This study has several limitations. Differences in testing behavior between vaccinated and unvaccinated individuals may influence the estimates. The assumption that all infections within 2-14 days of the index case resulted from the index case might lead to an underestimation of VE-infectiousness. The lack of data on the behavioral responses of household members after the index case's positive test could also affect the interpretation of VE-infectiousness. The reliance on self-reported data, except for index case vaccination status, introduces potential recall bias. However, the short time interval between infection and questionnaire completion might minimize this bias.