Objectives To determine whether time elapsed since the second injection of the Pfizer-BioNTech BNT162b2 mRNA vaccine was significantly associated with the risk of covid-19 infection after vaccination in people who received two vaccine injections.

Participants Adults aged ≥18 years who had received a reverse transcription polymerase chain reaction (RT-PCR) test between 15 May 2021 and 17 September 2021, at least three weeks after their second vaccine injection, had not received a third vaccine injection, and had no history of covid-19 infection.

Results 83 057 adults received an RT-PCR test for SARS-CoV-2 during the study period and 9.6% had a positive result.

Time elapsed since the vaccine injection was significantly longer in individuals who tested positive (P<0.001).

Conclusions In this large population of adults tested for SARS-CoV-2 by RT-PCR after two doses of mRNA BNT162b2 vaccine, a gradual increase in the risk of infection was seen for individuals who received their second vaccine dose after at least 90 days.

Individuals who are immune have some protection against reinfection and a reduced risk of severe clinical consequences.1 Recovering seropositive adult individuals have about 90% protection from SARS-CoV-2 reinfection after 150 days,2 and distinguishing reinfection from the effects of the initial infection is challenging before 90 days.3 In contrast, vaccination has been reported to be 50-95% effective at various time points.456 Nevertheless, the memory B cell humoral response and spike specific CD4+ cellular immune responses to SARS-CoV-2 are predictably diminishing over time.78 Therefore, concerns have been raised about a possible reduction in SARS-CoV-2 immune protection in vaccinated populations and in those who have recovered from covid-19 disease.9.

A large proportion of the population were quickly immunised, partly because of the centralised health services and the decision to use only one vaccine, achieving early control over the spread of the virus.101112 Up to 26 July 2021, more than 5.2 million Israelis were fully vaccinated with two doses of the Pfizer-BioNTech vaccine.13 Since June 2021, however, a resurgence of individuals with SARS-CoV-2 has been seen, which could be at least partially because of decreasing levels of anti-SARS-CoV-2 antibodies in vaccinated people.14 Here, we describe the results of a large scale study measuring the association between time elapsed since administration of the second dose of the Pfizer-BioNTech BNT162b2 vaccine and the risk of covid-19 disease.

We conducted a test negative design study in adult members of Leumit Health Services (aged ≥18) who underwent a SARS-CoV-2 reverse transcription polymerase chain reaction (RT-PCR) test between 15 May 2021 and 17 September 2021, after receiving two vaccine injections.

The test negative design is similar to a nested case-control design, with individuals tested for the presence of a pathogen of interest serving as the nesting cohort.1516 Individuals who had been diagnosed as having covid-19 before the study period, or who had received a third (booster) vaccine dose before the RT-PCR test, were excluded.

Time between the second vaccine injection and the RT-PCR test was divided into 30 day intervals, after the initial 90 days (reference): 90-119 days, 120-149 days, 150-179 days, and ≥180 days.

Time elapsed between the second dose of the vaccine and the RT-PCR test was categorised by intervals of 30 days after the initial 90 days (reference).

To account for variations in the infection rate and circulating strains throughout the study period, we defined a matched cohort, with a predetermined ratio of 1:5 positive (individuals who tested positive) to negative (controls), for each combination of test week, age group, and demographic group.

Figure 1 shows the flow diagram used for selection of the cohort: the cohort was based on 83 057 adults who received a RT-PCR test for SARS-CoV-2 during the study period, after two (and only two) vaccine injections, at least three weeks after their second vaccine injection, and with no evidence of previous covid-19 infection.

The median time between the second dose of the vaccine and the RT-PCR test was 164 days (interquartile range 138-185 days).

In each age group, we found a significant difference in elapsed time (in days) since the date of the second dose of the vaccine between individuals who tested positive and those who tested negative (P<0.001, standardised mean difference >0.6).

The rate of positive results increased with time elapsed since the second vaccine dose, and the increase was significant (P<0.01) in all age groups after ≥180 days.

Odds ratio for risk of SARS-CoV-2 according to time elapsed since second SARS-CoV-2 vaccine in the pre-matched cohort.

Compared with the initial 90 days after the vaccine, we found an increased risk of infection with time elapsed since vaccination: adjusted odds ratio 2.37 (95% confidence interval 1.67 to 3.36) after 90-119 days; 2.66 (1.94 to 3.66) after 120-149 days; 2.82 (2.07 to 3.84) after 150-179 days; and 2.82 (2.07 to 3.85) after ≥180 days (P<0.001 for all).

In this large population of individuals who received two doses of the Pfizer-BioNTech BNT162b2 vaccine, we found a significantly (P<0.001) higher risk of SARS-CoV-2 infection with time elapsed since the second vaccine dose, after the initial 90 days (reference category).

In the Leumit Health Services organisation, RT-PCR tests are provided with no limit on the number of requests from patients or physicians, allowing a high detection rate of covid-19 infection before and during the study period.

In initial analyses, we found that among individuals who had evidence of past infection and who received a RT-PCR test during the same study period, the infection rate was 4.7% for unvaccinated individuals and 3.8% for vaccinated individuals (v 9.6% infection rate among vaccinated individuals with no evidence of past infection).

In particular, we included vaccinated individuals who chose to request an RT-PCR test for SARS-CoV-2 during the study period.

We used a multivariable conditional logistic regression model adjusted for potential risk factors for covid-19, including age, sex, socioeconomic status, and comorbid conditions, to assess the effect of time elapsed since vaccination on the risk of infection.

To quantify the increase in risk over time, we used the initial 90 days after the second vaccination as the reference and calculated the odds ratio of infection in subsequent intervals of 30 days.

Lastly, the study was designed to estimate the effect of the time elapsed since vaccination on the risk of a positive SARS-CoV-2 test.

In this retrospective large cohort study, performed in individuals who received two doses of the BNT162b2 mRNA vaccine, protection seemed to decrease over time, and the risk of breakthrough infection by SARS-CoV-2 increased progressively compared with the protection provided during the initial 90 days.

The efficacy of two injections of mRNA vaccines could reduce over time and statistical analysis of the association between time elapsed since the vaccine and risk of infection could provide important clues about the need for a third injection and its preferred timing.

The risk of SARS-CoV-2 infection in adults who received two vaccine injections increased with time elapsed since vaccination compared with the reference (individuals vaccinated in the last 90 days).