Risk for Reinfection After SARS-CoV-2: A Living, Rapid Review for American College of Physicians Practice Points on the Role of the Antibody Response in Conferring Immunity Following SARS-CoV-2 Infection

Abstract

Background: The strength and duration of immunity from infection with SARS-CoV-2 are important for public health planning and clinical practice.

Purpose: To synthesize evidence on protection against reinfection after SARS-CoV-2 infection.

Data sources: MEDLINE (Ovid), the World Health Organization global literature database, ClinicalTrials.gov, COVID19reviews.org, and reference lists.

Study selection: Longitudinal studies that compared the risk for reinfection after SARS-CoV-2 infection versus infection risk in individuals with no prior infection.

Data extraction: Two investigators sequentially extracted study data and rated quality.

Data synthesis: Across 18 eligible studies, reinfection risk ranged from 0% to 2.2%. In persons with recent SARS-CoV-2 infection compared with unvaccinated, previously uninfected individuals, 80% to 98% of symptomatic infections with wild-type or Alpha variants were prevented (high strength of evidence). In the meta-analysis, previous infection reduced risk for reinfection by 87% (95% CI, 84% to 90%), equaling 4.3 fewer infections per 100 persons in both the general population (risk difference, -0.043 [CI, -0.071 to -0.015]) and health care workers (risk difference, -0.043 [CI, -0.069 to -0.016]), and 26.6 fewer infections per 100 persons in care facilities (risk difference, -0.266 [CI, -0.449 to -0.083]). Protection remained above 80% for at least 7 months, but no study followed patients after the emergence of the Delta or Omicron variant. Results for the elderly were conflicting.

Limitation: Methods to ascertain and diagnose infections varied.

Conclusion: Before the emergence of the Delta and Omicron variants, persons with recent infection had strong protection against symptomatic reinfections for 7 months compared with unvaccinated, previously uninfected individuals. Protection in immunocompromised persons, racial and ethnic subgroups, and asymptomatic index case patients is unclear. The durability of protection in the setting of the Delta and Omicron variants is unknown.

Primary funding source: Agency for Healthcare Research and Quality. (PROSPERO: CRD42020207098).

Appendix Figure.. Evidence search and selection.

WHO = World Health Organization.

Figure 1.. Risk for reinfection from SARS-CoV-2: meta-analysis.

Positive indicates the group within a study where participants were polymerase chain reaction (PCR)–positive or seropositive at baseline. Negative indicates those within a study who were PCR-negative and/or seronegative at baseline. Estimates of relative risk (RR) ranged from 0.02 to 0.20; on the x-axis, 0.00 represents no effect, whereas 1.00 represents maximum protection. Study weighting and effect averaging were done using the empirical Bayes random-effects model. Continuity correction was used for counts of 0 (0.5 added to all counts). Studies were sorted alphabetically within categories by study author. Efficacy (1 − RR) can be interpreted as the proportion or percentage of infections that are prevented by the exposure. Median follow-up time was 8 mo (range, 4–13 mo). NA = not applicable; UK = United Kingdom; US = United States. * This plot shows estimates of protection, defined as 1 − RR.

Figure 2.. L'Abbé plot showing no indication of systematic deviation from the meta-effect, no outlying studies, and no study suggesting a qualitatively different effect size.

Point sizes are proportional to study precision.

Figure 3.. Scatter plots conveying the influence of various factors on the protective effect of prior SARS-CoV-2 infection.

The factors are infection proportion in the negative group, waiting period (in months), total follow-up (in months), median age of participants (in years), symptom status at baseline (any infections vs. symptomatic only), and cohort allocation criterion (serology, serology or PCR, and PCR). In all plots, the y-axis is efficacy (1 − relative risk), which can be interpreted as the proportion or percentage of infections that are prevented by the exposure. Points have been jittered slightly for visual display. PCR = polymerase chain reaction.