Estimation of introduction and transmission rates of SARS-CoV-2 in a prospective household study

• Published in: PLoS computational biology Vol. 20; no. 1; p. e1011832
• Main Authors: van Boven, Michiel, van Dorp, Christiaan H, Westerhof, Ilse, Jaddoe, Vincent, Heuvelman, Valerie, Duijts, Liesbeth, Fourie, Elandri, Sluiter-Post, Judith, van Houten, Marlies A, Badoux, Paul, Euser, Sjoerd, Herpers, Bjorn, Eggink, Dirk, de Hoog, Marieke, Boom, Trisja, Wildenbeest, Joanne, Bont, Louis, Rozhnova, Ganna, Bonten, Marc J, Kretzschmar, Mirjam E, Bruijning-Verhagen, Patricia
• Format: Journal Article
• Language: English
• Published: United States Public Library of Science 01.01.2024; Public Library of Science (PLoS)
• Subjects: Adolescents; Adults; Analysis; Biology and Life Sciences; Children; Children & youth; Cohort analysis; COVID-19; Development and progression; Disease transmission; Diseases; Epidemic models; Epidemics; Estimates; Hazards; Health aspects; Households; Infections; Infectious diseases; Infectivity; Medical research; Medicine and Health Sciences; Medicine, Experimental; Netherlands; Patient admissions; People and Places; Population studies; Public health; Severe acute respiratory syndrome coronavirus 2; Stochasticity; Survival analysis; Teenagers; Vaccination; Viral diseases; Netherlands
• ISSN: 1553-7358; 1553-734X; 1553-7358
• DOI: 10.1371/journal.pcbi.1011832

Household studies provide an efficient means to study transmission of infectious diseases, enabling estimation of susceptibility and infectivity by person-type. A main inclusion criterion in such studies is usually the presence of an infected person. This precludes estimation of the hazards of pathogen introduction into the household. Here we estimate age- and time-dependent household introduction hazards together with within household transmission rates using data from a prospective household-based study in the Netherlands. A total of 307 households containing 1,209 persons were included from August 2020 until March 2021. Follow-up of households took place between August 2020 and August 2021 with maximal follow-up per household mostly limited to 161 days. Almost 1 out of 5 households (59/307) had evidence of an introduction of SARS-CoV-2. We estimate introduction hazards and within-household transmission rates in our study population with penalized splines and stochastic epidemic models, respectively. The estimated hazard of introduction of SARS-CoV-2 in the households was lower for children (0-12 years) than for adults (relative hazard: 0.62; 95%CrI: 0.34-1.0). Estimated introduction hazards peaked in mid October 2020, mid December 2020, and mid April 2021, preceding peaks in hospital admissions by 1-2 weeks. Best fitting transmission models included increased infectivity of children relative to adults and adolescents, such that the estimated child-to-child transmission probability (0.62; 95%CrI: 0.40-0.81) was considerably higher than the adult-to-adult transmission probability (0.12; 95%CrI: 0.057-0.19). Scenario analyses indicate that vaccination of adults can strongly reduce household infection attack rates and that adding adolescent vaccination offers limited added benefit.