Fig 1.
Flowchart of study population.
Table 1.
Characteristics of proband children in study population, time-varying exposure status at end of follow-up period.
Fig 2.
Overall infection-related hospitalization sibling hazard ratios.
Adjusted for maternal age, parity, gestational age, birth weight, sex, season of birth, 5-minute Apgar score, birth mode, and birth year. Additional adjustments for SES (from Socio-Economic Index for Areas (SEIFA) data) and smoking during pregnancy. Smoking data available from 1997–2013. (A) Infection-related hospitalization sibling hazard ratios among probands whose sibling/s had infection-related hospitalizations compared to probands whose siblings/s did not have infection-related hospitalizations (reference group). (B) Infection-related hospitalization sibling hazard ratios excluding proband infection-related hospitalization admissions occurring within 30 days following a sibling infection-related hospitalization exposure.
Fig 3.
Infection-related hospitalization sibling hazard ratios by clinical infection groups.
Infection-related hospitalization sibling hazard ratios by type of clinical infection in proband (outcome) and sibling (exposure), compared to risk in probands whose sibling/s did not have infection-related hospitalizations (reference group). Adjusted for maternal age, parity, gestational age, birth weight, sex of the child, season of birth, 5-minute Apgar score, birth mode, birth year. Number of infections in sibling was analyzed as both dichotomous (Yes/No (reference)) and categorical (0 (reference), 1, 2+).
Fig 4.
Risk of 1st occurrence infection-related hospitalization in children <5 and 5–18 years old, by time-at-risk.
In the exposed group, ‘time-at-risk’ is the time from the first sibling infection-related hospitalization (occurring prior to the 1st proband infection-related hospitalization) to the 1st proband infection-related hospitalization or end of study, whichever occurred first. In the unexposed group, ‘time-at-risk’ begins with the start of follow-up. Follow-up times shorter than the ‘time-at-risk’ interval were not included in the interval risk calculations. ‘Cases’ refers to probands with infection-related hospitalizations; ‘Non cases’ refers to probands with no infection-related hospitalizations.
Fig 5.
Infection-related hospitalization sibling hazard ratios by family size and sibling characteristics.
Adjusted for maternal age, parity, gestational age, birth weight, sex of the child, season of birth, 5-minute Apgar score, birth mode, birth year. (A) Infection-related hospitalization sibling hazard ratios by family size and age of siblings. (B) Infection-related hospitalization sibling hazard ratios excluding proband infection-related hospitalization admissions occurring within 30 days following a sibling infection-related hospitalization exposure. (C) Infection-related hospitalization sibling hazard ratio by family size and distribution of infections among siblings. IRH = infection-related hospitalization.