Background Helicobacter pylori infection causes chronic digestive diseases
with a major public-health impact, yet the design of prevention measures is
hampered by limited knowledge of transmission pathways. We studied the eff
ect of family composition on H pylori prevalence among rural Colombian chil
dren aged 2-9 years.
Methods 684 children were screened for H pylori with the UC-urea breath tes
t, For each child, birth order, birth spacing, number of 2-9-year-old sibli
ngs, and number of H-pylori-positive 2-9-year-old siblings was recorded. Od
ds ratios were estimated by logistic regression, controlling for hygiene-re
lated exposures, socioeconomic indicators, and the number of children in th
e household.
Findings The odds of infection increased with the number of 2-9-year-old si
blings in the household (odds ratios 1.4, 2.3, 2.6, and 4.3 for one, two, t
hree, and four to five siblings, respectively). Compared with first-born ch
ildren, odds ratios for children born second and third to ninth were 1.8 (9
5% CI 1.0-3.3) and 2.2(1.0-4.3), respectively. Compared with children born
10 or more years after the next older household member, those born within 4
years were 4.1 times (CI 2.0-8.6) more likely to be infected; the age gap
to the next younger household member displayed a weaker effect. The number
of H-pylori-positive 2-9-year-old siblings had particularly strong effect g
radient (1.5, 3.2, 5.6, and 7.1, for one, two, three, and four positive sib
lings, respectively).
Interpretation Among rural Andean children younger than 10 years, H pylori
infection seems to be transmitted most readily among siblings who are close
in age, and most frequently from older siblings to younger ones.