Table 1: Antibiotic Exposure and Risk of Eczema in Childhood

Ref. Population Time of Antibiotic Prescription(s) Method Antibiotic Use Assessed by Number in Study Findings/ OR Linked to Eczema 95% CI Comments
McKeever et al.
2002 [27]
Children in the
English midlands, under 11 years
1st year of life Retrospective
cohort.
GP records 21,129,
72.3% of available records
(aHR) 1.22 Yes 1.12-1.34 Inconsistent
dose-response
relationship.
No data on
confounding variables
Von Mutius
et al. 1999 [40]
Dresden, children,
aged 5-11
1st year 1-2 courses Survey, clinical
examination,
venous bloods
for IgE antibodies
Retrospective
questionnaires to
parents
12 601 1.26 Yes 1.05-1.52
Floistrup et al.
2006 [34]
Steiner school
children and a
comparator group, age 5-13
1st year Survey questionnaires to parents
Blood samples for
IgE estimation
Retrospective questionnaires to parents. 6733 1.61 Yes 1.21-2.15
Blood samples for IgE estimation
Cohet et al. 2005
[39]
New Zealand children age 6-7, 1584 with previous notifiable infection plus a general population group 1st year Survey. Questionnaires mailed to parents Retrospective questionnaires to parents. 4123 1.4 Yes 1.21-1.62 Eczema not linked to infections
Kummeling et al. 2007 [47] Netherlands, children up to 2 years 1st 6 months, including administration to breastfeeding mothers Prospective cohort Retrospective questionnaires to parents. 2764 0.94 No 0.75-1.18
Faroogqi and Hopkin 1998 [38] Children up to 12 years in a family doctor practice in Oxfordshire, UK 1st 2 years Retrospective cohort using. GP & public health records GP records 1934 2.04 Yes 1.53-2.72 Stronger association with broad spectrum antibiotic
Droste et al. 2000 [41] Belgium, children age 7-8 years 1st year Survey questionnaires to parents. Skin prick tests Retrospective questionnaires to parents. 1206 1.3 95% CI includes 1,uncertain 1.0-1.80
Sariachvili et al. 2007 [44] Belgium. Infants age 1 year 1st year Prospective cohort. Parents´ reports of eczema and IgE antibody measurements Retrospective reports of mothers at 3, 6 and 12 months 976 1.1 no 0.8-1.5
Celedon et al. 2002 [52] Boston USA, at age 5 years 1st year Prospective cohort. Parents´ reports of illness Telephone questionnaires to parents, every 2 months for 2 years 498 1.1 no 0.4-3.1 1 parent with a history of asthma or allergy. No association with allergic rhinitis
Jedrychowski et al. 2006 [28] Krakow, age 1 year Pregnancy Prospective cohort Questionnaires to mothers every 3 months 102 2.3 No 0.91-5.80
McKeever et al. 2002 [37] Children in the English midlands, age 11 years Pregnancy Retrospective cohort Database of GP records 24 690 (aHR) 1.19 Yes 1.09-1.31 Risk increases with more courses. Similar association with hayfever
Sariachvili et al. 2007 [44] Belgium. Infants age 1 Pregnancy Prospective cohort. Parents´ reports of eczema Retrospective reports of mothers at 5 months´ gestation & 3 months of life 976 1.8 Yes 1.2-2.7

Papers are ordered by sample size.