Early life determinants of wheeze and allergic disease: a longitudinal study in an Ethiopian birth cohort

Amberbir, Alemayehu (2012) Early life determinants of wheeze and allergic disease: a longitudinal study in an Ethiopian birth cohort. PhD thesis, University of Nottingham.

PDF - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (29MB) | Preview



The hypothesis that paracetamol may increase the risk of asthma and other allergic disease has gained consistent support from epidemiological studies, but evidence from longitudinal cohort studies, particularly those looking at the timing and dose of exposure are lacking. Epidemiological studies have also reported an inverse relation between gastro-intestinal infections including Helicobacter pylori, commensal bacteria and geohelminths and asthma and allergic disease, however, data from longitudinal birth cohort study are scarce. This thesis has therefore investigated the effects of paracetamol, H. pylori and other gastro-intestinal infections on the incidence and prevalence of allergic diseases and sensitization in a low-income birth cohort in which confounding by social advantage and other medical interventions is unlikely to play a role.


In 2005/6 a population based cohort of 1065 pregnant women from Butajira, Ethiopia was established, to whom 1006 live singleton babies were born, and these children have been followed-up from birth to age five. At ages one, three and five, the International Study of Asthma and Allergies in Children (ISAAC) questionnaires were administered to the mothers to obtain data on wheeze, eczema and rhinitis. Allergen skin tests to Dermatophagoides pteronyssinus and cockroach were performed at ages three and five. Data on child's use of paracetamol, and various early life putative risk factors, including levels of Der p 1 and Bla g 1 allergen in the child's bedding and symptoms of respiratory tract infections were also measured. Stool samples were collected at ages three and five for analysis of H. pylori antigen using a rapid test (Medimar immunocard), as well as for geohelminths (at ages one, three and five) and selected commensal bacteria (at age three). Multivariate logistic regression was used to determine the independent effects of various markers of paracetamol use on the incidence of each outcome between age one and five, as well as on prevalence at age five. Similar analyses were also carried out to determine the independent effects of H. pylori, geohelminths and commensals on the incidence and prevalence of each outcome.


Effects of paracetamol

Of the 1006 children in the cohort at birth, 863 children were successfully followed up at age five (94% of surviving mother-child dyads). Wheeze and eczema incidence between the ages of one and five were reported in 5.9% (40/676) and 5.8% (39/700) of children respectively, and rhinitis and sensitization incidence between ages three and five were found in 3.9% (31/798) and 2.0% (15/766) of children respectively. Paracetamol use in the first three years of life was common, with 18% reported use at age one but not three, 23% at age three but not one and 21% at both time points. Use in the first year of life was significantly associated with a dose-dependent increased risk of incident wheeze between ages one and three (fully adjusted ORs, 95% CI, 1.77; 0.96, 3.26 for 1-3 tablets and 6.78; 1.89, 24.39 for ..?. 4 tablets in past month versus never), but not eczema. The risk of incident wheeze, eczema, rhinitis and sensitization between ages three and five was increased in those exposed, significantly so for incident eczema (p=0.02) and borderline significant for rhinitis (p=0.07), with fully adjusted odds ratios (ORs), including for symptoms of respiratory tract infections, for persistent exposure (ages one and three) versus never of 3.82 (95% CI 1.36, 10.73) and 3.10 (1.00, 9.57) respectively. Borderline significant trends were also seen between paracetamol dose in the first three years of life and incident eczema and rhinitis, with adjusted ORs for heavy reported use compared to low of 1.59 (0.44, 5.74; p trend=0.06) and 2.31 (0.72, 7.46; p trend=0.07) respectively, but not with incident wheeze (fully adjusted OR=3.64; 1.34, 9.90, p trend=0.11). Cross-sectional analysis at age five resulted in significant positive dose-response effects of lifetime use (use at ages one, three and five) in relation to the prevalence of all outcomes.

Effects of gastro-intestinal infection

H. pylori infection was found in 17% of the children at age three but not five, 21% at age five but not three years, and 25% at both ages. In the longitudinal analysis, H. pylori infection at age three was significantly associated with a decreased risk of incident eczema between ages three and five years (adjusted OR, 95% CI, 0.31; 0.10, 0.94, p=0.02), but the associations with incident wheeze, rhinitis and sensitization were not significant. In cross-sectional analysis at age three, H. pylori infection was associated with a borderline significant reduced risk of eczema (adjusted OR, 95% CI, 0.49; 0.24, 1.01, p=0.05) and D. pteronyssinus sensitization (adjusted OR, 95% CI, 0.42; 0.17, 1.08, p=0.07), and a significant inverse association between current exposure to H. pylori, and any sensitization at age five (adjusted OR, 95% CI, 0.26; 0.07, 0.92, p=0.02). However, no significant associations were seen for wheeze and rhinitis.

The prevalence and intensity of geohelminth infection (hookworm, Ascaris lumbricoides and Trichuris trichiura) were found to be low in this cohort, with only 4% of children infected at age one, 9% at age three and only 0.2% at both ages. The risk of new onset wheeze between ages one and three was lower in those infected at age one (3.6%) than uninfected (7.8%), but infection was insufficiently prevalent to compute estimates of effect. Exposure to geohelminth infections in the first three years of life was not significantly associated with the incidence of reported outcomes or sensitization. However, A. lumbricoides infection was associated with a borderline increased risk of incident eczema between ages three and five (adjusted OR, 95% CI, 2.86; 1.04, 7.86, p=0.07). Children at age three were commonly colonized with enterococci 38% (207/544), lactobacilli 31% (169/544) and bifidobacteria 19% (103/544).

However, none of these commensal bacteria were associated significantly with either incidence or prevalence of allergic outcomes.


This longitudinal study from a developing country birth cohort provides further support for an association between early life use of paracetamol and increased risk of wheeze and allergic disease, which is unlikely to be explained by aspirin avoidance, reverse causation or confounding by indication. Furthermore, among young children in this cohort, the study found novel evidence to support the hypothesis of a protective effect of H. pylori infection on the risk of allergic disease, but no evidence to support an etiological role for the microflora enterococci, lactobacilli or bifidobacteria. The power of the study to explore the role of geohelminth infection on wheeze and allergic disease was limited by few infected children, and therefore understanding on this particular relation has not been much further advanced.

Item Type: Thesis (University of Nottingham only) (PhD)
Supervisors: Venn, A.
Davey, G.
Britton, J.R.
Keywords: Asthma in children, Risk factors, Allergy in children, Acetaminophen, Gastro-intestinal infections
Subjects: W Medicine and related subjects (NLM Classification) > WF Respiratory system
Faculties/Schools: UK Campuses > Faculty of Medicine and Health Sciences > School of Community Health Sciences
Item ID: 29714
Depositing User: Blore, Mrs Kathryn
Date Deposited: 01 Sep 2015 08:20
Last Modified: 16 Oct 2017 14:26
URI: https://eprints.nottingham.ac.uk/id/eprint/29714

Actions (Archive Staff Only)

Edit View Edit View