Acetaminophen or Antibiotic Use and Childhood Allergic Diseases
Acetaminophen or Antibiotic Use and Childhood Allergic Diseases
A total of 263 620 live births from 1 January 1998 to 31 December 1998, and a total of 9910 live births from 1 January 2003 to 31 December 2003, separately, were included in this study. Similar demographic distribution was observed between the two study birth cohorts (Table 1). Specifically, 52% were male, nearly 30% were born in an urban area, and similar distributions across the four EC categories were observed for both birth cohorts. For the 1998 birth cohort: 67% had equal to or more than 13 ambulatory visits, 25% had one or more inpatient visit and 26% had a paediatric visit during the 1st year of life. For the 2003 birth cohort: 64% had equal to or more than 13 ambulatory visits, 29% had one or more inpatient visit and 28% had a paediatric visit during the 1st year of life (Table 1). In addition, the demographic distribution across the four different medication groups (no use of acetaminophen or antibiotics, only antibiotics, only acetaminophen, use of both drugs) between the two birth cohorts is provided in Supplementary Table 1 (available as Supplementary data at IJE online).
The study children were followed up from age 2–6 years in both the 1998 and 2003 birth cohorts. In detail, of the children in the 1998 birth cohort: 7.2% developed atopic dermatitis (N = 19 015); 21.7% developed asthma (N = 57 328); and 31.2% developed allergic rhinitis (N = 82 292) over the 5-year follow-up period. Likewise, in the 2003 birth cohort: 10.0% developed atopic dermatitis (N = 987); 28.0% developed asthma (N = 2773); and 37.2% developed allergic rhinitis (N = 3691) over the 5-year follow-up period. The incident peaks of atopic dermatitis were observed before age 3 years and gradually decreased across different medication groups in both birth cohorts, separately (Figure 1a and b). The incident peak of allergic rhinitis was observed at age 4–6 years across the four different medication groups in the 1998 birth cohort, and at age 4–5 years in the 2003 birth cohort, except for the group with no exposure to acetaminophen and antibiotics (Figures 1e and 1f). However, no clear pattern was observed for the incident peak of asthma, but an increase in asthma incidence was observed when age increased in the 2003 birth cohort, particularly around the age when children entered kindergarten (Figure 1d).
(Enlarge Image)
Figure 1.
Incidence rate of acetaminophen and/or antibiotic use for three examined atopic diseases (atopic dermatitis, asthma, allergic rhinitis) in the 1998 and 2003 birth cohorts, separately
Table 2a, Table 2b and Table 2c shows the results of univariate analysis of the temporal relationship between exposure to acetaminophen and/or antibiotics and the subsequent development of the three examined allergic diseases in the 1998 and 2003 birth cohorts, respectively. The results suggest that exposure to acetaminophen and/or antibiotics, gender (except for atopic dermatitis), EC at birth, geographical area at birth and healthcare utilization were important risk factors affecting the development of the examined allergic diseases in both the 1998 and 2003 birth cohorts. In addition, we also assessed the individual effect of acetaminophen and antibiotics, respectively, on the three examined allergic diseases. The results indicate that acetaminophen and antibiotic use, separately, have an individual effect on an increased risk of the three allergic diseases (data not shown). Moreover, when investigating the dose-response effect, Table 3 shows the dose-response relationship between exposure to acetaminophen and antibiotics, separately, during the 1st year of life and the subsequent development of the three examined allergic diseases in the 1998 birth cohort in the univariate analysis. A dose-response pattern was not observed, in particular for antibiotic use, in the 2003 birth cohort (Table 3).
Furthermore, when the results were adjusted for the risk factors mentioned above, exposure to acetaminophen and/or antibiotics during the f1st year of life remained positively associated with the development of atopic dermatitis, asthma and allergic rhinitis, separately, in the 1998 birth cohort [adjusted hazard ratio (Adj_HR): 1.98 and 95% CI: 1.89–2.07 for use of both acetaminophen and antibiotics in atopic dermatitis; Adj_HR: 1.73 and 95% CI: 1.66–1.81 for use of both acetaminophen and antibiotics in asthma; Adj_HR: 1.70 and 95% CI: 1.64–1.76 for use of both acetaminophen and antibiotics in allergic rhinitis] (Table 4). In addition, a protective effect of developing atopic dermatitis, asthma and allergic rhinitis, separately, was found among children in the higher EC group (lower SES group), and those residing in a rural area in the 1998 birth cohort (Table 4). Interestingly, similar results for EC at birth, and geographical area at birth, separately, were also observed in the 2003 birth cohort (Table 4), but the observed effects of drug exposure, especially for atopic dermatitis and asthma in the 2003 cohort, were lower than those observed in the 1998 cohort and were not statistically significant (Adj_HR: 1.03 and 95% CI: 0.86–1.24 for exposure to both acetaminophen and antibiotics in atopic dermatitis; Adj_HR: 1.11 and 95% CI: 0.95–1.29 for exposure to both acetaminophen and antibiotics in asthma). Additionally, when compared with the unadjusted HRs in Table 2, the adjusted HRs were lower in Table 4 than the unadjusted HRs in Table 2. Our results suggest that exposure to acetaminophen and/or antibiotics is weakly associated with an enhanced risk of the three examined childhood allergic diseases. Of note, concerning the accuracy of the asthma diagnosis, we further excluded children aged 2–4 years and repeated the adjusted analysis. Similar results were observed in Supplementary Table 2 (available as Supplementary Data at IJE online).
Results
Study Birth Cohorts
A total of 263 620 live births from 1 January 1998 to 31 December 1998, and a total of 9910 live births from 1 January 2003 to 31 December 2003, separately, were included in this study. Similar demographic distribution was observed between the two study birth cohorts (Table 1). Specifically, 52% were male, nearly 30% were born in an urban area, and similar distributions across the four EC categories were observed for both birth cohorts. For the 1998 birth cohort: 67% had equal to or more than 13 ambulatory visits, 25% had one or more inpatient visit and 26% had a paediatric visit during the 1st year of life. For the 2003 birth cohort: 64% had equal to or more than 13 ambulatory visits, 29% had one or more inpatient visit and 28% had a paediatric visit during the 1st year of life (Table 1). In addition, the demographic distribution across the four different medication groups (no use of acetaminophen or antibiotics, only antibiotics, only acetaminophen, use of both drugs) between the two birth cohorts is provided in Supplementary Table 1 (available as Supplementary data at IJE online).
Incidence Rate of Three Examined Allergic Diseases
The study children were followed up from age 2–6 years in both the 1998 and 2003 birth cohorts. In detail, of the children in the 1998 birth cohort: 7.2% developed atopic dermatitis (N = 19 015); 21.7% developed asthma (N = 57 328); and 31.2% developed allergic rhinitis (N = 82 292) over the 5-year follow-up period. Likewise, in the 2003 birth cohort: 10.0% developed atopic dermatitis (N = 987); 28.0% developed asthma (N = 2773); and 37.2% developed allergic rhinitis (N = 3691) over the 5-year follow-up period. The incident peaks of atopic dermatitis were observed before age 3 years and gradually decreased across different medication groups in both birth cohorts, separately (Figure 1a and b). The incident peak of allergic rhinitis was observed at age 4–6 years across the four different medication groups in the 1998 birth cohort, and at age 4–5 years in the 2003 birth cohort, except for the group with no exposure to acetaminophen and antibiotics (Figures 1e and 1f). However, no clear pattern was observed for the incident peak of asthma, but an increase in asthma incidence was observed when age increased in the 2003 birth cohort, particularly around the age when children entered kindergarten (Figure 1d).
(Enlarge Image)
Figure 1.
Incidence rate of acetaminophen and/or antibiotic use for three examined atopic diseases (atopic dermatitis, asthma, allergic rhinitis) in the 1998 and 2003 birth cohorts, separately
Temporal Relationship Between Acetaminophen and/or Antibiotic Use and Allergic Diseases
Table 2a, Table 2b and Table 2c shows the results of univariate analysis of the temporal relationship between exposure to acetaminophen and/or antibiotics and the subsequent development of the three examined allergic diseases in the 1998 and 2003 birth cohorts, respectively. The results suggest that exposure to acetaminophen and/or antibiotics, gender (except for atopic dermatitis), EC at birth, geographical area at birth and healthcare utilization were important risk factors affecting the development of the examined allergic diseases in both the 1998 and 2003 birth cohorts. In addition, we also assessed the individual effect of acetaminophen and antibiotics, respectively, on the three examined allergic diseases. The results indicate that acetaminophen and antibiotic use, separately, have an individual effect on an increased risk of the three allergic diseases (data not shown). Moreover, when investigating the dose-response effect, Table 3 shows the dose-response relationship between exposure to acetaminophen and antibiotics, separately, during the 1st year of life and the subsequent development of the three examined allergic diseases in the 1998 birth cohort in the univariate analysis. A dose-response pattern was not observed, in particular for antibiotic use, in the 2003 birth cohort (Table 3).
Furthermore, when the results were adjusted for the risk factors mentioned above, exposure to acetaminophen and/or antibiotics during the f1st year of life remained positively associated with the development of atopic dermatitis, asthma and allergic rhinitis, separately, in the 1998 birth cohort [adjusted hazard ratio (Adj_HR): 1.98 and 95% CI: 1.89–2.07 for use of both acetaminophen and antibiotics in atopic dermatitis; Adj_HR: 1.73 and 95% CI: 1.66–1.81 for use of both acetaminophen and antibiotics in asthma; Adj_HR: 1.70 and 95% CI: 1.64–1.76 for use of both acetaminophen and antibiotics in allergic rhinitis] (Table 4). In addition, a protective effect of developing atopic dermatitis, asthma and allergic rhinitis, separately, was found among children in the higher EC group (lower SES group), and those residing in a rural area in the 1998 birth cohort (Table 4). Interestingly, similar results for EC at birth, and geographical area at birth, separately, were also observed in the 2003 birth cohort (Table 4), but the observed effects of drug exposure, especially for atopic dermatitis and asthma in the 2003 cohort, were lower than those observed in the 1998 cohort and were not statistically significant (Adj_HR: 1.03 and 95% CI: 0.86–1.24 for exposure to both acetaminophen and antibiotics in atopic dermatitis; Adj_HR: 1.11 and 95% CI: 0.95–1.29 for exposure to both acetaminophen and antibiotics in asthma). Additionally, when compared with the unadjusted HRs in Table 2, the adjusted HRs were lower in Table 4 than the unadjusted HRs in Table 2. Our results suggest that exposure to acetaminophen and/or antibiotics is weakly associated with an enhanced risk of the three examined childhood allergic diseases. Of note, concerning the accuracy of the asthma diagnosis, we further excluded children aged 2–4 years and repeated the adjusted analysis. Similar results were observed in Supplementary Table 2 (available as Supplementary Data at IJE online).