RESULTS
In this study, a total of 59,484 children’s parents or guardians completed questionnaires, and the response rate was 86.5%. The average age of participants was 10.23 years, and 49.39% was girls (Table 1). The prevalence of childhood asthma was 7.77%. Of those 4619 participants diagnosed with asthma, 585 cases (12.67%) had one family member affected by asthma, 73 cases (1.58%) had two affected family members and 13 cases (0.28%) had at least three affected family members.
Fig. 1 shows the adjusted ORs for childhood asthma by the number of family members affected by asthma. The ORs were 2.58 (95%CI: 2.34-2.84), 4.63 (95%CI: 3.53-6.08) and 7.38 (95%CI: 3.70-14.73) when one, two and at least three family members were affected, and the trend was statistically significant (P <0.001). The crude ORs which are shown in Table S1 were slightly greater than those adjusted.
Among children with one family member affected by asthma, those with affected father had the highest adjusted OR (4.89, 95%CI: 3.72-6.42, Cases: 75), followed by those with affected mother (OR: 3.94, 95%CI: 2.99-5.20, Cases: 67), paternal grandfather (OR: 2.59, 95%CI: 2.14-3.13, Cases: 132), maternal grandfather (OR: 2.40, 95%CI: 1.93-2.99, Cases: 97), paternal grandmother (OR: 2.08, 95%CI: 1.71-2.53, Cases: 119) and maternal grandmother (OR: 2.08, 95%CI: 1.67-2.59, Cases: 95) (Fig. 2). The P value of the trend was statistically significant (P <0.001). The corresponding crude ORs are shown in Table S2.
Among children with two family member affected by asthma, the odds of asthma for children’s both parents were affected (OR: 15.92, 95%CI: 4.66-54.45, Cases: 5) or those whose father and paternal grandfather were affected (OR: 11.11, 95%CI: 5.77-21.38, Cases: 16) were higher than those with other family history types, including paternal grandparents affected (OR: 5.04, 95%CI: 2.45-10.37, Cases: 10), maternal grandparents affected (OR: 4.39, 95%CI: 1.88-10.27, Cases: 7), father and paternal grandmother affected (OR: 3.64, 95%CI: 1.35-9.83, Cases: 5), mother and maternal grandfather affected (OR=5.54, 95%CI: 1.46-21.07, Cases: 3), or mother and maternal grandmother affected (OR=6.72, 95%CI: 2.86-15.81, Cases: 8) (Fig. 3). The crude ORs shown in Table S3 were slightly higher than those adjusted.
Table 2 shows the proportion of the effect of grandparental asthma on their grandchildren’s asthma that was mediated by parental asthma. Paternal asthma significantly mediated the association of childhood asthma with paternal grandfather asthma (12%) and paternal grandmother asthma (11%), and maternal asthma significantly mediated the association of childhood asthma with maternal grandfather asthma (6%) and maternal grandmother asthma (9%).