RESULTS
The research flowchart is shown in Fig. 1. We identified 87015 people with constipation and 87015 matched controls between 1999 and 2013 from the LHID. Table 1 shows the demographic characteristics of the study participants. The constipated patients and non-constipation cohort were similar in age and gender distribution. There were no statistically significant differences between the constipation and non-constipation groups after propensity score matching. The median follow-up duration in constipated group is 7.9 years and 8.5 years in non-constipated groups, respectively.
As shown in Table 2, the incidence of AD revealed 4.7 per 1,000 person-years in the constipation group, which was higher than the rate of 2.2 per 1,000 person-years observed in the non-constipation group. After adjustment, people with constipation had a significantly higher risk of AD than those without constipation (aHR: 2.11, 95% C.I. 1.98-2.24, p<0.001) In addition, the age groups < 6 years and ≧65 years showed a relatively higher risk of AD. (aHR of < 6 years old group: 2.94, 95% C.I. 2.64-3.28, p<0.001; ≧65 years old group: 1.60, 95% C.I. 1.47-1.75, p<0.001). Compared with women, men revealed a non-significantly lower risk of AD (aHR, 0.96; 95% CI, 0.9-1.02; p=0.16). In term of comorbidities, we observed that people with hypertension, chronic liver disease, diabetes, or COPD had a relatively higher risk of AD. (hypertension:1.21, 95% C.I. 1.1-1.33, p<0.001; chronic liver disease:1.34, 95% C.I. 1.13-1.59, p=0.001; diabetes:1.20, 95% C.I. 1.07-1.35, p=0.002; COPD: 1.44, 95% C.I. 1.21-1.7, p<0.001). By contrast, patients using corticosteroids or antihistamines during the study for period at least 30 days had a lower risk of AD. (corticosteroids: 0.76, 95% C.I. 0.70-0.82, p<0.001; antihistamines: 0.53, 95% C.I. 0.50-0.57, p<0.001)
Subgroup analyses were applied to assess the association between constipation and AD based on demographic characteristics, as shown in Table 3. Patients in the constipation group aged 12-19 years, had a 2.34-fold greater risk of AD compared with the same age group in the non-constipation group (aHR; 95% CI,1.84-2.98, P < 0.001). Patients aged 20-39, 40-64, and ≥65 years in the constipation group had a 2.23, 2.19, and 2.08-fold greater risk of AD.(aHR; 95% CI, 1.98-2.50, 1.93-2.48 and 1.85-2.34; P < 0.001), respectively. Among females, compared with those without constipation, there was a 2.16-fold higher risk of AD in patients with constipation (aHR; 95% CI, 2.01-2.33; P < 0.001). Among males, there was 1.95-fold higher risk of AD in patients with constipation (aHR; 95% CI, 1.75-2.16; P < 0.001). Furthermore, constipated patients had a higher likelihood of AD, regardless of comorbidities. However, constipated patients with using corticosteroids or antihistamines disclosed a lower risk of AD compared with non-users.
The Kaplan–Meier curves are shown in Fig. 2. The cumulative incidence of AD showed significantly higher in constipated people than in non-constipated patients, and the log-rank test for the comparison of cumulative incidence curves resulted in the P-value of <0.001