TABLE 2.
All-cause HCRU in Patients With EoE and Matched Controls During Follow-up
| HCRU | Patients With EoE (n=16,094) [A] | Matched Controls (n=16,094) [B] | Unadjusted OR* (95% CI) | P † | Adjusted OR (95% CI)‡ | P ‡ [A] vs. [B] |
|---|---|---|---|---|---|---|
| Diagnostic endoscopies, n (%) | 11,890 (73.9) | 278 (1.7) | 160.9 (142.2-182.1) | <0.0001 | 184.3 (160.4-211.7) | <0.0001 |
| Inpatient admissions, n (%) | 1107 (6.9) | 605 (3.8) | 1.9 (1.7-2.1) | <0.0001 | 1.5 (1.3-1.7) | <0.0001 |
| ER visits | ||||||
| Any visit, n (%) | 4211 (26.2) | 2069 (12.9) | 2.4 (2.3-2.6) | <0.0001 | 2.2 (2.0-2.3) | <0.0001 |
| Food impaction | 1083 (6.7) | 7 (0.04) | 165.8 (78.8-348.7) | <0.0001 | — | <0.0001 |
| Esophageal stricture | 510 (3.2) | 1 (0.006) | 526.7 (74.0-3746.4) | <0.0001 | — | <0.0001 |
| Asthma | 403 (2.5) | 124 (0.8) | 3.3 (2.7-4.1) | <0.0001 | 2.1 (1.7-2.7) | <0.0001 |
| Food allergy | 81 (0.5) | 12 (0.1) | 6.8 (3.7-12.4) | <0.0001 | 4.9 (2.5-9.5) | <0.0001 |
| Dysphagia | 77 (0.5) | 0 (0.0) | 77.4 (10.8-556.3) | <0.0001 | — | <0.0001 |
| GERD | 509 (3.2) | 54 (0.3) | 9.7 (7.3-12.9) | <0.0001 | 5.8 (4.1-8.2) | <0.0001 |
| Outpatient visits | ||||||
| Any visit, n (%) | 16,042 (99.7) | 13,810 (85.8) | 51.0 (38.7-67.2) | <0.0001 | 24.2 (18.3-32.1) | <0.0001 |
| Gastroenterologist | 10,838 (67.3) | 723 (4.5) | 43.8 (40.4-47.6) | <0.0001 | 35.6 (32.5-38.9) | <0.0001 |
| Allergist | 6226 (38.7) | 534 (3.3) | 18.4 (16.8-20.2) | <0.0001 | 14.9 (13.5-16.4) | <0.0001 |
| Pulmonary specialist | 511 (3.2) | 291 (1.8) | 1.8 (1.5-2.1) | <0.0001 | 1.1 (1.0-1.4) | 0.1592 |
| Psychologist | 586 (3.6) | 286 (1.8) | 2.1 (1.8-2.4) | <0.0001 | 1.5 (1.3-1.8) | <0.0001 |
Unadjusted ORs were evaluated for binary variables (ie, at least 1 visit). ORs for ER visits associated with dysphagia assume 1 event for matched controls.
P-values were calculated using McNemar’s test or the exact McNemar’s test for binary variables. A P-value <0.05 was statistically significant.
Adjusted ORs were evaluated for binary variables using logistic regression controlling for index year, health plan at index date, comorbidities (GERD and atopic disease), prior treatment (medication for GERD and asthma) and HCRU (inpatient, outpatient, and ER) during the preindex period. A generalized estimating equation was used to control for correlation between pairs. The adjusted models for ER visits associated with dysphagia, esophageal stricture, and food impaction did not converge owing to low numbers of events.
CI indicates confidence interval; EoE, eosinophilic esophagitis; ER, emergency room; GERD, gastroesophageal reflux disease; HCRU, health care resource use; OR, odds ratio.