Table 5.
Weighted odds ratios of the number of hospitalizations and ICU admissions in the exposed group relative to the reference group a.
| Observed Data (N = 206,248) | Weighted Data (N = 199,952) c | Weighted OR | |||
|---|---|---|---|---|---|
| All-cause Hospitalization, N (%) b | Non-Exposed (N = 165,925) |
Exposed (N = 40,323) |
Non-Exposed (N = 160,897) |
Exposed (N = 39,055) |
(95% CI), % d |
| None | 110,657 (66.69) | 26,768 (66.38) | 107,139 (66.59) | 25,994(66.65) | Reference |
| 1–2 events | 40,692 (24.52) | 9819 (24.35) | 39,565 (24.59) | 9488 (24.29) | 0.99 (0.96 to 1.02) |
| 3–5 events | 11,204 (6.75) | 2750 (6.82) | 10,899 (6.77) | 2632 (6.74) | 1.00 (0.95 to 1.04) |
| ≥6 events | 3372 (2.03) | 986 (2.45) | 3294 (2.05) | 941 (2.41) | 1.18 (1.09 to 1.29) |
| All-cause ICU admission, N (%) b | |||||
| None | 165,725 (99.88) | 40,257 (99.84) | 160,705 (99.88) | 38,992 (99.8) | Reference |
| 1 event | 36 (0.02) | 12 (0.03) | 34 (0.02) | 12 (0.03) | 1.39 (0.71 to 2.71) |
| ≥2 events | 164 (0.1) | 54 (0.13) | 158 (0.1) | 51 (0.13) | 1.34 (0.97 to 1.83) |
Abbreviations: CI, confidence intervals; ICU, intensive care unit; IPTW, inverse probability of treatment weighting; OR, odds ratio. A significant difference (p < 0.001) is indicated in bold. a Clinical conditions recorded after 24 months were investigated unless otherwise stated. b Results are reported as n (%) unless otherwise indicated. c IPTW on the propensity score was employed to balance the groups for baseline health data. Seventy covariates were weighted (Table 1 and Supplementary Tables S3 and S4). Individuals in the reference group were weighted using stabilized weights to produce a sample with the same distribution of covariates as the exposed group. d Multinomial logistic regression was used to determine OR and their 95% CI for the number of hospitalization events and ICU admissions as dependent variables. Breastfeeding duration was considered an independent variable.