Table 3. Risk of autoimmune diseases after adjusting for competing mortality.
| Autoimmune disease | Total | Male | Female | |||
|---|---|---|---|---|---|---|
| aHRa (95% CI)b | p valuec | aHRa (95% CI)b | p valuec | aHRa (95% CI)b | p valuec | |
| All | 1.16 (1.07–1.26) | 0.0002* | 1.18 (1.04–1.33) | 0.0094 | 1.15 (1.04–1.28) | 0.0092 |
| Autoimmune thyroid disease | 1.18 (1.02–1.37) | 0.0295 | 1.37 (1.02–1.83) | 0.0389 | 1.13 (0.95–1.34) | 0.1809 |
| Uveitis | 1.23 (1.03–1.46) | 0.0190 | 1.26 (0.98–1.61) | 0.0694 | 1.19 (0.94–1.51) | 0.1487 |
| Psoriasis | 0.97 (0.79–1.20) | 0.7856 | 0.77 (0.58–1.01) | 0.0594 | 1.38 (1.00–1.90) | 0.0472 |
| Primary adrenocortical insufficiency | 1.10 (0.84–1.44) | 0.5040 | 1.14 (0.76–1.70) | 0.5256 | 1.07 (0.74–1.55) | 0.7240 |
| Ankylosing spondylitis | 0.98 (0.74–1.28) | 0.8615 | 1.26 (0.91–1.73) | 0.1696 | 0.58 (0.34–0.97) | 0.0392 |
| Autoimmune encephalomyelitis | 2.72 (1.84–4.02) | <0.0001* | 2.32 (1.41–3.83) | 0.0010* | 3.44 (1.84–6.41) | 0.0001* |
| Rheumatoid arthritis | 0.99 (0.68–1.44) | 0.9456 | 1.62 (0.73–3.62) | 0.2386 | 0.87 (0.57–1.34) | 0.5339 |
| Sjögren’s syndrome | 0.87 (0.53–1.41) | 0.5632 | 0.50 (0.06–4.09) | 0.5159 | 0.90 (0.54–1.48) | 0.6742 |
| Systemic lupus erythematosus | 1.86 (0.93–3.72) | 0.0817 | 1.87 (0.49–7.11) | 0.3606 | 1.86 (0.83–4.16) | 0.1334 |
| Myasthenia gravis | 1.36 (0.69–2.68) | 0.3744 | 1.80 (0.75–4.30) | 0.1874 | 0.96 (0.32–2.87) | 0.9445 |
| Guillain–Barré syndrome | 1.29 (0.61–2.72) | 0.5106 | 0.78 (0.26–2.35) | 0.6542 | 2.31 (0.81–6.58) | 0.1181 |
| Inflammatory bowel diseases | 2.01 (0.57–7.05) | 0.2748 | 2.31 (0.47–11.34) | 0.3024 | 1.11 (0.12–10.37) | 0.9256 |
| Post-infectious arthritis | 0.97 (0.33–2.82) | 0.9515 | 1.13 (0.32–4.01) | 0.8551 | 0.65 (0.08–5.04) | 0.6769 |
Abbreviations: aHR, adjusted hazard ratio; CI, confidence interval.
a Adjusted Cox proportional hazard model by sex, age, area of residence, income, urbanization, hypertension, diabetes mellitus, dyslipidemia, COPD, cerebrovascular accident, renal failure, liver cirrhosis, ischaemic heart disease, and malignancy.
b 95% CIs were not adjusted for multiple comparisons and thus cannot be directly used for hypothesis testing or inference.
c Post hoc adjustment for multiple comparisons by the Bonferroni method required a significance level of P<0.0012.
*Achieved statistical significance after Bonferroni correction for multiple comparisons (P<0.0012).