Introduction Mortality rates of patients hospitalised with ulcerative colitis (UC) were defined to identify important risk factors.
Methods A cohort study was undertaken using the Scottish national record linkage database. Three-year mortality rates of patients with a primary diagnosis of UC between 1998 and 2000 were determined according to the type of admission: colectomy elective, colectomy emergency, no colectomy elective and no colectomy emergency. Univariate and multivariate analyses were used to determine the influence of age, gender, comorbidity, interval from admission to surgery, social deprivation, length of stay and type of hospital.
Result Of 1078 patients identified (90.9% audit completeness), 177 were coded as colectomy elective, 100 as colectomy emergency, 244 as no colectomy elective and 557 as no colectomy emergency. Crude three-year mortality rates were 5.6%, 9.0%, 9.8% and 16.0%. Male gender, comorbidity and length of stay were each independently related to mortality on multivariate analysis. Age was the critical determinant of mortality; no patient aged <30 years died; mortality in the under 50s (10/587 (1.7%)) was significantly lower than for those aged 50–64 years (26/246 (10.6%)) (χ2=32.91, p<0.0000001) and 65 years and above (96/245 (39.2%)) (χ2=218.2, p<0.0000001). There was no significant difference in mortality by admission type under 65 years. At 65 years and beyond the three year mortality in the four admission groups was 29.4%, 33.3%, 28.1% and 44.7%.
Conclusion Three-year mortality after an incident admission for UC is high over 65 years and is highest for emergency admissions not undergoing surgery. Management protocols in the elderly should be reconsidered in the light of these data.