OBJECTIVES: Previous reports suggest diabetes as the most prevalent chronic condition among gastroenteritis cases admitted to hospital but little evidence is available on the extent to which diabetes is a risk factor for medically-attended acute gastroenteritis (MA-AGE).

METHODS: Patients registered in general practices contributing to CPRD (Clinical Practice Research Datalink) from January 2006 to December 2014 and eligible for linkage to HES (Hospital Episode Statistics) APC (Admitted Patient Care) data were included. The type-I diabetic, type-II diabetic and diabetes-free cohorts were defined based on diabetes-related diagnoses, treatments and test results. MA-AGE events were defined based on MA-AGE related Read and ICD-10 codes. Episodes were defined as events separated by a minimum 14-day disease-free period. Episode rates and 95% confidence intervals (CI) were calculated for the three cohorts overall, by age group, by gender and by healthcare setting (diagnosed in hospital vs by GP only). In addition, age-gender standardized episode rate ratios comparing the type-I and type-II diabetic cohorts to the diabetes-free cohort were calculated, overall and by health care setting.

RESULTS: We observed an episode rate of 60.96 per 1000 person-years (95%CI: 59.33-62.62) in type-I diabetics and 62.56 (95%CI: 62.10-63.03) in type-II diabetics, compared to 32.58 (95%CI: 32.51-32.66) in diabetic-free. This increase is seen across all age groups and both genders, although varying in magnitude, and is higher in hospital than for GP only. The age-gender standardized episode rates ratio comparing the type-I and type-II diabetic cohorts to the diabetes-free cohort are 1.99 (95%CI: 1.96-2.03) and 1.78 (95%CI: 1.73-1.83) overall, 5.03 (95%CI: 4.88-5.18) and 2.85 (95%CI: 2.71-3.00) in hospital and 1.43 (95%CI: 1.43-1.48) and 1.59 (95%CI: 1.54-1.64) for GP only.

CONCLUSIONS: MA-AGE episode rates are increased in the diabetic compared to the diabetes-free populations. The highest increase is seen for MA-AGE diagnosed in hospital among type-I diabetics.