Objective: To identify factors that contribute to older Australians admitted to hospital with diabetes being re-hospitalised within 30 days of discharge.

Design, setting and participants: A retrospective cohort study of Department of Veterans' Affairs administrative data for all patients hospitalised for diabetes and discharged alive during the period 1 January - 31 December 2012.

Main outcome measures: Causes of re-hospitalisation and prevalence of clinical factors associated with re-hospitalisation within 30 days of discharge.

Methods: Multivariate logistic regression analysis (backward stepwise) was used to identify characteristics predictive of 30-day re-hospitalisation.

Results: 848 people were hospitalised for diabetes; their median age was 87 years (interquartile range, 77-89 years) and 60% were men. 209 patients (24.6%) were re-hospitalised within 30 days of discharge, of whom 77.5% were re-admitted within 14 days of discharge. 51 re-hospitalisations (24%) were for diabetes-related conditions; 41% of those re-admitted within 14 days had not seen their general practitioner between discharge and re-admission. Factors predictive of re-hospitalisation included comorbid heart failure (adjusted odds ratio [aOR], 1.49; 95% confidence interval [CI], 1.03-2.17; P = 0.036), numbers of prescribers in previous year (aOR [for each additional prescriber], 1.06; 95% CI, 1.01-1.08; P = 0.031), and two or more hospitalisations in the 6 months before the index admission (aOR, 1.79; 95% CI 1.15-2.78; P = 0.009).

Conclusion: Older people hospitalised for diabetes who have comorbid heart failure, multiple recent hospitalisations, and multiple prescribers involved in their care are at greatest risk of being re-admitted to hospital within 30 days. _targeted follow-up during the initial 14 days after discharge may facilitate appropriate interventions that avert re-admission of these at-risk patients.