Poster Presentation The Annual Scientific Meeting of the Australian Diabetes Society and the Australian Diabetes Educators Association 2013

Effects of hospital acquired conditions on length of stay for diabetic patients. (#348)

Jayde E Cromarty 1 , Sumit Parikh 2 , Kwang Lim 2 , Shamasunder Acharya 3 , Terri J Jackson 2
  1. University of Newcastle, Newcastle, NSW, Australia
  2. Northern Clinical Research Centre, Northern Health, Epping, VIC, Australia
  3. Endocrinology, John Hunter Hospital, Newcastle, NSW, Australia

Inpatients with diabetes are known to have longer lengths of stay (LOS) when hospitalised. Understanding differences in the rates and patterns of hospital acquired conditions between diabetes and similarly-complex patient sub-populations may reveal preventative measures to improve patient welfare and minimise LOS.

To evaluate the rate of hospital acquired conditions amongst diabetic and non-diabetic patients, to characterise differences in the type of complications these patients are most at risk of developing, and any effects on LOS.

47615 admission episodes from The Northern Hospital database (1 July 2011 to 30 June 2012) were reviewed and divided into four groups: 1) non-diabetic patients (n=45299), 2) diabetic patients without end-organ complications (n=356), 3) diabetic patients with end-organ complications (n=1775), and 4) a subset of non-diabetic patients with a Charlson co-morbidity score ≥1 (n=9255). Hospital acquired conditions were defined using the Classification of Hospital Acquired Diagnoses (CHADx). Linear regression was used to analyse the impact of group membership and the number of CHADx, on LOS.

Almost 30% of admissions of diabetic patients with end-organ complications had at least one hospital acquired condition, compared to only 13% for all non-diabetic patients and 17.6% for non-diabetic patients with a Charlson co-morbidity score ≥1. While the types of hospital-acquired conditions in diabetic patients were similar to their non-diabetic counterparts, their rates were consistently higher than the rates of similarly complex non-diabetic patients. Linear regression demonstrated diabetes patients with end-organ damage stay longer than other patients, and that each complication in a diabetes episode had a larger effect on LOS.

We demonstrate that diabetic patients have consistently higher rates of hospital acquired diagnoses and excess length of stay. These findings may provide a foundation for future clinical and cost-effectiveness studies of preventative practices for this high-risk patient population.