Diabetes detection on the surgical wards in an area with a high prevalence of diabetes.

AIM. To describe the prevalence of diabetes and audit the processes used for diabetes detection on the surgical wards at Middlemore Hospital. METHODS. By the retrospective viewing of the notes of all admissions over a 12 month period. Comparison was made with routinely collected hospital discharge data. RESULTS. Notes from 10,893 of 11,004 (99.0%) admissions were viewed. 550 (5.0%) admissions involved a diabetic patient but diabetic patients were more likely to be readmitted (27.6% vs 13.5% diabetes vs no diabetes, chi 2 = 61.2, df = 1, p < 0.001). Only 18.2% of diabetic admissions and 24.5% of diabetic patients were identified as such on hospital discharge data. No screening for diabetes was undertaken in 34.2% of younger and 8.7% of older admissions and 7.5% of patients had either glycosuria or a glucose > or = 8.0 mmol/L but received no follow up. Only 17 new cases of diabetes were detected. Although no ethnic group difference in screening or identification of diabetes was found, in those aged > or = 60 years, 31% of Maori and 26% of Pacific Islands inpatients had diagnosed diabetes (vs 7.9% European). A similar excess of diabetes in Maori (18%) and Pacific Islands (16%) patients was found in those aged 40-59 years (Europeans 6%). CONCLUSION. In areas with a high proportion of Maori and Pacific Islands people, diabetes is an important medical condition to be considered in all surgical inpatients. Screening tests for diabetes need to be more closely followed up, perhaps by informing the general practitioner of the inpatient glucose.