Background: Obesity is a global public health concern associated with numerous comorbidities, increased healthcare utilisation and expenditure, and premature mortality. Despite its growing prevalence, the true extent of obesity amongst acute inpatients in Australian hospitals has not been well defined. Dedicated obesity services in Australia are largely confined to major centres and are arguably deficient to address increasing burden of disease. There are no publicly funded multidisciplinary obesity services available in the Geelong region of Victoria. This study aimed to determine the point prevalence of overweight and obesity in adult patients admitted to a tertiary Victorian regional public hospital compared to prevalence documented by hospital administrative coding. Additional aims included assessing proportion of patients with recorded current weight measurements and investigating for association between body mass index (BMI), length of stay, and cost of admission.
Methods: A cross-sectional audit was conducted of adults admitted under acute General Medical or General Surgical Units at University Hospital Geelong, Australia. Data collection occurred over three days in June 2024, including weight and height measurements to calculate BMI and demographic information. Hospital administrative data on ICD-10 coding, length of stay, and total admission costs was extracted for the cohort after hospital discharge.
Results: The point prevalence of obesity (BMI ≥ 30kg/m2) was determined as 28% in the hospitalised cohort (n=139). Prevalence of combined overweight and obesity (BMI ≥25kg/m2) was 56.1%, lower than Victorian and national Australian population data from 2022 (68.3% and 65.5% respectively, p <0.05). ICD-10 coding identified significantly fewer patients with obesity (prevalence 11.5%, p <0.01). Only 38% of patients had a current weight documented on their medication chart. No significant correlation was found between BMI and length of stay or cost of admission.
Conclusion: The prevalence of obesity in the hospitalised cohort was lower than the general Australian population but significantly higher than documented by hospital administrative coding. These findings suggest use of ICD-10 coding data is not representative of actual obesity prevalence highlighting the need for improved documentation to estimate demand for targeted multidisciplinary obesity services.