Inpatient healthcare utilisation in patients with alcoholic liver disease: what are the costs and outcomes?

• Published in: Internal medicine journal 2016-12, Vol.46 (12), p.1407-1413
• Main Authors: Williamson, K. D., Gill, M. G., Andrews, J. M., Harley, H. A. J.
• Format: Article
• Language: English
• Subjects: Adult; Aged; alcohol abuse; alcoholic; Australia - epidemiology; Case-Control Studies; Comorbidity; delivery of healthcare; Female; healthcare economics and organisation; Hospital Costs - statistics & numerical data; Hospital Mortality; Humans; Inpatients; Length of Stay - economics; Length of Stay - statistics & numerical data; liver disease; Liver Diseases, Alcoholic - economics; Liver Diseases, Alcoholic - mortality; Liver Diseases, Alcoholic - therapy; Male; Middle Aged; mortality; Outcome Assessment (Health Care); Patient Acceptance of Health Care; Patient Discharge; Patient Readmission - economics; Patient Readmission - statistics & numerical data; Retrospective Studies
• ISSN: 1444-0903; 1445-5994
• DOI: 10.1111/imj.13258

Background Alcoholic liver disease (ALD) carries a significant cost burden and often leads to inpatient care. It is unclear whether inpatient care for ALD is any more costly than admission for other reasons. Aims To compare the costs and outcomes of inpatient care for ALD to two groups: a control group of matched cases admitted in the same time frame and people admitted for other chronic liver diseases (CLD). Methods All admissions for ALD and other CLD in a 3‐month period were retrospectively identified. Five randomly identified gender‐ and age‐matched contemporaneously admitted controls were allocated. Length of stay (LoS), mortality, inpatient costs, blood product utilisation and discharge destination were compared. Results Of the 71 admissions due to CLD, ALD was the most frequent cause (53/71, 75%). ALD admissions cost more (median $10 100 vs $5294; P = 0.0012) and had greater LoS (median LoS 7.2 days (interquartile range (IQR) 0.2–40.7)) than controls (2.6 days (IQR 1.1–6.8); P = 0.0001). A larger proportion of the ALD cohort required blood transfusion and had a higher mortality than controls (24.5 vs 6.4%, P = 0.002 and 13.2 vs 0.2%; P < 0.0001 respectively). Self‐discharge was more common in the ALD group (13.2 vs 1.1%, P < 0.0001). Conclusions ALD inpatient hospital admissions have greater median total cost, longer LoS, greater blood product utilisation, higher mortality and greater rate of discharge against medical advice than age‐ and gender‐matched controls. These data emphasise the large inpatient care burden, high mortality and suboptimal engagement in those with ALD, which justifies the more active provision of services for ALD.