Multimorbidity patterns and their associations with healthcare services utilisation in inpatients with chronic hepatitis B infection from 2011 to 2023: a retrospective observational study.

[BACKGROUND] Multimorbidity among patients with chronic hepatitis B (CHB) infection has emerged as a priority for healthcare and public health systems worldwide.

[OBJECTIVE] This study aimed to characterise time-trends in multimorbidities among patients with CHB infection. We identified multimorbidity clusters and combinations and quantified their associations with healthcare services utilisation.

[DESIGN] A retrospective observational study, using electronic medical record data.

[SETTING] A large tertiary general hospital in China.

[PARTICIPANTS] The study included 23 137 patients with CHB infection admitted between 2011 and 2023.

[OUTCOME MEASURES] Latent class analysis and association rule mining (ARM) were performed to identify multimorbidity clusters and combinations, respectively. Multivariable logistic regression quantified associations between the identified multimorbidity patterns and length of stay (LOS), daily expense and 1-year readmission for liver-related conditions (OYRL).

[RESULTS] The mean number of multimorbidities among hospitalised patients with CHB infection was 2.82±1.89. From 2011 to 2023, mean age increased from 44.2±13.7 to 48.4±13.1 (p<0.001). The prevalence of cirrhosis (45.50%-57.10%), hepatocellular carcinoma (HCC) (13.10%-17.10%) and non-alcoholic fatty liver disease (3.15%-5.08%) increased over time. Similar trends were observed for non-liver multimorbidities, including diabetes mellitus (9.86%-11.90%), hypertension (7.34%-10.30%) and chronic kidney diseases (0.96%-1.58%). We identified three multimorbidity clusters: Cluster 1 (43.58%) included patients in the early phase of CHB infection with the lowest overall burden of multimorbidity. Cluster 2 (47.71%) was characterised prominently by cirrhosis and HCC. Patients in cluster 3 (8.70%) were the oldest and exhibited the highest probability of metabolic, circulatory and kidney-related multimorbidities. Three clusters demonstrated different association strengths with healthcare utilisation. Most multimorbidity combinations identified by ARM were significantly associated with higher LOS and OYRL, but lower daily expenses.

[CONCLUSIONS] Multimorbidity imposes a substantial burden on CHB-infected patients. Our findings highlight the importance of early diagnosis and treatment of CHB infection, as well as tailored integral strategies for multimorbidity management in individuals with CHB infection.