Disease spectrum and comorbidity patterns of malignant neoplasms: a multi-center hospital-based retrospective analysis of inpatient insurance claims data

Background Previous studies indicate a high comorbidity burden among patients with malignant neoplasms, but claims-based comorbidity patterns have not been systematically characterized at the regional multi-hospital level in China. Methods This study is a multi-center, hospital-based retrospective analysis of inpatient insurance claims. We analyzed anonymized inpatient medical insurance claims data (2016–2021) from 163 hospitals in Zhanjiang, China, focusing on hospitalized patients with a primary diagnosis of malignant neoplasms. Malignant neoplasms and comorbidities were identified using International Classification of Diseases, 10th Revision(ICD-10) codes. Disease spectrum was stratified by sex, age, and region. Comorbidity patterns were delineated using association rule mining and network analysis. Results Among 107,029 patients, the ten most common malignancies accounted for 75.96% of all cases. Hospitalizations were more frequent among rural populations, males, and individuals aged ≥65 years. The median number of co-diagnosed conditions across major malignancies was 5 (interquartile range [IQR]: 3–7). Network analysis revealed three major co-diagnosis clusters: 1) liver cancer with chronic viral hepatitis and hepatic fibrosis; 2) lung cancer with chronic obstructive pulmonary disease (COPD) and pneumonia; and 3) colorectal cancer with inflammatory bowel disease-related conditions and intestinal obstruction. Patterns varied across sex, age groups, and urban–rural residence. Conclusions This study demonstrates a high comorbidity burden among hospitalized cancer patients, with distinct malignancy-specific co-diagnosis patterns. These findings support the need for integrated clinical management and targeted healthcare resource allocation, particularly for older, male, and rural patient populations.