Hang Qiu1, Liya Wang2, Xianrong Zeng3, Jingping Pan4. 1. School of Computer Science and Engineering, University of Electronic Science and Technology of China, Chengdu, China; Big Data Research Center, University of Electronic Science and Technology of China, Chengdu, China. Electronic address: qiuhang@uestc.edu.cn. 2. Big Data Research Center, University of Electronic Science and Technology of China, Chengdu, China. 3. Department of Neurology, Sichuan Provincial People's Hospital, Chengdu, China. 4. Health Information Center of Sichuan Province, Chengdu, China.
Abstract
BACKGROUND: Depression is a psychiatric disorder with a high comorbidity burden; however, previous comorbidity studies predominately focused on a few common diseases and relied on self-reported data. We aimed to investigate the comorbid status of depression concerning the entire spectrum of chronic diseases using network analysis. METHOD: Totally, 22,872 depressed inpatients and one-to-one matched controls were enrolled in the retrospective study. Hospital discharge records were aggregated to measure the comorbidities, where those with a prevalence ≥ 1% were selected for further analysis. Based on the co-occurrence frequency, sex- and age-specific comorbidity networks in depressed patients were constructed and the results were compared with the controls. Louvain algorithm was used to detect the highly interlinked communities. RESULTS: Depressed patients had 4 comorbidities on average, and 84.4% had at least one comorbidity. The comorbidity network in depression cases was more complex than controls (connections of 839 vs. 369). Intricate but distinct communities appeared within the comorbidity network in depressed patients, where the largest community included cerebrovascular diseases, chronic ischaemia heart disease, atherosclerosis and osteoporosis. Sex-specific central diseases existed, and cardiovascular diseases were the major central diseases to both gender. The older the depressed patients, the more severe the central diseases in the comorbidity network. LIMITATIONS: The causality of the observed interactions could not be determined. CONCLUSIONS: The application of network analysis on longitudinal healthcare datasets to assess comorbidity patterns can supplement the traditional clinical study approaches. The findings would improve our understanding of depression-related comorbidities and enhance the integrated management of depression.
BACKGROUND: Depression is a psychiatric disorder with a high comorbidity burden; however, previous comorbidity studies predominately focused on a few common diseases and relied on self-reported data. We aimed to investigate the comorbid status of depression concerning the entire spectrum of chronic diseases using network analysis. METHOD: Totally, 22,872 depressed inpatients and one-to-one matched controls were enrolled in the retrospective study. Hospital discharge records were aggregated to measure the comorbidities, where those with a prevalence ≥ 1% were selected for further analysis. Based on the co-occurrence frequency, sex- and age-specific comorbidity networks in depressed patients were constructed and the results were compared with the controls. Louvain algorithm was used to detect the highly interlinked communities. RESULTS: Depressed patients had 4 comorbidities on average, and 84.4% had at least one comorbidity. The comorbidity network in depression cases was more complex than controls (connections of 839 vs. 369). Intricate but distinct communities appeared within the comorbidity network in depressed patients, where the largest community included cerebrovascular diseases, chronic ischaemia heart disease, atherosclerosis and osteoporosis. Sex-specific central diseases existed, and cardiovascular diseases were the major central diseases to both gender. The older the depressed patients, the more severe the central diseases in the comorbidity network. LIMITATIONS: The causality of the observed interactions could not be determined. CONCLUSIONS: The application of network analysis on longitudinal healthcare datasets to assess comorbidity patterns can supplement the traditional clinical study approaches. The findings would improve our understanding of depression-related comorbidities and enhance the integrated management of depression.
Authors: Juan A G Ranea; James Perkins; Mónica Chagoyen; Elena Díaz-Santiago; Florencio Pazos Journal: Genes (Basel) Date: 2022-06-17 Impact factor: 4.141