Pankush Kalgotra1, Ramesh Sharda2, Julie M Croff3. 1. Graduate School of Management, Clark University, 950 Main Street (Carlson Hall), Worcester, 01610 MA, USA. Electronic address: pkalgotra@clarku.edu. 2. Spears School of Business, Oklahoma State University, 102 Gunderson, Stillwater, 74078 OK, USA. Electronic address: ramesh.sharda@okstate.edu. 3. Oklahoma State University, 429 Willard Hall, Stillwater, 74078 OK, USA. Electronic address: julie.croff@okstate.edu.
Abstract
PROBLEM: Multimorbidity health disparities have not been well examined by gender. Co-occurring diseases may be mutually deleterious, co-occurring independently, or co-occurring from a common antecedent. Diseases linked by a common antecedent may be caused by biological, behavioral, social, or environmental factors. This paper aims to address the co-occurrences of diseases using network analysis. METHODS: In this study, we identify these multi-morbidities from a large electronic medical record (EMR) containing diagnoses, symptoms and treatment data on more than 22.1 million patients. We create multimorbidity networks from males and females medical records and compare their structural properties. RESULTS: Our macro analysis at the organ-level indicates that females have a stronger multimorbidity network than males. For example, the female multimorbidity network includes six linkages to mental health, wherein the male multimorbidity network includes only two linkages to mental health. The strength of some disease associations between lipid metabolism and chronic heart disorders is stronger in males than females. CONCLUSION: Our multimorbidity network analysis by gender identifies specific differences in disease diagnosis by gender, and presents questions for biological, behavioral, clinical, and policy research.
PROBLEM: Multimorbidity health disparities have not been well examined by gender. Co-occurring diseases may be mutually deleterious, co-occurring independently, or co-occurring from a common antecedent. Diseases linked by a common antecedent may be caused by biological, behavioral, social, or environmental factors. This paper aims to address the co-occurrences of diseases using network analysis. METHODS: In this study, we identify these multi-morbidities from a large electronic medical record (EMR) containing diagnoses, symptoms and treatment data on more than 22.1 million patients. We create multimorbidity networks from males and females medical records and compare their structural properties. RESULTS: Our macro analysis at the organ-level indicates that females have a stronger multimorbidity network than males. For example, the female multimorbidity network includes six linkages to mental health, wherein the male multimorbidity network includes only two linkages to mental health. The strength of some disease associations between lipid metabolism and chronic heart disorders is stronger in males than females. CONCLUSION: Our multimorbidity network analysis by gender identifies specific differences in disease diagnosis by gender, and presents questions for biological, behavioral, clinical, and policy research.