Elma Dervic has found a way to reconcile transparency and data privacy in medical research, showing how the two don’t have to be in conflict.
How reproducible can medical research be when data is under the highest level of protection? This question has driven CSH researcher Elma Dervic since the beginning of her scientific career. Now, she has published a data paper addressing precisely this challenge.
In it, she presents networks of hospital diagnoses, broken down by age, gender, and observation period into a total of 96 subgroups. “The raw data remains protected, but disease relationships become visible,” Dervic explains.
How Much Transparency Is Possible?
When researchers share their findings openly, others can reproduce their work—an essential quality criterion in science. After all, how else can we verify whether results hold up to reanalysis? But in the medical field, transparency is particularly challenging.
“Medical data is highly sensitive and protected by multiple security measures for a good reason—it belongs to real people,” says Dervic.
In one study, she mapped 45 million hospital stays over 17 years in Austria to identify common disease trajectories in multimorbid patients from birth to death, as well as the critical moments that significantly shape their health outcomes. Throughout this work, one principle remained non-negotiable: “I exchanged insights and results with colleagues, but the raw data always remained secured and inaccessible outside my research group,” Dervic emphasizes.
Data Sharing Without Compromising Privacy
The turning point came with a request from the University of Belgrade, when a new idea emerged: What she could share were comorbidity networks.
“A comorbidity network isn’t a patient record—it’s a matrix of relationships, a structure that reveals how diseases connect without exposing individual details. And just like that, I had my aha! moment. I could share valuable datasets for research while fully respecting privacy laws,” says Dervic.
“So, we created a dataset that allows other researchers to explore comorbidity networks without requiring deep expertise in their construction,” Dervic adds. The team also shared aggregated data, enabling other scientists to build their own networks.
Why Do We Need Comorbidity Networks?
Comorbidity networks have become a crucial tool in network medicine, offering a systematic understanding of diseases and making connections between conditions visible.
According to Dervic, this dataset is intended to facilitate new discoveries in public health, network medicine, and AI-driven health research. To make access even easier, she and her team have developed an interactive web app where anyone can explore these comorbidity networks.
