IJPDS International Journal of Population Data Science

A new paper published in the International Journal of Population Data Science (IJPDS) offers practical guidance for researchers grappling with a common challenge in health research: how to accurately measure patient complexity when using administrative health data.

Every day, hospitals, clinics, and physicians generate vast amounts of data when they bill for the health care services they provide. These routinely collected records are transformed into large administrative databases that support a wide range of research and policy work. Even though they offer a rich lens to measure population-level health outcomes, these data were not originally designed for research and sometime miss critical elements. For example, the medical complexity of individual patients can be hard to decipher. One solution to this problem is to use a comorbidity index.

The study reviews widely used indices as well as lesser-known tools that rely on diagnostic groupings or medication history. “There are many different comorbidity indices available and it can be overwhelming to navigate the options,” said lead author Boglarka Soos. “We wanted to create a practical manual that researchers can consult to simplify the choice.”

Furthermore, the paper walks researchers through real-world considerations for implementing these indices. For instance, how should comorbidities be coded—into a single score or as individual variables? Does using more data yield better results? And how do we ensure we accurately reflect the baseline health of patients?

“We want to offer advice on both choosing and using the right comorbidity index,” Soos explains. “Although, patient complexity is seldom the primary focus of the study, it is a crucial component of the overall picture. We need to ensure that researchers apply comorbidity indices thoughtfully.”

By offering practical advice, the guide aims to strengthen the design, consistency, and interpretability of future studies that rely on administrative health data.

 

Click here to read the full article

Boglarka Soos, PhD Candidate, Department of Community Health Sciences, University of Calgary, Canada

Soos, B., Williamson, T., McBrien, K., Wiebe, S., Tonelli, M., Southern, D., Eastwood, C., Li, B., Quan, H. and Ronksley, P. (2025) “Considerations for selecting and implementing comorbidity indices when using secondary data sources: a guide for health researchers”, International Journal of Population Data Science, 10(3). doi: 10.23889/ijpds.v10i3.2973.