The Importance of Real-World Data Linkages in Health Care Research

What is real-world data (RWD)? 

In the health care setting, RWD refers to the data relating to patient health status and/or the delivery of health care routinely collected or provided outside of a research setting. Examples of RWD include: 

• Medical and pharmaceutical claims
• Electronic health records (EHRs)
• Disease registry data
• Data gathered from other non-traditional sources, such as mobile devices, social media, and wearable devices.

RTI remains a leader in using RWD for cutting edge health research, from using EHRs to inform public health surveillance to using wearables to detect influenza before someone gets sick to understanding the age of JUUL’s Twitter followers to highlight the impact of social media marketing of tobacco products on youth.

Why do researchers use RWD to improve evidence-based research and health outcomes?

Real-world data possess many benefits to the researcher; they do not require additional data collection efforts, offer robust sample sizes, and often reflect a more complex set of patient demographics and risk profiles than found in clinical trials. Medical and pharmaceutical claims, for instance, are needed to process payments between providers and insurers or the government.

Claims are operational data that researchers leverage to answer a host of research questions, such as the cost of treating breast cancer or understanding polypharmacy among foster children. Also, because clinical trial dataset are typically smaller and subject to strict inclusion/exclusion criteria, RWD sources offer robust sample sizes needed for inference, which can lead to a broader, more diverse set of patients with a constellation of health needs.

How can RWD benefit from being linked to other data?

While RWD benefit from being inexpensive to gather, most RWD sources suffer from being limited in their perspective. Because RWD was not designed for research, it often fails to address critical information needed to answer a research question. Linking across two different RWD sources or between a RWD and a survey fills key gaps that a single source alone cannot answer. By itself, a survey could be a snapshot of a patient’s wellbeing at a point in time. When combined with a set of longitudinal EHRs, it becomes a patient journey.

Consider this specific research question: is coordinated care for people with cancer associated with lower health care costs? Asking patients directly about their care coordination experiences provides valuable insights about the quality and coordination of their treatment, but these surveys don't include information about medical expenses. Conversely, medical insurance claims contain detailed cost data, but they don't indicate whether patients received well-coordinated care between their different health care providers.

How RTI Uses RWD Linkages to Improve Health Research

A RWD linkage solves the above problem. RTI has been the leader in using the National Cancer Institute’s linked resource: Surveillance, Epidemiology and End Results (SEER) cancer registry data and the Centers for Medicare & Medicaid Services' (CMS) Medicare Consumer Assessment of Healthcare Providers and Systems (CAHPS®) patient surveys, which is a cancer registry linked with a patient survey linked with Medicare claims. After analyzing this linked resource, we found that coordinated care was associated with less expensive care.

RWD linkages represent essential infrastructure for modern evidence-based research, enabling comprehensive studies that would be impossible through traditional data collection methods. By connecting disparate health care data sources, we can transform how researchers understand disease patterns and health outcomes across a diverse set of populations. 

In the next post, we will explore how artificial intelligence (AI) can assist with record linkage to enhance quality control and accelerate matching, both which promote user confidence in the linked data as a source for research.