Real-world data (RWD) is an emerging concept with a broad range of applications for studying patient health data including treatment safety and efficacy, short and long-term health outcomes, and disease burdens. RWD can include data from electronic health records, wearable technology, disease registries, patient-reported outcomes, social media, and various electronic health services and is a growing contributor to clinical and biopharma research on treatment efficacy. Traditional randomized control trials (RCTs) remain the gold standard for assessing treatment or diagnostic efficacy, but they often involve a homogenous population due to their strictly controlled inclusion criteria and limited sample sizes. Thus, RCTs can be limited in their ability to represent patient groups of minority backgrounds—including racial and ethnic minorities, patients with comorbidities, and patients with rare conditions.
Coupled with analysis using AI and machine learning technologies, vast amounts of real-world data from millions of patients have been analyzed to generate real-world evidence (RWE) more rapidly and with more cost-effectiveness than traditional clinical trials. With greater sample sizes, RWD may be better able to summarize the range of patient outcomes by including patient subgroups and demographics that are more representative of the true populations seeking healthcare. Additionally, this data may then be used to further efforts in precision medicine. Companies such as Flatiron Health, which generate RWE for oncology treatments, use data on patient genetic mutational variants and other biomarkers in an attempt to optimize treatment. This ability to capture diverse covariates is particularly useful in fields such as oncology and mental health, where patient outcomes are often associated with a multitude of genetic or demographic factors.
However, there are a number of limitations in the current capacities of RWD. A major limitation is the varying quality of data and lack of data standardization: patient notes involving free-text, incomplete fields, and different reporting formats across clinics all pose challenges. Additionally, rural healthcare systems or clinics with limited resources may be less able to anonymize data to share with RWD companies, potentially leading to their exclusion from RWD studies, and more broadly, the exclusion of minority populations whom their data represents.
Though further research and innovation is needed to improve data quality and analysis methodology, RWD continues to be a growing area of research that is taking steps toward addressing bias and disparities in healthcare.
Edited by: Anne Sacks
Graphic Designed by: Soojin Lee
Citations
https://bmcmedresmethodol.biomedcentral.com/articles/10.1186/s12874-022-01768-6
https://www.iqvia.com/solutions/real-world-evidence/real-world-data-and-insights?utm_source=google&utm_medium=cpc&utm_campaign=2023_gadsRWDUS_RWS_RS&utm_content=150775213880&utm_term=real%20world%20data&gad=1&gclid=Cj0KCQjwtJKqBhCaARIsAN_yS_mdudgoTAm9rWhG_uJabVVJcxoXQPnhssagteFuwgyG5LIib1B_rr4aArMZEALw_wcB