Linking census data with electronic medical records for clinical research: A systematic review

This paper considers the evidence documenting the linkage of EMRs and census when conducting clinical research. Our systematic review included 25 studies. They collected information on an average of 434,541 study participants and 72% of the studies focused on adult populations. The findings include that the most common diseases evaluated were obesity, cancer, and diabetes. The most commonly used census variables were location, income, and education. Twelve of the studies linked only the census and the EMR, while 13 studies linked the census, EMR, and additional research resources. This linkage was most prevalently used to describe a problem rather than for quality improvement purposes. Efforts should be channeled to increase the use of the census for health disparities and social determinants of health.