Search All Research Studies
AHRQ Research Studies Date
AHRQ Research Studies
Sign up: AHRQ Research Studies Email updates
Research Studies is a monthly compilation of research articles funded by AHRQ or authored by AHRQ researchers and recently published in journals or newsletters.
Results1 to 1 of 1 Research Studies Displayed
Rodriguez PJ, Ward ZJ, Long MW
Applied methods for estimating transition probabilities from electronic health record data.
This study examined the usefulness of electronic health records (EHRs) in estimating transition probabilities for state-transition models. No guidelines exist on appropriate methods. The authors used 3 potential methods to estimate transition probabilities for EHR data, using pediatric eating disorders (EDs) as a case study. They mapped diagnoses to 3 ED health states: anorexia nervous, bulimia nervosa, and other specified feeding or eating disorder using data from PEDsnet, which includes 8 US children’s hospitals. They used 3 approaches: simple first-last proportions, a multistate Markov (MSM) model, and independent survival models. The first-last proportion approach estimated higher probabilities of remaining the same health state, while the other two approaches estimated higher probabilities of transitioning to a different health state. Published literature differed substantially from these estimates. Further research is needed to improve methods for using EHR data to inform transition probabilities.
Citation: Rodriguez PJ, Ward ZJ, Long MW . Applied methods for estimating transition probabilities from electronic health record data. Med Decis Making 2021 Feb;41(2):143-52. doi: 10.1177/0272989x20985752..
Keywords: Electronic Health Records (EHRs), Health Information Technology (HIT), Health Services Research (HSR), Research Methodologies