Recommendation generation for performance improvement by using cross-organizational process mining

Download
2015
Kahveci, Mehmet Kubilay
Post marketing surveillance for pharmaceutical drugs has been largely dependent on spontaneous reporting systems (SRSs) for quite some time. Although paper based reporting forms are broadly replaced by digital (online) counterparts, accessibility and usability of those systems still pose problems. Considering the fact that adverse drug event (ADE) reporting is mostly a voluntary action and it takes a lot of effort to complete an ADE report on current systems, outputs are usually in low quality and quantity. On the other hand, individual case safety reports (ICSRs), generated by SRSs, contain contextual information such as patient's active medications, past medical history or past drug therapies, most of which is already available in patient's electronic health records (EHRs). Therefore, seamlessly accessing EHR sources to pre-fill ICSR forms would be a major improvement for spontaneous reporting process. There have already been studies aiming to utilize EHR data for post market surveillance. However, rather than focusing on facilitating the reporting process, they target automated detection of adverse events and to the best of our knowledge, none of them aims mobile platforms or has usability concerns for the end user. In this thesis, we address the issue of under-reporting and interoperability of those systems and demonstrate that EHR systems can be exploited in mobile SRSs to generate high quality reports at high rates and provide a better experience to the reporter as well. We have developed a scalable platform, integrable to existing reporting systems and EHR sources of different content models, which can semi-automatically pre-fill ADE reports using medical summary of patient available in EHR systems. The quality of reports produced using our tool and the amount of time spent reporting is a significant improvement compared to existing mediums.
Citation Formats
M. K. Kahveci, “Recommendation generation for performance improvement by using cross-organizational process mining,” M.S. - Master of Science, 2015.