An accurate medication history is an essential part of patient assessment and can have vital impact on the persons care. However, manually-acquired histories are prone to inaccuracies. The problem is especially serious in emergency care and in disaster situations due to the lack of time, overloaded staff and special patient conditions (e.g. comatose or confused patients, unaccompanied minors or elderly patients). This study, which was partially funded by the Bethesda Hospitals Emergency Preparedness Partnership, focused on patients attended by the Emergency Department (ED) of a regional hospital (Suburban Hospital, Bethesda, MD) and evaluated the potential added value of prescription history information from SureScripts, a consortium of major pharmacy benefit managers, pharmacies and healthcare providers. We established secure connection between the hospital and SureScripts so that prescription-filling reports could be retrieved real-time from SureScripts, on presentation of four pieces of patient identifying information (name, date of birth, gender and zip code) gleaned from HL7 messages generated by the hospitals registration system. For three months we collected in parallel the SureScripts information and the medication history manually acquired by the ED nurse. We also retrieved demographic, administrative (e.g. class of insurance, mode of arrival) and clinical (e.g. vital signs, Glasgow coma score) information from the hospitals database. All the information was de-identified before being sent to NLM for analysis. This research only involved de-identified data collected for routine care purposes and there was no intervention or patient contact. To compare the two sources of information, we first did comprehensive mapping of the drug names to the U.S. standard drug terminology, RxNorm. The SureScripts report was more structured and the drug names were more regular. The manual drug history contained free text drug names. Mapping was done by a combination of text matching algorithms supplemented by manual inspection. About two-thirds of all ED patients were identified in the SureScripts database, and for about half of all patients some positive drug information was returned. Among patients with current medications data from both SureScripts and the ED medication history, we counted individual medication matches by patient. Considering only drugs in SureScripts that had expired for less than 3 months of the ED encounter, over 30% of all current medications were not captured in the ED medication history. On the other hand, 25% of the medications were missing from the SureScripts data set. Neither source is found to be complete, but SureScripts does add substantial information to the medication histories collected by the ED personnel. We also classified the missing drugs according to their potential impact on patient care. A significant proportion of missing drugs in the ED medication history were considered critical or important in patient care (e.g. anticoagulants). Since the SureScripts information is mainly derived from insurance-related sources, the availability of information varies with patient demographics and insurance coverage, among other factors. In order to assess the generalizability of the results of our study population, we have built a prediction model for the availability of SureScripts data based on patient characteristics. Our model shows that Caucasian, English-speaking patients on commercial insurance or Medicare are more likely to be found in SureScripts. Since the completion of the initial data collection, healthcare providers at the Suburban Hospital ED are now routinely getting a printed summary of the SureScripts information. The users of the reports generally find them very useful, particularly in patients who cannot remember the names of the drugs they are taking. In addition to listing the drugs that the patient is currently taking, since the SureScripts report also contains the full history of prescriptions being filled in the past year, it is also found to be very useful in spotting problematic behaviors (e.g. narcotic drugs abuse, poor drug compliance).
Showing the most recent 10 out of 16 publications
