Albert Tzeel1, Victor Lawnicki2, Kim R Pemble3. 1. National Medical Director, HumanaOne, Clinical Leadership and Policy Development, Humana, Inc, Milwaukee, WI. 2. Econometrician, Business Intelligence and Informatics Competency Center, Humana, Inc, Louisville, KY. 3. Executive Director and Chief Executive Officer, Wisconsin Health Information Exchange, Mequon, WI.
Abstract
BACKGROUND: Health information exchanges (HIEs) have already demonstrated direct value in controlling the costs associated with utilization of emergency department services and with inpatient admissions from the emergency department. HIEs may also affect inpatient admissions originating from outside of the emergency department. OBJECTIVE: To assess if a potential association exists between a community-based HIE being used in hospital emergency departments and inpatient admissions emanating from outside of the emergency department. METHODS: The study design was observational, with an eligible population of fully insured plan members who sought emergency department care on at least 2 occasions over the study period between December 2008 and March 2010. Utilization data, obtained from medical and pharmacy claims, were matched to a list of emergency department utilizers where HIE querying could have occurred. Of the eligible members, 1482 underwent propensity score matching to create two 325-member groups-(1) a test group in which the HIE database was queried for all members in all of their emergency department visits, and (2) a control group in which the HIE database was not queried for any of the members in any emergency department visit. RESULTS: A post-propensity matching analysis showed that although the test group had more admissions per 1000 members overall (199 more admissions per 1000 members) than the control group, these admissions might have been more appropriate for inpatient treatment in general. The relative risk of an admission by the time of a first emergency department visit was 28% higher in the control group than the test group, although by the time of a second emergency department visit, it was only 8% lower in the control group. Moreover, test group admissions resulted in less time spent as inpatients, which was denoted by bed days per 1000 members (771 fewer bed days per 1000 members) and by average length of stay (4.27 days per admission for all admissions and 0.95 days per admission when catastrophic cases were removed). CONCLUSIONS: Based on these results, HIE availability in the care of patients presenting to the emergency department is associated with fewer inpatient hospital days and a shorter length of stay, even when catastrophic cases are removed from the analysis. Although many factors can play a role in this finding, it is possible that HIE promotion of more appropriate hospital admissions from outside of the emergency department is one cause. Such "indirect" value shows that the return on investment found by HIEs may even be greater than previously calculated. Additional study is warranted to further the business case for HIE investment for the various stakeholders who are interested in supporting HIE sustainability.
BACKGROUND: Health information exchanges (HIEs) have already demonstrated direct value in controlling the costs associated with utilization of emergency department services and with inpatient admissions from the emergency department. HIEs may also affect inpatient admissions originating from outside of the emergency department. OBJECTIVE: To assess if a potential association exists between a community-based HIE being used in hospital emergency departments and inpatient admissions emanating from outside of the emergency department. METHODS: The study design was observational, with an eligible population of fully insured plan members who sought emergency department care on at least 2 occasions over the study period between December 2008 and March 2010. Utilization data, obtained from medical and pharmacy claims, were matched to a list of emergency department utilizers where HIE querying could have occurred. Of the eligible members, 1482 underwent propensity score matching to create two 325-member groups-(1) a test group in which the HIE database was queried for all members in all of their emergency department visits, and (2) a control group in which the HIE database was not queried for any of the members in any emergency department visit. RESULTS: A post-propensity matching analysis showed that although the test group had more admissions per 1000 members overall (199 more admissions per 1000 members) than the control group, these admissions might have been more appropriate for inpatient treatment in general. The relative risk of an admission by the time of a first emergency department visit was 28% higher in the control group than the test group, although by the time of a second emergency department visit, it was only 8% lower in the control group. Moreover, test group admissions resulted in less time spent as inpatients, which was denoted by bed days per 1000 members (771 fewer bed days per 1000 members) and by average length of stay (4.27 days per admission for all admissions and 0.95 days per admission when catastrophic cases were removed). CONCLUSIONS: Based on these results, HIE availability in the care of patients presenting to the emergency department is associated with fewer inpatient hospital days and a shorter length of stay, even when catastrophic cases are removed from the analysis. Although many factors can play a role in this finding, it is possible that HIE promotion of more appropriate hospital admissions from outside of the emergency department is one cause. Such "indirect" value shows that the return on investment found by HIEs may even be greater than previously calculated. Additional study is warranted to further the business case for HIE investment for the various stakeholders who are interested in supporting HIE sustainability.
Authors: J Marc Overhage; Paul R Dexter; Susan M Perkins; William H Cordell; John McGoff; Roland McGrath; Clement J McDonald Journal: Ann Emerg Med Date: 2002-01 Impact factor: 5.721
Authors: Mark E Frisse; Kevin B Johnson; Hui Nian; Coda L Davison; Cynthia S Gadd; Kim M Unertl; Pat A Turri; Qingxia Chen Journal: J Am Med Inform Assoc Date: 2011-11-04 Impact factor: 4.497
Authors: William R Hersh; Annette M Totten; Karen B Eden; Beth Devine; Paul Gorman; Steven Z Kassakian; Susan S Woods; Monica Daeges; Miranda Pappas; Marian S McDonagh Journal: JMIR Med Inform Date: 2015-12-15