Abstract 85: A Pilot Population-Based Outcomes Study using a Health Information Exchange: Preliminary Feasibility Results
Introduction: Post-stroke outcomes are traditionally determined by direct, follow-up interview of stroke victims. This approach is expensive, time consuming and can lead to a biased cohort, all of which limit the ability to conduct population-based outcomes studies. We evaluated the feasibility of facilitating outcomes assessment using a Health Information Exchange (HIE), a regional network for sharing healthcare data among providers. We hypothesized that HIE-facilitated cohort identification with phone-call follow-up would enable more efficient population-based outcomes determination compared to traditional approaches.
Methods: We identified all regional ischemic strokes during 2010 by screening ER/hospital admissions for stroke cases, first using a computerized algorithm that filtered by selected key words and zipcodes for potential strokes, then using human-screening of those records to identify actual stroke cases. Cases were then enrolled into one of two cohorts in a non-random fashion. First, 500 cases were assigned for “hot pursuit” and in-person interview (traditional cohort). Those cases not enrolled in the traditional cohort for logistical reasons were assigned to have outcomes assessment by HIE-data review and phone call interview (HIE cohort) at 3 months only. The feasibility of contacting patients and assessing outcomes were the primary endpoints of interest for the HIE cohort; time spent and cost were also assessed but will not be discussed herein.
Results: There were 17,671,258 HIE messages about 1,967,415 unique patients seeking care at hospitals within the study region in 2010. The computerized algorithm identified 5,082 potential strokes for human review. In-person interviews were attempted on 793 subjects; 104 died or entered hospice; 173 refused and 516 agreed to participate in the traditional cohort. Immediate cooperation rate was 75%. Of 516 participants, 408 completed a 3-month follow-up interview. For the HIE cohort, 581 subjects were identified for potential phone interviews. 491 calls were placed, and 320 were reached; 204 completed the interview while 116 refused. For 3 month phone calls, the cooperation rate was 52% and the completion rate was 64%. Data sufficient to determine outcomes were similar using both approaches.
Conclusions: These results suggest that using HIE-data and phone-call follow-up for outcomes determination is feasible and comparable to hot-pursuit and in-person interviewing. The use of the HIE greatly facilitated both processes. While basic data collected appear similar, further study is needed to: 1.) fully assess for any differences in the quality of outcomes determination, and 2.) determine the feasibility of performing a truly population-based outcomes study using either approach or a combination of the two.
- © 2012 by American Heart Association, Inc.