Revisiting sociotechnical systems in a case of unreported use of health information exchange system in three hospital emergency departments Mustafa Ozkaynak PhD 1 and Patricia Flatley Brennan RN PhD 2 1 Postdoctoral fellow, Industrial Engineering Worcester Polytechnic Institute, Worcester, MA, USA 2 Moehlman Bascom Professor, Industrial and Systems Engineering and School of Nursing, University of Wisconsin, Madison, WI, USA Keywords emergency departments, health information exchange, social shaping, sociotechnical systems Correspondence Mustafa Ozkaynak Worcester Polytechnic Institute 3009 Gateway Park, 60 Prescott Street Worcester, MA 01605-2661 USA E-mail: mozkaynak@wpi.edu Accepted for publication: 9 January 2012 doi:10.1111/j.1365-2753.2012.01837.x Abstract Background Health information exchange (HIE) allows clinicians to access patient level health care information. HIE can potentially improve patient care in emergency departments. Methods We present a selected piece of evaluation of an HIE implementation in three Midwestern emergency departments. Data were collected through over 210 hours of direct observations and short interviews with 13 clinicians. Results The results suggest that the usage rate of the HIE was low. Moreover, two ways of unreported use of the HIE system by clinicians was uncovered: (1) The HIE system was being used mostly for patients only with specific characteristics. (2) The information from the HIE system could be used to confront with the patients. Discussion This study provides a case of how social system may shape a HIE technology. In order to fully benefit from HIE, understanding organizational and social context during the HIE design and implementation is needed. Such an understanding will also allow us to identify and detail required additional resources and organizational interventions that will complement HIE such as a case management strategy. Introduction Health information exchange (HIEs) can benefit emergency departments (EDs), because ED patients may possibly have a history in other care delivery organizations [1]. Despite the impor- tant promises of HIE, using system logs, Vest et al. found that the HIE system was accessed for only 2.3% of all encounters in the ED [2]. Novel usage patterns were more likely for more complex (with more than one chronic health complaint) patients. The odds of HIE usage were also lower in the face of time constraints or when the patient was unfamiliar to the facility. More exploratory studies are needed to better understand the motivation to use HIE and how clinicians use the information retrieved from HIE systems. Patient care in EDs involves significant challenges that may directly and indirectly affect the use of HIE. EDs provide a wide range of medical care to patients with various health conditions. EDs were originally established for stabilizing patients and trans- fer to another setting for a specialized care [3]. However, the role of the EDs has expanded. Currently, EDs also provide primary care [4,5]. There has been an ongoing discussion on the level of care that should be provided in ED [6–8]. Lack of clarity of the level of services expected from EDs is a challenge in preparing effective clinical guidelines that will provide decision support to clinicians when they are working in challenging conditions with a high time pressure. Moreover, EDs are a victim of fragmented care. Lack of needed information requires clinicians make more subjective inter- pretation about patients in EDs. All these challenges can cause compromise the quality of care and lead to suboptimal practice performance and decision making in this highly demanding and sensitive environment [9]. The HIE can contribute to such com- promise under the moderation of social factors by providing only limited information about a patient. Harrison et al. developed a framework called Interactive Socio- technical Analysis to study health information technologies (HITs) [10]. The framework includes four components: (1) new informa- tion technology; (2) social system; (3) HIT-in-use; and (4) techni- cal and physical infrastructure. One of the five interaction types that they defined between these components is ‘Social system mediates HIT use’. Social shaping and reinterpretations about new HIT often lead to different uses and practices from those intended by HIT’s designers. An important reason why HIT-in-use so often diverges from designed HIT is that the original designs fail to reflect ongoing features of the work system, broader context, and Journal of Evaluation in Clinical Practice ISSN 1365-2753 © 2012 Blackwell Publishing Ltd, Journal of Evaluation in Clinical Practice 1