Understanding how social influence and social networks affect the implementation of an electronic medical record system

Health information technology (IT), particularly in the form of electronic medical records (EMRs), is increasingly recognized as an important tool for improving patient safety and quality of care. It is widely believed that EMRs, defined as computerized medical information systems that collect, store and display patient information, will lead to major health care savings, reduced medical errors, and improved health. However, many health care organizations have not been able to realize the intended benefits of EMR systems because of failed implementation efforts, in which health care practitioners use the practice less frequently, less consistently, or less assiduously than required for the potential benefits of the system to be realized. Theories of social influence, such as social information processing theory and social learning theory, suggest that social influence may play critical role in enabling – or hindering – the implementation process. However, there is a paucity of empirical research that examines whether, and if so, how health care practitioners influence one another's acceptance of new technology. In response to this pressing gap in the literature, this dissertation draws on theories of social influence and network theory to inform an empirical investigation of the role of social networks (i.e., a set of actors connected by a set of social ties) in influencing health care practitioners' beliefs and use of EMRs. Employing a mixed methods approach that leverages both quantitative (i.e., pre- and post-implementation surveys) and qualitative (i.e., interviews and observations) methods, this dissertation uses primary data collected from six clinical units of a large academic hospital that implemented a new EMR system. The first paper examines the effects of network beliefs (i.e., the mean beliefs of an individuals' social network) on individual nurses' use of the new EMR system. Using survey and EMR usage data collected 3-5 months after the system installation, I find that network beliefs about ease of use, but not about usefulness, are positively associated with individuals' EMR use. In the second paper, I use survey data collected before and after the system installation to examine: (1) the effect of network beliefs on nurses' beliefs about the usefulness of the system; and (2) the potential moderating effect of network churn (i.e., the change in composition of an individual's network cause by the entry of new ties and exit of existing ties). The findings suggest that: (1) networks influence individuals' beliefs about the usefulness of technology; and (2) less network churn in the form of fewer tie deletions enhances the effect of network influence, whereas network churn in the form of tie additions has no effect on network influence. The third paper examines how health care practitioners trained as super users (i.e., nurses who receive extra training on the EMR system so that they can provide frontline support to their peer users) influence others. Through a comparative case study design, I identify the mechanisms of influence used by super users on two clinical units that differed with respect to how super users were chosen (i.e., volunteered vs. selected by managers). On the unit in which super users volunteered, super users were more proactive, provided more comprehensive explanations for their actions, used positive framing, and shared information more freely. Taken together, the analysis in this dissertation provides evidence of the significant effect of social networks on health care practitioners' beliefs and use of EMRs (Papers 1 and 2) and identifies behaviors used by health care practitioners to influence others (Paper 3). The implications of this work include contributions to network theory and models of technology acceptance as well as practical insight into how to leverage social networks to enhance individuals' beliefs, and ultimately, use of EMRs.