Date of Award
Doctor of Philosophy (PhD)
Becker, Robert H
Hamilton , Alfred A
Smith , Robert W
Ulbrich , Holley H
To understand problems related to a policy of implementing a lifelong longitudinal electronic health record (EHR) more fully, this dissertation examines compliance with changes in policy over time. We analyze drivers of compliance with a required electronic medical record (EMR) by hospital clinicians completing the records for deployed service members. This study examines compliance as an outcome of principal-agent (PA) relationships, with the EMR modeled as the measure of success between one level of bureaucratic principal (i.e. medical command) with control over the necessary mechanisms in order to ensure compliance of agents (i.e., medical professionals).
Policy compliance is operationalized in three ways: 1) the total number of inpatient EMRs completed; 2) the date on which new records are started; 3) the average number of days to close an inpatient EMR. For each of these dimensions, 'EMR' refers to what clinicians categorize as treatment for a disease non-battle injury or battle injury.
The first independent variable concept for this study is change in the level of information asymmetry between principal and agent, operationalized as the time a superordinate medical command (MEDCOM) is directly in control over hospitals. The second concept is the alignment of goals in order to reduce goal conflict. This is operationalized as a technology upgrade allowing hospital EMR to be used for both implementing the larger EHR as well as in providing real-time clinical notes necessary for the care of patients being evacuated to the next level of medical care. Finally, the concept of principal control mechanisms are operationalized in this study as the introduction of increased monitoring policy and sanctions at the clinician level during hospital transition periods.
We use quantitative data in the form of completed electronic medical records and utilize a quasi-experimental research design. The specific design chosen for the study is the interrupted time-series. The population for this study is all United States military service members seen as inpatients in deployed military hospitals directly supporting Operation Iraqi Freedom. The study period is 105 weeks. Overall, this research meets the objectives outlined in Chapter 1 (Introduction). The study examined two important questions regarding clinician compliance with completing EMRs for deployed service members. First, this study addressed if there was a change in policy compliance over time. By conducting an analysis of policy interventions, we established changes in policy compliance. Compliance was defined as the fluctuation in inpatient records started, records completed, and changes in the average time to complete records. Secondly, this study examined what factors influenced the performance of hospital clinicians and how significant these drivers' impact was on record completion. The analysis consisted of graphing the changes over time and examining changes that were most likely due to policy interventions. We further analyzed the changes over time utilizing ANOVA and least squares regression.
The results supported many of the hypotheses. Technology upgrades not only led to greater completion rates but also reduced the amount of variation in records completed week to week. The introduction of the monitoring policy also increased both record completions and records started. Finally, sanctioning showed the greatest impact on completing records.
This research is important for four reasons. First, this study provides a method to analyze policy implementation at different levels within one federal department. Second, this research enhances the body of knowledge in the inter-disciplinary evaluation of policy implementation. Third, this dissertation examines the role of specific control mechanisms, namely monitoring and sanction, not previously reported in the EHR implementation literature. Finally, this study provides real-world implications for implementing EHR policies in deployed environments.
This study determines that the time a MEDCOM is in charge, technology upgrades, monitoring, and sanctions do have an effect on policy compliance but are reliant on the measurement of compliance. As an example, technology upgrades significantly increase the number of EMR completed at hospitals, but they are not statistically significant in increasing or decreasing the number of new encounters started at the hospital. In addition, patient categories influence the significance between the independent and dependent variables.
Mellott, Mark, "DEPLOYED ELECTRONIC MEDICAL RECORD POLICY COMPLIANCE: AN INTRA-DEPARTMENT PRINCIPAL-AGENT PERSPECTIVE" (2010). All Dissertations. 566.