Date of Award

Spring 3-1-2012

Document Type

Dissertation

Degree Name

Doctor of Science in Information Systems

Department

Business and Information Systems

First Advisor

Omar El-Gayar

Second Advisor

Dorine Bennett

Third Advisor

Amit Deokar

Fourth Advisor

Surendra Sanikar

Fifth Advisor

Cecelia Wittmayer

Abstract

This study established and achieved three objectives which include development of a valid instrument for evaluation clinical reasoning performance with an electronic health record, an extension to TTF theory into the clinical domain and EHR context and confirmation of the on-going viability of TTF theory to explain the factors influencing individual performance. The motivation and rationale for the study is based on a clear need for a valid instrument that captures user evaluations for research and organizational purposes. The lack of a validated instrument for evaluating the impact of EHR use on clinical reasoning performance is another reason for revisiting the user evaluation construct. The study took place in a mid-size health system in the mid-western United States. The target population of the study included physicians, certified nurse practitioners and physician assistants with experience using an EHR. Sampling accurately represented the population of clinicians under study in this context. A 62.4% response rate was achieved for the survey. Semi-structured interviews were conducted to ensure that the theoretical constructs made sense to clinicians, and to identify any additional constructs that might be critical to TTF or task and technology characteristics. The interview process also identified potential problems with the utilization construct, and further suggested that clinicians did not view authorization as an important dimension of TTF. Pre-testing of the instrument assisted with clarification of the wording on survey questions, resulting in eleven changes to survey questions. Quantitative analysis was accomplished by means of exploratory factor and partial least squares analysis. The survey results suggest that for TTF, data quality, authorization, compatibility, and clinical informatics/IT relationship with users were important dimensions. Data locatability, production timeliness and system reliability all had significant but negative relationships with performance. Analysis of the survey data also suggested that task characteristics (task complexity, task uncertainty and task significance) effectively “moderate” the strength of the link between specific characteristics of EHR’s. The results give support to many dimensions of the theoretical model and raise some interesting questions of their own. For example, why do locatability and systems reliability not test well, or, are there other dimensions of task and technology characteristics that should be considered? Also, some of the hypothesized relationships between task and technology characteristics did not perform as expected. Indeed, these issues are excellent candidates for future research.

Download

Share

COinS