Evaluating a web-based health risk assessment with tailored feedback: what does an expert focus group yield compared to a web-based end-user survey?

BACKGROUND: Increasingly, Web-based health applications are developed for the prevention and management of chronic diseases. However, their reach and utilization is often disappointing. Qualitative evaluations post-implementation can be used to inform the optimization process and ultimately enhance their adoption. In current practice, such evaluations are mainly performed with end-user surveys. However, a review approach by experts in a focus group may be easier to administer and might provide similar results. OBJECTIVE: The aim of this study was to assess whether industrial design engineers in a focus group would address the same issues as end users in a Web-based survey when evaluating a commercial Web-based health risk assessment (HRA) with tailored feedback. METHODS: Seven Dutch companies used the HRA as part of their corporate health management strategy. Employees using the HRA (N=2289) and 10 independent industrial designers were invited to participate in the study. The HRA consisted of four components: (1) an electronic health questionnaire, (2) biometric measurements, (3) laboratory evaluation, and (4) individually tailored feedback generated by decision support software. After participating in the HRA as end users, both end users and designers evaluated the program. End users completed an evaluation questionnaire that included a free-text field. Designers participated in a focus group discussion. Constructs from user satisfaction and technology acceptance theories were used to categorize and compare the remarks from both evaluations. RESULTS: We assessed and qualitatively analyzed 294 remarks of 189 end users and 337 remarks of 6 industrial designers, pertaining to 295 issues in total. Of those, 137 issues were addressed in the end-user survey and 148 issues in the designer focus group. Only 7.3% (10/137) of the issues addressed in the survey were also addressed in the focus group. End users made more remarks about the usefulness of the HRA and prior expectations that were not met. Designers made more remarks about how the information was presented to end users, quality of the feedback provided by the HRA, recommendations on the marketing and on how to create more unity in the design of the HRA, and on how to improve the HRA based on these issues. CONCLUSIONS: End-user surveys should not be substituted for expert focus groups. Issues identified by end users in the survey and designers in the focus group differed considerably, and the focus group produced a lot of new issues. The issues addressed in the focus group often focused on different aspects of user satisfaction and technology acceptance than those addressed by the survey participants; when they did focus on the same aspects, then the nature of issues differed considerably in content.

Data reduction and representation in drug discovery.

Pre-clinical drug discovery relies increasingly on huge volumes of inter-related multivariate data. To make sense of these data and enable quality decision-making based on this plethora of information they must be presented in an interpretable form. Reducing the dimensionality of the data often leaves a data set that is too complex to interpret readily, so intuitive visualization methods are needed. Bioinformatics has provided much of the impetus for visualizing complex data, the cheminformatics community has been aggressive with the data-reduction problem. The increasing appreciation of the inter-related multifactorial nature of pre-clinical drug discovery makes visualization a burgeoning and active field that spans biosciences, mathematics and visual psychology.