JAMIA: Web-based diabetes mgmt tools mostly ineffectivefor now
 |
“Existing diabetes websites have wide variations in the quality of evidence provided and offer didactic information at high reading levels with little interactive technology, social support or problem-solving assistance,” wrote lead author Catherine Yu, MD, who is affiliated with the University of Toronto and St. Michael’s Hospital in Toronto, and colleagues.
“Similarly, although healthcare providers increasingly use online resources for patient care, the volume, breadth, editorial quality and evidence-based methodology upon which they were developed are highly variable,” she continued.
Researchers searched medical literature databases Medline, EMBASE, CINAHL and the Cochrane Central Register of Controlled Trials to locate web-accessible tools for evaluation using the following search terms: diabetes, hypertension, smoking cessation, weight reduction, online, computer-based and internet. Additionally, researchers entered preselected phrases into Google’s search engine and chose the first 30 results from each search that met their definition of a web-accessible tool for diabetes management.
Based on their search results and their criteria for inclusion, researchers identified 92 web-accessible tools and 57 corresponding studies to include in their review. According to the report, the 57 studies of web-accessible tools differed in terms of participants, settings and outcomes. “Although there were a number of studies with positive results, these results must be viewed with caution because of concerns about the reported study designs,” the authors wrote.
Researchers developed a rating system for the clinical usefulness and sustainability of tools using the framework from the literature they reviewed and input from experts. They graded web-accessible tools on a scale of zero to five, with a score of zero denoting “not clinically useful” and a score of five denoting “clinically useful answers are available most of the time, and are easily accessible and readable within a few minutes.”
The usability of web-accessible tools was graded according to a set of 27 criteria. Researchers graded each tool against each usability criterion with “yes,” “no” or “not applicable.”
With respect to the tools, researchers determined that 25 percent had easily accessible, clinically useful answers most of the time, 6 percent were free of usability errors and 60 percent had three or more usability errors, which commonly included limited use of visual elements and lack of intuition in navigation.
Additionally, researchers suggested that high attrition rates, or inconsistent use of web-accessible tools, limited their ability to produce outcomes.
Based on their results, Yu et al suggested that future research consider the correlation between consistent use of a web-accessible tool and patient outcomes, and mechanisms to minimize website attrition through the use of interactivity and feedback.
“Web-based tools have the potential to improve health outcomes and complement healthcare delivery, but their full potential is hindered by limited knowledge about their effectiveness, high prevalence of usability errors and high attrition rates,” Yu and colleagues wrote.
“A development and research agenda should include: developing strategies to reduce website attrition in order to maximize clinical outcomes; standardizing website quality indicators; and transparent reporting of these indicators in order to allow patients and clinicians to make informed decisions about website choice,” they concluded.