The Epimetrics Epicenter platform and the robust tools upon which it is built give us the capability to create interesting and novel special applications running within a collaborative environment. These applications all leverage our particular interest in collecting and analyzing data. Here are some examples from sites done to date that illustrate the kinds of custom tools we can deploy.

Custom Tools

Health-related Quality-of-Life Instruments

Health-related Quality-of-Life (HRQOL) instruments are basically questionnaires that have been validated against other standard clinical measures so that HRQOL scores correlate with — and predict — clinical outcomes. This means that instead of subjecting patients to expensive and potentially dangerous tests, their clinical course can be tracked by simply administering the questionnaires. Of course this kind of technique doesn’t treat people or deal with emergencies, but it is ideal for long-term monitoring because it can uncover early signs of problems when they’re easier to deal with.

We’re worked for a long time with one of the world leaders in the cardiovascular realm of this field, Dr. John Spertus. The latest collaboration produced the web site MyHealthOutcomes, which is a “personal health record” that allows patients to track several HRQOL measures that John has devised: the Seattle Angina Questionnaire, the Kansas City Cardiomyopathy Questionnaire, and the Peripheral Artery Questionnaire.

The MyHealthOutcomes site is basically a community site like any Epicenter site, with a number of tools that allow patients to complete the various questionnaires and track their trends. The site enables patients to share these results online with their physicians if they choose to do so. It also generates clear printable summaries of results that patients can take with them to their appointments. You can see this system in action if you try out the demos.

Floristic Quality Assessment

Floristic Quality Assessment (FQA) is a technique of measuring the ecological health of a particular biological community by calculating a “floristic quality index” from the plant species found and their ecological significance (called the “coefficient of conservatism”). One-time measures give a snapshot about the status of a parcel of land, while serial measures over time reveal the incursion of invasive species, the results of weed-abatement efforts, and the like.

Ordinarily FQAs are done by hand or with a special-purpose spreadsheet, but for the Remnant Prairies Association in the midwest, we created an online data management system that permits landowners to define regions of interest on their properties using Google Map tools and then to track serial FQIs for these parcels. The system then makes it easy for landowners to record the species they’ve identified, optionally importing prior lists to reduce the amount of data entry. Once all entries are made, the system automatically calculates the many parameters that comprise an FQA. You can see more of this system at this demo location.

For the Remnant Prairies group we also created a plant database that feeds into the FQA process and a database of organizations to assist building links with other groups in other states.

License Management

Within the CV Outcomes web site — another collaboration with John Spertus that runs his multi-center research program — we built a system that handles licensing of his various HRQOL instruments. Previously John had tried to accomplish this by hand, sending out three-ring binders with the information, but as the use of his questionnaires exploded (now involving about 1500 studies at nearly 10,000 sites and nearly 200,000 patients) this obviously did not scale.

The system we developed provides tools for interested parties to set up an account for their organization, specify the particulars of their project and the specific instruments and other items they need, and submit their requests online. The system then allows John’s admins to process the requests, generate agreements and invoices, and track the status of each project as signed agreements are returned and invoices are paid. Each step generates automated emails to keep everyone apprised of the current status of a project, and automatic reminders go out when projects near the end of their terms and renewal is due. John and his staff can review in real time cumulative stats on all aspects of the use of his instruments via a compact dashboard interface.

Species Tracking

A Bay Area environmental advocacy organization called Wild Equity Institute runs a biennial project called “The Big Year”. This focuses attention on threatened and endangered species through educational materials and a series of proctored trips during which participants endeavor to see the species of interest and perform some activity to benefit them. Participants then record their sightings and activities. To foster enthusiasm, the Big Year compiles totals and announces winners.

As part of the Wild Equity web site, we created a system that enables the Big Year staff to create new “big years”, manage the selected species for the year and the activities to be done, schedule all the trips and map them to the species to be seen on each trip, and collect from participants who are logged into the site their data. The system then provides real-time personal results as well as collective results (a "leaderboard") to all participants.


Q-Methodology is a semi-obscure form of preference assessment used primarily in the social sciences, notably public relations and psychology. The “Q” process consists of having subjects make a two-level sorting of statements about some topic. First they separate them into three categories “Agree”, “Uncertain”, and “Disagree”, and then they rank the statements in each category by degree. The process defines particular numbers of slots for each ranking that constrain subjects’ choices. The results then undergo factor analysis to discern patterns in the data.

We previously developed a complete online system for conducting Q studies that solved the main logistic problem — insuring that subjects make the proper pattern of selections for each portion of the sort — and handled all the enrollment tasks — email invitations/reminders — plus the data reporting/downloading. We have now rewritten Q-Assessor from the ground up to leverage Rails capabilities and to add real-time analysis of study results, among many other important features. Q-Assessor is still in a stealth-mode beta phase, but bookmark it if you’re interested in seeing how to do Q the modern way!

Do you have ideas for unusual projects you’d like to bring to life? Let us know and we’ll discuss them!

Created: September 15, 2009 12:48
Last updated: June 20, 2010 15:06


There are no comments so far.

If you signup and login, you can post comments.