Talk to the palm

Clinical trial research is discovering the value of
using handheld electronic patient diaries
to outwit human error

By Elsie Wagner

"How do you feel on a scale of one to five?" Self-reported answers to clinical-trial questions are fast shifting from paper-based to electronic data capture, using personal digital assistant (PDAs). Thanks to their advent, particularly those tools developed by Palm, e-companies are lobbying for e-source collection as the ideal way to get faster and more accurate results.

Medical research -- usually quick to embrace high-tech advances -- lags behind other industries, such as banks (eg, ATMs) and governments (eg, online tax returns), in using paperless models. Asking people to fill out forms in pen and ink is notoriously slow, painstaking, error-prone and expensive. In the clinical-trial world, paper diaries have long reigned as essential tools to report assessment of response to therapy. But this methodology is riddled with human drawbacks: patients often fill in a week's worth of diaries in the car just before seeing the doctor, check the wrong boxes or miss questions, transpose numbers when writing in dosages and don't record times properly -- or fabricate them. Accuracy can then be further compromised when the already cumbersome information is keyed manually into a database.

To address this inherent flaw, the PHT Corporation (www.phtcorp.com), based in Charlestown, Massachusetts, teamed up with the Palm Solutions Group to create LogPad, an electronic patient diary (EPD). Each participant in a study is given a handheld to reply to questions. Subjects have to complete all questions on each screen before moving on to the next, which removes the possibility of incomplete answers. Each LogPad is tailored to the specifications required by a trial.

FEATURES GALORE
These EPDs are loaded with features to minimize compliance problems: subject-reported data are not only time-stamped at entry, but the device also has preset reminder alarms to cue its carrier when it's time to reply; any values entered that vary significantly from baseline are prompted for correction (eg, inputting a weight of 310 lbs instead of 130); they can be equipped with two-way communication so a researcher can access a patient directly -- or consult and download the data in real time. Unlike its paper counterpart, an EPD transmits results directly to a central server for data processing, eliminating the fine art of deciphering handwriting.

Even given that the EPD can correct for these variables, how does it hold up against the paper test as a viable way to record clinically relevant information? A study published in the May 2002 issue of Pain, written by Dr RN Jamison from Harvard Medical School, compared the accuracy of pain reports using the validated visual analogue scale (VAS) in 24 healthy volunteers. Using either the usual paper test or a VAS converted to a Palm Pilot IIIxe (said EPD), subjects were asked to rate levels of pain triggered by cognitive or sensory stimuli from "none" to "worst possible," and to mark an anatomical figure with "where does it hurt?" After calculating 2,016 stimulus variables, the team found virtually no difference: electronic data collection was at least as reliable as using paper.

LOWERING COSTS
Pharmaceutical companies and academic institutions alike are looking for ways to cut costs by redesigning the clinical-trial process for increased efficiency and faster decision-making. According to a 2003 white paper released by the Gantry Group, an independent strategic intelligence firm (www.gantrygroup.com), the price of developing a new drug or biologic and bringing it to market averages $802 US million and takes 10 to 15 years, with the drug-testing stage alone accounting for over eight years from lab to human.

EPDs can help slash costs substantially by lowering patient reporting times from an average of 40 days in a study period to only two to three days. With a minimal lag time between

a patient's entry and the researcher's access to it, the decision whether or not to kill or continue the trial can be determined much earlier: sponsors usually wait months or years after the last patient is out the door while the data is being processed, especially true of multicentre, international trials.

GROWING IN POPULARITY
Increased accuracy of mathematically complex recorded data also translates into improved statistical power, which in turn may allow a trial to be conducted with 12% to 20% fewer subjects --and given the cost of about $6,000 US to enrol a single patient, savings really add up.

Adoption of the technology is gaining momentum, and is currently being used in over 75 clinical trials worldwide. In Driving the Paradigm Shift, released in November 2003, PHT reported on the numbers of users signed onto this new technology. These numbers included more than 40,000 patients, at just under 6,000 sites; in 27 languages, in 39 countries; and counted among its clients nine of the world's top-10 drug research and development companies. To underline the value of the EPD strategy, PHT cites a sample case of a "US based biotechnology company" which "needed a more efficient and statistically reliable way to deploy a complex measurement tool for an international Phase III study."

The disease is Crohn's; the measurement of symptom severity vs. experimental drug being used is the Crohn's Disease Activity Index (CDAI). The hurdles: the CDAI not only requires analyzing over 60 fields to supply statistical strength in over 300 patients at 50 sites in Europe and the US, but this trial also insists on collecting quality of life and other information on irritable bowel syndrome in the same subjects, often comorbid with Crohn's. In previous phase studies, scientists at the sponsoring company had been frustrated by the huge delays caused by wrestling with the task of cleaning up self-reported results contaminated with human error. Solution: perform the next phase using EPDs. The LogPads do all the computation for the index scales and there aren't any paper forms to manually monitor and fix. At last report, the study is still in progress.

There's resistance to using EPDs, but not where you might expect it. Surprisingly, it's the participants in clinical trials who overwhelmingly say they prefer using handheld gadgets to filling out paper questionnaires. Instead, the use of PDAs is challenged by some researchers, who think it's an "inflexible" process. Promoters believe that a psychologic adjustment to change is the culprit, and that it may take many years before these clinicians accept it as easily as most Canadians have, say, online banking.