Results of a pilot study of a mobile app meant to improve medication compliance among chronically ill, low-income patients are mixed but encouraging, the pilot's facilitators announced at an event MobiHealthNews attended in Washington, DC this week.

The study, funded by a grant from Qualcomm’s Wireless Reach initiative and conducted at the George Washington University Medical Center, found that a small group of Medicaid patients with chronic hypertension were quite willing to receive and act on medication reminders sent to their cell phones. But actual adherence to prescribed medication regimens was harder to attribute to the technology.

In a seven-month study that ended in August, average user satisfaction was 4.3 of a possible 5 for the Pill Phone, an FDA-cleared medication management app from Del Mar, Calif.-based Vocel. Researchers report similarly high scores to questions about whether the Pill Phone has helped patients keep better track of their meds and whether they would use a similar app in the future.

The Pill Phone provides alerts to Web-enabled mobile phones—not necessarily smartphones—with text and images. Patients must indicate whether they’ve taken the drug or skipped a dose, or they can choose a “snooze” function that delays the reminder. Content comes from The Pill Book, a guide with information and FDA-approved pictures of 1,800 commonly prescribed medications.

The GWU internal medicine, renal and cardiology clinics identified 350 District of Columbia Medicaid patients with hypertension by mining the Allscripts EHR used by the faculty practice, and eventually got 50 participants for the study. Patients, all of whom had been prescribed at least two drugs for their condition, had four clinic visits, first to establish a baseline, then to measure progress with and without the Pill Phone app.

Wireless Reach teamed with low-cost wireless carrier Cricket Communications to supply 3G-enabled “feature phones” to program participants at no charge. More than half of the study subjects, with a median age of 53, had diabetes in addition to hypertension, and 42 percent were smokers at the time the study began. Participants were on an average of 8.1 medications. Half had blood-pressure monitors at home, but 37 percent of those used their monitors less than once a month.

Of the more than 15,000 scheduled dosages, just 94 came back as skipped and patients hit “snooze” 134 times. Researchers didn’t see a single instance of a patient initiating Pill Phone changes.

Fourty-eight of the 50 test subjects completed the study, and only three of the 48 turned off medication reminder alarms, including one whose app stopped working. Adherence, as measured by pharmacy refill rates, topped 60 percent during the early weeks of the three-month Pill Phone usage period, but did drop below 55 percent in the final three weeks, based on patient-reported data.

More than 80 percent filled their medication orders as prescribed during the time the Pill Phone was on, but the refill rate returned to the baseline of just over 70 percent after the app was taken away. However, the effort did show measurable results in terms of controlling blood pressure.

In the study period, 17 patients reported at least one ER visit and 7 had to be hospitalized. There was no significant difference in Pill Phone usage among those needing to go to the ER or be admitted and those who did not.

“This pilot program had high acceptance, sustained use, patient satisfaction and retention rates in this high-risk population,” Dr. Samir Patel, an internist in the GWU Division of Renal Diseases and Hypertension, said Wednesday at an event on the GWU campus in the nation’s capital.

“The importance of the Pill Phone research study is that it offers a model for disease self-management that can be applied to at-risk communities,” said Dr. Richard Katz, director of cardiology at The George Washington University Hospital.

However, Patel noted the limited nature of the research. “This is a small study. We really can’t make a lot of inferences,” he said. “Longer studies will be needed.”

All but two of the participants were African-American, and only 21 percent were employed full-time. Just 17 percent of subjects had college degrees and nearly 44 percent had health literacy at or below an 8th grade reading level, but they generally found the Pill Phone app user-friendly.

“It was really easy to use because I’m not good with the cell phone,” said Dolores Smith, a study participant and herself a medical assistant. “I usually have to turn to the kids.”

Smith said it was easy for her to remember to take her first pill of the day when she first woke up, but it was easy to forget to take later dosages.

“Adherence really plummets when you take that second pill of the day,” Katz noted.

The research team found no significant differences among patients based on employment, education, age, gender or total number of medications prescribed.

Meantime, Vocel continues to refine its Pill Phone app. “Wouldn’t it be nice if the clinic had real-time information on adherence?” wondered Vocel CEO Carl Washburn. He also would like to add real-time monitoring so the clinic can intervene as necessary.

Katz, for one, said there is plenty of room for improvement. “We’re really early into this,” he said. “This is the [version] 1.0 of m-health.”