Compliance monitors: A place in a program?
One use: To identify nonadherent patients
According to what he calls the "500 year plan," his struggle to persuade the medical establishment to consider the role of medication monitors is right on schedule, says Thomas Moulding, MD, a clinical professor of medicine at Harbor/ UCLA Medical Center in Torrance, CA,
A physician who’s been championing such devices for over 30 years, Moulding is joking about the time frame, but serious about the monitors. Although they may or may not ultimately prove to be useful in helping patients to improve their pill-taking, the devices definitely can be used to find out how many pills a patient actually takes. And that, Moulding says, means they can be used to reliably separate adherent patients from those likely not to be adherent.
"It’s silly to put everyone on directly observed therapy [DOT] when there is this alternative," Moulding says.
"We can definitely find the patients who are noncompliant with these devices," agrees John Fowler, co-president of Anderson Clinical Technologies in Rolling Meadows, IL. Fowler’s electronic blister-pack monitor records the date and time each pill is dispensed, and can be programmed to continuously feed the captured data into a computer.
"That way, the health care professional sees that Joe Jones is missing three days of his doses," says Fowler. The device Fowler’s company markets right now, EXACT CM Compliance Monitor, is being used chiefly for drug trials where direct observation isn’t needed, but it’s beginning to find another use as well: as a test of psychographic "instruments" developed to predict compliance. One place it hasn’t been used is in TB control programs, he adds.
Moulding’s prototype monitor is somewhat similar to Fowler’s instead of a blister pack, it’s a small box (about the size of a bar of soap) that dispenses pills. The event is recorded when an electric eye "sees" that a light beam has been broken; a digital watch notes the time; and the data are recorded on a computer chip.
Research on the EXACT CM shows patients fall into three categories, no matter what their regimen, Fowler says. "About 50% to 60% of patients are compliant, and take at least 80% of their pills. Ten percent to 15% are not compliant, no matter what you do. And 30% to 40% are partially compliant."
Unlike Moulding’s device, which features optional add-ons to prompt the patient (with beepers or visual displays), Fowler’s monitor doesn’t attempt to change patient behavior, just to record it. Whether having a beeping medication monitor in the house will improve pill-taking behavior isn’t yet clear, Fowler says. "The jury’s out on prompting devices. They may help the forgetful patient, but what about the others?"
There isn’t a lot of data available on whether using a medication monitor affects behavior, Fowler concedes. The little information that does exist suggests that monitors simply do a good job of recording the behavior the patient would have exhibited anyway, device or no device, he says.
At the Centers for Disease Control and Prevention in Atlanta, Rick O’Brien, MD, chief of the clinical research branch of the Division of Tuberculosis Elimination, is politely skeptical on the subject. "With the current emphasis both in this country and globally on DOT, having something like this is less than essential," O’Brien says. "Yet I certainly understand the potential for its role in a program that gives selective DOT, and I think it would be interesting to evaluate it."
Before such devices are used, some questions deserve study, Moulding says. They include:
• Can medication monitors be used to identify patients likely not to adhere to their regimens? Assuming a monitor shows a patient is demonstrating good or bad compliance early on in treatment, is that a reliable predictor of the same sort of compliance later as well? Various studies have shown there is a positive correlation, Moulding says, but to what degree needs to be clarified.
• What are the implications for treatment success or failure when the medication monitor shows the patient is not doing a good job of taking his pills? How much irregular drug ingestion can take place before treatment will fail? Does it work to prolong treatment to compensate for missed doses?
• What course of action should be taken when the monitor reveals poor adherence? Some studies have shown that making an effort to better educate such patients can improve their compliance, Moulding says. Other alternatives, logically, would include placing the patient on DOT.
• Finally, will the medication monitors help patients remember to take their pills and thus improve compliance?
Whether patients will "cheat" by taking out pills but not swallowing them, worries neither Moulding or Fowler. "Sometimes patients dispense all the medication right before they come to the clinic," Fowler says but the monitor duly records that fact.
Patients almost never dispense the pill and then simply throw it away, both men say. In one study, only two urine samples out of 158 (taken from 58 patients with good monitor compliance records) were negative for drugs the patients were supposed to have been ingesting, Moulding says.1
Moulding has been tinkering with pill calendars and electronic devices since 1957, when, as a resident physician at the University of Michigan Hospital, he contracted a mild case of pulmonary TB. Sentenced by his own doctor to nine months of bed rest in a hospital, Moulding found he was desperately bored after three months, and talked his physician into letting him go home to take his medication unobserved.
Empty bottle did the trick
Feeling guilty about pulling rank to get special treatment, he determined to make sure he didn’t forget and miss part of his daily treatment ration of nine pills, taken three times a day.
His solution was to "invent" what might have been the first medication monitoring device: an empty bottle, which he filled with pills at the start of each day. Halfway through the day, if the bottle was empty, that meant he’d forgotten to fill it that morning; by nightfall, any remaining pills were those still left to be taken.
The device worked, and one of Moulding’s lifelong interests was launched.
Anderson Clinical Technologies rents EXACT CM Compliance Monitors at a cost of between $50 and $100 per patient per month (assuming a pharmacy does the packaging and loads the monitors), plus a one-time fee of between $100 and $200 per patient to cover the transfer of data.
Moulding’s Container Type Medication Monitor should be somewhat cheaper once it hits the market, but is not yet commercially available, he says.
For more information, call Anderson Clinical Technologies at (847) 392-9190; fax: (847) 392-9194; or Thomas Moulding, MD, at (310) 375-5359.
Reference
1. J Kilburn, R Beam, H David, et al. Reagent-impregnated paper strip for detection of metabolic products of isoniazids in urine. Am Rev Resp Dis 1972; 106:923.
Special Report: Directly Observed Therapy
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content