The "REMIND Trial" - Do Pillboxes Actually Work?

The Bottom Line:

Simply handing out 'Low-Cost' tools/devices (e.g. standard pillbox, pill bottle cap with a timer, pill bottle strip with toggles) to help patients remember to take their medications doesn't really seem to accomplish anything. 

The Background:

For anyone in health care, the topic of medication adherence is perhaps a constant consternation that many practitioners and organizations struggle with. It's a problem that has been pegged to be due to forgetfulness up to 60% of the time based on patient feedback. Nonetheless, the question has become, how do we help patients remember to take their medications then? CVS sought to tackle this problem with their recently concluded trial, published in JAMA Internal Medicine, entitled the REMIND Trial (Randomized Evaluation to Measure Improvements in Nonadherence from Low-Cost Devices).

The REMIND Trial - Synopsis

The trial was conducted from February 12, 2013, through March 21, 2015. A total of 53,480 patients were enrolled to be randomized into different intervention arms, including:

  • A pill bottle with toggles that can be selected on the day the medication was taken
  • A pill bottle cap with a timer that displays time since the last dose taken
  • A standard plastic pillbox (you know, the ubiquitous one handed out by so many organizations with their logo on it...)

The population selected to be included in the trial was quite stringent, with only patients who had medical plan sponsors continously for the previous 12 months included. With that being the case, the age group was limited to those between 18-65 (as those over 65 would have been [likely] Medicare patients). Eligible patients then had to meet one of the following three criteria:

  1. Filled between 1-3 oral maintenance medications for the treatment of cardiovascular disease;
  2. Another nondepression chronic condition (e.g. BPH, Parkinson Disease, Seizure);
  3. Or depression in the previous 12 months.

(Right away, we can point out that these are not very complex patients overall, and the rationale for why the limits of medications were chosen was due to the provided interventions themselves)

The patients selected to be included were screened for a history of nonadherence based upon identified medication possession ratio (basically tracking refill histories via insurance claims or documentation). Patients displaying an MPR between 30-80% were included in the trial. In total, 53,480 patients were identified to meet the inclusion criteria and were split into four arms of the trial (control vs one of the three interventions), then in March 2014 were mailed the intervention. I won't go into to much detail about the randomization of the groups and allocations, but it is worth a look if you are curious. 

Outcomes measured were the adherence of the patients based on their MPR after a 12-month follow-up period, with secondary outcomes being optimal adherence measurements in the population groups themselves (i.e. cardiac, chronic condition, depression). Patients were deemed optimally adherent if their MPR was greater than 80%.

So What did they find?

Not much. Basically, there was no meaningful statistical change in patients MPR split between all four groups, with the control group having 15.1% becoming optimally adherent, the pillbox group having 15.5% becoming optimally adherent, the digital timer cap group having 15.1% becoming optimally adherent, and the pill bottle strip group having 16.3% becoming optimally adherent.In comparison between the intervention groups, having a pillbox had a 10% higher odds of being more adherent. Lastly, no difference was found between any of the different patient groups. 

The Researchers Take on their Findings:

In this pragmatic, comparitive-effectiveness randomized clinical trial of more than 50,000 individuals who took up to 3 longterm medications to treat chronic conditions but were nonadherent to these therapies, 3 low-cost devices - pill bottle strip with toggles, digital timer cap, and standard pillbox - did not improve medication adherence. - REMIND TRIAL Authors

The authors laid bare the fact that their interventions really made no difference, and even the control arms demonstrated an observed 12-18% improvement in adherence. They stated that it is mostly the fact that these interventions are not meant for everyone, and should be tactfully used in those patients who were more likely to be nonadherent in the future. Follow-up surveys conducted in patient groups did not seem to demonstrate that patients mailed these interventions didn't necessarily give up on them, with up to 68% reporting using the devices 3-months from the start of the trial. Perhaps most dismally, the authors discuss the fact that these interventions may just not work. Rather, in order to work, they most likely need to be used in conjunction with additional support mechanisms. 

My personal take:

Nonadherence is a multifactorial problem that isn't just solved by throwing a patient a pillbox and hoping that they will suddenly become adherent. If that was the case, this problem would have been solved a long time ago. I like that the authors put some emphasis on the fact that many patients believe themselves to be adherent when based on MPR at least, it is quite the converse, which I believe is just the overall human element of being asked to take so many damned medications a day. The other item the authors mention is the involvement of pharmacists to help with adherence by being involved in patient care (which I wonder if CVS will integrate more in future as a more pliable intervention). 

Some items that I found interesting and that may be applicable to companies looking to tackle adherence was the issue with the pill bottle cap timers, which the authors believe may not have fit all pill bottles and led to misuse. The other item was that the pillbox (while not a major difference) may have performed better as it was more native to what patients already have used. In other words, going completely against the norm for innovation may not be intuitive for a lot of patients. 

I really think low-cost interventions can work, but only if healthcare professionals take the time to explain their purpose (in-person) and demonstrate how to use them. But this takes time, and in the hectic clinical environment of today, that may be hard to pull off (unless we have dedicated staff and materials for it). If someone was to pursue such a program (such as in a clinic or hospital) I would say the time would be best be used in those that are predicted to become nonadherent in the future or demonstratable nonadherence based on some objective data. 

Other considerations regarding this trial, which I think are more telling, is that these were really not complicated patients in the grand scheme of things. These are not my heart failure patients on 10-20 medications a day, or a patient with advanced chronic conditions requiring more than just oral medications. I question the impact of these interventions on such complicated patients, and what we need to do to even approach that population, which I think these currently investigated products really do not meet. 

Thanks for the read, and as always, I welcome your comments below!

Article Link

Choudhry NK, Krumme AA, Ercole PM, et al. Effect of Reminder Devices on Medication Adherence: The REMIND Randomized Clinical Trial. JAMA Intern Med. Published online Feb 27, 2017.