Every clinical trial participant with poor medication adherence will have their own set of complex reasons for not taking their medications as prescribed – meaning there is no one size fits all solution.
Instead, sponsors should take a data-driven approach that utilizes focused feedback, guides individualized interventions, and has been shown to be capable of improving patient adherence rates by up to 50%.1
A Stubborn Challenge
Poor patient adherence is a significant, long-standing problem in clinical trials. Studies have shown that by day 100, 20% of participants have stopped following the dosing protocol, and a further 12% have sub-optimal adherence.2
The consequences are obvious. It negatively impacts safety, data quality, data integrity, and research outputs. It leads to underestimations of drug efficacy and drains study power, which can result in the need for costly re-recruiting processes or even trial failure.
Researchers have been highlighting the issue for decades, but it has proved to be a stubborn challenge, in large part due to its complexity. The reasons for suboptimal adherence in clinical trials are multiple. Any number of factors, including forgetfulness, concerns about side effects, perceptions that the drug confers no benefit, or the burden associated with clinical trial roll-out, can act as barriers.
The literature tends to agree that the reasons broadly fall into the two categories of ‘intentional’, or related to a lack of motivation to take the regimen, and ‘non-intentional’, linked to a lack of resources.3 In truth however, there is usually a great deal of interplay between the two.
Traditional methods of patient adherence monitoring have proved to be less than ideal. Approaches such as pill count, self-report, or pharmacokinetic sampling are not only inaccurate, but they also provide only a snapshot of medicine taking behavior between visits.
If investigators are to understand the context of a trial participant’s poor adherence patterns, they need a detailed, holistic view of patient adherence behavior to inform individualized interventions.
It’s about moving the conversation, from just measuring adherence, to managing it: from simply averaging missed doses to understanding the root cause of the problem and using that knowledge to get – and keep – people on track.
Navigating the Path from Reflective to Automatic
The goal of modern medication adherence management is to help people initiate, implement, and persist with their treatment regimen. This process has three distinct phases: taking the first dose,
executing a routine of medicine taking behavior and following the defined regimen until the end of the trial.
According to the COM-B human behavior model, the key is helping people move from a “reflective” motivation at initiation, or at the start of a trial when they are most engaged, to an “automatic” motivation, in which they take their medicine as a matter of course.4
Multiple studies have shown that the quicker people adopt a strong medicine taking habit, the higher their persistence rates.1,5 Helping people to build that habit, which usually takes between 60 and 90 days,6 then, can contribute to higher adherence. The question is how.
With so many complex and interacting reasons for poor patient adherence, there isn’t a single solution. Sending text reminders, for example, has emerged as a popular intervention in recent years. However, this assumes the message recipients are non-adherent due to forgetfulness. It does nothing to answer any questions or concerns the patient may have around efficacy or safety, for example. In addition, persistent reminders can become intrusive, leading to disengagement or even discontinuation.
Only by understanding the root cause of an individual’s poor adherence can study teams hope to provide the appropriate intervention.
One solution is the DMAIC (define, measure, analyze, improve and control) element of the Six Sigma model. First developed in the telecommunications industry, the framework applies the scientific method to existing systems and processes.
When used in the context of adherence, an accurate, detailed record of past medicine taking behavior allows study teams and patients to work together to understand the barriers, and then implement the appropriate solutions.
The approach is perfectly suited to clinical trials, which have the clearly defined start (initiation) and end (discontinuation) points, as well as scheduled, pre-arranged visits (or virtual touch points) in which to deliver feedback and review the results.
That which can be measured can be improved, and patients’ awareness of their adherence patterns has been shown to change their dosing behavior. To date, more than 30 studies have shown that data-driven adherence feedback works.
One review and meta-analysis of 79 randomized clinical trials (RCTs) that electronically compiled drug dosing histories found that feeding back dosing patterns was the biggest factor influencing adherence. Interventional studies that included these focused discussions were 8.8% more effective than those without feedback, resulting in an average 20% overall improvement in adherence. In some patients, this figure stood as high as 50%.7
The approach works by combining medication adherence packaging, which timestamps medication administration, and powerful algorithms, which analyze the information to spot erratic dosing patterns.
Study investigators then discuss the data during the scheduled appointments, with a view to understanding the reasons behind the patterns. Establishing the reasons for poor medicine taking behavior allows for individualized interventions – whether they be reminder messages for the forgetful or education for the concerned.
The approach can be further enhanced by training the investigators in methods such as motivational interviewing, though this is not essential to success.
Four Steps to Success
There are four key elements to making this model work. Firstly, the data collection method must be frictionless, i.e., it must not create additional barriers to adherence by placing further burden on patients. medication adherence packaging is a good option here as it records dose information in the background, without asking study participants to do any more than take their medicine.
Next, the data collection method must be capable of generating a rich record of past medicine taking behavior that teams can rely on. Again, connected packaging is a good candidate as it has been shown to be 97% accurate. That compares to 70% for drug levels and markers, 60% for pill counts, 50% for HCP ratings, and just 27% for patient self-reporting, including electronic patient diaries.8
It also needs to focus on habit building. That means enabling HCPs to ensure patients have everything they need to initiate, implement, and persist with their medication at each stage of the journey. Key here is providing the tools, in the form of data and educational resources, they need to understand and react to all non-adherence drivers.
Finally, the approach should be seamlessly integrated into clinical trial execution, with focused feedback being provided and acted upon at regular intervals.
From Research to Practice
To date, the evidence for adherence feedback has focused on clinical trials, which have a defined start and end date and scheduled, regular HCPs / patient contact points. But there is also a need in routine practice.
Between 40% and 50% of people with a chronic condition, such as diabetes or hypertension, have poor adherence to their medication. This leads to an estimated 100,000 avoidable deaths and $100 billion in preventable medical costs every year.9
There is no reason to think helping people to build strong medicine taking habits in the community is unfeasible.
Rather, it would require some simple adaptations of current routine patient pathways to establish a feedback infrastructure. This may involve a more wide-spread adoption of digital adherence monitoring, and training nurses, pharmacists, or allied health professionals to deliver feedback and interventions, for example.
Measure to Manage and Improve
Improving adherence has benefits for every player in the healthcare and research ecosystem, from patients and HCPs to payers and pharma.
It is, however, a challenging problem to solve, and doing so relies on acknowledging its complexity. There are hundreds of reasons for poor adherence, and each one has a solution. The trick is matching the right intervention to the right patient, and a data-driven adherence feedback approach is the key.
Accurate dosing pattern data provides the basis for change. Once fed back to patients, it helps them to understand their behavior and gives them opportunities to ask for help, while enabling HCPs to design and deliver individual interventions.
- Demonceau, J., Ruppar, T., Kristanto, P., Hughes, D. A., Fargher, E., Kardas, P., … & Vrijens, B. (2013). Identification and assessment of adherence-enhancing interventions in studies assessing medication adherence through electronically compiled drug dosing histories: a systematic literature review and meta-analysis. Drugs, 73(6), 545-562.
- Eliasson, L., Clifford, S., Mulick, A., Jackson, C., & Vrijens, B. (2020). How the EMERGE guideline on medication adherence can improve the quality of clinical trials. British Journal of Clinical Pharmacology, 86(4), 687-697.
- Bae, S. G., Kam, S., Park, K. S., Kim, K. Y., Hong, N. S., Kim, K. S., … & Choe, M. S. P. (2016). Factors related to intentional and unintentional medication nonadherence in elderly patients with hypertension in rural community. Patient preference and adherence, 10, 1979.
- Jackson, C., Eliasson, Â. L., Barber, N., & Weinman, J. (2014). Applying COM-B to medication adherence: a suggested framework for research and interventions. European Health Psychologist, 16(1), 7-17.
- Blaschke, T. F., Osterberg, L., Vrijens, B., & Urquhart, J. (2012). Adherence to medications: insights arising from studies on the unreliable link between prescribed and actual drug dosing histories. Annual review of pharmacology and toxicology, 52, 275-301.
- Van der Weiden, A., Benjamins, J., Gillebaart, M., Ybema, J. F., & De Ridder, D. (2020). How to form good habits? A longitudinal field study on the role of self-control in habit formation. Frontiers in Psychology, 11, 560.
- Vrijens, B. (2019). A Six Sigma framework to successfully manage medication adherence. British Journal of Clinical Pharmacology, 85(8), 1661.
- Vrijens, B., Urquhart, J. Methods for measuring, enhancing, and accounting for medication adherence in clinical trials. (2014). https://pubmed.ncbi.nlm.nih.gov/24739446/
- Kleinsinger, F. (2018). The unmet challenge of medication nonadherence. The Permanente Journal, 22.