Contributed Commentary by Andrew D. Pucker, OD, PhD, Lexitas 

December 30, 2024 | We have all heard it often noted that treatment outcomes in real-world practice don't always reflect those observed in clinical trials, largely attributing these differences to a lack of adherence to prescribed therapy. Yet the uncomfortable truth is that therapeutic nonadherence happens in clinical trials too—with far-reaching consequences. 

Nonadherence refers to any deviation from the prescribed dosing regimen, this includes everything from not taking the correct dose, skipping doses, and taking the treatment at the wrong time. When subjects do not take their treatments as prescribed, it complicates the study's ability to assess the true efficacy and safety of the treatment. There are other issues as well. Subjects won't be as likely to benefit, therefore leading to an underestimation of the treatment effect. At the same time, subjects will be less likely to experience side effects, which can potentially cause an overestimation of the treatment's safety.  

Without careful monitoring of a subject's adherence, data are difficult to interpret, with increasing variability that influences the statistical power of the trial. All of us involved in clinical trials must strive to ensure a subject's maximum adherence to treatment. At risk is usable data to perform the efficacy analyses needed for the treatment to ultimately receive approval.  

This is not a small problem. A database study of electronically compiled dosing history data from 95 clinical studies found that half of the almost 17,000 subjects substantially deviated from the dosing regimen outlined in the study protocol. Only a small fraction of subjects adhered to a 24-h dosing schedule with once-daily medications, and by day 100, 20% of subjects had stopped taking treatment. A further 12% of subjects displayed suboptimal treatment adherence. There are myriad reasons for suboptimal treatment adherence, but the main culprits are a lack of efficacy (often a perception), poor treatment tolerance, complicated treatments, duration or high frequency of treatment, or subject misunderstanding. With these issues in mind, sponsors should strive to keep their treatment regimens as simple as possible.  

More severe issues like skipped dosing or complete discontinuation of medication are frequently observed, with "drug holidays" being a particular concern. Overdosing is also common in trials, which is characterized by the intake of additional doses or the consumption of doses too closely together and is a cause for concern as it increases the likelihood of encountering side effects.  

Measuring Treatment Adherence  

Medication adherence within drug trials is complex, but focusing on it throughout every stage of the trial can help to increase success. There are several common methods for measuring adherence, but not all apply equally well to ophthalmic trials.  

Direct Observation 

One of the only real-time measures of adherence that can guarantee drug administration is to directly observe subjects receiving or administering the treatment. This works primarily only when the subjects are being treated in a clinical setting, otherwise it would be too resource intensive and intrusive. In ophthalmology, treatments delivered via intravitreal injection are common in retina trials. It is certainly easy to measure compliance in these situations, as the only potential errors that could be introduced are if the study team provides the wrong treatment or if the subject fails to report to the clinic on schedule when multiple doses are needed.  

Pill/Dose/Vial Count 

Manual counting of the investigational product is convenient and is common for drug accountability in many clinical trials. As reflected in our and others' experience though, the count method can easily be censored by subjects and only provides an aggregate summary of adherence between visits. In other words, it wouldn't work for precise estimates of initiation, implementation. Because eye care drugs are mostly delivered topically, vial counting would only be feasible for single-use vial studies that use a new vial for each dose. Even then, we have found that subjects may re-use or lose vials. We learned the hard way that the vials can get very dirty during storage making investigational product reconciliation unpleasant for the study team! Additionally, interruptions in dosing or the timing of taking the investigational product cannot be fully evaluated during statistical analyses. 

Patient Self‑Reporting 

In ophthalmology clinical trials, the most commonly used method and the one we have the most experience with is self-reported adherence data. Subject diaries can be used to track different measures and methods of administration.  

Admittedly, there is a large variation in validity and reliability, as these instruments are sensitive to subject recall and desirability biases and, like pill counting, have often been found to overestimate adherence. While traditionally we recommend subjects use pen and paper and to record the compliance, we do find these data can be time consuming for Contract Research Associates (CRAs) to monitor for accuracy. Because of this, electronic data capture (i.e., subjects directly entering information into the system) is becoming more common in clinical trials.  

We let sponsors know that an advantage of self-report is the ability to explore why a subject has or has not adhered, and we can design reports to capture reasons for nonadherence. CROs are also increasingly using exit interviews to assess adherence. In our experience, subject self-reporting of treatment adherence is often used in combination with investigational product reconciliation. While tracking compliance with say both a paper diary and vial counting provides a backup method, we also make sure sponsors understand they will still need to designate one method as the primary method of measuring compliance. This is important because there is a potential for discordant reporting between multiple methods, and it is important to remove the potential for bias by selecting the method that may give the study a better result.  

Electronic diaries can time stamp the records and systems can employ reminders via text messages to help improve compliance in real time. Nevertheless, we have found these systems to be costly, and depending on the subject population, subjects could have difficulty completing them. Even if they are used, we always recommend paper diaries as a backup. System connectivity issues cannot be predicted, and those backup paper diaries will be helpful in case of technical problems. Since the action of recording the event, however, is often disconnected from the treatment, exaggerated levels of adherence can be reflected. Requiring the subject to capture a video of each use is another approach, but not one generally implemented in U.S. Food and Drug Administration regulated trials. 

Drug/Metabolite Monitoring 

Monitoring drug or drug metabolites in plasma, urine, or hair can provide a snapshot of adherence. It is, however, reported to be sensitive to bias when subjects ingest the investigational drug before a trial visit, even if they may have been nonadherent in the period running up to the visit ("white-coat adherence"). It would also be restricted to active arms of the trial or pharmacologically inactive biologic markers could be used. Unless daily sampling is undertaken it could measure adherence implementation but not determine initiation or persistence. Ultimately, most topical drops have limited systemic absorption; therefore, this approach will have limited utility in ophthalmology. 

Electronic Monitoring 

Electronic monitoring involves integrating a microchip into pharmaceutical packaging. The technology has been expanded to include not just pill bottles but also injectables, inhalers, cream tubes, and eye drop containers. The chip automatically timestamps each action required to access or administer the medication, providing real-time tracking of dosing histories. Removing the drug from its package does not necessarily indicate ingestion, however. Studies have demonstrated that electronically recorded dosing histories do align with bioanalytical measures in 97% of cases. Though possibly advantageous, our experience is that this technology has not been generally used in U.S. Food and Drug Administration-regulated trials. The lack of use is likely tied to the relatively recent emergence of this technology, and we anticipate this technology will gain traction as methods are developed.  

Ingestible Sensors and Electronic Detection of Ingestion 

Emerging technological approaches to measuring adherence include breathalyzers, monitors for inhalers, and ingestible sensors. Although they may not be feasible for all drugs or in all contexts, they can confirm that the drug has been ingested by the by the subject and provide insights into patterns of adherence and nonadherence. The technology may be considered overly invasive and therefore objectionable to the user, and these methods will also need to be fully vetted before they are suitable for use in U.S. Food and Drug Administration-regulated trials. 

Cost of Nonadherance  

Ultimately the monitoring method chosen has to be affordable, simple, and take into consideration the trial subjects' characteristics.  

When talking about compliance with medications, the focus is usually on clinical care but avoidable nonadherent behavior also exists in clinical trials. Without reliable measures of treatment compliance in place for clinical trials, conclusions drawn from these studies can be problematic with massive implications for clinical care.  

It also has financial ramifications to the sponsor, such as increased costs of recruiting, replacing, and retaining subjects, longer trial durations, and higher data collection costs. To mitigate the risks of nonadherence, trials tend to overenroll, leading to higher costs. With this in mind, we urge sponsors to strongly consider their strategies to enhance adherence early on in the trial design process. These plans should be in line with the investigational product being evaluated.    

Addressing this substantial and costly issue benefits all stakeholders. Most importantly, patients will be more likely to have access to safe and effective treatments, which is an outcome we are all seeking. A detailed and effective plan for ensuring subject compliance will likewise make it more likely for a drug to make it to market and for a company to ultimately be successful.  

Andrew D. Pucker, OD, PhD, FAAO Diplomate, FSLS, FBCLA, is Executive Director of Medical and Clinical Sciences at Lexitas Pharma Services, Inc. He earned a BS in Cellular and Molecular Biology from the University of Wisconsin-La Crosse, and OD, MS, and PhD degrees from The Ohio State University. His independent research career has focused on dry eye disease, contact lenses, and myopia development, and he has authored over 50 peer-reviewed publications and over 160 editor-reviewed columns. He can be reached at andrew.pucker@lexitas.com.