This feature is part of a series focused on drug adherence. To view other posts in the series, check out the spotlight page.
When preparing to launch a new rocket or satellite into space, the checklist of considerations NASA scientists make is mind-boggling. They have to plot the vessel’s trajectory, determine the amount of fuel it needs, equip it to handle intense heat and aerodynamic forces. Weighing on each of those decisions, both literally and figuratively, is gravity. Though electrical mishaps or math miscalculations are problematic in themselves, it’s gravity that ultimately brings everything crashing back down to Earth.
Each branch of science has to deal with its own kind of gravity — something that, albeit accounted for, stands to bench a project before it ever gets off the ground. For pharmaceutical researchers, it’s patient adherence. After all, how helpful can a pill really be if its target population won’t swallow it? But unlike the laws of physics, adherence is neither exact nor easily measurable.
While it’s rare for a clinical trial to fail solely because participants didn’t properly take their medications, non-adherence still looms over most, if not all, investigations, threatening data accuracy and often the future of promising therapies. Drugmakers acknowledge the detriment non-adherence can have on bottom lines and disease treatment options, and have turned to modern and mobile technologies as one potential fix. Still, industry insiders question whether there is enough focus on stamping out the problem altogether.
For decades, clinical trial researchers determined adherence with only a few tools at their disposal. The two norms were diaries and pill counts. The former supplied investigators with logs that showed when a patient took the drug being tested, as well as other health information. The latter allowed investigators to see how much medication a patient went through between trips to the clinic, providing evidence as to whether the established dosing schedule was being followed.
The credibility of both methods suffered, however, because patients had the power to fudge their adherence; there was no way to tell if they were actually taking the medication and when. Healthcare stakeholders debate exactly how frequently such events take place and the extent to which they affect a study’s outcomes — part of the reason why diaries and pill counts are still widely used in the clinical setting. What they do agree on, though, is the potential to improve compliance through the incorporation of newer, more tech-savvy approaches.
Pfizer, for instance, rolled out a program called mClinical in 2015 that leverages mobile devices to keep clinical trial participants adherent. The program’s offerings include electronic informed consent documents, information about upcoming clinical visits and apps that send patients notifications about taking their medicines. It also uses the video function on smartphones and tablets to conduct clinical visits remotely, and sensing equipment such as wearables to more accurately assess patient lifestyle habits.
On the other hand, Novo Nordisk employs email and text reminders too, though more so with younger trial participants, according to the company’s Chief Medical Officer for North America Todd Hobbs. For the Danish drugmaker, a key strategy for better clinical adherence is having more regular interactions with participants.
"If they do stop taking the drug, that's their choice and that's fine, but we need to understand why they did it," Hobbs told BioPharma Dive. "So we do a whole lot of contacting them through multiple methods so we at least know what's going on with them in the trial."
Technology adds much needed variety to the ways in which adherence is measured, yet also carries unique challenges that have made drug developers slow to adopt it.
Some patients, as Hobbs described, are less likely to check email messages or message via texting. More worrisome is the idea that those tactics may upend most clinical trials’ primary objective: to evaluate a drug’s safety and efficacy.
"What happens when we start to change patients' behavior so much that the control arm isn't really a control arm anymore — that these are no longer just patients receiving standard of care, but [being] coached and reminded and guided so much so that their behavior is now different?" Craig Lipset, Pfizer’s head of clinical innovation, said in an interview.
A safeguard against altering behavior too radically is a well-constructed study, according to Lipset. Pfizer’s clinical procedures try to limit lifestyle modifications to only those imperative to patient safety and trial integrity, such as requiring participants to undergo certain diagnostic tests or stop taking other medications.
"Keeping a focus on adherence to the protocol helps to control some of those risks that we're going to modify our control arm's behavior and jeopardize the power of the study," Lipset said.
While an adherence cure-all isn’t on the horizon, investigators are getting better at identifying situations where participants are more likely to fall off the treatment regimen.
A slew of research has shown patients are more compliant to a drug’s prescription when it requires less frequent dosing. A meta-analysis published in 2015 in the journal PLOS One found both adults and children are more inclined to take a drug that’s once-daily versus twice-daily, a drug that’s twice-daily versus thrice-daily, and so on.
Other studies have concluded that patients are more likely to forget proper treatment methods outside the clinic and become non-adherent the longer they have to take a drug. The mode in which a medicine is administered and the disease it targets also hint at what adherence obstacles investigators may run into.
"We know clearly that adherence is much better for pills than it is for shots and injections," Hobbs said, a trend that makes adherence particularly difficult to achieve in trials testing oral versus subcutaneous treatments.
"Pain, tremor, those types of things people can see and feel, and make people much more self-aware as far as taking the medication than" with diseases like hypertension or high cholesterol, which require long-term medications but don't have daily symptoms that patients can feel, Lipset said.
Drug companies have also focused their efforts on weeding out patients who aren’t serious about subscribing to protocols at the onset of a trial.
"Most patients enrolled in clinical trials will initiate treatment because they have been highly recruited to initiate treatment," Bernard Vrijens, chair of the European Society for Patient Adherence, Compliance and Persistence, said.
Effects on data
Despite adherence being an omnipresent challenge in the clinic, the times it’s blamed for a trial’s failure are few and far between.
Drugmakers claim that well-designed study designs, particularly those that are randomized, keep the amount of adherence relatively equal between experimental and control arms. Couple randomization with an improved ability to spot non-compliant patients early in the recruitment process and a handful of tools that check protocol deviations when they arise, and it becomes uncommon for adherence to spoil readouts.
"I won't say adherence has been fantastic, but at least it has not been a concern to where it affects the data quality,” Hobbs said, adding that Novo’s weight loss management program for Saxenda (liraglutide) has been one minor exception, as escalating doses lead to greater side effects and higher dropout rates.
But some healthcare researchers contend that the current strategies for dealing with non-adherence just aren’t good enough.
A 2013 study published in the journal Molecular Interventions, for example, criticized the use of intent-to-treat analysis — which includes all randomized participants regardless of withdrawal rates and noncompliance — in clinical trials.
"Proponents also argue that the ITT approach provides randomization that prevents biased allocation of subjects to treatment groups and accordingly prevents biased results," the study said. "Nonetheless, the ‘all-comers’ approach ignores the impact of medication compliance on outcome, giving subjects who dutifully take their medications no more weight than to subjects who fail to adhere to trial protocols."
Vrijens, meanwhile, points out that while most investigations successfully get patients started on treatment, they have much more trouble achieving daily implementation and longer-term persistence.
"It goes undetected today because they're using pill counting, which is easy to be censored by the patients and it's easy to create a good-looking adherence record," he said. "They should use a much more advanced technique to measure adherence, and especially daily implementation of the dosing regimen to make sure that patients are taking their medication in trials — because it's far away from the reality."