Telepharmacy Program Improves Treatment Adherence, But Not Clinical Outcomes

Deborah Brauser

November 20, 2017

ANAHEIM, CA — Although a novel, pharmacist-based, remotely delivered program is associated with a modest improvement in treatment adherence in patients with hypertension, diabetes, or hyperlipidemia, that doesn't translate into improved disease-control measures, new research suggests[1].

The randomized, proof-of-concept Study of a Tele-pharmacy Intervention for Chronic Diseases to Improve Treatment Adherence (STIC2IT)  assessed more than 4000 patients with poor baseline adherence. The intervention included a telephone consultation/interview with a practice-embedded pharmacist, patient-tailored text messages, mailed progress reports, and feedback from the pharmacists to the primary-care physicians.Results showed that the participants who went through the intervention had almost 5% greater treatment adherence 12 months later than those who received usual care.

However, there was no significant difference between the groups in disease-modifying outcomes, such as lower LDL or HbA1c levels.

Dr Niteesh Kumar Choudhry

Dr Niteesh Kumar Choudhry (Brigham and Women's Hospital, Boston, MA) told Medscape Cardiology that he was "more surprised than disappointed" with the results.

"We targeted the intervention to patients we thought needed the most help, and it did improve adherence, but when we looked at whether we could influence disease control, it turns out we could not," said Choudhry.

"One explanation for this discrepancy could be that we didn't improve adherence enough, which tells us what bar we need to achieve in the future."

He presented the results at the American Heart Association (AHA) 2017 Scientific Sessions.

Major Public-Health Problem

Half of all patients with cardiometabolic conditions do not adhere to prescribed therapy, leading to annual preventable health spending costs of $100 billion to $300 billion in the US alone, reported Choudhry. "This is a major public-health problem, and each year in the US there are about 125,000 deaths attributable to this."

Through the years, some interventions have shown only modest improvements, likely because they did not take into account a patient's specific barriers to adherence and/or they were delivered at just one time point, he added.

"Even effective interventions are difficult to sustain because they were imprecisely targeted to patients who do not need adherence assistance and are often expensive," said Choudhry.

For STIC2IT, investigators enrolled 4078 patients (55% men; mean age 59 years) from 14 primary-care practices in the US between August 2015 and July 2016. All of the participants were nonadherent to prescribed oral glucose-lowering, antihypertensive, or statin medications, according to claims data. They also had poor disease control, according to electronic health record data.

All patients filled out a Patient Activation Measure (PAM) questionnaire about their knowledge of and commitment to managing their healthcare. They were then randomly assigned to receive either the intervention (n=2038; baseline adherence rate 57.2%) or usual care (n=2040; baseline adherence rate 57.0%).

A pharmacist-initiated phone consultation/interview (mean time 25 minutes) was the foundation for the intervention. During each talk, patients were asked about potential adherence barriers and their willingness to modify behaviors. Patient and pharmacist then developed an action plan for improving both adherence and disease control based on what was discussed and on the patient's PAM level.

That said, only 52% of the intervention-assigned patients elected to do the phone consultation.

Still, all intervention-group members at PAM level 1 or 2 were offered intensive adherence strategies, such as:

  • Daily texts with adherence reminders or motivating messages.

  • Pillboxes with up to multiple-times-per-day dosing allowed.

  • Follow-up consultations.

  • One-on-one video chats.

Patients at PAM level 3 or 4 were offered less intensive strategies, such as weekly pillboxes and texting.

Regardless of PAM level, patients were mailed automated, individual progress reports containing biometric and adherence information at 6, 9, and 12 months. Discussions were also held between pharmacists and the patients' primary-care physician.

Primary, Secondary Outcomes

The primary outcome was medication adherence 12 months after randomization on an intention-to-treat (ITT) basis. The main secondary outcome was disease control at 12 months, defined as HbA1c less than 8%, blood pressure less than age-specific targets, and LDL that was less than targets from ATP III guidelines.

In the ITT analysis, adherence was 4.7% greater at 1 year in the intervention group than in the usual-care group (P<0.001).

"This effect size was similar to those achieved by more labor-intensive interventions," noted Choudhry.

The intervention-group members with hypertension or hyperlipidemia also had significantly greater adherence (by 8.5% and 4.6%, respectively) than the usual-care group members (both comparisons P<0.001).

While those with diabetes receiving the intervention had -0.2% less adherence, this was not significantly different (P=0.86).

The only prespecified subgroup analysis that showed significant differences in adherence rates was between the sexes, with the intervention more effective in men than women (P=0.03).

There was an identical rate of "good disease control" for those with all eligible conditions—achieved by 23.4% of each of the two intervention groups. The rates were also very similar (28.0% vs 27.7%, respectively) for good disease control for those with at least one eligible condition (P=0.84).

Will Investigators STIC2IT?

"Basically, we saw no effect at all on clinical outcomes," said Choudhry, citing several potential reasons for the findings. This included that the investigators used routinely collected data, "which might have included inaccuracies; so we might have overestimated adherence or underestimated the actual clinical impact."

He also noted that the adherence improvement may have just been too small to translate into clinical benefits or that patients may have required therapeutic intensification.

"Our results have more implications for future interventions. They should be more intensive while still pragmatic, they should focus on a more impactful patient population, and they simultaneously need to address adherence and other barriers to optimal disease control," he summarized.

During the press briefing, Choudhry admitted that questions remain about the study's generalizability, "and certainly these results would not be applicable in other less-integrated systems. But there are many jurisdictions throughout the country for which these results may apply."

To Medscape Cardiology, Choudhry said the investigators will not be abandoning this program. "This is too important of a problem and it's fundamentally about changing human behavior. But we need to figure out how to increase the use of these interventions," he said.

"This infrastructure offers promise: it's embedded, it's sustainable, and it's practical. Now we need to tweak the actual implementation of the intervention."

Opportunities for Improvement

At the briefing, official discussant Dr Tracy Y Wang (Duke Clinical Research Institute, Durham, NC) noted that strengths of STIC2IT were that it included a "low-cost, scalable intervention," it had a population-level pragmatic study design, and the change seen in adherence was sustained over time.

With all of that, though, why didn't it show a risk reduction in clinical outcomes?

"Would we expect a less than 5% improvement in adherence to actually reduce risk? I think the answer is: probably not. Also, the intervention was not embraced by all of the patients," said Wang, underlining that the phone consultation was used by only half of that group.

"Also, one wonders if there are disease-specific factors," she added, noting that the intervention worked less well in those with diabetes.

Other questions that remain unanswered include how "risk-factor control" was actually defined—and whether the study was even powered to see risk-factor control, she said.

The top take-home message, said Wang, is that there are opportunities for improvement. "We still need to figure out how to substantially move the needle on adherence."

Implementation Science

Session chair Dr Ivor J Benjamin (Medical College of Wisconsin) told Medscape Cardiology that treatment adherence in this patient population is a serious challenge.

"There's this whole new area called 'implementation science.' How are you going to be get people, not withstanding evidence, to be able to think about such things as computerized methods and reminders? As you can see, the pickup rate [for prescriptions] is not so good," said Benjamin, who is also the president-elect for the AHA.

"This means that we have to think about other types of interventions, because disease burden is still there," he added. "I'm speculating here, but how can we better use medical navigators, how can we bring people into support groups, and who are better able to identify that someone really needs to be adherent? Because clearly electronic devices, as much as they are ubiquitous, seem to not be the sole answer."

He added that patients in past studies have shown increased adherence, yet reverted back to nonadherence once the studies ended; and he agreed with the current investigators that their study's 5% increased adherence rate wasn't large enough to translate into clinical outcomes.

"As we think about the right patient, the right drug, and the right price, what wasn't discussed here was often patients have side effects from medications," said Benjamin.

"As we work on precision medicine, we need to work to define a better profile for patients where they can really see the intended benefit. That might be the most powerful way of reinforcing 'why I need to take this.' It's helping them to see that the potential benefit exceeds the potential downsides of not sticking with it."

The study was funded by the National Institutes of Health National Heart, Lung, and Blood Institute. Choudhry, the other study authors, and Benjamin report no relevant financial relationships.

Follow Deborah Brauser on Twitter:@MedscapeDeb. For more from | Medscape Cardiology, follow us on Twitter and Facebook.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.