How health IT can guide chronic disease patients to enhanced medication adherence
While the abuse of opioids and other prescription drugs has been getting a lot of important attention from healthcare stakeholders and policymakers, lately, a quieter but perhaps more long-standing problem involves how to address the problem of patients who stop taking their medications altogether.
Higher costs and poorer health are but two of the consequences of medication non-adherence, and in a recent article at Catalyst NEJM, Leah L. Zullig, PhD, and Hayden Bosworth, PhD, both of Duke University, analyze the processes related to medication adherence, discuss what happens when those processes break down and offer potential solutions.
As they describe it, there are “three distinct processes” connected to medication adherence: initiation, when the patient takes the first dose; implementation, which is how well the patient’s actual dosing regimen matches what was prescribed by the provider, and: discontinuation, when a patient stops taking the medication for any reason, either with or without the guidance of their healthcare provider.
There are, of course, many factors that can contribute to non-adherence, including socioeconomic circumstances, patient perceptions and motivations, the medical condition being treated and the healthcare system in which the person is receiving care.
And, at the same time, there’s no shortage of tools and strategies available to to promote medication adherence, but it’s important to remember that each patient and situation is unique.
Which is where the strategic use of health IT comes in.
In Zullig and Bosworth’s view, a key initial step toward improving medication adherence involves proactively screening patients for potential problems related to medication adherence by using predictive analytics and integrated data systems to identify patients who are “at risk for medication non-adherence on the basis of past behaviors and personal characteristics.”
An important next step, they say, is “to collect patient-reported outcomes (PROs), such as real-time symptoms and perceptions of medications . . . The combination of predictive analytics, electronic medical records data, and PROs will allow for the efficient allocation of resources to the right patients at the right times.”
As an example, they point to a project at Duke University in which investigators “are combining medical records data and PROs to identify and personalize an intervention to improve adjuvant endocrine therapy (AET) among women with breast cancer.” In the trial, “400 women who are taking AET are being randomized to a general health education program or an intervention involving a combination of phone calls, tailored interactive voice messaging based on information exchanged during phone sessions, and real-time adherence data obtained from wireless smart medication bottles.”
There are certainly other elements to the effort to improve medication adherence, but it’s clear that much of the effort’s success will rely on a more deft and strategic of the IT tools that are increasingly available.
While “there is no universal solution to improve adherence,” the writers point to the growing evidence, including the effective incorporation of health IT, that “suggests that combining approaches that are tailored to address a patient’s specific adherence barriers or challenges may equip patients with the understanding and tools they need to successfully engage in medication adherence.”