Medication Adherence - Is It Really the Patient’s Fault?

Medication adherence continues to be one of the most frustrating problems facing clinicians today. The problem is particularly vexing for blood pressure control, since hypertension is mostly asymptomatic and most patients require more than one medication for adequate control.¹ A typical perspective, if somewhat oversimplified, is that if patients took their medication as prescribed, their blood pressure would be well controlled. But it is not that simple. Medication adherence is complicated, often driven by a number of factors, and studies have shown that interventions focused on only one factor are often not effective.² In fact, the binary and static concept of adherence, that patients are either adherent or non-adherent, is likely not a useful construct.³ The American Heart Association (AHA) Scientific Statement, “Medication Adherence and Blood Pressure Control,” describes this complexity well, summarizing reasons patients may not be adherent to their antihypertensive regimen and highlighting evidence showing which interventions may increase adherence. The categorization of intervention strategies by stakeholder - patient, clinician, pharmacy and health insurer - is a fruitful construct, in large part because it directs accountability to components of the health care system beyond the patient. In their seminal work on medication adherence in 2003, the World Health Organization (WHO) promoted the concept that “patients need to be supported, not blamed.”⁴ Despite this and the growing evidence to the contrary, providers and health care systems tend to focus on patient-related factors.

The AHA Scientific Statement details how best to measure medication adherence, identify risk factors and predictors of nonadherence, and strategies to improve adherence. The first step, measurement of medication adherence continues to be challenging. Self-report and patient surveys are likely inaccurate or difficult to implement, respectively.⁵ A variety of innovative techniques such as digital sensors, electronic drug monitors and measurement of drug levels are generally not feasible in clinical practice. Pharmacy refill data may be the most practical and accurate approach for measuring medication adherence, as long as the health care system is able to capture all relevant pharmacy data.

Predictors of adherence are complex and multifactorial. The Scientific Statement categorizes these predictors using the WHO five dimensions of adherence framework: health care team and system-related factors, socioeconomic and demographic factors, therapy-related factors, patient-related factors and condition-related factors.⁴ While these predictors are numerous and often beyond the control of the provider or system, many can be directly influenced or modified, such as the complexity of the medication regimen and provider communication skills. Factors related to the socioeconomic and demographic dimension merit further discussion. Data from the National Health and Nutrition Examination Survey (NHANES) indicate that overall blood pressure control rates in the United States (US) fell significantly from 53.8% in the 2013-2014 survey to 43.7% in the 2017-2018 survey.⁶ Control rates in Black and Hispanic patients also decreased significantly, and now stand at 38.5% and 36.8% respectively. Medication nonadherence is a well described contributor.⁷ A retrospective study from Kaiser Permanente Northern California showed that Black and Hispanic patients were 56% and 46% more likely to be nonadherent to their antihypertensive medication, respectively.⁸ These differences were somewhat decreased after adjusting for use of mail order pharmacy and reduction of medication copayments, both of which were shown to significantly decrease nonadherence. Nevertheless, there likely are factors beyond socioeconomic status and ease of medication refill that contributed to the disparity. A recent study employing a large pharmacy claims database showed that adherence to antihypertensive medication was 7.7% and 7.9% lower in Black and Hispanic patients, respectively.⁹ After controlling for socioeconomic status, demographics, health status, out-of-pocket costs, and convenience of refilling prescriptions, these differences were attenuated by only 30% to 50%. This suggests that other factors, including patient-provider communication, may be involved.¹⁰

The Scientific Statement details numerous interventions that demonstrate modest improvements in adherence including patient education and counseling, medication regimen management, refill synchronization, reminder and feedback systems, and incentives. It’s worthwhile to highlight a few that are easy to implement and offer significant potential. First, health systems must have a reliable method of medication adherence assessment and provide that assessment at the point of service; pharmacy refill databases appear to be the best approach for this. With availability of real-time assessment, education and counseling in an open-ended manner, ideally using motivational interviewing, can be undertaken. Simplification of the medication regimen, especially with use of a fixed-dose combination (FDC) pill, is a great place to start. Now advocated by the majority of hypertension guidelines,^11-13 but vastly underutilized,¹⁴ FDCs have clear evidence for improving adherence, improving blood pressure control, and decreasing time to control. In a recent analysis of NHANES data, use of a FDC occurred in approximately 19% of patients in 2013-2016 and had not significantly changed since 2005-2008.¹⁴ In a meta-analysis of 12 retrospective studies, adherence was estimated to be 8% higher for patients naive to prior antihypertensives and 14% higher for non-naive single pill combination (SPC) patients compared with corresponding free equivalent combination (FEC) patients. Persistence in the SPC groups was twice as likely as the FEC groups.¹⁵ Similar results have been found in other reviews.^16,17 Self-monitoring of home blood pressure is an intervention that works on multiple levels. With proper training, home blood pressure measures are more predictive of cardiovascular outcomes than office blood pressures,¹⁸ and should be used in both diagnosis of hypertension and medication titration.¹² Studies assessing self-monitoring of blood pressure also show improvement in medication adherence. A systematic review of 13 studies found a small but significant benefit of self-monitoring of blood pressure on medication adherence.¹⁹ A more recent prospective observational study of 7,751 patients showed that patients given a home blood pressure monitor had decreased days per week in which medication was skipped or reduced.²⁰ In addition, medication adherence increased as the frequency of home blood pressure monitoring increased. Further, the Scientific Statement raises the importance of treatment intensification in the setting of suboptimal adherence. Lack of adherence may be one the biggest drivers of clinical inertia in hypertension.²¹ Providers frequently postpone intensification if adherence is not satisfactory, leading to revolving cycles of delays and rechecks. Not only is intensification safe and effective with suboptimal adherence, but it decreases time to control, which is an important factor in reducing cardiovascular outcomes.^22-24

We’ve seen direct confirmation of the effectiveness of many of these interventions in Kaiser Permanente Southern California (KPSC). KPSC improved their hypertension control rates from 54% to 84% from 2004 to 2010 due to a number of key factors.²⁵ Perhaps the most important factors were the use of a FDC medication as the initial step in the hypertension treatment algorithm and a no-hassle, no-cost follow-up blood pressure visit. Between 2005 and 2011, use of the preferred FDC lisinopril/hydrochlorothiazide pill increased from 18% to 37%, paralleling the increase in control rates.²⁵ Less well described is use of a pharmacy database (since most refills are filled within KPSC pharmacies) that enables providers to have adherence data at the point-of-service (in essence a “5th vital sign”), as well as a robust mail order pharmacy system for refills. Providers are not only encouraged to have adherence discussions with patients, based on these data, but to also consider treatment intensification when appropriate in selected patients with suboptimal adherence. When the medication regimens are simplified and the refill process is made easy, adherence improves. Furthermore, having real time data to open the conversation with the patient, is likely to uncover patient-related factors leading to suboptimal adherence.

Patients do have a large role to play in medication adherence but shifting to a systems approach will likely reap bigger benefits in the long run. Simple evidence-based measures to assess for and improve adherence are feasible and can be undertaken by all clinicians and health systems. The AHA Scientific Statement on Medication Adherence and Blood Pressure control provides a straightforward blueprint for how to do this and represents an important step forward in improving the approach to blood pressure control.