Hypertension is one of the most widespread modifiable cardiovascular risk factor worldwide. It is estimated that about a third of US adults have hypertension. A recent report from the American Heart Association estimated that hypertension accounts for greater than $50 billion in healthcare cost annually . Research over the last 30 years has clearly demonstrated a causative effect of hypertension on cardiovascular and cerebrovascular disorders. A host of pharmaceutical agents lower blood pressure and exert a beneficial effect in improving outcomes. More recent studies have demonstrated that intensive control of hypertension may have incremental benefits even in patients with low baseline cardiovascular risk . Although there are a wide array of drugs for lowering blood pressure, the greatest obstacle to achieving optimal control of blood pressure is prescription compliance. Every physician across the world who prescribes antihypertensive medication faces this challenge. Factors that influence prescription behaviors and compliance are less well understood creating a barrier in healthcare delivery and translating randomized trial data into routine clinical practice. Any intervention that can improve patient compliance of medications and facilitate effective blood pressure control would thus have a positive impact on overall health and patient outcomes.
In this context, Krousel-Wood et al. report results of a prospective study using a simplified, easily available, open access four-point hybrid assessment tool (K-Wood-MAS-4) in assessing patient compliance to antihypertensive drugs and relation to long-term cardiovascular outcomes . The K-Wood-MAS-4 tools incorporates four domains of adherence behavior – self-efficacy, physical function, intentional medication taking and forgetfulness. A score at least 1 on this scale indicates low adherence. The investigators rightfully compared outcomes using the metric of proportion of days covered (PDC) from pharmacy disposition databases. PDC is a widely used metric that assesses patient compliance with filling and refilling medications. The study population included greater than 1500 adults with no prior cardiovascular events and the primary outcome was cardiovascular events (stroke, myocardial infarction, congestive heart failure, or death). Over a 3-year follow-up, rate of low adherers was 38.7%. Rate of incident cardiovascular event was 8.9% with rates being significantly higher in low adherers compared with high adherers (12.8 vs. 6.4%; hazard ratio 2.29). This difference was consistent across different age groups, race and sex. Among the four domains included in the K-Wood-MAS-4 tool, the most significant predictors of outcomes were physical function and medication taking self-efficacy. Intentional (feeling well and not taking) or unintentional (forgetfulness) nonadherence did not pan out as significant predictors of poor outcomes. Analysis of outcomes based on PDC revealed cardiovascular event rates of 11.7 and 8.7% in low and high adherers, respectively. Interestingly, survival curves separated earlier using the K-Wood-MAS-4 when compared with PDC.
The results of this study are easy to understand and very provocative for clinical practice. First, the authors need to be congratulated for a rigorous methodology involving prospective patient surveys and strict adjudication of outcomes. Several factors influence medication adherence practices, especially in the elderly. Polypharmacy, sociobehavioral factors, drug side effects and overall well being are heavy obstacles, we encounter in clinical practice day in and day out. Several aspects of the current study can help clinicians identify and target high-risk patient populations in addressing these barriers. Firstly, sociodemographic factors associated with low adherence by the K-Wood-MAS-4 were female sex, black race, polypharmacy, affordability, depressive symptoms, low social support, high stress and lower educational level. These patients represent a challenging cohort in clinical practice and outcomes are influenced by level of health literacy and social support systems. Efforts to identify and closely follow patients with these risk factors would be beneficial. Secondly, among the four domains, physical function had the highest predictive value in predicting adherence and long-term cardiovascular outcomes. This in our opinion is the strongest message of the study. We often encounter patients with limited or no physical function who require multiple medications for various comorbidities. These results demonstrate that this represents a high risk group for low adherence. Addressing uncontrolled hypertension in this group of patients depends more on understanding barriers to adherence and less on adding new classes of drugs. Thirdly, the most common metric used for pharmacy compliance (PDC) does not appear to predict true compliance.
Validating compliance assessment scales such as K-Wood-MAS-4 and incorporating them in quality metrics of patient outcomes would be the way forward to utilize existing therapies to the maximum potential. This also may be the most cost effective way of improving healthcare outcomes compared with the cost of development of a new therapy. It is time for the clinician to think beyond the clinic.
Conflicts of interest
There are no conflicts of interest.
1. Benjamin EJ, Blaha MJ, Chiuve SE, et al. American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Heart disease and stroke statistics: 2017 update: a report from the American Heart Association [published corrections appear in Circulation
. 2017; 135(10): e646 and 2017; 136(10): e196]. Circulation
2. Attar A, Sayadi M, Jannati M. Effect of intensive blood pressure lowering on cardiovascular outcomes based on cardiovascular risk: a secondary analysis of the SPRINT trial. Eur J Prev Cardiol
2018; [Epub ahead of print]. doi: 10.1177/2047487318800741.
3. Krousel-Wood M, Peacock E, Joyce C, Li S, Frohlich E, Re R, et al. A hybrid 4-item Krousel-Wood Medication Adherence Scale predicts cardiovascular events in older hypertensive adults. J Hypertens