At least 1 in 7 cardiology patients now reports nonadherence to prescribed medications, potentially leading to negative outcomes across a broad range of cardiovascular diseases. This nonadherence can begin as early as the time of prescription or any time thereafter and occurs for a variety of reasons, including communication difficulties, polypharmacy, and a variety of objective and perceived side-effects. Among elderly, low-income, and disabled patients, drug costs represent a growing source of medication nonadherence and can be markedly reduced through the use of drug assistance programs and low-cost generic medications without sacrificing evidence-based therapy. Depression also contributes strongly to nonadherence and is widely prevalent in cardiovascular populations. Improvements in depression are mirrored by improvements in adherence. A systematic screening to identify the presence of nonadherence and many of its causes can be accomplished with minimal impact on visit length. In conclusion, once specific concerns are recognized, options frequently exist to help patients and providers address many of the most common difficulties.
Today, the cardiac cornerstones of hypertension, hyperlipidemia, heart failure, and coronary artery disease are medically managed more effectively than ever before, in large part because of the application of multidrug therapies. Continued improvements in the long-term success of many percutaneous interventions and cardiac surgeries have also come about, in part, as a direct result of improvements in pre- and postoperative drug regimens. The contemporary practice of cardiology has necessarily become one of polypharmacy, and success in identifying and managing adherence to medications has become essential to daily practice.
Select Consequences of Nonadherence in Specific Disease States
Certain diseases have seen decades of treatment on the basis of polypharmacy, and in these disease states, the consequences of nonadherence to medical therapy have been particularly well documented.
Systolic heart failure as a condition is especially challenging because of the synergistic importance of adherence to drug and lifestyle therapies. Van der Wal et al identified 18 studies of medication adherence in heart failure from 1988 to 2002, demonstrating a consistent overall adherence rate with medical therapy of about 80%. Nine studies looking at diet and fluid restriction found self-reported compliance to be 50% to 75% in most cases. Translated into outcomes, across 7 studies encompassing 1,500 patients with heart failure, nearly 1/3 of all hospitalizations were identifiably due to nonadherence to either medication or diet. In studies in which the data were available, uncontrolled hypertension accounted for another 15% of admissions, placing issues of medical therapy at blame for ≥1/2 of these hospitalizations.
Acute coronary syndromes have seen the remarkable success of stent therapy, which is inherently dependent on continued adherence to the dual-antiplatelet therapies of aspirin and thienopyridines. A 2006 study of data from the Prospective Registry Evaluating Myocardial Infarction: Events and Recovery (PREMIER) found that of 500 patients treated with drug-eluting stents, nearly 1 in 7 had prematurely stopped their clopidogrel <30 days after infarction, despite clear instruction to continue therapy for ≥3 to 6 months. This nonadherent group saw 1-year mortality of 7.5%, almost 10-fold higher than their adherent peers, with a rate of cardiac rehospitalization roughly 50% higher as well. A similar study looking at aspirin discontinuation in stented patients found that rates of ST-segment elevation myocardial infarction in those who had discontinued aspirin within the previous month was double that of those who continued taking aspirin.
Hypertension and hyperlipidemia therapies have been proven prospectively to substantially reduce mortality. Indeed, the Scandinavian Simvastatin Survival Study (4S) trial became one of the first in cardiology to demonstrate a statistically significant decrease in all-cause mortality from a single-drug therapy. In those with <20% adherence to study medications, published data from the Anglo-Scandinavian Cardiac Outcomes Trial–Lipid-Lowering Arm (ASCOT-LLA) showed a doubling of event rates for stroke, myocardial infarction, and angina pectoris. On the basis of statistical analysis, it has been estimated that increasing compliance with hypertensive and lipid therapies from partial (20% to 80%) to ideal (>80%) rates would avoid 800 myocardial infarctions and 600 strokes per 100,000 patients, or roughly 1 major event per 70 patients with improved compliance.
Patient Perceptions
The reasons for suboptimal adherence to medications vary from survey to survey. A sampling of the most common reasons given for nonadherence from a survey of >150 nonadherent patients with heart failure is listed in Table 1 . Among nonadherent patients with hypertension, side effects were the single largest concern, followed by a general dislike of medications and a feeling that the medicines were unnecessary. Recognition of the diversity of these concerns serves as a backdrop against which specific issues may then be further illustrated.
| Reason | Frequency |
|---|---|
| Cannot remember to take on time | 20% |
| Too expensive | 16% |
| Too many medications | 10% |
| Don’t know how or when to take | 9% |
| Side effects | 8% |
| Other patient education–related issues | 15% |
| Physically unable or too ill | 10% |
Patient Perceptions
The reasons for suboptimal adherence to medications vary from survey to survey. A sampling of the most common reasons given for nonadherence from a survey of >150 nonadherent patients with heart failure is listed in Table 1 . Among nonadherent patients with hypertension, side effects were the single largest concern, followed by a general dislike of medications and a feeling that the medicines were unnecessary. Recognition of the diversity of these concerns serves as a backdrop against which specific issues may then be further illustrated.
| Reason | Frequency |
|---|---|
| Cannot remember to take on time | 20% |
| Too expensive | 16% |
| Too many medications | 10% |
| Don’t know how or when to take | 9% |
| Side effects | 8% |
| Other patient education–related issues | 15% |
| Physically unable or too ill | 10% |
Types of Medication Nonadherence
Although most practitioners would agree that medication nonadherence can be loosely defined as a failure to take medications as prescribed, further classification can be helpful to separate the circumstances into 3 distinct categories.
Defaulting is a failure to start a therapy entirely, either after hospital discharge or on completion of a clinic visit. Causes include failure of communication by facility or provider, loss of prescription, and doubts about the efficacy of the therapy itself. In clinical practice, a default rate of 10% to 20% has been repeatedly demonstrated.
Nonpersistence is failure to continue reliably (usually defined as taking >80% of prescribed doses) with a therapy already started without a rational reason to do so. Causes include the complexities of polypharmacy, absence of motivation caused by depression, issues of medication costs, and losses to follow-up with providers.
Rational nonadherence is the cessation of a prescribed therapy because of concern for, or the presence of, medication side effects.
Each of these factors plays a significant role in the overall problem of nonadherence and the complications that follow.
Specific Causes of Defaulting and Nonpersistence
Provider communication is where any discussion of adherence must begin. Defaulting, nonpersistence, and rational nonadherence can all be affected by the effectiveness of communication between doctors and patients. In a 2006 study, Tarn et al objectively evaluated recorded sessions in which physicians were prescribing new medical therapy. In patients being seen for cardiovascular complaints, some measures were particularly well addressed, including the names and purposes of medications (>95%). Patients were much less likely to receive any discussion of the frequency or timing of dosing (63%), adverse effects (67%), or the duration for which they should continue therapy (17%). The results were similar whether primary care providers or cardiologists were educating patients. It has been demonstrated that 1-time education at the time of prescription or discharge halves the rate of initial nonadherence, but this “teaching moment” must be an effective one.
Polypharmacy continues to evolve alongside new additive and synergistic therapies for heart disease. In just the period from 1998 to 2001, the number of drugs prescribed for elderly patients with heart failure at hospital discharge increased by 12%, with the average patient taking 7 or 8 separate drugs for a total of 11 daily doses. At the upper end of the curve, 10% of these patients were prescribed >18 doses per day.
Studies evaluating dose frequency have shown rapid decreases in adherence rates with increasing dose frequencies ( Table 2 ). Patients asked to take 1 medication at 3 separate dosing times are, on average, only 65% compliant, a substantial decrease from the 79% compliance with once-daily medications. This problem is compounded further by the matter of timing. Even when taking the correct number of doses, <1/2 of all patients taking thrice-daily medications are able to take all 3 doses within 2 hours of the prescribed times. As a result, doses are often doubled when remembered or otherwise missed entirely.
| Dosing | Took Most Doses | Took on Time |
|---|---|---|
| Once daily | 79% | 74% |
| Twice daily | 69% | 58% |
| 3 times daily | 65% | 46% |
| 4 times daily | 51% | 40% |
Similarly, the number of medications taken together directly affects patient success in taking drugs as prescribed. In a study of >8,000 patients prescribed medications for hypertension and hyperlipidemia, only 1/3 were able to take >80% of their medications at 6 months. Of those who did, adding just 2 additional drugs to the regimen decreased overall compliance by another 28%, with further reductions as more medications were added.
Conventional strategies that have been identified to improve compliance in the setting of multidrug regimens include starting drugs during the course of a hospitalization, starting multiple drugs together to minimize the overall number of changes to the medication list, and reducing the number of conditions under treatment at any given time.
A recent trend in the effort to improve compliance with difficult regimens has been the creation of combination drugs, such as the now generic pill combining lisinopril and hydrochlorothiazide. A meta-analysis of such drugs showed a 20% relative reduction in the rate of nonadherence compared to component drugs taken separately. More than 20 combination drugs for cardiovascular disease are now on the market.
Taking this concept 1 step further is the so-called polypill. After Yusuf published a review illustrating how aspirin, β blockers, angiotensin-converting enzyme (ACE) inhibitors, and statins together could theoretically decrease worldwide cardiovascular events by 75% in at-risk patients, Wald and Law extrapolated further, theorizing that a pill containing low doses of multiple medications could be safely given to patients deemed at risk by age alone, regardless of other risk factors. According to their analysis, doing so could reduce worldwide burdens of cardiovascular disease by ≥80%. That conclusion, based on a meta-analysis, was highly controversial and sparked heated discussion across the field of cardiology. Nonetheless, a combination of thiazide, atenolol, ramipril, simvastatin, and aspirin called Polycap (Cadila Pharmaceuticals Ltd., Ahmedabad, India) has recently been assessed in a phase II clinical trial. With >2,000 patients assigned to either Polycap or 1 of 8 arms containing only a portion of its components, each of the 9 groups was followed for 12 weeks. The medication was well tolerated, discontinued in 16% for Polycap compared to 14.6% for aspirin alone and 14.8% across all arms. Efficacy appears to be high as well, with Polycap achieving reductions in systolic and diastolic blood pressure, heart rate, and cholesterol as substantial as any other arm of the study. These results were durable throughout the duration of active treatment. Further studies remain ongoing.
Depression is widely associated with heart disease and can have substantial effects on medication adherence. One in 3 patients with congestive heart failure, recent myocardial infarction, or acute coronary syndromes will meet criteria for either major or minor depression, a figure consistent across all 3 conditions.
Even mild depression is sufficient to dramatically alter compliance with essential therapy. A study of patients with acute coronary syndromes found that the nonadherence rate by electronic medication monitoring in those without depression was 15%, doubling to 30% in those with mild depression. Those with moderate to severe depression were only slightly less adherent at 37%, suggesting that the presence or absence of depression is far more important to medication adherence than the degree of depression ( Figure 1 ). Few studies have looked at the longitudinal association of compliance and adherence, but 1 study found that each point of improvement over time in the Beck Depression Inventory was associated with a 1% absolute improvement in medication adherence.
Medication costs represent a key source of nonadherence in all fields of medicine, particularly in patients with low or fixed incomes, those with chronic medical conditions, and those on disability. In 2007, the United States spent an average of almost $750 per resident, 10% of all medical costs, on prescription drugs. Since 1960, this figure has grown an average of 7.4% per year, keeping identical pace with the national gross domestic product, but far outstripping growth in household and individual income. Thus, as medication costs are increasingly shifted from government and employers (through insurance) to individuals (through copayments and noncovered drugs), it is little surprise that even employed, insured patients are reporting an inability to keep up with increasing costs. Today, 1 in every 7 Americans is not taking a prescribed drug therapy because of financial constraints.
Treated according to current evidence-based guidelines, a newly stented patient with ischemic cardiomyopathy could see drug costs in excess of $3,900 over 1 year ( Figure 2 , Table 3 ). Indeed, actual costs of medications for elderly patients with heart failure were $3,823 per year in 2001 and have increased by an average of 8% per year.
