Adherence to Drug Therapy

Introduction

This book describes the panoply of medications available to treat a broad array of skin diseases. The outcomes of these treatments are not entirely predictable. In clinical trials, treatment response varies among research subjects. This variability is even greater in the clinic setting and has resulted in the development of the field of pharmacogenomics; but, a key underlying and often underappreciated determinant of the outcome of dermatologic treatment is simply how well patients use their medications, a concept subsumed under the term ‘adherence.’

Q4.1 Adherence (previously known as ‘compliance’) is the degree to which a patient’s behavior coincides with the recommendations of their healthcare provider.

Poor adherence is ubiquitous across all fields of medicine. Q4.2 Improving patients’ adherence behavior may be a quick, low-cost effective way to improve medical outcomes. Nonadherence to medication regimens is responsible for 10% of all hospital admissions and costs the US health care system US$100 to $300 billion annually.

The problem of nonadherence manifests in many ways in the treatment of dermatologic disease, particularly with chronic topical treatment. The problem of tachyphylaxis (‘the more you use the medicine, the less it works’) is commonly a manifestation of poor adherence (‘the less you use the medicine, the less it works’). Other phenomena that are best explained by patients’ adherence (or lack of adherence) include the resistance of scalp psoriasis to topical treatment and the tendency of children with atopic dermatitis previously ‘resistant’ to topical treatment to clear rapidly—in just 2 or 3 days—with topical treatment administered in the hospital setting.

This chapter will (1) describe how adherence is measured, (2) examine the magnitude of the problem of adherence, (3) describe the variables that influence adherence behavior, and (4) identify principles and practical strategies to improve patients’ adherence to treatment.

Measures of Adherence

Assessing adherence can be challenging. Self-reporting of adherence is not a particularly reliable methodology. A more objective measure to monitor adherence is blood test monitoring for drug levels. However, this approach is not helpful for locally acting topical treatments and often overestimates adherence to systemic medications because patients tend to be better about using their medications around the time of an office visit, a phenomenon called ‘white coat compliance’ (one of the best examples of this phenomenon is the tendency for people to floss more before going to the dentist).

In the past, studies on adherence in dermatology relied on surveys, pill counts, or medication weights. In an anonymous survey of psoriasis patients, roughly 40% reported nonadherence to their treatment regimen; we suspect most of the other 60% were not completely honest. Counting pills or weighing a topical medication may appear to be a more objective way to assess adherence, but patients may dump their medication to hide their poor adherence behavior. Some creativity may help. To encourage honest responses, the physician may ask, ‘It can be hard to use the medicine every day. How often are you missing doses: every day or every other day?’ or, for self-injected medications, ‘Are you keeping the extra syringes you’ve accumulated refrigerated like you are supposed to?’ To improve the reliability of pill counts, instead of prescribing 60 tablets for a twice a day dosing and having the patient return in a month, either prescribe 70 tablets or prescribe 60 but have patients come back before the month is over. That way, if the bottle is empty, the physician can distinguish a bottle that has been emptied as a result of good adherence from one that was emptied to hide poor adherence.

Other approaches have been used to more reliably assess patients’ use of their medications in research studies. Q4.3 Pharmacy data can be used to tell if and when patients filled a prescription. Data on refills can establish an upper limit on patients’ use of medication by describing how many days’ worth of medication the patient obtained over the course of treatment. The standard metric for this is the ‘medication possession ratio.’ The medication possession ratio is calculated as the number of days of supply for dispensed prescriptions divided by the number of days between prescription refills.

Pharmacy data are useful for determining whether patients procure medication, but they are a step removed from whether the patient uses the treatment. Q4.4 Objective electronic monitors in the caps of medication containers—such as the Medication Electronic Monitoring System (MEMS)—provide more direct information on patients’ use of medication. These devices record the date and time each time the bottle or tube is opened. Electronic monitoring of adherence can be used to elucidate the various patterns of poor adherence. The initial phase of adherence is ‘acceptance,’ when the patient fills and begins treatment. Patients with poor ‘acceptance’ may not fill the medication or may begin taking it after some delay. ‘Persistence’ refers to how long patients stay on therapy. Poor persistence implies that patients discontinued their treatment early. During the period when patients are using their medication, ‘quality of execution’ refers to how well the patient is using the medication. Patients may skip doses at regular intervals, take drug holidays, or overtreat. The typical patient probably is doing some combination of all of these.

The Magnitude of Poor Adherence in Dermatology

Electronic monitors and prescription refill data have opened up new vistas in our understanding of how well patients use the medication we prescribe. Pharmacy data reveal that many prescriptions are never filled. A groundbreaking Danish study found that 4 weeks after the doctor visit, 30% of dermatologic prescriptions had not been filled; 50% of the psoriasis prescriptions had not been filled. In a study population of North Carolina Medicaid patients (for whom the costs for procuring medication were very low), the medication possession ratio was only 35% for nonbiologic psoriasis treatments. For biologics, the medication possession ratio was considerably better, at 66%, but this indicates that on average those patients missed taking at least one-third of recommended doses.

Studies of pharmacy data inform us about how much medication the patient received; electronic monitors provide a better assessment of how often medication was used. In a study of patients with psoriasis, adherence to twice-daily 6% topical salicylic acid was measured over 8 weeks using self-report diaries, medication weights, and MEMS caps. Although the subjects reported adherence rates of 90% to 100% in their treatment diaries and medication weights, the electronic monitors revealed an overall average adherence of only 55%, with a drop in adherence of approximately 20% every 5 weeks. Electronic medication monitors have revealed that patients’ adherence behavior is far worse than previously thought.

Certain patient populations are notoriously nonadherent, particularly teenagers. In one small study of teenage acne patients, electronically monitored adherence to a once-daily treatment regimen was 82% on day 1, dropping to 45% at the end of 6 weeks. In a study of children with atopic dermatitis, electronic monitors were used to assess adherence to topical triamcinolone over 8 weeks. Mean adherence was an abysmal 32%, with a drop of 60% to 70% in adherence over the first 3 days of the study.

Much of the data on adherence come from patients who are knowingly enrolled in studies of adherence (although they usually do not know how adherence is being measured). Thus, patients in these studies are probably motivated to be more adherent than they would under normal circumstances. Yet, even in these clinical studies, lasting no more than a few weeks to months, adherence rates are suboptimal. One can imagine just how poor adherence must be in regular dermatology practice, particularly among patients with chronic diseases who require daily medications, not just for a few weeks or months, but for a lifetime.

Factors that Influence Adherence Behavior

Adherence behavior is affected by many interacting variables. There is a complex relationship between adherence behavior, treatment, and outcome. Q4.5 The factors that affect adherence can be classified as internal and external ( Table 4.1 ). A major internal factor affecting adherence is the patient’s motivation to get better. Some patients may not be particularly motivated to get better, or they may not be bothered by their condition. Other patients may be seeking secondary gain from their condition. One might expect that patients with severe disease would be very motivated to use their medications, but patients with worse quality of life can be less likely to be adherent. Perhaps these patients feel hopeless and simply resign themselves to their condition. Sometimes patients want to get better but may not have understood the medication instructions, or they may be forgetful or unable to physically apply the medication.

Patients may have psychiatric comorbidities, such as depression, that interfere with their ability to carry out their treatment. Age is another important factor, as children and teenagers are less likely to adhere to treatment. Additionally, patients’ understanding of their disease and the expectations for treatment affect their adherence behavior, as does their understanding of the treatment itself, in particular expectations about adverse effects (AE). Of profound importance is the patient’s trust in their doctor and the quality of the physician–patient relationship. Overall, 70% of adherent patients reported that they were using their medication because they believed their provider was a compassionate advocate.

Q4.6 External factors affecting adherence behavior include (1) complexity of regimen, (2) cost of and access to treatment, (3) vehicle choice in topical medications, and (4) AE. In addition, (5) the time it takes a medication to work is critical, as is (6) the patient’s response to treatment. For example, if a medication takes weeks to work but the patient is expecting a ‘quick fix,’ then that patient may stop using the medication, thinking that it failed. Conversely, if the patient has rapid initial success with a particular treatment, he or she may be more likely to continue using it. Other important considerations over which physicians have considerable control include the quality of the physician–patient relationship, plans for follow-up visits, and clarity of instructions provided.