Hypertension is a disease that can be treated but rarely cured. Thus, patients are usually informed that therapy will be maintained quasi for the rest of their life (Chap.​ 31). The long-term management of hypertension actually generates many obstacles that may affect the capacity of patients to stay on therapy. The first obstacle is the acceptance of the diagnosis and the difficulty for some patients to initiate a course of lifelong treatment for a silent asymptomatic disease. Depending on the quality of the information they received, patients may not be motivated. Thereafter, patients may be confronted with the first side effects of the prescribed drugs. In the absence of symptoms before treatment, side effects may not be accepted and tolerated. With time, the treatment scheme may become more and more complex as it is well recognized that 70 % of patients will need more than one drug to reach the target BP recommended by guidelines (Chaps.​ 30 and 41). Later, patients may be exposed to the delayed side effects such as sexual dysfunction. These factors will further decrease the motivation to remain under treatment, and for many patients, the necessity to pursue the treatment will be questioned leading in most cases to a discontinuation of therapy due to a low persistence. A recent analysis by Simpson et al. has actually shown that patients with an excellent adherence to therapy have a global positive attitude toward all preventive recommendations for a better health [21]. Thus, this type of patients has a better survival even if they receive a placebo as demonstrated recently in the post hoc analysis of a heart failure trial [22]. However, it is also important to notice that in some cases, a good adherence may have damaging effects if the drug has serious adverse effects, and patients continue to take them as, for example, nonsteroidal anti-inflammatory drugs or antiarrhythmic agents (Chap.​ 42).

Besides these therapy-related factors, several other barriers and risk factors for interrupting treatments have been identified (Table 43.1). These barriers come not only from the patient him-/herself but also from the family, the physician, the nature of the treatment, and the healthcare system. These include depression, other comorbidities, personal as well as family beliefs on hypertension and on the necessity of treatment, lack of knowledge about hypertension, cost of medications, the use of alternative medicines, memory problems in elderly patients, and poor quality of life.

Several studies have clearly identified the characteristics of patients at high risk of not being adherent to their treatment. These are essentially young, active men, hypertensive patients of Afro-American origin, and elderly patients with cognitive dysfunction [23]. Gender does not seem to be a major determinant although some studies have suggested that adherence to therapy is better in females than males. The same is true of the educational level. The health system per se and particularly the copayment has also a major impact on the long-term persistence of patients treated for hypertension [24]. Thus, it is very important for physicians to identify these high-risk factors, as strategies may be developed to prevent any interruption of therapy as will be discussed later in this review.

It is well recognized that adherence to medication is an important determinant of the BP response to treatment and hence of the clinical benefits of antihypertensive therapies. In the last two decades, many clinical studies have investigated drug adherence in patients with hypertension in most cases using the medication event monitoring system (MEMS) which enables to follow the adherence on a day-to-day basis for a long time period. These studies aimed not only at defining adherence in hypertension but also at investigating the ability of the monitoring system to support drug adherence [25, 26]. Recent reviews have confirmed that adherence is highly variable but relatively high when measured in hypertension. This high adherence may be due to the measurement bias, as adherence tends to increase as soon as it is measured. In addition, a careful monitoring of adherence might help. However, in one control study, it did not really improve BP control but rather prevented from changes in drug therapies [26]. Nonetheless, many interesting aspects of the adherence process have been described with this approach. The first and probably the most important one is that drug adherence is a very dynamic process, which can hardly be summarized with a single number. Indeed, patients may be adherent for some periods and less adherent in other circumstances, for example, during weekends and holidays. Second, it is very difficult to define a cutoff point above which drug adherence is sufficient and acceptable. Indeed, in the literature, an arbitrary cutoff of 80 % is used to define a good adherence, but it is obvious that 80 % can be achieved in many ways with different clinical impacts, for example, missing 1 day every 5 days or 1 week every 5 weeks. Moreover, the impact of missed doses on BP control and risk reduction depends largely on the pharmacological profile of the prescribed drugs [27]. This is the reason why it is generally recommended to use long-acting medications in patients at risk of poor adherence. Thirdly, many investigators have tried to demonstrate a correlation between the level of BP achieved under treatment and the percentage of doses taken. These correlations have sometimes been obtained [28] but they have generally been very weak. In fact, there are many good reasons why this type of correlation should not be very high. Indeed, high BP values can be obtained in nonadherent hypertensive patients as well as in adherent patients insufficiently or inadequately treated.

Despite many of these limitations, increasing evidence gathered from large databases has shown that patients with a good adherence have a better BP control [29]. Moreover, highly persistent hypertensive patients have a greater reduction of their cardiovascular risk than patients interrupting their treatment [30]. At last, a good adherence to antihypertensive medications has been associated with a significant reduction of the risk of coronary heart disease and congestive heart failure and cerebral diseases [31–33].

Apparent resistant to therapy is one clinical situation in which the question of drug adherence becomes particularly important. Indeed, when the prescribed antihypertensive drugs do not lower BP as expected, physicians are confronted to two crucial questions: is the patient a nonresponder to therapy or is the patient not taking his/her drug as recommended thus being a nonadherer? In the absence of any adequate tools to measure drug adherence, physicians are left with only one option, i.e., to consider the patient as a nonresponder. Thus, physicians will probably increase the drug doses or add another drug in order to achieve the therapeutic targets. However, it is obvious that if poor adherence is the real problem, adding new drugs is a completely wasted action, which will only aggravate the situation.

With the recent development of renal denervation and baroreflex activation to treat patients with resistant hypertension, a sudden interest for resistant hypertension and its causes has emerged [34] (Chap.​ 42). Several analyses have investigated the potential causes of apparent resistance to drug therapy in hypertension. Not surprisingly, poor drug adherence has been identified as one important factor of pseudo-resistance. Yet the true incidence of poor adherence remains largely unknown in this patient population because adherence was rarely measured adequately. Thus, published figures range between 10 and 50 % [16, 17, 35–39]. In our experience, using electronic monitoring of drug adherence, about one third of patients with resistant hypertension had adherence problems leading to uncontrolled hypertension [40]. More recently, three studies have been conducted among hypertensive patients with uncontrolled BP or candidates to renal denervation to assess drug adherence using measurements of either urinary or blood concentrations of drugs [16, 17, 19]. Two of these studies reported almost 50 % of partial or complete nonadherence and the other 23 % of nonadherence in renal denervation candidates. These figures are important and may well be underestimations as they do not take into account the white coat adherence phenomenon discussed earlier in this review [41]. Figure 43.3 illustrates the limits of this approach on the nature of adherence. Interestingly, poor adherence to the medication has been found in inpatients as well as in outpatients [19]. In hypertensive patients with coronary heart disease, resistant hypertension also appears to be common (38 %) and to be associated with a worst cardiovascular prognosis [42].

In any case, it is now increasingly accepted that poor adherence is a real concern in resistant hypertension. Despite the availability of adequate noninvasive methods, adherence to therapy remains largely underdiagnosed and almost never measured in clinical practice. In recent studies, some investigators have used the directly observed treatment (DOT) or “tablet feed” which is commonly used in the management of tuberculosis, to evaluate the role of poor adherence in mediating uncontrolled BP [43, 44]. Although a small number of patients were enrolled in some of these studies, it clearly showed that many patients actually normalized their BP when the treatment was given under control. Thus, taken together, these new data suggest that adherence to the drug regimen is clearly a critical issue in resistant hypertension, and we believe that a good assessment of adherence in patients not responding to drug therapy is essential to take rational therapeutic decisions [40]. However, today, physicians are still limited by the lack of cheap and easy to use methods of monitoring adherence in clinical practice.

Multiple ways to improve drug adherence have been investigated in hypertension and chronic diseases in general. The proposed strategies concern the patient, drug therapy, pharmacist, physician, and the entire healthcare system.

The first and certainly the most important step is providing careful information to the patients on the goals of therapy and on the means to achieve the objectives. Indeed, in order to be motivated to stay on therapy, patients need to know the risks linked to their disease and to consider them as serious. They have to know the BP goals and the effects expected from therapy. They also have to be informed on the means to achieve these goals to acquire specific competences and to develop a set of strategies to reach the targets, to cope with the disease, and to autoregulate their actions. Finally, patients have to believe in their personal efficacy. Thus, the empowerment of the patient is critical [45]. In this respect, recent studies have suggested that self-monitoring and even self-uptitrating of drugs by the patients may be helpful to improve BP control even in high cardiovascular risk patients [46]. Thus, home BP monitoring may be a good recommendation in patients who have difficulties with their treatment and are at high risk of not staying on therapy [47]. On their side, physicians also need to be motivated and should avoid medical inertia in order to show patients that they are eager to achieve the BP goals in order to preserve their patients’ health [48].

One can also act on the prescribed regimen to support drug adherence over time. Indeed, it is well recognized that the number of drugs patients received and the dosing frequency have a major impact on persistence [49–51]. Several studies have demonstrated that simplifying the drug regimen with the use of fixed-dose combinations has a positive impact on adherence and persistence [52–55]. A recent meta-analysis confirmed that simplification of therapy is an effective therapeutic approach in hypertension that improves adherence though the impact on BP control is relatively modest [54]. This finding was actually confirmed in a Cochrane analysis of different strategies to improve adherence in hypertension [56]. Interestingly, it seems that beyond the number of tablets per day, the perception of the burden of the disease and of the therapy is predominant [57]. Indeed, patients may accept to follow a treatment including a high number of tablets per day provided they are convinced of the clinical benefits. Regarding the adaptation of drug therapy, it is also recommended to prefer the prescription of combinations of long-acting drugs which blunt the effect of isolated missed doses. Studies have shown that BP is more likely to remain adequately controlled despite the omission of one or two doses if the antihypertensive drug has a long half-life [58].