Monitoring of drug adherence in hypertension

by M. Burnier, Switzerland

Service of Nephrology and
Hypertension, University
Hospital Lausanne

It is today well accepted that poor adherence is a real concern in hypertension. With the absence of new drugs to treat hypertension, it is crucial to improve blood pressure control with available antihypertensive agents by working on adherence. Unfortunately, nonadherence to therapy remains largely underdiagnosed in clinical trials, despite the availability of adequate new noninvasive methods, and it is almost never measured in clinical practice. In order to make good therapeutic choices about the treatment of resistant hypertension, it is vital to systematically assess adherence in patients who fail to respond adequately to blood pressure–lowering medication. The emergence of new, less expensive, and less complicated ways of monitoring adherence (eg, adherence apps) should help in this objective. It is of the utmost importance to remember that hypertensive patients on therapy, but with inadequate adherence are at greater cardiovascular risk. Conversely, good adherence to antihypertensive medication is associated with a lower incidence of cardiovascular and renal complications. Lastly, poor adherence has an enormous negative impact on health care costs. All efforts made toward improving drug adherence therefore represent a win-win situation that satisfies all parties in the health care network, ie, patients, physicians, other health care providers, and payers.

The prevention of cardiovascular and renal complications due to hypertension, dyslipidemia, diabetes, or obesity is based essentially on the recommendation of lifestyle changes—such as increasing physical activity, lowering body weight, smoking cessation, and reducing salt intake—and of drug treatment.1 To this purpose, physicians prescribe antihypertensive therapies, lipid-lowering drugs, and antidiabetic agents to achieve the blood pressure, lipid, and glucose levels recommended by international guidelines.

Epidemiological surveys have, however, repeatedly demonstrated that only a fraction of patients treated for hypertension, dyslipidemia, or diabetes actually reach therapeutic goals and, even in the best cases, only around 50% of treated patients reach targets. There are several potential explanations for these poor results. Among them, one can cite: medical inertia (when physicians do not intensify treatment when necessary); and deficiencies of health care systems in their approach to chronic diseases, leading to limited access to effective treatments and poor adherence to therapy.1 Thus, in a recent analysis of drug adherence in a large group of patients treated for primary and secondary prevention of cardiovascular diseases, low adherence to long-term therapy (around 50% or less) was found with antihypertensive agents as well as with lipid-lowering treatments and aspirin, suggesting that poor adherence is common in this clinical context.2 The implicit consequence of these observations is that uncontrolled patients remain at high risk of developing cardiovascular complications, sometimes at levels similar to those of untreated patients.3

Drug adherence in hypertension

It is obvious that adherence to antihypertensive therapy is an important determinant of the blood pressure response to treatment. In the last two decades, drug adherence has been measured in many studies. These have not only aimed to characterize drug adherence in hypertension, but also to investigate how adherence could be enhanced or supported.4,5 Lessons learned from these studies are that adherence is highly variable, but relatively high when quantified in hypertension. This latter observation was rather surprising, but these instances of high adherence may be due to measurement bias. Indeed, as soon as patients are told that their adherence will be monitored, their assiduity in taking their pills regularly improves substantially, at least during the first months of monitoring. With the use of devices that enable us to obtain a dosing history, many interesting aspects of the adherence process have come to light. The first and probably the most important one is that drug adherence is a dynamic process that is hard to summarize with a single number. Patients may be adherent at certain periods and less so at other times, for example, during weekends, holidays, or when problems associated with family life or work occur.

Because of the variability and difficulty in quantifying adherence, establishing a cut-off level for acceptable treatment adherence is not easy. In medical literature, the cut-off arbitrarily chosen to define “good” adherence is 80%. This cutoff can be obtained by missing one day’s treatment every five days or equally by missing one week’s treatment every five weeks; however, the clinical consequences of the missed treatment in these two scenarios are unlikely to be the same. It should be noted that the impact of different antihypertensive drugs on blood pressure control and cardiovascular risk reduction depends on their individual pharmacological profiles, with long-acting drugs preferable for patients who find adherence problematic.6

At an individual study level, some studies have been unable to demonstrate a relationship between adherence and blood pressure level achieved with treatment while, in others that have shown a relationship to exist, it was weak.7 One reason for this weakness is that high on-treatment blood pressure values may also be caused by unsuitable treatment or too low a dosage of an appropriate medication. Large-scale analyses have, however, concluded that patients with good adherence do have better blood pressure control and a reduced cardiovascular risk.8,9 The risk of coronary heart disease, heart failure, and cerebrovascular disease is reduced in hypertensive patients who adhere well to treatment.10-12

The recent revival of drug adherence

In the field of hypertension, the problem of poor adherence to drug therapy has not received much attention in the past and, hence, has been rather overlooked. For several decades, physicians and researchers focused on the development of new drugs and combination therapies that were supposed to improve blood pressure control in hypertensive populations. This is no longer the case. Adherence has recently come back to the forefront of hypertension research with the development of interventional therapies, such as renal denervation13,14 or carotid baroreflex stimulation,15 for the management of patients with apparent resistant hypertension.

Determining whether a patient with resistant hypertension is taking their antihypertensive medication as prescribed is critical. If antihypertensive medication fails to reduce blood pressure, two explanations are possible: the patient is nonresponsive to therapy; or the patient is nonadherent and not taking his medication as advised. Up until recently, patients with resistant hypertension would have automatically been categorized as nonresponders rather than nonadherent because of the lack of means of and great difficulty in assessing medication adherence. This automatic categorization as “nonresponsive” would have then led to the addition of other antihypertensive agents and/or an increase in dosage. Neither of these measures would have any effect in a patient whose true problem was nonadherence, and this absence of effect would be incorrectly attributed to “resistance” rather than nonadherence, further complicating an already muddled situation. Moreover, before intervening with an expensive procedure, it is important to be sure that candidates for renal denervation are truly resistant.16 When potential causes of apparent resistance to drug therapy in hypertension were investigated, poor drug adherence was identified as one of the most important factors of pseudoresistance. Despite its importance, the true incidence of poor adherence in resistant hypertension remains imprecise because adherence has rarely been measured adequately. Published figures range between 10% and 50%.17-23

Monitoring drug adherence: a clinical challenge

Measuring drug adherence in clinical practice has never been a simple issue, whereas in clinical trials measurement is standardized: assessment of adherence is essentially based on pill counts and the return of unused pills. In fact, there are many other published techniques that can be used in different clinical contexts. Physicians use some of them, but we ought to acknowledge that the majority of physicians dedicate very little time to this issue even though they recognize that adherence is an important parameter to assess in chronic diseases. Interviewing a patient about his or her medication and treatment adherence is not particularly accurate (≈30% chance of detecting nonadherence), but it is easy and relatively quick. A quick interview is particularly valuable as a prompt in patients looking for an opportunity to engage in discussion; for example, it will induce certain patients—those who never started treatment or those who stopped treatment—to explain why they decided to do what they did. That said, a physician’s intuition by itself is unfortunately not a reliable gauge of adherence, nor are other indirect markers of treatment adherence, like drug-induced adverse events, clinical response, or consultations attended. Nonetheless, a careful, nonjudgmental discussion about adherence to therapy should remain the first step in any investigation on drug adherence. Physicians should be better trained on how to formulate their questions in order to leave the patient space and time to eventually communicate their problems with drug therapy. Table I shows examples of questions that can be used by physicians, as suggested by Brown et al.4

Table I. Types of question that physicians could ask in order to
assess a patient’s medication adherence.
Adapted from reference 24: Brown and Bussell. Mayo Clin Proc. 2011;86(4):
304-314. © 2011, Mayo Foundation for Medical Education and Research.

Several questionnaires have been created to assess adherence. These questionnaires are generally used in clinical studies, but they could also be used in clinical practice. Their results tend to overestimate adherence, as patients often forget the episodes when no medication was taken. The most popular questionnaire is the Morisky questionnaire,25 which is easy and rapid to use in its simplified version. The four items of the simplified questionnaire are: (i) Do you ever forget to take medications? (ii) Are you ever careless in taking your medications? (iii) Do you ever miss taking your medications when you are feeling better? and (iv) Do you ever miss taking any of your medications because you are feeling sick? Each item has a yes/no response option and one “yes” provides one point. A score of ≥2 is indicative of low adherence to drug therapy. When compared to electronic measurements of adherence that provided a dosing history, the Morisky questionnaire was found to overestimate drug adherence.

Pill counting, a favorite and generally effective method of assessing drug adherence in clinical trials, overestimates adherence to treatment too because patients tend to give back empty boxes. For calculating long-term treatment adherence in epidemiological studies, the percentage of days covered by prescriptions can be determined by regular checking of prescription refills using the records from electronic pharmacy dispensing systems.26 This type of continuous registry is useful for estimating persistence with therapy and establishing risk factors associated with poor adherence.27,28 Weaknesses of this method, however, are the assumptions that patients take their medicine as they should and that patients always use the same pharmacy. Patient education and comprehensive monitoring of all medication distribution channels could help address these weaknesses.

In the future, drug adherence monitoring in clinical trials might be performed using a new system of ingestible sensor included in every single pill. The Food and Drug Administration has accepted this system, developed by Proteus Digital Health, as a new approach to monitor adherence in clinical studies.29 It has the advantage of proving drug ingestion has occurred, but it still relies on patient compliance: the patient has to apply a patch to his or her skin in order for data to be transmitted (Figure 1, page 310).

Continuous electronic pill-box monitoring as well as intermittent direct measurements of plasma or urinary drug levels can also provide relevant and trustworthy information about drug adherence. In the first instance, if the pill box is opened it is assumed the drug is ingested. I believe that some researchers worry unduly about the ability to prove that medication has actually been taken with electronic systems. Over the long term (several months), a patient is unlikely to routinely open an electronic pill box for the express purpose of systematically discarding his or her medication, so perhaps this worry is misplaced or at least not as great as one might believe. Long-term monitoring is likely to be dependable.

In the second instance, with plasma and urine drug monitoring, a positive result proves that a drug was ingested, but not when or how often or how many doses were forgotten. A negative result is easy to understand: the medication was simply not taken. Direct measurement of drug levels, though expensive and arduous, is becoming more popular in resistant hypertension,20,21 but white-coat adherence bias can exaggerate the real level of adherence. Clinical trial investigators are normally obliged to inform patients that they will be taking blood or urine samples for testing and some patients, in anticipation of the tests, will become more adherent just before and after a planned visit.30

In studies where patients were not informed of blood and urine testing, treatment adherence in patients with resistant hypertension has been found to be especially limited.20,21,31

Figure 1. Illustration of how an ingestible sensor works in wirelessly observed therapy.
Abbreviation: TB, tuberculosis.
From reference 29: Belknap et al. PLoS One. 2013;8(1):e53373. © 2013, Belknap et al.

About half of resistant hypertension patients and about a quarter of renal denervation candidates are nonadherent (partially or completely). Our study of electronic monitoring of drug adherence showed adherence problems affected approximately one in three patients with resistant hypertension.32 Yet, measuring drug levels and communicating the results to patients may have a favorable impact on adherence, as was observed in one clinical trial involving patients with resistant hypertension.33 In a recent analysis, Schmieder et al have clearly shown that some patients change from being adherent to being partially or totally nonadherent after renal denervation.34

Finally, in recent studies, some investigators have used 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 blood pressure.35,36
Although a small number of patients were enrolled in some of these studies, these clearly showed that blood pressure actually normalized in many patients when the treatment was given under controlled conditions. The strategy used was to couple directly observed treatment with the measurement of 24-hour ambulatory blood pressure. Patients were asked to come to the unit with their drugs. These were taken under supervision and blood pressure was monitored thereafter over 24 hours.37 In some cases, this approach may be useful and may avoid the use of drug level measurements, which are more costly.

As mentioned in a recent review, the ideal method to assess drug adherence in clinical practice should “provide a reliable capture, storage, analysis, and communication of dosing history data in ways that make it difficult or impossible for patients or trial staff to censor or otherwise manipulate the data.”38 Nowadays, three methods are close to fulfilling these criteria: the retrospective analysis of prescription refill records for epidemiological studies, the analysis of chemical markers of drug exposure, and the automatic electronic monitoring of adherence.38

The future of drug adherence monitoring will certainly include the use of mobile health technologies and apps, connecting patients with their physicians or with other members of the health care team. Today, several apps are available on the market, but so far none of them has really been convincing in terms of diagnosing poor adherence. They might be rather more useful in supporting adherence in the long-term management of hypertensive patients. Today, a large clinical trial is ongoing with an app, the Eurohypertension app, developed in part by the European Society of Hypertension. Preliminary results with this approach have suggested an improvement in blood pressure control, implying better adherence to therapy among patients using the app.39


New antihypertensive drugs are lacking so making the best use of those we already have, which includes optimizing adherence to therapy, is a logical move. Though nonadherence is underdiagnosed in clinical trials and rarely measured in prac- tice, it is a real and costly challenge in hypertension. Hypertensive patients who take their treatment irregularly are at greater cardiovascular risk,9 while those with good adherence see their risk of outcomes diminish.10-12,40 In patients with resistant hypertension who respond unsatisfactorily to blood pressure–lowering medication, the systematic assessment of adherence is particularly crucial.32 The ongoing development of modern, inexpensive, and simple instruments for monitoring adherence promises to benefit patients, physicians, and health care payers. ■

1. Mancia G, Fagard R, Narkiewicz K, et al. 2013 ESH/ESC guidelines for the management of arterial hypertension: the Task Force for the Management of Arterial Hypertension of the European Society of Hypertension (ESH) and of the European Society of Cardiology (ESC). Eur Heart J. 2013;34(28):2159- 2219.
2. Naderi SH, Bestwick JP, Wald DS. Adherence to drugs that prevent cardiovascular disease: meta-analysis on 376,162 patients. Am J Med. 2012;125(9): 882-887.
3. Mancia G, Messerli F, Bakris G, Zhou Q, Champion A, Pepine CJ. Blood pressure control and improved cardiovascular outcomes in the International Verapamil SR-Trandolapril Study. Hypertension. 2007;50(2):299-305.
4. Christensen A, Osterberg LG, Hansen EH. Electronic monitoring of patient adherence to oral antihypertensive medical treatment: a systematic review. J Hypertens. 2009;27(8):1540-1551.
5. Wetzels GE, Nelemans PJ, Schouten JS, et al. Electronic monitoring of adherence as a tool to improve blood pressure control. A randomized controlled trial. Am J Hypertens. 2007;20(2):119-125.
6. Burnier M, Brede Y, Lowy A. Impact of prolonged antihypertensive duration of action on predicted clinical outcomes in imperfectly adherent patients: comparison of aliskiren, irbesartan and ramipril. Int J Clin Pract. 2011;65(2):127- 133.
7. Mallion JM, Baguet JP, Siche JP, Tremel F, de Gaudemaris R. Compliance, electronic monitoring and antihypertensive drugs. J Hypertens Suppl. 1998; 16(1):S75-S79.
8. Bramley TJ, Gerbino PP, Nightengale BS, Frech-Tamas F. Relationship of blood pressure control to adherence with antihypertensive monotherapy in 13 managed care organizations. J Manag Care Pharm. 2006;12(3):239-245.
9. Corrao G, Parodi A, Nicotra F, et al. Better compliance to antihypertensive medications reduces cardiovascular risk. J Hypertens. 2011;29(3):610-618.
10. Perreault S, Dragomir A, Roy L, et al. Adherence level of antihypertensive agents in coronary artery disease. Br J Clin Pharmacol. 2010;69(1):74-84.
11. Perreault S, Dragomir A, White M, Lalonde L, Blais L, Bérard A. Better adherence to antihypertensive agents and risk reduction of chronic heart failure. J Intern Med. 2009;266(2):207-218.
12. Kettani FZ, Dragomir A, Côté R, et al. Impact of a better adherence to antihypertensive agents on cerebrovascular disease for primary prevention. Stroke. 2009;40(1):213-220.
13. Schlaich MP, Sobotka PA, Krum H, Lambert E, Esler MD. Renal sympathetic- nerve ablation for uncontrolled hypertension. N Engl J Med. 2009;361(9): 932-934.
14. Bhatt DL, Kandzari DE, O’Neill WW, et al; SYMPLICITY HTN-3 Investigators. A controlled trial of renal denervation for resistant hypertension. N Engl J Med. 2014;370(15):1393-1401.
15. Victor RG. Carotid baroreflex activation therapy for resistant hypertension. Nat Rev Cardiol. 2015;12(8):451-463.
16. Calhoun DA, Jones D, Textor S, et al; American Heart Association Professional Education Committee. Resistant hypertension: diagnosis, evaluation, and treatment: a scientific statement from the American Heart Association Professional Education Committee of the Council for High Blood Pressure Research. Circulation. 2008;117(25):e510-e526.
17. Burnier M. Managing ‘resistance’: is adherence a target for treatment? Curr Opin Nephrol Hypertens. 2014;23(5):439-443.
18. de la Sierra A, Segura J, Banegas JR, et al. Clinical features of 8295 patients with resistant hypertension classified on the basis of ambulatory blood pressure monitoring. Hypertension. 2011;57(5):898-902.
19. Persell SD. Prevalence of resistant hypertension in the United States, 2003- 2008. Hypertension. 2011;57(6):1076-1080.
20. Jung O, Gechter JL, Wunder C, et al. Resistant hypertension? Assessment of adherence by toxicological urine analysis. J Hypertens. 2013;31(4):766-774.
21. Tomaszewski M, White C, Patel P, et al. High rates of non-adherence to antihypertensive treatment revealed by high-performance liquid chromatographytandem mass spectrometry (HP LC-MS/MS) urine analysis. Heart. 2014;100 (11):855-861.
22. Persu A, Jin Y, Baelen M, et al; European Network Coordinating research on REnal Denervation Consortium. Eligibility for renal denervation: experience at 11 European expert centers. Hypertension. 2014;63(6):1319-1325.
23. Judd E, Calhoun DA. Apparent and true resistant hypertension: definition, prevalence and outcomes. J Hum Hypertens. 2014;28(8):463-468.
24. Brown MT, Bussell JK. Medication adherence: WHO cares? Mayo Clin Proc. 2011;86(4):304-314.
25. Morisky DE, Ang A, Krousel-Wood M, Ward HJ. Predictive validity of a medication adherence measure in an outpatient setting. J Clin Hypertens (Greenwich). 2008;10(5):348-354.
26. Halpern MT, Khan ZM, Schmier JK, et al. Recommendations for evaluating compliance and persistence with hypertension therapy using retrospective data. Hypertension. 2006;47(6):1039-1048.
27. Hasselström J, Zarrinkoub R, Holmquist C, et al. The Swedish Primary Care Cardiovascular Database (SPCCD): 74 751 hypertensive primary care patients. Blood Press. 2014;23(2):116-125.
28. Qvarnström M, Kahan T, Kieler H, et al. Persistence to antihypertensive drug treatment in Swedish primary healthcare. Eur J Clin Pharmacol. 2013;69(11): 1955-1964.
29. Belknap R, Weis S, Brookens A, et al. Feasibility of an ingestible sensor-based system for monitoring adherence to tuberculosis therapy. PLoS One. 2013; 8(1):e53373.
30. Cramer JA, Scheyer RD, Mattson RH. Compliance declines between clinic visits. Arch Intern Med. 1990;150(7):1509-1510.
31. Strauch B, Petrák O, Zelinka T, et al. Precise assessment of noncompliance with the antihypertensive therapy in patients with resistant hypertension using toxicological serum analysis. J Hypertens. 2013;31(12):2455-2461.
32. Burnier M, Schneider MP, Chioléro A, Stubi CL, Brunner HR. Electronic compliance monitoring in resistant hypertension: the basis for rational therapeutic decisions. J Hypertens. 2001;19(2):335-341.
33. Brinker S, Pandey A, Ayers C, et al. Therapeutic drug monitoring facilitates blood pressure control in resistant hypertension. J Am Coll Cardiol. 2014;63(8): 834-835.
34. Schmieder RE, Ott C, Schmid A, et al. Adherence to antihypertensive medication in treatment-resistant hypertension undergoing renal denervation. J Am Heart Assoc. 2016;5(2). pii: e002343. doi: 10.1161/JAHA.115.002343.
35. Fadl Elmula FE, Hoffmann P, Larstorp AC, et al. Adjusted drug treatment is superior to renal sympathetic denervation in patients with true treatment-resistant hypertension. Hypertension. 2014;63(5):991-999.
36. Bunker J, Callister W, Chang CL, Sever PS. How common is true resistant hypertension? J Hum Hypertens. 2011;25(2):137-140.
37. Eskås PA, Heimark S, Eek Mariampillai J, Larstorp AC, Fadl Elmula FE, Høieggen A. Adherence to medication and drug monitoring in apparent treatment- resistant hypertension. Blood Press. 2016;25(4):199-205.
38. Burnier M, Wuerzner G, Struijker-Boudier H, Urquhart J. Measuring, analyzing, and managing drug adherence in resistant hypertension. Hypertension. 2013;62(2):218-225.
39. Albini F, Xiaoqiu L, Torlasco C, et al. An ICT and mobile health integrated approach to optimize patients’ education on hypertension and its management by physicians: The Patients Optimal Strategy of Treatment (POST) pilot study. Conf Proc IEEE Eng Med Biol Soc. 2016;2016:517-520.
40. Muntner P, Judd SE, Krousel-Wood M, McClellan WM, Safford MM. Low medication adherence and hypertension control among adults with CKD: data from the REGARDS (Reasons for Geographic and Racial Differences in Stroke) study. Am J Kidney Dis. 2010;56(3):447-457.

Keywords: adherence; adherence app; adherence monitoring; directly observed treatment; drug monitoring; electronic pill box; hypertension; ingestible sensor; Morisky questionnaire; pill counting; resistant hypertension