This drug-utilization study in a prescription database of more than 50,000 patients analyzed compliance, persistence, and switching behavior for ACE inhibitors and ARBs.
To investigate compliance, persistence, and switching patterns for angiotensin-converting enzyme (ACE) inhibitors and angiotensin receptor blockers (ARBs).
Drug-utilization analysis using a large prescription database.
Prescription data for more than 50,000 incident users of ACE inhibitors or ARBs were collected, cumulating close to 200,000 patient-years of medication use. Incidence, drug dosage, 1-year compliance, long-term persistence, and switching patterns were analyzed. The specific drugs investigated were captopril, enalapril, lisinopril, perindopril, ramipril, and fosinopril (ACE inhibitors), and losartan, valsartan, irbesartan, candesartan, and olmesartan (ARBs). Results were adjusted for age, sex, starting date, and comorbidities.
The 1-year compliance (88.3% vs 88.3%, P = .996) and 3-year persistence (81.9% vs 82.4%, P = .197) rates were similar between ACE inhibitors and ARBs. Users of ACE inhibitors more often switched therapy (24.2% vs 13.1%, P <.001), primarily to an ARB. Variations in compliance, persistence, and switching behavior were detected between specific ACE inhibitors, but not between specific ARBs.
Although residual confounding and indication bias cannot be ruled out, this study showed that compliance, persistence, and switching behavior varied between specific ACE inhibitors but not between specific ARBs. These results support prescribing of cheap generic ARBs as opposed to expensive ARBs. Apart from factors leading to therapy switches, compliance and persistence were similar between ACE inhibitors and ARBs.
(Am J Manag Care. 2011;17(9):609-616)
This drug-utilization study in a prescription database of more than 50,000 patients analyzed compliance, persistence, and switching behavior with angiotensin-converting enzyme (ACE) inhibitors and angiotensin receptor blockers (ARBs). The 2 drug classes were compared, as well as specific drugs within the drug classes.
Antihypertensives are a cornerstone in the prevention and treatment of cardiovascular and renal diseases.1 Agents that inhibit the renin-angiotensin system (RAS), which include angiotensin-converting enzyme (ACE) inhibitors and angiotensin receptor blockers (ARBs), are especially important. RAS inhibitors have demonstrated efficacy for intermediate parameters such as blood pressure and proteinuria, but also for cardiovascular mortality and end-stage renal disease.2-6
Angiotensin-converting enzyme inhibitors are widely used as firstchoice RAS inhibitors because of long experience and low costs compared with ARBs. These advantages are decreasing, however, because of present and upcoming patent expirations of ARBs. Furthermore, ARBs are associated with placebo-like tolerability,7,8 which may improve therapy compliance and persistence. On a group level, ARBs are sometimes proposed to be superior to ACE inhibitors.9 However, comparative studies often fail to demonstrate clinically relevant differences between ACE inhibitors and ARBs, and guidelines commonly suggest they are equivalent for nearly all indications.10
Complicating these matters is the debate surrounding the comparative effectiveness of specific ACE inhibitors and ARBs.11,12 For the specific drugs there is no conclusive evidence on differences in drug efficacy and tolerability. A recent meta-analysis of 32 placebo-controlled trials suggested that all ACE inhibitors have similar efficacy for reducing mortality in congestive heart failure.13 Results from observational studies, on the other hand, present conflicting evidence regarding the existence of a class effect.14,15 Similarly, recent reviews were unable to draw conclusions on the comparative efficacy of specific ARBs.12,16 Real-life drug-utilization patterns can supplement evidence from clinical trials.11,17 First, drug compliance and persistence are recognized markers of drug efficacy and tolerability.18 Second, therapy switches are signs of unsatisfactory treatment response and unacceptable adverse effects.19,20
The objective of our study was to investigate drug compliance, persistence, and switching patterns of RAS-inhibiting agents in newly treated patients.
Prescription data between 1999 and 2010 were retrieved from the IADB.nl database, which holds a representative sample of the Dutch population of more than 500,000 individuals. Each prescription record contains basic patient characteristics and information on drug, dosage, prescriber (general practitioner or specialist hospital doctors), and dispensing date. The IADB.nl prescription database has been validated for drug-utilization studies21,22 and has previously been used for such studies.23,24 Due to high patient-pharmacy commitment in the Netherlands,25 complete medication histories of individuals could be retrieved or constructed through linking pharmacy registries. Drugs were systematically classified using the Anatomical Therapeutic Chemical ATC) Classification System of the World Health Organization. 26 In the Netherlands, healthcare insurance is semiprivatized. Risks for insurance companies are regulated by a national equalization pool. Risks for the public are minimized by the obliged purchase of coverage and by government-mandated acceptance for basic insurance plans. The medications included in these analyses are all fully reimbursed without restriction.
Patient Population and Drugs
Incident users of RAS inhibitors (ATC C09) older than 18 years were included. The following drugs were investigated: captopril, enalapril, lisinopril, perindopril, ramipril, and fosinopril (ACE inhibitors), and losartan, valsartan, irbesartan, candesartan, and olmesartan (ARBs). Fixed-dose combinations with diuretics were also included. Combined, these drugs constituted 96% of all RAS inhibitors in the database. Comorbidities were recorded by proxy of comedication, prescribed before or at maximum half a year after initiating RAS inhibiting therapy. Diabetes mellitus (DM) therapy was identified by prescription of glucose-lowering drugs (ATC A10).27 Dyslipidemia therapy was identified by prescription of lipid-lowering drugs (ATC C10).27 Ischemic heart disease (IHD) therapy was identified by prescription of either nitrates (ATC C01DA) or platelet aggregation inhibitors (ATC B01AC).28 Heart failure (HF) therapy was identified by prescription of either digoxin (ATC C01AA05) or loop diuretics (ATC C03C).29 Chronic obstructive pulmonary disease (COPD) therapy was identified by incident use of adrenergic inhalants (ATC R03A) or anticholinergic inhalants (ATC R03BB) in patients 55 years or older.30 Incident use of COPD therapy was defined as the first use of an inhaler while being known in the database for at least 1 year. Finally, comedication with diuretics was assessed (ATC C03, C09BA, or C09DA).
Drug-utilization patterns were investigated: incidence, dosage, 1-year compliance, long-term persistence, and switching behavior. The drug that was most commonly prescribed within its class was used as the reference drug (enalapril for the ACE inhibitors and losartan for the ARBs).
Incidence and Dosage. Incidence was defined as the first drug used after being present in the database for at least 1 year.23,24 Because up-titration is common and necessary to achieve optimal blood pressure control,31 the dosage was measured 6 months after drug initiation. The dosage was expressed in defined daily doses (DDDs); 1 DDD is the mean dose per day for a drug used for its main indication in adults.32
Compliance. Drug compliance (ie, adherence) is defined as “the extent to which a patient acts in accordance with the prescribed interval and dose of a dosing regimen.”33 A common method is the proportion of days covered, calculated as the number of days the patient had access to the drug divided by the number of days in a specified time period.34 This time period was 1 year, starting at therapy initiation. Based on empiricalstudies to predict hospitalizations for hypertension and congestive HF,35 a threshold of 80% was used to dichotomize between compliant and noncompliant patients. Patients who discontinued therapy or switched to a different drug were excluded, as this behavior was assessed in separate analyses. Differences in compliance were analyzed compared with the reference drug, adjusting for age, sex, year of initiating therapy, and comorbidities.
Persistence. Whereas drug compliance refers to treatment intensity, drug persistence focuses on treatment duration. Drug persistence is defined as “the duration of time from initiation to discontinuation of therapy.”33 Persistence was measured using the refill-sequence method. The time between the first prescription and the point at which an unacceptable prescription gap occurs was measured.36 The length of this unacceptable gap or “grace period” was 90 days.36,37 In case of overlapping prescriptions, the second prescription was shifted forward to account for drug stockpiling.38 Patients were censored when lost to follow-up or when switching therapy, as switching was analyzed separately. Differences in persistence were analyzed compared with the reference drug, adjusting for age, sex, year of initiating therapy, and comorbidities.
Switching. A switch was defined as an RAS-inhibiting agent permanently substituting for the initial drug therapy.39 Specific analysis was performed for switches from an ACE inhibitor to an ARB, which can be related to adverse events, in particular angioedema and dry cough.8,24 Differences in switching patterns were analyzed compared with the reference drug, adjusting for age, sex, year of initiating therapy, and comorbidities.
All continuous variables are presented as mean ± standard deviation, unless noted otherwise. Differences in compliance were tested using logistic regression. Differences in persistence and switching patterns were plotted using Kaplan-Meier plots and tested using the log-rank test and Cox proportional hazard analysis. All statistical analyses were performed using R, version 2.5.1 (the GNU Project, www.r-project.org/).
Prescription data for 53,000 incident users of ACE inhibitors and ARBs were collected. A total of 51,181 patients initiated therapy on the predefined drugs. These patients cumulated close to 200,000 patient-years of medication use. Baseline characteristics of these patients are shown in Table 1. The type of medical prescriber was similar for users of ACE inhibitors and ARBs (percentage of general practitioners was 66.5% for users of ACE inhibitors vs 66.9% for users of ARBs, P = .465). ACE inhibitor users were older than ARB users (63.2 ± 14.1 years vs 61.5 ± 13.7 years) and more often male (48.8% vs 42.8%) (both P <.001). Comorbidities were more common in users of ACE inhibitors compared with ARB users, including DM (19.8% vs 14.5%), dyslipidemia (38.3% vs 30.6%), IHD (40.7% vs 30.3%), HF (21.6% vs 14.3%), and COPD (3.8% vs 3.0%); users of ACE inhibitors also were more likely to comedicate with diuretics (57.6% vs 55.5%) (all P <.001). Patient characteristics varied among users of different ACE inhibitors (Table 1), while users of different ARBs were largely similar.
Incidence and Dosage
The most frequent prescribed ACE inhibitor was enalapril (37.2%) and the most frequent prescribed ARB was losartan (34.5%); these drugs were used as reference drugs. The median prescribed dosage corresponded to the DDD (Table 1, ). The 2 exceptions were captopril, which was prescribed below the DDD of 50 mg in 65% of patients, and ramipril, which was prescribed above the DDD of 2.5 mg in 70% of patients.
After excluding 24,805 patients who discontinued or switched treatment, 20,236 ACE inhibitor users and 6140 ARB users were analyzed for 1-year compliance. By design, none of these patients had switched or permanently discontinued RAS therapy. Higher patient age and comedication for dyslipidemia increased the chance of being compliant (9.4% and 25.6% over 10 years, respectively, P <.001), while comedication for COPD and later year of initiating therapy decreased the chance of being compliant (-24.3% per year [P = .005] and -1.5% per year [P = .035], respectively). The compliance of ACE inhibitor and ARB users was 88.3% (P = .996) for both classes. There was variation in compliance between the specific molecules (), both without and with adjustment for age, sex, year of initiating therapy, and comorbidities. Compliance among users of ramipril (90.4%, P = .05) and fosinopril (91.6%, P = .017) was higher compared with compliance among users of enalapril (87.9%). Within the ARB group, users of candesartan were found to be significantly less compliant than users of losartan (86.1% vs 88.8%, P = .027).
Persistence data are shown in Table 2 and and. Higher age (hazard ratio [HR] = 0.91 per 10 years, P <.001), later year of initiating therapy (HR = 0.71 per year, P <.001), comedication for IHD (HR = 0.90, P = .001), and comedication for HF (HR = 0.75, P <.001) increased the chance of being persistent, while comedication for dyslipidemia (HR = 1.24, P <.001), comedication for COPD (HR = 1.26, P = .001), or use of diuretics (HR = 1.15, P <.001) decreased the chance of being persistent. After 3 years of treatment, persistence with ACE inhibitors and ARBs was not significantly different both without and with adjustment for possible confounders (81.9% vs 82.4%, P = .197). Between the different ACE inhibitors, persistence differed significantly (overall P <.001). Enalapril users had the lowest persistence rate after 3 years, namely 80.8%, which was significantly lower than the rate with other ACE inhibitors. Users of ramipril and fosinopril showed the highest persistence: 85.8% and 83.4%, respectively (P <.001 and P = .047 vs enalapril, respectively). In contrast, there were no significant differences in persistence among ARB users (overall P = .073). The use of different grace periods, such as 60 days or 120 days, did not change the relative order of persistence.
Users of ACE inhibitors switched drugs more than ARB users. After 3 years of therapy, 24.2% of ACE inhibitor users had switched therapy, compared with 13.1% of ARB users (P <.001). This difference in switching rates was not dependent on the year of starting therapy or any other possible confounders. Compared with users of enalapril, users of perindopril switched less often, while users of captopril switched significantly more often. Most ACE inhibitor switchers started using an ARB (75.0%). Users of perindopril and captopril switched significantly less often to an ARB compared with users of enalapril. Users of candesartan switched less often to another RAS inhibitor compared with users of losartan.
In the present study we analyzed drug-utilization patterns of RAS inhibitors. Apart from therapy switches, compliance and persistence were similar between ACE inhibitors and ARBs. On the drug level, several differences between the ACE inhibitors were detected. Ramipril and fosinopril users had higher compliance and persistence rates than users of the other ACE inhibitors, possibly indicative of more favorable drug tolerability profiles. Users of ARBs, on the other hand, were similar in terms of compliance, persistence, and switching behavior.
The most frequently prescribed RAS inhibitors were enalapril and losartan. These drugs are among the first marketed members in their classes, underlining the emphasis that is placed on prescribing experience in the Netherlands. The prescribed dosage often corresponded to the DDD. As an exception, ramipril was often prescribed at a higher dose, 5 mg/day, than the DDD of 2.5 mg. Clinical trial data in cardiovascular disease5 and renal disease6 also showed that ramipril is often prescribed at doses above 2.5 mg/day.
A novel finding of our study is that, apart from factors leading to therapy switches, compliance and persistence were similar between ACE inhibitors and ARBs. These results at first glance seem to disagree with results of previous studies, including one study in 15,000 hypertensive patients that reported superior persistence with ARBs.9 However, in our study, patients were censored at the time of switching. No such censoring was used in other studies, and as a consequence, these studies failed to detect the similarity in compliance and persistence between drug classes. Indeed, our results showed that switching was more frequent among users of ACE inhibitors compared with users of ARBs, in agreement with previous studies.9 Reasons for the difference in switching patterns between ACE inhibitors and ARBs at the class
level deserve careful attention. One possible explanation is the well-known existence of ACE inhibitor—specific adverse events such as angioedema and dry cough,8,24 as well as the placebo-like tolerability of ARBs.7,8 A large meta-analysis of randomized controlled trials with ACE inhibitors and ARBs found only minor differences in discontinuation rates due to adverse drug events40; however, real-life observational studies have found discontinuation rates due to ACE inhibitor adverse events to be as high as 19%.41 Another possible explanation is strong marketing of the newer ARBs, although year of therapy initiation was not an influential confounder in the analyses. Prescription sales of antihypertensive drugs have been shown to be correlated with marketing efforts of pharmaceutical companies.42 Regardless of the reasons for switching, long-term persistence can be negatively influenced by switching therapy43; this should be a topic for further research.
There was variation in drug-utilization patterns between the specific ACE inhibitors. The average prescribed dosage of captopril was below the DDD and did not increase over time. Captopril users often switched to a different ACE inhibitor. Together, these findings suggest that patients and physicians prefer to switch drugs rather than increase the pill burden of captopril. This is in accordance with evidence that once-daily antihypertensive dosing regimens are associated with superior compliance.44 Users of ramipril and fosinopril showed high rates of compliance and persistence, which might indicate favorable drug tolerability profiles compared with other ACE inhibitors. These results are in accordance with a previously published study analyzing compliance and persistence in more than 6000 ACE inhibitor users, which also found the highest compliance and persistence for ramipril and the lowest for enalapril.45 In contrast to ACE inhibitors, the specific ARBs had very similar patterns of drug utilization. Candesartan users were less compliant and switched less often compared with users of other ARBs. The difference in compliance was small, however (86.1% vs 88.8%), and previous studies found no differences in adverse event rates between ARBs across the approved dosage ranges.16 Therefore, a confounding effect of indication bias or residual confounding cannot be ruled out. Our results support a recent cost-effectiveness analysis that recommended generic cheaper ARBs over more expensive branded ARBs, as the differences in efficacy are small.46 Our study showed that differences in compliance, persistence, and switching behavior between ARBs are also small, thereby providing even less reason to prescribe expensive ARBs.
Our study has several limitations. First, our analysis used prescription data, which did not necessarily reflect actual drug use. Validation studies, however, showed good correlation between prescription claims and actual drug use.47 Second, the indication for prescribing was not registered in our database. Although we adjusted the results for several comorbidities by proxy of comedication, the possibility of residual confounding, influence of treatment history (such as chronic kidney disease), or indication bias remains. In addition, some comorbidities are associated with underprescribing, such as cholesterol-lowering therapy.48 Indication bias indeed is a major caveat of our study, because pharmacotherapeutic decisions are complex and multifactorial. Although the differences between ACE inhibitors found in our study are supported by the literature and are indicative of differences in drug tolerability profiles, there is no proven causality. For the same reason, frequency of medication administration (eg, once daily, twice daily) could not be analyzed because of indication bias. Temporal confounding, for example through publication of new trial evidence, might have influenced drug-utilization patterns. These effects have been described previously (eg, for nonantihypertensive medications after discovery of serious side effects49). We adjusted for year of therapy initiation in our study; this did not influence the results. Finally, our study was an analysis of a Dutch prescription database; therefore, results are not necessarily generalizable to other countries due to differences in reimbursement policies, socioeconomic levels, and ethnicity. Still, the real-life drugutilization patterns of our study should provide valuable data to supplement evidence from clinical trials.11
In conclusion, although residual confounding and indication bias cannot be ruled out, this study showed that compliance, persistence, and switching behavior varied among users of different ACE inhibitors, but not among users of different ARBs. In terms of drug-utilization characteristics, there appears to be no reason for prescribing more expensive branded ARBs rather than cheaper generic ARBs. Apart from factors leading to therapy switches, compliance and persistence were similar between users of ACE inhibitors and ARBs.
Author Affiliations: From Department of Pharmacy (SV, NHN, STV, LJ, MJP, CB), University of Groningen, Groningen, the Netherlands.
Funding Source: No funding was received for this study. Researchers have previously received independent research grants from sanofi-aventis (manufacturer of ramipril, irbesartan, and losartan) and Daiichi-Sankyo (manufacturer of captopril and olmesartan).
Author Disclosures: The authors (SV, NHN, STV, LJ, MJP, CB) report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.
Authorship Information: Concept and design (SV, NHN, STV, LJ, MJP, CB); acquisition of data (SV, NHN, STV); analysis and interpretation of data (SV, NHN, STV, CB); drafting of the manuscript (SV, MJP, CB); critical revision of the manuscript for important intellectual content (SV, STV, LJ, MJP, CB); statistical analysis (SV, NHN, STV); administrative, technical, or logistic support (STV); and supervision (LJ, MJP, CB).
Address correspondence to: Stefan Vegter, PharmD, Unit of Pharmaco-Epidemiology & PharmacoEconomics, Department of Pharmacy, University of Groningen, Antonius Deusinglaan 1, 9713 AV Groningen, The Netherlands. E-mail: firstname.lastname@example.org.
1. Mancia G, Laurent S, Agabiti-Rosei E, et al. Reappraisal of European guidelines on hypertension management: a European Society of Hypertension Task Force document. J Hypertens. 2009;27(11):2121-2158.
2. The SOLVD Investigators. Effect of enalapril on survival in patients with reduced left ventricular ejection fractions and congestive heart failure. N Engl J Med. 1991;325(5):293-302.
3. Brenner BM, Cooper ME, de Zeeuw D, et al; RENAAL Study Investigators. Effects of losartan on renal and cardiovascular outcomes in patients with type 2 diabetes and nephropathy. N Engl J Med. 2001; 345(12):861-869.
4. Dahlöf B, Devereux RB, Kjeldsen SE, et al; LIFE Study Group. Cardiovascular morbidity and mortality in the Losartan Intervention For Endpoint reduction in hypertension study (LIFE): a randomised trial against atenolol. Lancet. 2002;359(9311):995-1003.
5. Yusuf S, Sleight P, Pogue J, Bosch J, Davies R, Dagenais G. Effects of an angiotensin-converting-enzyme inhibitor, ramipril, on cardiovascular events in high-risk patients. The Heart Outcomes Prevention Evaluation Study Investigators [published correction appears in N Engl J Med. 2000;342(10):748]. N Engl J Med. 2000;342(3):145-153.
6. Ruggenenti P, Perna A, Loriga G, et al; REIN-2 Study Group. Bloodpressure control for renoprotection in patients with non-diabetic chronic renal disease (REIN-2): multicentre, randomised controlled trial. Lancet. 2005;365(9463):939-946.
7. Bohm M, Baumhäkel M, Mahfoud F, Werner C. From evidence to rationale: cardiovascular protection by angiotensin II receptor blockers compared with angiotensin-converting enzyme inhibitors. Cardiology. 2010;117(3):163-173.
8. Cicardi M, Zingale LC, Bergamaschini L, Agostoni A. Angioedema associated with angiotensin-converting enzyme inhibitor use: outcome after switching to a different treatment. Arch Intern Med. 2004;164(8):910-913.
9. Conlin PR, Gerth WC, Fox J, Roehm JB, Boccuzzi SJ. Four-year persistence patterns among patients initiating therapy with the angiotensin II receptor antagonist losartan versus other antihypertensive drug classes. Clin Ther. 2001;23(12):1999-2010.
10. Miller AE, Cziracky M, Spinler SA. ACE inhibitors versus ARBs: comparison of practice guidelines and treatment selection considerations. Formulary. 2006;41:274-284.
11. Hernandez AF, Harrington RA. Comparative effectiveness of angiotensin-converting-enzyme inhibitors: is an ACE always an ace? CMAJ. 2008;178(10):1316-1319.
12. Siragy HM. Comparing angiotensin II receptor blockers on benefits beyond blood pressure. Adv Ther. 2010;27(5):257-284.
13. Garg R, Yusuf S. Overview of randomized trials of angiotensin-converting enzyme inhibitors on mortality and morbidity in patients with heart failure: Collaborative Group on ACE Inhibitor Trials [published correction appears in JAMA. 1995;274(6):462]. JAMA. 1995;273(18): 1450-1456.
14. Pilote L, Abrahamowicz M, Eisenberg M, Humphries K, Behlouli H, Tu JV. Effect of different angiotensin-converting-enzyme inhibitors on mortality among elderly patients with congestive heart failure. CMAJ. 2008;178(10):1303-1311.
15. Hansen ML, Gislason GH, Kober L, et al. Different angiotensin-converting enzyme inhibitors have similar clinical efficacy after myocardial infarction. Br J Clin Pharmacol. 2008;65(2):217-223.
16. Smith DH. Comparison of angiotensin II type 1 receptor antagonists in the treatment of essential hypertension. Drugs. 2008;68(9): 1207-1225.
17. Hasford J, Mimran A, Simons WR. A population-based European cohort study of persistence in newly diagnosed hypertensive patients. J Hum Hypertens. 2002;16(8):569-575.
18. Dusing R. Adverse events, compliance, and changes in therapy. Curr Hypertens Rep. 2001;3(6):488-492.
19. Ambrosioni E, Leonetti G, Pessina AC, Rappelli A, Trimarco B, Zanchetti A. Patterns of hypertension management in Italy: results of a pharmacoepidemiological survey on antihypertensive therapy. Scientific Committee of the Italian Pharmacoepidemiological Survey on Antihypertensive Therapy. J Hypertens. 2000;18(11):1691-1699.
20. Chen K, Chiou CF, Plauschinat CA, Frech F, Harper A, Dubois R. Patient satisfaction with antihypertensive therapy. J Hum Hypertens. 2005;19(10):793-799.
21. Schirm E, Monster TB, de Vries R, van den Berg PB, de Jong-vanden Berg LT, Tobi H. How to estimate the population that is covered by community pharmacies? an evaluation of two methods using drug utilisation information. Pharmacoepidemiol Drug Saf. 2004;13(3):173-179.
22. Tobi H, van den Berg PB, de Jong-van den Berg LT. The interaction database: synergy of science and practice in pharmacy. In: Brause RW, Hanisch E, eds. Medical Data Analysis. Berlin, Germany: Springer-Verlag; 2000:206-211.
23. Vegter S, de Jong-van den Berg LT. Choice of first antihypertensive—comparison between the Irish and Dutch setting. Br J Clin Pharmacol.2008;66(2):313-315.
24. Vegter S, de Jong-van den Berg LT. Misdiagnosis and mistreatment of a common side-effect—angiotensin-converting enzyme inhibitorinduced cough. Br J Clin Pharmacol. 2010;69(2):200-203.
25. Leufkens HGM, Urquhart J. Automated pharmacy record linkage in the Netherlands. In: Strom BL, ed. Pharmacoepidemiology. Chichester, UK: John Wiley & Sons Ltd; 2008:347-360.
26. WHO Expert Committee. The selection and use of essential medicines. World Health Organ Tech Rep Ser. 2009;958:1-242.
27. Tu K, Manuel D, Lam K, Kavanagh D, Mitiku TF, Guo H. Diabetics can be identified in an electronic medical record using laboratory tests and prescriptions. J Clin Epidemiol. 2011;64(4):431-435.
28. Gray J, Majeed A, Kerry S, Rowlands G. Identifying patients with ischaemic heart disease in general practice: cross sectional study of paper and computerised medical records. BMJ. 2000;321(7260):548-550.
29. Udris EM, Au DH, McDonell MB, et al. Comparing methods to identify general internal medicine clinic patients with chronic heart failure. Am Heart J. 2001;142(6):1003-1009.
30. Penning-van Beest F, van Herk-Sukel M, Gale R, Lammers JW, Herings R. Three-year dispensing patterns with long-acting inhaled drugs in COPD: a database analysis. Respir Med. 2011;105(2):259-365.
31. Segura J, Christiansen H, Campo C, Ruilope LM. How to titrate ACE inhibitors and angiotensin receptor blockers in renal patients: according to blood pressure or proteinuria? Curr Hypertens Rep. 2003;5(5):426-429.
32. Bergman U. The history of the Drug Utilization Research Group in Europe. Pharmacoepidemiol Drug Saf. 2006;15(2):95-98.
33. Cramer JA, Roy A, Burrell A, et al. Medication compliance and persistence: terminology and definitions. Value Health. 2008;11(1):44-47.
34. Peterson AM, Nau DP, Cramer JA, Benner J, Gwadry-Sridhar F, Nichol M. A checklist for medication compliance and persistence studies using retrospective databases. Value Health. 2007;10(1):3-12.
35. Karve S, Cleves MA, Helm M, Hudson TJ, West DS, Martin BC. Good and poor adherence: optimal cut-point for adherence measures using administrative claims data. Curr Med Res Opin. 2009;25(9):2303-2310.
36. Caetano PA, Lam JM, Morgan SG. Toward a standard definition and measurement of persistence with drug therapy: examples from research on statin and antihypertensive utilization. Clin Ther. 2006;28(9):1411-1424; discussion 1410.
37. Sharma PP. Predictors and Consequences of Nonadherence to Antihypertensive Medication [dissertation]. Rutgers, The State University of New Jersey; 2007.
38. Vink NM, Klungel OH, Stolk RP, Denig P. Comparison of various measures for assessing medication refill adherence using prescription data. Pharmacoepidemiol Drug Saf. 2009;18(2):159-165.
39. Corrao G, Zambon A, Parodi A, et al. Discontinuation of and changes in drug therapy for hypertension among newly-treated patients: a population-based study in Italy. J Hypertens. 2008;26(4):819-824.
40. Law MR, Wald NJ, Morris JK, Jordan RE. Value of low dose combination treatment with blood pressure lowering drugs: analysis of 354 randomised trials. BMJ. 2003;326(7404):1427.
41. Morimoto T, Gandhi TK, Fiskio JM, et al. An evaluation of risk factors for adverse drug events associated with angiotensin-converting enzyme inhibitors. J Eval Clin Pract. 2004;10(4):499-509.
42. Vitry A, Lai YH. Advertising of antihypertensive medicines and prescription sales in Australia. Intern Med J. 2009;39(11):728-732.
43. Caro JJ, Speckman JL, Salas M, Raggio G, Jackson JD. Effect of initial drug choice on persistence with antihypertensive therapy: the importance of actual practice data. CMAJ. 1999;160(1):41-46.
44. Frishman WH. Importance of medication adherence in cardiovascular disease and the value of once-daily treatment regimens. Cardiol Rev. 2007;15(5):257-263.
45. Gogovor A, Dragomir A, Savoie M, Perreault S. Comparison of persistence rates with angiotensin-converting enzyme inhibitors used in secondary and primary prevention of cardiovascular disease. Value Health. 2007;10(5):431-441.
46. Grosso AM, Bodalia PN, Macallister RJ, Hingorani AD, Moon JC, Scott MA. Comparative clinical- and cost-effectiveness of candesartan and losartan in the management of hypertension and heart failure: a systematic review, meta- and cost-utility analysis. Int J Clin Pract. 2011;65(3):253-263.
47. Enlund H. Measuring patient compliance in antihypertensive therapy—some methodological aspects. J Clin Hosp Pharm. 1982;7(1):43-51.
48. Gumbs PD, Verschuren WM, Mantel-Teeuwisse AK, et al. Drug costs associated with non-adherence to cholesterol management guidelines for primary prevention of cardiovascular disease in an elderly population: the Rotterdam study. Drugs Aging. 2006;23(9):733-741.
49. Vegter S, Kölling P, Töben M, Visser ST, de Jong-van den Berg LT. Replacing hormone therapy—is the decline in prescribing sustained, and are nonhormonal drugs substituted? Menopause. 2009;16(2):329-335.