Persistence and adherence with antihypertensive drug therapy in a German sickness fund population

The objective of this retrospective cohort study was to examine persistence and adherence (compliance) with newly initiated antihypertensive therapy in the real-life setting of a German sickness fund. Persistence was highest after an index prescription for angiotensin receptor blockers (ARBs) (52.7%), followed by angiotensin-converting enzyme (ACE) inhibitors (34.5%), calcium channel blockers (CCBs; 34.4%), diuretics (26.0%) and β -blockers (13.6%).

Poor persistence to and adherence with drug therapy represents the most important cause of uncontrolled blood pressure,4 which is a well-recognized risk factor for coronary heart disease, heart failure, stroke and renal failure.1 With effective blood pressure-lowering therapy, the risk of such fatal and non-fatal events can be decreased.2, 3 The analysis of pharmacy claims data has been used to assess patients’ medication-taking behaviour for many years.5, 6, 7, 8, 9, 10, 11 ARBs have been studied in greater detail in the United States,8, 10 but have not been investigated across adequate timeframes and in sufficient numbers of patients within Europe and Germany in particular. Therefore, a retrospective cohort study analysing claims data for a large representative statutory German sickness fund (Gesetzliche Krankenkasse) was completed to investigate differences in adherence and persistence with the following classes of antihypertensive drugs: ARBs, ACE inhibitors, CCBs, β-blockers and diuretics.

Sickness fund members over 18 years of age and with at least one prescription for an antihypertensive drug between 1 January 2000 and 31 December 2003 were included, if they fulfilled the criteria of a defined index prescription (see below). Exclusion criteria were:

  1. 1

    hospitalization associated with a diagnosis of coronary heart disease, myocardial infarction, or heart failure before the index date (timeframe surveyed at least 180 days),

  2. 2

    an index prescription for a fixed or free combination of antihypertensive drugs, because in general, they are not chosen to start an antihypertensive therapy,

  3. 3

    and only one prescription for the index drug class following the index prescription, because they were assumed to be not adherent and the inclusion of such beneficiaries might have diluted the results.

Antihypertensive drugs that are seldom used today (for example minoxidil, clonidine, dihydralazine and α-blockers) were not investigated in the study.

An index prescription indicating a newly initiated antihypertensive therapy was defined as (i) no prescription for an antihypertensive agent 6 months before, and (ii) claims data available for at least 12 months after the index prescription. Treatment discontinuation was assumed, if there was no refill within 60 days after the end of the dispensed supply of a prescription (calculated from the number of defined daily doses (DDD)).12

Persistence rate with the index drug class (proportion of patients without therapy discontinuation at 12 months) and adherence (medication possession ratio (MPR): number of dispensed DDD per 12 months divided by 365) were assessed. Moreover, the time of continuous prescription for any antihypertensive drug following the index prescription was determined in patients with prescriptions for additional antihypertensive drug classes, and patients with prescription of the index drug class only.

The distribution of study participants across the index drug classes was tested for homogeneity with χ2-tests.

The influence of index drug class, age, gender and diabetic comorbidity (indicated by a diagnosis recorded with a hospitalization or sick leave and/or at least two antidiabetic drug prescriptions) on duration of continuous therapy with the index drug class was analysed using a Cox proportional hazard model. The differences of the mean values adjusted for age and sex for the MPR at 360 days, and the duration of continuous prescription of any antihypertensive drug following the index prescription were determined by analysis of variance (ANOVA), comparing patients with an index prescription of ARBs against all other drug classes. All statistical analyses were performed using the software package SPSS, version 12.0. The α-level of significance was set at 0.05.

A total of 62 754 patients met our inclusion criteria (see Supplementary Figure w1). Supplementary Table w1 summarizes the baseline characteristics of the study population. The mean age of the cohort was 54.1±13.3 years, 58.7% were men and 10.5% had diabetes. Most index prescriptions were for β-blockers (50.2%), followed by ACE inhibitors (24.4%), diuretics (10.6%), CCBs (9.4%) and ARBs (5.5%).

The Cox proportional hazard model (Table 1) showed that compared with the duration of continuous prescription of an ARB, the relative risk for discontinuation of prescription was significantly higher for all other drug classes. The relative risk for discontinuation of index drug class prescription was significantly lower in men, patients with diabetes and those older than 40 years.

Table 1 Results of the Cox proportional hazard model analysing the influence of index drug class, age, gender and diabetic comorbidity on the relative risk for an interruption of therapy with the index drug class

The highest persistence rate at 12 months was found after an index prescription for ARBs (52.7%), followed by ACE inhibitors (34.5%), CCBs (34.4%), diuretics (26.0%) and β-blockers (13.6%). Adherence rates (MPR) showed good agreement with the persistence findings (Supplementary Table w2). They were significantly higher for ARBs (0.70), followed by ACE inhibitors (0.56), CCBs (0.54), diuretics (0.53) and β-blockers (0.39). Results on patients not only receiving a monotherapy is given in the Supplementary Tables w3 and w4.

A major strength of this study is that the sickness fund database contains records of all prescriptions that were dispensed in a pharmacy, thereby providing a complete and detailed recording of all prescriptions for an individual member. An important limitation of our study is that the database does not contain blood pressure measurements and other clinical information, so we were not able to determine the impact of differences in medication persistence and adherence on blood pressure control and cardiovascular outcomes. We also have to consider that, while each recorded prescription can be considered as the intention of the patient to take the dispensed drug, it could not be ascertained that the drugs were actually taken. Another limitation of the study is that there was no information on the actual amount of days supplied with medication. We therefore used DDDs as an estimation. Owing to differences between dosage forms and recommended dosages, we probably have underestimated persistence and adherence for the β-blockers, and made an overestimation for the ARBs.

Moreover, it cannot be ruled out that differences in medication-taking behaviour observed in our study were at least partly due to some kind of selection bias. Overall, the findings of the present study are in general agreement with other studies that have analysed persistence and adherence with antihypertensive drugs from claims data.6, 7, 8, 9, 10, 11

In conclusion, our study provides estimations of patients’ persistence to and adherence with newly initiated antihypertensive therapy in the real-life ambulatory care setting of a German sickness fund. Given that good persistence and adherence with antihypertensive therapy is necessary for blood pressure control and, in turn, cardiovascular risk reduction, these findings should be considered in medical decisions relating to the treatment of patients with hypertension. Probably, the differences in persistence to and adherence with antihypertensive drug therapy may not primarily depend on the prescribed drug class. There may be more important factors, which are affected by the patients’ characteristics, and which should be analysed in further studies.

Conflicts of interest

A Höer, H Gothe and G Schiffhorst are employees (and B Häussler is the Director) of IGES – Institute for Healthcare and Social Research, an independent research organization that received funding from Novartis for this study. Z Khan and G Vincze are employees of Novartis, which manufactures products for the treatment of hypertension (angiotensin receptor blockers).


  1. 1

    Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, Izzo Jr JL et al. The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: the JNC 7 report. JAMA 2003; 289: 2560–2572.

  2. 2

    Psaty BM, Smith NL, Siscovick DS, Koepsell TD, Weiss NS, Heckbert SR et al. Health outcomes associated with antihypertensive therapies used as first-line agents. A systematic review and meta-analysis. JAMA 1997; 277: 739–745.

  3. 3

    Turnbull F, Blood Pressure Lowering Treatment Trialists’ Collaboration. Effects of different blood-pressure-lowering regimens on major cardiovascular events: results of prospectively-designed overviews of randomised trials. Lancet 2003; 362: 1527–1535.

  4. 4

    World Health Organization. Adherence to Long-term Therapies. Evidence for Action. World Health Organization: Geneva, 2003.

  5. 5

    Steiner JF, Koepsell TD, Fihn SD, Inui TS . A general method of compliance assessment using centralized pharmacy records: description and validation. Med Care 1988; 26: 814–823.

  6. 6

    Conlin PR, Gerth WC, Fox J, Roehm JB, Boccuzi SJ . Four year persistence patterns among patients initiating therapy with the angiotensin II receptor antagonist losartan versus other antihypertensive drug classes. Clin Ther 2006; 23: 1999–2010.

  7. 7

    Degli Esposti E, Sturani A, Di Martino M, Falasca P, Novi MV, Baio G et al. Long-term persistence with antihypertensive drugs in new patients. J Hum Hypertens 2002; 16: 439–444.

  8. 8

    Wannemacher AJ, Schepers GP, Townsend KA . Antihypertensive medication compliance in a Veterans’ Affairs healthcare system. Ann Pharmacother 2002; 36: 986–991.

  9. 9

    Hasford J, Mimran A, Simons WR . A population-based European cohort study of persistence in newly diagnosed hypertensive patients. J Hum Hypertens 2002; 16: 569–575.

  10. 10

    Wogen J, Kreilick CA, Livornese RC, Yokoyama K, Frech F . Patient adherence with amlodipine, lisinopril, or valsartan therapy in a usual-care setting. J Manag Care Pharm 2003; 9: 424–429.

  11. 11

    Erkens JA, Panneman MM, Klungel OH, van den Boom G, Prescott MF, Herings RM . Differences in antihypertensive drug persistence associated with drug class and gender: a PHARMO study. Pharmacoepidemiol Drug Saf 2005; 14: 795–803.

  12. 12

    WHO Collaborating Centre for Drug Statistics Methodology.

Download references

Author information

Correspondence to A Höer.

Additional information

Supplementary Information accompanies the paper on the Journal of Human Hypertension website (

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Höer, A., Gothe, H., Khan, Z. et al. Persistence and adherence with antihypertensive drug therapy in a German sickness fund population. J Hum Hypertens 21, 744–746 (2007).

Download citation

Further reading