Beliefs about medicines in Dutch acenocoumarol and phenprocoumon users

Abstract

Aims

Adherence to the generally complex regimen of coumarin derivatives is vital in order to keep patients in the adequate International Normalized Ratio range. Patients' beliefs about medicines are associated with the level of therapy adherence. Our first aim was to assess beliefs about coumarins. Secondly, we compared the beliefs about coumarins with the beliefs about other cardiovascular drugs.

Methods

The Beliefs about Medicines Questionnaire was used to assess medication beliefs. The questionnaire was completed by new users of coumarins indicated for venous thromboembolism or atrial fibrillation. A necessity score and a concerns score were calculated for all patients. The analyses were repeated for users of antihypertensive drugs or statins (not using coumarins).

Results

Three hundred and twenty patients were included in the analysis of the beliefs about coumarins. The mean necessity score was 15.3, the concerns score 12.3 and the necessity–concerns differential 3.0. Patients with venous thromboembolism (n = 71) had higher necessity scores than patients with atrial fibrillation (n = 249; 16.8 vs. 14.9, P < 0.001). The mean necessity score in 493 users of other cardiovascular drugs was 16.1, the concerns score 13.5 and the necessity–concerns differential 2.6. The necessity score was higher in chronic cardiovascular drug users (n = 192) than in new users (n = 301; 17.9 vs. 14.9, P < 0.001).

Conclusions

Coumarin users score higher on the necessity scale than on the concerns scale, which is also the case in users of other cardiovascular drugs. Patients with atrial fibrillation have a less positive attitude towards these drugs than patients with venous thromboembolism, and could therefore benefit more from specific attention.

WHAT IS ALREADY KNOWN ABOUT THIS SUBJECT

• Several studies have shown that higher concerns about medication as well as lower necessity beliefs are associated with higher non-adherence.

• The Beliefs about Medicines Questionnaire can be used to assess the beliefs that patients may have about their medication and has not been used in patients using coumarin derivatives.

WHAT THIS STUDY ADDS

• Patients who recently began using coumarin derivatives scored higher on the necessity scale than on the concerns scale, which indicates a positive attitude towards these drugs.

• Patients with atrial fibrillation had a less positive attitude than patients with venous thromboembolism and might benefit from extra attention to improve their adherence.

Introduction

Patient beliefs about medicines are an important factor in the adherence to the therapy. Patients can have concerns about, for example, the side-effects of the drug, or they can believe that the drug is not really necessary for their health. Several studies have shown that higher concerns about medication as well as lower necessity beliefs are associated with higher non-adherence [1–5]. Non-adherence is a significant challenge to clinical practice and, for some patients, extra education might be useful to increase adherence. Knowledge about patient beliefs might help to identify patients who would benefit from additional counselling. The Beliefs about Medicines Questionnaire (BMQ) was developed to simplify the wide range of beliefs that patients may have about their medication [6].

The BMQ has not yet been applied in studies focusing on anticoagulant therapy with coumarin derivatives. The coumarin derivatives acenocoumarol, phenprocoumon and warfarin are frequently used for the treatment and prevention of thromboembolic events in patients with, for example, atrial fibrillation (AF) or venous thromboembolism (VTE) [7]. These drugs have a narrow therapeutic range and show large inter- and intra-individual variation in dose requirements. Frequent monitoring of the level of anticoagulation and adjustment of the dose is required. This makes the treatment with coumarins more burdensome than most other cardiovascular drugs. However, it is possible that adherence to the therapy is higher for these drugs than for other cardiovascular drugs, because of the frequent checks. Adherence was shown to contribute significantly to warfarin drug response in a recent study [8]. Another recent study in Australia demonstrated that many patients with atrial fibrillation discontinue their warfarin treatment [9]. Analysing the beliefs of patients using coumarin derivatives can provide insight into their attitude towards their treatment. The primary aim of this study was, therefore, to assess the beliefs about acenocoumarol and phenprocoumon in patients initiating therapy with these drugs for AF or VTE. A secondary aim was to compare the beliefs about coumarin derivatives with the beliefs about other cardiovascular drugs (antihypertensive drugs or statins).

Methods

Participants

The European pharmacogenetics of oral anticoagulation (EU-PACT) trial aims to assess the effectiveness of pharmacogenetic-guided dosing of coumarin derivatives and includes patients starting acenocoumarol, phenprocoumon or warfarin therapy for either AF or VTE [10]. This trial was conducted in The Netherlands and in Greece, but for the present study only Dutch patients were included. Patients who did not use any coumarin derivative before inclusion in the trial were recruited at four different anticoagulation clinics in The Netherlands. More details about the EU-PACT trial can be found elsewhere [10]. After ∼1 week of therapy with either acenocoumarol or phenprocoumon, patients were asked to fill in the BMQ.

To compare the beliefs of coumarin users with users of other cardiovascular drugs, data from the study of van Geffen et al. [11] were used. In that study, users of an antihypertensive drug or a statin were included, and they all received the BMQ by mail. Patients using a coumarin derivative were excluded from our analysis. The study included new users who had not used any antihypertensive drug or statin in the previous 2 years and chronic users who had been prescribed at least 40 prescriptions in the previous 3 years. More information about the study can be found elsewhere [11]. Given that all coumarin users in the EU-PACT study were new users, we looked at new and chronic users in the data of van Geffen et al. [11] separately.

Beliefs about Medicines Questionnaire

The BMQ-Specific used in this study focused on beliefs about acenocoumarol and phenprocoumon or other cardiovascular medicines. Participants of EU-PACT trial were asked to give their beliefs about coumarin derivatives and participants of the study of van Geffen et al. [11] were asked about all their cardiovascular drugs. The BMQ-Specific questionnaire consists of two scales [6]. The necessity scale is focused on patients' beliefs about the necessity of using their medicines. The concerns scale is focused on the concerns that patients may have about their medicines. Each item is scored on a 5-point Likert scale, as follows: 1, strongly disagree; 2, disagree; 3, uncertain; 4, agree; or 5, strongly agree. The scores obtained from the individual questions in each scale were summed, divided by the total number of statements in the scale and then multiplied by five [5]. The range of possible scores on each scale was 5–25. Higher scores represented stronger necessity or concern beliefs. A necessity–concerns differential was calculated by subtracting the concerns scores from the necessity scores (range −20 to 20).

Statistical analysis

Data were analysed using IBM SPSS Statistics version 19 (IBM Corp, Armonk, NY, USA). Patients were excluded from the analysis if more than two answers of the BMQ were missing. Missing item scores of included patients were replaced by a score of 3 (uncertain) on the Likert scale. To confirm the psychometric properties of this Dutch version of the BMQ used in coumarin users, a principal component factor analysis was performed with Varimax rotation. Internal consistency of different parts of the BMQ was estimated using Cronbach's α.

Mean and median necessity and concerns scores and the necessity–concerns differential were calculated for patients with AF and for patients with VTE. The scores for the necessity and concerns scales were split at the scale midpoint to create the following four belief groups: indifferent (low necessity beliefs, low concerns); accepting (high necessity beliefs, low concerns); sceptical (low necessity beliefs, high concerns); and ambivalent (high necessity beliefs, high concerns). These groups have been used in earlier studies also [5,11,12]. Chi-square tests were used to compare the percentage of patients with a specific attitude for AF and VTE patients or new and chronic users.

Although the EU-PACT study was not designed to assess the association between the beliefs about coumarin anticoagulants and the percentage of time spent in the therapeutic International Normalized Ratio (INR) range (PTIR), we nevertheless looked at the correlation between the necessity and concerns scores and the PTIR and also checked for differences in PTIR between the different belief groups.

Results

Participants

In The Netherlands, 340 patients were included in the EU-PACT trial. Of these, 20 patients did not complete the BMQ. One answer was missing (and replaced by a score of 3) in two (0.6%) patients; there were no patients with more than one answer missing. Table 1 shows the characteristics of the 320 patients included in this study.

Table 1.

Characteristics of included patients

	EU-PACT trial [10] (n = 320)		Van Geffen study [11] (n = 493)*
Characteristics	AF (n = 249)	VTE (n = 71)	New users (n = 301)	Chronic users (n = 192)
Age [years; mean (range)]	68 (37–90)	54 (22–83)	59 (19–88)	67 (39–98)
Sex [n (%)]
Male	156 (62.7%)	37 (52.1%)	149 (49.5%)	111 (57.8%)
Female	93 (37.3%)	34 (47.9%)	152 (50.5%)	81 (42.2%)
Medication [n (%)]
Acenocoumarol	119 (47.8%)	44 (62%)	–	–
Phenprocoumon	130 (52.2%)	27 (38%)	–	–
Other antithrombotic	–	–	57 (18.9%)	71 (37%)
No antithrombotic (one or more other cardiovascular drugs)	–	–	244 (81.1%)	121 (63%)
Comorbidity [n (%)]
Hypercholesterolaemia	63 (25.3%)	3 (4.2%)	107 (37.5%)	101 (54.3%)
Hypertension	121 (48.6%)	14 (19.7%)	179 (61.1%)	148 (79.6%)
Diabetes	38 (15.3%)	3 (4.2%)	40 (13.7%)	74 (39.8%)
Angina pectoris	8 (3.2%)	0 (0%)	20 (7.6%)	41 (23.7%)
Myocardial infarction	21 (8.4%)	1 (1.4%)	12 (4.5%)	22 (12.7%)
Transient ischaemic attack	7 (2.8%)	1 (1.4%)	14 (5.3%)	15 (8.7%)
Stroke	7 (2.8%)	1 (1.4%)	11 (4.2%)	5 (2.9%)

Abbreviations are as follows: AF, atrial fibrillation; VTE, venous thromboembolism.

^*Coumarin users were excluded from this analysis.

Beliefs about acenocoumarol and phenprocoumon

Principal component factor analysis confirmed the original structure, with two components for necessity beliefs and concerns. Internal consistency was similar amongst the components (α = 0.66 for concerns and α = 0.67 for necessity). Factor loadings are reported in Table 2.

Table 2.

Principal component analysis using Varimax rotation

	Concerns	Necessity
My health, at present, depends on my coumarin	–	0.673
Having to take coumarins worries me	0.548	–
My life would be impossible without my coumarin	–	0.692
I sometimes worry about the long-term effects of my coumarin	0.724	–
Without my coumarin I would be very ill	–	0.682
My coumarin is a mystery to me	0.400	–
My health in the future will depend on my coumarin	–	0.663
My coumarin disrupts my life	0.656	–
I sometimes worry about becoming too dependent on my coumarin	0.734	–
My coumarin protects me from becoming worse	–	0.542
This coumarin has unpleasant side-effects	0.566	–

Factor loadings >0.30 are reported.

The necessity and concerns scores and the necessity–concerns differential of acenocoumarol and phenprocoumon users are shown in Table 3. The mean necessity score for the entire population was higher than the scale midpoint (15.3), while the mean concerns score was lower than the scale midpoint (12.3). This led to a positive mean necessity–concerns differential of 3.0. The highest mean score was seen in the item ‘My health, at present, depends on my coumarin’. For this item, 49.7% of the patients agreed or strongly agreed. The lowest score was seen in the item ‘My coumarin disrupts my life’, for which 80.3% of the patients disagreed or strongly disagreed.

Table 3.

Mean scores ± SD on the BMQ-specific

	EU-PACT trial [10] (n = 320)		Van Geffen study [11] (n = 493)
Beliefs	AF (n = 249)	VTE (n = 71)	New users (n = 301)	Chronic users (n = 192)
Necessity	14.9 ± 2.9	16.8 ± 3.5*	14.9 ± 4.2	17.9 ± 3.6*
Concerns	12.5 ± 3.1	11.9 ± 3.1	13.3 ± 4.1	14.1 ± 4.1*
Necessity–concerns	2.4 ± 3.9	4.9 ± 3.7*	1.5 ± 4.4	3.8 ± 4.4*

Abbreviations are as follows: AF, atrial fibrillation; VTE, venous thromboembolism.

^*P < 0.05 for AF vs. VTE or new users vs. chronic users.

Patients with AF had significantly lower necessity beliefs (14.9) than patients with VTE (16.8, P < 0.001). Also the necessity–concerns differential was higher in VTE patients than in AF patients (4.9 vs. 2.4, P < 0.001).

Figure 1 shows the necessity and concerns scores of patients with VTE or with AF and the percentage of patients in each belief group. Most patients had low concerns; 85% of the patients were either in the indifferent or in the accepting group. Fifty-eight per cent of the patients with VTE were accepting, vs. 34% of the patients with AF (P < 0.001). The AF patients were more frequently in the indifferent group than VTE patients (51 vs. 31%, P = 0.003). There were no significant differences in INR control (PTIR) between the belief groups (P = 0.58), and there was no significant correlation between necessity and PTIR (Pearson correlation −0.015, P = 0.79) or concerns scores and PTIR (Pearson correlation −0.024, P = 0.67).

Figure 1.

Scatter plot of the necessity and concerns scores by indication. In the boxes, the percentage of venous thromboembolism (VTE) and atrial fibrillation (AF) patients that have the specific attitude towards the therapy is shown. ◊, VTE; , AF

Comparison with beliefs about other cardiovascular drugs

Data of 578 patients were available from the study of van Geffen et al. [11]. Of these, 529 patients filled in all of the questions of the BMQ. One answer was missing in 18 (3.3%) patients; two answers were missing in four (0.7%) patients. In total, 27 patients missed more than two items, and these patients were excluded from the present analyses. As we wanted to compare coumarin users with users of other cardiovascular drugs (and other antithrombotics), we excluded 58 coumarin users from the study of van Geffen et al. [11]. Table 1 shows the characteristics of the 493 patients included in the analysis of the study of van Geffen et al. [11]. There was a greater proportion of females in this study compared with the EU-PACT study, and cardiovascular comorbidity was also more prevalent. Principal component factor analysis with Varimax rotation resulted in the same two-component structure as in the EU-PACT study.

The necessity and concerns scores and the necessity–concerns differential of cardiovascular drug users are shown in Table 3. The mean necessity score of the entire population was higher than the scale midpoint (16.1), while the mean concerns score was lower than the scale midpoint (13.6). This led to a positive mean necessity–concerns differential of 2.4. The highest mean score was seen in the item ‘My cardiovascular medicines protect me from becoming worse’. For this item, 59.2% of the patients agreed or strongly agreed. The lowest score was seen in the item ‘My cardiovascular medicines disrupt my life’, for which 77.7% of the patients disagreed or strongly disagreed.

Chronic users of cardiovascular drugs had significantly higher necessity scores (17.9 vs. 14.9, P < 0.001), higher concern scores (14.1 vs. 13.3, P = 0.046) and a higher necessity–concerns differential (3.8 vs. 1.5, P < 0.001) than new users. New users of cardiovascular drugs did not have a significantly different necessity score from new users of coumarin derivatives (14.9 vs. 15.3, P = 0.13); however, these patients did have a higher concerns score (13.3 vs. 12.3, P = 0.01) and a lower necessity–concerns differential (1.5 vs. 3.0, P < 0.001) than coumarin users.

Figure 2 shows the necessity and concerns scores of new and chronic users of cardiovascular drugs and the percentage of patients in each belief group. Most patients had low concerns; 69% of the patients were either in the indifferent or in the accepting group. Forty-six per cent of the chronic users were accepting, vs. 26% of the new users (P < 0.001). New users were more frequently in the indifferent group than chronic users (45 vs. 20%, P < 0.001).

Figure 2.

Scatter plot of the necessity and concerns scores by duration. In the boxes, the percentage of new and chronic patients that have the specific attitude towards the therapy is shown. ◊, new user; , chronic user

Discussion

Users of coumarins display stronger necessity beliefs compared with concerns, which is also the case in users of other cardiovascular drugs. Patients with VTE score higher on the necessity scale and the necessity–concerns differential than patients with AF. This is also true for chronic users of antihypertensives or statins if compared with new users of these drugs. The fact that patients score higher on the necessity scale than on the concerns scale would indicate a positive attitude towards these drugs. The most positive attitude is found in coumarin users with VTE or chronic users of cardiovascular drugs. Patients with AF have a less positive attitude and might benefit from extra attention to improve their adherence.

A possible limitation to this study is that the beliefs about coumarin derivatives were measured among patients participating in a randomized, controlled trial. This could have led to a selection bias, through the selection of more patients with a positive attitude towards the drugs. In this trial, however, no new drug was being investigated, but rather a new dosing strategy. On the contrary, patients with more concerns about the adverse effects of coumarins could have been more willing to participate, hoping that genotype-guided dosing would reduce the risk of adverse events. The study of van Geffen et al. [11] was an observational study. The results of the necessity and concerns scores in that study were similar to the results from the EU-PACT trial (necessity above scale midpoint, concerns below scale midpoint and necessity–concerns differential ∼3). The comparability between the two studies could also be hampered by the differences in the way the questionnaire was administered (during a visit to the anticoagulation clinic vs. mailing the questionnaire) and the timing of the assessment. The effect of data collection method on the observed differences between these two groups remains unclear. Also, the two studies differed in patient characteristics, indication for the drugs and the kind of prescriber. A further limitation is the single assessment of the beliefs, which might change after several months of treatment.

To our knowledge, this is the first study investigating the beliefs about coumarin derivatives. In the study by van Geffen et al. [11], the beliefs about cardiovascular medication were investigated in relationship to satisfaction with information. In their study, some coumarin users were also included, but not analysed separately. In our study, the mean necessity score was above scale midpoint. This is in agreement with the necessity scores in the other studies. In a study on the beliefs about inhaled corticosteroids in The Netherlands, the necessity score was 15.6, which is similar to the score of 15.3 in the present study [5]. Necessity scores were even higher in a study on Norwegian patients with a mental disorder (17.2) [2] or in British patients with rheumatoid arthritis (19.2) [4]. In those two studies, the concerns scores were also above the midpoint (17.9 and 15.8 respectively). In other studies, the concerns were lower, as in the present study [3,5,13]. The necessity–concerns differential was positive in most cases, except for patients with a mental disorder, in whom this differential was −0.70 [2].

In many of these studies, the beliefs about medicines were shown to be related to the adherence [1–5,12]. According to Clifford and colleagues [1], intentional non-adherence was associated with lower necessity beliefs and higher concerns. Unintentional non-adherence was not associated with the beliefs about medicines. Aikens et al. found that adherence was associated with the necessity–concerns differential and was highest in patients in the ‘accepting’ group (high necessity beliefs, low concerns) [12]. Most of the coumarin users with VTE are in this group (58%). Coumarin users with AF, however, are more often in the ‘indifferent’ group (low necessity beliefs, low concerns). This could mean that patients with AF are at increased risk of poor adherence. Patients using coumarins for VTE generally have more complaints (pain) and therefore see the usefulness of coumarins more than patients using these drugs for AF. This could explain the higher necessity scores and necessity–concerns differential in this group. Although patients with AF often have complaints, these are generally controlled with rate or rhythm control. The additional antithrombotic treatment might therefore be perceived as less important. It would be interesting to look at the differences in beliefs between patients with complaints related to AF or with a previous transient ischaemic attack or stroke and patients without any complaints. In this study, this information was not available, and the groups of patients with a previous transient ischaemic attack or stroke were too small to perform a subgroup analysis. It is, however, important to note that comorbidity was more frequent in patients with AF (Table 1). This may have led to confounding by the burden of disease or polypharmacy, because these are more often seen in AF patients than in VTE patients.

Van Geffen et al. showed that the beliefs about medicines were associated with the patients' needs for information and counselling [11]. The BMQ could be used to identify the patients who would benefit most from extra patient education. This education could address any questions they might have and dispel any misunderstanding or concerns. This could decrease the concerns, and if the necessity of the treatment can be explained clearly to the patient, this might increase the necessity beliefs.

As warfarin is not used in The Netherlands, we did not include warfarin users in this study. However, no differences were found between the two coumarins investigated in this study, and it is therefore likely that similar results would be seen for warfarin. The choice between acenocoumarol and phenprocoumon in The Netherlands is mainly based on the region, rather than on patient characteristics, so no selection bias is expected regarding the prescribed coumarin. All patients in this study attended an anticoagulation clinic, which is standard practice in The Netherlands. This might have caused a slightly different view of the drugs than patients treated by, for example, the general practitioner, because these clinics have considerable experience with the drugs and provide relevant information and education to patients. We also have no data on patients doing self-management or self-monitoring of the treatment, which could change their beliefs about the treatment. The EU-PACT study took place in a stable socioeconomic environment, and patients were well educated, which might make the results less transferable to a setting where these characteristics are different. However, socioeconomic factors have not consistently been linked to either beliefs about medicines or medication adherence.

Recently, new oral anticoagulant drugs (direct thrombin inhibitors and factor Xa inhibitors) have been developed, and these are considered to be good alternatives to coumarin derivatives. Unlike coumarins, these new drugs do not require monitoring, which could lead to a higher risk of poor adherence. However, the adherence to these drugs has not yet been investigated extensively and is still reason for concern [14]. Given that the adherence in clinical trials is generally assumed to be higher than in clinical practice, the effectiveness of the new drugs could be less favourable than what is currently seen in these trials. The results of this study show that patients using coumarin derivatives generally have a positive attitude towards their therapy. As a result of this positive attitude and the fact that frequent monitoring is required, the overall risk of poor adherence with coumarin derivatives is probably low. In some groups, however, this risk is higher than in others (for example patients with AF compared with VTE patients). Patients with AF could benefit from extra attention, for instance with patient education.

Future studies can investigate the association of the beliefs with patient adherence in coumarin users. In this study, we could not demonstrate an association between INR control and beliefs, although the quality of INR control was shown to be associated with adherence in the past [15]. In a larger study, the correlation with age, educational levels or the presence of comorbidities could also be tested, because the subgroups in this study were small. In clinical practice, the BMQ could be used as a simple tool to measure patient beliefs and subsequently identify patients with a negative attitude towards anticoagulant therapy. Knowledge of the beliefs the patient has about the medication can be taken into account during contact with patients to improve their attitude towards the treatment and perhaps subsequently improve their therapy adherence.

Competing Interests

All authors have completed the Unified Competing Interest form at http://www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare: TIV, WKR, RMFvS, AdB and AHMvdZ had support from the European Community's Seventh Framework Programme for the submitted work; no financial relationships with any organizations that might have an interest in the submitted work in the previous 3 years; and no other relationships or activities that could appear to have influenced the submitted work.