INTRODUCTION

Phosphorus (F) control in patients on haemodialysis constitutes one of the most important issues facing nephrologists today. Various studies have demonstrated the difficulty of achieving the goal of F proposed by the K/DOQI (P <5.5 mg/dL) 1-3 and its impact on morbidity and mortality.4-6 Additionally, the goals for phosphorus control are steadily becoming more ambitious and, in fact, figures are being proposed that come ever closer to normal levels.6

In all likelihood, the lack of compliance with phosphorus binder (PB) treatment plays an important role in the difficulty in achieving the objectives of serum phosphorus level control,7 although the causes for this are not clearly understood.

PBs present certain characteristics that distinguish them from other drugs, in particular, they have to be taken with food and their interference with an individual’s lifestyle and social habits impacts on the lack of treatment compliance.8 A study performed on AIDS patients9 showed that the percentage of patients who did not comply with the treatment increased considerably when the definition of compliance also took dietary considerations into account.

Today we have a wide array of PBs10-14 with different presentations and characteristics. One key concept is to understand in what measure each of the different types of PB is adapted to an individual’s habits and behaviour as well as their expectations, beliefs and preferences.15 The emergence of increasingly expensive PBs16 makes non-compliance have a significant economic impact, given that it affects the efficiency (cost-effectiveness) of treatments, which hinders achieving clinical effectiveness with the lowest consumption of resources possible.

The aim of this study was to expand our understanding of patients’ preferences and beliefs regarding PBs and their influence on compliance and in achieving therapeutic goals.

PATIENTS AND METHODS

This was an observational transversal study. A total of 121 out of 165 dialysis patients from our unit were included. Forty-four patients were excluded from the study: 9 patients were unable to answer the questionnaire due to dementia or significant hearing loss; 12 patients were not prescribed PBs and 23 took a single PB, which prevented comparison with other types of binders.

The 121 patients who were receiving a combination of two (85 patients) or three PBs (36 patients) answered a questionnaire specifically related to their preferences regarding the PBs. The specific questions about PBs referred to drugs that the patients regularly took (more than 3 months of treatment). Thirty-six patients took three types of PB: aluminium hydroxide (Pepsamar®), calcium acetate (Royen®) and sevelamer (Renagel®). Eighty-five patients took two: calcium acetate (Royen®) and sevelamer (Renagel®). In January 2008, 45 patients who previously had been treated with high doses of sevelamer had their medication changed to lanthanum carbonate (Fosrenol®), which allowed this new medication to be included in the preferences analysis along with the rest of the PBs after having been used for at least two months by the patient. Data was collected between March and April 2008.

Sixty-three percent of the patients were men with a mean age of 64.8 ± 14.2 years and a mean time on haemodialysis of 154.6. ± 192.4 months (3.74 to 931.5 months). The median Charlson comorbidity index was 8 (P25-P75: 6-10). 

The average number of daily drugs prescribed for this population was 9.8 ± 3.0 and the average number of daily tablets was 18.6 ± 7.9, of which 51% were in PB form (average number of daily PB tablets: 9.5 ±6.0). All patients had 100% compliance for attending the prescribed dialysis sessions. The average number of daily tablets for patients taking calcium acetate (Royen®) was 4.9 ± 2.9; 3.4 ± 1.7 for aluminium hydroxide (Pepsamar®); 7.1 ± 3.5 for sevelamer (Renagel®); and 3.0 ± 0.0 for lanthanum carbonate (Fosrenol®).

Methods for assessing treatment compliance

In general, drug treatment compliance was measured using the SMAQ compliance questionnaire, which has been validated for the Spanish AIDS population17 (Table 1). The answer choices were dichotomous; any response leaning toward non-compliance was considered as not following treatment. This questionnaire has been shown to have sufficient internal consistency (Cronbach’s alpha = 0.75) in patients infected with HIV, it is easily reproducible (overall agreement 88.2%; kappa = 0.74)17 and it has been shown to have sufficient internal consistency in haemodialysis patients (Cronbach’s alpha for this population = 0.70).8 The specific compliance with PB treatment was analysed by means of a specific non-validated questionnaire that asked about PB treatment compliance, the preferred type of PB and the reasons why, the least-liked type of PB and the reasons why, as well as a series of questions about their knowledge of the usefulness and importance of the drug (“What use have PBs?”, “Why do you think they are important?”) and about patient habits (outings, social life).

Measurement of clinical consequences of non-compliance

The consequences of non-compliance with PB treatment were estimated indirectly by determining serum phosphorus levels. Failure to reach the objective occurred when the average for all monthly phosphorus measurements performed in the last 6 months exceeded 5.5 mg/dl. Phosphorus levels were measured using UV spectrometry (normal range 2.7 to 4.5 mg/dl).

Related variables

We studied possible factors associated with the lack of compliance, and analysed the following variables that might affect it:

1. Age.

2. Sex.

3. Time in haemodialysis.

4. Charlson co-morbidity index.18

5. Number of drugs prescribed.

6. Number of tablets per day prescribed.

7. Number of PB tablets per day prescribed.

8. Degree of acceptance of each type of PB. Patients were questioned about which PB they preferred and which they liked the least and the reasons why.

9. Knowledge of the usefulness of the drug and its importance.

10. Influence of habits (outings, social life).

Statistical analysis

The statistical analysis was performed using SPSS 12.1. We used the Student’s t-test for the comparison of independent samples when the data followed a normal distribution. For those cases in which the data did not, we used the Wilcoxon test. The qualitative variables were compared using the chi-squared test and the extent of the association was quantified using the odds ratio calculation with a confidence interval of 95%. Logistic regression analysis was used in order to evaluate the possible influence of each of the independent variables on PB treatment compliance. The dependent variables were the degree of compliance estimated with the SMAQ questionnaire, the degree of PB treatment compliance estimated with the specific questionnaire, and the average serum phosphorus levels. The independent variables were age, time on haemodialysis, comorbidity, use of vitamin D, the number of different types of oral medication prescribed daily, the number of tablets prescribed daily and the total number of PB tablets. Statistical significance was considered to be at p < 0.05.

RESULTS

SMAQ compliance questionnaire

According to the SMAQ compliance questionnaire, 40% (49/121) of the patients assessed were defined as not following some of the prescribed treatments.

In the 6 months prior to assessing compliance, 14% of patients studied (17/121) had an average phosphorus level > 5.5 mg/dl and 7.4% (9/121) had >6 mg/dl. The 39.1% of patients who complied with treatment and 60.9% of those who did not had serum phosphorus levels > 5.5 mg/dl (OR = 4.8; IC 95%, 1.0-6.6; p = 0.02). The group of patients that did not comply with the treatment according to the SMAQ questionnaire had an average serum phosphorus level significantly higher than the group of patients that complied with the treatment (4.7 ± 0.9 versus 4.4 ± 0.7; p <0.01).

There were no significant differences in the averages for Kt/V and nPCR between patients with P >5.5 mg/dl (1.45 ± 0.54 and 1.0 ± 0.5 g/dl, respectively) and with p <5.5 mg/dl (1.46 ± 0.54 and 1.1 ± 0.4 g/dl, respectively)

Questionnaire for specific compliance for phosphorus binders

Some 21.4% of the patients admitted non-compliance with therapy specifically for PBs. When questioned about which drug they would not take if they could choose, 53 patients said they would not stop taking any of the drugs they were taking (43.9%) and 12 patients said they would stop them all (9.9%). The group of drugs that the largest percentage of patients would choose to stop taking were the PBs (20.6%), followed by Resincalcio® (4.9%) and the hypotensive drugs (3.3%). The rest (17.5%) corresponded to other types of drugs in smaller rates.

Patients who specifically did not comply with the PB treatment had a greater probability of having average phosphorus readings >5.5 mg/dl (OR = 4.7; IC 95%, 1.07-6.5; p = 0.03). Table 2 shows the factors related to lack of PB treatment compliance.

The logistic regression analysis showed that none of the independent variables (age, time on haemodialysis, comorbidity, use of vitamin D, the number of different types of oral medication prescribed daily, the number of tablets prescribed daily and the total number of PB tablets) were independently associated with a lack of overall or specific compliance or with the average levels of serum phosphorus.

Preferences, knowledge and habits relating to phosphorus binders

Some 43.8% (53/121) of the patients had no preferences for any particular PB and 6.6% (8/121) did not reply to this item in the survey. Some 54.5% of the patients had a PB prescribed that they did not like. Table 3 lists the preferences of the patients for the different types of binders and Table 4 shows the reasons justifying them.

Some 24.4% of patients did not take the medication when leaving the house or staying with friends, posing a greater risk of not meeting the goal of P < 5.5 mg/dl (OR = 10.5; IC 95%, 1.8-16.4; p < 0.001). These patients were younger (59.3 ± 15.8 versus 66.8 ± 14.5; p <0.01) and had a greater number of daily PB tablets (12.03 ± 7.9 versus 8.7 ± 5.5; p <0.01)

Patient knowledge and beliefs about the usefulness and importance of PB are shown in Table 5.

Patients who took a PB that they did not like had a higher risk of having P readings that were out of control (>5.5 mg/dl) (OR = 13.3; IC 95%, 1.1-1.5; p = 0.0001). Paradoxically, patients who did not comply with phosphorus binder treatment had a better understanding of the usefulness (OR = 17.3; IC 95%, 2.2-10.1; p <0.0001) and importance of the drug (OR = 10,4; IC 95%, 1.5-6.6; p = 0001).

DISCUSSION

Compliance with CKD treatment in its various aspects is essential and a large part of its results depend on it. This study shows a lack of treatment compliance with prescribed drugs in haemodialysis patients that reaches 40%. In 21% of cases, there was a specific lack of compliance for PB, which was associated with higher average serum phosphorus levels. There are few references that address this issue specifically and the data reported vary, according to the studies, between 22 and 74% of patients, with a median of 50%.7,19 It is well known that the lowest compliance levels are observed in chronic patients in whom there is no discomfort or immediate risk and in whom the treatment requires a lifestyle change. Patients with CKD belong to this group of patients.20

The factors that determine treatment compliance in general are diverse. Firstly, clear and consistent relationships have not been found between compliance and socio-demographic variables or population characteristics7 except in the case of age (young patients are more likely not to comply). The profile of a patient that does not comply with PB treatment was a young patient with lower comorbidity and a greater number of PB tablets prescribed.

What has been demonstrated is that the complexity and demands of the treatment regimen influence the lack of compliance; consequently, the number of daily doses is inversely proportional to the level of compliance and this, in turn, is associated with a lower patient quality of life.19 Patients on haemodialysis are prescibed a high number of daily tablets of which, in our study, more than half are PBs. Thus, in a previously conducted study by our group,22 with 51 patients with difficulty controlling phosphorus and who received large doses of sevelamer (more than 9 tablets/day), changing sevelamer for 3 daily tablets of lanthanum carbonate helped achieve a better control of serum phosphate (the percentage of patients with serum phosphate <5.5 mg/dl increased from 46.9 to 72.5%; p<0.01). The reduction in the number of tablets may somehow influence patient compliance with treatment and therefore, produce better clinical outcomes, as other studies have suggested.19

Interference with daily habits, whether during work hours or within the context of a patient’s social life, may motivate some of them to stop taking medication or to take it at the wrong time. This occurs more frequently with young patients who have a more active social life. PBs require a very strict dosage regimen since they must be taken during or immediately after each meal and, generally, three times a day, which may be a factor in promoting non-compliance with treatment with these drugs as opposed to other drugs.8,19 Indeed, a quarter of our patients admitted not taking the medication when out of the house or when being with friends. A study performed on AIDS patients23 showed that the percentage of patients that did not comply with treatment increased considerably when the definition of compliance also took dietary considerations into account.

Simplifying the therapy regimen has a certain positive effect but does not solve the problem of non-compliance since it is not the only factor involved. Another factor involved, which was not analysed by our study but should be the subject of future research, was the influence of the amount of time a patient had been taking a particular type of binder, since taking a certain medication for a long period of time may cause the patient to tire of doing so. In this sense, lanthanum carbonate may have a certain “advantage” in our study since it was introduced only two months prior to its initiation. For example, in our previously-mentioned study on the conversion from sevelamer to lanthanum carbonate, after 12 months and despite the potential advantages in reducing the number of tablets, 31% of patients preferred to go back to sevelamer.20

Additionally, an adequate understanding of the drugs’ actions may positively influence compliance. Our study demonstrated the significant lack of knowledge about the usefulness and importance of PBs. Almost half of the patients studied did not know what the drugs were for or the extent of their importance. Not one patient related these drugs with cardiovascular disease. Adequately conveying the importance of these drugs is essential and may contribute to increasing treatment compliance.24 However, knowledge per se does not ensure compliance. In fact, non-compliant patients taking PB show, paradoxically, a greater knowledge of their use and importance, possibly due to dealing with patients who, faced with the lack of phosphorus control, had been presssured more towards compliance and had received more explanations from the doctor. The influence of age cannot be ruled out, given that younger patients tend to better understand the indications of chelators.

It is known that a high percentage of patients quit treatment to avoid side effects, regardless of the clinical relevance that these may have, which means that drugs that frequently induce digestive symptoms may constitute a greater risk of non-compliance. In our population, digestive intolerance was among the negative side effects of PB, which include constipation, flatulence, nausea, abdominal discomfort and dyspepsia. The PB that most digestive symptoms were attributed to was aluminium hydroxide (18%), followed by lanthanum carbonate and sevelamer (6%). Calcium acetate caused the least number of adverse digestive symptoms (4%).

More than half of the patients in our population did not like the PB they were prescribed, and patients who took a PB that they did not like had a higher risk of having F readings that were out of control (>5.5 mg/dl). Today we have a wide array of PBs with different characteristics. The majority of patients studied received various type of binders and almost half took two or more types. The PB most positively regarded among patients that took it was calcium acetate, followed by lanthanum carbonate, sevelamer and, finally, aluminium hydroxide. The reasons given by the patients for negatively rating the PBs were the form, taste, number of tablets and gastric intolerance. Sevelamer received complaints about its form for its excessively large tablets and for requiring a large number of them, which required the patient to ingest large quantities of water. Some 17.7% of patients that took lanthanum carbonate did not like the chewable form and, in fact, some elderly patients had trouble chewing them. Almost a quarter of the patients that took aluminium hydroxide reported unpleasant taste.

The first step required for improving therapy compliance is proper diagnosis of the problem. In some cases, certain objective parameters, such as the levels of phosphorus, will clearly indicate that we are faced with a case of non-compliance. Compliance is often subjectively assessed in clinical practice. This prevents the identification of many patients who may be non-compliant, missing the opportunity to intervene in their behaviour. This also facilitates the adoption of therapeutic attitudes with a high degree of empiricism, which in many cases causes an unnecessary increase in the doses of prescribed drugs. This is justifed by a theoretical lack of response to treatment even though the underlying truth is a lack of compliance.

The compliance questionnaires are tools that require few resources and are affordable and adaptable to the characteristics of each centre. Research on the use of compliance questionnaires is a continually advancing issue. It would be desirable to incorporate new instruments that are validated for these types of patients. The SMAQ questionnaire has been validated for HIV+ patients in the Spanish population but could also be used with CKD patients. In our study we decided to use a combination of methods to evaluate compliance: the average levels of serum phosphate and the responses to the SMAQ questionnaire on compliance with the medication in general and a specific non-validated questionnaire in which the patients were asked about their compliance with PBs. We found that the factors associated with these three dependent variables were similar, with a statistically significant association between patients with average levels of P >5.5 mg/dl in the previous months and the lack of treatment compliance in accordance with both questionnaires. These correlations seem to additionally validate the use of these questionnaires with patients on haemodialysis. Additionally, the SMAQ showed sufficient internal consistency in this group of patients (Cronbach’s alpha = 0.70). Knobel et al17 showed a similar internal consistency (Cronbach’s alpha = 0.75) among HIV+ patients as well as a satisfactory reproducibility (88.2% overall agreement, kappa 0.74). The SMAQ can show its validity for evaluating treatment compliance in patients on haemodialysis: it shows adequate levels of sensitivity and specificity when compared to other more objective measurements; is correlated with phosphorus levels; is reliable; has sufficient internal consistency and reproducibility; is easy to apply (takes about 5 minutes); and is inexpensive.  

Knowing the patients’ preferences regarding the prescribed drugs may be another major factor in achieving good treatment compliance. It is important for both nephrologists and the industry to understand these aspects. The pharmaceutical industry, based on their knowledge of patient preferences, should try to develop drugs that adjust to these preferences and develop presentations that are better tolerated and that pass through the patients as unnoticed as possible. In addition, as long as there are no medical contraindications, nephrologists can use in each case those PBs that each patient prefers or combinations thereof in order to minimise side effects. Some authors have suggested25 that a combination of low doses of different drugs may be a more cost-effective method to ensure tolerability, efficacy and compliance. In our study, patients who took binders that they did not like showed poorer control of serum phosphorus and were the ones prescribed the highest doses of PB. If lack of compliance is not explored, it is current practice for the doctor to increase the prescribed doses, which fails to achieve the objective.19 The relationship that is established between health care providers and patients is of utmost importance. Trust, continuity, accessibility, flexibility and confidentiality are all factors that favourably influence compliance. Supervision of therapy compliance by the team that cares for the dialysis patient is a measure that may contribute favourably to patient compliance with the prescribed treatment. Probably one of the most important factors is providing detailed and realistic information, as well as joint decision making in a framework of mutual trust (patient-centered model).26

To summarise, short and long-term compliance is the result of a complex process that is developed through various stages: acceptance of the diagnosis, perception of the need to perform correct treatment, the motivation to do so, the provision and training of skills to carry it out, the capacity of overcoming the barriers and difficulties that may arise, and the maintenance of achievements over time. Treatment must be individualised and adapted to the needs and preferences of each patient. It is essential to understand the patient’s daily habits, personal resources and family, and analyse their understanding of the disease and the degree of awareness and confidence they have in order to start the treatment. The assessment of possible risk factors for achieving an optimal treatment compliance should serve in planning specific interventions for each patient, interventions that will generally be multifaceted and multidisciplinary.27

In conclusion, patients who took PBs that they found unpleasant, for various reasons, had poorer phosphorus control despite having a larger quantity of these drugs prescribed for them, which leads us to believe that the reason for this lack of control may lie in a lack of treatment compliance. A greater understanding of the patients’ preferences, the use of different drugs in lower doses, a better doctor-patient relationship and detailed and accurate information as part of a joint decision-making process may be key factors in achieving phosphorus control objectives.28,29

ANNEX. QUESTIONNAIRE ON PHOSPHORUS BINDER PREFERENCES

1. Which of the medications that you take would you choose to drop?

2. Which of the medications that you take for controlling phosphorus (Pepsamar®, Royen®, Renagel® and Fosrenol®) do you like the most? Why?

3. Which of the medications that you take for controlling phosphorus (Pepsamar®, Royen®, Renagel® and Fosrenol®) do you like the least? Why?

4. Of the medcations you are taking for phosphorus, indicate which ones you like or dislike:

                 * Pepsamar® LIKE / DISLIKE

                 * Royen® LIKE / DISLIKE

                 * Renagel® LIKE / DISLIKE

                 * Fosrenol® LIKE / DISLIKE

5. What do you think these medications are for?

6. Why do you believe these are important to take?

7. When you leave the house or are with friends, do you still take the medications?

10275_108_8384_en_w4777103751210275_18107_1558_es_10275_18107_1558_es_tabla1_copy1_en.doc

Table 1. SMAQ Compliance questionnaire

10275_108_8385_en_w4777103751110275_18107_1559_es_10275_18107_1559_es_tabla2_copy1_en.doc

Table 2. Factors related to treatment compliance or non-compliance of treatment with phosphorus binders (n = 121)

10275_108_8386_en_w4777103751510275_18107_1560_es_10275_108_1560_es_tabla3_en.doc

Table 3. Patient prefences with respect to phosphorus binders

10275_108_8387_en_w4777103751410275_18107_1561_es_10275_108_1561_es_tabla4_copy1_en.doc

Table 4. Reasons for justifying not taking various phosphorus binders

10275_108_8388_en_w4777103751610275_18107_1562_es_10275_108_1562_es_tabla5_copy1_en.doc

Table 5. Responses to questions related to beliefs about the usefulness and importance of phosphorus binders (n = 121)