A Comparative Study among Different Treatment Adherence Methods in Hypertensive Patients

Tania Pavão Oliveira Rocha1, José Albuquerque de Figueredo Neto2, Darci Ramos Fernandes1, Ewaldo Eder Carvalho Santana3, Jerusa Emídia Roxo Abreu4, Raimundo Luís Silva Cardoso4, Jorgiléia Braga de Melo5 1Universidade Federal do Maranhão Programa de Pós-graduação em Ciências da Saúde São Luís, MA Brazil 2Universidade Federal do Maranhão Departamento de Medicina I São Luís, MA Brazil 3Universidade Federal do Maranhão Departamento de Engenharia Elétrica São Luís, MA Brazil 4Universidade Federal do Maranhão Programa de Pós-graduação em Saúde e Ambiente São Luís, MA Brazil 5Universidade Federal do Maranhão Programa de Pós-graduação em Saúde Materno-infantil São Luís, MA Brazil


Introduction
Systemic hypertension (SH) is a multifactorial clinical condition characterized by high and sustained levels of blood pressure (BP).It is often associated with functional and/or structural changes to target organs (heart, brain, kidneys and blood vessels) and metabolic changes, with consequent increased risk of fatal and non-fatal cardiovascular events 1 .
In Brazil, a review study on SH prevalence, carried out in various cities, found SH prevalence rates ranging from 19% to 44% (depending on the criteria adopted and the assessment process), with higher rates among individuals aged over 60 and with lower educational level 2 .
Non-adherence to treatment, both in relation to patient's lifestyle changes and whether physician's prescription is followed, directly affects the control of blood pressure levels 3 .Moreover, the absence of symptoms and the chronic nature of hypertension are two aspects that strongly contribute to poor adherence to treatment 4 .
Adherence to hypertension treatment may be considered the coincidence degree observed between the patient's behavior and healthcare professionals' therapeutic recommendations 5 .Blood pressure control, reduced incidence or delayed occurrence of complications and improved quality of life are expected as a result 6 .
Although there are several strategies available to evaluate adherence to hypertension treatment, there is no consensus on which to consider the "golden standard" among them 7 .Methods can be either direct such as the dosage of the drug active ingredient/metabolite, or indirect such as the pill count and user's self-report 8 .Some authors have developed generic questionnaires to assess adherence to treatment in chronic diseases, which have also been applied to hypertensive individuals 9 .The use of validated instruments to measure the construct of adherence allows subjective data on the routine care provided to hypertensive individuals to be accurately measured 9 .
Non-adherence to hypertension treatment is a challenge for those who work and investigate this area.According to Mascarenhas et al. 10 , in order for adherence to occur, a patient should be knowledgeable about their condition, conscious of the importance of blood pressure control and have access to health care services, which must be maintained for the patient's entire lifetime.
Considering the relevance of this subject matter, this research aimed to make a comparison between three tools used for assessment of adherence to hypertension treatment, taking the rates found and blood pressure control into consideration.To check the relation between adherence to treatment and sociodemographic and clinical characteristics of patients studied.

Methods
This is a descriptive, cross-sectional analytical study conducted in São Luís, MA, Brazil.
Data was collected from August 2011 to September 2012, with 502 hypertensive individuals randomly drawn from different public health districts of the municipality 11 .The following inclusion criteria were used: male/female patients with systemic hypertension, who are users of the Estratégia Saúde da Família (Family Health Strategy) health care model, in the capital city of São Luís, state of Maranhão, over 18 years old, being monitored for at least six months.
A structured questionnaire was used to record the sociodemographic data (gender, age, marital status, educational level, occupation and family income); lifestyle habits (smoking, alcohol consumption and physical activity); diagnosis and treatment (elapsed time between diagnosis and treatment; if under hypertension treatment; type of treatment and return to physician for a follow-up visit) and clinical data (blood pressure measure, waist circumference, body mass index (BMI)).
For treatment adherence to be assessed, the following questionnaires were used: Qualiaids team's Medication Adherence Questionnaire (MAQ-Q), Morisky-Green test (MGT) and Haynes' questionnaire.
MAQ-Q 7 consists of three questions.It addresses the action (if the individual takes the medication and how much), process (how the medication is taken within a seven-day period; if doses are skipped or irregularly spaced; if doses are suspended for determined periods), and adherence results (in this case, if blood pressure was under control).Respondents are deemed non-adherent if they fail to take the correct dosage (80%-120% of the prescribed doses), or take it in a manner that was not recommended (with "suspended periods", "irregularly spaced doses", treatment abandonment or "partial adherence"), or report any alteration in blood pressure.
The Morisky-Green test 12 is based on the misuse of drugs that, according to the authors, occurs due to one or all of the following events: forgetfulness, carelessness, drug interruption due to improvement or worsening of the patient's health condition.It consists of four questions, and a positive answer to any of these questions classifies the individual as non-adherent.This test has been chosen because of its availability in Portuguese language, small number of questions, ease of measurement and understanding, and as it offers the possibility to check patient's behavior towards drug intake.
Haynes' questionnaire 13 consists of a single question whose positive answer classifies the individual as non-adherent.
This study protocol was approved by the Ethics Committee of Hospital Universitário Clementino Fraga Filho of Universidade Federal do Rio de Janeiro, under no.38/11.All participants signed an Informed Consent Form, in compliance with the Helsinki Declaration.

Results
Table 1 shows patients' answers to the three questionnaires used.Regarding the manner in which the medication is taken, obtained by the MAQ-Q, 106 patients (21.12%) reported not having taken the medication or taken at least one extra tablet, and 261 patients (51.99%) reported alterations in their blood pressure in the most recent blood pressure reading.As for MGT, of the four evaluated questions, the highest percentages of positive attitudes towards medication intake were for "When you feel better, do you sometimes stop taking medications?"-"No" (91.24%), and for "Sometimes, when you feel worse, do you stop taking your medicine?"-"No" (91.83%).In Haynes' questionnaire, most patients (86.85%) reported that they had no difficulty taking their medications in the last 30 days.Concordance assessed whether patients who were considered non-adherent by a method were also considered so by the other.
Table 3 shows the association between sociodemographic and clinical variables with non-adherence, considering each of the tests used.Concerning income, nonadherence was higher among those who received less than the minimum wage, with association only for Haynes (20.0%)."Smoking" and "return to physician for a follow-up visit" variables were significant in all three questionnaires; non-adherence was proportional among smokers and those who rarely attended follow-up visits.Among variables with statistical significance for both Haynes and Morisky, nonadherence rate was higher for retired individuals with paid work, non-alcoholic individuals and for those who used monotherapy as a form of treatment.

Discussion
The term adherence is used to refer to the compliance with therapy recommendations.Antihypertensive treatment adherence rates vary widely in the literature 7,[15][16][17][18][19][20] .Worldwide statistics show that treatment adherence rates range from 7-65%, with Japan being the country with the highest adherence rate and Slovakia 21 with the lowest.
Studies are not often comparable as they address different individual profiles and use different adherence identification methods.
In recent years, therapeutic adherence has become one of the biggest problems in medical practice due to its complexity.About 40-60% of patients do not use prescribed medication 22 .This percentage increases when lifestyle is evaluated 21 .
The use of questionnaires validated to assess adherence has increased in recent years.An integrative review study 9 showed this increase, especially in the years 2008 and 2009; a likely explanation for this growth is the interest of professionals in understanding the adherence/ non-adherence phenomenon.
The data obtained by this study has revealed that patients identified as hypertensive by MGT reported having had positive attitudes towards medication intake when answering the questions "When you feel better, do you sometimes stop taking medications?-"No" (91.24%) and "Sometimes, when you feel worse, do you stop taking your medicine?"-"No" (91.83%).These results were similar to those from other studies 17,18 .The highest number of negative answers (25.1%) was given for the question "Have you ever forgotten to take your medicine?"-"Yes", and 21.12% of the patients reported being careless about the time to take their medications.
A study conducted by Plaster 19 assessed adherence by means of the Morisky-Green test.That study found, as in this study, the highest rate of "yes" answers for questions 1 and 2. Adherence questions based on forgetfulness, carelessness and medications taken when patients feel better or worse enable our perception of patients from areas in which non-adherence is higher and lower.
Another tool used to assess adherence was the Haynes' test.In this case, the question about therapeutic adherence or non-adherence is made in a friendly manner, trying to express the lowest possible pressure.According to this test, 87.85% of the individuals reported no difficulty taking their antihypertensive medications.These results were similar to those from the study conducted by Melchiors 23 , Hamilton 18 and Santa-Helena et al. 7 , which obtained 70.2%, 84.0%, 88.0%, respectively.This may reflect that these tools have low sensitivity to detect non-adherence, which was also pointed out in a previous study 3 .
A research on adherence to antihypertensive drug treatment, conducted in the city of São Luís, State of Maranhão, found a non-adherence rate of 32.7% 24 and used as a measure the sum of the number of tablets taken in the last five days.
Non-adherence prevalence rate measured, in this study, by MGT, MAQ-Q and Haynes was, respectively, 29.28%, 60.16% and 13.15%.Non-adherence levels measured by MGT and Haynes, reported in this study, were lower than those found in other researches 7,[16][17][18]20,25 . Afterestablishing a correlation between hypertension control with indirect adherence tests, MAQ-Q and Haynes were found to have statistical significance, whereas among non-adherent patients, MAQ-Q was the test that identified the highest rate for uncontrolled hypertension.
In another study, this relation was not revealed 23 .Tests had a good concordance level, in contrast to what was obtained in the study by Bastos-Barbosa et al. 15 When used, the MAQ-Q found a non-adherence prevalence rate of 60.16%.This result was higher than that found in southern Brazil 7 where MAQ-Q resulted in a non-adherence rate of 47.8%.When compared to other methods, this tool obtained lower adherence rates, because when it is used, its cutoff point for non-adherence is increased as a manner to improve sensitivity, as it considers adherent those who took the correct dosage of their medicines, in the prescribed manner and reporting a favorable effect.
Findings in this research have shown a statistically significant relationship between variables: income, occupation, smoking, alcoholism, form of treatment and return to physician for a follow-up visit, with nonadherence in at least one of the tests.Income was described as a factor negatively associated with adherence to treatment of chronic diseases 26 , coinciding with results from this study, in which the non-adherent patients earned less than the minimum wage as monthly income.However, this association was not observed in other studies 17,20,27 .
Regarding occupation variable, it has been shown to have statistical significance with non-adherence in both Haynes and Morisky tests, in which retirees with paid work were those who had the highest non-adherence rates.Other studies 20,27 , however, show that retirement has a positive impact, as it provides patients with more time to dedicate themselves to the treatment.
Ungari and Dal Fabro 17 found no significant association between non-adherence level and the socioeconomic variables.Most authors unanimously report that association between therapeutic adherence level and sociodemographic factors is not consistent.
With regard to lifestyle habits such as smoking, significant differences were found in relation to nonadherence in the three methods used, with smokers being identified as those who least adhered to the treatment.These results were similar to those obtained in other studies 28,29 .It is worth of stressing that changing hypertensive patients' habits and lifestyle should be emphasized as part of therapeutic care, as it significantly contributes to the reduction of blood pressure.
Regarding treatment method, individuals with lower adherence rates were those using monotherapy, with statistical significance in Haynes and MGT, contrary to the results of other studies reporting that a higher adherence rate is associated with lower amounts of prescribed drugs and adoption of simple therapeutic treatments 30,31 .
Return to physician for a follow-up visit was associated with non-adherence in the three questionnaires, with patients who had the lowest adherence rate being those who rarely attended the follow-up physician visits.
According to a study by Santos et al. 32 , patients' attendance at scheduled physician appointments has positive correlation with treatment adherence, i.e., the more regularly patients attend to appointments, the greater will be their adherence to treatment, which is consistent with the results obtained in this research.
As limitations to this study, the lack of consensus on the optimal method for assessing treatment adherence can be mentioned, as well as the varied methods used in the literature, which rendered comparison difficult.Despite such limitations, it was possible to estimate adherence/ non-adherence to treatment in a sample of hypertensive patients from northeastern Brazil, contributing to a body of evidence that supports interventions aimed at this group of patients.
It is also inferred that the limitations of this study lie in the impossibility of listing the causes of non-adherence, suggesting that future research can foster an understanding of related factors, thus contributing by means of actions to increase the effectiveness of hypertension control programs.
It is also worth pointing out that this study was developed within the public service specific reality, suggesting the need for studies involving different contexts.

Conclusions
This study showed that in general the three study adhesion tests had a good concordance level.When establishing the relationship between questionnaires and blood pressure control, both MAQ-Q and Haynes has shown to be statistically significant, while MAQ-Q identified the largest number of individuals with uncontrolled pressure among those who did not adhere to treatment.MAQ-Q had the highest non-adherence rate.Variables such as smoking and patients' return to physician for a follow-up visit had an association with non-adherence to treatment in all three tests.

Table 1 Frequency rate of answers to the questionnaires used in the assessment of adherence to antihypertensive treatment MAQ
-Q -Qualiaids team's Medication Adherence Questionnaire n %1.In the past seven days, didn't/did you take at least one extra tablet of this medication?Haynes' Questionnaire 1.Many people have some kind of difficulty taking their medication.Over the last 30 days, have you had any difficulty taking your blood pressure medication?

Table 2
shows the prevalence rates of adherent/nonadherent patients, according to the tests used and blood pressure control.The MAQ-Q had 60.16% non-adherence, followed by MGT, with 29.28%, and Haynes, with 13.15%.

Table 3 Frequency rate of patients who do not adhere to antihypertensive treatment, according to the methodology used and sociodemographic and clinical characteristics
MAQ-Q -Qualiaids team's Medication Adherence Questionnaire; MW -Minimum Wage; Retiree w/ paid activ.-Retiree with paid activity; Pens.benef.-pension beneficiary a chi-square test; *Fisher's exact test; # 1 minimum wage: BRL 622.00