Non-Targeted Self-Measured Blood Pressure and Hypertension Control in Public and Private Health Systems in Brazil

Abstract Background: It is estimated that more than 30% of the Brazilian population has systemic arterial hypertension (SAH), and mostly as an uncontrolled disease. The most recent Brazilian Guideline of Hypertension recommends the practice of self-measurement of blood pressure (BP) as one of the strategies for a better control of SAH, but there is no consensus about the efficiency of this tool. Objective: To assess the control of SAH and the practice of non-targeted self-measured BP (SMBP) among hypertensive users of the Unified Health System (SUS) and the Supplementary Network (SN). Methods: This is a cross-sectional, observational, analytical study, with a stratified probability sample. One thousand volunteers were investigated, being 500 from SUS and 500 from the SN. Uni and multivariate analyses were performed considering a 5% significance level. Results: Patients from SUS presented inferior sociodemographic data (schooling, social status) in relation to those of the SN (p < 0.001), and showed lower control of SAH (p = 0.014), as well as more visits to the emergency room in the past year due to hypertension (p = 0.002), and fewer regular appointments with the cardiologist (p = 0.004). SMBP was equally present in both assessed groups (p = 0.567), even though users of the SN have been more advised to not conduct such a practice (p = 0.002). SMBP (p < 0.001) was an independent factor for uncontrolled SAH both in SUS (OR = 3.424) and in the SN (OR = 3.474). Conclusion: Patients in SUS presented lower SAH control. The practice of SMBP, mostly practiced with an uncalibrated digital device, was equally present in both groups and became an independent factor of uncontrolled SAH.


Type of study and data collection
This is an observational, cross-sectional and analytical study that included 1,000 patients assisted at the outpatient cardiology clinics of private and public hospitals between June, 2017, and October, 2019 in Aracaju-Sergipe. The sample was defined by convenience and selected in a non-probability consecutive manner, and evaluated 500 individuals in each group.
We included individuals diagnosed with SAH, defined according to the Brazilian Guidelines of Hypertension, 6 and aged above 18 years. We excluded the ones who presented with psychiatric conditions. The information was obtained through a standardized questionnaire that included sociodemographic and clinical data of the patient, information about SMBP and self-medication, besides information in the medical record.
SAH control was assessed with the ambulatory blood pressure monitoring (ABPM) or the mean of the two last As one of the strategies to promote higher adherence to treatment, the most recent Brazilian Guideline of Hypertension recommends the practice of blood pressure (BP) self-measurement, called BPSM. 6 However, even though this methodology is attractive because it is easy to acquire and use the devices, especially digital ones, it presents practical limitations, such as: lack of calibration and unsatisfactory quality of some instruments; non-standardized BP measurement technique; interference of stressful situations, among others. There is no consensus in the literature as to the real benefit of this method in incrementing the adherence to antihypertensive treatment for disease control; besides, there are few studies approaching the impact of this strategy. Therefore, this investigation was conducted aiming at assessing the control of SAH, as well as the influence of non-targeted self-measured blood pressure (SMBP) in chronic hypertension users of the Unified Health System (SUS) and the Health Supplementary Network (SN).

Ethical aspects
All volunteers signed the Informed Consent Form according to resolution 466/2012, allowing the use of their information, as long as the identification data were confidential. This study was approved by the Human Research Ethics Committee, CAAE number:

Statistical Analysis
The collected data were stored and analyzed using SPSS Statistics 22.0. The quantitative variables were described as mean and standard deviation, according to the normal distribution of the sample, and qualitative variables were absolute number and frequency. The Shapiro-Wilk test was used to verify the normality of the sample distribution.
Pearson's chi-square or Fisher Exact tests were used, when adequate, to analyze associations, besides the Student's t-test for independent samples, to verify the difference between groups regarding age. Finally, the univariate and multivariate analyses were conducted, which included logistic regression. We considered admission of a model as p = 0.25, and permanence, p = 0.05. The method of choice was the backward stepwise, which enabled reaching adjusted odds ratio and 95% confidence intervals. For statistical significance, we considered a two-tailed p of 0.05 for all tests.
The variables included in the regression model were: sex, color, age, schooling, social status, marital status, depression, peripheral obstructive arterial disease, diabetes mellitus, dyslipidemia, chronic kidney disease, coronary disease, cerebrovascular disease, chronic heart failure, practice of SMBP and self-medication.

Descriptive analysis
One Thousand volunteers were assessed, being 500 from SUS and 500 from the SN, with mean age of 60.9 ±11.9 years and 61.14 ±13 years, respectively, without differences between both groups (p = 0.618). The total sample was formed by 57.1% of women and 42.9% of men, with no difference between the types of health services (p = 0.085).
There were more black people in SUS and more white people in the SN (p < 0.001), and no differences regarding yellow and brown users. The prevalent social classes in SUS were D and E, whereas in the SN it was A, B and C (p < 0.001). Besides, there was prevalence of higher education in the SN, whereas illiteracy and lower schooling levels were prevalent in SUS (p < 0.001), as demonstrated in Table 1.
As to the comorbidities, the patients in SUS presented with diabetes (p = 0.021) and dyslipidemia more often (p < 0,001) than users in the SN. However, it was not possible to observe discrimination between groups regarding the occurrence of coronary disease, heart failure, peripheral obstructive arterial disease, cerebrovascular disease, chronic kidney disease, chronic obstructive pulmonary disease and depression ( Table 1).

Control of SAH and SMBP in the SN and SUS
As to the behavior of SAH, the patients cared for in SUS had less control of the disease, more visits to the emergency room in the past year due to hypertension, and attended fewer regular appointments with the cardiologist (p < 0.05). Besides, patients in the SN selfmedicated more often (p < 0.001) ( Table 2).
Of the total population, 44.7% did SMBP, mostly using a digital device (93.8%). Besides, 21.1% were aware of the need for the annual calibration of the device, which was performed in only 10.7% of the devices.
There were no differences regarding the practice of SMBP between groups (p = 0.567), which was performed by 43.8% of SUS users and 45.6% of SN users. However, in the SN there were more discussions about the positive and negative aspects of selfmeasurement with an assistant physician, as well as more advice against that practice (p < 0.05). In general, most patients were advised to do SMBP, both in SUS (88%) and in the SN (83.2%).
The practice of SMBP was associated with more symptoms of hypertension and self-medication, both in SUS and in the SN. Among users of the SN, those belonging to higher social classes performed less SMBP. Among volunteers of SUS, the discussion with the physician about SMBP was more common among those who practiced it (Tables 3 and 4).

Variables that are independently associated to the non-control of BP in the groups
In multivariate analyses, SMBP and the female gender were predictors of uncontrolled SAH in patients of SUS and the SN. In SUS, another predictor of uncontrolled SAH was self-medication (Tables 5 and 6).

Discussion
The main findings of this investigation were: a) the practice of SMBP was an independent predictor for the non-control of BP, both in individuals assisted by the public and the private health system; b) there was a recommendation for SMBP for most members in both groups, and the encouragement to not practice it was more prevalent in those belonging to the SN; c) most users are unaware of the need for the annual calibration of the device, and finally, d) SMBP was associated with more self-medication in both assessed groups -Central Figure. The most recently published Brazilian Guidelines of Hypertension (2020) indicates self-measured BP as one of the possible strategies for the control of BP, with level of recommendation I. 6 In fact, randomized and standardized studies, with only one type of specific and well-calibrated device, and skilled volunteers regarding the technique of BP checking and frequency of measurements, back up this methodology for better BP control. 13,14 However, in the daily practice, self-measurement is not advised and is carried out with different types of devices, many of which are not properly calibrated. In our study population, of those who did SMBP, 93.8% used a digital device, and only 21.1% of them were aware of the need for its annual calibration, which was performed in only 10.7% of the devices.
Similar findings were observed in other investigations which were not able to demonstrate the benefits of SMBP for the control of pressure levels. 15,16 We can speculate that these observations would be a result of false BP values generated by an unsatisfactory measurement technique and/or use of uncalibrated devices, 16 which encourage improper behaviors, such as self-medication, especially in anxious individuals. 15 Therefore, it is important that the patient be aware of the technical details of BP measurement, as well as the frequency with which it should be done, besides aspects regarding tensiometers (type, calibration), before recommending SMBP.
Besides, it was observed that black individuals, those with lower schooling levels, in lower social classes were      more prevalent in SUS, which was the group associated with 172% more chances of not controlling SAH. Also, they attended fewer regular appointments with the cardiologist and attended the emergency room more often in the past year due to hypertension. The existence of inequalities regarding both BP control and access to health is a wellestablished reality. [17][18][19][20] Certainly, the investment in more intense actions in primary care may help reduce the overload of more complex sector, for example, reducing the visits to the emergency room. 21,22 The limitations of the study are intrinsic to a crosssectional study, such as the inability of pointing out causes to the outcomes, which would be possible in a cohort study. Besides, other factors not considered in the analysis may impact the assessed events. Finally, the results refer only to patients assisted in cardiology services of private and public hospitals of Aracaju-Sergipe, which limits the generalization of data.

Conclusions
In the assessed sample, patients in SUS, with lower socioeconomic indexes, showed lower SAH control, fewer regular appointments with the cardiologist and more visits to the emergency room due to hypertension.
The non-targeted SMBP, mostly performed with an uncalibrated digital device, was equally present among SUS and SN users, and presented itself as an independent predictor of uncontrolled SAH in both groups, besides being associated with the practice of self-medication.

Potential Conflict of Interest
No potential conflict of interest relevant to this article was reported.

Sources of Funding
There were no external funding sources for this study.

Study Association
This study is not associated with any thesis or dissertation work.

Ethics Approval and Consent to Participate
This study was approved by the Ethics Committee of the CEP da Universidade Federal de Sergipe (UFS) under the protocol number 60473316.9.0000.5546. All the procedures in this study were in accordance with the 1975 Helsinki Declaration, updated in 2013. Informed consent was obtained from all participants included in the study.