Hypertension (HT) remains the leading cause of death worldwide. Despite this, it remains underdiagnosed and undertreated (1). In Brazil, it is estimated that 35% of the adult population has HT and that about 20% of these have blood pressure (BP) values within the targets recommended for the reduction of cardiovascular (CV) risk (2).

However, when we evaluate the rates of HT control in patients treated by cardiologists and in private services, the values found are better than those described in the databases of patients assisted in the Unified Health System (SUS) and can reach 60.6% (3–6).

Further, it is well established that both diagnosis and assessment of HT control by BP monitoring methods are more accurate than office measurement and should be performed whenever possible (2, 7, 8).

Given this scenario this study aimed to compare BP values and BP control among hypertensive patients followed up in public referral center (PURC) and private referral centers (PRRC), located in two regions of Brazil (Midwest and Northest) under the coordination of the same medical team and following the same treatment protocols established by the Brazilian Guidelines on Hypertension (2).

A total of 2.956 participants were evaluated, of which 1.789 (60.5%) and 1.167 (39.5%) were from the PRRC and PURC groups respectively with no age difference. The frequency of males was higher in the private service (Table 1).

As for nutritional status, only 21.7% of the sample was not overweight with a higher prevalence of obesity in the PURC compared to the PRRC.

Regarding antihypertensive drugs, the use of angiotensin-converting enzyme inhibitors (ACEI) and angiotensin receptor blockers (ARB) was similar in both groups. However, all other classes have a higher frequency of use in patients followed up in the public service. The mean number of antihypertensives used was 2.0 ± 1.3 for the total sample; 1.4 ± 0.7 for PRRC and 2.9 ± 1.5 for PURC (p < 0.001) (Table 1).

All mean blood pressure values, both from office measurements and HBPM in the PURC and PRRC are described in Table 2. The frequency of uncontrolled HT was higher in the PURC both by office measurements and by HBPM (Figure 1).

Regarding the possible phenotypes of uncontrolled BP in treated patients with HT, the frequency of sustained hypertension (SH), when both office BP and HBPM are above the recommended targets, was higher in the SRPU. However, masked hypertension (MH) was more frequent in PRRC (Figure 2).

Furthermore, we evaluated all 904 patients (30.6%) who were using 3 or more classes of antihypertensive drugs in combination. Further, we found this subgroup had a higher prevalence of uncontrolled HT in the PURC (Figure 3).

The sample evaluated consisted of people with HT with an average age of just under 60 years and an overweight average body mass index. We observed a higher prevalence of females in the total sample, with an even higher prevalence in the PURC. This characteristic of higher frequency of medical care by females has already been described in other publications and denotes the lack of attention to men's health concerning HT (10, 11).

Another important aspect observed in the sample was the higher overweight prevalence in the PRRC (43.0%) and obesity in the PURC (42%). Such characteristics may denote a poorer quality diet and an increased association with comorbidities or severity of hypertensive disease in patients in the public service because it is known that increased BMI and abdominal circumference are risk factors for both increased BP and major cardiovascular diseases (12–14). Moreover, socioeconomic and cultural barriers, in general, are associated with the differences in BMI found both in our sample and in other analyses already published (15, 16).

PURC patients used a greater number of antihypertensive medications. Moreover, except for the classes that block the renin-angiotensin-aldosterone system (ACEI or ARB), the other classes were used more frequently in the PURC. As for antihypertensives available for use in the basic health network in our country, most of them have short half-lives and we do not have fixed combinations in a single pill. This reality leads to the need for the use of a greater number of drugs, pills and daily doses. This is known to be associated with lower adherence to treatment and resulting worsening in BP control (2, 5, 17, 18). We should also consider that there was a higher prevalence of obesity in the PURC, a factor associated with pressure levels that are more difficult to control (2, 5, 17, 18).

Despite the greater number of AH used in PURC, the mean SBP and DBP values, whether measured in the office or by HBPM remained significantly higher than in PRRC. The consequence of a higher number of uncontrolled patients is an increase in the incidence of major associated CV outcomes. Effectively improving the control of HT means dramatically reducing the incidence of stroke, acute myocardial infarction, heart failure and chronic kidney disease (19–21).

When we evaluated the possible phenotypes of uncontrolled HT, we found a higher prevalence of SH in the PURC. This phenotype is the one with the worst prognosis relating to cardiovascular outcomes in hypertensive patients since it denotes a lack of BP control in both office measurements and HBPM (22). This aspect reinforces the importance of assessment, whenever possible, through BP monitoring methods. This way, we will have a broader understanding of the BP control status and the best strategies for pharmacological treatment (23). In a previous study of patients with HT treated by specialist physicians and evaluated by HBPM, SH rates were 33.7% (6); in our sample we found values of 50.7% and 26.7% in PURC and PRRC, respectively (p < 0.001).

When evaluating the subgroup using three or more AH drugs, we found higher levels of patients off target by both the office measure and HBPM which was to be expected and similar to the total sample, those followed up in the PURC had worse control rates than the PRRC group.

This study has some limitations since we have not evaluated socio economic level and adhesion with specific tools, however, by evaluating a population of more than 3,000 hypertensive patients using public and private reference centers under the same coordination, we believe it is possible to evaluate with less bias the aspects related to drug use in both scenarios and the impact on BP control. Furthermore, the information obtained both by the office measurement and by the HBPM allows greater accuracy in the evaluation of the recommended goals.

It was possible to observe that patients followed up in public referral centers have a higher prevalence of obesity and use more than double the number of antihypertensive drugs than private services.

The blood pressure control rates assessed by both office measurement and HBPM are always worse in the PURC and on average 15.3 mmHg and 12.1 mmHg higher than in the PRRC for SBP and DBP, respectively, in the office measurements.

The poorer control of HT is associated with a higher incidence of the main cardiovascular outcomes. Therefore, there is an urgent need to reassess strategies so that, in the end, we can decrease the disparities between public and private services and increase our hypertensive population's cardiovascular protection.