Educational intervention performed by nurses for blood pressure control: a systematic review with meta-analysis

Objetivo: to assess the effect of an educational intervention performed by nurses for blood pressure control in people with arterial hypertension, when compared to usual care. Método: a systematic review with meta-analysis of randomized clinical trials, conducted in six databases. The studies included were those in which an educational intervention was performed by nurses on people with arterial hypertension. The risk of bias was assessed by means of the Risk of Bias Tool, the meta-analysis was performed in the Review Manager software and certainty of the evidence was calculated in the Grading of Recommendations Assessment, Development and Evaluation system. Resultados: a total of 1,692 studies were found, which were peer-reviewed, including eight of them in the meta-analysis. The meta-analysis was calculated for the “systolic blood pressure” and diastolic blood pressure” outcomes, in subgroups by time and by intervention performance type. For the in-person educational intervention, performed individually combined with a group activity, the effect estimate was -12.41 mmHg (95% Confidence Interval: from -16.91 to -7.91, p<0,00001) for systolic pressure and -5.40 mmHg (95% Confidence Interval: from -7.98 to -2.82, p<0,00001) for diastolic pressure, with high certainty of evidence. Conclusión: the educational intervention performed by nurses, individually and combined with a group activity, presents a statistically significant clinical effect. PROSPERO registration No.: CRD42021282707.


Introduction
Hypertension is recognized as one of the most important risk factors for all-cause mortality, in addition to being the leading cause of cardiovascular morbidity and mortality and disability worldwide. Thus, it becomes necessary to establish a care plan focused on three dimensions: therapeutic actions involving and not involving medications, educational actions and self-care (1) .
Maintaining the patient's motivation to adhering to the treatment is perhaps one of the most arduous challenges faced by health professionals in relation to the care of people with arterial hypertension, especially in the Primary Health Care context, which is why it is always necessary to recognize the person's will to undergo the treatment and understand their motivations and reasons for non-adherence to the therapy (2)(3) . It is in the Primary Health Care environment that it is possible to develop comprehensive patient care, including health promotion and protection, in addition to being the gateway to the Unified Health System (Sistema Único de Saúde, SUS) and the communication center for the entire Health Care Network to the SUS (4) .
In order to reach the goals that contribute to therapeutic adherence in people with arterial hypertension, and thus lead to improving their quality of health and life, nurses have been seeking support in interventions that collaborate for effective clinical Nursing care, which may contribute to achieving blood pressure control in people with arterial hypertension. Nursing interventions in Primary Health Care seek to enable an improvement in the patient's clinical condition and to improve care promotion in order to contribute to health care and to the prevention of arterial hypertension cases (5) .
Educational interventions are prioritized among the ones that nurses can perform to care for people with arterial hypertension. According to the Nursing Interventions Classification (NIC), the health education intervention aims at developing and providing instructions and learning experiences to facilitate a voluntary behavioral adaptation that promotes the health of individuals, families, groups or communities (6) .
Among the educational strategies to be adopted by nurses, those that are carried out with longer monitoring periods, that perform interactions between small groups and that are based on the care and partnership model deserve attention (7) .
Educational interventions carried out for people with arterial hypertension should aim at reducing the blood pressure levels since, according to data from a meta-analysis performed in 2016, it was verified that a 10 mmHg reduction in systolic blood pressure significantly decreases by 20% the risk of major cardiovascular disease events (relative risk: 0.80, 95% confidence interval: 0.77-0.83) similarly across several population subgroups, regardless of high blood pressure onset (8) .
Therefore, in view of the diversity of educational interventions that can be carried out by nurses on people with arterial hypertension aiming at blood pressure control, it is pertinent to seek scientific evidence on the theme in question, justifying this review. In this perspective, the objective of this study is to assess the effect of an educational intervention performed by nurses for blood pressure control in people with arterial hypertension, when compared to usual care.

Type of study
This is a systematic literature review with metaanalysis written according to the items proposed in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) (9) as well as to the recommendations set forth in the Cochrane Handbook for Systematic Reviews of Interventions in Health (10) . This type of study is a planned process, which summarizes diverse evidence from primary studies, with pre-specified eligibility criteria, to answer a specific research question and provide the best evidence about a given intervention (10) .
The protocol of this review was previously published in the International Prospective Register of Systematic Reviews (PROSPERO) platform.

Eligibility criteria
In order to define the eligibility criteria, the Population, Intervention, Control, Outcome and Study type (PICOS) strategy (10)

Information source
The search was conducted in the following databases:

Search selection
The results from each database were e-imported into the EndNote Basic reference manager, online version, to later remove the duplicates. The titles and abstracts of the publications were examined independently by two researchers. Subsequently, both reviewers dually and independently evaluated the full texts of the potentially eligible studies regarding compliance with the eligibility criteria. Any and all disagreements were solved by means of a discussion between the reviewers or with the participation of a third reviewer.

Data extraction
Data extraction from the studies selected for the final sample was in charge of two independent researchers for a subsequent comparison. There were no divergences in the consensus meeting. A clinical form prepared by the researchers was used, containing the following:

Bias assessment
The Revised Cochrane risk-of-bias tool for randomized trials (RoB 2) was adopted to analyze the risk of bias of the primary studies. The following domains were Rev. Latino-Am. Enfermagem 2023;31:e3930. evaluated for internal validation: bias resulting from the randomization process; bias due to deviations from the intended interventions; bias due to missing result data; bias in measurement of the results; and bias in selection of the reported result. Finally, as a result of the risk assessment, each study was categorized as follows: low risk of bias, uncertain risk of bias or high risk of bias (10) .
Each stage was in charge of two independent reviewers and, in case any disagreement arose, they were solved by means of a discussion between the reviewers or with the participation of a third reviewer. The RoB 2 tool, version 22 of August 2019, was used to synthesize the risk of bias analysis of the studies. Heterogeneity between studies was statistically evaluated using Pearson's chi-square test (χ²) for its significance and the I-square (I²) inconsistency test for its magnitude. In relation to χ², a more conservative significance level (p<0.10) was considered to evidence the presence of heterogeneity. For the I² value, the scale rating according to Cochrane was considered (10) : from 0% to 40%: heterogeneity may not be important; from 30% to 60%: it can represent moderate heterogeneity; from 50% to 90%: it can represent substantial heterogeneity; and from 75% to 100%: considerable heterogeneity.

Synthesis of the results
Regarding the statistical model chosen for the metaanalysis, it was determined by exploring heterogeneity, according to Cochran's Q test and to the I² statistical value. The random effect model was chosen when heterogeneity was between substantial and considerable, and fixed effect when it was moderate or unimportant (11) .
As an effect measure, the continuous variable of the SBP and DBP values was used. For the means difference (MD) calculation a 95% confidence interval (CI) was considered, where an MD value below zero means a favorable outcome for the intervention. The overall effect of the intervention was considered statistically significant when p-values<0.05 were identified.

Certainty of the evidence assessment
To classify certainty of the evidence, the Grading of Recommendations Assessment, Development and Evaluation (GRADE) system from Cochrane (12) was used.
The criteria evaluated were methodological limitations (risk of bias), inconsistency, indirect evidence, imprecision, and publication bias. Certainty of the evidence was classified as follows: high, moderate, low or very low.
In this process, the online GRADE Working Group tool was used, which can be accessed free of charge at http:// www.gradeworkinggroup.org.
Among these publications, the study conducted in Brazil (15) presented a statistically significant difference regarding gender between the intervention and control groups at baseline.
As for the way in which the in-person educational intervention was carried out by the nurses, it was identified that it was both individually, in the office (13,(15)(16)19) or during home visits (14) , and combined with a group activity (17)(18)20) .
In relation to the time between completion of the intervention and post-intervention data collection time, it was identified that five studies (62.5%) carried out data collection immediately after finishing the intervention (13,(15)(16)(19)(20) , two studies (25.0%) had a 1-month time interval (17-18) and one study (12.5%) had a 6-month interval (14) after completing the intervention.
As for the SBP and DBP outcomes, four studies (50.0%) reported how the blood pressure checks were performed (15,(17)(18)(19) , one study (12.5%) reported that it followed the American Heart Association Guidelines (13) , one study (12.5%) asserted following a standard protocol of the institution where patients undergo treatment (20) , and another two studies (25.0%) did not report how the SBP and DBP measurements were performed, only informing the respective values (14, 16) .
Other outcomes were also identified in the studies, In the DBP outcome, only the study conducted in Sweden (16) did not present any reduction in such value when comparing the intervention and control groups after the intervention. However, when the SDP reduction is verified, in each of the study groups it was evidenced that the reduction in the Intervention Group (-9.4 mmHg) was higher than in the Control Group (-7.1 mmHg). Five studies (62.5%) confirmed a statistically significant difference in DBP reduction between the groups (13)(14)(18)(19)(20) .  The studies carried out in Palestine and Turkey (17)(18) were assessed as with low risk of bias; those performed in the United States, Nigeria, Sweden, China and

Risk of bias assessment
Iran (13)(14)16,(19)(20) as with uncertain risk of bias; and the one conducted in Brazil (15) as with high risk of bias.
The study classified as with high risk of bias was due to concerns about the randomization process for not having concealed the allocation sequence and presenting baseline differences between the groups (Domain 1) and referring to the deviation from the intended interventions for not reporting in the study whether there was non-adherence to the intervention that might have affected the outcome and whether an appropriate analysis was used to estimate the adherence effect (Domain 2).

Meta-analysis
The eight RCTs selected for this review were included in the meta-analysis, creating subgroups for the PAS and PAD outcomes, as well as subgroups by intervention longer than 3 months and by way of carrying out the in-person educational intervention. Figure  for the "systolic blood pressure" and "diastolic blood pressure" outcomes. Teresina, PI, Brazil, 2022 Rev. Latino-Am. Enfermagem 2023;31:e3930.
In this meta-analysis, with eight studies, the in-

Certainty of the evidence assessment
Certainty of the evidence was assessed according to the SBP and DBP outcomes for the types of interventions performed by nurses, resorting to GRADE (12) . For the meta-analysis, in the comparison between the in-person educational intervention and usual care to reduce SBP, quality of the evidence was considered low (confidence in the effect estimate is limited) for presenting the risk of bias classified as high in the evaluation of the study carried out in Brazil (15) , in addition to presenting a serious inconsistency, that is, substantial heterogeneity (I 2 =54%). For the DBP outcome, quality of the evidence was considered moderate (moderate confidence in the estimated effect), also due to the high risk of bias rating in the same study.
In the meta-analysis of the comparison of in-person educational intervention performed by nurses, individually and in the office combined with group activity versus usual care, quality of the evidence, both for the "reduction in the SBP value" and "reduction in the DBP value" outcomes and according to GRADE (12) , was classified as high (there is strong confidence that the true effect is close to the estimated one).

Discussion
In this systematic review with meta-analysis, different forms of educational interventions carried out by nurses, in person, that contribute to blood pressure control in people with hypertension were identified.
However and patient, performed in the office (13,(15)(16)19) , individually during the home visit (14) , and individually in the Nursing office combined with group activity meetings carried out by nurses (17)(18)20) . In the cases in which the intervention took place individually in the office, two studies (13,15) started, at baseline, with their samples presenting blood pressure classified as pre-hypertension and, in the other two studies (16,19) , blood pressure was classified as Stage 1

The study conducted in Nigeria
hypertension. As for the time interval between one inperson consultation and another, it varied from 2 to 6 weeks and each consultation lasted a mean of 30-40 minutes in two studies (13,19) , whereas the other two studies did not report this information (15)(16) .
The subgroup of the in-person educational intervention performed by nurses individually, without considering the intervention length in time (13,(15)(16)19) , presented a mean reduction effect estimate of - in the study conducted in China (19) and mean drops of -16.2 mmHg in SBP and -9.4 mmHg in DBP in the study conducted in Sweden (16) .
As for the in-person educational intervention performed by nurses, combining individual meetings with group activity moments, the studies were carried out in Palestine (17) , Turkey (18)  When evaluating quality of the evidence, according to GRADE, this subgroup considering time, the assessment of the evidence was classified as high both for the SBP and DBP outcomes.
The meta-analysis of the subgroup of the in-person educational intervention performed by nurses, individually combined with group activity, lasting at least 3 months, shows a statistically significant reduction in SBP and DBP, as well as, according to a meta-analysis carried out in 2016 (8) , it proves a clinically significant effect, thus confident for the reduction in the risk of major cardiovascular events.
As It is also important to reinforce the need for clinical trial studies, with this intervention and of good methodological quality, to be carried out in Brazil, as this is the only way to be certain about the quality of the intervention in our country. It is also suggested that the cost of the intervention carried out by nurses be evaluated, as it is relevant to consider whether this intervention is financially beneficial or not to this population group. It is believed that implementing interventions with a preventive approach can stand out in relation to the high financial costs resulting from hospitalizations associated with arterial hypertension in the SUS.

Conclusion
It was evidenced that the in-person educational intervention performed by nurses, especially when carried out individually, through the Nursing consultation and combined with group activities, has a considerable clinical impact on reducing the SBP and DBP values in people with arterial hypertension, being extremely useful to controlling the blood pressure levels and, consequently, to reducing comorbidities, hospitalizations and premature deaths due to cardiovascular diseases.
It is expected that the findings of this systematic review with meta-analysis will encourage managers Falcão LM, Guedes MVC, Borges JWP, Silva GRF. and public health services to provide resources so that nurses can carry out the intervention, as well as sensitize nurses to recognize and apply in Primary Health Care the interventions that have shown proven benefits in reducing systolic and diastolic blood pressure in people with high arterial hypertension.