Introduction

Spirituality has been currently defined as ‘a construct of moral, mental, and emotional values that guide an individual’s thoughts, behaviors, and attitudes in both intra- and interpersonal relationship, amenable to scientific evaluation through proper methodology’ (1). Although often utilized as a therapeutic tool in clinical medicine, high quality evidence supporting the efficacy of spirituality as an intervention is lacking (2).

Arterial hypertension (HTN) is a major cause of morbidity and mortality worldwide, significantly impacting life expectancy, quality of life, and the global economy (3, 4). Despite the proven benefits and affordability of blood pressure-lowering medications, HTN often remains poorly controlled (5) with adverse consequences for the vessels, heart, brain, and kidneys (6). Additional non-pharmaceutical interventions that improve blood pressure (BP) and outcomes would be desirable (7).

Spirituality-based interventions that stimulate edifying feelings and reduce the harmful ones have shown promise (8, 9, 10) but have undergone only limited evaluation in randomized trials. We performed a randomized trial to evaluate the efficacy of a spirituality intervention for treatment of HTN. Our specific objectives were to determine whether, in patients with mild or moderate HTN, a spirituality-based intervention compared with control reduces BP and improves endothelial function after 12 weeks.

Methods

Study design and participants

The trial methods have been published previously (11) and the trial is registered at ensaiosclinicos.gov.br (RBR-7 m7ct53). Briefly, FEEL was an open randomized controlled trial, evaluating the use of a spirituality-based intervention in stage I and II hypertension. The trial was approved by the institutional review board from the Clinical Hospital from Federal University of Goias (registration number: 5.487.621), and all patients provided written informed consent. Trial was reported according to Consolidated Standards of Reporting Trials (CONSORT) 2010 updated statement (12). The trial was coordinated by and performed at the Hypertension Unit, Federal University of Goiás, Brazil, a HTN reference center, and was scientifically supported by the Spirituality and Cardiovascular Medicine Department of the Brazilian Cardiology Society (DEMCA/SBC).

Eligibility criteria

Patients were eligible if they had been previously diagnosed with hypertension, stage I (Systolic blood pressure (SBP) 140–159 mmHg, and/or diastolic blood pressure (DBP) 90–99 mmHg) or II (SBP 160–179 mmHg and/or DBP 100–109 mmHg), with or without antihypertensive medication. Patients were excluded if office BP was ≥ 180/110 mmHg, any changes had been made to antihypertensive therapy in previous 30 days, or the patient was unable to receive or read messages via WhatsApp.

Randomization

Patients were randomized using a 24-hour, centralized automated, internet-based system (www.randomizer.org) in a 1:1 ratio to receive spirituality-based intervention or control (routine care only).

Data collection

Patients’ assessment

All participants were evaluated in two office visits, being the first (V1) at the beginning of the study and the final (V2) after ± 3 days of the 12-week intervention period. No changes were made to regular medication and lifestyle habits (physical exercise practice, alcohol intake, or smoking) during the inter-visit interval for the patients who completed all the study procedures.

One follow-up phone call was scheduled at 42 days after V1, to ask about symptoms, reinforce the maintenance of lifestyle habits and medication, and, for experimental group, encourage adherence to activities.

History and physical examination

At the first visit (V1), all participants underwent a standardized cardiology consultation, including recording of demographics (age, sex), medical history (diabetes, dyslipidemia, other), lifestyle, including smoking (never, current smoking or former smoking), regular alcohol intake (no or yes, but no quantity was specified), physical activity (no, yes irregular, when <150 minutes/week, yes regular, when >150 minutes/week), medications (dosage, frequency) and religion (Catholic, Protestant, Spiritist, other) and performance of anthropometric measurements (weight, height, and body mass index).

Measurement of blood pressure and flow mediated dilation

Office BP was measured three times, one minute apart with an OMROM HBP1100, according to Brazilian hypertension guidelines (13). The average of the two last measurements was used to determine office BP.

Central BP was measured with a triple shot protocol using a Cardios Arteris@ device in C1 calibration, obtaining peripheral SBP (SBPp) and DBP (DBPp), central SBP (SBPc) and DBP (DBPc), pulse wave velocity (PWV) and augmentation index (AiX) corrected for heart rate. The averages of the three measurements were calculated for these variables (14).

Flow-mediated dilation (FMD) was assessed by only one investigator, to minimize examiner bias (15), with the gold standard device Unex EF38, which uses Doppler ultrasound attached to an articulated robotic arm to measure the diameter of the brachial artery before and after ischemia caused by inflating a cuff 50 mmHg above the individual’s SBP for five minutes (16). Increases of 10% or more in basal diameter of the artery are considered normal. Assessment of FMD was carried out due to its relevance as an outcome for short-term non-pharmacological interventions (17).

All research subjects were instructed on how to perform the Home Blood Pressure Monitoring (HBPM) using an OMRON HEM-9200T device, according to the Brazilian guideline (14). Over the next four days, BP was measured three times, one minute apart, in the morning and evening, always with an empty bladder, and before meals or taking antihypertensive medication.

Intervention procedures

Subjects randomized to the spirituality-based intervention received a daily WhatsApp message interspersed with rest days to facilitate adherence (for schedule, see Table 1). Each message required no more than five minutes to complete. Messages included a short video, a written message for self-reflection, or an activity based on content addressed in the previous video. Messages were aimed at stimulating forgiveness, ‘considered a process of releasing resentment’ (18); gratitude, defined as recognizing and responding with kindness ‘to the roles of other people’s benevolence in the positive experiences and outcomes that one obtains’ (19); optimism, ‘defined as a stable, generalized expectancy for the occurrence of good outcomes in life’ (20); and purpose in life, ‘defined as a self-organizing life aim that stimulates goals, manages behavior, and provides a sense of meaning’ (21).

Table 1

Schedule of messages’ content for the intervention group.

DAYTASKDAYTASKDAYTASKDAYTASK
1Video 122Activity 643Day off64RM 22
2RM 123RM 844Video 965Activity 21
3Activity 124Day off45RM 1566Day off
4Day off25Video 646Activity 1467Video 12
5Video 226RM 947RM 1668RM 23
6RM 227Activity 748Activity 1569Activity 22
7Activity 228RM 1049RM 1770RM 24
8Day off29Activity 850Activity 1671Activity 23
9Video 330Day off51Day off72Day off
10RM 331Video 752Video 1073Video 13
11Activity 332RM 1153RM 1874RM 25
12Day off33Activity 954Activity 1775Activity 24
13Video 434Activity 1055RM 1976RM 26
14RM 435RM 1256Day off77Day off
15Activity 436Activity 1157Video 1178Video 14
16RM 537Activity 1258RM 2079RM 27
17Day off38Day off59Activity 1880RM 28
18Video 539Video 860RM 2181Day off
19RM 640RM 1361Activity 1982Video 15
20Activity 541Activity 1362Day off83RM 29
21RM 742RM 1463Activity 2084Activity 25

RM: reflection message.

Adherence was evaluated by WhatsApp read receipts (two blue ticks) for all messages and, for activities, received answers were also considered.

The videos contained a script written by the research group and narrated in a neutral tone by a professional voice actress and neutral graphic art developed by a professional graphic designer.

Messages for self-reflection were quotes related to the topic, all from known authors, properly referenced.

Activities included answering questions related to the content of the videos or completion of related tasks: for example, preparing and sending a message to someone to be forgiven, or writing down personal life purposes.

Participants received all messages from the same investigator number. They were informed that no conversation could take place between them, and no doubts could be cleared, to avoid investigator bias.

Statistical analysis

Sample size calculation

With 35 individuals per group, the study was designed to have 90% statistical power to detect a difference of 3 mmHg in the mean BP between Intervention and Control groups, assuming a common standard deviation of 3.82 mmHg, using a two-sided unpaired t-test with significance level (alpha) of 5% and allocation ratio 1:1 (22). The sample size calculation was performed using the EZR (version 1.64) (23) statistical software application for R (24). Considering accommodating for a maximum drop-out rate of 30%, the total sample size was increased to 100 individuals (50 per group) (22).

Statistical analysis

Baseline characteristics were reported as counts and percentages for categorical variables and means and standard deviations (SD) for normally distributed variables. Differences between randomized groups at baseline characteristics were assessed using unpaired two-sample t-tests for quantitative variables and using chi-square test or Fisher exact test, when appropriate, for categorical variables (25,26,27).

Intergroup comparisons were assessed using unpaired two-sample t-tests for all quantitative outcomes (28,29,30). These outcomes were defined as absolute differences between 12 weeks (final visit) value minus baseline (initial visit) value. Analysis of covariance (ANCOVA) was also performed while controlling for the effects of gender and BMI (baseline) (28,29,30). The adjusted means (with 95% CI) were reported for ANCOVA models. All assumptions for ANCOVA, including homogeneity of regression slopes, equality of variances and normality of residuals, were checked and met (28,29,30,31,32,33). Normality assumptions for ANCOVA was assessed by visual inspection of histogram plot and Q-Q plot of residuals and application of Shapiro-Wilk (SW) normality test (30,31,32,33). The assumption for homogeneity of regression slopes was assessed by rerunning the ANCOVA models, but this time including covariate-independent variable interactions (30, 33). The assumption of homogeneity of variance for unpaired two-sample t-test and ANCOVA was evaluated with Levene’s test (30, 31).

Intragroup comparisons were assessed by two sample paired t-tests (28,29,30). The assumption that paired differences should be normally distributed was evaluated using plots (Q-Q plot and histogram) and SW test (28, 30,31,32).

Additionally, as a sensitivity analysis, statistical analyses were also performed after conducted missing imputation using a machine learning model-based imputer (simple tree) approach. This requires a model to be created for each input variable that has missing values. The default model is a 1-NN (single nearest neighbor) learner (34, 35). The Machine learning methods are known to be the best to impute missing data to improve accuracy (36).

All hypothesis tests were two-sided and a p-value < 0.05 was considered statistically significant. Statistical analyses were performed using R 4.2.3 (R Foundation, Vienna, Austria) (37), Jamovi v2.5.6 (38) and JASP v0.18.3.0 (39).

Results

From 07/27/2022 through 11/25/2022, 100 patients were enrolled (49 in the control and 51 in the intervention group). The baseline characteristics of participants were relatively well balanced between the groups (Table 2). Mean age was 57.3 (SD 11.5) years, 71 (71%) were women and most prevalent risk factors were sedentary lifestyle (47%), alcohol intake (25%) and current smoking (12%).

Table 2

Baseline characteristics.

INTERVENTION (N = 51)CONTROL (N = 49)p-VALUE
Gender, n (%)0.0111
Male9 (17.6%)20 (40.8%)
Female42 (82.4%)29 (59.2%)
Diabetes, n (%)0.7281
No37 (72.5%)34 (69.4%)
Yes14 (27.5%)15 (30.6%)
Physical activities, n (%)0.4271
No23 (45.1%)24 (49.0%)
Yes—irregular16 (31.4%)10 (20.4%)
Yes—regular12 (23.5%)15 (30.6%)
Marital status, n (%)0.1932
Single10 (19.6%)11 (22.4%)
Married28 (54.9%)18 (36.7%)
Living together6 (11.8%)4 (8.2%)
Divorced4 (7.8%)10 (20.4%)
Widow3 (5.9%)6 (12.2%)
Smoking, n (%)0.7941
Never39 (76.5%)35 (71.4%)
Former6 (11.8%)8 (16.3%)
Current6 (11.8%)6 (12.2%)
Religion, n (%)0.8852
Protestant22 (43.1%)21 (42.9%)
Catholic20 (39.2%)22 (44.9%)
Spiritist2 (3.9%)1 (2.0%)
Others7 (13.7%)5 (10.2%)
Alcohol intake, n (%)0.2991
No36 (70.6%)39 (79.6%)
Yes15 (29.4%)10 (20.4%)
Age0.2043
Mean (SD)55.8 (10.6)58.8 (12.3)
BMI0.0433
Mean (SD)32.0 (5.9)29.7 (5.4)
Office SBP0.4703
Mean (SD)130.1 (15.3)127.9 (15.1)
Office DBP0.4183
Mean (SD)79.6 (10.9)77.9 (8.8)
Central SBP0.2203
Mean (SD)117.5 (12.9)114.4 (12.6)
Central DBP0.9603
Mean (SD)81.8 (10.6)81.9 (9.5)
HBPM SBP0.7653
Mean (SD)122.5 (10.8)123.3 (12.6)
HBPM DBP0.7653
Mean (SD)80.6 (9.6)80.1 (7.3)
PWV0.3023
Mean (SD)8.0 (1.5)8.3 (1.7)
FMD0.7733
Mean (SD)9.8 (5.1)10.2 (6.1)

1P-value from Chi-Square test;2P-value from Fisher Exact test;3P-value from unpaired two-samples t-test with equal variance.

BMI—body mass index, SBP—systolic blood pressure, DBP—diastolic blood pressure, HBPM—home blood pressure monitoring, PWV—pulse wave velocity, FMD—flow mediated dilation.

Of the 100 patients randomized, 75 participants completed both baseline and end of study BP and FMD measurements and contributed to the primary analyses (Figure 1), accordingly to the previous sample size calculation. As the timeframe for the final visit was strict (±3 days after the 12-week intervention period), this can explain part of the high dropout rate. This and all other reasons are detailed in Figure 1.

Figure 1 

Flow diagram of randomized patients, beginning to end of the study.

The main results are presented in Tables 3 and 4. Intervention group had significant improvement in the office SBP, central SBP and FMD, while control group had a worsening of office DBP and FMD. When these differences were compared between the groups, office SBP and FMD showed statistically significant benefit in the spirituality-based intervention group and, when adjusted for gender and BMI, office DBP, central SBP and FMD were statistically improved in the spirituality-based intervention group by the end of the follow-up period. Missing data imputation strengthened and improved previous results, demonstrating statistically significant benefits for office and central SBP, office DBP and FMD (Tables 5 and 6).

Table 3

Study outcomes: within group comparisons between first and final visits.

OUTCOMESPIRITUALITY-BASED INTERVENTION GROUP (G1) n = 39CONTROL GROUP (G2) n = 36
MEANSDLOWER 95%CIUPPER 95%CIP-VALUEMEANSDLOWER 95%CIUPPER 95%CIP-VALUE
Office SBP (V2)121.3713.09126.6315.55
Office SBP (V1)128.9715.60127.1815.30
Office SBP (V2-V1)–7.60–11.72–3.48< 0.001–0.56–6.405.290.848
Office DBP (V2)80.5810.8381.658.39
Office DBP (V1)79.3910.0877.289.08
Office DBP (V2-V1)1.19–1.724.100.4124.381.846.920.001
Central SBP (V2)113.4211.79115.1213.97
Central SBP (V1)117.5212.84113.7412.33
Central SBP (V2-V1)–4.11–7.87–0.340.0341.38–3.666.410.582
Central DBP (V2)81.1510.7980.3810.37
Central DBP (V1)81.8210.1981.539.47
Central DBP (V2-V1)–0.67–3.171.820.588–1.14–4.722.430.520
HBPM SBP (V2)121.6910.27122.3112.87
HBPM SBP (V1)123.1310.34121.9212.51
HBPM SBP (V2-V1)–1.44–4.021.150.2680.39–1.932.700.735
HBPM DBP (V2)80.5610.1679.816.38
HBPM DBP (V1)80.879.2579.397.94
HBPM DBP (V2-V1)–0.31–1.861.250.6910.42–1.372.200.638
PVW (V2)7.771.248.421.61
PWV (V1)7.821.378.281.69
PVW (V2-V1)–0.05–0.220.130.5840.14–0.070.350.179
FMD (V2)14.286.677.235.71
FMD (V1)10.165.2410.576.31
FMD (V2-V1)4.121.416.820.004–3.34–6.36–0.330.031

SD – standard deviation, V1 – initial visit, V2 – final visit, SBP - systolic blood pressure, DBP - diastolic blood pressure, HBPM – home blood pressure measurement, PWV – pulse wave velocity, FMD – flow mediated dilation.

Table 4

Study outcomes: between groups comparison.

OUTCOMEGROUPnUNADJUSTED ANALYSIS VIA t TESTADJUSTED ANALYSIS VIA ANCOVA
MEANSD95% CIp-VALUE1 MEAN95% CIp-VALUE2
Office SBP (V2 – V1)Control36–0.5617.26(–6.4; 5.29)0.047–0.38(–5.56; 4.81)0.063
Intervention39–7.6012.71(–11.7; –3.48)–7.37(–12.96; –1.79)
Office DBP (V2 – V1)Control364.387.51(1.83; 6.91)0.1024.49(1.72; 7.27)0.031
Intervention391.198.98(–1.72; 4.1)0.14(–2.85; 3.13)
Central SBP (V2 – V1)Control361.3814.88(–3.65; 6.41)0.0781.77(–2.71; 6.25)0.038
Intervention39–4.1011.62(–7.87; –0.34)–5.00(–9.82; –0.17)
Central DBP (V2 – V1)Control36–1.1510.56(–4.7; 2.43)0.824–1.06(–4.13; 2.00)0.684
Intervention39–0.687.71(–3.17; 1.82)–1.96(–5.26; 1.34)
HBPM SBP (V2 – V1)Control360.396.84(–1.93; 2.70)0.2930.68(–1.84; 3.20)0.579
Intervention39–1.447.98(–4.02; 1.15)–0.32(–3.04; 2.39)
HBPM DBP (V2 – V1)Control360.425.27(–1.37; 2.2)0.5350.52(–1.20; 2.25)0.722
Intervention39–0.314.80(–1.86; 1.25)0.09(–1.77; 1.94)
PWV (V2 – V1)Control360.150.62(–0.06; 0.36)0.1550.17(–0.03; 0.37)0.134
Intervention39–0.050.55(–0.22; 0.13)–0.04(–0.25; 0.17)
FMD (V2 – V1)Control36–3.348.90(–6.36; –0.33)<0.001–3.01(–5.95; –0.07)<0.001
Intervention394.128.35(1.41; 6.82)4.94(1.77; 8.11)

1p-value from unpaired two-samples t-test with equal variance; 2p-value from ANCOVA models adjusted for gender and BMI (baseline).

SD – standard deviation, V1 – initial visit, V2 – final visit, SBP - systolic blood pressure, DBP - diastolic blood pressure, HBPM – home blood pressure measurement, PWV – pulse wave velocity, FMD – flow mediated dilation.

Table 5

Study outcomes: within group comparisons between first and final visits (missing imputation analyses).

OUTCOMESPIRITUALITY-BASED INTERVENTION GROUP (G1) n = 51CONTROL GROUP (G2) n = 49
MEANSDLOWER 95%CIUPPER 95%CIP-VALUEMEANSDLOWER 95%CIUPPER 95%CIP-VALUE
Office SBP (V2)121.9511.67125.0513.80
Office SBP (V1)130.1415.29127.9315.15
Office SBP (V2-V1)–8.19–11.85–4.52< 0.001–2.88–7.832.070.248
Office DBP (V2)81.0910.0282.678.40
Office DBP (V1)79.5710.9277.958.84
Office DBP (V2-V1)1.52–1.014.060.2344.722.746.70<0.001
Central SBP (V2)114.1110.56114.8012.15
Central SBP (V1)117.5412.91114.4012.59
Central SBP (V2-V1)–3.43–6.880.010.0510.40–3.714.520.845
Central DBP (V2)80.869.4980.469.02
Central DBP (V1)81.8010.6081.919.50
Central DBP (V2-V1)–0.95–3.321.420.426–1.44–4.351.460.324
HBPM SBP (V2)121.7910.63123.2412.44
HBPM SBP (V1)122.6010.73123.2812.46
HBPM SBP (V2-V1)–0.8–2.941.340.455–0.04–1.961.880.965
HBPM DBP (V2)80.519.4479.905.77
HBPM DBP (V1)80.549.5080.157.28
HBPM DBP (V2-V1)–0.03–1.421.370.968–0.25–1.701.210.735
PVW (V2)7.951.338.451.60
PWV (V1)7.971.478.301.72
PVW (V2-V1)–0.02–0.170.130.7690.15–0.030.320.103
FMD (V2)14.086.227.105.01
FMD (V1)9.835.1310.156.12
FMD (V2-V1)4.251.956.56<0.001–3.05–5.46–0.650.014

SD – standard deviation, V1 – initial visit, V2 – final visit, SBP - systolic blood pressure, DBP - diastolic blood pressure, HBPM – home blood pressure measurement, PWV – pulse wave velocity, FMD – flow mediated dilation.

Table 6

Study outcomes: between groups comparison (missing imputation analyses).

OUTCOMEGROUPNUNADJUSTED ANALYSIS VIA t TESTADJUSTED ANALYSIS VIA ANCOVA
MEANSD95% CIp-VALUE1 MEAN95% CIp-VALUE2
Office SBP (V2 – V1)Control49–2.8817.24(–7.83; 2.07)0.085–2.457(–6.858; 1.944)0.045
Intervention51–8.1913.03(–11.8; –4.52)–8.959(–13.775; –4.143)
Office DBP (V2 – V1)Control494.726.91(2.735; 6.70)0.0504.869(2.549; 7.19)0.019
Intervention511.529.03(–1.01; 4.06)0.828(–1.712; 3.367)
Central SBP (V2 – V1)Control490.4014.33(–3.71; 4.52)0.1531.09(–2.645; 4.826)0.042
Intervention51–3.4312.25(–6.88; 0.01)–4.519(–8.607; –0.43)
Central DBP (V2 – V1)Control49–1.4410.11(–4.35; 1.46)0.791–1.114(–3.758; 1.53)0.736
Intervention51–0.958.43(–3.32; 1.42)–1.766(–4.66; 1.128)
HBPM SBP (V2 – V1)Control49–0.046.69(–1.96; 1.88)0.5980.273(–1.794; 2.339)0.847
Intervention51–0.807.60(–2.94; 1.34)–0.019(–2.28; 2.243)
HBPM DBP (V2 – V1)Control49–0.255.07(–1.7; 1.21)0.828–0.035(–1.483; 1.413)0.634
Intervention51–0.034.96(–1.42; 1.37)0.469(–1.116; 2.054)
PWV (V2 – V1)Control490.150.61(–0.03; 0.32)0.1460.162(–0.002; 0.327)0.08
Intervention51–0.020.53(–0.17; 0.13)–0.05(–0.229; 0.13)
FMD (V2 – V1)Control49–3.058.37(–5.46; –0.65)<0.001–2.738(–5.137; –0.338)<0.001
Intervention514.258.18(1.95; 6.55)4.587(1.961; 7.214)

1p-value from unpaired two-samples t-test with equal variance; 2p-value from Ancova models adjusted for gender and BMI (baseline).

SD – standard deviation, V1 – initial visit, V2 – final visit, SBP - systolic blood pressure, DBP - diastolic blood pressure, HBPM – home blood pressure measurement, PWV – pulse wave velocity, FMD – flow mediated dilation.

One study participant reported anger as the only harm resulted from the intervention.

Discussion

This is a pilot study, at the forefront of research on this topic. Larger, multicentric studies are already planned. We performed an innovative randomized controlled trial to evaluate the effects of a readily applicable, low cost, spirituality-based intervention on BP and endothelial function in patients with mild or moderate hypertension. The main findings were that the spirituality-based intervention improved SBP control and FMD.

Most previous studies of the possible benefits of spirituality on clinical outcomes are observational. Small, randomized trials have evaluated the effects of meditation (40), psychotherapy (41), and religious interventions (42). Most of them showed a protective effect on human health.

The exact physiological pathway through which the benefits related to the practice of edifying feelings occur is unknown and is likely to be multifaceted. There are hypotheses related to improvements in lifestyle habits, such as reducing smoking, alcohol consumption, drug addiction, or better adherence to healthy diets, exercise practices, or medication treatment (43). None of these hypotheses apply to the group studied here, as all these factors were evaluated by questionnaire at the beginning and end, ensuring that they were not modified during the inter-consultation interval. Another empirical and generical hypothesis encompasses immunological, inflammatory, endocrine, and autonomic mechanisms in the possible cardiovascular modulation (44).

The spirituality intervention that we evaluated focused on four edifying feelings: forgiveness, gratitude, optimism, and life purpose (45, 46, 47, 48). Individuals with a greater capacity for forgiveness, whether innate or acquired through training or therapy have shown benefits in lower left ventricular workload and myocardial oxygen consumption (49), and ultimately lower cardiovascular stress and cardioprotection (50). Gratitude has been related to reduced mental stress, lower cardiovascular risk (51). Optimism has been associated with a lower risk of cardiovascular disease (CVD), lower cardiovascular and all-cause mortality (52). Life purpose or a meaningful life has been linked with lower risk of cardiovascular events and all-cause mortality (53), improve coping and reduce stress in people with CVD (54).

In relation to HTN, multiple mechanisms have been proposed to explain the possible beneficial effects of edifying feelings, mostly related to lower stress levels. Gratitude seemed to work on lowering SBP through a stress-buffering effect on both reactions to and recovery from acute psychological stress (55), and optimism seemed to achieve lower BP by reducing stress intensity and frequency, together with better sleep quality (56). Furthermore, forgiveness, achieved through specific training, seemed to have taken similar paths in HTN treatment, especially in those with elevated levels of anger, probably due to deactivation of sympathetic system, that was previously activated by anger (57). Purpose in life was able to improve BP control, probably because of better sleep, mood, and increase in physical activity practice (58).

The baseline characteristics indicate a population of hypertensive patients with other associated risk factors and mostly adequate BP control. This profile presents a challenge in demonstrating the effectiveness of a potential new treatment strategy to lower blood pressure, especially in the short term. To help detect an effect of our spirituality-based intervention on BP we performed multiple measurements of BP and also measured FMD, which can detect subtle changes in vascular reactivity following short term interventions, mediated by endothelial nitric oxide production (59, 60).

It has been previously reported that each percentage point increase in FMD corresponds to a reduction of about 10% in cardiovascular risk (17). The increase of 4.1 percentage points in FMD in the spirituality-based intervention group was greater than that reported with physical activity (61) or increased intake of phytosterols (62).

Limitations to be considered in the current study include a modest sample size, and that 25% did not undergo both baseline and 12-week measurement of BP and FMD, although this was balanced between the groups. This considerable loss happened due to several reasons, such as medications change during follow-up, not being able to return within the preestablished timeframe (±3 days after the 12-weeks intervention period), and for experimental group, not completing all the proposed tasks. Also, this was a single-center study restricted to patients with mild or moderate hypertension. Furthermore, due to the intervention design, total blinding was not possible. Nonetheless, BP was measured with electronic device, and the obtained number was copied directly to the data form, and we adopted a strategy for FMD-examiner blinding. Missing data imputation reinforced the results previously obtained—the intervention group obtained improvements on office and central SBP, plus FMD.

Conclusion

Spirituality-based interventions that promote edifying feelings can improve both office systolic blood pressure and endothelial function. These results suggest that spirituality-based interventions may be a useful adjunct to standard approaches for adult patients with mild or moderate hypertension, and that these topics could be addressed by physicians with their patients, with potential benefits for cardiovascular health.

Data Accessibility Statement

The data that support the findings of this study are available, upon request, at https://osf.io/adq5b/?view_only=06326e85b49e40a4a08494ae004a9206, reference number osf.io/adq5b.