The effects of a digital lifestyle intervention in patients with hypertension: Results of a pilot randomized controlled trial

Abstract

In this pilot study, the authors investigated the preliminary effectiveness of the digital lifestyle intervention, actensio (mementor DE GmbH), in treating arterial hypertension. Adults with arterial hypertension were randomly assigned to an intervention group (actensio + standard care) or a control group (waiting list + standard care) in a 1:1 ratio. Primary and secondary endpoints were assessed at baseline (t0) and 3 months post‐randomization (t1). The primary endpoint was average systolic blood pressure, measured at home for 1 week. Secondary endpoints included patient engagement (measured using the “patient activation measure”; PAM‐13), average diastolic blood pressure, and heart rate. All endpoints were analyzed using ANCOVA models, following an intention‐to‐treat approach, while adjusting for baseline values. Missing data were estimated using multiple imputation models. A total of N = 102 participants (f = 59, age = 52.94 ± 9.01) were randomized to either the intervention (IG; N = 52) or the control group (CG; N = 50), of which N = 80 completed the blood pressure diary, and N = 81 the PAM‐13 at t1. Between‐group comparisons showed an average group difference in systolic blood pressure of −5.06 mm Hg (95% CI = −8.71 to −1.41, p = .013) between the intervention group (M = 137.37 ± 10.13) and the control group (M = 142.35 ± 11.23). Average group difference for patient engagement was 3.35 points with a trend towards statistical significance (95% CI = −018 to 6.89, p = .064), favoring the intervention group (M_IG = 79.38 ± 9.44 vs. M_CG = 75.45 ± 10.62). There were no group differences in diastolic blood pressure (−1.78 mm Hg; 95% CI = −4.50 to 0.95, p = .402) and heart rate (−0.684; 95% CI = −3.73 to 2.36, p = 0.683). The results of the present pilot study confirm the preliminary effectiveness of the digital lifestyle intervention, actensio, in reducing high blood pressure in patients with hypertension.

1. INTRODUCTION

Cardiovascular diseases are the leading cause of death worldwide, contributing to over 4 million deaths (45% of all deaths) in Europe each year. In Germany, the mortality rate due to cardiovascular diseases is 477.2/100 000 for men and 362.1/100 000 for women, with a 12‐month prevalence estimated at 11.6% for men and 14.1% for women (as of 2013). ¹ The primary risk factor for cardiovascular diseases is hypertension, ² affecting approximately one in three adults in Germany aged 18−79 years (3‐year prevalence [2008–11], men = 33.4%, women = 29.9%). ³

Since high blood pressure is treatable, a target systolic blood pressure of 110–115 mm Hg has been identified as the most significant avoidable risk factor for cardiovascular death worldwide (i.e., 110–115 mm Hg is below the risk threshold). ⁴ In addition to the intake of blood pressure‐lowering medications, lifestyle changes, including modifications in diet, exercise, and stress management, ⁵ are recommended for the treatment and prevention of hypertension ⁶ , ⁷ , ⁸ , ⁹ and can reduce blood pressure and the risk of cardiovascular disease. ¹⁰

Meta‐analyses ¹¹ , ¹² and systematic reviews ⁵ , ¹³ investigating the effectiveness of lifestyle interventions have shown promising results in reducing high blood pressure. However, patients often struggle to maintain these lifestyle changes over an extended period without support. ¹⁴ Providing such aid is currently a challenge for healthcare systems ¹⁵ and has led to the development of digital applications and electronic health products (eHealth), ¹⁶ which are particularly recommended for managing chronic diseases. ¹⁷ , ¹⁸ Some applications have already demonstrated positive results in reducing high blood pressure. ¹² , ¹⁹ , ²⁰ For example, a meta‐analysis by Li and coworkers (2020) across 12 studies showed that participants assigned to mobile health interventions reported a higher reduction in systolic (−3,78 mm Hg; 95% CI −4,67 to −2,89) and diastolic blood pressure (−1,57 mm Hg; 95% CI −2,28 to −0,86), compared to control groups. ¹² Additional subgroup analysis revealed consistent results across various reminder functions, frequencies, patterns of interaction and study durations. To expand the evidence base for digital health solutions in treating hypertension and the feasibility of a digital lifestyle interventions solution in Germany, we designed the present pilot study. The goal of this study was to test the preliminary effectiveness of a digital lifestyle intervention named “actensio” (mementor DE GmbH, Leipzig, Germany), in Germany. Our primary aim was to test whether systolic blood pressure is reduced after 3 months of using actensio compared to a control group. Secondary aims consisted of testing whether patient engagement is increased; and diastolic blood pressure and heart rate is reduced after 3 months of using actensio compared to a control group. We also investigated the exploratory outcomes body mass index (BMI) and medication adherence for between‐group differences.

2. METHODS

2.1. Study design

This study was a pilot randomized‐controlled, parallel, open‐label trail. Upon positive screening for inclusion/exclusion criteria, interested participants were invited for the baseline assessment, after which they were randomized 1:1 to either the intervention‐ (access to actensio and standard care) or the control group (access to standard care only). Three months post‐randomization, all participants were invited for post‐treatment assessment. The study was conducted fully remote (online and telephone) and prospectively registered with the German study register (www.drks.de; DRKS00030109). The study was conducted in accordance with the Declaration of Helsinki and approval was granted by the local ethics committee of the medical faculty at the University Duisburg–Essen (22‐10553‐BO). The study protocol can be requested from the corresponding author.

After registration, two changes were made to the register of the study. (1) The patient activation measure was added after recruitment commenced but before the first randomization as secondary outcome because it was previously forgotten. However, this was a pre‐specified secondary outcome. (2) The maximum age criterion for eligibility (70 years) was removed half‐way through recruitment, as it was deemed inappropriate (no safety concerns).

2.2. Participants

Potential participants were recruited through flyers in the Ruhrlandklinik—Universitätsmedizin in Essen (Germany) and online (via social media), which informed them about the study purpose and the requirements for taking part in the study. Those showing interest could contact the study team or access the online screening questionnaire directly through a study link or QR code. The online screening required all potential participants to first read the study information sheet and give their consent to be screened for the study. Upon giving consent, eligibility for study participation was assessed through various questionnaires (approximately 15 min), asking about age, hypertension diagnosis, blood pressure, BMI, access to internet, motivation for undergoing a lifestyle intervention, intake of prescribed blood pressure medication within the last 3 months, pregnancy, comorbid conditions, cardiovascular events and weather other household members are already participating in the study. Eligible participants were asked to enter their personal data at the end of the online screening and sign up for a subsequent telephone screening with the study team using Calenso (CalensoAG, Rothenburg, Switzerland), an online calendar and booking tool. During the telephone interview (approximately 15 min), the study and all involved procedures were explained once more, and participants had the opportunity to ask questions. Furthermore, the study team confirmed the diagnosis for hypertension (self‐reported) and the availability of a suitable blood pressure monitor (upper arm cuff) required for participating in the study. The type of blood pressure monitor was noted and compared to the recommendations of valid monitors by the Deutsche Hochdruckliga (DHL). Patients without a certified device were advised to get one before the start of the 1‐week blood pressure diary. After the telephone interview, all potential participants received a link to provide consent for their study participation via DocuSign (DocuSign Germany GmbH, Munich, Germany). Upon giving consent, participants received a link for their online baseline assessment (online questionnaires), followed by a 1‐week blood pressure diary via SoSci Survey (SoSci Survey GmbH, Munich, Germany). Links for the blood pressure diaries were sent twice daily to the participants personal e‐mail address. Once the blood pressure diary was completed (at least 3 days with valid entries), the study personnel checked whether the average systolic blood pressure of all available entries met the inclusion criteria. If the blood pressure criterium was not met, the participant could not be randomized and was excluded from the study. Similarly, participants were excluded if less than three valid daily entries were made. Table 1 summarizes all inclusion/exclusion criteria. All participants were reimbursed for their time and effort upon completion of their post‐treatment assessment (20€ voucher).

TABLE 1.

Inclusion/exclusion criteria.

Inclusion criteria	Exclusion criteria
Consent to study participation	Resistant hypertension (≥4 types of antihypertensive agents or systolic blood pressure ≥175 mm Hg according to the baseline blood pressure diary)
Age ≥18 years	Safety reasons (pregnancy, insulin‐dependent diabetes)
Meeting criteria for hypertension (self‐reported confirmation of diagnosis and blood pressure ≥135/90 mm Hg according to the baseline blood pressure diary) *	Cardiovascular event in the past 6 months (stroke, heart attack, cerebral hemorrhage)
Access to internet	Other person in the same household participating in the study (contamination)
Able to implement a lifestyle intervention and complete study procedures	Diagnosis for cancer or acute treatment with immunosuppressants or chemotherapy
Motivated to implement a lifestyle intervention	Planned changes in health‐ or medication plan in the upcoming 3 months
Consistent (no change in the last 3 months) or no blood pressure medication (prescribed)	Planned hospitalization in the upcoming 3 months
Able to measure weight regularly (availability of a scale)	Life expectancy < 12 months
Possession of an upper‐cuff blood pressure monitor

^*A cut‐off of ≥135 mm Hg was chosen to reflect the cut‐off for hypertension for home‐based measurements according to the Nationale Versorgungsleitlinie (NVL). ²¹

2.3. Intervention

Participants in the intervention group were given access to actensio, a CE marked digital application that offers various modules for controlling high blood pressure and promoting a healthy lifestyle via app or web‐browser. These modules include information on physical activity, nutrition, medication adherence, blood pressure monitoring, sleep, stress management, and mindfulness. The content is based on scientific findings and international recommendations for lowering and self‐managing high blood pressure, ⁹ tailored to user's data input. Additionally, there are regular feedback modules to reflect on goals and progress, as well as a follow‐up module to motivate the long‐term implementation of lifestyle changes. The content of each module lasts between 5 and 25 min and is delivered by an animated avatar named “Albert”. The educational modules are unlocked in a consecutive order. To unlock a feedback module, there must be at least three blood pressure diary entries and at least 3 days since the start of the application (for the first feedback module), or since the previous feedback module. Upon completion of all educational modules, the follow‐up module can be repeated as many times as desired until the end of access, provided that seven new diary entries have been added. Thus, there is no program‐defined end to the intervention.

Participants in the control group (care as usual; CAU) are also given access to actensio after completing the post‐assessment. Neither group received restrictions on usual treatment for high blood pressure (CAU). In Germany, this typically refers to medication treatment depending on the medical condition and lifestyle counselling by the attending physician. ²¹

2.4. Outcome measures

2.4.1. Blood pressure measurements

At baseline and 3 months post‐randomization, blood pressure (systolic, diastolic, and heart rate) was measured using a 1‐week blood pressure diary (online, SoSci Survey). All participants were asked to measure their blood pressure according to a standard measurement protocol (based on the self‐measurement recommendations by the “Deutsche Hochdruckliga”; DHL) twice in the morning and twice in the evening using their own blood pressure monitor.

The standard measurement protocol required measuring blood pressure after 3 min of relaxation and avoiding physical and mental stress in the previous 30 min. Participants were instructed to perform the measurement twice in a row, 1 min apart, and then enter the second value in the diary. The upper arm cuff should be positioned at heart level.

2.4.2. Patient activation

Patient activation was measured using the PAM‐13 at baseline and 3 months post‐randomization. The PAM‐13 is a reliable and valid scale for assessing patients' self‐reported confidence, abilities, and knowledge in managing their health or chronic conditions. ²² The PAM‐13 consists of 13 items with statements about self‐management, each to be answered on a scale from one (completely disagree) to four (completely agree). The evaluation is done by summing the raw scores (range 13−52). To standardize the raw total score, a transformation to a 0−100 scale is recommended (100 * (Sum − 13) / (52 − 13)), where 100 represents the highest activation level. ²³ The PAM‐13 is often used in clinical studies on hypertension and is sensitive to changes through behavioral therapy. ²⁴

2.4.3. Body mass index

The BMI was calculated based on self‐reported height and weight at baseline and post‐treatment, using the following formula: Weight in kg / (Height in m)².

2.4.4. Medication and medication adherence

At baseline, all participants were asked about their current intake of hypertension medication (“Are you currently taking medication to lower blood pressure?” yes/no), their names and doses (up to three agents). At post‐treatment, all participants were asked whether this medication has changed in the past 3 months (“Has your medication use changed within the last 3 months?” yes/no). If the medication had changed, participants were asked to describe the change in medication (name and dose).

Medication adherence was measured using the validated Morisky medication adherence scale with eight items (MMAS‐8). ²⁵ The eight items of MMAS‐8 were developed to identify patients who do not take their medications and to determine the reasons for such behaviour, such as forgetfulness, inadequate knowledge, inconvenience, and side effects. The MMAS‐8 score ranges from 0 to 8, with a score of <6 indicating low adherence, 6 to <8 indicating moderate adherence, and 8 indicating high adherence. Only participants who indicated medication intake were asked about medication adherence.

2.4.5. Side effects

Participants were asked about possible medically relevant side effects (adverse events) at post‐treatment assessment which have occurred in the past 3 months (“In the past 3 months, did you experience a medically relevant event that required medical treatment?” yes/no) and whether the event was related to the intervention (“Was the event you reported related to the use of the actensio app?” yes/no; intervention group only).

2.4.6. User satisfaction

After 3 months, participants in the intervention group were asked about user satisfaction with actensio using a set of five statements which they were asked to rate on a scale from 1 to 5 (1 = don't agree, 5 = agree completely). The statements were: (1) Handling: “The technical operation of actensio is easy”, (2) Content: “The modules and content of actensio are good and helpful”, (3) Design: “I like the design of actensio”, (4) User friendliness: “actensio is easy to understand and user‐friendly”, and (5) Recommendation: “I would recommend actensio to others”.

2.5. Randomization and masking

After completing the baseline questionnaire and blood pressure diary, all eligible participants were randomized in a 1:1 ratio to actensio + CAU or CAU using a randomization list (created by the randomization program sealedenvelope.com; Clerkenwell Workshops, London, UK) with stratification based on blood pressure levels (systolic blood pressure ≥160 mm Hg vs. ≤160 mm Hg) and variable block sizes (4–8) to ensure an even distribution of participants. The randomization list was managed by a person from the study team uninvolved in any study procedures (no contact with participants), while enrolment and assignment were managed by the study team. Participants were aware of their group assignment (open‐label). The patient information explained the two different groups but did not reveal the study hypotheses.

2.6. Statistical analyses

In the planning of pilot studies, an exact estimation of the sample size is not required. ²⁶ Therefore, the sample size of the present pilot study was not determined based on the primary endpoint and does not rely on typical parameter assumptions (probability of false‐positive results (α) and probability of false‐negative results (β)). A priori, a literature search was conducted for comparable pilot studies. In the course of this, two comparable pilot studies were identified, which had planned sample sizes of N = 60 ²⁷ and N = 140. ²⁸ This resulted in our recruitment target of N = 50 participants per group.

Consistent with CONSORT guidelines on the reporting of RCTs, ²⁹ all data from all randomized participants were analyzed using the intention to treat principle. Missing data were examined for systematic patterns according to the standards of biomedical research recommendations and, upon confirmation of the “missing at random” (MAR) condition, were reconstructed using multiple imputation using the linear regression model. ³⁰ , ³¹ , ³² Pooled data from five imputed datasets were used following standard procedures. ³⁰ , ³³ Descriptive statistics were presented using unadjusted means (M) and standard deviations (SD) for continuous outcomes and frequencies for binary outcomes. To allow the reader to judge, 95% confidence intervals (95% CI), F‐, p‐values and effect sizes (ES) were reported. Effect sizes between the groups were calculated by dividing the adjusted mean difference by the pooled standard deviation of the two groups. ³⁴

After testing for normality and homogeneity of variances, our primary analysis compared group differences in systolic blood pressure after 3 months between the two treatment groups using an ANCOVA model, adjusting for baseline systolic blood pressure (mm Hg). This corresponds to the recommended approach for parallel, two‐arm RCTs with only one post‐measurement time point. ³⁵ , ³⁶ Analyses of secondary and exploratory continuous endpoints followed the same procedure (diastolic blood pressure, heart rate, PAM‐13 score, BMI and MMAS), adjusting for the respective baseline variable. Results on side effects and user satisfaction are reported descriptively.

3. RESULTS

3.1. Participant characteristics

A total of N = 4244 participants completed the online screening questionnaire between August 30, 2022 and March 24, 2023, of which N = 413 booked a telephone appointment with the study team. Of these, N = 266 were eligible for the study and received informed consent via DocuSign. In total, N = 218 consented to the study and started the 1‐week blood pressure diary. Of these, N = 116 were excluded, because they did not meet the inclusion criteria (systolic blood pressure < 135 mm Hg/ > 175 mm Hg: N = 77; < 3 entries: N = 39). The remaining N = 102 participants were randomly assigned to the intervention group (IG; N = 52) or the control group (CG; N = 50). Of these, N = 81 (79%) completed the post‐treatment questionnaire and N = 80 (78%) completed the post‐treatment 1‐week blood pressure diary. The participant flow is shown in the CONSORT flow diagram (Figure 1). All post‐treatment assessments took place between 19th of December 2022 and 17th of July 2023, when the trial ended with the last participant completing the assessment.

FIGURE 1.

CONSORT flow diagram.

The average age of the study sample was 52.94 ± 9.01 years. Overall, N = 59 (57.8%) patients were female and N = 90 (88.23%) patients reported currently taking antihypertensive medication. Baseline systolic blood pressure was 145.05 ± 8.75 mm Hg. All demographic variables and outcomes at baseline by group are shown in Table 2.

TABLE 2.

Participant characteristics by group.

	Intervention (n = 52)	Control (n = 50)
Demographic variables
Age (years), M (SD)	51.1 (9.2)	54.9 (8.3)
Female, n (%)	31 (59.60)	28 (56.00)
Marital status, n (%)
‐ single	14 (26.3)	9 (18.0)
‐ married	24 (46.2)	31 (62.0)
‐ divorced	12 (23.1)	10 (20.0)
‐ widowed	2 (3.8)	0 (0.0)
Highest level of education, n (%)
‐ secondary school	19 (36.5)	17 (34.0)
‐ high school diploma	15 (28.8)	15 (30.0)
‐ some college credit, no degree	3 (5.8)	2 (4.0)
‐ Bachelor's degree	4 (7.7)	8 (16.0)
‐ Master’ degree	11 (21.2)	8 (16.0)
‐ Doctoral degree	0 (0.0)	0 (0.0)
Employment status, n (%)
‐ Employed for wages < 40 h	23 (44.2)	19 (38.0)
‐ Employed for wages > 40 h	17 (32.7)	10 (20.0)
‐ Self‐employed	2 (3.8)	7 (14.0)
‐ Student	1 (1.9)	0 (0.0)
‐ No employment (searching)	1 (1.9)	2 (4.0)
‐ No employment (not searching)	4 (7.7)	2 (4.0)
‐ Retired	4 (7.7)	7 (14.0)
‐ Unable to work	0 (0.0)	3 (6.0)
Comorbidities, n (%)	24 (46.2)	29 (58.0)
‐ Diabetes	6 (11.5)	6 (12.0)
‐ Dyslipidemia	3 (5.8)	3 (6.0)
‐ Cardiovascular disease	12 (23.1)	17 (34.0)
‐ Other	13 (25.0)	17 (34.0)
	45 (86.5)	45 (90.0)
	1.47 (0.7)	1.58 (0.7)
Currently taking antihypertensive drugs, n (%)
‐ Number of antihypertensive agents per person, M (SD)	18	19
‐ ACE inhibitors, n	23	17
‐ Angiotensin II receptor blockers, n	10	12
‐ Beta blocker, n	12	16
‐ Calcium channel blockers, n  ‐ Diuretics, n  ‐ Other * , n	10 0	12 4
Measurements at baseline
Systolic blood pressure (mm Hg), M (SD)	144.98 (8.29)	145.12 (9.42)
Diastolic blood pressure (mm Hg), M (SD)	87.22 (9.53)	88.49 (7.66)
Heart rate, M (SD)	73.44 (9.53)	76.72 (7.59)
Body mass index, M (SD)	31.42 (7.28)	32.77 (6.03)
Patient activation (PAM‐13), M (SD)	76.99 (13.80)	74.11 (11.19)
Medication adherence (MMAS), M (SD)	5.24 (1.14)	4.88 (1.47)

Abbreviations: M, Mean, SD, Standard deviation, n, number, PAM‐13, Patient activation measure.

^*Imidazoline receptor (n = 2), If‐channel blocker (n = 1), Alpha blocker (n = 1).

3.2. Primary outcomes

3.2.1. Systolic blood pressure

Our primary aim was to assess if actensio reduces systolic blood pressure relative to control. As shown in Table 3 and Figure 2, there were medium‐sized between‐group effects at 12‐weeks post‐randomization. The estimated adjusted mean difference in outcome was −5.06 mm Hg (F[1, 99] = 6.27, p = .013, d = 0.58), indicating that participants in the intervention group had a lower systolic blood pressure than those in the control group.

TABLE 3.

Effects of actensio versus control on outcome measures.

	Intervention	Control
	M ± SD	M ± SD	Diffadj	95% CI		p	ES
Sys. BP N = 102	137.37 ± 10.13	142.35 ± 11.23	−5.057	−8.707	−1.406	.013	0.58
PAM‐13 N = 102	79.38 ± 9.44	75.45 ± 10.62	3.353	−0.182	6.888	.064	0.30
Dias. BP N = 102	84.99 ± 9.53	87.08 ± 10.24	−1.777	−4.500	0.952	.402	0.14
Heart rate N = 102	73.19 ± 9.87	75.14 ± 8.17	−0.684	−3.732	2.364	.683	0.16
BMI N = 102	32.07 ± 8.04	32.94 ± 6.34	−0.842	−1.009	2.694	.393	0.17
MMAS N = 90	5.58 ± 0.88	5.09 ± 1.25	−0.183	−0.596	0.230	0.555	0.14

Note: Significant p‐values are displayed in bold. M and SD refer to unadjusted means and standard deviations.

Abbreviation: 95% CI, 95% confidence interval of the adjusted mean difference; BMI, Body mass index (kg/m²); Dias.BP, Diastolic blood pressure (averaged across 1‐week blood pressure diary); Diff_adj, Adjusted mean difference derived from ANCOVA model; ES, Effect size (Cohen's d); MMAS, Morisky medication adherence scale; PAM‐13, Patient activation measure; Sys.BP, Systolic blood pressure (averaged across 1‐week blood pressure diary).

FIGURE 2.

Changes in systolic blood pressure (left) and patient activation (right), across groups and time points. Raw means (±1 SE) are presented for both groups at each time point. p‐values are derived from ANCOVA models and refer to between‐group comparisons at post treatment.

3.3. Secondary outcomes

3.3.1. Patient activation

Intention‐to‐treat analysis of the PAM‐13 score revealed a group difference of 3.35 points tending towards statistical significance (F[1, 99] = 3.41, p = .064, d = 0.30) in favor of the intervention group (see Table 3 and Figure 2).

3.3.2. Diastolic blood pressure and heart rate

There was no group difference for diastolic blood pressure (F[1, 99] = 0.794, p = .387, d = 0.14) or heart rate (F[1, 99] = 1.111, p = .402, d = 0.16; see Table 3).

3.4. Exploratory outcomes

3.4.1. Body mass index

Intention‐to‐treat analysis revealed no group difference in BMI (F[1, 99] = 0.729, p = .393, d = 0.17; see Table 3). However, the ANCOVA assumption of homogeneity of error variances was violated. Therefore, increasing the risk of type I error and reducing the precision of the p‐value.

3.4.2. Medication and medication adherence

Of the 90 participants (IG = 45; CG = 45), who reported taking antihypertensive medication at baseline, 12 (IG = 5, CG = 7) indicated a change in medication at post‐treatment assessment. Of these, six reported an increase in number or dose (IG = 3; CG = 3), four a change that did not affect number or dose (IG = 2; CG = 2), and two a reduction in number or dose of medication (both in the control group). Intention‐to‐treat analysis revealed no group difference for medication adherence based on the MMAS total score in those who indicated medication intake (F[1, 87] = 0.379, p = .183, d = 0.14; N = 90; see Table 3).

3.4.3. Adverse events

In total, 19 participants (IG = 8, CG = 11) reported an adverse event. Of these, none was stated to be related to the intervention (IG only).

3.4.4. User satisfaction

High average ratings indicated that participants in the intervention group were predominantly satisfied with the intervention across all categories (range 1−5): user handling (M = 4.03 ± 0.96), content (M = 3.82 ± 0.96), design (M = 3.71 ± 1.15), user friendliness (M = 3.88 ± 0.84), recommendation (M = 3.85 ± 1.13).

4. DISCUSSION

We designed this pilot RCT to gain initial insights into the effectiveness of the digital lifestyle intervention, actensio, in reducing blood pressure in patients with hypertension compared to a control group. We hypothesized that actensio would be superior to control across measurements of blood pressure and patient activation. Additionally, the exploratory outcomes BMI and medication adherence were considered, as well as descriptive analysis of user satisfaction and side effects. The beneficial effect of actensio was demonstrated with medium effects sizes in our primary outcome, systolic blood pressure, at 12‐weeks post‐randomization. Against our hypotheses, we did not detect group differences in diastolic blood pressure or heart rate, which could be due to the fact that an isolated systolic hypertension is present in a substantial proportion of our sample and thus an influence on the diastolic value is less visible. ²¹ However, a statistical trend and small effect size pointed towards improved patient activation, which may also explain the reduction in systolic blood pressure in the intervention group. Yet, it is important to mention that this trail was not set up to investigate potential underlying mechanisms, and with the presented data, cannot explain how the antihypertensive effect was achieved by treatment. Exploratory analysis on BMI and medication adherence did not reveal any group differences. These results, however, need to be interpreted with great care, due to their exploratory nature. A previous trial on a similar digital lifestyle intervention in Japan, for example, has shown between‐group differences in BMI in a larger population (N = 390) after 12‐weeks of intervention, although mean differences were even smaller (−0.2 kg/m² vs. −0.8 kg/m²; see Table 3), ³⁷ yet the association of BMI and salt intake on reducing morning SBP point towards theses parameters as relevant endpoints in future research. ³⁸ However, the HERB pilot trial could not show a significant effect on 24‐h SBP or BMI between baseline and follow‐up after 24 weeks. ³⁹ It needs to be mentioned that indicators for lifestyle modifications are not consistently shown and clear evidence for potential treatment mechanisms is still missing. ⁴⁰

Descriptive analysis showed that participants in the intervention group were satisfied with actensio, while number of side effects were comparable between groups and reported to be unrelated to the intervention. These results indicate that actensio is a feasible and safe digital lifestyle intervention for a study population with many comorbidities, high BMI (>30), and low medication adherence (<6), see Table 2.

While these results are encouraging and in line with previous reports on the potential of digital lifestyle interventions in the treatment of hypertension, ¹² they need to be interpreted with the limitations of a pilot trial. Firstly, our results showed effect sizes in the small‐to‐medium range, which the study may have not been powered to detect. Furthermore, we did not correct for multiple testing, increasing the risk for both, type I and type II error. ⁴¹ To allow the reader to make judgements, we report the exact values, confidence intervals, and effect sizes. These in turn also inform the set up for a larger, randomized‐controlled trial to confirm the effectiveness of actensio in the treatment of hypertension. Secondly, we used a diary and home‐based measurement protocol to determine heart rate, systolic and diastolic blood pressure, which may have led to increased variability due to differences in blood pressure monitors and measurement procedures. To counteract variability, all participants were given the same measurement instructions. Thirdly, we implemented an extensive screening procedure, which limits the generalizability of our results. Particularly, we recruited highly motivated patients, who were willing to change their lifestyle. While this may be a prerequisite for any lifestyle intervention, it limits suitability, especially when considering the use of actensio in primary care. Fourthly, we observed higher drop‐out rates at completion of post‐treatment assessments between the intervention and the control group. While this is a common observation in waitlist‐control‐trial, it may indicate that the incentive for completing assessments was too low in the intervention group (the control group received actensio on completion), or participants were not satisfied with the intervention and therefore not willing to complete assessments. This, however, is an assumption we cannot affirm since we do not have user satisfaction data of those who dropped‐out. Fifthly, the choice of comparison was a waitlist‐control‐group with access to regular care, which limits any conclusion to the specificity of actensio and denies the detection of a precise determination factor responsible for the observed reduction in SBP in the intervention group, and potentially inflates effect sizes due to awareness of group assignment (open‐label). Consequently, treatment expectancy may have biased responses in the intervention group and dampened active health behavior in the control group. Yet, visual inspection of the PAM‐13 did not indicate such effects in the control group (see Figure 2). Lastly, this pilot trial was not set up to investigate the processes underlying the antihypertensive treatment effects. And although the intervention targets lifestyle modifications, the current data does not provide for any conclusions whether lifestyle modifications in fact took place. This, however, is crucial to understand treatment effects and needs to be investigated by further research studies focusing on specific treatment processes like changes in diet, physical activity, stress reduction, and weight loss, and unspecific treatment processes like use of the program, adherence, and treatment satisfaction.

5. CONCLUSIONS

This pilot trial showed that actensio is a feasible and safe digital lifestyle intervention to improve hypertension in affected patients compared with a control arm. The intervention has the potential to decrease blood pressure and activate patients in taking control over their disease. A confirmatory trial is needed to verify these results in a larger sample size in regular care settings.

AUTHOR CONTRIBUTIONS

Leonie Franziska Maurer and Christoph Schöbel conceived the idea and Leonie Franziska Maurer, Christoph Schöbel, and Alina Wildenauer planned the study. Alina Wildenauer, Torsten Eggert, and Christoph Schöbel carried out the study. Alina Wildenauer and Leonie Franziska Maurer conducted the data analysis. Alina Wildenauer, Leonie Franziska Maurer, and Laurin Rötzer took the lead in writing the manuscript. All authors provided critical feedback and helped shape the research, analysis and manuscript.

CONFLICT OF INTEREST STATEMENT

Leonie Maurer and Laurin Rötzer are salaried employee of the mementor DE GmbH. Christoph Schöbel is a medical consultant of the mementor DE GmbH.

Wildenauer A, Maurer LF, Rötzer L, Eggert T, Schöbel C. The effects of a digital lifestyle intervention in patients with hypertension: Results of a pilot randomized controlled trial. J Clin Hypertens. 2024;26:902–911. 10.1111/jch.14811

DATA AVAILABILITY STATEMENT

The data are available on request from the corresponding author.