Abstract
Background
Limited data exist on the benefits of lifestyle behavior change delivered using telehealth and web‐based applications with varied support on blood pressure (BP).
Methods and Results
We conducted a 2‐site randomized controlled trial at Geisinger (January 2019–March 2021) to compare the efficacy of 2 remotely delivered strategies using web‐based applications in participants with 24‐hour systolic BP 120–160 mm Hg and body mass index ≥25 kg/m2. Both arms received access to web‐based applications and the same lifestyle guidance per American Heart Association guidelines. One arm received minimal nonclinical staff support, and the other arm received dietitian support with motivational interviewing during weekly calls. The primary outcome was 12‐week change in 24‐hour systolic BP. A total of 187 participants were randomly assigned, with 156 (83.4%) completing the trial. In both arms, 24‐hour systolic BP was reduced at follow‐up, but the difference in BP change was not significant (dietitian‐led arm, −6.73 mm Hg [95% CI, −8.64 to −4.82]; minimal‐support arm, −4.92 [95% CI, −7.01 to −2.77]; P comparing groups=0.2). The dietitian‐support arm had greater 12‐week improvements in the secondary outcomes sleep systolic BP (mean, −6.92 versus −1.45; P=0.004), sleep diastolic BP (−3.31 versus 0.73; P=0.001), and self‐reported physical activity (866 versus −243 metabolic equivalent task minutes per week; P=0.01) and tended to have improvements in weight loss (−5.11 versus −3.89 kg; P=0.1) and Healthy Eating Index–2015 score (9.23 versus 6.43 units; P=0.09).
Conclusions
Both the dietitian‐ and minimal‐support interventions reduced 24‐hour systolic BP similarly, although the dietitian‐led intervention led to greater improvements in several secondary cardiometabolic outcomes.
Registration
URL: https://www.clinicaltrials.gov; Unique identifier: NCT03700710.
Keywords: behavior change, blood pressure, diet, healthy lifestyle, mHealth, motivational interviewing, weight loss
Subject Categories: Diet and Nutrition, Lifestyle, Hypertension, High Blood Pressure, Exercise
Nonstandard Abbreviations and Acronyms
- ABPM
ambulatory blood pressure monitoring
- AOBP
automated office blood pressure
- HEI‐2015
Healthy Eating Index–2015
Clinical Perspective.
What Is New?
This study shows that dietitian‐led and minimal‐support remotely delivered interventions using web‐based applications reduced 24‐hour systolic blood pressure similarly.
The dietitian‐led intervention additionally reduced sleep systolic and diastolic blood pressure and increased physical activity.
Participant satisfaction with the interventions was high overall, but higher among participants in the dietitian study arm.
What Are the Clinical Implications?
The study intervention arms, delivered remotely using web‐based applications, are relatively low cost and could be deployed to individuals with elevated blood pressure to decrease blood pressure and future cardiovascular risk.
Additional studies are needed to test the effectiveness and economics of similar remote approaches using web‐based applications in diverse populations.
Hypertension is a leading cause of morbidity and mortality, affecting 121.5 million adults aged ≥20 years in the United States. 1 Unhealthy dietary patterns, high sodium intake, low physical activity, and obesity contribute significantly to the pathogenesis of hypertension and cardiovascular diseases. 2 , 3 The American College of Cardiology/American Heart Association 2017 Hypertension Guidelines recommend that patients with elevated systolic blood pressure (BP) (120–129 mm Hg) and hypertension (≥130/80 mm Hg) undergo lifestyle modification. Evidence from prior studies have demonstrated that weight loss, healthy dietary patterns (eg, the Dietary Approaches to Stop Hypertension or Mediterranean), and increased physical activity lower BP. 4 , 5 , 6 , 7 , 8
Helping patients lower BP through lifestyle modification is challenging in clinical practice as providers often lack the time and resources to deliver lifestyle interventions. 9 , 10 , 11 Telehealth interventions with mobile health applications and online programs could be useful in aiding patients to improve lifestyle behaviors and nutrition‐related outcomes by providing insights based on their own data. 12 Prior telehealth lifestyle behavior modification trials have been shown to be as effective as traditional, in‐person interventions for weight reduction.
Prior BP telehealth studies have combined remote monitoring with lifestyle modification and antihypertension medication treatment, making it challenging to assess the benefits of lifestyle modification through telehealth/online programs on BP. 13 , 14 Few data exist on the efficacy of these strategies in rural populations, and limited data exist on the benefits of dietitian medical nutrition therapy on BP, assessed by 24‐hour ambulatory BP monitoring (ABPM).
In this context, we conducted a randomized controlled trial comparing the effect of 2 strategies (minimal remote nonclinical staff support versus remote dietitian support with motivational interviewing), both using online programs and mobile applications to promote healthy behavior change, on 12‐week changes in 24‐hour systolic BP and other measures of healthy lifestyle. We hypothesized that the dietitian‐led approach would result in a greater decrease in 24‐hour systolic BP compared with the minimal‐support arm.
Methods
Overall Study Design
The data that support the findings of this study are available from the corresponding author upon reasonable request. The Healthy BP study was a parallel‐arm, randomized clinical trial conducted at 2 hospitals (Geisinger Medical Center, Geisinger Wyoming Valley) in the Geisinger Health System. The study intervention period was 12 weeks to minimize the chance that a participant in the study would undergo a change in antihypertensive medication and because prior studies have shown this time frame is sufficient to show an improvement in BP through lifestyle changes. 2 , 15 , 16 The study was approved by the Geisinger Institutional Review Board, and participants gave written informed consent. As described previously in the study protocol article, 17 the COVID‐19 pandemic occurred in the middle of the trial, necessitating protocol adaptations to preserve clinic space and personal protective equipment and minimize in‐person contact. Nevertheless, primary outcome data collection procedures were not altered throughout the trial period.
Study Participants
Eligibility criteria included the following: aged ≥18 years, body mass index >25 kg/m2, access to a telephone and either a computer or smartphone with internet access, 24‐hour ambulatory systolic BP between 120–160 mm Hg, and successful completion of the run‐in period (ie, enter dietary data for at least 5 of 7 days and enter weight on the online platform). Exclusion criteria included the following: inability to understand English; myocardial infarction, stroke, or atherosclerotic cardiovascular disease within the prior 6 months; planned or previous bariatric surgery; pregnancy, breastfeeding, or planned pregnancy before the end of participation; self‐reported average consumption of >21 alcoholic beverages per week or binge drinking; psychiatric hospitalization in the past year; current angina; plans to leave the area before the end of the study; current participation in another clinical trial; and principal investigator discretion (ie, concerns about safety, compliance). Providers were notified by the principal investigator (A.R.C.) via electronic message about their patients' participation in the study. For patients with 24‐hour or awake systolic BP ≥145 mm Hg, cases were discussed with patients and their providers on whether to start or escalate antihypertensive medication treatments before participation in the study. Participants who had BP medication changes were reevaluated for eligibility with repeat 24‐hour ABPM at least 2 weeks after a medication change.
Trial Conduct
Patients with elevated BP were identified primarily through electronic health records as well as via Geisinger Health Plan health screenings, referrals from primary care providers, and self‐referrals. Potential participants were then invited to complete a 24‐hour ABPM test to confirm that their BP was elevated outside the office as recommended by guidelines. 18 Enrollment began in January 2019 and ended in March 2021. Baseline data were collected during a visit when an ABPM device was placed and a screening visit (initially in‐person, then remote during the pandemic; Figure 1). During a run‐in phase, participants were instructed to enter dietary data (3 eating occasions per day) for at least 5 of 7 days and enter their weight into the online platform to be eligible for randomization. Randomization assignments were made centrally by a computer program with block randomization (random block sizes of 4 and 6) with no stratification variables. An unblinded staff member (not involved in intervention or data collection) opened a sealed opaque envelope with the randomized intervention. Eligible participants were randomly assigned 1:1 to 2 groups: (1) minimal support or (2) dietitian support. The outcome assessor remained blinded throughout the study. Follow‐up data were collected 12 weeks after randomization.
Figure 1. Study flow diagram.
ABPM indicates ambulatory blood pressure monitoring; BP, blood pressure; and SV, study visit.
Intervention Components Common to Both Arms
Both arms received the same lifestyle guidance per American Heart Association guidelines and access to web‐based applications, which included the Evolve (formerly known as BMIQ) online platform, an evidence‐based, Health Insurance Portability and Accountability Act–compliant program 19 , 20 (Table S1). An earlier version of the platform was used from January 2019 to December 2020 in which participants used a meal‐logging application (LoseIt) that integrated with the web platform. Industry‐led updates to the platform were phased in during December 2020 and went live by March 2021. The new platform included all of the same elements; meal‐logging occurred directly on the Evolve platform rather than requiring the use of an external application. Educational materials on the platform, with adaptations by the study team, included materials on promoting weight loss, eating a Dietary Approaches to Stop Hypertension–type diet (increased intake of fruits/vegetables, whole grains, protein sources from plants, lean meats, and seafood, and reduced intake of sodium, sweets, and saturated fat), increasing physical activity, setting goals, overcoming barriers, and preventing relapse. At baseline, participants received a nutrition report based on a food frequency questionnaire (Viocare) that provided personalized suggestions to improve dietary habits that align with national dietary guidelines. 21 Participants (and their providers) were encouraged to avoid BP medication changes during the 12‐week period when possible.
To encourage adherence with dietary data entry, all participants received virtual lottery tickets (lottery drawings for 10 $300 gift certificates) each week in which they entered at least 5 days per week of dietary data and 2 days per week of weights. All participants were offered the opportunity to participate in grocery store tours at a local grocery store (Weis Markets); after March 2020, grocery store tours were offered virtually.
Participant goals for both arms included (1) lose 3% of weight at 12 weeks, (2) consume a healthier dietary pattern (high in fruits; vegetables; whole grains; low‐fat dairy; vegetable, fish, and poultry sources of protein; and healthier sources of fat) and avoid sweets and salt, (3) reduce sodium intake to <2300 mg/d, and (4) participate in at least 180 min/wk of moderate/vigorous intensity physical activity.
Minimal‐Support Arm
Participants in the minimal‐support arm received no additional support unless they were not entering dietary data sufficiently. If they did not enter dietary data at all for a week or were not adherent with the recommended dietary data entry for 2 weeks, a research assistant reached out by phone to troubleshoot and encourage participation.
Dietitian‐Support Arm
Participants in the dietitian‐support arm additionally received weekly telephone calls (initial call 30–60 minutes, follow‐up calls 15–20 minutes) from a study dietitian. Our approach was informed by Carver and Scheier's control systems theory 22 and the situated learning theoretical framework. 23 Dietitians received motivational interviewing training bi‐annually by an expert (C.C.). Dietitians examined meal logs of participants with special focus on sodium, fruits, and non‐starchy vegetables. Dietitians explored participants' goals and values and helped participants set attainable goals around a health behavior for the week and assessed their progress towards those goals.
Data Collection Procedures
ABPM was conducted using SpaceLabs Ontrak devices, which were programmed to take 2 measurements per hour during both awake and sleep periods. Participants were asked to wear the device for 24 hours, record awake and sleep times, and repeat the 24‐hour ABPM if fewer than 14 awake measurements or 7 sleep measurements were obtained. For participants who failed to report awake or sleep times on the activity diary, we visually examined the SpaceLabs Ontrak activity monitor to determine awake and sleep times. 24 Before March 2020, weight, waist circumference, and automated office BP (AOBP) were measured by research staff using standardized procedures. After March 2020, visits were shifted to mostly remote visits to minimize potential exposure and preserve health care resources. Collection of the primary outcome data remained unchanged. Study staff collecting 24‐hour ABPM data remained blinded to randomization assignments. Weight and waist circumference were self‐measured at home by participants with scales and tape measures that were provided if needed. AOBP was not able to be measured during the pandemic. We administered the International Physical Activity Questionnaire Short Form 25 as well as the web‐based Viocare Food Frequency Questionnaire (FFQ) during in‐person visits initially. The Viocare FFQ has been validated previously 21 and can calculate Healthy Eating Index (HEI)–2015 scores (see Data S1 for full details). During the pandemic, an email with a link to the online Viocare FFQ was provided, and the International Physical Activity Questionnaire Short Form responses were collected by telephone.
Outcomes
The primary outcome was change from baseline to 12 weeks in 24‐hour systolic BP. Secondary outcomes included changes in awake and sleep systolic BP; 24‐hour, awake, and sleep diastolic BP; total HEI‐2015 score; weight; waist circumference; physical activity (total metabolic equivalent task minutes per week); and AOBP systolic and diastolic BP. Other outcomes included change in individual components of the HEI‐2015 score, end‐of‐study participant satisfaction, and post hoc outcomes of sodium:potassium ratio, weight loss >3%, and end‐of‐study mean awake BP <130/80 mm Hg.
Statistical Analysis
Primary analyses were complete case analyses, in which participants with missing data were excluded from analyses. Continuous outcomes were compared via linear regression, whereas dichotomized outcomes were compared via logistic regression. A 2‐sided α of 0.05 was used for analyses. We also tested for an interaction between treatment assignment and baseline 24‐hour systolic BP (≥ or <130 mm Hg). Because it is possible that the effects of the interventions or the quality of secondary outcome measurements could be different during the COVID‐19 pandemic, we conducted exploratory analyses testing whether changes from baseline to 12 weeks in 24‐hour systolic BP and other study outcomes differed in participants who completed the intervention before March 1, 2020, versus participants who completed the intervention after March 1, 2020. We also calculated group means for each hour of the day for systolic and diastolic BP at baseline and 12 weeks and then plotted hourly means by randomization group using lowess smoothing. Several sensitivity analyses were conducted: (1) an intention‐to‐treat analysis that imputed missing values for all 187 randomly assigned participants using multivariate normal data augmentation,26 (2) an intention‐to‐treat analysis using all available individual‐level BP readings (n=186; 1 participant's individual 24‐hour ABPM data were unavailable) from 24‐hour ABPM using generalized estimating equations, and (3) an analysis excluding nonadherent participants (<25% weeks with satisfactory completed dietary data entry, defined as at least 3 eating occasions ≥5 days per week).
Based on a sample size of 150 participants completing the study, we estimated that we would have >80% power to detect a difference of 4.6 mm Hg between study arms in change of 24‐hour systolic BP, a difference of 6.9 units in change of total HEI‐2015 score, and a 2.1 kg difference in weight change between the groups. All analyses were performed using Stata 15.1 (StataCorp, College Station, TX).
Results
Of 487 participants who were assessed for eligibility with 24‐hour ABPM, 220 attended study visit 1, and 187 were randomly assigned (Figure 1). Baseline characteristics were similar between both arms (Table 1). Mean age was 54.6 (SD, 13.1) years, mean body mass index was 34.5 (6.5) kg/m2, 52% were women, 97% were non‐Hispanic White participants, 7% had diabetes, and 25% had depression/anxiety. Baseline mean 24‐hour SBP and DBP were 132.7 (8.6) and 77.2 (7.7) mm Hg, 67% had awake BP ≥135/85 mm Hg, and 23% were on BP medications. A total of 156 (83.4%) completed a follow‐up 24‐hour ABPM measurement with 78 completers in each arm. The median numbers of awake and sleep BP readings were 31 (interquartile interval, 28–34) and 16 (14–16), respectively.
Table 1.
Baseline Characteristics
| Minimal support (n=93) | Dietitian support (n=94) | |
|---|---|---|
| Age, y | 54.5 (13.0) | 54.8 (13.2) |
| Female sex | 47 (51%) | 50 (53%) |
| Non‐Hispanic White participants | 90 (97%) | 91 (97%) |
| Education level | ||
| Less than high school | 0 (0.0) | 4 (4%) |
| High school | 18 (19%) | 12 (13%) |
| Some college or trade school | 32 (34%) | 40 (43%) |
| College degree | 27 (29%) | 25 (27%) |
| Graduate or professional school | 16 (17%) | 13 (14%) |
| Annual household income | ||
| <$25 000 | 6 (6%) | 6 (6%) |
| $25 000–$50 000 | 19 (20%) | 22 (23%) |
| $50 000–$75 000 | 14 (15%) | 18 (19%) |
| $75 000–$100 000 | 15 (16%) | 22 (23%) |
| ≥$100 000 | 38 (41%) | 25 (27%) |
| Not reported | 0 (0%) | 1 (1%) |
| Employment status | ||
| Full‐time | 56 (60%) | 52 (55%) |
| Part‐time | 12 (13%) | 10 (11%) |
| Unemployed | 3 (3%) | 7 (7%) |
| Retired | 22 (24%) | 23 (24%) |
| Disabled | 0 (0%) | 2 (2%) |
| Smoking status | ||
| Current | 4 (4%) | 9 (10%) |
| Former | 19 (20%) | 23 (24%) |
| Never | 69 (74%) | 62 (66%) |
| On BP medications | 24 (25.8%) | 19 (20.2%) |
| Diabetes | 5 (5.4%) | 8 (8.5%) |
| Dyslipidemia | 19 (20.4%) | 29 (30.9%) |
| Depression/anxiety | 24 (26%) | 22 (23%) |
| Body mass index, kg/m2 | 34.8 (6.7) | 34.2 (6.2) |
| Waist circumference, cm | 113.7 (15.3) | 113.2 (14.9) |
| SBP, mm Hg | ||
| 24 h | 133.1 (8.8) | 132.4 (14.9) |
| Awake | 138.6 (9.0) | 137.9 (8.7) |
| Sleep | 120.5 (11.3) | 120.2 (10.3) |
| DBP, mm Hg | ||
| 24 h | 77.1 (8.5) | 77.3 (6.8) |
| Awake | 81.3 (9.3) | 81.3 (7.2) |
| Sleep | 67.5 (8.1) | 68.6 (7.4) |
| HEI‐2015 score, units* | 59.7 (11.7) | 61.8 (10.4) |
| Moderate/vigorous activity, MET min per wk | 1589 (2312) | 1392 (2043) |
| Walking activity, MET min per wk | 1137 (1398) | 1006 (1227) |
Data are provided as mean (SD) or number (percentage). BP indicates blood pressure; DBP, diastolic blood pressure; HEI‐2015, Healthy Eating Index–2015; MET, metabolic equivalent task; and SBP, systolic blood pressure.
Total HEI‐2015 score range 0 to 100.
Intervention Fidelity
During the 12‐week intervention period, the median number of weeks with satisfactory dietary data was 10 (interquartile interval, 5–12). The proportion of participants who logged fewer than 25% of weeks was 11.7% in the dietitian‐support arm and 17.2% in the minimal‐support arm. Among dietitian‐support arm participants, the median number of weekly phone calls completed was 11 (interquartile interval, 9–12). Only 3 participants (2 dietitian support, 1 minimal support) had an antihypertensive medication change during the study period.
The primary outcome, 24‐hour systolic BP, decreased from baseline to 12 weeks in both the dietitian‐support arm (−6.73 mm Hg [95% CI, −8.64 to −4.82]) and the minimal‐support arm (−4.92 mm Hg [95% CI, −7.01 to −2.77]), with no significant difference between arms (−1.81 mm Hg [95% CI, −4.66 to 1.05]; P=0.2) (Table 2). However, sleep systolic BP decreased more in the dietitian‐support arm (−6.92 mm Hg [95% CI, −9.33 to −4.51]) than the minimal‐support arm (−1.45 mm Hg [95% CI, −4.28 to 1.38]), with a between‐group difference of −5.47 mm Hg (95% CI, −9.16 to −1.79; P=0.004). Similarly, sleep diastolic BP also decreased greater in the dietitian‐support arm (−3.31 mm Hg [95% CI, −4.83 to −1.78]) than the minimal‐support arm (+0.73 mm Hg [95% CI, −1.03 to 2.49]), with a between‐group difference of −4.04 mm Hg (95% CI, −6.35 to −1.73; P=0.001) (Table 2). Awake systolic and diastolic BP decreased similarly in both arms. Hourly mean systolic and diastolic BP for the 2 randomization groups are shown in Figure 2. At the end of the trial, 41.0% of the dietitian‐support arm had awake BP <130/80 mm Hg and 28.2% in the minimal‐support arm had awake BP <130/80 mm Hg (P=0.09) (Figure 3).
Table 2.
Primary and Secondary Outcomes
| Dietitian‐support arm | Minimal‐support arm | Between‐group difference in change, mean (95% CI) | P value between groups | |||||
|---|---|---|---|---|---|---|---|---|
| Baseline, mean (95% CI) | 12 wk, mean (95% CI) | Change, mean (95% CI) | Baseline, mean (95% CI) | 12 wk, mean (95% CI) | Change, mean (95% CI) | |||
| 24‐h SBP, mm Hg (n=156) | 132.11 (130.23 to 134.00) | 125.38 (123.34 to 127.43) | −6.73 (−8.64 to −4.82) | 133.54 (131.47 to 135.61) | 128.62 (125.92 to 131.31) | −4.92 (−7.01 to −2.77) | −1.81 (−4.66 to 1.05) | 0.2 |
| Awake SBP, mm Hg (n=156) | 137.79 (135.78 to 139.81) | 130.68 (128.27 to 133.09) | −7.12 (−9.16 to −5.07) | 139.35 (137.27 to 141.42) | 133.14 (130.53 to 135.75) | −6.21 (−8.28 to −4.13) | −0.91 (−3.81 to 1.99) | 0.5 |
| Sleep SBP, mm Hg (n=156) | 120.37 (118.15 to 122.59) | 113.45 (111.32 to 115.58) | −6.92 (−9.33 to −4.51) | 120.47 (117.79 to 123.16) | 119.03 (115.52 to 122.53) | −1.45 (−4.28 to 1.38) | −5.47 (−9.16 to −1.79) | 0.004 |
| 24‐h DBP, mm Hg (n=156) | 77.32 (75.84 to 78.80) | 73.86 (73.10 to 75.67) | −3.46 (−4.67 to −2.25) | 76.91 (74.93 to 78.89) | 74.91 (72.75 to 77.07) | −2.00 (−3.52 to −0.48) | −1.46 (−3.39 to 0.47) | 0.1 |
| Awake DBP, mm Hg (n=156) | 81.47 (79.87 to 83.08) | 77.63 (76.07 to 79.19) | −3.85 (−5.12 to −2.57) | 81.22 (79.03 to 83.41) | 78.37 (76.16 to 80.59) | −2.85 (−4.52 to −1.17) | −1.00 (−3.09 to 1.09) | 0.3 |
| Sleep DBP, mm Hg (n=156) | 68.91 (67.32 to 70.50) | 65.60 (64.12 to 67.08) | −3.31 (−4.83 to −1.78) | 66.90 (65.03 to 68.77) | 67.63 (65.20 to 70.05) | 0.73 (−1.03 to 2.49) | −4.04 (−6.35 to −1.73) | 0.001 |
| Weight, kg (n=159) | 101.34 (96.31 to 106.36) | 96.22 (91.41 to 101.03) | −5.11 (−6.15 to −4.08) | 101.69 (96.82 to 106.56) | 97.80 (92.88 to 102.71) | −3.89 (−4.92 to −2.86) | −1.22 (−2.68 to 0.23) | 0.1 |
| Waist circumference, cm (n=144) | 113.51 (110.06 to 116.96) | 108.57 (105.48 to 111.66) | −4.94 (−6.71 to −3.17) | 113.03 (109.55 to 116.52) | 109.02 (105.51 to 112.53) | −4.02 (−5.83 to −2.20) | −0.92 (−3.44 to 1.60) | 0.5 |
| HEI‐2015 score, units (n=154)* | 62.07 (59.61 to 64.53) | 71.30 (69.39 to 73.22) | 9.23 (6.79 to 11.68) | 60.52 (57.89 to 63.16) | 66.96 (64.48 to 69.43) | 6.43 (4.30 to 8.56) | 2.80 (−0.42 to 6.03) | 0.09 |
| Total MET min per week (n=162) | 2353 (1791 to 2916) | 3220 (2523 to 3917) | 866 (250 to 1483) | 2703 (2069 to 3338) | 2460 (1958 to 2963) | −243 (−869 to 382) | 1109 (238 to 1981) | 0.01 |
| AOBP SBP, mm Hg (n=71)† | 134.76 (130.50 to 139.01) | 128.38 (124.80 to 131.96) | −6.38 (−10.72 to −2.04) | 138.79 (133.28 to 144.31) | 135.47 (129.47 to 141.47) | −3.32 (−8.80 to 2.16) | −3.05 (−9.86 to 3.75) | 0.4 |
| AOBP DBP, mm Hg (n=71)† | 84.54 (81.78 to 87.30) | 79.41 (76.58 to 82.23) | −5.14 (−8.51 to −1.76) | 90.21 (85.89 to 94.52) | 84.97 (80.23 to 89.71) | −5.24 (−8.99 to −1.48) | 0.10 (−4.85 to 5.05) | 1.0 |
AOBP indicates automated office blood pressure; DBP, diastolic blood pressure; HEI‐2015, Healthy Eating Index–2015; MET, metabolic equivalent task; and SBP, systolic blood pressure.
Total HEI‐2015 score range 0 to 100.
AOBP was only taken on participants who completed the study before March 2020.
Figure 2. Hourly SBP and DBP measured by 24‐hour ABPM.
Mean hourly SBP and DBP for each study group were calculated at baseline and 12 weeks and plotted using lowess smoothing. ABPM indicates ambulatory blood pressure monitoring; DBP, diastolic blood pressure; and SBP, systolic blood pressure.
Figure 3. American Heart Association blood pressure categories at baseline and 12 weeks.
Breakdown of American Heart Association blood pressure categories at baseline and 12 weeks among participants with complete 24‐hour ambulatory blood pressure monitoring data. At the end of 3 months, 41.0% of the dietitian‐support arm and 28.2% of the minimal‐support arm had awake blood pressure at <130/80 mm Hg (P value comparing groups=0.09). DBP indicates diastolic blood pressure; and SBP, systolic blood pressure.
The dietitian‐support arm also had greater improvements in total metabolic equivalent task minutes per week (between‐group difference of 1109 [95% CI, 238–1981]; P=0.01), with numerically greater but nonsignificant improvements in HEI‐2015 score (2.80 [95% CI, −0.42 to 6.03]; P=0.09), weight (−1.22 kg [95% CI, −2.68 to 0.23]; P=0.1), and sodium:potassium ratio (−0.12 [95% CI, −0.28 to 0.03]; P=0.1) compared with the minimal‐support arm. Weight loss >3% was achieved by 50.0% in the dietitian‐support arm and 41.9% in the minimal‐support arm (P=0.3). The dietitian‐support arm had greater improvements in HEI‐2015 subcategories, including total vegetables, greens and beans, total protein foods, and refined grains (Table 3). Participants in the dietitian‐support arm also reported greater satisfaction with the research study than the minimal‐support arm (Figure 4).
Table 3.
Changes in Individual Components of HEI‐2015 Score, Sodium, and Potassium by Randomization Group
| Dietitian‐support arm | Minimal‐support arm | Dietitian–Minimal support | P value | |
|---|---|---|---|---|
| Total HEI‐2015 score | 9.23 (6.79 to 11.68) | 6.43 (4.30 to 8.56) | 2.80 (−0.42 to 6.03) | 0.09 |
| Total fruits | 1.25 (0.82 to 1.68) | 0.99 (0.68 to 1.31) | 0.26 (−0.27 to 0.79) | 0.3 |
| Whole fruits | 1.15 (0.77 to 1.52) | 1.04 (0.72 to 1.36) | 0.11 (−0.38 to 0.60) | 0.2 |
| Total vegetables | 0.69 (0.48 to 0.91) | 0.35 (0.13 to 0.57) | 0.34 (0.03 to 0.65) | 0.03 |
| Greens and beans | 1.05 (0.74 to 1.36) | 0.52 (0.19 to 0.84) | 0.54 (0.09 to 0.99) | 0.02 |
| Whole grains | 1.06 (0.28 to 1.85) | 0.81 (0.13 to 1.50) | 0.25 (−0.79 to 1.29) | 0.6 |
| Dairy | −0.41 (−0.97 to 0.16) | −0.11 (−0.69 to 0.47) | −0.30 (−1.09 to 0.50) | 0.5 |
| Total protein foods | 0.22 (0.01 to 0.43) | −0.08 (−0.30 to 0.14) | 0.30 (0.00 to 0.60) | 0.05 |
| Seafood and plant proteins | 0.42 (0.12 to 0.73) | 0.08 (−0.29 to 0.46) | 0.34 (−0.14 to 0.82) | 0.2 |
| Fatty acids | 1.45 (0.83 to 2.07) | 1.04 (0.48 to 1.60) | 0.41 (−0.42 to 1.24) | 0.3 |
| Refined grains | 1.09 (0.54 to 1.64) | 0.36 (−0.11 to 0.83) | 0.73 (0.01 to 1.45) | 0.05 |
| Sodium | −0.84 (−1.52 to −0.15) | −0.49 (−1.19 to 0.22) | −0.35 (−1.33 to 0.63) | 0.5 |
| Added sugars | 0.53 (0.13 to 0.93) | 0.57 (0.02 to 1.12) | −0.04 (−0.71 to 0.64) | 0.9 |
| Saturated fats | 1.55 (0.89 to 2.21) | 1.34 (0.73 to 1.95) | 0.21 (−0.69 to 1.10) | 0.6 |
| Total calories, kcal/d | −674 (−891 to −457) | −546 (−753 to −339) | −128 (−426 to 170) | 0.4 |
| Sodium, mg/d | −920 (−1252 to −587) | −845 (−1156 to −535) | −74 (−526 to 378) | 0.8 |
| Sodium density, mg/kcal per d | 0.13 (0.04 to 0.22) | 0.04 (−0.04 to 0.12) | 0.09 (−0.03 to 0.21) | 0.1 |
| Potassium, mg/d | −307 (−589 to −25) | −341 (−585 to −97) | 34 (−336 to 405) | 0.9 |
| Potassium density, mg/kcal per d | 0.43 (0.33 to 0.54) | 0.22 (0.13 to 0.31) | 0.21 (0.07 to 0.35) | 0.003 |
| Sodium:potassium molar ratio | −0.35 (−0.46 to −0.23) | −0.22 (−0.33 to −0.12) | −0.12 (−0.28 to 0.03) | 0.1 |
The first two columns show mean (95% CI) changes in dietary measures, from baseline to 12 months in the study arms. The third column shows the between‐group difference (95% CI) in changes in dietary measures. HEI‐2015 indicates Healthy Eating Index–2015.
Figure 4. Satisfaction by study arm.
χ2 test, P=0.004. BP indicates blood pressure.
There were no significant interactions between treatment assignment and baseline 24‐hour systolic BP (≥ or <130 mm Hg; P=0.4) for the primary outcome. Likewise, there were no interactions between treatment assignment and participation during the COVID‐19 pandemic with any of the study outcomes (Table S2). Among 71 pre–COVID‐19 period participants, changes in systolic BP measured by AOBP were −6.38 mm Hg (95% CI, −10.72 to −2.04) in the dietitian‐support arm and −3.32 mm Hg (95% CI, −8.80 to 2.16) in the minimal‐support arm (Table 2). Results were similar in a sensitivity analysis imputing missing outcome data, a sensitivity analysis using a generalized estimating equation approach with all individual 24‐hour BP data, and a sensitivity analysis excluding 12 participants who logged satisfactory dietary data <25% of the weeks (Tables S3 through S5).
There were 3 adverse events unrelated to the study interventions, including 2 new cancer diagnoses and a snow‐tubing accident resulting in multiple fractures.
Discussion
In this randomized controlled trial, we demonstrate that the remote delivery of lifestyle intervention using web‐based applications with minimal support versus dietitian support reduced 24‐hour systolic BP without significant difference. Although there were no significant differences in the primary outcome 24‐hour systolic BP, we observed significantly greater reductions in sleep systolic BP and diastolic BP in the dietitian‐support arm than the minimal‐support arm. In addition, the dietitian‐support arm significantly improved physical activity more than the minimal‐support arm and tended to better improve dietary quality and weight loss.
The observed greater improvements in sleep systolic and diastolic BP in the dietitian‐support arm are important to note as numerous studies have found that elevated nocturnal BP and a lack of fall in BP during sleep (ie, “nondipping”) are important risk factors for cardiovascular disease and mortality, even after adjusting for traditional risk factors and 24‐hour systolic BP. 27 , 28 , 29 One explanation for why the dietitian‐support arm experienced greater improvements in sleep BP but not awake BP could be that measurement during sleep is more standardized (laying down, relatively motionless) than awake BP. Increased physical activity during awake hours of BP measurement could have resulted in transiently higher awake BP related to physical activity, which was higher in the dietitian‐support arm than the minimal‐support arm.
In contrast to our findings, Lopes et al reported that a supervised aerobic training program successfully reduced 24‐hour systolic BP and daytime BP with no significant effect on nighttime systolic BP. 30 A key difference in their trial was that participants were instructed to not exercise during 24‐hour ambulatory BP measurements, whereas participants in our trial were instructed to continue their regular daily routine without any specific restrictions on exercise. In addition, because 12‐week weight loss was relatively large in both the dietitian‐support arm (−5.11 kg) and the minimal‐support arm (−3.89 kg), the strong effect of weight loss on awake BP may have reduced our ability to detect further benefit of BP from other aspects of the telehealth intervention, that is, diet and physical activity.
There were some limitations in our trial. Study participants were predominantly non‐Hispanic White (97%) participants, reflecting the demographics of the surrounding, predominantly rural area. In addition, the study population skewed on the more educated side. Although our study did not include a usual care control group, a prior study found that use of the Evolve platform alone, or in combination with nonclinical staff support, increased weight loss at 12 months (online platform alone, −1.9 kg; combined intervention, −3.1 kg) compared with usual care (−1.2 kg) and that these trends persisted at 18 months. 19 Another limitation was that we shifted to remote research visits after March 2020 out of necessity because of COVID‐19 pandemic‐related restrictions. This resulted in reliance on participants self‐measuring weight and waist circumference and an inability to measure AOBP on about half of our study participants, which may have adversely impacted data quality for these secondary outcomes.
There were a number of strengths in our study. Foremost, the use of 24‐hour ABPM enhanced the power to detect changes in mean BP through repeated measurements while allowing examination of nocturnal BP. We included a wide age range of participants with hypertension, ranging from 23 to 89 years. Unlike most BP telehealth intervention studies, we designed our study to minimize any changes in BP medications successfully and thus were able to quantify lifestyle‐induced changes in BP. Adherence to patient‐reported data entry and dietitian phone calls was excellent overall, although both intervention arms had received a modest financial incentive (lottery to increase adherence to dietary data entry).
The magnitude of the improvements in BP were substantial, similar to a BP medication, and in line with much more resource‐intensive feeding studies and lifestyle intervention trials. 2 , 4 , 15 , 30 Elevated nocturnal BP increases the risk for cardiovascular disease independent of daytime BP, and our secondary findings suggest that dietitian support combined with web‐based applications can reduce this important cardiovascular risk factor. 27 Both intervention strategies, using web‐based applications with remote support by dietitians or nonclinical staff, are scalable interventions that can be easily deployed remotely as the intervention only required internet access and telephone access (no smartphone required). According to the Pew Research Center, 93% of US adults report using the internet and 77% report having access to home broadband. 31 Although we did not plan a formal economic analysis, study costs were relatively inexpensive, with a per‐participant cost of $36 for digital health vendors Evolve and Viocare and ≈$400 per‐participant cost (directs only) for dietitian services. Additional studies are needed testing the effectiveness and economics of similar remote approaches using web‐based applications in diverse populations.
In conclusion, dietitian‐support and minimal‐support approaches using web‐based applications resulted in similar reductions in 24‐hour systolic BP. The dietitian‐support intervention lowered sleep BP and improved physical activity greater than the minimal‐support intervention.
Sources of Funding
This work was supported by Geisinger Health Plan. Chang received support from the National Institutes of Health/National Institute of Diabetes and Digestive and Kidney Diseases (K23DK106515). The funders had no role in the study design, collection, management, analysis, interpretation of data, or the decision to submit the report to publication.
Disclosures
None.
Supporting information
Data S1
Tables S1–S5
Acknowledgments
We thank the participants of this research study as well as Dr Aronne (Evolve) and R. Weiss (Viocare).
A version of this work was presented at the American Heart Association Epidemiology, Lifestyle Scientific Sessions, March 1–4, 2022.
Supplemental Material is available at https://www.ahajournals.org/doi/suppl/10.1161/JAHA.122.027213
For Sources of Funding and Disclosures, see page 10.
Preprint posted on MedRxiv April 3, 2022. https://doi.org/10.1101/2022.03.31.22273248.
References
- 1. Tsao CW, Aday AW, Almarzooq ZI, Alonso A, Beaton AZ, Bittencourt MS, Boehme AK, Buxton AE, Carson AP, Commodore‐Mensah Y, et al. Heart disease and stroke statistics‐2022 update: a report from the American Heart Association. Circulation. 2022;145:e153–e639. doi: 10.1161/CIR.0000000000001052 [DOI] [PubMed] [Google Scholar]
- 2. Sacks FM, Svetkey LP, Vollmer WM, Appel LJ, Bray GA, Harsha D, Obarzanek E, Conlin PR, Miller ER III, Simons‐Morton DG, et al. Effects on blood pressure of reduced dietary sodium and the Dietary Approaches to Stop Hypertension (DASH) diet. DASH‐sodium collaborative research group. N Engl J Med. 2001;344:3–10. doi: 10.1056/NEJM200101043440101 [DOI] [PubMed] [Google Scholar]
- 3. Whelton PK, Carey RM, Aronow WS, Casey DE Jr, Collins KJ, Dennison Himmelfarb C, DePalma SM, Gidding S, Jamerson KA, Jones DW, et al. 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA guideline for the prevention, detection, evaluation, and management of high blood pressure in adults: a report of the American College of Cardiology/American Heart Association task force on clinical practice guidelines. Hypertension. 2018;71:e13–e115. doi: 10.1161/HYP.0000000000000065 [DOI] [PubMed] [Google Scholar]
- 4. Appel LJ, Champagne CM, Harsha DW, Cooper LS, Obarzanek E, Elmer PJ, Stevens VJ, Vollmer WM, Lin PH, Svetkey LP, et al. Effects of comprehensive lifestyle modification on blood pressure control: main results of the PREMIER clinical trial. JAMA. 2003;289:2083–2093. [DOI] [PubMed] [Google Scholar]
- 5. Tuomilehto J, Jousilahti P, Rastenyte D, Moltchanov V, Tanskanen A, Pietinen P, Nissinen A. Urinary sodium excretion and cardiovascular mortality in Finland: a prospective study. Lancet. 2001;357:848–851. doi: 10.1016/S0140-6736(00)04199-4 [DOI] [PubMed] [Google Scholar]
- 6. Jennings A, Berendsen AM, de Groot LCPGM, Feskens EJM, Brzozowska A, Sicinska E, Pietruszka B, Meunier N, Caumon E, Malpuech‐Brugère C, et al. Mediterranean‐style diet improves systolic blood pressure and arterial stiffness in older adults. Hypertension. 2019;73:578–586. doi: 10.1161/HYPERTENSIONAHA.118.12259 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Davis CR, Hodgson JM, Woodman R, Bryan J, Wilson C, Murphy KJ. A Mediterranean diet lowers blood pressure and improves endothelial function: results from the MedLey randomized intervention trial. Am J Clin Nutr. 2017;105:1305–1313. doi: 10.3945/ajcn.116.146803 [DOI] [PubMed] [Google Scholar]
- 8. Ndanuko RN, Tapsell LC, Charlton KE, Neale EP, Batterham MJ. Dietary patterns and blood pressure in adults: a systematic review and meta‐analysis of randomized controlled trials. Adv Nutr. 2016;7:76–89. doi: 10.3945/an.115.009753 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Berman AH, Kolaas K, Petersén E, Bendtsen P, Hedman E, Linderoth C, Müssener U, Sinadinovic K, Spak F, Gremyr I, et al. Clinician experiences of healthy lifestyle promotion and perceptions of digital interventions as complementary tools for lifestyle behavior change in primary care. BMC Fam Pract. 2018;19:139. doi: 10.1186/s12875-018-0829-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Rubio‐Valera M, Pons‐Vigués M, Martínez‐Andrés M, Moreno‐Peral P, Berenguera A, Fernández A. Barriers and facilitators for the implementation of primary prevention and health promotion activities in primary care: a synthesis through meta‐ethnography. PLoS One. 2014;9:e89554. doi: 10.1371/journal.pone.0089554 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Kardakis T, Jerdén L, Nyström ME, Weinehall L, Johansson H. Implementation of clinical practice guidelines on lifestyle interventions in Swedish primary healthcare ‐ a two‐year follow up. BMC Health Serv Res. 2018;18:227. doi: 10.1186/s12913-018-3023-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Villinger K, Wahl DR, Boeing H, Schupp HT, Renner B. The effectiveness of app‐based mobile interventions on nutrition behaviours and nutrition‐related health outcomes: a systematic review and meta‐analysis. Obes Rev. 2019;20:1465–1484. doi: 10.1111/obr.12903 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Tucker KL, Sheppard JP, Stevens R, Bosworth HB, Bove A, Bray EP, Earle K, George J, Godwin M, Green BB, et al. Self‐monitoring of blood pressure in hypertension: a systematic review and individual patient data meta‐analysis. PLoS Med. 2017;14:e1002389. doi: 10.1371/journal.pmed.1002389 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Omboni S, McManus RJ, Bosworth HB, Chappell LC, Green BB, Kario K, Logan AG, Magid DJ, Mckinstry B, Margolis KL, et al. Evidence and recommendations on the use of telemedicine for the Management of Arterial Hypertension: an international expert position paper. Hypertension. 2020;76:1368–1383. doi: 10.1161/HYPERTENSIONAHA.120.15873 [DOI] [PubMed] [Google Scholar]
- 15. Miller ER 3rd, Erlinger TP, Young DR, Jehn M, Charleston J, Rhodes D, Wasan SK, Appel LJ. Results of the diet, exercise, and weight loss intervention trial (DEW‐IT). Hypertension. 2002;40:612–618. doi: 10.1161/01.HYP.0000037217.96002.8E [DOI] [PubMed] [Google Scholar]
- 16. Chang AR, Bailey‐Davis L, Hetherington V, Ziegler A, Yule C, Kwiecen S, Graboski E, Melough MM, Collins C, Anderson C. Remote dietary counseling using smartphone applications in patients with stages 1‐3a chronic kidney disease: a mixed methods feasibility study. J Ren Nutr. 2020;30:53–60. doi: 10.1053/j.jrn.2019.03.080 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Taher M, Yule C, Bonaparte H, Kwiecien S, Collins C, Naylor A, Juraschek SP, Bailey‐Davis L, Chang AR. Telehealth versus self‐directed lifestyle intervention to promote healthy blood pressure: a protocol for a randomised controlled trial. BMJ Open. 2021;11:e044292. doi: 10.1136/bmjopen-2020-044292 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. US Preventive Services Task Force , Krist AH, Davidson KW, Mangione CM, Cabana M, Caughey AB, Davis EM, Donahue KE, Doubeni CA, Kubik M, et al. Screening for hypertension in adults: US preventive services task force reaffirmation recommendation statement. JAMA. 2021;325:1650–1656. doi: 10.1001/jama.2021.4987 [DOI] [PubMed] [Google Scholar]
- 19. Baer HJ, Rozenblum R, De La Cruz BA, Orav EJ, Wien M, Nolido NV, Metzler K, McManus KD, Halperin F, Aronne LJ, et al. Effect of an online weight management program integrated with population health management on weight change: a randomized clinical trial. JAMA. 2020;324:1737–1746. doi: 10.1001/jama.2020.18977 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Barenbaum SR, Mathews SE, Saunders KH, Minero G, Mauer E, Aronne LJ, Shukla AP. Utility of BMIQ, a novel web‐based weight management programme, at an academic weight management centre. Obes Sci Pract. 2020;6:134–138. doi: 10.1002/osp4.393 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Kristal AR, Kolar AS, Fisher JL, Plascak JJ, Stumbo PJ, Weiss R, Paskett ED. Evaluation of web‐based, self‐administered, graphical food frequency questionnaire. J Acad Nutr Diet. 2014;114:613–621. doi: 10.1016/j.jand.2013.11.017 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Carver C, Scheier M. On the Self Regulation of Behavior. Cambridge University Press; 1998. doi: 10.1017/CBO9781139174794 [DOI] [Google Scholar]
- 23. Lave J, Wenger E. Situated Learning: Legitimate Peripheral Participation. Cambridge University Press; 1991. doi: 10.1017/CBO9780511815355 [DOI] [Google Scholar]
- 24. de Greeff A, Shennan AH. Validation of the Spacelabs 90227 OnTrak device according to the European and British hypertension societies as well as the American protocols. Blood Press Monit. 2020;25:110–114. doi: 10.1097/MBP.0000000000000424 [DOI] [PubMed] [Google Scholar]
- 25. Craig CL, Marshall AL, Sjöström M, Bauman AE, Booth ML, Ainsworth BE, Pratt M, Ekelund U, Yngve A, Sallis JF, et al. International physical activity questionnaire: 12‐country reliability and validity. Med Sci Sports Exerc. 2003;35:1381–1395. doi: 10.1249/01.MSS.0000078924.61453.FB [DOI] [PubMed] [Google Scholar]
- 26. Schafer JL. Analysis of Incomplete Multivariate Data. CRC Press; 1997.
- 27. Fagard RH, Celis H, Thijs L, Staessen JA, Clement DL, De Buyzere ML, De Bacquer DA. Daytime and nighttime blood pressure as predictors of death and cause‐specific cardiovascular events in hypertension. Hypertension. 2008;51:55–61. doi: 10.1161/HYPERTENSIONAHA.107.100727 [DOI] [PubMed] [Google Scholar]
- 28. Ohkubo T, Imai Y, Tsuji I, Nagai K, Watanabe N, Minami N, Kato J, Kikuchi N, Nishiyama A, Aihara A, et al. Relation between nocturnal decline in blood pressure and mortality. The Ohasama study. Am J Hypertens. 1997;10:1201–1207. doi: 10.1016/S0895-7061(97)00274-4 [DOI] [PubMed] [Google Scholar]
- 29. Yang WY, Melgarejo JD, Thijs L, Zhang ZY, Boggia J, Wei FF, Hansen TW, Asayama K, Ohkubo T, Jeppesen J, et al. Association of office and ambulatory blood pressure with mortality and cardiovascular outcomes. JAMA. 2019;322:409–420. doi: 10.1001/jama.2019.9811 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Lopes S, Mesquita‐Bastos J, Garcia C, Bertoquini S, Ribau V, Teixeira M, Ribeiro IP, Melo JB, Oliveira J, Figueiredo D, et al. Effect of exercise training on ambulatory blood pressure among patients with resistant hypertension: a randomized clinical trial. JAMA Cardiol. 2021;6:1317–1323. doi: 10.1001/jamacardio.2021.2735 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Internet/Broadband Fact Sheet. April 7, 2021. Available at: https://www.pewresearch.org/internet/fact‐sheet/internet‐broadband/. Accessed March 20, 2022.
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data S1
Tables S1–S5