MOBILE Intervention Pilot Study in College Students with Elevated Blood Pressure

Abstract

Background:

Hypertension is typically recognized in middle-aged and older adults but often overlooked in younger populations.

Objective:

We evaluated a mobile intervention for reducing blood pressure (BP) in college-age students over 28 days.

Methods:

Students with elevated BP or undiagnosed hypertension were assigned to an intervention or control group. All subjects completed baseline questionnaires and attended an educational session. For 28 days, intervention subjects sent their BP and motivation levels to the research team and completed assigned BP-reducing tasks. After 28 days, all subjects completed an exit interview.

Results:

We found a statistically significant decrease in BP in the intervention group only (P = .001) but no statistical difference in sodium intake for either group. Mean hypertension knowledge increased in both groups but was only significant for the control group (P = .001).

Conclusions:

The results provide preliminary data on BP reduction with greater impact on the intervention group.

Introduction

Hypertension is typically recognized in middle-aged and older adults but is often overlooked in younger populations.^1,2 Adjusted for age, hypertension among U.S. adults 20 and older was estimated to be 46.0% by the National Health and Nutrition Examination Survey (2013–2016).³ Further, it affects 1 in 8 adults between 20 and 40 years of age, with prime college years included in this range.⁴

When carefully examined, college students express multiple behaviors that increase their susceptibility to hypertension. Students endure higher degrees of stress, anxiety, and insufficient sleep.^5–6 Due to time constraints, students may be unable to exercise regularly and prioritize fast food over nutritious meals.⁵ Additionally, high alcohol consumption, which is prevalent among college students also contribute to weight gain.⁷ Students can prevent hypertension with lifestyle adjustments and regular blood pressure (BP) monitoring,^8–9 but regular monitoring is uncommon within this population, often due to a lack of insurance and time, the hassle of scheduling appointments, perceived invincibility, and a belief that medical care is unnecessary.¹⁰ Therefore, for this population, self-monitor BP (SMBP) is ideal and effective in controlling hypertension.

In a cross-sectional study on 13,512 young adults, 75% of college students were unaware they had high BP.⁹ Since hypertension is asymptomatic, most individuals do not feel its effects.¹ Without early detection and intervention, undiagnosed hypertension increases with age; by adulthood, it can contribute to an earlier onset of CVD such as coronary heart disease, heart failure, stroke, and transient ischemic attacks.⁴

This study aimed to highlight the importance of regular BP monitoring, encourage elevated BP and undiagnosed hypertensive college students to adopt BP-reducing habits, and determine if the mHealth to Optimize Blood Pressure Improvement (MOBILE) Intervention effectively lowers BP in undiagnosed hypertensive full-time college students between 18 to 29 years of age.

According to the Fogg Behavior Model (FBM), an individual who wants to change their habits must be motivated to perform the requisite tasks, and the tasks must align with their current motivation level.¹¹ In essence, difficult tasks require higher motivation, and easier tasks require lower motivation. FBM affirms that behavior change (B) relies on 3 elements: Motivation (M), Ability (A), and Prompt (P), where B = MAP.^11–12 Motivation refers to an individual’s desire to engage in an activity. Ability refers to an individual’s capacity or skills to perform and complete a task successfully. Prompt refers to the external task given to an individual to perform. According to FBM, the more motivated an individual is, the more likely they will perform a difficult task.

The MOBILE Intervention integrates FBM principles with technology (i.e., smartphones and mHealth) to encourage undiagnosed hypertensive college students to adopt positive, BP-reducing habits. The intervention considers each subject’s daily motivation level and tailors BP-lowering tasks to the self-reported states.

Methods

Study Design

We recruited full-time college students between 18 to 29 years of age to participate in the MOBILE intervention. This randomized controlled trial pilot study used mHealth technology to modify behavior and encourage healthy habits to lower BP. This study incorporated the subject’s current behavior and recommended behavior-modifying tasks tailored to their daily motivation level via SMS messaging. This approach required intervention subjects to monitor their BP daily using a provided Withings Bluetooth BP cuff. Approval from the Institutional Review Board was obtained prior to the start of the study.

The Formative Phase

The formative phase assessed and rated 83 potential SMS interventions using the motivation level required to complete them. We recruited 10 full-time college students 18 to 29 years of age and separated them into 2 groups of 5 students each. Each group participated in a 30-minute Zoom meeting and rated 83 SMS interventions on a scale of 1 to 5 and X, where 1 identified the SMS intervention as requiring low motivation; 3, moderate motivation; and 5, high motivation. Poorly written and/or unsuitable interventions were marked with an X to be discarded or reconsidered. Appropriate considerations were made following the formative phase, and adjustments were made accordingly and a total of 81 messages were kept. SMS interventions were created and developed by the research team using American Heart Association (AHA) hypertension facts, knowledge, and guidelines aimed at reducing BP. Intervention messages varied from physical activity tasks (e.g., “Let’s reach for 2,000 steps today”—low motivation), nutritional tasks (e.g., “How about only consume low fat dairy products today”—moderate motivation), motivational tasks (e.g., “Let’s recite some motivational self-talk today: I can do this, I can lower my high BP), and learning tasks (e.g., “Let’s get creative! Let’s make your own colorful salad today and compare how your body may feel differently as compared to a heavy greasy meal.”—high motivation).

Recruitment, Enrollment, and Measurements

We recruited full-time undergraduate (≥ 12 credits) or graduate (≥ 9 credits) students 18 to 29 years of age with (a) elevated BP (systolic BP [SBP] 120 to 129 mm Hg and diastolic BP [DBP] < 80 mm Hg) or undiagnosed hypertension stage I (SBP 130 to 139 mm Hg or DBP 80 to 89 mm Hg), as defined by the AHA,¹⁴ and (b) regular access to a smartphone with unlimited texting. We excluded students who were pregnant, lactating, planning to become pregnant during the study, taking antihypertensive medication, or diagnosed with a life-threatening illness or condition associated with hypertension.

Recruitment were completed within six months and recruitment strategies included posting flyers around campus, advertising via university media platforms, and communicating the study to student organizations and departments. Interested students who contacted the principal investigator (PI) were invited to a 30-minute Zoom meeting, where the PI explained the study, answered questions, and assessed the student’s eligibility to participate. Potential participants meeting all preliminary inclusion criteria then met with the research assistant in person to have their BP screened. At this meeting, two measurements were taken 5 minutes apart using an FDA-approved Bluetooth Withings BP cuff. The average of both measurements was used. If potential participants met the BP inclusion criteria, we obtained their informed consent, randomized them into either the control or intervention group using a random number generator, and notified them of their assigned group. The research assistant then scheduled the participants for a second in-person meeting, the educational session, where participants received information on BP and completed various questionnaires. Before the educational session began, all subjects’ BPs were taken again to ensure they still met the required BP level. If any subject’s BP was normal, they were no longer qualified to participate in the study.

During the educational session, all subjects self-reported their anthropometric measurements (i.e., height in inches, weight in pounds) and sociodemographic information (i.e., age, race/ethnicity, sex, education, marital status, and insurance coverage). We used the height and weight information to calculate each subject’s body mass index (BMI; < 18.5 = underweight, 18.5 to 24.9 = normal/healthy weight, 25.0 to 29.9 = overweight, and above 30.0 = obese).⁸ Subjects self-reported any history of diabetes mellitus, heart problems, high BP, and cigarette or e-cigarette use; current use of antihypertensive medications; and family history of heart disease. We also administered the Hypertension Knowledge-Level Scale (HK-LS), which has demonstrated reliability and validity for use with adults.¹³ The scoring for the HK-LS ranges from 0 and 22; each correct answer is worth 1 point. The higher the score, the higher the subject’s knowledge of hypertension. The Cronbach’s Alpha for the HK-LS was .764 for our sample of 29 participants. To measure nutrient and sodium intake, the research team used the Automated Self-Administered Recall System (ASA-24) Dietary Assessment Tool. Both the HK-LS and ASA-24 were administered to subjects at the educational session and exit meeting. All questionnaire/assessment responses were collected via Qualtrics survey.

Intervention Group

Intervention group participants communicated with the research assistant via SMS daily for 28 days. Participants sent the research assistant their motivation level (1 = low, 3 = moderate, 5 = high) and BP (measured using a provided Bluetooth Withings BP cuff). Participants were urged to measure their BP before exercising or consuming caffeinated beverages or food. After receiving the participant’s daily BP and motivation level, the research assistant used a random number generator to select an intervention task based on the subject’s reported motivation, then relayed the task to the participant. Participants could request a different intervention task and change their motivation level as needed. We also encouraged participants to record their BP, motivation, and assigned task (including whether they completed it) daily in a journal provided by the research team. Participants who did not provide their BP and motivation levels by 5:00 pm Pacific Time were sent a reminder by the research assistant.

Control Group

The control group completed all processes except the 28-day intervention. After the educational session, the control group participants completed the height, weight, and sociodemographic questionnaire; BP assessment; ASA-24; and pre-test HK-LS. Twenty-eight days later, these participants attended an exit meeting to complete the post assessment (i.e., height, weight, BP, ASA24, and post-test knowledge).

Data Analysis

We calculated descriptive statistics for all variables and compared baseline characteristics between intervention and control groups using Fisher’s exact test or chi-square analysis as appropriate. We also calculated pre-to-post comparisons for both groups using paired t-tests for continuous variables and exact McNemar-Bowker tests for categorical variables. To examine potential differences in outcomes for control and intervention groups, we used a repeated measures ANOVA for pre-to-post time periods and included age as a covariate in the models.

Results

Descriptive Characteristics at Baseline

With only six months of recruitment, 30 participants were enrolled, but only 29 participants completed the analysis with one participant identified later to not meet all the inclusion criteria. Therefore, twenty-nine participants enrolled in the trial (intervention: n = 15; control: n = 14), and we compared baseline characteristics between the intervention and control groups. Participants were 19 to 29 years old, averaging 24.1 years (SD = 3.04) for the intervention group and 21.7 years (SD = 2.40) for the control group. Sixty percent of intervention participants were female compared to 50% of the control group. Race and ethnicity were asked separately. In the intervention group, 53.3% of participants were white, 26.7% mixed or other, and 20% Asian, whereas the control group was 42.9% white, 35.7% Asian, 14.3% mixed or other, and 7.1% black. In both groups, approximately 40% were Hispanic or Latino, more than 73% were single, and more than 80% had insurance. In terms of health, 26.7% in the intervention group reported a family history of heart disease compared to 57.1% in the control group. We did not observe any significant difference in the above characteristics; however, the intervention group had a significantly higher percentage of graduate-level participants (60%) than the control group (P = .001).

Clinical and Other Data at Baseline

Data from the baseline assessments evaluated multiple variables to capture BMI (kg/m²), BP, caloric intake, sodium intake, and hypertension knowledge. At baseline, the mean BMI was 28.7 (SD = 7.53) and 29.0 (SD = 6.04) for intervention and control groups, respectively. At baseline, 46.7% of intervention and 28.6% of control participants were classified as overweight, 26.7% of intervention and 42.9% of control participants as obese. The mean BP readings at baseline (based on at least 2 readings) were 131.2 mm Hg (SD = 11.30) for SBP and 80.6 mm Hg (SD = 8.70) for DBP for the intervention group and 125.5 mm Hg (SD = 12.04) for SBP and 80.9 mm Hg (SD = 9.01) for DBP for the control group. Using American Heart Association classifications,³ 40.0% of intervention and 28.6% of control participants had elevated BP, 40.0% and 42.9% hypertension stage I, and 20% and 28.6% hypertension stage II, respectively. The mean caloric intake (kcal) reported using the ASA24 recall at baseline was 2446.5 (SD = 1425.86) for the intervention group and 1874.2 (SD = 752.16) for the control. The mean sodium intake reported at baseline was 3955.4 (SD = 2530.53) for the intervention group and 3765.5 (SD = 1953.08) for the control. Lastly, the mean hypertension knowledge scores at baseline were 17.9 (SD = 3.07) for the intervention group and 17.1 (SD = 2.48) for the control. There were no differences in measurements between intervention and control groups at baseline (all P > .05).

Clinical and Other Data Post Intervention

We collected post-intervention data 28 days following baseline data collection. The mean BMI was 28.8 (SD = 7.19) and 29.0 (SD = 6.09) for intervention and control groups, respectively; 46.7% of intervention and 14.3% of control participants were classified as overweight, 26.7% of intervention and 50.0% of control participants as obese. The mean BP readings after 28 days were 118.7 mm Hg (SD = 11.32) for SBP and 73.7 mm Hg (SD = 7.60) for DBP for the intervention group and 124.0 mm Hg (SD = 12.31) for SBP and 77.9 mm Hg (SD = 7.90) for DBP for the control group. In AHA classification terms, 66.7% of intervention and 28.6% of control participants had normal BP, 6.7% and 14.3% elevated BP, 20.0% and 50.0% hypertension stage I, and 6.7% and 7.1% hypertension stage II, respectively. The mean caloric intake reported post assessment was 1977.7 (SD = 1053.72) for the intervention group and 1631.7 (SD = 631.13) for the control. The mean sodium intake reported post assessment was 3799.4 (SD = 2540.81) for the intervention group and 3046.2 (SD = 1095.85) for the control. The mean hypertension knowledge scores were 19.3 (SD = 1.35) for the intervention group and 19.1 (SD = 1.54) for the control.

Primary and Secondary Outcomes: BP, Sodium Intake, and Hypertension Knowledge

We examined the MOBILE intervention’s impact on BP reduction along with sodium intake and hypertension knowledge improvement after 28 days (Table 1). In the repeated measures ANOVA models, age was not a significant covariate (all P > .05), except for the total HK-LS score (P = .011). The intervention group’s BP decreased significantly (P = .001) compared to the control, which led to a significant time-by-group interaction effect (P = .011). Using the ASA24, we found no statistical significance in sodium intake for intervention or control groups after 28 days. Hypertension knowledge for both groups improved after 28 days but was only significant for the control group (P = .001). A larger sample size may further explain this relationship.

Table 1.

Comparison of Pre- and Post-Clinical Data for Intervention and Control Groups^a

Variables	Intervention Group			Control Group
	n = 15				n = 14
	Pre	Post	P valueb	Pre	Post	P valueb
Continuous Measures
BMI	28.7 (7.53)	28.8 (7.19)	.859	29.0 (6.04)	29.0 (6.09)	.774
Systolic Blood Pressure, mmHg	131.2 (11.30)	118.7 (11.32)	.001**	125.5 (12.04)	124.0 (12.31)	.638
Diastolic Blood Pressure, mmHg	80.6 (8.70)	73.7 (7.60)	.001**	80.9 (9.01)	77.9 (7.90)	.188
Caloric Intake, cal	2446.5 (1425.86)	1977.7 (1053.72)	.128	1874.2 (752.16)	1631.7 (631.13)	.305
Sodium Intake, mg	3955.4 (2530.53)	3799.39 (2540.81)	.539	3765.5 (1953.08)	3046.2 (1095.85)	.273
HK-LS Score	17.9 (3.07)	19.3 (1.35)	.066	17.1 (2.48)	19.1 (1.54)	.001**
Categorical Measures
BMI, kg/m2c			1.000			.995
Underweight	0 (0)	0 (0)		1 (7.1)	1 (7.1)
Normal or Healthy Weight	4 (26.7)	4 (26.7)		3 (21.4)	4 (28.6)
Overweight	7 (46.7)	7 (46.7)		4 (28.6)	2 (14.3)
Obese	4 (26.7)	4 (26.7)		6 (42.9)	7 (50.0)
Blood Pressure‡			.016			.197
Normal	0 (0)	10 (66.7)		0 (0)	4 (28.6)
Elevated	6 (40.0)	1 (6.7)		4 (28.6)	2 (14.3)
HTN Stage I	6 (40.0)	3 (20.0)		6 (42.9)	7 (50.0)
HTN Stage II	3 (20.0)	1 (6.7)		4 (28.6)	1 (7.1)

^aData shown as N (%) or mean (SD).

^bP values represent pre-to-post comparisons within groups using either a 2-sided paired t-test or McNemar’s test.

^cRefer to the methods section for the BMI and blood pressure categories.

^**Correlation is significant at the .01 level (2-tailed).

Discussion

This study evaluated the feasibility of the MOBILE intervention in full-time college students with elevated BP or undiagnosed hypertension. We found that the intervention group experienced a significant reduction in BP compared to the control. At the educational session, zero participants had normal BP levels. After 28 days, more than half of the intervention participants had normal BP while only 28.6% of the control group had normal BP. Despite the short study period, the intervention group results demonstrate the feasibility and practicality of the MOBILE intervention for reducing BP in this population. Our findings are comparable to Saptharishi et al.,¹⁵ who implemented a similar intervention for 8 weeks and found a significant decrease in BP in the intervention group but not the control. Similarly, Nidich et al. found using transcendental meditation with college students significantly reduced BP in the intervention group while BP increased in the control.¹⁶

While there were no significant differences between the sodium and caloric intakes of study groups, these intakes decreased for both groups over 28 days. This finding may be explained by the participants’ increased awareness of their consumption habits after the educational session and/or the Hawthorne Effect—that participants alter their behaviors while being monitored.¹⁷ Based on the post-intervention survey, both groups reported decreased sodium intake, increased water intake, and increased fruit and vegetable consumption during the study. Future investigation is needed to determine the actual effect of the independent variables on behavior change.

We observed no weight change in the intervention group, but participants were able to maintain weight for 28 days. In contrast, one control group participant initially categorized as overweight was obese at post data collection. Similarly, West et al. implemented a 9-week intervention focusing on weight gain prevention using technology to modify diet in 58 undergraduate students and found no significant differences between the intervention and control groups; instead, both groups maintained their weight.¹⁸ Buscemi et al. completed a brief motivation intervention (1 educational session,1 follow-up phone call) on obese college students for 3 months and found this intervention was insufficient to encourage weight loss; the participants lost between ½ and 1 kg in weight.¹⁹ Likewise, Hayes et al. implemented a 4-week intervention focusing on behavioral modifications using text messaging and frequent educational sessions and found a minimal decrease in weight loss for all groups.²⁰ Despite the lack of significant weight change observed in the study groups, the MOBILE intervention may help determine whether weight loss and obesity prevention are possible using this approach. By preventing additional weight gain and potentially assisting students in developing healthy eating habits, these positive changes can impact heart health and, most importantly, prevent hypertension and CVD. Further studies with extended intervention periods and intentional weight loss outcomes are warranted to ascertain the actual effect of the MOBILE intervention.

Interestingly, the control group, but not the intervention group, increased its average hypertension knowledge after 28 days. One explanation is intervention group participants may have relied on the research team to feed them this information daily. It is also possible the control group was willing to seek extra knowledge, which could explain the significant increase in this group. Another possibility is that the participants have greater interest in improving their HTN knowledge due to a higher percentage of (57.1%) of family history of heart disease compared to the intervention group. Further future trials will be able to clarify this point.

Of interest, previous research provides mixed messages regarding how much involvement is appropriate to encourage behavior change in college students. Napolitano et al. implemented private groups on social media in combination with weight loss interventions, showing students had a positive experience engaging with the research team via the private group and receiving morning text message reminders and tailored weekly progress reports.²¹ In contrast, Kanstrup et al. evaluated a weight loss treatment program in overweight college students and learned many students preferred to monitor their dietary and physical activity monitoring infrequently.²² The students found the digital health tracker burdensome, anFd when they did not meet their goals, they experienced decreased motivation and a sense of failure.²²

Limitations

This study has several limitations. The lack of effects for sodium and caloric intake for both groups is likely due to our data collection method. We only collected ASA24 data pre and post study, which may be insufficient to examine changes and eating patterns; however, this approach was more suitable for this population because it was non-invasive and convenient. We also found BP decreased in the intervention group despite no significant changes in caloric and sodium intakes. Physical activity may have influenced BP levels in the intervention group; however, we did not monitor the participants’ physical activity. Our MOBILE intervention was also short and the sample size small, but the effectiveness of the intervention in reducing the BP for intervention group participants compared to the control group suggests that expanding the program in duration and sample size will be beneficial. We did not examine sustainability as this study assessed and evaluated the feasibility of the intervention for this population.

Conclusions

The MOBILE intervention is the first randomized controlled trial to evaluate the feasibility of reducing BP in college students. This pilot study highlights the importance of engaging full-time college students with elevated BP or undiagnosed hypertension and the potential efficacy of the MOBILE intervention. The findings, which demonstrate promising results for BP-reducing approaches, warrant further examination concerning this intervention and its long-term effects.