Texting and Educating on Lifestyle Therapy: A Pilot Study in a Spanish-Speaking Population in North Carolina

,


Introduction
The Hispanic/Latinx population comprises the largest minority group in the United States; the majority of this population speaks Spanish at home. 1,2,3When compared to the non-Hispanic white population, this ethnic group is at a greater risk for cardiovascular disease, likely due to various cardiovascular risk factors such as uncontrolled hypertension, diabetes, obesity, and smoking. 1,2These elements have led heart disease to be the leading cause of death in this population.A study published in 2012 by the Journal of the American Medical Association showed that 25% of Hispanic/Latino men and 24% of Hispanic/Latina women have hypertension, but only 15% reported taking prescribed anti-hypertensive medications. 1,4The Hispanic/Latinx population also has a higher risk for type 2 diabetes; compared to 11.3% of non-Hispanic/Latinx population, 22.6% of the Hispanic/Latinx population has a diagnosis of type 2 diabetes.Moreover, death related to diabetes was 51% higher in the Hispanic/Latinx population than in the non-Hispanic/Latinx white population. 2,5The prevalence of hypertension and diabetes, as well as the increased prevalence of diabe-tes-related death, highlight the need for better management of cardiovascular risks in the Spanish-speaking population.
Lifestyle modifications such as exercising, weight loss, and consumption of a healthy diet have been proven effective in decreasing the risk of developing type 2 diabetes and improving blood pressure control. 6,7However, a study conducted in North Carolina showed that only 37.4% of Spanish-speaking participants met recommendations for physical activity, which could be associated with increased triglycerides, insulin resistance, and poor control of hypertension. 8,9Furthermore, as the median income for the Hispanic/Latinx population is substantially lower than their non-Hispanic counterparts, the typical diet may include more processed and high-calorie foods. 10,11Feeding America, a nonprofit charity organization composed of a nationwide network of food banks that feed underserved populations, reports that 17% of Hispanics in the United States participate in the program annually, compared to 10% of the white non-Hispanic population. 12Lack of physical activity and access to healthy food pose as barriers to optimal control of cardiovascular risks in this population. 13n addition to lifestyle habits, the Hispanic/ Latinx population reports low self-confidence in managing cardiovascular risks due to poor health literacy and lower socioeconomic status. 2,14A retrospective study showed that Spanish-speaking patients with type 2 diabetes were less likely to understand healthcare instructions such as prescription labels, which was shown to adversely affect glycemic control. 14The cost of managing cardiovascular risks is an additional barrier.Low income and a lack of health insurance were identified as primary reasons for Hispanic/Latinx patients not seeking medical care upon initial diagnoses of diabetes or hypertension. 2,15Resources in the Spanish language, cost-effective interventions reinforcing motivation for exercise, and providing tips on incorporating healthier foods may improve health outcomes by providing more accessible and understandable information.
Several organizations in North Carolina provide preventative healthcare to Spanish-speaking populations, including, but not limited to, blood pressure and blood glucose screenings.Our organization, the Student Health Action Coalition (SHAC) Outreach, is a student-led interdisciplinary organization at the Eshelman School of Pharmacy at the University of North Carolina at Chapel Hill.SHAC Outreach provides healthcare services to underserved populations, including the Spanish-speaking congregation of a local Catholic church.SHAC Outreach provides monthly screenings with the goal of promoting awareness and increasing access to preventative care; however, more continuous support may be more effective in longitudinal health management. 15In an effort to improve health outcomes within the population we serve, the SHAC Outreach team began to explore the utility of mobile text messaging as a means to provide more frequent health coaching.
As access to mobile phones is becoming increasingly commonplace, recent studies have shown the success of cellular interventions in underserved populations through the reinforcement of lifestyle modifications. 16,17,18In particular, the TEXT ME study found that sending four text messages per week for six months to an underserved population with coronary heart disease significantly decreased low-density lipoprotein levels, body-mass index, and systolic blood pressure while significantly increasing physical activity. 17ur study, which was modeled after the TEXT ME study, aimed to evaluate the effectiveness of using cellular communication to enhance health outcomes in a community clinic situated in a Spanish-speaking Catholic church in suburban North Carolina.The primary objective of this study was to determine the impact of an educational text messaging program on participants' blood pressure and confidence in managing their own health.A secondary objective was to evaluate participant interest in continuing the program beyond the study.

Methods
This Institutional Review Board-approved prospective cohort study was conducted at a local Catholic church in Chapel Hill, North Carolina.Study participants were recruited through announcements at a Spanish-speaking Catholic mass for six consecutive weeks.The following inclusion criteria were used: age of 18 years or older and ownership of a mobile phone with unlimited text messaging (as assessed through interview with the participant).Those without unlimited text messaging were excluded in the study to avoid placing a financial burden on participants.Participants provided consent prior to study enrollment.At the time of enrollment, participants completed a 7-item investigator-developed survey in either Spanish or English to self-report gender, age, diagnosis of type 2 diabetes mellitus (T2DM), hypertension (HTN), and/or hyperlipidemia (HLD), and preference for English or Spanish text messages.The pre-survey also assessed the frequency with which participants had difficulty understanding medical conditions and needed help reading prescriptions on scales of 1 to 5 (with 1 indicating never and 5 indicating always), as well as their confidence in self-management of their health on a scale of 1 to 10 (with 1 indicating least confident and 10 indicating most confident).The investigators manually measured and recorded each participant's baseline sitting blood pressure after participants rested for at least 5 minutes.
Following a rolling enrollment strategy, participants immediately began their 20-week intervention at the time of informed consent.The investigators sent two text messages per week using a secure texting platform, EZ Texting.Text messages informed the participants about simple lifestyle modifications.Investigators developed the recommendations using resources from the American Heart Association, American Society of Hypertension, National Institute of Health, and the National Heart, Lung, and Blood Institute (Appendix 1).For example, one text message read, "Whole grains are better for your heart, try whole wheat instead of white bread!" (Figure 1).All participants received the same messages in the same order, regardless of enrollment date.For example, the first message was always sent on the Tuesday following enrollment at 2:17pm.After the 20-week intervention, participants repeated the initial survey with three additional questions.The additional questions assessed the frequency with which participants read the messages on a scale of 1 to 3 (with 1 indicating never and 3 indicating always), the helpfulness of the messages on a scale of 1 to 10 (with 1 indicating not helpful and 10 indicating very helpful), and the desire to continue receiving messages (indicated with a yes or no).The investigators manually measured and recorded each participant's final sitting blood pressure after participants rested for at least 5 minutes.Survey questions with scaled responses were assessed with the Siegel-Tukey test; blood pressure readings were assessed using the Wilcoxon signed-rank test.

Results
Thirty-three participants enrolled in the study.One participant elected to withdraw from the study prior to initiation of the intervention.Of the 32 participants remaining, 97% (n=31) requested to receive Spanish-language messages.All study participants completed the 7-item pre-survey.Demographic data is found in Table 1.Participants of the study were 61% female (n=21) with a mean age of 48 years.Self-reported cardiovascular risk factors included 25% with T2DM (n=8), 22% with HTN (n=7), and 28% with HLD (n=9).
At the conclusion of the study, 7 persons completed the post-survey; 25 persons were lost to follow-up.Participants were sent a maximum of four messages as reminders to complete a post-survey.The median baseline score for participant difficulty understanding medical conditions was 2 and remained 2 upon study completion (1 indicating never, 5 indicating always).The median baseline score for participant needing assistance reading prescriptions was 2 and increased to 3 upon study completion (1 indicating never, 5 indicating always).The median baseline score for participant confidence managing health was 7 and remained 7 upon study completion (1 indicating least confident, 10 indication most confident).Of participants who completed both surveys at baseline and upon study completion, the mean systolic blood pressure decreased by 13 mmHg (95% confidence interval [CI] -27 to 0.7 mmHg, p=0.06) while the mean diastolic blood pressure decreased by 1 mmHg (95% CI -8 to 7 mmHg, p=1.00).
The median score for how often messages were read was 3 (1 indicating never, 3 indicating always).The median score in helpfulness of the messages was 9 (1 indicating least helpful, 10 indicating very helpful).Of participants who completed both preand post-surveys, a total of 6 persons (86%) expressed desire to continue the text-messaging program beyond study completion.

Discussion
The baseline characteristics of our study population are similar to those reported in national studies, especially in terms of the percentage of the population with cardiovascular risks such as T2DM and HTN.However, the mean baseline blood pressure of this population was within normal limits, which may indicate that this population is appropriately managed with medications.Future health interventions therefore should survey specific medications taken by participants to ensure that lifestyle modifications are combined with optimal medication management.Additionally, this study suggests that interventions should be conducted in a patient's native language, as the majority of participants elected to receive the intervention in Spanish.
A decrease in mean systolic blood pressure by 13 mmHg was found at the conclusion of the text messaging intervention.While blood pressure may fluctuate based on various factors including the consumption of caffeine, recent exercise, sitting with legs crossed, and changes in equipment or personnel involved, the marked decrease in mean systolic blood pressure may indicate that text messages reinforcing lifestyle changes make a positive impact on blood pressure control. 19As the study design did not control for the effect of any outside factors (new medications, etc.), the intervention cannot definitively take credit for the demonstrated blood pressure lowering.Future studies could improve upon this data by increasing the number of blood pressure readings obtained per participant, increasing the number of participants who return for follow-up, standardizing the measurements, and adding a control group.Furthermore, other markers of increased risks for heart disease may also be measured such as pre-and post-cholesterol levels as well as atherosclerotic cardiovascular disease risk status.
A primary objective of this study was to determine the impact of a text-messaging program on participants' confidence in managing their own health.The data shows no change after the intervention, with the mean confidence score remaining at 7/10 (with 10 being most confident).The results may be due to the small sample size and the method of measurement.Increasing the size of the study and improving rate of follow-up would allow for a more accurate approximation of the intervention's effect on confidence in managing health.Refining the questionnaire to include questions beyond the basic "How confident are you in managing your own health?"would improve the study's ability to demonstrate effect in this area.Additionally, this study only looked at the participants' confidence in maintaining their health and did not evaluate their retention or application of lifestyle modification information beyond blood pressure measurements.To further improve participant confidence, future studies should focus on individualizing the messages by identifying and targeting participants' specific health management needs.
The intervention received very positive feedback from participants who followed-up.Patient interest in and enthusiasm for the management of his or her health is a crucial step in creating a successful lifestyle modification program. 20There was appreciable interest in continuing to receive text messages beyond the end of the study, and no interested participants were excluded from the study due to a lack of unlimited texting.This indicates that future text messaging interventions in this Spanish-speaking population will be both feasible and acceptable.
A limitation of this study was the substantial percentage of patients who did not complete follow-up procedures despite multiple text message reminders, which may have introduced a participation bias.One possible explanation includes the lack of incentives to complete the post-survey.Because we made no contact beyond the biweekly messaging during the 20-week intervention, participant interest in measuring their progress (or in the text messages) may have waned.Incorporating incentives may be able to increase follow-up.
In conclusion, this study is the first to investigate the effect of a Spanish-language text messaging platform on the blood pressure and confidence in personal health maintenance of a Spanish-speaking population in suburban North Carolina.Despite an unchanged confidence in managing personal health, participant interest for the program, combined with the low financial burden, indicate that a texting intervention is a feasible method for student-run clinics to provide more frequent health coaching for Spanish-speaking populations.

Figure 1 .
Figure 1.Example of text messages received by participants

Table 1 .
Baseline characteristics N = 32; SD: standard deviation Journal of Student-Run Clinics | Texting and Educating on Lifestyle Therapy: A Pilot Study in a Spanish-Speaking Population in North Carolina Create a rewards system that helps motivate you to live a healthy lifestyle.Treat yourself to an activity that you enjoy after you accomplish an exercise goal.Cree usted un sistema de recompensa que le motiva a vivir una vida saludable.Premíese con una actividad que usted goza después de lograr una meta de ejercicio journalsrc.org | J Stud Run Clin4;1 | 11 NA NIH: National Institutes of Health; ASHP: American Society of Health-System Pharmacists; NA: Not Applicable