Abstract
This article describes the impact of an 8-week community program implemented by trained volunteers on the hypertension self-management of 185 patients who were batch randomized to intervention or wait-list control groups. Compared with control group participants, a higher proportion of treatment group participants moved from the cognitive to behavioral stages of motivational readiness for being physically active (P < .001), practicing healthy eating habits (P = .001), handling stress well (P = .001), and living an overall healthy lifestyle (P = .003). They also demonstrated a greater average increase in perceived competence for self-management, F(1.134) = 4.957, P = .028, η2 = .036, and a greater increase in mean hypertension-related knowledge, F(1.160) = 16.571, P < .0005, η2 = .094. Enduring lifestyle changes necessary for chronic disease self-management require that psychosocial determinants of health behavior are instilled, which is typically beyond standard medical practice. We recommend peer-led, community-based programs as a complement to clinical care and support the increasing health system interest in promoting population health beyond clinical walls.
Introduction
The project described in this article, Expanded Health Coaches for Hypertension Control (EHCHC), is closely aligned with the Healthy People 2020 goal: HDS-12—Increase the proportion of adults with hypertension who have it under control from 43.7% to 61.2%, as it addresses the predictors of hypertension self-management behaviors such as self-efficacy and stage of readiness for change.1,2 EHCHC demonstrates how primary care practices can collaborate with a community-based program to effectively promote hypertension self-management of those aged 45 years and older. We hypothesized that participants in the treatment group would demonstrate statistically significant improvements in hypertension self-management knowledge and self-efficacy, and readiness to improve health behaviors compared with the wait-list control participants.
Methods
Theoretical Framework
EHCHC is guided by the chronic care model, which was developed in response to the Institute of Medicine report, “Crossing the Quality Chasm” calling for improvements in patient care. A fundamental tenet of the chronic care model is empowering patients to take an active part in their care. 3 The model identifies the community, the health system, and self-management support as essential for effective disease management. In EHCHC, trained community volunteers mentored those self-referred or referred by primary care practices in building self-management skills and connecting participants to resources in their local community.
Community Health Workers
Community health workers (CHWs) have proven effective in improving patient adherence with chronic disease management recommendations by providing a trusting and culturally appropriate extension of the provider-patient relationship, a leading predictor of adherence. 4 Specifically, research has shown that CHWs are effective in improving hypertension management and health outcomes.5,6 Our CHWs are trained, community volunteers called Health Coaches, based on a preference for this name that emerged from focus groups conducted before our first Health Coaches project 5 and which we continued to use in our most recent project. 7 EHCHC also builds on our work to develop culturally appropriate lay-led physical activity programs for rural areas. 8 These projects considered health literacy challenges of our population and all materials were written at a fifth-grade level with a heavy reliance on pictures. Development of our original hypertension control project, which was adopted by a local program targeting low-income, underinsured, or uninsured populations, is more fully described elsewhere. 7 Through various communication channels, we recruited community members to be trained as volunteer Health Coaches. Those who passed the criminal background check, achieved at least an 80% on a knowledge test based on the training program, and demonstrated effective small group teaching skills, were assigned to lead a series of 8 weekly small-group classes lasting for 90 minutes each. The 6 female and 4 male Health Coaches were all white and older than 50 years, which closely matched the demographic profile of our participants (see Table 1) and residents of the county of interest where 89.5% of residents are white.
Participant Demographics and Cardiac-Related Health Issues.
Self-reported diagnosis of specific hypertension-related health issue(s), including diabetes, stroke, congestive heart failure, myocardial infarction, and kidney disease.
Program Participants
Nationally, South Carolina is 1 of 7 states with the highest prevalence rates of hypertension with an increase from 26.8% in 1997 to 33.2% in 2008. This rate is higher than the nationwide median of 27.0 % and the Healthy People 2020 target objective of 26.96%. 2 Remarkably, the rate of hypertension in the intervention site, Oconee County, located in the Appalachian upstate of South Carolina, is 34.6% which is even higher than the state average. This rate could be due to the fact that Oconee County residents have higher rates of tobacco and alcohol use and are more obese than their state and national counterparts. These behaviors could reflect the fact that residents have a lower percentage of high school and/or college graduates, lower household income levels, and a higher poverty level as compared with the state overall. 9
Participant recruitment strategies, which were supported by other studies10 -12 included using a variety of community channels with opportunities for self-referral, as well as increasing physician referrals by making individual presentations to physician practice staff and sending messages to physicians and patients through hospital communication channels used by the CEO and Wellness Center Director. Our most effective recruitment strategy was a mail-out to all patients from a physician practice group recommending participation in the EHCHC program if the patient had been diagnosed with hypertension.
As referrals were received, the Project Coordinator determined eligibility based on age of 45 years or older, hypertension diagnosis, and residence in Oconee County. Participant responses to an item on the Health Risk Appraisal (HRA) used in the study verified that 99.93% had been diagnosed by a physician to have hypertension. Because we recruited participants over time, we did not have a complete list of eligible participants prior to randomization. After the eligibility screening, the Project Coordinator employed a batch randomization procedure to assign ~20 participants at a time to a condition (intervention or wait-list). 11 Tracking of batches and assignment were documented over time to check equivalence of groups on important factors such as gender, age, and race/ethnicity.
EHCHC Educational Modules
Using a scripted manual, the Health Coaches provided 8 core EHCHC modules consisting of the following: Development of Action Plans and Monitoring Behaviors with a Personal Health Diary; Basics of Hypertension Control; Nutrition (which includes weight control); Physical Activity (which includes weight control); Tobacco Use Cessation; Stress Management; Medication Management; and Development of a Long-term Action Plan to continue hypertension control activities post-intervention. Participants were provided with a notebook containing session activities and additional information as well as supplies such as a blood pressure monitor, pedometer, cookbook, and relaxation CD. Session activities focused on building self-efficacy to perform specific behaviors such as measuring blood pressure, incorporating more physical activity in daily lives as measured by a pedometer, modifying favorite recipes or using those featured in the provided cookbook, and using the visualization CD to promote relaxation. The program focused on building self-efficacy as this perception is critical to moving from consideration of a behavior change to initiating the change.
In the first implementation of this project, we conducted a process evaluation to assess program completeness, fidelity, and participant satisfaction. 7 Based on process evaluation findings, we refined the original program by lengthening some sessions and adding more flipchart pictures during session discussions. We report the results of implementation of the refined program in this article.
Measures
Data were collected at baseline from the intervention and wait-list groups and measures were repeated at the end of each 8-week intervention period. The internal review board of Clemson University approved all aspects of the study including data collection and storage protocols. Measures included an HRA, the Personal Wellness Profile (PWP) by Wellsource, Inc, and surveys used in our previous studies7,13 that measure psychosocial variables of knowledge and self-efficacy for hypertension self-management, and motivational readiness for change, in addition to fasting clinical measures of systolic and diastolic blood pressure, weight, waist circumference, total cholesterol, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, triglycerides, and glucose.
The PWP HRA has been certified by the National Committee for Quality Assurance (NCQA) as a high-quality HRA. Fitness items are based on the American College of Sports Medicine (ACSM) and the American Heart Association (AHA) guidelines. Coronary risk factor items are based on National Heart, Lung, and Blood Institute (NHLBI) guidelines, and diabetes risk factors are based on American Diabetes Association’s guidelines.
A pool of knowledge survey items was developed by project staff after reviewing NHLBI and AHA materials. Survey items were then evaluated for health literacy appropriateness and selected through consensus of staff and Health Coaches. After pilot-testing, the final survey consisted of 17 items that included 2 to 5 items, each with 1 correct response and 3 distracters, related to program content in hypertension, physical activity, nutrition, stress management, tobacco use, and medication management.
The Perceived Competence Scale is a short 4-item scale, based on constructs from self-determination theory, and designed to assess feelings of competence at carrying out a treatment regimen. Response options range from 1 = “Not at all true” to 7 = “True” and the alpha reliability for the items in a combined analysis across 4 items is .90. We adapted the Perceived Competence Scale for hypertension self-management and have used this scale and others in previous projects.7,13
Statistical Methods
All analyses were conducted using IBM SPSS Statistics Version 23. 14 Descriptive statistics were used to examine the distribution of continuous and categorical variables. McNemar’s Test was used to examine differences in the proportion of participants in cognitive versus behavioral stages of motivational readiness over time. Mixed analysis of variance was used to examine group differences (between-subjects factor) in knowledge, perceived competence for hypertension self-management, and clinical values over time (within-subjects factor). An apriori power calculation was computed using G*Power 3.1.2. 15 For a repeated-measures analysis of variance test, with error probability (α) fixed at .05, power (1 − β) fixed at 0.80, assuming a correlation of 0.5 among repeated measures, and with an estimated small-to-medium (f = 0.15) effect size, a total sample size of 74 was determined to be necessary to detect differences, including within-between interactions (group * time).
Results
Study participants were predominantly female (67.6%) and Caucasian (87.0%). Approximately half had no comorbidities and about one-quarter had been diagnosed with one or more conditions, including congestive heart failure (4.0%), kidney disease (6.1%), stroke (7.3%), myocardial infarction (9.3%), and diabetes (18.2%).
Of the 185 participants, 162 (89.2%) completed the knowledge test, 136 (73.5%) completed the perceived competence for hypertension self-management measure, and 112 (60.5%) to 121 (65.4%) completed the physical activity motivational readiness items, at both baseline and posttest. Comparisons between completers and noncompleters revealed no statistical differences in gender, age, race/ethnicity, and number of comorbidities.
During the study, there was a higher proportion of treatment group participants moving from the cognitive to behavioral stages of motivational readiness for being physically active (P < .001), practicing healthy eating habits (P = .001), handling stress well (P = .001), and living an overall healthy lifestyle (P = .003).
We observed a group by time interaction for perceived competence for hypertension self-management, F(1.134) = 4.957, P = .028, η2 = .036. Specifically, treatment group participants demonstrated a greater average increase in perceived competence for self-management compared with control group participants during the study. For the perceived competence in hypertension self-management scale, Cronbach’s alpha coefficient ranged from .841 at baseline to .911 at posttest indicating good to excellent internal consistency. In addition, there was a group by time interaction for hypertension-related knowledge, F(1.160) = 16.571, P < .0005, η2 = .094. Specifically, treatment group participants demonstrated a greater increase in mean hypertension-related knowledge compared with control group participants (Figure 1). Omnibus (multivariate) tests for changes in clinical values revealed no group by time interaction (P = .057) and no main effect for group (P = .569); however, a main effect for time (P < .001) was observed. Univariate analyses pinpointed small changes in systolic blood pressure (P = .001), diastolic blood pressure (P = .018), weight (P = .048), waist circumference (P < .001), and high-density lipoprotein cholesterol (P = .036) for both the treatment and control groups. Although not statistically significant, mean changes from baseline to 8 weeks postintervention are of greater magnitude in the treatment group and some changes may be clinically significant (Table 2).
Motivational Readiness Changes From Baseline to 8 Weeks.
Answers to stage of change items were dichotomized into a cognitive (Early Stage) category, where Precontemplation = “I haven’t thought about changing,” Contemplation = “I plan to change in the next 6 months,” Preparation = “I plan to change this month”) versus a behavioral (Late Stage) category, where Action = “I recently started doing this,” Maintenance = “I do this regularly for the past 6 months”). Table 2 values are presented as number and percent in each stage of change category at each measurement time point. Exact 2-tailed P values are based on McNemar’s test.
Pre-post changes in self-competence for hypertension control and hypertension-related knowledge.
Discussion
Optimal population health promotion requires more than clinical care and pharmacological intervention. Enduring lifestyle changes necessary for chronic disease self-management require that psychosocial determinants of health behavior, such as self-efficacy and readiness to change health behaviors, are addressed, which is typically beyond standard medical practice. We recommend peer-led, community-based programs as a complement to clinical care and support the increasing health system interest in promoting population health beyond clinical walls.
Primary care providers can greatly enhance patient outcomes by partnering with CHWs, called Health Coaches in our project. 16 In our study, we found that local retirees were a rich source of Health Coaches. The retirees enthusiastically committed their time and energy to helping others in their community and expressed the benefits they experienced from meaningful civic engagement. Another example of a community that capitalized on local human resources to augment efforts of primary care is found in Wisconsin where members of Veterans of Foreign Wars posts were trained to be peer health educators to members of 14 posts. 17
In an intervention including education, skill acquisition, peer mentoring, social support, and home blood pressure monitoring, it is difficult to determine which component contributed to what percentage of the improvement in readiness for physical activity and weight management, and perceived competence for hypertension self-management. We agree with others18,19 that home blood pressure monitoring is important to effective self-management. Self-monitors have fewer office visits and lower costs per year than those receiving usual care. 20 A Cochrane review 21 found that self-monitoring was associated with reductions in systolic and diastolic blood pressure. In our study, Health Coaches provided instructions and observed participant skill in using an Omicron blood pressure monitor that was provided to each person along with a Personal Health Diary in which to record daily blood pressure readings. Studies have concluded that encouraging patients to become active participants in their care and providing them with skills and confidence for active self-management improves hypertension control.22,23
We believe our combination of peers providing education, skills, and social support to self-monitor blood pressure, follow a DASH (Dietary Approaches to Stop Hypertension) dietary plan, increase physical activity, manage stress, manage medications, and end tobacco use increased participant self-efficacy to better manage their hypertension and readiness to change physical activity, manage stress, and manage weight. Warren-Findlow et al 22 found that hypertension self-management self-efficacy was strongly associated with adherence to 5 of 6 prescribed self-care activities among African Americans with hypertension. The Community Outreach and Cardiovascular Health (COACH) trial found that CHWs were effective in improving participants’ hypertension control further supporting the use of community-based education and outreach to supplement standard therapy for hypertension and cardiovascular risk. 24
Cost-Effectiveness
A US national probability sample survey, the Medical Expenditure Panel Survey, found that in 2001 the mean incremental annual per capita direct cost for a hypertensive individual was $1131 with prescription medicines, inpatient visits, and outpatient visits accounting for more than 90% of expenditures. 25 The cost of providing the community-based educational program described in this paper to an individual is $180 for materials and supplies. The yearly salary of $40 000 for a full-time program coordinator to recruit, train, and direct 5 volunteer community Health Coaches who can serve 200 individuals each year adds $200 to individual costs for a total of $380 per person. We are not suggesting that a community-based, peer-led program can substitute for clinical care, but we have found our program improved predictors of hypertension self-management behaviors of those already under a physician’s care and thus would add value to clinical care. Improving hypertension self-management behaviors could contribute to reduction in mortality due to stroke and coronary heart disease which can be impacted by even small changes in blood pressure of 2 to 3 mm Hg. 26
Limitations
Our study attrition rate was 22%, but when examining differences between those who participated in 8-week postintervention measures and those who did not, we found that there were no statistical differences in demographic or comorbidity variables. We also found that our rate compared favorably to other studies of hypertension self-management with attrition rates of 30%. 17 Another limitation of the study is single measure/item response rates at follow-up, which were reasonable for the knowledge test and perceived competence scale but were lower for the motivational readiness items. Regardless, the study still had adequate power for our planned comparisons and analyses.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The project described in the article was funded by the United States Department of Agriculture–National Institute of Food and Agriculture Award: 2012-46100-20122.