Self‐Monitoring With Coping Skills and Lifestyle Education for Hypertension Control in Primary Care

ABSTRACT

Self‐monitoring with support, lifestyle modifications, and emotion management improves blood pressure (BP). Patients with hypertension need continual support to modify behaviors, but time pressures limit lifestyle education in primary care settings. Using mixed methods, we aimed to study the feasibility and acceptability of an innovative 6‐week program that combined self‐monitoring with coping skills and lifestyle education for patients with uncontrolled hypertension. Patients with uncontrolled hypertension interested in lifestyle modifications before intensifying medications were enrolled from primary care clinics. Patients self‐monitored emotions, behaviors, and BPs and received education from medical providers and mind‐body therapists through shared medical appointments (SMAs) with an option of weekly printed materials. Over 6 months, 31 eligible participants completed the program with higher uptake (21/41) from physician referrals (74.2% women, 41.9% Black, median household income $100 000). Fourteen participants opted for weekly educational materials due to upcoming SMA sessions being fully booked or personal schedules. Pre‐ to post‐intervention paired t‐test showed improvement in systolic BP of 11.6 mmHg (95% CI, 6.6–16.6, p < 0.0001), and hypertension control rate improved by 36% (11/31) post‐intervention. Higher baseline systolic BP was associated with higher BP reduction (p < 0.001). Thematic analysis showed the perceived benefit of self‐awareness, education, and peer support, whereas time constraints were perceived as challenges. Self‐monitoring with education on coping skills and lifestyle modification is feasible and improved BP and hypertension control across diverse primary care patients interested in lifestyle modifications; however, few low‐income patients enrolled. Less burdensome and community‐based interventions may improve participation in low‐income patients.

1. Introduction

The US Surgeon General released a call to action in 2020 that recommends self‐monitoring of blood pressure (BP), lifestyle changes, and medication management to improve the declining hypertension control rates, but support for self‐monitoring and lifestyle counseling is challenging in primary care [1, 2, 3]. National Health and Nutrition Examination Survey (NHANES) data from 2015 to 2018, when compared to 2009 to 2014 data, showed hypertension control reduced across age and race‐ethnicity groups with no change in access to healthcare and increased prevalence of obesity—concluding the need for better quality of care and support for lifestyle modifications [2]. Self‐monitoring is a known behavior change mechanism in interventions targeting cardiovascular disease risks [4]. Self‐monitoring is essential for self‐regulation, while negative emotions can affect cardiovascular health (CVH) behaviors and BP [5, 6]. Self‐monitoring with coping skills and lifestyle education may potentially improve CVH behaviors and hypertension outcomes but has not been studied.

Hypertension is managed primarily by primary care physicians (PCPs) who see their patients with hypertension once or twice a year. However, PCPs have limited time, and continued support for self‐monitoring and lifestyle modifications is needed to change health behaviors successfully [7]. Shared medical appointments (SMAs) can provide knowledge to a group of patients with overlapping health conditions or outcome goals over several weeks. SMAs are reimbursable by most insurance companies, so they can be feasible and cost‐effective approaches in resource‐constrained primary care [8, 9]. Evidence synthesis of qualitative and quantitative research shows patients and physicians benefit from improved support for education with SMAs [10, 11, 12]. However, no previous SMA program for hypertension has explored coping skills education for emotional self‐management combined with self‐monitoring to enhance self‐awareness. This study is the first to focus on self‐monitoring of emotions and lifestyle behaviors combined with education on coping skills and lifestyle modifications for patients with uncontrolled hypertension.

Ecological momentary assessments (EMA), multiple daily assessments collected in real‐life contexts using a smartphone, have been used to monitor daily emotions in psychopathological disorders. In this study, EMAs were used for self‐reports of health behaviors and emotional experiences, and BPs to encourage self‐assessments in patients’ daily contexts [13].

The intervention design, development, and evaluation were guided by the self‐regulation model from Bandura's social cognitive theory (SCT) of agentic perspective [14]. In SCT, human behavior is regulated by self‐influence through three secondary mechanisms: self‐monitoring of behavior alongside its antecedents and effects, judgment of one's behavior in relation to personal standards or environmental contexts, and affective self‐reaction. In our intervention, we use SCT constructs of self‐regulation (self‐monitoring of behavior and outcome of behavior), behavioral capability (knowledge and coping skills training), and observational learning (self‐awareness, shared learning) to improve self‐efficacy (ability to improve CVH behaviors) [15]. The SCT focuses on continuous interactions and influences between individual factors, environment, and behaviors. EMAs encouraged self‐assessment of associations between individuals, environments, and behaviors in our study.

In this study, we explore (1) Is it feasible for patients with uncontrolled hypertension from primary care to participate in a 6‐week program of self‐monitoring with coping skills and lifestyle education before considering medication intensification to improve hypertension control? (2) Is the intervention acceptable to patients with uncontrolled hypertension as measured by retention and patient feedback?

2. Methods

We used a single‐group, mixed‐methods study design approach, and our protocol was approved and registered by Cleveland Clinic IRB. We used an explanatory sequential mixed methods design where post‐intervention qualitative data collection was intended to explain the feasibility and acceptability of self‐monitoring and education intervention components. Explanation of data was done through the integration of qualitative and quantitative data to draw inferences about the feasibility and acceptability of our intervention. Reporting of the integrated results was done through a narrative and contiguous approach where quantitative measures of feasibility and acceptability were followed by qualitative reports of patients’ perceptions about the intervention components and recommendations.

2.1. Participants and Setting

We enrolled patients with uncontrolled hypertension who preferred lifestyle modifications instead of medication intensification, which is a preference‐sensitive choice, so a control group or randomization was not considered feasible. SMA programs can have varying schedules, and patients get charged applicable copays as SMAs are reimbursable medical appointments. We adhered to patient preferences for SMA versus weekly educational material options rather than randomizing, as the choice would depend on the patient's personal schedules, preferences, and program participation costs.

We set up two suburban primary care practice sites in the Cleveland Clinic health system to loan home BP monitors and enroll patients. Each site had 7–8 primary care providers providing care for around 4000 patients with hypertension annually. We included patients with uncontrolled hypertension based on the last clinic BP measurement of systolic BP > 130 or diastolic BP > 80 mmHg, and patient desires or PCP recommended a trial of lifestyle modifications instead of medication intensification. We excluded patients who were pregnant, had a terminal illness, had clinic BP > 180/110, had severe cognitive impairment, had recent CVD in the last 6 months, had arm circumferences that exceeded the limit for the largest home BP monitor cuff, or any physical or mental impairment that would affect patients’ ability to participate in the study.

2.2. Recruitment

Eligible patients with uncontrolled hypertension at the last clinic visit who were seen in the previous 6 months and during recruitment were recruited through PCP messages to the research team and outreach through portal messages, letters, or phone calls. Additionally, weekly clinic lists of patients with uncontrolled hypertension were screened, and eligible patients (without medication intensification) were outreached through portal messages. Research staff repeated the BP measurement to ensure the patient's BP continued to be > 130/80 when loaning home BP monitors during enrollment.

2.3. Intervention Components

2.3.1. Self‐Monitoring

We encouraged self‐assessments through EMA three times a day for at least 4 days of each week during the 6‐week intervention period to facilitate self‐awareness and self‐judgment of any fluctuations and associations between emotions, behaviors, and BPs in real‐time. Daily morning, afternoon, and evening EMA prompts were sent to patients’ smartphones so patients could choose the days that worked for them to complete the assessments. EMA is a well‐studied methodology in populations with mental health disorders [16] and for monitoring BP and heart rate fluctuations with stress arousals [17]. We selected a home BP monitor brand (Greater Goods) that is included in the US BP Validated Device Listing (VDL) compiled by the American Medical Association (AMA) and comes with a wall charger and an adjustable cuff for most arm sizes. In addition, PI and research staff validated all home BP monitors by individually comparing BP monitor readings with validated clinic BP machines. Research staff taught participants proper BP measurement techniques using checklists and teach‐back techniques from primary author's previous study and gave them a printed participant guide with written instructions [18].

Research staff gave patients verbal and written information on self‐monitoring of cardiovascular behaviors included in the American Heart Association's (AHA) life's essential eight (number of sleep hours, number of minutes of moderate or vigorous physical activity, serving sizes of vegetables, fruits, whole grains, fish, and soda intake) [19]. We used the components of AHA's CVH metrics as they are associated with various important health outcomes, from cardiovascular diseases to cancer [19]. Additionally, patients were given information on daily monitoring of emotions using the Positive and Negative Affect Schedule (PANAS), which is a validated self‐report measure made of two mood scales—one 10‐item scale measuring positive affect (PA) and the other 10 items measuring negative affect (NA) [20, 21]. All participants received a printed participant guidebook that included instructions on self‐monitoring BP, CVH behaviors, and emotions. Patients were assisted with downloading the MyCap mobile application, which is a free, secure web‐based platform provided by REDCap for EMA data capture [22, 23]. Participants completed one self‐monitoring assessment in the presence of research staff for any troubleshooting or questions. Participants received feedback through weekly graphic summaries of their daily average home BPs and their CVH behaviors.

2.3.2. Education

We chose a 6‐week intervention duration based on the increased dropout rates in SMA attendances noted by the Cleveland Clinic Center for Integrative Lifestyle and Medicine Center after 6–7 weeks of participation over the last decade. The 6‐week SMA program and materials were created by clinicians with expertise in lifestyle medicine and holistic psychotherapists with the goal of supporting patients to achieve optimal health for themselves. The program focused on four key evidence‐based components for good health—nutrition, physical activity, stress relief, and restorative sleep. The program was delivered as 90‐min weekly sessions where each session included three components (1) Medical visit with individual assessments and education delivered by a medical provider; (2) healthy coping skills education and exercises delivered by a holistic psychotherapist; and (3) gentle stretching exercises with chair yoga in last 10–15 min by yoga therapist. There were sessions offered at various times, including evenings, with two in‐person options and three virtual session options. All of the five SMA session options started in the same week every 6 weeks so if patients cannot attend their regular session, they have the option of attending another session in the same week as all sessions delivered similar content each week. From a sustainability perspective, the SMA program for optimal health emphasizes awareness of stress‐triggered health behaviors and is not disease‐specific. Hence, the SMA was available for patients without hypertension to maintain the clinical productivity of SMA providers, but non‐study participants did not engage in self‐monitoring or study assessments. Every 6‐weeks, a new cohort of patients attends the SMA programs throughout the year with a break between mid‐December to mid‐January to avoid disruptions due to holiday seasons. As we were recruiting patients with uncontrolled hypertension and no medication intensification, we offered the option of weekly educational materials if all the upcoming SMA sessions within 6 weeks were full.

We offered printed SMA educational materials for anyone interested in educational materials or if participants could not afford copays for SMA or the SMA times did not work, or SMA participation would not occur within 2 months (as enrollees had uncontrolled hypertension). Educational materials for non‐SMA participants were the weekly SMA presentation handouts and homework, including links to audio and video of guided imagery, cooking videos, and yoga videos sent as sequenced weekly scheduled emails for 6 weeks.

After the 6‐week intervention completion, study participants returned BP monitors and completed an open‐ended post‐intervention survey or interview. Patient's physicians received post‐intervention clinic BP measurements, and patients were instructed to make follow‐up appointments with their physicians for the management of persisting uncontrolled hypertension.

3. Data Collection

Demographic, social, and clinical data were collected from participants and their medical records. The primary outcomes were the feasibility and acceptability of the intervention.

3.1. Primary Outcomes

3.1.1. Feasibility and Acceptability

Intervention uptake and feasibility were measured from the proportion enrolled through primary care referrals or outreach messages through patient portals or letters during the active recruitment period from October 2022 until April 2023. We assess acceptability based on the proportion of patients engaged in self‐monitoring or SMA or educational materials for 4 or more of 6 weeks. Feasibility and acceptability were assessed qualitatively through post‐intervention by asking what participants liked, disliked, or would recommend being changed for self‐monitoring, education, and clinical services based on their program participation experience.

3.2. Secondary Outcomes

3.2.1. BP and Hypertension Control Measures

The last measured clinic BP before enrollment was used as pre‐intervention BP, and clinic BP was measured at an office visit after program participation, which was used as post‐intervention BP. Research staff measured BP post‐intervention, but to maintain a pragmatic evaluation of the program, we preferred the use of post‐clinic BPs when available. Hypertension control was calculated as the proportion of patients with systolic BP < 130 and diastolic BP < 80 mmHg. We also explored change in hypertension control with a change in average home BP from the first 3 days to the last 3 days of reporting beyond 4 weeks of participation.

3.2.2. Other Measures

We calculated the CVH metric out of 6 points (Table S1) for the measured daily components of diet, physical activity, and sleep from the first week to the last week of participation if self‐reporting extended beyond 4 weeks of participation. We calculated the mean PA and mean NA from the PANAS measure over the first 3 days to the last 3 days of reporting beyond 4 weeks of participation to explore change in PA and NA. We collected the following pre‐, post‐, and 6‐month post‐intervention validated surveys: Mindful Attention Awareness Scale (MAAS), which is a validated scale for baseline assessment of an individual's mindfulness trait [24, 25], the Insomnia Severity Index, which is a validated scale for sleep [26, 27], and PROMIS Self‐efficacy for Managing Chronic Conditions [28].

3.2.3. Analysis

Our primary goal was not to detect significant improvements in clinical or patient‐reported outcomes but instead to field‐test our intervention to improve the feasibility for future upscaling and efficacy testing of the intervention [29]. Descriptive statistics for all variables are reported.

3.2.4. Quantitative Analysis

Statistical analysis was done using SAS software.

We compared characteristics of gender, age, race/ethnicity, BP, and BMI of study participants with eligible outreached populations from the two primary care clinic sites using a Chi‐square for categorical variables and a t‐test for continuous variables.

We used paired t‐tests and Wilcoxon signed rank tests to compare continuous variables (BP, average home BP, CVH metric, PA, NA, Self‐efficacy, MAAS, and insomnia index) before and after treatment and paired proportion tests to compare changes in the proportion of patients with controlled hypertension. We created multivariable linear regression models to assess whether pre‐intervention BP and BMI are associated with changes in BP.

For sample size calculation, home BP monitoring with education and support can improve hypertension control rates by 27%–56%, including in the primary author's previous home BP monitoring study [18, 30, 31]. Before recruitment, we calculated a conservative sample size of 54 would provide greater than 80% power to detect significant improvement in our main outcome of hypertension control by 15% using a paired proportions test with a two‐tailed alpha of 0.05; so, we initially aimed for this conservative sample size of 60 but setting up EMA, SMA scheduling and grant funding logistics limited final sample size to 33 of which 31 participants were eligible for analysis. A sample size of 31 has > 80% power to detect a 24% improvement in hypertension control, closer to the effect size noted in the literature.

3.2.5. Qualitative Analysis

Post‐intervention feedback included questions on what participants liked, disliked, or would recommend changing for each intervention component (self‐monitoring and education) as well as the intervention and clinical services (e.g., what did you like about the self‐monitoring part of the program?). Additionally, participants were asked their opinions on how their mood, diet, physical activity, and BP affect each other based on their participation in the study. See Supporting Document 2 for post‐intervention open‐ended survey questions. Participants were given the option to provide post‐intervention feedback as an open‐ended survey or interview. Twenty‐eight participants completed an online post‐intervention survey, and feedback was provided through an interview by one participant (93% response rate). Open‐ended surveys and interviews were organized in REDCap and extracted in Microsoft Excel to analyze using the framework method, a thematic analysis method that allows for structured reduction of textual data to produce a highly structured output of summarized data. The framework method uses matrices in which rows correspond to cases, columns correspond to survey or interview questions, and cells contain textual data. As all but one participant provided qualitative feedback through surveys, data were extracted in matrix format from REDCap in Excel. For the one participant post‐intervention interview, the answers to all the interview questions were transferred verbatim to the matrix.

The following questions guided the qualitative analysis: What did participants like/dislike about the intervention components? How could the intervention and clinical practices be improved in the future? Does self‐monitoring and education facilitate self‐awareness and self‐efficacy? One author (S.J.P.) created the primary codebook after reviewing all survey transcripts, and two research team members (E.U. and I.T.) were trained in coding transcripts to identify themes. Two co‐authors double‐coded each transcript. Thematic analysis identified themes related to the questions. Discrepancies between coders were resolved through synchronous and asynchronous discussions. Finally, each coder's transcript summary was combined to achieve consensus on consolidated themes for each question.

4. Results

Figure 1 shows our recruitment and follow‐up completion status in a modified CONSORT flow diagram.

FIGURE 1.

Modified CONSORT (consolidated standards of reporting trials) flow diagram showing screening, enrollment, follow‐up, and analysis of participants. PCP indicates primary care physician.

4.1. Primary Outcome‐Feasibility and Acceptability Outcomes

Of the 880 eligible participants, 33 were enrolled, of which one dropped out, and one was ineligible as they did not follow up for any primary care clinic visits within the Cleveland Clinic health system. Table 1 shows differences between the target eligible population and study participant characteristics. Enrollees were more likely to be female, Black, and in the age range between 40 and 60 years compared to outreached populations. Fifty‐three percent (21/41) of eligible participants referred by PCPs enrolled and the enrollment rate from outreach messages was around 1%. Participants were predominantly female, with a mean age of 56.0 (SD 11.1) years, mean baseline systolic BP of 144.3 (SD 12.5), and mean BMI of 31.8 (SD 5.9). The enrolled participants’ systolic BP varied from 131 to 178 mmHg. Table 2 shows baseline participant characteristics. Sixteen participants were married, the rest were single/divorced/widowed. Six participants had childcare and five participants had adult care responsibilities. The median annual household income of enrolled participants was $100 000 (IQR $40 000 to $150 000). All patients completed baseline demographics and surveys. The post‐intervention feedback completion rate was 93% (29/31).

TABLE 1.

Differences between target eligible population (outreach) and study participant characteristics.

Participant characteristics	Eligible patients (target population)		Study participants		p value
	N	%	N	%
	880		31
Age (years)					0.1360
< 40	87	9.9	1	3.2
40–59	364	41.4	18	58.1
≥ 60	429	48.8	12	38.7
Gender					< 0.0001
Male	432	49.1	7	22.6
Female	447	50.8	23	74.2
Nonbinary	1	0.1	1	3.2
Race					0.0115
African American/Black	185	21.0	13	41.9
Caucasian/White	637	72.4	18	58.1
Other	58	6.6	0	0.0
Last clinic systolic BP value (mmHg)					0.4000
≤ 140	553	62.8	17	54.8
141–150	181	20.6	6	19.4
≥ 151	146	16.6	8	25.8
Last clinic diastolic BP value (mmHg)					0.9401
≤ 90	718	81.6	26	83.9
91–100	125	14.2	4	12.9
≥ 101	37	4.2	1	3.2
BMI					0.4943
< 30	373	42.4	13	41.9
30–34.9	243	27.6	12	38.7
35–39.9	140	15.9	3	9.7
≥ 40	120	13.6	3	9.7

The bold values indicate statistical significant differences between the respective groups of participant characteristics.

TABLE 2.

Sample demographics overall and by White and Black participants.

	Overall (N = 31)	White (N = 18)	Black (N = 13)	p value
Age, mean (SD)	56.0 (11.1)	57.9 (11.8)	53.5 (9.8)	0.28
				0.13
< 60 years, N (%)	19 (61.3)	9 (50.0)	10 (76.9)
≥ 60 years, N (%)	12 (38.7)	9 (50.0)	3 (23.1)
Gender				0.14
Male	7 (22.6)	6 (33.3)	1 (7.7)
Female	23 (74.2)	11 (61.1)	12 (92.3)
Other	1 (3.2)	1 (5.6)	0 (0.0)
Education				0.80
High school graduate, diploma, or the equivalent (GED)	3 (10.0)	1 (5.9)	2 (15.4)
Some college credit (no degree)	3 (10.0)	1 (5.9)	2 (15.4)
Associate degree	2 (6.7)	1 (5.9)	1 (7.7)
Bachelor's degree	12 (40.0)	8 (47.1)	4 (30.8)
Master's degree	10 (33.3)	6 (35.3)	4 (30.8)
Annual household income $, median (IQR)	100 000 (27 000, 450 000)	125 000 (29 000, 450 000)	54 000 (27 000, 166 000)	0.091
Employed				0.24
Yes	22 (71.0)	11 (61.1)	11 (84.6)
No	9 (29.0)	7 (38.9)	2 (15.4)
Marital status				0.83
Married	16 (51.6)	9 (50.0)	7 (53.8)
Single/widowed/divorced	15 (48.4)	9 (50.0)	6 (46.2)
Caregiver for child < 18 years of age	6 (19.4)	2 (11.1)	4 (30.8)	0.21
Caregiver for adults	5 (16.7)	3 (17.6)	2 (15.4)	0.99
Baseline SBP, mean (SD)	144.3 (12.5)	148.6 (13.8)	138.4 (7.2)	0.013
Baseline DBP, mean (SD)	84.9 (8.2)	85.3 (9.2)	84.2 (6.8)	0.72
BMI, mean (SD)	31.8 (5.9)	31.5 (7.1)	32.3 (3.9)	0.68
Baseline PROMIS self‐efficacy score, mean (SD)
Emotions	47.4 (8.5)	47.2 (8.2)	47.6 (9.3)	0.90
Medications and treatments	48.3 (9.4)	49.4 (10.0)	47.0 (8.9)	0.55
Social interactions	46.5 (8.8)	45.9 (9.4)	47.2 (8.5)	0.73
Symptoms	51.6 (9.3)	50.0 (8.7)	53.7 (10.0)	0.34
Daily activities	51.4 (7.8)	50.1 (7.6)	53.0 (8.2)	0.36
Baseline MAAS scores, mean (SD)	4.6 (0.8)	4.5 (0.7)	4.7 (0.9)	0.70
Baseline Insomnia sleep index, mean (SD)	8.6 (5.9)	8.1 (5.6)	9.2 (6.5)	0.66
Smoking status				0.99
Current smoker	3 (11.1)	2 (13.3)	1 (8.3)
Former smoker (last smoked > 12 months)	5 (18.5)	3 (20.0)	2 (16.7)
Never smoker	19 (70.4)	10 (66.7)	9 (75.0)

Fourteen of the 31 participants opted for educational materials due to personal SMA scheduling challenges (6 participants) or the upcoming SMAs were fully booked by the time of participant recruitment (8 participants). The SMA attendance was < 50% and self‐monitoring was not completed by three participants (10%; one Black male, one White male, and one White female). Twenty‐five (80%) participants engaged in at least 3 days per week of self‐monitoring for at least four weeks, whereas 12 (39%) of participants had 100% adherence to self‐monitoring for all of the six weeks. Post‐intervention survey completion rate was 93% and for 6‐month post‐intervention survey completion ranged from 42% (13/31) for PROMIS measures to 48% (15/31) for MAAS measures.

4.1.1. Change in BP and Hypertension Control

See Table 3 for all the results of changes from pre‐intervention to post‐intervention for all secondary outcomes.

TABLE 3.

Results for secondary outcomes.

Outcomes	N	Pre‐intervention value	Post‐intervention value	Change from baseline	p value
Systolic BP (SBP) (mmHg) mean (SD)	31	144.3 (12.5)	132.7 (12.2)	−11.6 (13.8)	< 0.0001
Diastolic BP (mmHg) mean (SD)	31	84.9 (8.2)	79.2 (9.2)	−5.7 (8.8)	0.0011
Proportion at goal (%) Systolic BP:
SBP < 140 mmHg N (% of 31)	31	16 (51.6)	24 (77.4)	8 (25.8)	0.0337
SBP < 130 mmHg N (% of 31)	31	0 (0.0)	15 (48.4)	15 (48.4)	< 0.0001
Proportion at goal (%) Diastolic BP (DBP):
DBP < 90 N (% of 31)	31	23 (74.2)	27 (87.1)	4 (12.9)	0.1985
DBP < 80 N (% of 31)	31	6 (19.4)	16 (51.6)	10 (32.2)	0.0079
Average home SBP (mmHg) mean (SD)	24	133.5 (12.3)	130.1 (9.2)	−3.4 (9.9)	0.1068
Average home DBP (mmHg) mean (SD)	24	84.7 (9.2)	82.7 (7.8)	−2.0 (6.7)	0.1609
Average PA a mean (SD)	24	19.0 (11.9)	21.9 (9.5)	2.9 (10.8)	0.1800
Average NA b mean (SD)	24	10.2 (4.3)	11.0 (2.9)	0.8 (3.6)	0.6261
CVH metric over week c (out of maximum possible 6 points) mean (SD)	22	3.15 (0.67)	2.80 (0.76)	−0.35 (0.64)	0.0182
6‐week PROMIS self‐efficacy score d mean (SD)
Emotions	24	47.3 (8.7)	49.7 (9.0)	2.3 (7.5)	0.1414
Medications and treatments	24	48.6 (9.5)	48.5 (8.3)	−0.1 (10.3)	0.9505
Social interactions	24	46.7 (9.0)	49.4 (9.5)	2.8 (5.9)	0.0476
Symptoms	24	51.8 (9.4)	53.9 (8.9)	2.1 (6.1)	0.1439
Daily activities	24	51.7 (7.8)	52.3 (8.0)	0.6 (5.1)	0.6233
MAAS scores e mean (SD)	29	4.6 (0.8)	4.6 (0.70)	−0.03 (0.72)	0.8116
Insomnia sleep index mean (SD)	25	8.6 (6.0)	7.0 (5.3)	−1.7 (3.7)	0.0078
6‐month PROMIS Self‐efficacy for managing mean (SD)
Emotions	13	47.7 (9.3)	52.8 (10.0)	5.2 (9.7)	0.0787
Medications and treatments	13	47.9 (8.1)	52.6 (7.0)	4.7 (8.3)	0.0646
Social interactions	13	46.3 (7.1)	53.4 (9.2)	7.1 (8.1)	0.0083
Symptoms	13	51.7 (7.6)	56.5 (8.9)	4.8 (8.2)	0.0781
Daily activities	13	51.9 (7.9)	54.4 (7.8)	2.5 (7.2)	0.2314
MAAS scores mean (SD)	15	4.6 (0.7)	4.9 (0.8)	0.4 (0.7)	0.0585
Insomnia sleep index mean (SD)	14	6.5 (5.5)	5.5 (4.8)	−1.0 (3.8)	0.3406

Abbreviations: CVH, cardiovascular health; MAAS, Mindful Attention Awareness Scale; NA, negative affect; PA, positive affect; PANAS, positive and negative affect schedule.

^a,b PA and NA are calculated from the PANAS as the sum of PA and NA

^c CVH metric was measured over 7 days as a measure of fish intake and minutes of physical activity are weekly so a change from week 1 to the last reported week beyond 4 weeks was reported.

^d PROMIS Self‐Efficacy for Managing Chronic Conditions item banks are well‐validated and comprise five domains, Self‐Efficacy for Managing: Daily Activities, Symptoms, Medications and Treatments, Emotions, and Social Interactions.

^e MAAS Mindful Attention Awareness Scale; For CVH metric, PROMIS self‐efficacy, and MAAS scores—higher scores are better. For the Insomnia Severity Index lower scores are better; statistics are presented as mean (SD), N (column %).

The bold values indicate statistical significance in the comparison between groups.

Baseline systolic BP was 144.3 (SD 12.5), and Diastolic BP was 84.9 (SD 8.2). Post‐intervention change in systolic BP was −11.6 (SD 13.8; p < 0.0001), and in Diastolic BP was −5.7 (SD 8.8; p = 0.002). The proportion of patients with controlled hypertension increased from 0% to 35.5% (p = 0.0003). See Table 4 for comparisons of pre‐ and post‐intervention secondary outcome changes between Black and White participants. Baseline systolic BP was lower in enrolled Black participants compared to White participants. There were no significant differences in change in BP (p = 0.16) between Black and White participants. In the first week of intervention, 16.7% of patients had controlled hypertension (< 130/80) based on their average home BP, and 20.8% of patients had controlled hypertension based on the average home BP in the last week of participation beyond 4 weeks. See Table S2 for comparisons of pre‐ to post‐intervention change in secondary outcomes between participants that attended SMA and educational materials, which shows no statistically significant difference in change in BP between these groups of participants.

TABLE 4.

Outcome comparisons between Black and White participants.

Factor	Total (N = 31)	White (N = 18)	Black (N = 13)	p value
Age (years)	56.0 ± 11.1	57.9 ± 11.8	53.5 ± 9.8	0.28 a
Pre‐intervention systolic BP (mmHg)	144.3 ± 12.5	148.6 ± 13.8	138.4 ± 7.2	0.013 b
Post‐intervention systolic BP (mmHg)	132.7 ± 12.2	134.0 ± 13.2	130.9 ± 10.9	0.50 a
Change in systolic BP (mmHg)	−11.6 ± 13.8	−14.6 ± 14.9	−7.5 ± 11.2	0.16 a
Pre‐intervention diastolic BP (mmHg)	84.9 ± 8.2	85.3 ± 9.2	84.2 ± 6.8	0.72 a
Post‐intervention diastolic BP (mmHg)	79.2 ± 9.2	78.9 ± 10.3	79.5 ± 7.8	0.85 a
Change in diastolic BP (mmHg)	−5.7 ± 8.8	−6.4 ± 8.5	−4.7 ± 9.4	0.59 a
Post‐intervention‐controlled hypertension N (%)	11 (35.5)	7 (38.9)	4 (30.8)	0.72 c

Note: Statistics are presented as Mean ± SD, N (column %).

^a t‐test.

^b Satterthwaite t‐test.

^c Fisher's exact test.

Multivariable linear regression models analyzing the effect of baseline BP and BMI on change in BP or hypertension control outcomes did not show any difference by BMI (coefficient = −0.1025, SE = 0.2552, p = 0.6879), whereas higher baseline systolic BP was associated with increased improvement in systolic BP (coefficient = −0.6822, SE = 0.1217, p < 0.0001) and higher baseline diastolic BP was associated with increased improvement in diastolic BP (coefficient = −0.43, SE = 0.206, p = 0.0368) See Table S3 for results of the multivariable regression models. Table 5 shows comparisons between changes in BP and hypertension control with baseline systolic BP < 140, between 141 and 150, and >150 mmHg. There were no statistically significant differences in the odds of controlled hypertension by race/ethnicity, mode of education (SMA and educational materials), and baseline BP. Table S4 shows the odds ratios of controlled hypertension by each characteristic.

TABLE 5.

Outcome comparisons between participants with differing baseline systolic BP.

Factor	Baseline systolic BP ≤ 140 (N = 17)	Baseline systolic BP 141–150 (N = 6)	Baseline systolic BP > 150 (N = 8)	p value
Age (years)	52.6 ± 9.9	59.2 ± 12.6	60.9 ± 11.1	0.17 a
Gender				0.39d
Female	14 (82.4)	3 (50.0)	6 (75.0)
Male	2 (11.8)	3 (50.0)	2 (25.0)
Non‐binary	1 (5.9)	0 (0.00)	0 (0.00)
Pre‐intervention systolic BP a (mmHg)	135.7 ± 2.7	145.2 ± 4.4	161.9 ± 10.1	< 0.001 a
Post‐intervention systolic BP a (mmHg)	130.4 ± 10.0	130.0 ± 9.5	139.8 ± 16.4	0.17 a
Change in systolic BP a (mmHg)	−5.4 ± 9.6	−15.2 ± 8.5	−22.1 ± 17.8	0.009 a
Pre‐intervention diastolic BP a (mmHg)	83.6 ± 8.1	84.3 ± 8.9	88.0 ± 8.0	0.46 a
Post‐intervention diastolic BP a (mmHg)	79.5 ± 7.5	75.2 ± 10.5	81.5 ± 11.6	0.45 a
Change in diastolic BP a (mmHg)	−4.1 ± 8.6	−9.2 ± 9.1	−6.5 ± 9.3	0.48 a
Post‐intervention‐controlled hypertension N (%)	6 (35.3)	3 (50.0)	2 (25.0)	0.78d

^a Data not available for all subjects. Missing values: Pre_Systolic = 1, Post_Systolic = 1, change_SBP = 1, Pre_Diastolic = 1, Post_Diastolic = 1, change_DBP = 1, Post_controlled = 1.

Statistics are presented as mean ± SD and N (column %).

p‐values: a = ANOVA, d = Fisher's Exact test.

The bold values indicate statistical significance in the comparison between groups.

4.1.2. Qualitative Analysis

Tables 6, 7, 8, and S5 show the themes from our thematic analysis with the frequency of themes and quotes. Table 6 shows self‐monitoring themes, which included progress monitoring, self‐awareness, attributions, self‐efficacy, challenges of EMA survey choices, and the practicality of checking BP throughout the day. Table 7 shows themes for education received through printouts and SMA, which included observational learning, education that was self‐efficacy‐inducing, and challenges of time and cognitive efforts needed for SMA. One White and one Black female participant was uncomfortable sharing in group sessions and desired sharing to be optional. Table 8 shows themes for feedback on participants' recommendations to change intervention and clinical services, which included acceptability, time burden, communication and individualized support for behavior change, decision‐making with accurate BP assessments, and referrals for additional support. Table S5 shows themes related to weekly summary reports, which included visual feedback for observational learning, graphs for progress monitoring, and the need for additional support for interpretations.

TABLE 6.

Themes from likes and dislikes for self‐monitoring component.

Self‐progress‐monitoring (Frequency: 8)

“Liked all of it…your different emotions affect your blood pressure.” “Liked monitoring and writing down what the blood pressure was throughout the day. Liked monitoring mood as well.” “Liked that it kept me on track with my blood pressure. Liked that I could fill out what I was eating every day.”

Self‐awareness (judgment) (Frequency: 7)

“Just being aware how during parts of the day your blood pressure may go up or down depending on what you eat or if you're getting up to move.” “I feel like looking at my emotions and paying attention to why I'm having the emotion and things to do with certain emotions was very helpful.” “made me realize importance of everything I do—eat, drink, exercise, sleep, mood, talking with someone or work emails affect my BP. I pay more attention to my body and habits”

Attribution (Frequency: 4)

“liked all of it your different emotions affect your blood pressure if angry, it affects blood pressure, so don't let things bother me” “I like that it is always at the forefront of your mind which forces you to be cognizant of all that affects your blood pressure.”

Self‐efficacy (ability to change behaviors) (Frequency: 3)

“I thought the constant monitoring made me much more aware of blood pressure, emotions, diet, and activity. Being aware of such things also made me change my behavior, especially in the area of improving my diet.” “The program was very helpful and will try to continue eating right and working out.”

Challenges: EMA emotion measures (PANAS) and diet choices (Frequency: 16)

“Emotions part was difficult because many of them I did not feel at all.” “meat choices (did not like that the only meat choice was fish)…emotions—answering every single one, did not like that—found it was cumbersome” “I liked the idea of performing a regular emotional check‐in, but wasn't sure the categories were as useful…” “I think there needed to be a space in each category to enter my own additional data. It would have been helpful to be able to add the foods I ate or the mood I was in when they were not any of the options in the survey.” “constant reminders made me realize what I needed to do but doing the right things were totally discouraging” (negative reinforcement)

Challenges: Time and practicality of frequent BP checks (Frequency: 10)

“I work nights so having to take my blood pressure 3 times a day was hard.” “It was difficult to monitor throughout the day because of work.” “The blood pressure being taken 3 times a day was somewhat stressful.”

TABLE 7.

Themes from likes and dislikes for education content: SMA and educational printouts.

Peer support‐observational learning (Frequency: 13)

“really enjoyed the shared medical appointments, they were comforting to see that there were other people that have similar struggles to myself ” “liked being with other people in similar situations. liked listening to how they cope, their strategies and their ideas.” “liked the fact that we were able to do a group setting and able to hear everyones challenges or successes…kept me motivated” “It was great to hear others like me share what they are going through. This was great in that it tells me I am not alone”

Education‐self‐efficacy benefits of printed and in‐person education (Frequency: 6)

“really appreciated that they sent the powerpoint bc I printed it out and was highlighting it on a physical copy went…doing the recipes. post study is now implementing 2 small changes per week” “printed out the home materials and loved the link for the meditations was able to use the meditation links to help fall asleep” “liked having …(medical provider) there to answer any medical questions. liked …(psychotherapist) for mental health and …(yoga specialist) for yoga.”

Challenges: Time and cognitive efforts needed for shared medical appointments (Frequency: 6)

“Time consuming. not given enough individualized attention.” “Maybe the time because I was usually getting off that morning from working the night before and wanted to sleep. The evening time didn't work because I had to be at work at 7 pm.” “90 minutes was too long.” “Sharing with others in group and hearing their issues some non‐relevant to what needed to be heard… others just talked to much… did not like sharing some of my personal info either… some of this was just too time consuming and could of been omitted I believe.” “Not forcing someone to talk especially when others were successful and you're not. very discouraging.”

TABLE 8.

Recommendations for changes to intervention and clinical services based on study participation.

Overall acceptability: No additional changes (Frequency: 13)

“I think the program was very beneficial. The mods were great, offered great tools and insights on how to better yourself and your habits” “I appreciate being about to participate in the classes. I initially participated in the in person class and later needed to switch to the online class so I got both experiences. Both were great and I liked the online group/class the best! This was a surprise! I do online work all the time and was looking forward to a “live” class.” “really enjoyed the program and the process enjoyed the Zoom and enjoyed the information that was given in the packets and surveys would do it again”

Time burden for self‐monitoring and health education components (Frequency: 6)

“SMA should be 60 minutes at most. Shorten the after BP check questionnaire” “lot to ask of people to do for 6 weeks…3 times a day of filling out the information” “Overall, I enjoyed participating in the study. My only suggestions for improvement might be in the area of better defining the emotional component of the testing”

Communication and individualized support for behavior change (Frequency: 7)

“A once a week call to check on me would have been okay with me.” “Great question! Here are a few thoughts. I would like some feedback on what you saw in my weekly reports and any other feedback on my daily questionnaires.” “when doctors are communicating with patients, like weight loss would be beneficial for the patient's health to say to a patient ‘you're obese, you need to lose weight’ and they ship them out the door, it is not helpful would be more helpful to share information with patients and not overloading them.” “inserting into the questioning, how much meditation or mindfulness are you doing? what are your calming or centering practices? recommending these practices and a certain number of times a week.”

Decision‐making with accurate BP assessment (Frequency: 3)

“Not solely looking at the blood pressure in the office but take into account what it is while I'm at home. Not jumping to conclusions of well she's not doing what she's supposed to be doing. I know that schedules are tight with appts but to allow the pt to sit for the 5 min after they are in the exam room before taking the blood pressure.” “More at home monitoring and follow up to see improvements.”

Referral for additional support (Frequency: 2)

“if doctors dont see that progress is happening, push patients to join studies or to start making changes for hypertension and weight loss” “I would appreciate your inquiring about how I am doing with striving to improve my ability to exercise and perhaps a possible referral to an exercise physiologist (?) or other specialist to help me”

5. Discussion

We are the first to evaluate the feasibility of self‐monitoring of emotions and lifestyle behaviors combined with education on coping skills and lifestyle modifications for patients with uncontrolled hypertension who preferred lifestyle modification trials before medication intensification. It was feasible to enroll patients with uncontrolled hypertension in our program with a higher enrollment rate from PCP referrals compared to outreach messages. Enrolled patients' baseline systolic BP varied from 131 to 178, with higher baseline BP showing higher BP improvements, though PCPs more frequently referred patients with mild hypertension (systolic BP between 131 and 139). Interprofessional collaborative practices with nurses and pharmacists have frequently used telemonitoring or SMA models and shown improved diabetes and BP levels, but most studies have focused on the medical management of diabetes and/or hypertension with an infrequent focus on self‐regulation and emotional coping skills like our study [8, 32, 33, 34].

Twenty‐five (80%) participants engaged in at least 3 days per week of self‐monitoring for at least 4 weeks, whereas 12 (39%) of participants had 100% adherence to self‐monitoring for all of the 6 weeks. Most study participants were employed, educated, and had higher annual household incomes along with higher baseline self‐efficacy and mindfulness traits, which may explain some of the positive outcomes [24, 28, 35]. Themes for SMA feedback from Black participants focused on the value of peer support, while White participants valued knowledge and access to mind‐body therapists. Patients reported increased self‐awareness of interactions between their emotions, health behaviors, and BPs with self‐monitoring due to participation in the study. However, SMA time scheduling challenges, the number of emotions in the PANAS scale, and checking BP three times a day were perceived as challenges by participants. Using a few simplified questions instead of longer scales for EMA may improve their practicality in patients with uncontrolled hypertension [36]. The primary author's previous home BP monitoring study participants reported that home BP monitoring was feasible twice daily [18].

Our secondary outcomes of BP and hypertension control improved in study participants from pre‐intervention to post‐intervention. Three patients with systolic BP > 170 preferred to participate in the program instead of medication intensification and reduced their systolic BP by 20–40 points. Patients with uncontrolled hypertension who are reluctant to intensify medications may benefit from enrolling in self‐monitoring and education programs for lifestyle modifications, irrespective of their baseline BP. The insomnia sleep index improved from average subthreshold insomnia to the borderline score for the absence of insomnia, but we did not find significant differences in mindfulness, self‐efficacy, and CVH metric scores, but this was a feasibility study and most measures moved toward a healthier direction over 6 months. Though not statistically significant due to our feasibility study design, moderate effect size for improvement in PA was noted from pre‐ to post intervention with self‐monitoring and coping skills education. Positive emotions are known to undo the aftereffects of negative emotions according to the broaden‐and‐build theory of positive emotions [37]. Negative emotions lead to deteriorative biological and behavioral processes, whereas positive emotions can accelerate cardiovascular recovery and improve BP [6]. Emotional self‐monitoring and coping skills education may potentially promote problem‐solving for controllable stressors and seeking for other coping mechanisms for uncontrollable stressors, thus buffering the effects of stress on BP and overall CVH [38].

Our study shows a higher proportion of uptake of a readily available program in Black patients with uncontrolled hypertension. Acceptability was mostly similar among White and Black participants. Having programs available and accessible to all hypertensive patients, even though not tailored to specific populations, may still be an effective way to reduce healthcare access disparities. Nevertheless, the baseline BP of enrolled Black participants was lower than that of enrolled White participants. Black patients with higher baseline BPs may need individualized or alternative outreach approaches.

SMA schedules were limiting for enrollment, and 14 of the 31 participants opted for weekly educational materials. Few patients disliked sharing or listening to others’ stories of their progress during SMAs but deferred the option of switching to weekly educational materials when offered. SMAs are reimbursable and cost‐efficient as they allow medically necessary services to be delivered in the presence of other group members, but several participants desired individualized feedback [39]. There was no significant difference in hypertension outcomes between participants receiving educational materials and interactive SMA intervention or between Black and White patients, but our small sample size was not powered to detect differences between these groups. A recent review of healthcare delivery interventions for hypertension mentions that educational materials and pamphlets were less effective than interactive education in underserved populations [40]; however, our study included an additional self‐monitoring component.

Most of the study participants were females, which may reflect the increased prevalence of psychosocial stress in this population, as the program flyers specified the inclusion of lifestyle and coping skills education. The primary author's previous work showed women with uncontrolled hypertension had worsening BP compared to men despite participation in a 2‐year nurse‐led clinic‐based care coordination program [41]. Co‐addressing mental health and CVH may potentially improve gender‐based disparities in cardiovascular disease outcomes [42, 43].

Our study's strengths include using mixed methods to measure feasibility and acceptability as participants gave feedback on how the intervention worked for them and why the intervention was liked or disliked. Additionally, insomnia sleep index, PANAS, MAAS, and self‐efficacy scales are valid self‐report measures, and qualitative data augmented our understanding of how personal experiences matched participants' self‐awareness of the interactions between their emotions, health behaviors, and BPs.

5.1. Limitations

Our study has the limitations of a single‐arm study without control as we chose preference‐sensitive choice of lifestyle modifications versus medication intensification for uncontrolled hypertension. A recent randomized controlled trial of telephone‐based lifestyle counseling for diet modifications compared to enhanced pharmacotherapy protocol showed improved long‐term hypertension control in Black adults, which aligns with our study findings [44]. Clinic BPs are known to be the least accurate, but hypertension control rates improved from 16.7% to 20.8% based on the first and last week of average home BP readings supporting the direction of our results. We offered a lifestyle modification program but did not screen or address any social needs that may interfere with lifestyle modifications. Participants joining the study were educated with higher self‐efficacy, so they were likely more motivated to make lifestyle modifications to improve their hypertension control. Hence, our findings may not apply to individuals with low self‐efficacy, social disadvantages, and enforced program participation. Our study's 6‐month post‐intervention survey response rates were < 50%, which could be due to being disconnected from the program for 6 months. We did not use any telephone or email reminders to increase 6‐month response rates, which may yield better response rates in the future.

5.2. Future Directions

Patients with uncontrolled hypertension desire support for self‐monitoring, lifestyle modifications, and emotional self‐regulation. Self‐monitoring support with education by interdisciplinary team members is feasible and acceptable for patients with uncontrolled hypertension. However, frequency and measures for daily monitoring need to be simplified to improve patient‐centeredness and apply across diverse health literacy levels [36, 45]. SMA programs are reimbursable and can provide education and self‐monitoring support, but time burden and scheduling issues may limit patient participation. Flexible, individualized support through weekly phone calls or text messages for emotion, lifestyle, and BP self‐monitoring may improve intervention uptake [46, 47]. Most participants in our study were educated and there was limited participation by low‐income patients. Tailored SMA programs with peer or lay health coaches may have the potential to address social needs and support lifestyle modifications with linkages to required resources in the community [3]. Future studies addressing social determinants of health and digital literacy that affect the ability to engage in self‐monitoring and lifestyle modification programs are needed. Improving self‐efficacy for managing emotions, lifestyle behaviors, and increasing positive emotions may have the potential for improved long‐term hypertension control and healthy aging due to overlapping modifiable risk factors between dementia and cardiovascular disease risks [19, 48, 49].

6. Conclusions

Our study demonstrates that self‐monitoring of BP, CVH behaviors, and emotions with coping skills and lifestyle education is feasible and effective in improving hypertension control across diverse primary care populations interested in lifestyle modification trials. However, few low‐income patients participated, and time schedules limited participation. Flexible, less time‐intensive programs that consider social determinants of health and patient‐centric self‐monitoring measures may further improve participation in these interventions.

Author Contributions

S.J.P.: Conceptualization, funding acquisition, data curation, formal analysis, investigation, methodology, project administration, resources, supervision, visualization, writing—original draft preparation, writing—review and editing writing. N.G.: Data curation, investigation, validation, formal analysis (quantitative), software, visualization, writing—original draft preparation, writing—review and editing writing. I.T. and E.U.; data curation, validation, formal analysis (qualitative), writing—review and editing. All authors have read and agreed to the published version of the manuscript.

Ethics Statement

The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board (or Ethics Committee) of Cleveland Clinic (protocol code CC 22–895 and date of approval 8/29/22).

Consent

Any research article describing a study involving humans should contain this statement. Informed consent was obtained from all subjects involved in the study.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting information

Data Availability Statement

The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request and permission from Cleveland Clinic IRB.