Patients with Heart Failure: Internet Use and Mobile Health Perceptions

Abstract

Background  Heart failure is a complex clinical syndrome noted on approximately one in eight death certificates in the United States. Vital to reducing complications of heart failure and preventing hospital readmissions is adherence to heart failure self-care routines. Mobile health offers promising opportunities for enhancing self-care behaviors by facilitating tracking and timely reminders.

Objectives  We sought to investigate three characteristics of heart failure patients with respect to their heart failure self-care behaviors: (1) internet use to search for heart failure information; (2) familiarity with mobile health apps and devices; and (3) perceptions of using activity trackers or smartwatches to aid in their heart failure self-care.

Methods  Forty-nine heart failure patients were asked about their internet and mobile health usage. The structured interview included questions adapted from the Health Information National Trends Survey.

Results  Over 50% of the patients had utilized the internet to search for heart failure information in the past 12 months, experience using health-related apps, and thoughts that an activity tracker or smartwatch could help them manage heart failure. Qualitative analysis of the interviews revealed six themes: trust in their physicians, alternatives to mobile health apps, lack of need for mobile health devices, financial barriers to activity tracker and smartwatch ownership, benefits of tracking and reminders, and uncertainty of their potential due to lack of knowledge.

Conclusion  Trust in their physicians was a major factor for heart failure patients who reported not searching for health information on the internet. While those who used mobile health technologies found them useful, patients who did not use them were generally unaware of or unknowledgeable about them. Considering patients' preferences for recommendations from their physicians and tendency to search for heart failure information including treatment and management options, patient–provider discussions about mobile health may improve patient knowledge and impact their usage.

Background and Significance

Heart failure is a complex chronic condition in which the heart's function to fill or eject blood is impaired. 1 2 With an estimated prevalence of 6 million adults, it is a serious public health concern in the United States. 3 Hospitalization among heart failure patients is particularly concerning as it has substantial financial implications related to health care expenditures. 4 5 It constitutes 79% of the lifetime costs of heart failure care. 6 Moreover, there are approximately 1.25 million hospitalizations with a primary diagnosis of heart failure per year, which costs the U.S. hospitals $13.6 billion annually, making it the fifth most expensive treated condition. 3 7

Lifestyle risk factors for heart failure include physical inactivity, obesity, and smoking. 8 Physicians advise physical activity, weight management, smoking cessation, and medication adherence to heart failure patients as part of their self-care regimen following their diagnosis. 9 However, 83% of all heart failure patients are hospitalized at least once, and 43% are admitted to the hospital at least four times. 10 Hospital readmission rates surpass 20% within the first 30 days of discharge and approach 50% within 6 months of discharge. 11 12 13 While readmissions are complicated by various factors including comorbidities and social determinants of health, they may be greatly reduced by focusing on patient education to address patients' lack of health information and comprehension as well as their adherence to prescribed self-care. 14 15

With the rise in internet and mobile device usage, there have been significant increases in patient use of the internet to find health information and mobile apps to manage their health. Between 2008 and 2017, the proportion of American adults who looked for health information was greater than two-thirds. 16 Patients, particularly those with poorer health status, 17 18 have become more active consumers of health information. 16 19 20 Although online health information can influence patients to consult physicians and ask them about the information they have come across on the internet, 21 the prevalence of online health misinformation and disinformation poses a threat to public health. 22 23

Objectives

Previous studies have researched various home monitoring interventions including websites, telephone services, and wireless sensors to prevent complications of heart failure and improve care efficiency. 24 25 26 And while recent studies have explored mobile health (mHealth) usage and perceptions among heart failure patients, 27 28 29 literature on heart failure patients' experiences using the internet and perceptions of mHealth, particularly activity trackers and smartwatches, for their heart failure self-care remains limited. We sought to investigate three factors in this study: (1) extent to which heart failure patients use the internet to search for heart failure information; (2) extent to which they are familiar with mHealth apps and devices; and (3) their perceptions of using an activity tracker or smartwatch to assist them in managing their heart failure self-care.

Methods

Study Design

This study was part of a multimethod study that comprised qualitative interviews and quantitative surveys about heart failure patients' use of and attitudes toward the internet and mHealth. 30 A sample size of 50 patients was determined to facilitate data saturation for the qualitative component and adequate power for the quantitative component. The multimethod study was designed to function as a bridge to our larger study that seeks to utilize a novel mobile app to issue adherence reminders (e.g., medication, exercise, and weight monitoring) and allow patients to monitor their heart failure self-care. 31

Recruitment

In collaboration with the internal medicine, cardiomyopathy, and cardiology outpatient clinics, study personnel prescreened all patients diagnosed with heart failure at an urban university-based health system in California between February 2018 and September 2018. Patients who were between 50 and 80 years of age and were scheduled for an appointment at any of the three outpatient clinics were eligible for the study. The exclusion criteria included having a known cognitive disability (e.g., dementia), inability to communicate in English, and having visual or auditory impairments to the extent that one could not use a smartphone.

Eligible patients were contacted over the phone prior to their appointments and were provided information about the study. Study personnel continued recruitment until 50 patients agreed to participate in the study. The verbal consent process was conducted over the phone in English with those who were interested in participating. Each participant's New York Heart Association (NYHA) classification, which places heart failure patients into one of four categories based on their physical activity limitations, 32 and their ejection fractions were reported upon enrollment in the study to describe the severity of their heart failure. This information was documented to characterize the study participants according to the severity of their heart failure as NYHA class III or higher and ejection fraction less than 40% are suggestive of marked limitation. Enrolled subjects received a $20 gift card for completing the interview.

Data Collection

Participants were asked about their internet and mHealth usage based on questions adapted from the Health Information National Trends Survey (HINTS), which was formulated by the National Cancer Institute to assess access to and use of health information for cancer patients. 33 Questions regarding internet use were chosen based on two established models of information seeking behavior: Sense-Making Theory and Berry-Picking Model. The Sense-Making Theory describes an individual's tendency to seek new information due to an information gap. 34 The Berry-Picking Model considers the information search process to be a dynamic process that generally does not employ a specific strategy. 35

Our structured interview also included questions about smartphone, tablet, and activity tracker usage and perceptions. These questions were selected from the HINTS with the Technology Acceptance Model, which suggests that perceived ease of use and perceived usefulness determine whether an individual accepts and uses a given technology, as the guiding theoretical framework. 36 In total, the interview guide consisted of 32 potential questions, including 14 open-ended questions ( Supplementary Appendix 1 , available in the online version). Interviews were recorded using a digital voice recorder, and the interviewers wrote patient responses on paper to prevent loss of information from potential recording errors.

Data Analysis

Recorded interviews were manually transcribed using Microsoft Word. Handwritten interview notes that included words and phrases taken by the interviewers (E.L. and K.A.) were used to generate the codes and codebook. For qualitative data analysis, interview transcripts were imported into Dedoose 9.0.90, a qualitative data analysis software application. Handwritten interview notes were analyzed for interviews that had recording errors or were difficult to transcribe due to poor audio quality of the recordings. The first three interviews were coded by A.S. and C.W.A. after which any discrepancies were resolved through a consensus discussion. Remaining transcripts were coded by a single researcher (A.S.).

Through conceptual analysis of the data, general trends and patterns were identified. Potential relationships among the codes were also examined to determine themes across the categories of information. Emergence of patterns in the data facilitated grouping codes into themes. These themes were conceptualized to represent participants' collective experiences and perspectives through their own words.

Results

Demographics

Structured interviews, which took between 2 and 15 minutes, were conducted with 49 heart failure patients as one patient did not complete the interview. Thirteen interviews had recording errors, and poor audio quality prevented transcription of seven interviews. The average age of the participants was 64.2 years (standard deviation: 8.1; range: 50–78; Table 1 ). The majority of the participants were male (67%), of non-Hispanic or non-Spanish origin (81%), and White (66%). Most (57%) of the participants had ejection fractions less than 40%, whereas 34% were determined to be NYHA class III or higher.

Table 1. Demographics of study population.

Characteristics	Values
Age (y; n  = 49), mean (SD)	64.2 (8.1)
Sex ( n  = 49), n (%)
Male	33 (67)
Female	16 (33)
Hispanic or Spanish origin ( n  = 48), n (%)
No	39 (81)
Yes	9 (19)
Race or ethnicity ( n  = 47), n (%)
White	31 (66)
Black or African American	11 (23)
Asian	5 (11)
American Indian or American Native	0 (0)
Native Hawaiian or other Pacific Islander	0 (0)
Education ( n  = 49), n (%)
High school	9 (18)
Some college, associate degree, or trade school	22 (45)
Bachelor's degree	10 (20)
Master's degree or above	8 (16)
Annual income ($; n  = 49), n (%)
0-25,000	15 (31)
25,001–50,000	8 (16)
50,001–75,000	8 (16)
≥75,001	18 (37)
New York Heart Association class ( n  = 44), n (%)
I	4 (9)
II	25 (57)
III	15 (34)
IV	0 (0)
Ejection fraction ( n  = 49), n (%)
≤40%	28 (57)
41–49%	3 (6)
≥50%	18 (37)

Abbreviation: SD, standard deviation.

Internet Use and Website Type

Twenty-six of the 49 participants reported using the internet to look for heart failure information in the past 12 months ( Table 2 ). General search engines ( n  = 14; e.g., Google, YouTube) were the most common type of website used followed by commercial websites ( n  = 7; e.g., WebMD), academic websites ( n  = 6; e.g., Mayo Clinic, American Heart Association), patient resources ( n  = 3; e.g., patient portal), and government websites ( n  = 1; e.g., MedlinePlus, National Institutes of Health). Google was the most mentioned search engine for heart failure information ( n  = 13), whereas WebMD was the most mentioned website ( n  = 6).

Table 2. Types of websites used by participants.

Type of website ( n  = 26)	Values, n (%)
Commercial website	7 (27)
Academic website	6 (23)
Government website	1 (4)
Search engine	14 (54)
Patient resources	3 (12)

Half of the participants who reported using the internet to look for heart failure information recalled using heart failure-related terms such as symptoms ( Table 3 ). Management of heart failure was another common category as four participants looked for information about medications (e.g., Entresto) and two looked for information about monitoring devices (e.g., CardioMEMS, Kardia device). Six participants performed web searches related to procedures and treatments (e.g., pacemaker, defibrillator, stem cell therapy).

Table 3. Categories of information searched by participants.

Category of information ( n  = 26)	Values, n (%)
Heart failure-related medical terms	13 (50)
Medication	4 (15)
Monitoring device	2 (8)
Procedures/treatments	6 (23)

Trust in Their Physicians

For participants who expressed not looking for heart failure information on the internet in the past 12 months, qualitative analysis revealed one overarching theme. Twenty-three participants reported not conducting web searches, and seven of them attributed it to trusting their physicians. When asked about why he has not used the internet to look for heart failure information, one participant responded, “I guess I have faith in my doctors.”

Trust in their physicians to provide recommendations was an important factor as it allowed participants to circumvent the existence of conflicting information on the internet. During his interview, one participant explained,

I trust the doctors who tell me what [I] need to do and what [I] need to look at. And so, I listen to them … [There are] only different opinions [on the internet], and it makes it more confusing and scar y.

Another participant provided similar comments:

I don't [look for heart failure information on the internet] because there's too much controversy. I prefer to go directly to like a doctor or something like that. I mean, I have in the past. You know, it's been way back. But you know, one researcher says one thing, and the other researcher says something else then you really don't know. Being a patient, you really don't know how to put that together, so I would prefer to go directly to a medical doctor who can decipher it and explain it to me.

In addition to recommendations, physicians often provide printed educational materials and resources that are intended to support patients manage their diseases. The trust five participants had in their physicians allowed these materials and resources to become reliable sources of information outside of the hospital. As a result, web searches for heart failure information were thought to be unnecessary for these participants:

I've gone to [the patient portal] to look at different information and messages on there. I haven't done any research.

Mobile Health App and Device Usage

Of the 49 participants, 25 described using health-related apps ( Table 4 ). Eleven of the 25 reported having experience using fitness tracking apps (e.g., Fitbit, Garmin). Ten stated using the preinstalled health app (e.g., Apple Health, Samsung Health) on their smartphone. Nutrition, heart-related, and patient portal (i.e., electronic health record, health information resources, communication with medical care team) apps were used by four, three, and four participants, respectively. Two of the four participants who were unaware of health-related apps prior to their interview expressed interest in learning more about such apps and how to use them.

Table 4. Types of health-related apps used by participants.

Type of health-related app ( n  = 25)	Values, n (%)
Nutrition	4 (16)
Physical activity	11 (44)
Health	10 (40)
Heart-related apps	3 (12)
Nutrition + physical activity	4 (16)
Patient portal	3 (12)
Unaware of apps	4 (16)

Alternatives to Mobile Health Apps

There were participants who used other strategies to track their nutrition and physical activity besides mobile apps. Many of these alternative methods to fitness and diet tracking focused on paper logs: “I use a spiral notebook for my daily weight, blood pressure, and all.” It was particularly common for tracking nutrition: “the food and stuff like that I do those manually.”

Lack of Need for Mobile Health Devices

Thirty-six participants stated they did not own an activity tracker or smartwatch, and 16 of them indicated that they did not see the need to own one. One participant mentioned, “I haven't really had the need for it,” whereas one other said, “I never thought I needed one.” Two different participants used the same words to describe how their physical inactivity contributed to their lack of interest: “I'm not that active.” The latter elaborated on her reasoning:

I just have not [moved into] maintaining that kind of documentation because I can't—I don't walk well, and there's not much to track .

Two other participants used the same words in providing a reason for their lack of activity tracker and smartwatch ownership by indicating that they “don't wear a watch.”

Financial Barriers to Activity Tracker and Smartwatch Ownership

The cost of activity trackers and smartwatches was an inhibiting factor for six participants. One participant complained, “I think they are expensive, and I don't want to spend money on it.” Despite the variety of available activity trackers and smartwatches, financial barriers prevented them from owning the devices: “I looked into all of that, but they're very expensive and I can't afford it.”

Perceptions of Activity Trackers and Smartwatches

Twenty-seven of the 49 participants thought that an activity tracker or smartwatch could help them manage heart failure. On the other hand, 12 participants did not think that they could help them. Five answered with a maybe or possibly, whereas five others expressed that they were unsure.

Benefits of Tracking and Reminders

Thirteen of those who thought that an activity tracker or smartwatch could help them manage heart failure alluded to their specific activity tracking features such as steps. One participant who owns a Garmin activity tracker said,

I don't get in as many [steps] as [my doctors] would like to see me get in, but [it] kinda makes me look at it to say, 'Okay, can I get up and walk for a few minutes?'

Their ability to monitor heart rate and provide exercise reminders was mentioned by seven and five participants, respectively. Among those who mentioned exercise reminders, one participant who owns a Fitbit activity tracker mentioned,

Just tracking my day-to-day to see if, when I'm behind on things and tries to give me a little more umph to do more. And each week, it gives me an overall assessment of what I did for that week, which day was bad, so it just tries to help me with my activity.

Similar sentiments were expressed by another participant who owns a Fitbit activity tracker:

I like it that [it reminds] me … It's a great thing to just try to motivate you. Now, I highly recommend that baby bigtime.

Uncertainty of Their Potential due to Lack of Knowledge

A total of 10 participants did not answer the question about whether activity trackers and smartwatches could help manage heart failure with a “yes” or “no.” There were also five participants who were not knowledgeable about activity trackers or smartwatches and their functions: “I don't like really know what it could do for me, what it's good for, [or] what you use it to do;” “I don't know anything about them;” and “I'm just now hearing about it.” Additionally, one participant who answered that an activity tracker or smartwatch could help said, “If someone tells you how to use it and what you're looking for and stuff, it could be helpful.”

Discussion

Principal Findings

Older adults and patients with chronic conditions have been known to trust their providers for health information and rely on them for medical decision-making, particularly those who do not use the internet for health information. 37 38 Our results are consistent with these trends and our study participants' information seeking behaviors can be explained by another study that found heart failure patients rely heavily on their providers for heart failure information. 39 Our findings also support the Sense-Making Theory, since 26 participants sought information to better understand heart failure. Contrary to a previous HINTS study reporting commercial websites are the most used type of website among people surveyed, 40 participants in this study most commonly used search engines ( Table 2 ). It may suggest a change in health information seeking behaviors as a more recent study found 97% of health information seekers beginning their web searches with search engines. 20 This finding along with their use of various heart failure-related medical terms indicate the lack of a search strategy among our participants ( Table 3 ), which is consistent with the Berry-Picking Model's consideration that information searches are not satisfied by a single search or set of documents but across multiple searches.

Similar to the findings of a prior study informed by the Technology Acceptance Model, 41 results of this study suggest that heart failure patients do not determine their intention to use mHealth solely based on perceived usefulness and perceived ease of use. Participants who do not use mHealth apps to track their nutrition and physical activity utilize other methods such as paper logs as they may be more comfortable and familiar with doing so. In regard to activity trackers and smartwatches, 22 participants were not interested in owning them or were unaware of their existence, which is in concordance with results of another study investigating patient preferences for mHealth resources. 42 For those who had considered purchasing an activity tracker or smartwatch but did not have one at the time of their interview, the cost of activity trackers and smartwatches was a barrier that discouraged them from owning one, which alludes to how technology can exacerbate health inequities. 43 As some health insurance programs offer free or discounted activity trackers, information about these programs may help patients overcome the financial barrier.

Participants who use an activity tracker or smartwatch spoke positively of their ownership citing their ability to track physical activity and send reminders that encourage physical activity. On the other hand, there were eight participants who had not heard of activity trackers or smartwatches and did not know how they could help them manage heart failure. Whereas our findings indicate that heart failure patients use the internet to search for heart failure information, they may not come across information about mHealth solutions during their web searches because mHealth is not considered routine care at this time. 9 Furthermore, the impact of mHealth educational interventions on heart failure knowledge is currently uncertain, 44 and activity tracker data may stimulate fear, uncertainty, and anxiety. 45 However, activity trackers may have the potential to promote heart failure self-care and decrease its severity by providing a method to monitor and encourage physical activity. 45 46 47

Although the literature on physician perspectives on mHealth is limited, one previous study found that physicians including cardiologists think mHealth apps could reduce delays in diagnosis and improve patient outcomes. 48 As indicated by our results and those of previous studies, 28 29 heart failure patients are receptive to mHealth solutions and their features such as tracking and reminders, especially following the recommendation from their physician. 41 Since heart failure patients trust their physicians for recommendations and seek information about treatment and management options, patient–physician discussions about mHealth technologies may improve patient knowledge about their features and how they can help monitor their adherence to heart failure self-care behaviors. Introducing heart failure patients to mHealth may lead to an increase in its usage for heart failure self-care as mHealth users in this study found the tracking feature of mHealth apps and devices useful.

Limitations and Future Directions

Enrollment in this study was restricted to patients from a university-based health system and those between 50 and 80 years of age. Generalization of our results may be affected by the number of interview recordings that had recording errors or were of poor quality and skewed demographics, such as sex, race, and education, in the study sample ( Table 1 ). English proficiency was crucial to comprehend the directions and questions as qualitative interviews were only conducted in English, so the study was confined to English language speakers. This exclusion criterion may explain the disproportionate representation. Future studies should include translation of this interview guide into other languages such as Spanish. Another limitation is the study's completion in 2018 prior to the coronavirus disease 2019 pandemic. Whereas the internet and mHealth may not have advanced significantly in technology or pricing since then, follow-up studies should be conducted to determine any changes in patients' sources of health information, with particular focus on social media and artificial intelligence, and mHealth usage.

Conclusion

This study demonstrated that participants were generally familiar with conducting searches on the internet and the functions of mHealth. Those who used mHealth apps and devices thought they were useful as well. However, there were many factors that influenced their decision to search for health information on the internet and use mHealth. Both participants who did and did not engage in those activities alluded to the importance of understanding their value prior to doing so. One major factor in influencing their self-care regimen was their relationships with their physicians. Owing to the participants' trust in their expertise, many participants relied on recommendations and resources from their physicians. These findings suggest that patient–physician discussions about mHealth in regard to heart failure self-care may improve patient knowledge about their features with potential to influence patient perceptions and usage. Future studies should investigate physicians' perspectives on mHealth with respect to improving adherence to heart failure self-care behaviors and their recommendations.

Clinical Relevance Statement

This study underscores the pivotal role of patient trust in health care providers, which can influence patient self-care in heart failure management. We also focus on the type of mHealth apps and devices heart failure patients use to assist them in managing their heart failure self-care, as well as any barriers to their use. The identification of barriers to the adoption of mHealth such as lack of awareness suggests the importance of patient education in increasing patient knowledge about mHealth-based self-care strategies.

Multiple-Choice Questions

1. What was the most common type of website used by heart failure patients to search for heart failure information on the internet?

   1. Academic websites (e.g., Mayo Clinic, American Heart Association)

   2. Commercial websites (e.g., WebMD)

   3. Search engines (e.g., Google, YouTube)

   4. Government websites (e.g., MedlinePlus, NIH)

   Correct Answer : The correct answer is option c. Participants most commonly used search engines for their web searches, and Google was specifically mentioned as the most used search engine for heart failure information. While academic and commercial websites were also used, search engines were the most frequent choice for obtaining heart failure information online.

2. What was the one reason some heart failure patients reported not searching for health information on the internet?

   1. Lack of access to the internet

   2. Trust in their health care providers

   3. Inability to use the internet

   4. Financial barriers to computing device ownership

   Correct Answer : The correct answer is option b. Many participants who did not use the internet to search for health information attributed this behavior to their trust in their physicians. They believed their physicians provided reliable information, negating the need for online searches.

Funding Statement

Funding This study was supported by the National Heart, Lung, and Blood Institute under grants R56HL135425 and R01HL141773.