Association of Mobile Health (mHealth) Use with Health Status and COVID-19-Related Concerns by People with Mobility Impairments

Abstract

Purpose:

mHealth technology has increased dramatically in the wake of the pandemic. Less research has focused on people with mobility impairing (PMI) disabilities. This study determined the prevalence of mobile health (mHealth) use among PMI adults during the COVID-19 escalation and examine demographic, health, and COVID-19 concerns correlates.

Methods:

PMI adults (N=304) completed an online survey investigating mHealth use and COVID-19 concerns related to food access in June of 2020. Smartphone and mHealth use were measured with an adapted version of the survey used in the Pew Internet & American Life project. Descriptive and multivariable analyses were conducted to determine associations of demographics, health status, and COVID-19 concerns with mHealth use. About two thirds (N=201) of the sample were mHealth users (owned a smartphone and engaged in health promoting behaviors with the smartphone; e.g., sought online information, tracked health behaviors, used patient portals).

Results:

Having hypertension was associated with higher mHealth use, and having higher COVID-19 concerns about food access was associated with higher mHealth use. Those who used mHealth were also more engaged with smartphone apps for communication, services and entertainment. Only the association between educational attainment and mHealth use remained significant after adjusting for other covariates in multivariable logistic regression models.

Discussion:

People with mobility impairments continue to need support in the use of mHealth technology to help maximize access to potentially important tools for rehabilitation and health management. There is a need to continue to investigate mHealth and its applications for people with disabilities.

Introduction

People with mobility impairment (PMI) disabilities often have limited access to health services and health information, which placed them at elevated risk from morbidity and mortality associated with COVID-19 [1,2]. PMI face increased risk of severe infection, and more severe disease because of existing comorbidities. Higher risk of health compromising conditions increases the need for services; yet, barriers such as transportation, greater need for support and stigma create additional barriers [3,4]. Increasing availability and innovation of mobile health (mHealth) technology presents a possible solution to bridge the health care gap for hard to reach communities, people with disabilities, and other marginalised and disadvantaged populations [5–10].

Although mHealth technology was developed prior to the pandemic, its use has increased dramatically in the wake of COVID-19-related stay at home recommendations [11]. Providers can conduct examinations and consultations via videoconferencing technology, order health records, receive telemetry, provide health education, and send orders and prescriptions over mobile devices [9,12,13]. In turn, patients can ask questions, check on results, schedule appointments, receive reminders, ask for prescription refills and receive other kinds of health support using mHealth technology [9,12]. In particular, studies on the management of chronic conditions, such as type 2 diabetes, using mHealth applications has grown dramatically in the last decade [14], providing promising evidence. Despite recognised barriers to using mHealth technology (e.g., communication, vision) and subsequent calls for improved access [15], research on the current state of mobile healthcare for people with disabilities is still quite limited [6,16,17].

Population prevalence studies have historically noted lower use of technology among people with disabilities compared to those without disabilities [18]; however, this is rapidly changing. In particular, there has been a dramatic jump in smartphone ownership and use among all segments of the US population [19], including people with disabilities, over 90% of whom own a wireless device, which serves as a key hub of information and communications infrastructure [20]. This may reflect greater accessibility of devices and technology drawing on improvements for universal access [15], in particular, for a growing senior population. In one study investigating smartphone use by disability or impairment type among adults (age 18 and over) with disabilities, ≥75% used devices for multiple functions, including texting, searching the internet, receiving and sending email, and using mobile apps; use of devices to access social media and maps/GPS was slightly lower (≥55%) [20].

Past research on mHealth focusing on PMI is limited, and the pandemic presented a unique opportunity to investigate this relationship during the health challenge of the pandemic. Challenges to PMI’s access to health care may have been exacerbated during the pandemic due to lack of adequate transportation [3], infeasibility of exercising preventive behaviors (e.g., mask wearing, hand washing), or inability to avoid coming into close contact with others who may be infected (support providers, family members), all of which can impede receiving care via traditional means [2,21]. With containment measures in place, more PMI may be increasingly dependent on mHealth technology affording new opportunities to determine how PMI are using mHealth technology to guide future research and practice. The purpose of this study was to determine the prevalence of mobile health (mHealth) use among PMI adults during the COVID-19 escalation and examine potential correlates including demographic factors, health status, and COVID-19 concerns.

Method

Study design

Data were collected using an online survey format, as part of a larger study to develop the Food Environment Assessment Survey Tool (FEAST) instrument [22].

Participants

Eligible (N=304) adults (1) had a chronic health condition or disability causing impaired mobility for a minimum of one year, (2) were at least 18 years of age, (3) live in the United States, and (4) have access to a working phone or computer with internet access. Participants were recruited from social media platforms (Facebook, Instagram, and Twitter), notices via community-serving organizations, and by referral from previous research studies. As previously reported, 58.4% (N=177) heard about the research study from Facebook, 21.1% (N=64) from Twitter, 12.9% (N=39) from the FEAST website, 5.3% (N=16) from Instagram, 1.7% (N=5) from the study Community Advisory Board (CAB), and 0.7% (N=2) from the previous participants; one participant did not respond to the question [22].

After responding to recruitment materials, potential participants were emailed a website link that directed them to the FEAST consent form and screening questions. Those who met inclusion criteria received a separate email invitation with a link to the survey. Eligible participants received up to five emailed prompts to complete the survey over a two-week period. All data were collected in June of 2020.

Measures

Demographic information

Participants completed items assessing gender, ethnicity/race, education, age, and neighborhood area (e.g., urban, rural).

Smartphone Usage

Smartphone use was measured with using 14 yes/no items derived from the Pew Internet & American Life project [23], which we adapted to reflect greater variety of available apps on smartphones. Items measured four domains of respondents’ smartphone usage: mHealth (4 items, e.g., track health behaviors), communication (3 items, e.g., send or receive email), services (2 items, e.g., request transportation services like Uber, Lyft, taxi), and entertainment (5 items, e.g., play games like Candycrush). Participants were coded as mHealth users if they reported using smartphones for at least one of the following: 1) tracking health behaviors such as exercising, sleeping, or eating; 2) looking for health or medical information online; 3) refilling medical prescriptions; or 4) using patient portal apps to set-up doctors’ appointments.

mHealth Tools

Mobile health users indicated the frequency of use and completed five additional items about their mHealth use. The questions measured 1) whether they received any text updates or alerts about health or medical issues, 2) whether they had any software applications or apps that helped them track or manage health, 3) types of health apps they used, 4) whether they had a wearable device to track health or fitness, and 5) if they had had used their cell phone to stay updated on COVID-19 through applications, such as Twitter and Instagram.

Participants who did not endorse mobile phone ownership were not asked about smartphone ownership or mHealth use and were coded as “no” for these items. Likewise, those who did not endorse smartphone ownership were not asked about mHealth use and were coded as not using mHealth tools.

Current chronic health conditions

Participants were asked about the existence of current health problems or conditions, i.e., diabetes, high blood pressure, bronchitis, emphysema, or other lung issues, heart disease, heart failure, or heart attack, cancer, and any other ongoing health problems.

COVID-19-related concerns

Five items assessed COVID-19-related concerns using on a 5-point, Likert-type response scale with response options of completely disagree, somewhat disagree, neither disagree or agree, somewhat agree, and completely agree. Three items assessed difficulties in getting healthy food during the pandemic, safety concerns during shopping, and use of food delivery services. Two items assessed perceived barriers due to a lack of help for grocery shopping and food preparation, and financial concerns caused by job loss or a reduced work hours.

Data Analysis

Descriptive analyses were conducted for demographic information (gender, ethnicity/race, education, age, and residential location), health status, and COVID-19-related concerns. Bivariable analyses using chi-square tests were conducted to determine how mHealth users differed from non-mHealth users in terms of demographics, health status and COVID-19-related concerns. A multivariable logistic regression model was then constructed to determine how mHealth use might be associated with demographics, health status, and COVID-19-related concerns in PMI. The dependent variable was mHealth user status (yes or no), and the covariates were added successively to the model in three blocks: (1) demographics, (2) health status, (3) COVID-19-related concerns. These variables were entered to account first for consistent demographic associations, then health status, then novel COVID-19-related concerns.

Results

Sample Characteristics

Of 304 participants who completed the FEAST Survey [22], about two thirds were male (66.0%; n=200), Caucasian (68.8%; n=209), and lived in a city or large town (62.7%; n=190). Less than half finished some college or junior college, but did not graduate from a four-year college (41.6%; n=126; See Table 1). About half of the participants (52.4%; n=158) reported that they had good or excellent health (Table 3). A considerable proportion had ongoing health problems, such as bronchitis, emphysema, or other lung conditions (20.1%; n=61), high blood pressure (14.8%; n=45), any other problem that was not mentioned (10.9%; n=33), heart disease, heart failure, or heart attack (8.6%; n=26), diabetes (7.6%; n=23), or cancer (0.7%; n=2).

Table 1.

Demographic characteristics of study participants

Variable	mHealth User		Non-mHealth User		Total
	M or n	SD or %	M or n	SD or %	M or n	SD or %	p
Age in years (M/SD)	36.18	8.46	35.77	9.56	36.07	8.72
Gender
Female	71	35.3	32	31.4	103	34.0
Male	130	64.7	70	68.6	200	66.0
Neighborhood Area
A city or large town	126	62.7	64	62.7	190	62.7
A suburb or just outside a city large town	47	23.4	27	26.5	74	24.4
A small town	24	11.9	9	8.8	33	10.9
In the country or a long way from town	4	2.0	2	2.0	6	2.0
Race/Ethnicity
African American/Black	18	9.1	11	11.8	29	9.5
Native American	6	3.0	3	3.2	9	3.0
Asian	8	4.0	2	2.2	10	3.3
Latino/Hispanic	20	10.1	14	15.1	34	11.2
White/Caucasian	146	73.7	63	67.7	209	68.8
Educational Level							*
Some high school, but I did not graduate	15	7.5	13	12.7	28	9.2
High school	48	23.9	33	32.4	81	26.7
Some college or junior college	97	48.3	29	28.4	126	41.6
College graduate or higher	41	20.4	27	26.5	68	22.4

Note.

^*p < 0.05.

Table 3.

Descriptive statistics for smartphone usage by mHealth user status.

	mHealth User		Non-mHealth User		Total
Response	n	%	n	%	n	%	p
Communication
Send or receive email	137	68.2	41	39.8	178	58.6	***
Send or receive text messages	155	77.1	53	51.5	208	68.4	***
Use WhatsApp, Facebook messenger, or other messaging app	112	55.7	17	16.5	129	42.4	***
Services
Request transportation services like Uber, Lyft, Taxi or others	91	45.3	4	3.9	95	31.3	***
Check your bank account balance or do any online banking	102	50.7	13	12.6	115	37.8	***
Entertainment
Play games like Candy crush, Fortnite, Pokémon or others	68	33.8	6	5.8	74	24.3	***
Use a game or app to learn a language like Duolingo	68	33.8	6	5.8	74	24.3	***
Take a picture	136	67.7	30	29.1	166	54.6	***
Access the Internet	139	69.2	33	32.0	172	56.6	***
Interact with social media services like Twitter, Facebook, Reddit, Instagram	137	68.2	32	31.1	169	55.6	***

Note.

^*p < 0.05.

^**p < 0.01.

^***p < 0.001.

Two thirds (n=201) of the sample were mHealth users, endorsing at least one of the four mHealth use-related items. The most frequently endorsed item was the use of online health or medical information (n=141), followed by tracking health behaviors (n=107), the use of patient portal apps (n=92), and refill of medical prescriptions (n=81).

mHealth Users vs. non-mHealth Users

Demographic associations

A chi-square test comparing mHealth users to non-mHealth users revealed a significant association between the highest education level in the household and mHealth use ([X²=3, N=303]=11.369, p=0.010). Education levels were higher among those who were classified as mHealth users than non-mHealth users. No other demographic variables (i.e., age, gender, living in a city, race/ethnicity) were related to mHealth user status.

Health and COVID-19-related concerns associations

Participants with high blood pressure were significantly more likely to use mHealth (17.9% vs. 8.7%, χ²[1, N=304]=4.543, p<0.05), and those with chronic lung conditions were significantly less likely to use mHealth (15.4% vs. 29.1%, χ² [1, N=304]=7.973, p<0.01). Those endorsing more COVID-19 concerns tended to be mHealth users, as well. mHealth users more strongly agreed with finding it difficult to get healthy food (t[301]=2.455, p<0.05), feeling less safe going shopping (t[301]=5.127, p<0.001), using food delivery services more frequently (t[301]=3.147, p<0.01), as well as perceiving fewer people to help with shopping or food preparation (t[301]=1.978, p<0.05) or less money for food (t[301]=2.216, p<0.05; See Table 2.)

Table 2.

Descriptive statistics for current chronic health conditions and COVID-19-related concerns

Response	mHealth User		Non-mHealth User		Total
	N	%	N	%	N	%	p
Current Chronic Health Conditions
Diabetes	17	8.5	6	5.8	23	7.6	-
High blood pressure	36	17.9	9	8.7	45	14.8	*
Bronchitis, emphysema, or other lung conditions	31	15.4	30	29.1	61	20.1	**
Heart disease, heart failure, or heart attack	15	7.5	11	10.7	26	8.6	-
Cancer	2	1.0	0	0	2	0.7	-
Any other ongoing health problem or condition not already mentioned	17	8.5	16	15.5	33	10.9	-
COVID-19-Related Concerns (M/SD)
I have found it more difficult to get healthy food during the COVID-19 pandemic	3.67	1.05	3.34	1.15	3.56	1.09	*
I feel less safe going shopping now that there are social or physical distancing and face covering orders	3.78	0.97	3.16	1.06	3.57	1.04	***
I have used food delivery services (for example, Doordash, Uber Eats) more frequently since the pandemic started	3.76	1.10	3.35	0.96	3.62	1.07	**
I have fewer people to help me shop or prepare food since the pandemic started	3.68	1.03	3.43	1.01	3.59	1.03	*
I have less money for food due to a job loss or a cut in hours caused by COVID-19	3.84	0.95	3.59	0.85	3.75	0.93	*

Note. Responses of “prefer not to answer” and “don’t know” were treated as missing

^*p < 0.05.

^**p < 0.01.

^***p < 0.001.

Smartphone usage associations

As shown in Table 3, notable group differences were found across all dimensions of smartphone usage (communication, services, entertainment); with mHealth users more likely to use smartphones for these purposes. mHealth use was positively associated with types of smartphone communication: emails (68.2% vs. 39.8% for users vs. non-users, respectively; χ²[1, N=304]=22.559, p<0.001); text messaging (77.1% vs. 51.5%; χ²[1, N=304]=20.750, p<0.001); and other messaging apps (55.7% vs. 16.5%; χ²[1, N=304]=42.876, p<0.001). mHealth users were also more likely than non-users to utilise transportation services (45.3% vs. 3.9%, χ²[1, N=304]=54.304, p<0.001) and online banking (50.7% vs. 12.6%, χ²[1, N=304]=42.089, p<0.001). Also, mHealth users were more likely than non-users to have played games (33.8% vs. 5.8%, χ²[1, N=304]=29.003, p<0.001), learned a new language (33.8% vs. 5.8%, χ²[1, N=304]=29.003, p<0.001), taken pictures (67.7% vs. 29.1%, χ²[1, N=304]=40.798, p<0.001), accessed the Internet (69.2% vs. 32.0%, χ²[1, N=304]=38.187, p<0.001), and interacted with social media services (68.2% vs. 31.1%, χ²[1, N=304]=37.952, p<0.001).

Multivariable associations

The logistic regression analysis of associations of participants’ demographics, health status, and COVID-19-related concerns is summarised in Table 4. In Model 1, mHealth use was negatively related to living in a city and positively related to having at least a college degree. After adjusting for health status and COVID-19-related concerns (Model 3), having a college degree was still significantly related to mHealth use (OR = 8.18, 95% CI = 2.84-23.55), but living in a city was not.

Table 4.

Hierarchical logistic regression analysis: Associations of mHealth use with participant demographic characteristics, health status, and COVID-19-related concerns

	Model 1		Model 2		Model 3
Covariate	OR	95% CI	OR	95% CI	OR	95% CI
Age (years)	0.99	(0.94, 1.03)	0.99	(0.95, 1.04)	1.00	(0.95, 1.05)
Gendera	2.40	(0.95, 6.10)	2.29	(0.90, 5.81)	1.68	(0.61, 4.61)
Race/Ethnicityb	0.45	(0.20, 1.02)	0.38	(0.16, 0.90)	0.45	(0.18, 1.11)
Living in a cityc	0.38*	(0.16, 0.88)	0.33*	(0.14, 0.79)	0.56	(0.21, 1.49)
Education
Less than high school (ref)	-	-	-	-	-	-
High school diploma or GED	2.34	(0.52, 10.48)	2.32	(0.52, 10.34)	3.54	(0.70, 17.92)
Some college	1.56	(0.54, 4.48)	1.59	(0.55, 4.62)	2.08	(0.65, 6.60)
College graduate or higher	7.01***	(2.74, 17.88)	7.72***	(2.94, 20.30)	8.18*	(2.84, 23.55)
Living with any chronic health conditionc	-	-	2.06	(0.93, 4.54)	1.86	(0.78, 4.41)
Difficulty getting healthy food during the pandemicc	-	-	-	-	1.40	(0.90, 2.17)
Feeling less safe going to shoppingc	-	-	-	-	1.39	(0.88, 2.20)
Used delivery services more frequentlyc	-	-	-	-	1.34	(0.87, 2.06)
Fewer people to help me shop or prepare foodc	-	-	-	-	0.94	(0.62, 1.45)
Less money for food due to job loss or cut in hoursc	-	-	-	-	0.93	(0.59, 1.45)

Note.

^a0=Female, 1 = Male = 0.

^b0 = Other, 1 = Caucasian/white.

^c0 = No, 1 = Yes.

^*p < 0.05.

^***p < 0.001

Discussion

The findings of this study showed that the transition to greater technology usage during the pandemic may have exacerbated the demographic differences between PMI mHealth users and PMI who are non-mHealth users. Individuals who had graduated college lived in a city were more likely to be mHealth users, consistent with previous reports [19]; however, after adjusting for the health status and COVID-19 concerns, location of residence was not related to mHealth use. Location of residence was not significant after controlling for all independent variables in the final model. Of note, the odds of being an mHealth user were 8 times greater among those who had at least graduated college than for those who did not. This study found low rates of mHealth among individuals with low education, offering an avenue to improve access to the mHealth resources to ultimately minimise social and health disparities.[24] Increased accessibility is needed for mHealth technologies to be more user-friendly and tailored to the needs of people with disabilities [15,24], such as personalised designs [25,26] and features in those devices that allow feasible remote interactions [27]. To the authors’ knowledge, this is the first study to examine mHealth usage among PMI in the United States considering of the impact of the COVID-19 pandemic.

Another contribution of this study is the finding that those who use smartphones for mHealth also tend to get the most out of their smartphones by using communication, services, and entertainment. High engagement is important for success managing chronic conditions with mHealth strategies [28]. mHealth users are more likely to utilise all three features of smartphones that were measured, compared to their non-mHealth user counterparts, suggesting higher mHealth engagement. In general, people with disabilities face greater challenges with day-to-day activities than those without disabilities; increased engagement with the services offered by their phone could have notably improved health and wellbeing both during the pandemic and as we enter the “new normal.” Community education or focused marketing to help users with disabilities exploit the full range of services of their smartphones might help increase access to care, social connectivity and other important goods and services to those who may not realise the power of the broadly available technology [9].

We found no differences in COVID-19-related concerns between mHealth users and non- users. People in both groups had concerns about the pandemic, regardless of mHealth usage. At the relatively early stage of the pandemic during data collection (June, 2020), technology usage may have only been beginning to accelerate (e.g., heavier use on telecommuting, more applications for home-based exercise, social connections) [19].

These findings imply that using mHealth apps may not have alleviated concerns during the pandemic. Jones et al. [6] pointed out in their literature review that mainstream mHealth apps are not fully studied for their usability and effectiveness, a concern echoed by others [9,26,29]. and that most mobile apps are not tailored to people with disabilities, despite the fact that approximately one in four American adults lives with some type of disability [30].

Regardless of the pandemic, reported use of mHealth was mixed among participants with current health conditions. For example, high blood pressure was more common among mHealth users, while lung conditions were more common among non-mHealth users. It may be that some conditions are so common and well-studied with clear remedies, such as hypertension, that mHealth applications for their management are more readily available [7,10,13]. Future research could explore if and how different health conditions or co-morbid symptoms are linked with mHealth usage among PMI, as has been done in the case of type 2 diabetes [14]. There are clear opportunities for innovation in translating best practice care to tailored mHealth application across a range of chronic health conditions affecting the public [6].

Limitations of this study include that a majority of study participants were White and lived in a city or large town, which may not represent findings from indigenous groups or people of color or other underserved communities disproportionally vulnerable to COVID-related health consequences. Similar to a previous study [6], data collection methods of this study may limit participation in the survey for those who have limited access to a working phone or computer with internet capability due to socioeconomic disadvantage. The study only asked about mobile platforms and did not control for the use of other supplementary resources that may affect mHealth usage, such as non-mobile web-based platforms, home-based services, and other tracking devices [31]. People with disabilities may use different types of healthcare resources depending on feasibility, accessibility, and availability of the resources.

On the other hand, there are several strengths that help offset limitations. This study includes data from people from over 80% of states in the US. Also, this study is one of few published empirical studies that provides insights on mHealth usage among people with disabilities [32]. In addition to the prevalence of mHealth usage by study participants, this study examined distinct services via mHealth technology that could improve the day-to-day activities and quality of life of PMI. Our study advances the relevant literature to inform future research, practice and policy.

Conclusion

The limited use of mHealth among people with disabilities is concerning in the midst of a global health crisis [24]. In this study, having less education was associated with lower mHealth use. Results suggested that those who used mHealth tended to have higher engagement with other smart phone apps. This article adds important information to this nascent field investigating mHealth use among PMI [6,9]. Future studies should focus on the use and effectiveness of various mHealth resources on health for PMI by incorporating measures for health outcomes and address healthcare needs to be met via mHealth resources [6,17]. There is a pressing need to continue investigating mHealth, its applications and its effectiveness on health outcomes in the context of disability. Such empirical efforts could result in marked improvements in the health and quality of life of the underserved and underrepresented population of people with disabilities.

Implications for Rehabilitation.

• Many people with mobility impairing disabilities may be missing opportunities for mHealth rehabilitation and health care.

• COVID-19 has widened existing gaps in access and use of mHealth technology among people with mobility impairing disabilities.

• Focused education is needed to help people with disabilities exploit the full range of services of their smartphones to increase access to care, social connectivity and other important goods and services to enhance rehabilitation and health management.

Data availability statement:

The data that support the findings of this study are available from the corresponding author, REL, upon reasonable request.