Acceptance and Use of eHealth/mHealth Applications for Self-Management Among Cancer Survivors

Abstract

Cancer survivors’ acceptance and use of eHealth/mHealth applications for self-management can be unique and are not fully understood. We used data from the Health Information National Trends Survey 4 Cycle 4 to examine cancer survivors’ acceptance and use of eHealth/mHealth applications for key self-management processes, and conducted logistic regression and Rao-Scott design-adjusted Chi-square tests to assess bivariate associations between potential predictors and actual use. Potential factors were selected based on the Individual and Family Self-Management theory. High acceptance of eHealth applictions was identified, and adoption of mHealth was relatively low. Younger, higher educated, married, employed, and higher income survivors tended to use eHealth/mHealth applications for self-management. Survivors who were newly diagnosed or still on treatment were more likely to look for cancer information online or communicate with health providers electronically. BMI and rural residency were associated with use of mHealth apps to achieve a health-related goal and treatment decision-making.

Introduction

Advances in cancer diagnosis and treatment have led to longer survival. More than 15 million Americans diagnosed with cancer were alive by January 1, 2016 [1]. Similar to persons with other chronic conditions, cancer survivors are expected to perform self-management starting at the point of diagnosis, in order to achieve their long-term care goals. However, cancer survivors may lack the confidence or skills to perform self-management of symptoms, take medications, implement lifestyle changes, and deal with other consequences of cancer [2]. Many factors can affect their long-term engagement in self-management, such as individual, condition-specific, family, and environmental factors [3]. Conceptual descriptions of self-management have identified five core self-management processes, including problem solving, decision-making, resource utilization, partnerships with healthcare providers, and taking action [4]. To accomplish those processes, cancer survivors will need support from family and friends, healthcare professionals, communities, health systems, and possibly information and communication technologies [5].

The development of Web- and mobile-based health-related applications is increasing, with the goal of facilitating health behavior changes and support of patients in chronic disease self-management [6, 7], including limited applications in cancer self-management [8, 9]. Some applications have integrated behavior change techniques, such as goal setting, self-monitoring, and decision support feedback, potentially making them convenient and powerful tools to support self-management behaviors and improve health outcomes [10]. However, a significant body of scientific evidence has not been established for the effectiveness of current web- and mobile-based self-management interventions [6–8, 11]. It has been suggested that the eHeatlh/mHealth system development process should fully assess user perceptions and address their unique needs [9]. More generally, the feasibility and acceptability of using web- or mobile-based technology for health self-monitoring needs additional investigation [7].

The purpose of this study was to understand cancer survivors’ perceptions and actual use of Internet and mobile technology to support their cancer self-management needs, with a specific focus on four self-management processes: (1) resource utilization, operationalized as cancer information seeking and access to personal health information; (2) treatment decision making; (3) taking action, operationalized as achieving health-related goals; and (4) partnership with healthcare providers, operationalized as exchanging medical information with health care professionals. Further, factors associated with survivors’ actual use of technology for self-management processes were explored.

Methods

Data used in the study were from the National Cancer Institute’s Health Information National Trends Survey 4 (HINTS 4) Cycle 4, which is the first cycle of HINTS that included questions about adoption of mHealth applications [12]. The HINTS targets American adults aged 18 and older, to assess their knowledge of, attitudes toward, and use of cancer- and health-related information [12]. Cycle 4 used a single-mode mail survey, with a two-stage sample design, including a stratified sample of addresses and a selected adult within each sampled household [12]. The data were collected from 3,677 respondents from August to November 2014, with an overall 34.04% response rate [12]. This study focused on the subpopulation of individuals who have ever been diagnosed with cancer, of which there were 542 respondents in HINTS 4 Cycle 4.

Guided by the Individual and Family Self-Management Theory [3], potential predictive factors were categorized as contextual factors and self-management process factors, including socio-demographic characteristics, clinical characteristics (diagnosis and treatment), health status, psychological distress, knowledge and beliefs (self-efficacy, belief in cancer cause), self-regulation skills and abilities (preference of shared decision making), and social facilitation (social support and regular exercise pressure from others) (Table 1). As shown in Table 2, outcome measures included cancer survivors’ acceptance of the Internet and mHealth apps, and actual use of the Internet and mHealth apps for health self-management processes.

Table 1.

Summary of Potential Predictive Factors (NH: Non-Hispanic, HS: High School).

Factors	% (Weighted)	Factors	% (Weighted)
Age		Race/Ethnicity
18–34	3.8	NH-White	80.4
35–49	15.1	NH-Black	6.2
50–64	34.8	Hispanic	11.2
65+	46.3	Other	2.2
Gender		Marital Status
Male	37.6	Married	65.2
Female	62.4	Not Married	34.8
Employment		Income
Employed	36.6	< $50,000	46.3
Un-employed	63.4	$50,000+	53.7
Education		BMI
<HS	15.7	Underweight	9.2
HS/Some College	47.9	Normal	31.1
College+	36.4	Overweight/obese	59.7
Rural		Regular Provider
Yes	5.6	Yes	81.3
No	94.4	No	18.7
Cancer type		Co-morbidity
Breast	15.4	0	20.7
Prostate	8.3	1	26.9
Others	76.3	2+	52.4
Years since Diagnosis		Time of Last Treatment
< 1 Year	12.4	On Treatment	12.0
2–5 Years	19.9	< 1 Year	11.5
> 5 Years	67.8	1–5 Years	22.1
Psychodistress		5+ Years	54.3
None	71.3	General Health
Mild	16.6	Good+	80.5
Moderate/Severe	12.1	Fair/Poor	19.5
Self-Efficacy		Social Support
Completely/Very	67.2	A little/Not at all	9.2
Somewhat	26.3	Somewhat	15.2
A little/Not at all	6.5	A Lot	75.6
Everything Causes Cancer		Exercise Pressure
Agree	56.6	A little/Not at all	81.6
Disagree	43.4	A lot/Some	18.4
Decision Making (moderate survival)		Decision Making (low survival)
Self-leading	39.9	Self-leading	50.8
Shared	47.9	Shared	38.7
Doctor-leading	12.2	Doctor-leading	10.5

Table 2.

Acceptance and Use of eHealth/mHealth for Self-Management.

Measure	Level	n (%, weighted)
Acceptance
Access to the internet or send and receive e-mail	Yes	375 (73.9)
	No	154 (26.1)
Trust of online cancer information	Some/A lot	334 (68.9)
	Not at all/A little	146 (31.1)
Inportance of access personal health information electronically	Very	347 (67.7)
	Somewhat	109 (19.4)
	Not at all	62 (12.9)
Having mHealth apps installed in mobile devices	Yes	92 (23.4)
	No apps	168 (36.2)
	No device	232 (40.4)
Interested in exchanging medical information with a HCP electronically	Some/Very	210 (44.7)
	Not at all/A little	260 (55.3)
Actual Use
Seeking online cancer information in the past 12 months	Yes	147 (31.8)
	No	218 (41.7)
	No access to Internet	154 (26.5)
Access to personal health information online in the last 12 months	Yes	126 (29.4)
	No	402 (70.6)
Exchanging medical information with a HCP electronically in the past 12 months	Yes	133 (29.8)
	No	376 (70.2)
Using mHealth apps to achieve a health-related goal such as losing weight, or increasing physical activity	Yes	41 (13.0)
	No	47 (10.1)
	No apps	168 (36.4)
	No device	232 (40.5)
Using mHealth apps to make a decision about how to treat an illness or condition	Yes	31 (7.6)
	No	58 (15.6)
	No apps	168 (36.3)
	No device	232 (40.5)

All statistical analyses considered the complex design of the HINTS 4 sample. The final sample weight variable was used to calculate population estimates, and 50 replicate weights were used to calculate accurate standard errors of the weighted estimates using the jackknife replication method [12]. Specifically, descriptive statistics were used to summarize the cancer survivor population characteristics and their acceptance and use of eHealth/mHealth technologies for self-management. Bivariate logistic regression analyses were used to assess relationships between each potential factor and cancer survivors’ seeking cancer information via the Internet, access to online patient portals (PHR), and exchanging medical information with a health care professional through various technologies. Regression coefficients were estimated by pseudo-maximum likelihood estimation methods and odds ratios were reported. Rao-Scott design-adjusted Chi-square and F-tests were used for cross-tabulations to examine associations between each potential factor and cancer survivors’ using mHealth apps for achieving a health-related goal, and for making a treatment decision. All statistical analyses were conducted using Stata (version 14, StataCorp LP., College Station, TX). The level of significance was 0.05.

Results

Most adult cancer survivors in the US were more than 65 years old, female, non-hispanic white, at least high school educated, currently married, unemployed, and reported their household income to be over $50,000. More than half of survivors were overweight or obese, had the diagnosis of cancer more than 5 years ago, and had at least 2 other chronic conditions in addition to cancer. The majority felt good general health and no distress, were very confident in taking good care of their own health, and had support from their friends or family (Table 1).

Regarding cancer survivors’ acceptance of eHealth/mHealth applications for self-management, most survivors had access to the internet (73.9%), trusted online cancer information (68.9%), and considered it very important to be able to access to their own health information electronically (67.7%). However, more than one-third of survivors did not have a smartphone or tablet (40.4%), and only 23.4% of survivors had the mHealth apps. In addition, less than half of survivors (44.7%) reported that they were somehow or very interested in exchanging medical information with a health care professional electronically (Table 2).

In the past 12 months, a limited number of cancer survivors acually used the Internet to look for cancer information for themselves (31.8%), accessed their personal health information online (29.4%), and had exchanged medical information with a health care professional through email, text message, mobile apps, video conference or social media (29.8%). There were even fewer survivors using mHealth apps to help achieve a health-related goal (13.0%) or make a decision about how to treat an illness or condition (7.6%) (Table 2).

As shown in Table 3, bivariate logistic regression results indicated that age, education, marital status, employment status, income, cancer type, years of diagnosis, time of last treatment, and belief that everything causes cancer significantly predicted survivors’ seeking cancer information online. Specifically, the odds of seeking online cancer information were 4.02 and 2.77 times higher among survivors aged 35–49 and 50–64 years than those aged 65 years and older. The odds were also higher among higher educated, married, employed, higher income survivors. Breast cancer survivors have 2.08 times higher odds of seeking cancer information online than other cancer survivors, while the odds for prostate cancer survivors were 65% lower than other cancer survivors. Cancer survivors who were recently diagnosed or still on treatment had higher odds of seeking online information than those who had been diagnosed for more than 2 years or had their last treatment more than 1 years ago. Other factors, such as gender, race, BMI, number of co-morbidities, health status, self-efficacy, and social support, were not found to be significant correlates.

Table 3.

Significant Predictive Factors of Using eHealth Applications for Self-Management PHR (PHR: Personal Health Records, HCP: Health Care Professionals, HS: High School, OR: Odds Ratio).

Factors	Seeking Online Cancer Inform		Access to PHR		Exchanging Medical Inform. with HCP
	F-Test		F-Test		F-Test
	OR	p	OR	p	OR	p
Age (ref: 65+)
	F(3,45)=6.1; p<0.001		F(3,45)=2.5; p=0.07		F(3,45)=12.3; p<0.001
18–34	5.74	0.22	2.75	0.47	5.22	0.16
35–49	4.02	0.02	2.72	0.12	5.03	0.01
50–64	2.77	<0.01	2.31	0.01	4.20	<0.01
Education (ref: < High School)
	F(2,46)=3.9; p=0.03		F(2,46)=0.2; p=0.80		F(2,46)=3.5; p=0.04
HS/Some College	5.95	0.01	1.55	0.54	5.48	0.02
College+	6.71	<0.01	1.63	0.50	5.99	0.01
Marital Status (ref: Not Married)
	F(1,47)=12.9; p<0.001		F(1,47)=5.4; p=0.02		F(1,47)=8.0; p=0.007
Married	2.99	<0.01	2.65	0.02	2.28	<0.01
Employment (ref: Un-employed)
	F(1,47)=6.9; p=0.01		F(1,47)=5.9; p=0.02		F(1,47) =4.3; p=0.04
Employed	2.53	0.01	2.42	0.02	2.09	0.04
Income (ref: < $50,000)
	F(1,47)=21.0; p<0.001		F(1,47)=15.3; p<0.001		F(1,47)=31.2; p<0.001
$50,000+	3.60	<0.01	4.53	<0.01	4.76	<0.01
Cancer Type (ref: Others)
	F(2,46) =5.9; p=0.005		F(2,46)=0.5; p=0.61		F(2,46) =0.8; p=0.47
Breast	2.08	0.05	1.54	0.32	1.62	0.25
Prostate	0.35	0.04	1.05	0.93	0.87	0.81
Years since Diagnosis (ref: < 1 Year)
	F(2,46)=6.01; p=.005		F(2,46)=1.37; p=0.27		F(2,46)=3.78; p=0.03
2–5 Years	0.26	0.01	0.55	0.18	0.26	0.02
> 5 Years	0.18	<0.01	0.53	0.12	0.29	0.01
Time of Last Treatment (ref: Still on Treatment)
	F(3,45) =7.35; p<0.001		F(3,45)=2.01; p=0.13		F(3,45)=1.23; p=0.31
< 1 Year	1.52	0.47	3.03	0.09	0.67	0.54
1–5 Years	0.24	<0.01	0.84	0.76	0.38	0.11
5+ Years	0.23	<0.01	1.03	0.95	0.41	0.09
Beliefs that Everything Causes Cancer (Ref: Agree)
	F(1,47)=4.66; p=0.04		F(1,47)=0.45; p=0.50		F(1,47)=0.05; p=0.81
Disagree	0.54	0.04	0.81	0.50	0.93	0.82

Access to PHR and exchanging medical information with HCP electronically had similar associations with the socio-demographic predictors, that is, married, employed, and higher income survivors had higher odds of using eHealth applications for self-management. However, education was not found to be associated with survivors’ access to PHR. Cancer survivors who were diagnosed less than 1 year ago had higher odds of using technology for exchanging medical information with HCP. Other cancer-related characteristics, such as cancer type and yeas of diagnosis, and beliefs that everything causes cancer were not found to be associated with either access to PHR or exchanging medical information electronically (Table 3).

The assessment of using mHealth apps for self-management focused on achieving a health-related goal and treatment decision making. As shown in Table 4, younger, higher educated, married, employed, and higher income cancer survivors were more likely to use mHealth apps for self-management. In addition, obese survivors were more likely to use mHealth apps to help achieve a health-related goal, such as losing weight or increasing physical activity, while survivors living in rural areas were more likely to use mHealth to help make treatment decisions.

Table 4.

Factors Associated with Use of mHealth Applications.

Factors	For Achieving a Health-Related Goal		For Making a Treatment Decision
	% of YES Response	Rao-Scott F-Test	% of YES Response	Rao-Scott F-Test
Age
18–34	62.7	F(6.8, 319.4)=4.97; p<.0001	27.8	F(6.4, 299.3)=3.92; p=.0007
35–49	30.4		20.8
50–64	13.2		6.8
65+	2.0		2.8
Education
<HS	0	F(4.6, 215.8)=3.83; p=.003	0	F(4.3, 199.8)=4.18; p=.002
HS/Some College	17.8		12.7
College+	10.6		4.6
Marital Statius
Married	17.1	F(2.5, 116.9)=10.0; p<.0001	9.9	F(2.8, 133.7)=7.9; p=.0001
Not Married	5.3		3.5
Employment
Employed	29.3	F(2.9, 136.5)=10.8; p<.0001	14.5	F(2.7, 128.8)=10.2; p<.0001
Un-employed	4.5		4.0
Income		F(2.8, 132.3)=15.2; p<.0001		F(2.8, 132.3)=15.8; p<.0001
< $50,000	6.5		2.8
$50,000+	18.6		11.6
BMI
Under-weight	11.5	F(7.3, 341.9)=2.03; p=.048	5.0	F(7.2, 336.0)=1.98; p=.055
Normal	10.6		4.5
Overweight	7.0		4.5
Obese	22.4		15.2
Rural		F(2.2, 104.6)=2.68; p=.07		F(2.3, 106.0)=5.14; p=.005
Yes	47.8		47.8
No	11.5		5.5

Discussion

This study described cancer survivors’ acceptance and actual use of web- and mobile-based health applications for self-management, and explored potential factors associated with their actual use. The dataset was from a national mail survey with a complex sample design. Findings of this study are expected to be generalized to the US adult cancer survivor population.

Findings of the study indicated high acceptance of eHealth applications and a relatively low adoption of mHealth apps for self-management. It is believed that the eHealth/mHealth system can provide convenient platforms for patients to engage in self-management, especially in personal health information seeking and management, health communications, and health decision support [13]. However, cancer survivors’ self-management and technology adoption behaviors can be complex and have not been fully understood. This study found that less than one-third of cancer survivors used eHealth or mHealth applications for cancer information seeking, access to personal health information, or exchanging medical information with health care professionals. Cancer survivors have a relatively lower mobile device ownership than American adults (60% vs. 68%) [14]. One possible explanation may be because that cancer survivors tend to be old (46.3% are 65+ years old). Among all cancer survivors, the percentage of mHealth apps owners is low (23%). However, among those survivors who had mobile devices, they were actually more likely to install mHealth apps than general population who have mobile devices (64% vs. 58%) [15]. The proportion of cancer survivors who used mHealth apps to help achieve health goals or treatment decision-making is similar as that in the general US adult population [16].

Although predictive factors of using eHealth/mHealth applications for self-management were slightly different depending on the type of technologies and self-management processes, survivors’ technology use were mainly associated with their socio-demographic characteristics, such as age, education, marital status, employment status, and household income. Younger, higher educated, married, employed, and higher income survivors were more likely to use eHealth/mHealth applications for self-management, which is congruent with the literature [13–16]. Some cancer-specific factors were found to be associated with survivors’ online cancer information seeking, and communication with health care professionals. It is understandable that newly diagnosed cancer survivors and survivors who were still on active treatment have tremendous needs for health information in order to understand the disease, and deal with consequences of diagnosis and treatment [17]. This study also found that cancer survivors who believed that everything causes cancer were more likely to look for cancer information online. This finding is inconsistent with a previous report that beliefs in everything causes cancer was not associated with cancer information seeking in general population [18]. The potential interpretation may be because cancer survivors are more motivated to obtain cancer information in order to understand their diagnosis of cancer than the general population.

Cancer survivors’ use of mHealth apps for self-management was found to be associated with their BMI and rural residency. As more than half of cancer survivors were overweight or obese, and many weight loss apps were integrated in smartphone or available for free download [19], it is understandable that cancer survivors would use mhealth apps to help achieve a health-related goal, such as losing weight or increasing physical activity. Although rural residency was not found to be associated with mHealth app use in the general population [16], cancer survivors living in rural areas seem to benefit from their increased access to health services and health information through mHeath technology [20].

This study did not find any association between race and use of eHealth/mHealth applications. This is not consistent with research on the US general population, which indicates that African American have higher odds of using the mHealth application for achieving health goals and treatment decision making than whites [16]. It is unclear whether cancer health disparities play a role in survivors’ use of mHealth technologies for self-management, which will need to be further explored. It is interesting to notice that self-efficacy, one of potential mechanisms indicated in many self-management interventions, was not found to be associated with cancer survivors’ use of eHealth/mHealth applications for self-management in this study. This finding is inconsistent with the report from the general population [16]. Cancer surviors’ self-efficacy to self-manage can vary widely according to the illness-related tasks [21]. However, cancer-related self-efficacy was not specifically measured in the HINTS survey, which was a potential limitation of the study.

There are a few other limitations in this study. First, only bivariate analyses were conducted to explore associations between potential factors and outcome variables. Some survey questions had a small number of responses from cancer survivors. For example, only 41 survivors reported that they had used mHealth apps for achieving a health-related goal. Therefore, the jackknife replication method was not able to be conducted in a multivariate regression model to calculate design-adjusted standard errors [12]. Although findings of the study tend to be preliminary, significant factors identified by bivariate analyses would contribute to the final selection of potential predictors in future multivariate analyses. Second, although a self-management theory was used to guide the consideration of potential factors for cancer survivors’ use of eHealth/mHealth applications for self-management, due to the limited number of survey questions available, some constructs in the theory could not be operationalized by measures in the survey. It is also possible that some potential factors may not match exactly with theoretical concepts. Last, survey data were collected about two years ago, which may undermine survivors’ technology adoption for the present day.

Conclusion

This study described cancer survivors’ acceptance and actual use of eHealth/mHealth applications for self-management, and revealed high acceptance of eHealth applications and relatively low adoption of mHealth apps for self-management. Less than one-third of cancer survivors used eHealth/mHealth applications for self-management, which may be limited by availability of technologies. A few predictive factors of using eHealth/mHealth applications among cancer survivors were different from those identified for the general population, with the exception of socio-demographic predictors, such as age, marital status, employment, and income. In addition, cancer-specific factors were identified to be significantly associated with survivors’ use of eHealth/mHealth applications, indicating that cancer survivors may have unique needs for their use of technologies for self-management. Future studies with a larger sample size will be needed to further explore the predictive model for cancer survivors’ acceptance and use of eHealth/mHealth applications for self-management.