Exploring the predictors of home telehealth uptake by elderly Australian healthcare consumers

Abstract

Background

Despite the significant access and cost-saving potential of telehealth, the uptake of telehealth services in Australia has been sporadic. Understanding the factors that drive the uptake of home-telehealth services from the consumer perspective has received scant attention in the literature.

Aim

The aim of this study was to explore how a comprehensive set of factors may influence the intention of older Australians to adopt home telehealth services.

Methods

A survey of 306 Australians aged between 50 and 68 years was conducted to examine the influence of six categories of predictors on the intention of older Australians to adopt home telehealth: (a) demographics, (b) health status and usage, (c) mobility and ease of access to healthcare, (d) technology usage and anxiety with technology, (e) telehealth attitudes, and (f) personality traits.

Results

Hierarchical regression analysis revealed that significant predictors were: trust in telehealth (β = 0.35); the technology acceptance model (β = 0.27); healthcare habits (β = −0.20); dissatisfaction with traditional healthcare (β = 0.19) and online behaviors (β = 0.09). The model explained 63% of the variance in intention to adopt home telehealth.

Conclusion

This study is the first of its kind in Australia and provides valuable insight into the factors which impact consumer’s intention to adopt telehealth services.

Keywords

Telehealth uptake consumer predictors

Introduction

The ageing population in Australia is poised to impact significantly on health care services through increasing demand and rising healthcare costs. Delivering such services into the home will become paramount not just to contain costs, but to meet the expectations of older Australians who prefer to remain in their own homes, particularly those in remote areas or with restricted mobility.¹ Telehealth has the potential to contribute significantly to the delivery of health care to the ageing population, addressing equity of access to services and the needs and expectations of this large segment of the Australian population.

Despite the significant access and cost-saving potential of telehealth, the uptake of telehealth services to date has been sporadic and generally small in scale. There is often underutilization of telehealth, even when the requisite infrastructure has been made available.² In Australia, the use of telehealth outside the public hospital system is negligible, with very little uptake in the homecare health sector.²

Considerable research has been conducted to understand the key factors that influence and drive practitioner use of telehealth services.^3,4 However, research focused on understanding the factors that drive uptake from the consumer perspective, beyond the technical aspects of the interaction, has received limited attention in the literature. A series of research papers from Taiwan, Slovenia, and China involving in-depth focus groups and structural equation modelling have found key factors such as perceived ease of use of a telehealth system,^5–7 the perceived usefulness of the telehealth intervention,^5–7 effort expectancy⁵ and anxiety and security concerns⁵ are related to the intention to adopt telehealth services. To date, to the best of our knowledge, no empirical research has been conducted in Australia to examine and model broad consumer determinants which predict intent to adopt telehealth services.

With a focus on Australian aged health care consumers, this study explores how a comprehensive set of factors may influence the intention to adopt home telehealth services. Drawing on the extant telehealth and consumer psychology literatures, we identified and examined six categories of predictors: (a) demographics; (b) health status and usage; (c) mobility and ease of access to healthcare; (d) technology usage and anxiety; (e) telehealth attitudes; and (f) personality traits. Understanding consumer preferences regarding home telehealth and the factors that drive the adoption of this service delivery mode is critically important in order to develop services to meet this emerging health care challenge both locally and globally.

Methods

Participants

A total of 306 community dwelling men and women aged between 50 and 68 years (baby boomers born between 1946 and 1964) were recruited into the study. Qualtrics, a web-based data collection and analytics company, was engaged to provide a panel of consumers across all Australian states and territories with an equal gender split and an equal number of Australians from metropolitan versus regional, rural or remote areas. Apart from age, no additional inclusion or exclusion criteria were applied. Data collection was administered through an online survey tool in December 2014 and the study was approved by the appropriate ethics committee.

Data collection

A comprehensive explorative survey was developed to examine potential factors which may impact on aged consumers’ intentions to adopt home telehealth services, which served as the dependent variable (DV) of interest. Intention to adopt telehealth services was calculated as the composite score of five items (seven-point Likert scale) related to the possibility, likelihood, interest, and intention of the participant to adopt telehealth services if they were available. The internal consistency of the items that made up the DV was found to be high (α = 0.96). An example item is “I intend to adopt home telehealth.”

The survey included a broad range of potential predictor variables drawn from existing research on telehealth and consumer psychology. These variables represented six key categories: (a) demographics; (b) health status and usage; (c) mobility and ease of access to healthcare; (d) technology usage and anxiety with technology; (e) telehealth attitudes; and (f) personality traits. Details of the survey measures are provided in Table 1.

Table 1.

Details of survey measures.

Predictor variables	Description
Group 1 – Demographics
Geographical Location, work status, education, income	A series of questions to elicit simple demographic details from participants.
Group 2 – Health status and usage
Chronic health condition	The presence and number of chronic health condition/s the participants suffers from. Recoded into a dichotomous indicator of presence of chronic illness (yes = 1, no = 2).
Consults a medical specialist	Whether the participant regularly consults with a specialist medical specialist (yes = 1, no = 2).
Traditional healthcare habits (α = 0.90)	Three items assessing the participant’s tendency to choose conventional face-to-face healthcare options.
Dissatisfaction with traditional healthcare (α = 0.74)	Three items assessing dissatisfaction with conventional face-to-face healthcare services such as waiting time.
Group 3 – Mobility and ease of access to healthcare
Distance to hospital, travel confidence, lives alone/with others	Questions assessing the distance of the participant from a hospital (in km), how confident they are in travelling to services (1 = “not confident at all” to 5 = “completely confident”) and if they live alone or with others (alone = 1, with others = 2).
Group 4 – Technology usage and anxiety with technology
Online behaviours (α = 0.51)	Seven items assessing frequency of engaging in a range of online activities including email, social media, communication tools (e.g. Skype), search, and games. A seven-point Likert scale ranging from ‘never’ (1) to ‘daily’ (7) was used. An example item is “How often do you engage in the following: Search the Internet for information.”
Technology anxiety (α = 0.93)	A six-item scale assessing feelings of apprehensiveness and anxiety with technology and the tendency to avoid or hesitate using technology.⁸ An example item is “I feel apprehensive about using technology.”
Group 5 – Telehealth attitudes
Technology acceptance model (TAM) (α = 0.96)	The TAM is an information systems theory related to how users come to accept the use of a technology.⁹ A seven-item measure adapted to assess the perceived ease of use, usefulness, social interaction, and affordability of home telehealth services was used. An example item is “Learning to use home telehealth technology appears easy.”
Convenience	A single item measure assessing the potential for telehealth services to increase the convenience of receiving medical services.
Trust in telehealth (α = 0.98)	Nine items adapted from the Behavioral Trust Inventory¹⁰ to examine the willingness to trust in telehealth services. An example item is “I would feel comfortable depending on health services delivered through telehealth.”
Group 6 – Personality traits
Risk aversion (α = 0.80)	Six items assessing comfort with unfamiliar situations and the predisposition to take chances in situations where the outcome is unknown. These questions examine an individual’s predisposition or attitude towards taking risks.¹¹
Personal proactivity (α = 0.93)	Bateman and Crant’s¹² 17-item measure of proactive personality was used to assess an individual’s propensity to proactively influence their environment. An example item is “I am constantly on the lookout for new ways to improve my life.”

Data analysis

Hierarchical multiple regression was conducted using the SPSS statistical package. Data were first checked for normality, outliers, and multi-collinearity. All scales had acceptable internal reliability (Cronbach’s alpha >0.70), with the exception of online behaviour (alpha = 0.51). This was a measure of the frequency of engaging in common online behaviours and hence not expected to have a high internal reliability. Variables were entered into a hierarchical regression in three blocks to identify the relative influence of each block of variables. To control for demographic variables, these were first entered in Block 1. Block 2 consisted of variables assessing health status and usage, mobility and ease of access to healthcare, and technology usage and anxiety. Block 3 consisted of variables assessing telehealth attitudes and personality traits (see Table 1). A significance level of p < 0.05 was set for all analyses.

Results

Sample

A total of 306 Australians between the ages of 50 and 68 years participated in the study. An equal number of men and women participated, with 50% from metropolitan regions and the remaining 50% split between regional (30%), rural (19%) and remote (1%) regions, representing all Australian states and territories. Most of the participants were able to independently drive (75%), had a regular general medical practitioner (GP) (89%) and lived within 10 km of a hospital (54%) or GP (80%). Approximately two-thirds of participants had at least one chronic health condition (64%) with 18% of participants reporting three or more chronic conditions. Approximately one-third of participants had a regular specialist doctor (38%). All participants with the exception of one had access to the Internet at home with the highest proportion having an Asymmetric digital subscriber line (ADSL) Internet connection (73%) followed by mobile (3 G/4 G) (24%). The most prevalent technology available in the home was the laptop computer (70%) and desktop computer (68%), followed by smart phones (55%) and tablet computer technology (36%). Participants represented a range of educational backgrounds (41% high school or less, 34% vocational training, 25% university degree), income (<AUD25 k to >AUD100 k) and work status (58% retired, 37% working, 5% unemployed).

The results of the hierarchical regression are shown in Table 2. Model 1 was not statistically significant (F_4,301 = 1.55, p = 0.19) indicating that the demographic variables were not significant predictors, explaining just 2% of the variance in the dependent variable. Model 2 consisted of all Block 1 factors plus the factors related to health status and usage, mobility and ease of access to healthcare, and technology usage and anxiety with technology. Model 2 was statistically significant (F_9,292 = 12.46, p < 0.001) and explained 29% of the variance in the dependent variable, representing a 27% increase over Model 1. The final model consisted of all previous variables plus factors associated with telehealth attitudes and personality traits. This model was significant (F_5,287 = 27.65, p < 0.001) and explained a total of 63% of the variance in the DV, representing an increase of 34% over Model 2. Five variables were significant predictors in this final model: trust in telehealth (β = 0.35, p < 0.001); the technology acceptance model (TAM)(β = 0.27, p < 0.001); healthcare habits (β = −0.20, p < 0.001); dissatisfaction with traditional healthcare (β = 0.19, p < 0.001) and online behaviours (β = 0.09, p = 0.018).

Table 2.

Hierarchical multiple regression models determining the impact of multiple factors on the intention to adopt home telehealth services.

Independent variables	Dependent variable
	Model 1		Model 2		Model 3
	β	SE	β	SE	β	SE
Metro vs regional, rural, remote	−0.04	0.16	−0.02	0.15	0.02	0.11
Work status	−0.04	0.05	−0.06	0.05	0.01	0.04
Income	0.07	0.07	0.03	0.07	0.00	0.05
Education	0.06	0.07	−0.03	0.06	0.01	0.04
Chronic health condition			−0.08	0.16	−0.07	0.12
Consults a specialist medical practitioner			−0.07	0.16	−0.07	0.12
Traditional healthcare habits			−0.33 ^a	0.07	−0.20 ^a	0.05
Dissatisfaction with traditional healthcare			0.22 ^a	0.06	0.19 ^a	0.05
Distance to hospital			0.06	0.10	0.04	0.08
Travel confidence			0.04	0.08	−0.06	0.06
Lives alone or with others			0.02	0.18	−0.00	0.13
Online behaviours			0.20 ^a	0.07	0.09 ^b	0.05
Technology anxiety			−0.04	0.07	0.06	0.06
TAM					0.27 ^a	0.09
Convenience					0.08	0.04
Trust in telehealth					0.35 ^a	0.08
Risk aversion					−0.02	0.06
Personal proactivity					0.05	0.07
R²	0.02		0.29 ^a		0.63 ^a
Δ R²			0.27 ^a		0.34 ^a
F	1.55		9.27		27.65

: standardized regression coefficients; SE: standard error; TAM: technology acceptance model.

p < 0.01, ^bp < 0.05.

Discussion

This study examined a comprehensive set of factors for their influence on the intention of older Australians to adopt home telehealth services. Results of hierarchical regression analyses support the view that a range of factors significantly predict the aged healthcare consumers’ intention to adopt home telehealth services. Specifically, we found the strongest predictors were the TAM (β = 0.27, p < 0.001) and trust in telehealth (β = 0.35, p < 0.001). These results are consistent with a study by Tsai⁷ who also found that trust and the TAM had a strong relationship with telehealth usage intention in Taiwan. Our results extend understanding by also showing that traditional healthcare habits, dissatisfaction with traditional healthcare, and online behaviours also predict telehealth adoption intentions.

Our findings are perhaps unsurprising considering the technical nature of telehealth solutions: the TAM is specifically designed to evaluate the acceptance of new technologies and it is intuitive that those who more frequently engage in online activities are more likely to embrace technology for health interactions. Likewise, it makes good sense that consumers who feel that telehealth is a trustworthy platform through which to receive health care services will be more likely to adopt home telehealth, and that patients who are dissatisfied with traditional health care may seek alternatives whereas those that habitually seek traditional face to face health services will be less inclined to adopt telehealth.

However, our study also reveals several counter-intuitive and surprising results. Telehealth is considered to be most relevant for those living rurally and remotely given the greater difficulty accessing conventional in-person services. Our modelling shows however, that neither geographical location (metro vs regional, rural or remote), nor distance from a hospital, were significant predictors of intention to adopt telehealth services. Equally surprising was the finding that the presence of a chronic disease – which implies increased dependence on medical care – was unrelated to adoption intentions, as was the need to consult with a specialist medical practitioner. Convenience is another factor often associated with telehealth interventions. While convenience was correlated with adoption intentions (r = 0.49, p < 0.001), this factor was non-significant in the regression model. Finally, our findings suggest that personality factors that might be expected to be associated with telehealth adoption (risk aversion and proactivity) are not significant predictors.

This study had a number of limitations which must be noted. First, the data collection for this study was performed online thus biasing the sample frame to members of the community who have an online presence. It can be argued that these are the very people for whom home telehealth is most relevant, however it is conceivable that the predictors for older Australians who do not have an online presence may be quite different from those reported in this study. Second, our sample frame was restricted to those aged between 50 and 68 years of age. As health care usage increases with age it is also important to understand the predictors associated with telehealth uptake in the more elderly population. Future research should use a varied data collection strategy and consider increasing the age range of the sample frame.

This study is the first of its kind in Australia and provides valuable insight into the factors which impact on older consumer’s intention to adopt home telehealth services. This research has relevance for both the planning and the marketing of new homecare telehealth services in Australia and beyond.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

References

Morris

Mueller

Jones

. Tomorrow’s elders with disabilities: What the wireless industry needs to know. Journal of Engineering Design 2010; 21: 131–146.

Smith

Armfield

Croll

. A review of Medicare expenditure in Australia for psychiatric consultations delivered in person and via videoconference. J Telemed Telecare 2012; 18: 169–171.

Jacobs

Weiner

Reeve

. Determining the predictors of innovation implementation in healthcare: A quantitative analysis of implementation effectiveness. BMC Health Serv Res 2015; 15: 6–6.

Wade

Eliott

Hiller

. Clinician acceptance is the key factor for sustainable telehealth services. Qual Health Res 2014; 24: 682–694.

Cimperman

Brencic

Trkman

. Older adults' perceptions of home telehealth services. Telemed J E Health 2013; 19: 786–790.

Deng

Liu

. Comparison of the middle-aged and older users' adoption of mobile health services in China. Int J Med Inform 2014; 83: 210–224.

Tsai

. Integrating social capital theory, social cognitive theory, and the technology acceptance model to explore a behavioral model of telehealth systems. Int J Environ Res Public Health 2014; 11: 4905–4925.

Meuter ML, Bitner MJ, Ostrom AL, et al. Choosing among alternataive service modes: An investigation of customer trial of self service technologies. J Mark 2005; 69: 61–83.

Davis FD. Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quart 1989; 13: 319–340.

10.

Gillespie N. Measuring trust in organizational contexts: An overview of survey-based measures. In: Lyon F, Mollering G and Saunders M (eds) Handbook of Research Methods on Trust. Cheltenham: Edward Elgar, 2012, pp.175–188.

11.

Bell DE. Risk, return, and utility. Management Science 1995; 41: 23–30.

12.

Bateman TS and Crant JM. The proactive component of organizational behavior: A measure and correlates. J Organ Behav 1993; 14: 103–118.