Promoting new users’ online health consultation services usage behavior strategically

Abstract

Online consultation services have the potential to reduce the workload of healthcare staff, provide timely care to patients, and improve doctor-patient relationships. The COVID-19 pandemic has accelerated the development of these services and platforms, but it remains to be seen whether the general public will continue to use them after the pandemic is under control. This research proposes a framework to examine the factors contributing to UK adults’ continued usage of online healthcare consultation services after COVID-19 restrictions have been lifted. A total of 430 new users completed surveys, and the results indicate that expectation confirmation, system quality, and information quality can positively impact users’ self-efficacy toward using online consultation services. This, in turn, can influence their continued usage behavior. Furthermore, the results suggest that participants’ perception of health risks can moderate the relationship between self-efficacy and continued usage behavior. The strategic implications of these findings are discussed.

Keywords:

• Online health consultantion services
• perceived health risk
• protection motivation theory
• self-efficacy
• self-regulation theory
• usage behavior

Introduction

Online healthcare consultation services have received attention from policy-makers, researchers, and practitioners because they have the potential to change the way healthcare information is exchanged, improve care quality, build better doctor-patient relationships, and reduce the workload of medical staff (Almathami et al., 2020; De Witte et al., 2021; Liu et al., 2022; Lu et al., 2011). For example, a study conducted by Choi and Kim (2014) found that virtual technologies can help patients with high blood pressure manage their diet and lifestyle more effectively than face-to-face appointments. In addition, studies confirmed that telehealth interventions, which can be one of OHCSV’s functions, can effectively reduce patient waiting time (Caffery et al., 2016; Xiong & Zuo, 2022). In this research, the term "online healthcare consultation services" (OHCSV) refers to an internet-based application or website that supports communication between patients and medical staff remotely (Liu et al., 2022). These services usually include ICT (Information and communication technology) functions that enable users to share, transfer, and communicate data or information in realtime from their homes with care providers (e.g., doctors), at a clinical site (Almathami et al., 2020).

Before the start of the COVID-19 pandemic, the number of OHCSV users has increased gradually (Casey et al., 2017). There are several reasons for this growth. First, technological advancements have enabled new e-consultation services to include more useful functions, such as scheduling appointments (Liu et al., 2022). Second, software developers have become more experienced in constructing services for healthcare consultations, making them more user-friendly (Zanaboni & Fagerlund, 2020). Third, users, including the public and medical staff, have become more familiar with digital technology applications, including OHCSV (De Witte et al., 2021; Gagnon et al., 2010; Jung & Padman, 2014). In addition to the three factors mentioned above, the COVID-19 pandemic and the corresponding policies implemented to suppress the spread of the virus, such as travel restrictions and quarantine requirements, have accelerated the use of online services for medical professionals and the public (Darley et al., 2022; Mann et al., 2020). For example, in New York City, the use of telemedicine increased by 8,729% during the COVID-19 pandemic (Ramaswamy et al., 2020). In the UK, the number of users having online doctor consultations has increased from 1.19 million in 2019 to 1.83 million in 2022 (Statista, 2022).

Some studies have investigated the public’s OHCSV usage behavior (e.g., Boehm et al., 2020; De Witte et al., 2021; Kumar, 2022; Yang et al., 2023). However, there is still room for further research. First, existing studies on OHCSV usage have emphasized identifying why some patients, such as those with breast cancer and high blood pressure, adopt or reject these services (e.g., Almathami et al., 2020; Choi & Kim, 2014; Jung & Padman, 2014; Lu et al., 2011; Zanaboni & Fagerlund, 2020). Furthermore, the COVID-19 pandemic has prompted additional research on how it has impacted the usage of OHCSV by both the public and medical professionals (e.g., Boehm et al., 2020; De Witte et al., 2021; Zhou et al., 2020). These studies have contributed to our understanding of OHCSV users, but further research grounded in theories when perceived health risk is high may have added value to the literature and relevant practices. The findings of such research may be more generalizable and transferable. To address gaps in the literature, this study plans to propose a framework to investigate the antecedents of users’ continued OHCSV usage behavior during the COVID-19 pandemic, incorporating the theories of self-efficacy and protection motivation.

Second, since Bandura’s (1977) work, the concept of self-efficacy has received additional attention from researchers in a wide range of subjects, such as education studies, leisure research, and healthcare studies (Brown et al., 2016; Johnson, 1999; Sharma et al., 2022; Shea & Bidjerano, 2010). However, researchers have also suggested that self-efficacy and its antecedents can be further explored in the online service context and healthcare context (Hale et al., 2022; Sharma et al., 2022). Specifically, a gap exists in connecting new users’ personal factors and social surroundings with their OHCSV continued usage behavior. Also, how users’ perception of health risks can moderate the relationship between self-efficacy and continued usage behavior is a question that can be further explored.

Third, the COVID-19 pandemic and the corresponding policies aimed at suppressing its spread, such as travel restrictions and quarantine requirements, have led to an acceleration of the public’s and medical professionals’ use of OHCSV for exchanging medical and healthcare-related information (Casey et al., 2017; Mann et al., 2020; Ramaswamy et al., 2020). However, with COVID-19 vaccination programs being rolled out in some countries, such as the UK and the USA, patients may consider visiting medical facilities to see their doctors in person. Further investigation is needed to understand whether patients will continue using these services for obtaining healthcare information, as they do offer benefits, such as convenience (Almathami et al., 2020; Liu et al., 2022; Lu et al., 2011). Moreover, reports have indicated that medical staff in certain regions, such as the UK, the US, and Japan (Fletcher et al., 2023; Ishikawa, 2023; US Department of Health & Human Services, 2022), have experienced high levels of overwork, and OHCSV might alleviate this by reducing their workload while improving their health and wellbeing.

This research aims to narrow the gap in the online consultation service literature and healthcare communication studies by examining new users’ continued usage behavior of OHCSV once COVID-19-related restrictions have been lifted. This research has three objectives. First, based on the self-regulation theory and protection motivation theory, this study proposes a model to analyze the public’s OHCSV continued usage behavior. Second, the impact of expectation confirmation, system quality, information quality, and social influence on self-efficacy will be analyzed, and the effect of self-efficacy on continued usage behavior will be tested. Third, it plans to examine the ability of perceived health risks to moderate the public’s OHCSV continued usage behavior. Fourth, the implications of this study for medical professionals, software developers, and policymakers will be discussed.

Research context

The workload of primary care service staff in the UK has increased significantly since 2007 (Fletcher et al., 2023). GPs have struggled to meet these challenges, which is partially due to a shortage of staff. This situation has led to increased patient dissatisfaction, declining health and wellbeing of practitioners, and poorer patient-doctor relationships (Edwards et al., 2017). Studies have shown that a third of GPs have considered quitting the profession, and 21% of UK patients have to wait more than a day to see a GP. To address these challenges, the NHS released the "General Practice Forward View" report to encourage GPs to use internet technologies, such as online consultation systems, to meet their needs and reduce their workloads. Various initiatives, such as the GP Access Fund and Clinical Commissioning Group budgets, have been implemented to support GPs in incorporating IT and internet technology services into their practices (Casey et al., 2017). Because of the above initiatives, the usability of later-developed OHCSV has improved significantly when compared to earlier versions, and they are more commonly used by UK medical professionals and the general public (Darley et al., 2022). According to Statista (2022), the number of users having online doctor consultations has increased from 0.48 million in 2017 to 1.83 million in 2022. Based on the discussion above, it is suitable to focus on UK adults when studying the factors that contribute to users’ continued usage of OHCSV after COVID-19-related restrictions have been lifted.

Theoretical background and definitions

Self-regulation theory and protection motivation theory will be used as this research’s overarching theories when studying an individual’s OHCSV continued usage behavior. The theory of self-regulation is rooted in the field of social psychology (Hall & Fong, 2007). According to Vancouver (2008), self-regulation refers to the ability to maintain desired internal states within oneself. Higgins (1998) describes self-regulation as an individual’s capacity to choose and pursue desirable outcomes despite obstacles. Based on these ideas, self-regulation theory and related frameworks explain the outcomes that individuals seek to achieve and the processes they use to attain those outcomes (Pihie & Bagheri, 2013; Vancouver, 2008). Furthermore, when trying to attain desired outcomes and depending on the outcomes’ nature, individuals will utilize different resources, including past experiences, physical abilities, cognitive evaluations, and social support from friends and others (Pihie & Bagheri, 2013; Hall & Fong, 2007).

Researchers propose that the self-regulation theory is applicable to individuals managing their health (Hall & Fong, 2007; Johnson, 1999; Leventhal et al., 2016) and learning new knowledge (Brown et al., 2016; Shea & Bidjerano, 2010). This is because patients and learners have a unique interpretation of their experiences and need to formulate strategies that are suitable for them, such as how to manage pain or use feedback provided by lecturers (Brown et al., 2016; Johnson, 1999). They also decide how satisfied they are with the outcome (e.g. level of discomfort and grade). Additionally, Brown et al. (2016) and Johnson (1999) suggest that confidence in one’s ability to manage the situation – whether it is their health or their academic achievement – is an important factor to consider.

When using the self-regulation theory to study individuals’ behavior, self-efficacy has been identified as a key factor in understanding human behavior, such as how patients cope with physical illness and how learners use online learning platforms (Brown et al., 2016; Johnson, 1999; Shea & Bidjerano, 2010). This is because self-efficacy is an individual’s subjective evaluation of their level of competence in executing certain behaviors and achieving future outcomes, and individuals are more likely to be successful in tasks, such as using OHCSV, when they have confidence in their abilities (Shea & Bidjerano, 2010). In this study, self-efficacy refers to an individual’s confidence regarding his/her ability to use OHCSV (Bandura, 1977; Wood & Bandura, 1989). The consequences of self-efficacy will be explored through a focus on continued usage behavior, which will be measured by the number of times individuals use OHCSV to obtain healthcare information and the amount of time they spend on these services after COVID-19-related restrictions have been lifted (Chen & Peng, 2012; Lau, 2017; Lin et al., 2014)

To examine the antecedents of self-efficacy, this study focuses on system quality, information quality, social influences, and expectation confirmation (Alruwaie et al., 2020; Bates et al., 2006; Demir et al., 2021; Oghuma et al., 2016). In this research, social influence refers to the extent to which others’ (e.g., friends and others) beliefs affect an individual’s decisions to use OHCSV to obtain healthcare information (Alruwaie et al., 2020). Expectation confirmation measures the extent to which an individual’s actual experience using OHCSV confirms their initial expectations (Oghuma et al., 2016). System quality is defined as the ability of OHCSV platforms to offer services that meet users’ needs during consultation sessions and before and after sessions, such as scheduling a consultation session and receiving post-consultation follow-up information (Demir et al., 2021). In terms of information quality, it is defined as the extent to which users believe that the information on and delivered through OHCSV is useful, accurate, current, and trustworthy (Bates et al., 2006; Benotsch et al., 2004; Dedeoglu, 2019).

As documented in the literature, some members of the public avoided visiting medical facilities during the COVID-19 pandemic due to fear of catching the virus. This results in a significant increase in the uptake of OHCSV (Darley et al., 2022; Mann et al., 2020; Ramaswamy et al., 2020). To consider the public’s health concerns, this study adopts the protection motivation theory along with the self-efficacy theory. The protection motivation theory was proposed by Roger (1975) and later refined by Maddux and Rogers (1983). These authors suggest that when making decisions with risky or highly uncertain outcomes, individuals will be motivated to evaluate the severity of the risk, the likelihood of it happening, and their ability to mitigate the risk and protect themselves.

The protection motivation theory has been adopted and applied to various research contexts, such as tourists visiting high-crime destinations and consumers canceling restaurant reservations due to fear of catching COVID-19 (Peng & Chen, 2021; Wong & Yeh, 2009). Although these studies’ contexts and proposed frameworks are different, the central proposition by Roger (1975) and Maddux and Rogers (1983) remains the same: individuals are more likely to engage in protective behavior when faced with a decision that involves a high level of risk and/or has a highly uncertain outcome. This theory is relevant to the current research context as vaccinated individuals can still contract the COVID-19 virus when visiting hospitals and GPs during the COVID-19 pandemic (CDC, 2022). As a result, some members of the public might continue to engage in protective behavior by using OHCSV when seeking medical and healthcare-related information.

This present study focuses on the perception of health risk as a way to understand the public’s motivation to protect themselves by using OHCSV to obtain medical and healthcare-related information during the COVID-19 pandemic (Foroudi et al., 2021; Peng & Chen, 2021; Shin & Kang, 2021). The perception of health risk is relevant in this research context because it applies to situations where insufficient information about behavior may negatively affect an individual’s health and safety (Chen & Chang, 2012; Peng, 2020; Shin & Kang, 2021). Reports have shown that some members of the public have delayed or canceled their appointments, while others have used OHCSV to obtain information due to the perceived health risk associated with visiting medical facilities during the COVID-19 pandemic (The Health Foundation, 2020). According to Shin and Kang (2021) and Peng and Chen (2021), the impact of the perceived health risk on individual behavior can be further explored due to the impact of the COVID-19 virus on an individual’s health. This present research extends the existing literature by examining the prolonged influence of perceived health risk and its ability to moderate individuals’ behavior. In this research, perceived health risk refers to the public’s perception of the impact of contracting the COVID-19 virus as a result of visiting medical facilities (e.g., GP and hospitals) during the pandemic (Shin & Kang, 2021).

Research framework and hypotheses

Based on the theories this research adopts, the following research framework and hypotheses will be tested. To examine users’ OHCSV continued usage behavior, the effect of self-efficacy will be investigated. To examine the antecedents to self-efficacy, this study focuses on the influence of information quality, social influence, system quality, and expectation confirmation. In addition, this study will test perceived health risk’s ability to moderate self-efficacy’s influence on continued usage behavior (Figure 1).

Figure 1. Research framework and hypotheses.

The first hypothesis of this study examines the influence of expectation confirmation on OHCSV users’ self-efficacy. Previous research on technology product continued usage behavior has shown that consumers’ confidence in their ability to learn to use new gadgets and whether their actual experience confirms their initial expectations are two influential factors (Chiu et al., 2021; Susanto et al., 2016). Studies on health-related behavior (e.g., quitting smoking and pain management) have suggested that outcome expectancy can affect an individual’s self-efficacy (Baker & Kirsch, 1991; McDonald et al., 2010); however, further investigation is still needed to understand the nature of this relationship (Williams & Rhodes, 2016). Because newer OHCSV is more useful and has better interfaces than earlier services (Liu et al., 2022; Zanaboni & Fagerlund, 2020), some users who used these services during the COVID-19 pandemic for the first time might find their usage experience better than expected, which, in turn, makes them feel certain about their ability to using OHCSV. The following hypothesis (H1) will be tested:

• H1: Expectation confirmation has a positive effect on self-efficacy.

The second hypothesis tested in this research is the relationship between the system quality of OHCSV and users’ self-efficacy. Previous studies on social network sites or mobile applications have confirmed that system quality is important to users’ subsequent reactions and responses (Chen & Lin, 2014; Hajiheydari & Ashkani, 2018; Lin et al., 2014). Additionally, Yu et al. (2014) suggest that self-management websites with good system quality might enhance patients’ self-efficacy. Although few studies have investigated the relationship between the system quality of OHCSV and users’ self-efficacy, research on OHCSV has shown that poor system quality can deter patients from using these services (Healthwatch, 2021; Zanaboni & Fagerlund, 2020). In other words, some patients might feel they cannot master how to use these services because of poor system quality. Based on the reviewed literature, this research proposes that OHCSV with good system quality, such as easy navigation, can contribute to individuals’ belief that they will be able to quickly learn how to use OHCSV (H2).

• H2: System quality has a positive effect on self-efficacy.

The third hypothesis of this study examines the relationship between social influence and OHCSV users’ self-efficacy. Researchers studying individuals’ participation in health-related behaviors (e.g. sports) and use of DVLA’s (Drivers and Vehicle Licensing Agency) online services have found that social influence is a strong predictor of an individual’s self-efficacy (Alruwaie et al., 2020; Guan & So, 2016; Keith et al., 2015). Social influence, such as social ties, can be an important factor in a patient’s ability to manage their health (Saltzman et al., 2020; Tang et al., 2021), but some researchers suggest it can also be linked to a patient’s ability to engage with online-based health-related information (Anderson-Bill et al., 2011). Given the research discussed above and the increase in OHCSV users since 2019 (Darley et al., 2022; Mann et al., 2020; Ramaswamy et al., 2020), this study proposes that individuals who pay attention to how others obtain healthcare-related information are more likely to believe that they can use OHCSV well (H3).

• H3: Social influence has a positive effect on self-efficacy.

The fourth hypothesis of this research examines the relationship between the information quality of OHCSV and users’ self-efficacy. Existing literature on online services suggests that information quality can be considered as the core value proposition of these services (Alruwaie et al., 2020; Chen & Chang, 2012). Moreover, in the context of eGovernment services, Alruwaie et al. (2020) confirmed that information quality can contribute to citizens’ confidence in overcoming challenges they encounter while using eGovernment services. Researchers have highlighted the importance of perceived information quality when studying online medical and healthcare services (Tao et al., 2017). This is because one of the reasons the general public use OHCSV is to seek accurate and reliable healthcare-related information (Jung & Padman, 2014; Lu et al., 2011). Based on the discussion above, this study proposes that individuals will be more certain about their ability to use OHCSV fluently if they believe that these services provide clear, accurate, and reliable healthcare information (H4).

• H4: Information quality has a positive effect on self-efficacy.

The fifth hypothesis of this research aims to test the effect of self-efficacy on OHCSV users’ continued usage behavior. Previous studies on the adoption of online applications, such as online banking services, have confirmed that individuals who are confident in their ability to use such services will have stronger intentions to use them (Makki et al., 2016; Upadhyay et al., 2022). Experts have suggested that this might also be applicable to the use of OHCSV (Liu et al., 2022; Zhang et al., 2016). Although some studies have examined the effect of self-efficacy, further investigation is necessary because the public’s OHCSV continued usage behavior might have implications for policymakers and medical professionals. Moreover, this research aims to contribute to the literature by focusing on OHCSV users’ actual continued usage behavior, which differs from their behavioral intentions (Norton et al., 2017; Saltzer, 1981). This research proposes that individuals who are confident about their ability to use OHCSV fluently will be more likely to continue to use these services after the lifting of COVID-19 restrictions (H5):

• H5: Self-efficacy has a positive effect on continued usage behavior.

The sixth hypothesis of this research examines the ability of perceived health risk to moderate the relationship between self-efficacy and OHCSV continued usage behavior. Studies have shown that the COVID-19 pandemic and policies implemented to suppress its spread have accelerated public usage of OHCSV (Darley et al., 2022; Mann et al., 2020; Ramaswamy et al., 2020). However, with the lifting of COVID-19-related restrictions, individuals now have the option to visit medical facilities in person, even though it remains an ongoing global health issue Some individuals may still perceive being in an enclosed environment with other people to be risky to their health (Peng & Chen, 2021), even though they have been vaccinated against COVID-19 (CDC, 2022). In the literature on perceived risks, researchers have confirmed that this factor can moderate individuals’ behavioral intentions and actual behaviors (Akram et al., 2019; Lin et al., 2012; Yin et al., 2020). However, whether perceived health risks can moderate the relationship between self-efficacy and behavior in the context of OHCSV continued usage remains to be explored. To contribute to the literature, this study proposes that the effect of self-efficacy on OHCSV users’ continued usage behavior will be moderated by their perception of the impact of contracting the COVID-19 virus as a result of visiting medical facilities during the pandemic (H6):

• H6: The effect of self-efficacy on OHCSV continued usage behavior will be different between individuals with higher perceived health risks and those with lower perceived health risks.

Research methods

Sampling and data collection methods

To achieve the aim of this research, which is to examine the antecedents that contribute to the general public’s continued usage of OHCSV after COVID-19-related restrictions have been lifted, we used Amazon Mechanical Turk (MTurk) to collect data. MTurk provided easy access to a wide and diverse range of participants during the COVID-19 pandemic when social distancing was encouraged by some public health authorities. MTurk is also considered a reliable sample source (Santos & Giraldi, 2017; Sugathan & Ranjan, 2019). Furthermore, because this research focuses on the general public’s usage of online services, using MTurk increased the likelihood of reaching individuals with this experience.

We prepared an online survey before data collection and posted it on MTurk for individuals to participate. Participants who agreed to take part and passed the screening questions were able to fill out the survey after we presented the research’s purpose to them. At this stage, we also provided participants with a description of OHCSV and some of its functions. Other than living in the UK and being over the age of 18 years, eligible participants needed to have used OHCSV to communicate (e.g., virtual consultation and communicating through private messaging) with medical staff (e.g., NHS GPs and nurses) at least once between March 2020 and April 2022 and not have used OHCSV to communicate with medical staff before March 2020. The sampling method we used was non-probability purposive sampling. During the six-week data collection period, we collected a total of 430 usable questionnaires. The demographic breakdown of the sample set can be found in Table 1.

Table 1. Characteristics of the participants (N = 430).

Variables	Traits and purchase behavior	%
Gender	Male	56.5
	Female	43.5
Respondent’s age	Between 18–30 years old	27.2
	Between 31–40 years old	30.7
	Between 41–50 years old	20.5
	Between 51–60 years old	16.5
	61 and above	5.1
Highest education level (completed)	Other / prefer not to say	1.6
	Secondary school up to 16 years	26
	Higher or secondary or further education	24
	College or university	33.5
	Post-graduate degree	14.9

Survey design and measurement items

The study participants completed a survey that consisted of two sections. The first section collected information on their demographic backgrounds, while the second section included 27 questions on the constructs used in this research. Specifically, it included seven questions on information quality adapted from Alruwaie et al. (2020), four questions on social influence from Hung et al. (2012), three questions on perceived health risk from Shin and Kang (2021), four questions on system quality from Lin et al. (2014), three questions on expectation confirmation from Oghuma et al. (2016), three questions on self-efficacy from Chen et al. (2015), and three questions on continued usage behavior from Chen and Peng (2012) and Lau (2017). All questions were measured on a seven-point Likert-type scale. The statements used in this study can be found in Table 2.

Table 2. Measurement items.

Variable/Adopted from	Measurement items
System quality (SQ)/ Lin et al. (2014)	SQ1: The online healthcare consultation services that I used were easy to navigate.
	SQ2: The online healthcare consultation services that I used operated reliably.
	SQ3: The online healthcare consultation services that I used provide various features for me to interact with healthcare experts and staff.
	SQ4: The pages on the online healthcare consultation services that I used are well-formatted.
Perceived health risk (PHR)/ Shin and Kang (2021)	Due to COVID-19-related health concerns…
	PHR1: I feel nervous about going to my GP for a face-to-face appointment.
	PHR2: going to my GP for a face-to-face appointment is a risky decision for my health.
	PHR3: I feel uncomfortable about going to my GP for a face-to-face appointment because of my health.
Social influence (SI)/ Hung et al. (2012)	SI1: Before using online healthcare consultation services, it is important to know what impression it will make to others.
	SI2: My friends and I tend to use the same method to obtain healthcare information from GPs.
	SI3: Before using online healthcare consultation services, it is important to know what others think of people who use them.
	SI4: I tend to pay attention to how others obtain healthcare information from their GPs.
Information quality (IQ)/ Alruwaie et al. (2020)	IQ1: Through online healthcare consultation services, I get the information I need in time.
	IQ2: Information provided by online healthcare consultation services meets my needs.
	IQ3: Information provided by online healthcare consultation services is in a useful format.
	IQ4: Information provided by online healthcare consultation services is clear.
	IQ5: Information provided by online healthcare consultation services is accurate.
	IQ6: Information provided by online healthcare consultation services is up to date.
	IQ7: Information provided by online healthcare consultation services is reliable.
Expectation confirmation (EC)/ Oghuma et al. (2016)	EC1: My experience with using online healthcare consultation services was better than I expected.
	EC2: The service level or function provided by online healthcare consultation services was better than I expected.
	EC3: Overall, most of my expectations from using online consultation services were confirmed.
Self-efficacy (SE)/ Chen et al. (2015)	SE1: I have mastered how to use online healthcare consultation services.
	SE2: I am certain I can use online healthcare consultation services fluently.
	SE3: I was able to quickly learn how to use online healthcare consultation services.
	SE4: I use online healthcare consultation services as well as I would like.
Continued usage behavior (CUB)/ Chen and Peng (2012); Lau (2017)	After COVID-19 restrictions have been lifted,…
	CUB1: I used online consultation services to get healthcare advice (1). Never, …, (7) Always
	CUB2: I used online consultation services to get healthcare advice ___ time(s). (1). 0, …, (7) More than 6 times
	CUB3: I spent ___ at using online consultation services to get healthcare advice. (1). I did not spend any time using online consultation services to get healthcare advice, …, (7) More than 6 hours

Data analysis and results

Model measurement

The software used to analyze the collected data was IBM SPSS Statistics 25 and IBM SPSS AMOS 25. Anderson and Gerbing (1988) two-step approach, which includes confirmatory factor analysis (CFA) and structural equation modeling, was applied. In the first step, CFA was conducted, and the factor loading of the items used exceeded the recommended threshold of 0.7 by Fornell and Larcker (1981). Moreover, the reliability of this research’s items was supported by the squared multiple correlations. The composite reliability (CR) scores, as shown in Table 3, ranged from 0.83 to 0.94, exceeding Hair et al. (2012) recommended threshold of 0.7. To confirm convergent validity, the average variance extracted (AVE) of each construct was reviewed, and they were all above 0.5. Convergent validity was confirmed. Additionally, discriminant validity was established because the shared variance between pairs of constructs was less than the AVE of each variable (Fornell & Larcker, 1981). To rule out the influence of common method variance (CMV), the common latent factor (CLF) method was applied (Podsakoff et al., 2003). The differences between the regression weights with and without the latent variable were similar (<0.20), suggesting that this research was not significantly affected by common method bias.

Table 3. Correlations and descriptive statistics.

	Mean	SD	CrA	CR	AVE	EC	SQ	IQ	SI	SE	CUB
EC	5.49	1.12	.83	.83	.63	.79
SQ	5.46	1.19	.87	.87	.69	.74	.80
IQ	5.46	1.13	.86	.86	.67	.79	.80	.82
SI	2.70	1.28	.94	.94	.70	−.75	.-.77	−.77	.83
SE	5.42	1.22	.90	.90	.75	.79	.79	.80	−.83	.87
CUB	5.57	1.08	.87	.87	.63	.78	.80	.82	−.80	.74	.79

Bold numbers on the diagonal elements are the square root of each construct’s AVE value.

AVE = Average variance extracted; CUB = continue usage behavior; CR = Composite reliability; CrA = Cronbach’s Alphas; EC = expectation confirmation; IQ = information quality; SE = self-efficacy; SI = social influence; SQ = system quality.

Structural model and hypothesis testing results (H1-H5)

Once the measurement model was confirmed (χ2/df = 2.34; RMSEA = 0.055; CFI = 0.968; GFI = 0.905; NFI = 0.946), the structural model was estimated before examining the hypotheses. The results showed that the proposed model fits the data well (χ2/df = 2.82; RMSEA = 0.064; CFI = 0.956; GFI = 0.900; NFI = 0.923). After the structural model was validated, the proposed hypotheses were tested. Expectation confirmation’s positive influence on self-efficacy was confirmed (t = 2.33; β = .46; p < 0.01); therefore, H1 was supported. The result confirms system quality has a positive and significant impact on self-efficacy (t = 2.07; β = 0.44; p < 0.01); therefore, H2 was supported. H3 was not supported because social influence did not have a significant effect on self-efficacy (t = –.32; β = –.11; p > 0.05). The result confirms information quality has a positive and significant effect on self-efficacy; therefore, H4 was supported (t = 5.66; β = 0.35; p < 0.001). Lastly, self-efficacy was shown to have a positive and significant effect on continued usage behavior (t = 19.26; β = 0.86; p < 0.001). Therefore, H5 was supported. A summary of the hypothesis testing results (H1-H5) is provided in Table 4.

Table 4. Hypotheses tests (H1–H5).

Path	Standard estimate (β)	t-value	Results
H1: EC→SE	.46	2.33**	Support
H2: SQ→SE	.44	2.07**	Support
H3: SI→SE	−.11	−.32	Not support
H4: IQ→SE	.35	5.66***	Support
H5: SE→CUB	.86	19.26***	Support

EC = expectation confirmation; CUB = continue usage behavior; IQ = information quality; SE = self-efficacy; SI = social influence; SQ = system quality.

***p < .001; **p < .01; * p < .05.

Moderating effect of perceived health risk (H6)

To investigate the moderating effect of perceived health risk (H6), a multiple group analysis was conducted. Because the mean respondent perceived health risk score was 5.56, participants with perceived health risk scores above 5.56 were classified into the high perceived health risk group (N = 233) and participants with perceived health risk scores below 5.56 were classified into the low perceived health risk (N = 207). Through analyzing the χ2 difference between the unconstrained and constrained models, we confirmed the high perceived health risk group and the low perceived health risk were different at the model level for outcomes related to continued usage behavior (Δchi-square = 56.01, Δdf = 22, p < 0.001). We then confirmed the two groups were significantly different on the hypothesized path (i.e. self-efficacy and continued usage behavior) because the difference in the regression coefficients between the unconstrained and constrained models was significant (Δchi-square = 15.53, Δdf = 1, p < 0.001). Perceived health risk could moderate self-efficacy’s effect on individuals’ OHCSV continued usage behavior at the path level. H6 was supported (Table 5).

Table 5. Perceived health risk’s moderating effect (H6).

Path estimated	Low perceived health risk group β(t)	High perceived health risk group β(t)	Δχ2, Δdf	Moderating effect	Result
SE→ CUB	.48(7.01)***	.93(10.716)***	15.53, 1	Significant	Supported

CUB = continue usage behavior; SE = self-efficacy.

**p < .01. ***p < .001.

Discussion and implications

Because OHCSV has the potential to improve the quality of care, foster better doctor-patient relationships, and alleviate the workload of medical staff, these services have received attention from policy-makers, researchers, and practitioners since the late-2000s. However, despite these benefits, the number of OHCSV users by medical professionals and the public only increased rapidly after the start of the COVID-19 pandemic. As COVID-19-related restrictions are being lifted in some countries where medical staff is facing heavy workloads, including the UK, it is necessary to investigate whether the public, especially individuals who started to use OHCSV during the COVID-19 pandemic, will continue to use these online services.

To achieve this research aim, this study proposes a framework based on the self-regulation theory and the protection motivation theory (Maddux & Rogers, 1983; Pihie & Bagheri, 2013; Roger, 1975; Vancouver, 2008). The findings of this study generally support the notion that individuals will utilize different resources, such as experience and cognitive evaluations, when trying to achieve a desired outcome, which is to manage their health through OHCSV. Moreover, individuals are more likely to engage in protective behavior when faced with a decision that involves a high level of risk and/or has a highly uncertain outcome, such as contracting the COVID-19 virus while visiting hospitals and GPs. After discussing how the findings of this study align with these two theories, the next section further elaborates on the theoretical implications of our hypothesis testing results.

Self-efficacy’s antecedents and consequence

In terms of the antecedents to OHCSV users’ self-efficacy, this research confirms this factor can be affected by users’ perception of information quality, system quality, and previous positive experiences. The findings of this research on information quality demonstrate that individuals who believe they are receiving clear and useful advice through OHCSV are more likely to be confident in their ability to continue to use these services effectively. This study’s findings confirm the proposition that accurate and reliable information is essential for online services to retain users (Alruwaie et al., 2020; Tao et al., 2017). In addition, it aligns with Johnson’s (1999) research that quality information on coping with events that occur during illness provided by medical professionals can be helpful to patients’ self-efficacy in managing their illness. This present study is one of the few to test the influence of information quality on self-efficacy and demonstrate that perceiving good quality healthcare information from OHCSV can improve users’ confidence in using these services effectively.

Although individuals may not choose to use an OHCSV solely because of its good quality system, having an easy-to-navigate interface and reliable platform can lead users to believe they can use it effectively. This study’s results largely support previous research on online platforms, which highlights the importance of having systems that are well-formatted and have various features (Chen & Lin, 2014; Hajiheydari & Ashkani, 2018; Healthwatch, 2021; Lin et al., 2014; Zanaboni & Fagerlund, 2020). Previous literature on healthcare has suggested that expectations can affect a patient’s self-efficacy (Baker & Kirsch, 1991; McDonald et al., 2010). The results of this research demonstrate that expectation confirmation can have a positive influence on self-efficacy in the context of OHCSV. New users in this study were more confident in their ability to use these systems effectively when their usage experience exceeded their expectations. This finding also contributes to the discussion regarding whether or not expectation can affect self-efficacy (Williams & Rhodes, 2016).

Contrary to the hypothesis of this research and deviating from the proposition of the self-regulation theory that individuals will utilize different resources, including other people’s experience and support, when trying to achieve a desired outcome, social influence had no significant impact on the participants’ level of self-efficacy at using OHCSV. Because the process of managing one’s health can be a unique experience for each individual and recently developed OHCSV have many functions, paying attention to how others obtain healthcare-related information might have little impact on users’ beliefs regarding how quickly they can learn to use OHCSV. However, further research will be needed to explore the validity of this explanation.

In terms of self-efficacy’s influence, this research found that self-efficacy can have a positive effect on OHCSV users’ continued usage behavior (H5). In other words, although the general public has the option of visiting medical facilities in person after COVID-19-related restrictions have been lifted, individuals are more likely to use OHCSV frequently if they believe they have mastered the use of OHCSV. This finding is consistent with research on other online service users, such as banking and finance (Makki et al., 2016; Upadhyay et al., 2022). However, by examining actual usage behavior rather than behavioral intentions, this research makes an incremental contribution to the literature on online consultation services and healthcare communication studies.

Perceived health risk’s moderating effect

Perhaps the most interesting aspect of this research is the ability of perceived health risk to moderate the influence of self-efficacy on continued usage behavior. The findings from our analysis show that new users of OHCSV who have a high perceived health risk score differ from those with a low perceived health risk score. However, self-efficacy will have a significant influence on continued usage behavior in either case. From the perspective of the protection motivation theory, users who perceive visiting medical facilities to see their doctors in person as risky to their health will use their ability to mitigate the risk and protect themselves, such as using OHCSV to communicate with their GPs (Foroudi et al., 2021; Peng & Chen, 2021; Shin & Kang, 2021). This can explain the results gathered from participants with a high perceived health risk score. As for participants with a low perceived health risk score, being in an enclosed environment with other people might increase the chance of contracting the COVID-19 virus, and contracting this virus could still have negative health implications for vaccinated individuals (CDC, 2022). These individuals might still find visiting medical facilities in person risky when compared to pre-COVID-19 pandemic times; therefore, they will continue to use OHCSV frequently when they have high confidence in their ability to use these services well.

Alternatively, it is possible that the COVID-19 pandemic served mainly as an accelerator in the growth of OHCSV usage, as the number of users had already been increasing in some regions, including the UK, before the pandemic (Casey et al., 2017; Statista, 2022). While the public may differ in their perception of health risks, the study suggests that once they have developed a strong sense of self-efficacy in using OHCSV, they will continue to use these services regardless. Further research is needed to explore the moderating effect of perceived health risk in more detail.

Practical implications

By incorporating marketing practices into the development and promotion of OHCSV, experts in the medical and healthcare industry can become more effective at raising awareness of health and medical-related issues, changing behaviors, and widening access to healthcare. Health marketing has the potential to contribute to the United Nations’ Sustainable Development Goal 3, which aims to ensure healthy lives and promote well-being for all at all ages (Parkinson & Davey, 2023). In addition to contributing to the literature, this research has four implications for practitioners and policymakers.

First, this study shows that self-efficacy has a significant influence on the continued usage of OHCSV. Considering this result with existing studies’ findings on the benefits of using OHCSV, such as reducing healthcare staff workload, providing timely care to patients, and improving doctor-patient relationships, it seems that the advantages of adopting or expanding the use of OHCSV in hospitals and medical clinics outweigh the drawbacks. For those who have already implemented OHCSV, monitoring users’ perceptions of their ability to use the system can help forecast future usage behavior. Those who are considering purchasing or leasing OHCSV could compare users’ confidence levels when evaluating different systems. Practitioners who promote OHCSV to hospitals and GP clinics, such as sales representatives, can emphasize their product’s ability to enable quick learning, which can contribute to continued usage behavior. Furthermore, since some members of the public have not yet used OHCSV, and the uptake of this technology may continue to increase in the near future, there are still business opportunities for service providers (e.g., selling/leasing OHCSV).

Secondly, to promote the self-efficacy of OHCSV users, medical facilities’ information technology staff and service providers need to ensure that their systems are reliable (e.g., do not crash frequently) and well-formatted (e.g., using appropriate fonts for readability). Additionally, they need to collaborate closely with medical staff, such as doctors and caretakers, as well as patients, to identify useful features and areas for improvement or removal. Furthermore, it is essential to have interface designs that are not only easy to use but also adaptable to the circumstances of different populations, such as incorporating translation functions to accommodate users from diverse language backgrounds.

Some users might want to access OHCSV on their mobile devices (e.g., mobile phones and tablets); therefore, software developers and service providers may also need to ensure that their system is compatible with different operating systems and devices. Another way to improve users’ self-efficacy is to deliver high-quality information through OHCSV, which should be timely, up-to-date, and accurate. Communicating healthcare information online can be different from communicating in person. For example, it might be challenging for doctors to read patients’ subtle body language as the camera tends to focus more on users’ upper body during virtual consultation. Medical facilities using OHCSV should provide regular training sessions for their staff on how to use OHCSV effectively in various situations from patients’ perspectives, such as virtual consultations and responding to patient questions through OHCSV's messaging function.

Thirdly, because OHCSV is still an evolving means of communication with medical professionals, adopting some of the marketing practices used to promote new products when marketing these services to the public could be advantageous. For example, based on the technology adoption curve, medical and healthcare facilities could target individuals who have demonstrated confidence in their ability to use technology products and have consistently been early adopters of new technologies. This information could be obtained through research methods used to study consumers’ experiences and journeys, such as surveys and focus groups.

Once these early adopters have been identified, methods used to promote new technology products could be devised. Hosting workshops and seminars might be beneficial. Furthermore, during these events, practitioners could offer free trials through promotional codes if the service or some of its functions are not free. Alternatively, other incentives, such as coupons from reputable health food stores or gyms, could be distributed. Additionally, collaborating with relatable influencers and encouraging existing users to share positive reviews about the system’s reliability and information quality might be effective methods to market OHCSV to users.

Fourthly, by exploring the effect of perceived health risk on the relationship between self-efficacy and continued usage behavior, it may be beneficial for policymakers, such as legislators and health ministers, to continue investing in and supporting medical professionals in incorporating IT and internet technology services into their practices. Regardless of users’ perception of health risks, they will continue to use OHCSV if they are confident in their ability to use these services. Therefore, the number of OHCSV users may not decline once the global health emergency caused by the COVID-19 pandemic is declared over by the WHO. New funding initiatives to encourage hospitals and clinics to install OHCSV and subsidies for upgrading existing systems may still be useful in regions where the workload of medical staff has significantly increased. Moreover, with advancements in fields such as smart healthcare and precision medicine, it is likely that more digital technologies might be integrated into OHCSV. It will be crucial for policymakers to discuss and update relevant legislation to account for these changes.

On the other hand, policymakers and practitioners still need to be cautious when interpreting perceived health risk’s effect. A global health emergency like the COVID-19 pandemic is not an event that occurs regularly; therefore, additional research is needed to further explore this factor’s effect under different circumstances. Moreover, data collected from the US and Europe suggests the growth of online doctor consultation users has slowed down and many members of the public have not yet used these services before (Statista, 2023a, 2023b). Hospitals and medical clinics likely will need to have a hybrid approach when communicating with patients in the foreseeable future; therefore, investment in conventional (e.g., in-person) consultation facilities/training needs to continue while more patients and medical staff are familiar with OHCSV as well as these services’ limitations and benefits.

Conclusion

This research contributes to the literature and practices relating to OHCSV. However, it has some limitations. First, we collected our data using an online sampling technique during the COVID-19 pandemic, which might have affected the data in unintended ways. Additionally, we cannot provide insight into why some members of the public never use online consultation services to obtain healthcare advice. Therefore, future research may consider testing this proposed model using different data collection methods, such as on-site sampling. Second, individuals aged 60 years or older are underrepresented in this research. Studies have shown that these individuals continue to face challenges and barriers when accessing online services (Xiong & Zuo, 2022). Future studies could focus on these individuals because the digital divide could widen social disparities and lead to feelings of alienation. Third, this research focuses on users in the UK, who have been exposed to OHCSV since the mid-2000s. Future research may want to apply this study’s framework to other economies that are just starting to develop their systems, such as China, India, and Taiwan, as this could improve the framework’s generalizability.

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.