Analysis the Critical Factors of M-government Service Acceptance: An Integrating Theoretical Model between TAM and ECM

The development of smartphones has an essential role in digital services, which have the advantage of access to health services that can be performed anywhere and anytime. Unfortunately, the diffusion process of smartphone technology in the provision of health services faces the problem of decreasing the actual use of mobile health applications. Previous studies on m-government health services have also focused on using TAM and UTAUT to predict individual user behavior. A gap in the literature has been identified in previous studies. This study investigates the critical factors for the individual acceptance of m-government health services, especially the Peduli Lindungi mobile health service. The two theoretical models, TAM and ECM, were integrated to enhance the body of knowledge in predicting individual behavior in the m-government healthcare context. A quantitative method was used to analyse a total of 200 data. PLS-SEM was used to analyze the outer and inner models. The findings of this study support all hypotheses. The study's findings show that Perceived Usefulness, Perceived Easy to Use, and Satisfaction significantly affect on Intention Use; Perceived Usefulness, Perceived Easy to Use, and Expectation Confirmatory significantly affect on Satisfaction, Perceived Easy to Use has a significant effect on Perceived Usefulness. Expectation confirmation significantly affected Perceived


Introduction
S m a r t p h o n e s p l ay a n essential role in modern health services (Lee et al., 2018;Saare et al., 2019). According to Wang et al. (2019), there are three reasons which has an impact on changes in product purchase transaction patterns to meet needs. Many studies have shown that smartphones can provide more economical remote healthcare as technological devices (Alaiad et al., 2019;Breil et al., 2019;Lee et al., 2018;Saare et al., 2019). The existing literature shows that the use of smartphones in healthcare manifests itself in implementing m-health applications. It is a revolution in e-government, defined as the adoption of ICTs in the implementation system (M. Huda & Yunas, 2016;Napitupulu, 2016Napitupulu, , 2017Rahmadany, 2021).
E-government is a new form of bureaucratic reform to improve the quality of public services for citizens. In addition, e-government aims to improve government performance and meet the public's need for transparency and accountability of government financial information to realize good governance (Napitupulu, 2016(Napitupulu, , 2017. M-health applications are health services that utilize smartphone technology devices through internet access. According to Breil et al. (2019), an M-health application supports self-management of health and offers positive benefits, such as prevention and reminder features, that are more promising and cost-effective. In addition, M-health services have advantages in terms of access to health services that can be provided anywhere and anytime. In developing countries, the m-health application is a solution to the digital divide in health service problems (Hussain et al., 2018). Unfortunately, using smartphones in health services also has a negative impact that causes excessive changes in a person's behavior (Lee et al., 2018). Apart from the negative impact, using smartphones in health services is an innovation to realize good health service governance.
Research on m-health services has recently attracted considerable attention. This is because m-government services have the advantage of rebuilding the performance of e-government services to make them more efficient, effective, and open (Rossel et al., 2006). Unfortunately, despite the potential benefits of m-government health services, few studies have demonstrated their effectiveness (Byambasuren et al., 2018).
M-government health services are facing many challenges, the actual use of health applications that users download from time to time, starting to decline because of problems related to trust, suitability, personalization, and accessibility (Shemesh & Barnoy, 2020) and uses that include narrow functionality, low user engagement, and non-compliance with the target public for m-health applications (Nunes et al., 2019). This provides a negative experience for many patients accessing digital health services (Xie et al., 2020b).
Previous research has predicted the behavior of users in adopting and accepting mobile health services. Some researchers have used the TAM variable (Lee et al., 2018;Nusairat et al., 2021;Saare et al., 2019;Shemesh & Barnoy, 2020;Wang et al., 2019;Xu et al., 2022), UTAUT (Alaiad et al., 2019;Breil et al., 2019;Garavand et al., 2020;Nunes et al., 2019), UTAUT 2 (Yu et al., 2021), and some of these integrated models between TAM and UTAUT (Lee et al., 2018;Saare et al., 2019) Wang et al. (2019) conducted an empirical investigation to explain user acceptance of the mobile health insurance base and the TAM. Only empirical research has investigated user adoption by integrating the TPB, PMT, and personal health differences (Zhang et al., 2020 A recent study needs to be more extensive in exploring the relationship between TAM and ECM variables in the mobile payment context (Sarassina 2022). In addition, this study was conducted in a developing country in Indonesia.
The experience of a developing country is worth reviewing because developing countries use ICT to address critical issues, such as bridging knowledge interests, social access disparities, and decreasing negative consequences on long-term national development (Malisuwan et al., 2016 (Davis, 1989).
Although the TAM has substantial explanatory power in predicting the use of technological products (Nyoro et al., 2015), the construction emphasizes the technological aspect of using new technological products. As a result, TAM construction for predicting the use of technology products needs to be improved. The weakness of TAM, PU, and PEOU is their inability to determine the driving factors influencing their confidence in technology use (Yousafzai et al., 2007). This leads to low descriptive richness in predicting the use of technology products and is considered a significant drawback of the TAM (Poong & Eze, 2015;Poong & Eze, 2008). The TAM must be more robust in ignoring other assessment aspects of new technology products (Yuen et al., 2021). Therefore, the modification of TAM is required.
Previous studies have offered different research frameworks for modifying TAM using other variables or theoretical models. For example (Lee et al. (2018) and Saare et al. (2019) integrated the TAM and UTAUT, and Sarassina ( 2022;Xie et al., 2020) integrated the TAM and ECM.
The expectancy-confirmation model is used to analyze intentional behavior for sustainable use. ECM is a model built based on TAM and ECT (Xie et al., 2020b). According to Sarassina (2022a), the ECM and TAM have the same construct. ECM is a theory that can predict the continued behavior of users when using a new system. Despite the similarities between TAM and ECM, the two models have differences that require attention (Sarassina, 2022a). ECM theory has advantages in explaining the user belief factor in using  Figure 1 shows the variable name abbreviations used in this study to simplify and assign a specific label to each calculated variable.
In the empirical investigation of previous studies, the TAM variable has proven to be a strong model for predicting the individual behavior to adopt and accept technology-based services, such as mobile health applications (Lee et al., 2018;Nusairat et al., 2021;Saare et al., 2019;Shemesh & Barnoy, 2020;Wang et al., 2019;Xu et al., 2022).
In addition, the integration of TAM and ECM was confirmed in a previous study (Sarassina, 2022a;Xie et al., 2020b). His research proves that PU has a powerful influence on satisfaction (β = 0.497, t-value = 12.380, p-value = 0.000), while PEOU also has a strong influence on satisfaction (β = 0.378, t-value = 9.520, p-value = 0.000). Based on this explanation, we propose the following hypothesis: H1: There is a significant relationship between user satisfaction and intention to use m-government services.
H2: There is a significant relationship between perceived usefulness and the intention to use m-government services.
H3: There is a significant relationship between perceived ease of use and the intention to use m-government services.
H4: There is a significant relationship between perceived usefulness and user satisfaction.
H5: There is a significant relationship perceived easy to use and user satisfaction.
H6: There is a significant relationship between expectation-confirmation and user satisfaction.
H7: There is a significant relationship between perceived ease of use and usefulness.
H8: There is a significant relationship between expectation-confirmation and perceived ease of use.
H9: There is a significant relationship between expectation-confirmation and perceived usefulness.

Methods
A quantitative method using an online survey was used. The quantitative method is appropriate for performing an explanatory factor analysis of the relationship between

Figure 2. Recent Investigation between TAM and ECM
Sources: integration of the models from previous works (Sarassina, 2022;Xie et al., 2020)  Sources: adopted and modified from previous works to participate and felt safe. We also choose to use convenience participation in this survey because the respondents who provided the data had to meet the respondents' criteria.
The PLS-SEM approach was used to analyze the outer and inner models using SmartPLS ). An AVE limit value of 0.50 to fulfill the minimal requirements of construct validity (Khwaja & Zaman, 2020;Xie et al., 2020b). The minimum limit value of Acceptable Alpha, more significant than 0.7, was used in prior studies (Wang et al., 2019;Xie et al., 2020b;Zaman et al., 2019). To accept the minimum sample for testing the hypothesis, the number of research samples followed the R-square method guidelines (Hair et al., 2014;Jr. et al., 2021). According to Hair et al. (2014, p. 21), the R-square method calculates the number of hypotheses tested with the minimum R-square chosen. The result of calculating the minimum sample to test the hypothesis was 181 samples. Another study recommended a sample size of 181 to achieve a valid and adequate test using multivariate statistical analysis (Kock, 2018).
The survey collected 209 responses over a period of three months. Before testing the SmartPLS application, we conducted a first test to select outlier data and eliminate the data. The second PLS algorithm test seeks to analyze the selected data.

Respondent Information
Data were collected for three months, and 209 responses were obtained. However, the first PLS Algorithm testing phase found as many as nine data points as outlier data, and 200 data were analyzed for research purposes using SmartPLS software. The results of the descriptive analysis are presented in Table 2 Table 3 presents the values of the loading factor test results for each construct item. Convergent validity was determined based on the loading factor and AVE, while reliability was determined based on CR and CA, as shown in Table 3. The results showed that all the constructs achieved an acceptable score that exceeded the threshold. In the next stage, discriminant validity was measured using the Fornell-Larker criterion test, as shown in Table 4 Table 5 shows the results of testing Source: primary data collection  services. In addition, technology enables users to accomplish given tasks faster, enhances job performance, boosts productivity, raises work effectiveness, and makes work more comfortable (Davis, 1989). The third factor drives intention to use e-government services. The third factor also reflects the user's behavioral intent when it comes to adopting information technology. These findings are consistent with those of previous studies (Eid and Selim, 2020;Sarassina, 2022;Wirtz et al., 2019;Xie et al., 2020). Thus, the increase in individual behavioral intentions toward accepting technology is substantially On the other hand, the findings show that perceived ease of use significantly affects perceived usefulness (H7). Perceived ease of use refers to the user's efforts to learn and use technology (Davis, 1989), whereas perceived usefulness focuses on enhancing performance through technology (Davis, 1989). The analysis of variance suggests that user interface design influences the association between perceived ease of use and perceived usefulness by 43% (Cho & Hung, 2009). The research findings on the influence of PEOU on PU align with those of previous research (Cho & Hung, 2009;Xie et al., 2020b). Research (Xie et al., 2020b) shows that the easy use of shared-nurse innovations can increase their usefulness in health services. The test results show that the CR value is 6.888, with a p-value of less than 0.001, indicating that perceived ease of use strongly influences perceived usefulness. The assumption is that a governmental health service provides excellent benefits if citizens' efforts to obtain this service are minimal. This means that digital services require minimal effort because the operation is designed to be easy to use. Consequently, the benefits of using digital services are also very high. Therefore, the relationship between perceived ease of use and perceived usefulness is justified by the assumption that the less effort required to use a digital service, the more valuable it is.
Furthermore, PU, PEOU, and EC significantly affected satiation (H4, H5, and H6). This study confirms that the TAM variables, including PU and PEOU, encourage user satisfaction with the These results indicate that the satisfaction level of m-health application users will increase if they are satisfied with the technology, which is easy and useful to use. In particular, the perceived benefits of digital services positively impact satisfaction with digital services. There is "added value" gained, or even users feel that there is an increase in performance when using digital services.
Consequently, satisfaction with digital services has increased. The results of this study explain how perceived usefulness and ease-of-use significantly affect user satisfaction. The findings of this study show a positive influence between PEOU and Satis, in line with previous research (Xie et al., 2020), while the positive influence between PU and Satis is in line with previous research (Wilson et al., 2021). TAM variables positively affect satis, in line with previous research (Sarassina, 2022a).
According to Wilson et al. (2021) Handling Committee can conduct accessibility audits to assess the convenience of government digital services. According to Huda et al. (2022), accessibility audit methods can be performed using Web content accessibility guidelines, functional accessibility evaluators, and Nielson usability guidelines. Accessibility audit activities are also a preventive action by the government to prevent problems related to trust, suitability, personalization, and accessibility (Shemesh & Barnoy, 2020). The central government and ministries that develop applications must pay attention to the "added value" of the technology being implemented. Digital services must focus on citizens' needs. The resulting "added value" can build a positive perception of citizens that digital services provide better benefits than traditional services. Added value is also a fundamental reason motivating them to continue sustainably using