In this study, HUTAUT, D&M models and DOI are based on prior quantitative instruments connected to the research variables and the local context of South African public hospitals adoption was used to build the important factors for the SCT user survey that was conducted. A 5-point Likert scale was used to collect responses, with 1 representing ‘strongly disagree’ and 5 representing ‘strongly agree’. According to what is implied by this scale, the respondents have a lot of selection flexibility (Hung, Tsai & Chuang 2014). After analysis, a survey questionnaire should be included for increased reliability and validity (Odhiambo 2015). The study further reviewed the questionnaires before data collection by distributing them to nurses.

For the new questionnaire, a pilot test was carried out for 50 nurses at Steve Biko Academic Hospital (six emergency nurses, 13 midwives, 14 paediatric nurses, 15 neonatal surgical nurses, and 2 other nurses from different departments). The result was further analysed by principal component analysis (PCA) using the Statistical Package for the Social Sciences (SPSS), specifically, for all measured constructs.

The Kaiser–Meyer–Olkin (KMO) and Bartlett’s tests were the second result of the PCA factor analysis. The KMO statistic ranges from 0 to 1. For factor analysis to be effective, the value of KMO should be close to 1 rather than 0. A value close to 1 indicates that correlation patterns are relatively compact, implying that factor analysis should yield distinct and reliable factors. Accepting values greater than 0.5 is recommended by Kaiser (1974). Furthermore, values between 0.5 and 0.7 are considered mediocre, values between 0.7 and 0.8 are considered good, values between 0.8 and 0.9 are good and values greater than 0.9 are considered superb. Therefore, the KMO statistic value for the data used in this study was found to be 0.918.

A purposive sampling technique was employed to collect the data due from nurses from different wards. This study used the cross-sectional data collection method based on structured questionnaires. Structured questionnaires were provided to healthcare personnel at Kalafong Tertiary Hospital, Steve Biko Academic Hospital, Tshwane District Hospital and Pretoria West District Hospital. These hospitals were chosen based on a few characteristics, including their location in Gauteng and the communities or areas they serve. Furthermore, each hospital is classified as academic, tertiary or district-based hospital. Data were collected over 9 months because of the coronavirus disease 2019 (COVID-19) pandemic. A total of 486 survey questionnaires were distributed; however, the responses were 406 in total. Steve Biko Academic exhibited the highest usable responses (60%) followed by Kalafong Tertiary hospital (15%), Tshwane District hospital (10%) and Pretoria West hospital (5%).

Quantitative analysis is used to determine the strength of correlations between various components in the context of an event. The predictions of a theory are used in this procedure to see whether the hypothesis is viable. The greatest strength of quantitative analysis lies in providing repeatable evidence that supports the theoretical prediction. Furthermore, quantitative analysis can detect minor associations, which would be undetected by human data analysis. The questionnaire data were collected, analysed and validated with SPSS (v26). The variables were then defined and labelled. In addition, before using SPSS, data were imported into Microsoft Excel to check for errors (Esewe 2017). For complex statistical data analysis, a multitude of academics utilise SPSS. The SPSS software package was created for the management and statistical analysis of social science data. Respondents from the listed hospitals in the Tshwane district completed all questions. The study adopted variables and relationships for statistical techniques using SPSS software. For the benefit of data analysis, exploratory factor analysis (EFA), confirmatory factor analysis (CFA) and structural equation modelling (SEM) were applied for a hybrid of factor analysis and path analysis (Maguraushe 2021).

Ethical clearance to conduct this study was obtained from the University of South Africa CAES Health Research Ethics Committee (reference no. 2020/CAES_HREC/081).

Using descriptive statistics, we investigated and analysed summary information on the distribution and central tendency of continuous variables. Mean, minimum, maximum and skewedness values were used to analyse and comprehend the distribution of the collected data. The mean value represents the midpoint of the available range. Skewness is a measure of asymmetry in a set of statistical data from the normal distribution (Pallant 2020). Skewedness is classified into two types: negative and positive (Pallant 2020). The females had the highest representation of 63%, while their male counterparts were 37% of the total sample. Generally, it was difficult to have an equal representation of both gender; therefore, these statistics are acceptable so that the conclusions drawn from this study can be generalised across gender.

A total of 406 usable responses from the total distributed 486 surveys, representing a 74% response rate, were subjected to descriptive analysis in SPSS. Figure 2 depicts the profile of the respondents. The sample exhibited unequal representation of female (63%) and male (37%) nurses because of a larger percentage of female nurses in the surveyed hospitals. There was an unbalanced number of respondents because of COVID-19-related illnesses among these nurses affected the response rate.

Respondents were asked to specify the department they work in. The results as summarised in Table 1 indicated that 32.5% of the respondents work in the surgical department, which is the most represented. The midwifery department followed this with a total of 31.5% of the sample. Emergency and paediatric departments represent 20.2% and 10.6%, respectively. The least represented departments are the neonatal and other departments that were not explicitly mentioned in the questionnaire, each with 4.9% and 0.2%, respectively.

In this study, the outcome of the structural equation model highlighted the relationship between the investigated constructs. The summary was extracted from Analysis of a Moment Structures output of the standardised significance levels obtained after running the hypotheses in Table 2. The levels of the hypothesised relationships between latent variables compromised the underlying causal structure of the SCT implementation. The hypothesised relationship was established according to a ratio of 1.96. This simply means that hypotheses are significant or supported, and if their constructs should be greater than 1.96, then it is rejected.

The study’s findings showed that the effectiveness of information and service had a major influence on the adoption of SCT in healthcare. The study’s findings also imply that strict policies must be adopted to ensure the effective and efficient use of SCT in public healthcare. This study’s findings confirmed that regardless of whether a service is provided by internal agencies or by outside contractors, policies and procedures must guide the service provider. According to Hsu, Yen and Chung (2015), service quality makes it possible to assess the caliber of ICT services offered by healthcare professionals. As a result, it has been made clear how important service quality is when applying SCT in public healthcare, especially when dealing with patients’ vital health information. In order to maximise the use of smart card technologies, adequate regulations are needed to control the operational requirements and service quality of healthcare providers.

The researchers draw the conclusion that the absolute quality of the information in the system is necessary for the adoption of smart cards in public healthcare to be successful. According to Kilsdonk, Peute and Jaspers (2017), the idea of information quality is the integration of various elements, such as human, organisational and technical variables that provide a variety of information in the system. Furthermore, the Department of Health needs to create a clear policy guide based on the study’s background which would make it easier for the public and private sectors to use computer technology in their different settings while maintaining high data quality.

An environmental management system is a set of management principles, operational procedures and performance standards for improving a business’s environmental performance (Angeles 2014). Infrastructure, privacy and security issues must be addressed while making wise decisions to maintain healthcare performance. Without first educating healthcare professionals about the technology, the Department of Health would not be able to decide on the adoption of SCT. Furthermore, of the people in their environment, healthcare professionals are socially conditioned to respond in a variety of ways (Howell et al. 2016). The extensive use of performance construct adopted in HUTAUT, which directly impacts implementation, is said to be directly impacted by technology.

The research’s findings were thoroughly investigated and it was found that BI, communication, compatibility and trialability positively impact the adoption of SCT. According to the studies of the following theories, it can be inferred that these findings have direct theoretical implications:

Maeko’s (Maeko & Van der Haar 2018) HUTAUT model shows that BI affects how people use technology. The mediating construct that is significant for the adoption of SCT is the BI construct as used in the UTAUT model (as HUTAUT is derived from UTAUT theory). In addition, BI is the primary indicator of SCT’s actual use and demonstrates the healthcare professional’s willingness to use it. Intriguingly, the study discovered that BI significantly affects SCT in public healthcare and that it is a valid construct in the HUTAUT model with a significant value of 0.01. As a result, there are higher standards for SCT adoption, including behavioural intent and commitment to using technology to improve the adoption environment for SCT.

System use is a crucial aspect of adopting SCT in public healthcare and has been extensively used in studies by the Delone and McLean IS success model (1992, 2003). Because system use makes healthcare operational processes simpler, it is believed to significantly impact SCT in public healthcare. In addition, it has sophisticated search features, multiple access points and the capacity to digitally combine data from numerous databases worldwide (Acquah-Swanzy 2015). In line with technological adoption and usage, the findings highlight communication, compatibility and trialability as essential elements of Rogers’ (2003) DOI theory. For SCT to be adopted in public hospitals, communication, compatibility and trialability are all crucial components with important values (p = 0.01). This statement backs up Rogers’ DOI model’s theoretical stance. According to the DOI theory, communication is the primary means by which news about innovations such as the SCT is disseminated through contemporary media. Rogers (2003) asserts that the degree of confidence that is dependable on existing principles and the anticipated user requirements determine compatibility. The results of this study corroborate Rogers’ findings from 2003, who discovered that compatibility can accommodate individual preferences using SCT systems, making innovation more adaptable in public healthcare institutions.

The concept and theory that encourages healthcare professionals to adopt novel approaches and adopt goals and values that are compatible with their current work practices are strengthened by this study. As a result, advantages such as data protection, storage capacity, data consistency, access authorisation, data ownership, device usability and privacy concerns become apparent early on in the adoption process (Cripps, Standing & Prijatelj 2012). The results of this study also support Rogers’ (2003) DOI theory, according to which compatibility is a key factor in the introduction of new products such as the SCT in public healthcare facilities.

In addition, according to Rogers (2003), trialability – or a system’s ability to be tested or trialled before making firm decisions – is a crucial element for technology implementation. According to the study’s findings, trialability significantly affects SCT implementation, which is consistent with Rogers’ (2003) research. This study addressed the importance of trialability in SCT adoption and confirms the work of Dos Santos Brito (2014). It was found that trialability in DOI addresses the adoption of technology in healthcare from data inputs, data storage and data processes on portable devices and servers to maintain personal data for smart health. The findings of this study are also consistent with those of Tsai and Chang (2016), who discovered that a system’s adoption is the only factor that determines whether it succeeds or fails. This covers the creation and administration of policies as well as software planning and development.