1 The feasibility of telemedicine based consultation in the primary healthcare sector in Johannesburg Karishma Jivan Mistry WITS Business School Research report presented in partial fulfilment for the degree of Master of Business Administration to the Faculty of Commerce, Law, and Management, University of the Witwatersrand April 2021 2 DECLARATION I Karishma Jivan Mistry declare that this research report entitled ‘The feasibility of telemedicine- based consultation in the primary healthcare sector in Johannesburg’ is my own unaided work. I have acknowledged, attributed, and referenced all ideas sourced elsewhere. I am hereby submitting it in partial fulfilment of the requirements of the degree of Master of Business Administration at the University of the Witwatersrand, Johannesburg. I have not submitted this report before for any other degree or examination to any other institution. Karishma Jivan Mistry Signed at Johannesburg on 30 April 2021 Name of candidate Karishma Jivan Mistry Student number 451825 Telephone numbers 071 048 9009 Email address karishmajivan@yahoo.com First year of registration 2019 Date of proposal submission 16 November 2020 Date of report submission 30 April 2021 Name of supervisor Patrick Zhuwao and Kambidima Wotela 3 ABSTRACT Author: Karishma Mistry Supervisor Patrick Zhuwao Thesis title: The feasibility of telemedicine based consultation in the primary healthcare sector in Johannesburg Healthcare system in South Africa is overburdened and telemedicine is considered a useful tool to make quality healthcare more accessible. In South Africa, regulations imposed by Health Professional Council of South Africa (HPCSA) and restricted access to technological, educational, economic and sociocultural factors has limited the ease of accessing and using telemedicine. In order for telemedicine to gain prominence, it needs to be integrated into everyday practice and be seen as an alternative to face-to-face consultation. Hence, the purpose of this study was to evaluate a combination of factors, in the context of Covid-19 pandemic, namely telemedicine as an engaging platform, seeking written or verbal informed consent for clinical procedures and using information and communication technology (ICT) and availability and knowledge of electronic resources to conduct virtual consultation. In addition, the method used to store patient information was examined to understand the prominence it has in the context of telemedicine. A quantitative, cross-sectional research strategy was used to collect data from 80 family physicians practicing in either private, state or both sectors. This study reported that a large number of participants have used some form of telemedicine, although only one third use it daily. On average the study population reported to agree that telemedicine is an engaging platform, but half the respondents found that the explanation concerning the disease or treatment is not better. Verbal consent is used more than written consent for activities relating to clinical examination or when using ICT. While, majority of the respondents indicated that they do have an electronic device to implement telemedicine consultation, only half the respondents had access to quality broadband network. The study also reported that about two-thirds of the respondents reported to stored patient information manually. Overall, more attention is required to make the functionality of telemedicine more noticeable. Telemedicine may not be valued so much when there is good health infrastructure, however, when health infrastructure is poor, telemedicine is unable to stand on its own. Hence, associated economic, education and socio cultural factors needs to be evaluated further so that behavioural barriers are reduced and telemedicine is used more frequently. 4 TABLE OF CONTENTS DECLARATION ....................................................................................................................................... 2 Abstract .......................................................................................................................................... 3 Table of contents ........................................................................................................................................ 4 List of tables .......................................................................................................................................... 6 List of figures .......................................................................................................................................... 7 ACKNOWLEDGEMENTS .................................................................................................................... 8 Definition of key terms and concepts ..................................................................................................... 9 Introduction to the research ................................................................................................................... 10 1.1 Background and context ..................................................................................................... 10 1.1.1 Defining telemedicine ........................................................................................... 13 1.2 Research conceptualisation ................................................................................................. 13 1.2.1 The research problem statement ......................................................................... 13 1.2.2 The research purpose (aim and objectives) statement ..................................... 15 1.2.3 Research questions ................................................................................................ 16 1.3 Significance of the research study ...................................................................................... 16 1.4 Delimitations and assumptions of the research study .................................................... 17 1.5 Limitations ............................................................................................................................. 18 1.6 Preface to the research report ............................................................................................ 18 2 Literature review............................................................................................................................... 19 2.1 Research problem analysis .................................................................................................. 19 2.1.1 Introducing telemedicine ...................................................................................... 20 2.1.2 Challenges to implement telemedicine in South Africa .................................. 21 2.2 Research knowledge gap analysis ....................................................................................... 24 2.2.1 Performance expectancy: Engaging platform ................................................... 25 2.2.2 Social influence: Written informed consent ...................................................... 26 2.2.3 Facilitating condition: Availability & knowledge of resources to conduct telemedicine .................................................................................................................................. 27 2.2.4 Effort expectancy: Storing patient information ................................................ 28 2.3 Unified Theory of Acceptance and Use of Technology (UTAUT) ............................. 29 2.4 Summary and conclusion .................................................................................................... 32 2.4.1 Summary of literature reviewed .......................................................................... 32 3 Research strategy, design, procedure and methods .................................................................... 33 3.1 Research strategy .................................................................................................................. 33 3.2 Research design .................................................................................................................... 34 3.3 Research procedure and methods...................................................................................... 34 3.3.1 Research data and information collection instrument(s) ................................ 34 3.3.2 Research target population and selection of respondents .............................. 36 3.3.3 Ethical considerations when collecting research data ...................................... 38 3.3.4 Research data and information collection process ........................................... 38 3.3.5 Research data and information processing and analysis ................................. 39 3.4 Research strengthens—reliability measures applied ....................................................... 40 3.5 Research weaknesses—technical and administrative limitations .................................. 41 4 Presentation of research results ..................................................................................................... 42 5 4.1 Socio-Demographic Characteristics of Participants ....................................................... 42 4.2 Telemedicine usage .............................................................................................................. 43 4.3 Telemedicine as an engaging platform .............................................................................. 46 4.4 Informed consent preference (written or verbal) to perform a clinical examination or when using ICT ............................................................................................................................... 48 4.5 Availability and quality of essential resources to conduct telemedicine ...................... 51 4.6 Modalities used to store patient information ................................................................... 53 4.7 Associations - one way Analysis of Variance (ANOVA)............................................... 54 4.7.1 Differences of engagement means between age groups ................................. 54 4.7.2 Differences of engagement means between gender groups ........................... 56 4.7.3 Contingency Analysis of gender by ease of use ................................................ 58 4.7.4 Contingency Analysis of age by ease of use ...................................................... 59 5 Discussion of reserch findings ....................................................................................................... 61 5.1 Introduction .......................................................................................................................... 61 5.2 Socio-Demographic Characteristics of Participants ....................................................... 61 5.3 Telemedicine as an engaging platform .............................................................................. 63 5.4 Informed consent preference (written or verbal) to perform a clinical examination or when using ICT ............................................................................................................................... 65 5.5 Availability and quality of essential resources to conduct telemedicine ...................... 67 5.6 Modalities used to store patient information ................................................................... 70 6 Summary, conclusions, limitations, and recommendations ...................................................... 73 Introduction ...................................................................................................................................... 73 6.1 Conclusions ........................................................................................................................... 73 6.2 Limitations ............................................................................................................................. 74 6.3 Recommendations (action plan) ........................................................................................ 75 6.4 Areas for future research ..................................................................................................... 76 References ........................................................................................................................................ 78 Appendix 1: Questionnaire ..................................................................................................................... 80 Appendix 2: Letter to participate ........................................................................................................... 85 Appendix 3: Ethics approval ................................................................................................................... 86 Appendix 4: Histograms: Differences of engagement means between age groups ....................... 87 Appendix 5: Histograms - Differences of engagement means between gender groups ............... 90 Appendix 6: Knowledgeability of telemedicine amongst age groups ............................................... 92 Appendix 7: Bio ........................................................................................................................................ 93 6 LIST OF TABLES Table 1: Socio-demographic representation of the study population Page 42 Table 2: Telemedicine usage of the study population Page 43 Table 3: Telemedicine as an engaging platform Page 46 Table 4: Information exchange during a telemedicine consultation Page 47 Table 5: Summary Statistics on written number of procedure Page 49 Table 6: Summary Statistics on verbal number of procedure Page 50 Table 7: The efficiency of telemedicine software Page 51 Table 8: Preference for ease of using telemedicine, learning to operate and usefulness Page 52 Table 9: Storage of patient information Page 53 Table 10: Efficiency of storing patient information Page 53 Table 11: Tests that the Variances are Equal Page 54 Table 12: Oneway Anova Summary of Fit for Engaging platform and age Page 55 Table 13: Means for Oneway Anova Page 55 Table 14: Means and Standard Deviation Page 55 Table 15: Analysis of Variance Page 55 Table 16: Connecting Letters Report Page 56 Table 17: Tests that the Variances are equal Page 56 Table 18: Pooled t Test Page 57 Table 19: Contingency Table measuring ease of use and gender Page 58 Table 20: Pearson correlation test Page 58 Table 21: Contingency Table measuring ease of use and age Page 59 Table 22: Pearson Correlation Tests Page 60 7 LIST OF FIGURES Figure 1: Root causes to deteriorating health care system in South Africa Page 12 Figure 2: UTAUT model Page 30 Figure 3: How often telemedicine is used Page 44 Figure 4: Modalities used by study population Page 44 Figure 5: Usefulness of telemedicine as per the study population Page 45 Figure 6: Study population knowledge about telemedicine Page 45 Figure 7: Percentage of respondents that find telemedicine to be an engaging platform Page 47 Figure 8: Study population preference on informed consent Page 48 Figure 9: Written number of Procedures Page 49 Figure 10: Verbal number of Procedures Page 49 Figure 11: Study population representation on tools used by doctors for telemedicine Page 51 Figure 12: Study population representation on tools used by patients for telemedicine Page 52 Figure 13: Test to determine if variances are Equal Page 54 Figure 14: Oneway Analysis of Engaging platform by Age Page 55 Figure 15: Test to determine if variances are equal Page 56 Figure 16: Mosaic Plot measuring ease of use and gender Page 58 Figure 17: Mosaic Plot measuring ease of use and age Page 59 8 ACKNOWLEDGEMENTS A special thanks goes out to all medical doctors who participated in the survey. When the survey was launched (January 2021), South Africa was experiencing its second-wave of the Covid-19 pandemic, putting doctors under immense pressure in treating patients. While this was a challenging period, my heartfelt gratitude goes to each and every participant for taking time out to answer the survey. I would also like to take the opportunity to thank Mr Hennie Gerber for his assistance and guidance in analysing the data. Lastly, I would like to express my sincere gratitude and appreciation to my supervisor, Mr Patrick Zhuwao for his valued advice, guidance and support. DEDICATION I dedicate this work to my late father who always believed in me and supported me in every milestone achieved. 9 DEFINITION OF KEY TERMS AND CONCEPTS 1. Telemedicine: the practice of medicine using electronic communications, information technology or other electronic means between a health care practitioner in one location and a patient in another location (Chifamba, 2018). 2. Informed consent: According to the Health Professions Council of South Africa Guidelines For Good Practice In The Health Care Professions Seeking Patients’ Informed Consent: The Ethical Considerations (Book 4, page 14), informed consent is an exercise of an informed choice by a patient who has the capacity to give consent: in instances where there are multiple options or alternatives to treatment; or in making a decision whether to withhold or disclose information or allow someone else to disclose information on their medical condition to a defined third party; or in making a decision for purposes of reimbursement by a Medical Scheme. 10 INTRODUCTION TO THE RESEARCH 1.1 Background and context The healthcare system in South Africa is overburdened due to underlying factors associated from the apartheid period between 1948 and 1993 (Delobelle, 2013). To date, majority of the South African population rely on public health facilities and unfortunately access to quality service is scarce for this population (Delobelle, 2013). As part of the National Development Plan aimed to achieve equity, efficiency, effectiveness and quality of healthcare provision by 2030; the Department of Health released a National Digital Health Strategy for South Africa in 2019 (page 8) that sets out a vision of ‘better health for all South Africans enabled by digital health’ (National Digital Health Strategy for South Africa 2019 - 2024, 2019) One technique of digital health, called telemedicine, a form of information and communication technology used to deliver a health care service virtually, can be considered a useful tool to make quality health care more accessible (Cilliers & Flowerday, 2013) and an effective solution to alleviate the issues faced by the current health care system in South Africa. There is a prospect that telemedicine can, insofar, open the door to healthcare service delivery that is of better quality, highly accessible, and more efficient (van Dyk, 2014) to South African communities, particularly the rural and underserved areas. This can be supported by Jack and Mars (2013) research that stated that telemedicine is considered to be a cost-effective and efficient tool to deliver health care to under-resourced areas. However, implementing telemedicine does come with diverse challenges (Cilliers & Flowerday, 2011). The literature review (chapter 2) conducted for this study identified the root causes for the deteriorating healthcare system that are shown in Figure 1 (page 13). According to Cilliers and Flowerday (2011), limited access to technology can be attributed to the costs associated with required ICT infrastructure such as internet connectivity, high-end videoconferencing systems and sophisticated medical devices. Cilliers and Flowerday (2011) further mentions that the current state of infrastructure often leads to frequent interruptions of electricity supply, poor connectivity and low bandwidth, making telemedicine unreliable. 11 Another factor to low use of telemedicine is inequality (Petersen, Brown, Pather, & William, 2019) resulting from the South African history that still mirrors the South African population. Inequality had developed a socio-economic divide (Delobelle, 2013) leading to a ripple effect on survival needs, such as quality healthcare and lack of basic computer literacy. Petersen et al (2019) research stated that economic factors (such as the digital divide between the rich and poor), educational factors (such as lack of technology literacy) and sociocultural factors (such as lack of support from family and peers) are contributors to low use of telemedicine. Moreover, guidelines developed by the Health Professional Council of South Africa (HPCSA) namely seeking written informed consent, quality of care, data protection and prior doctor-patient relationship, is a barrier towards the implementation of telemedicine, especially in an asynchronous or telephonic format (Townsend, Scott and Mars, 2019). 12 Figure 1: Root causes to deteriorating healthcare system in South Africa 13 From the onset of Covid-19, this research aimed to evaluate a combination of factors that limit the ease of accessing and use of digital health namely: telemedicine consultation as an engaging platform, seeking written informed consent for clinical procedures and using ICT, and availability and knowledge of resources to conduct virtual consultation. In addition, the method used to store patient information was introduced into the research to examine availability and sharing of patient information in the context of telemedicine. Concepts from the Unified Theory of Acceptance and Use of Technology (UTAUT) model was used as a guide in developing the research questions to examine the factors that limit the use of telemedicine. The following section expands on the research problem statement, research purpose statement and research propositions. 1.1.1 Defining telemedicine According to the American Telemedicine Association, telemedicine is a form of information and communication technology in which medical information between a doctor and patient is exchanged remotely (Chifamba, 2018). Telemedicine is divided into two categories namely asynchronous and synchronous. Asynchronous, also called store and forward telemedicine, is when information is shared between two parties, independent of time and at their own convenience (Chifamba, 2018). For example, images are stored and emailed to a referred medical professional. Synchronous or real-time telemedicine is when communication between a doctor and patient is live and interactive. This form of communication is usually undertaken on a videoconference call (Chifamba, 2018). 1.2 Research conceptualisation 1.2.1 The research problem statement Jack & Mars (2013) point that in order for telemedicine to gain prominence, it needs to be integrated into everyday practice and be seen as an alternative to face-to-face consultation. Hence, the importance to look into the factors that contribute to the low use of telemedicine. Whilst, there has been research done to examine the factors affecting healthcare workers’ acceptance and use of telemedicine, there has been minimal academic research on measuring a combination of the key contributors of accessibility and usefulness of telemedicine during Covid- 14 19. Hence, Section 1.2 will cover the following factors and its weaknesses that limit the accessibility and use of telemedicine (a) written informed consent, (b) if telemedicine provides an engaging platform, and (c) availability of resources and knowledge to implement telemedicine. The fourth factor is the method used to store patient information and how this influences the use of telemedicine. 1.2.1.1 Written informed consent In the event of a telemedicine consultation, the HPCSA General Ethical Guidelines For Good Practice in Telemedicine (Booklet 10, page 12) stipulates that healthcare practitioners are required to seek consent in writing from patients to conduct clinical procedures virtually, such as diagnosis and treatment (Human Rights, 2014). This becomes a challenge for patients because of required paperwork and e-signatures; and a challenge for doctors in the need of asynchronous or telephone consultations. Thus, the need to examine from family physicians if written or verbal informed consent is routinely obtained from patients during clinical examination or when using ICT, before and during Covid-19. 1.2.1.2 Engaging platform As per the HPCSA General Ethical Guidelines For Good Practice in Telemedicine (Booklet 10, page 18), the level of engagement during a telemedicine consultation, between a medical doctor and patient, is highly important to ensure that the patient is provided with sufficient information and fully understands the diagnosis and treatment provided (Human Rights, 2014). In their research, Jack and Mars (2013), indicated that there is sufficient evidence that a patient would obtain the same level of quality in a virtual consultation (be it synchronous or asynchronous) and a face-to-face consultation. However, there is not sufficient theory to justify the doctor-patient relationship as technology and telemedicine practice evolve. 1.2.1.3 Availability of resources and knowledge to implement telemedicine Restricted access to resources such as electronic devices and good internet connection as well as lack of knowledge needed to operate the telemedicine system is another mechanism that makes the implementation of telemedicine unsuccessful. This is a resultant from the South African history, which still mirrors some parts of the population, in that inequality led to a socio-economic divide (Cilliers & Flowerday, 2011). Hence, the need to investigate if electronic resources and digital literacy is still an implication to low use of telemedicine. 15 1.2.1.4 Patient information The South African Public healthcare system still follows the traditional method of record keeping that has negative consequences such as patients’ folders being misplaced and incorrect patient diagnosis. Hence, the National Digital Health Strategy (2019 – 2024) has prioritised to establish an integrated platform and architecture for the health sector information system, which will ensure interoperability and linkage of existing patient-based information systems. This will have a positive implication on telemedicine because instant accessibility of patient information will improve the quality and continuity of care. Additionally, the transmission of documents between different venues poses a health risk because viruses can easily be transmitted, a learning from the Covid-19 pandemic. In consequence, the examination of the method used by family physicians to store patient information, its success rate on National Digital Health Strategy and influence it will have on telemedicine. 1.2.2 The research purpose (aim and objectives) statement Following the weaknesses to gain maximum value of telemedicine as discussed in section 1.2.1, the aim of this research was to study a combination of key factors that limit the ease of accessing and using telemedicine by medical doctors specialising as Family Physicians in Johannesburg, in the context of Covid-19. The four factors examined in this research study included: a) If telemedicine results in an engaging platform in a doctor-patient relationship, especially as technology and telemedicine practice evolve; b) If written or verbal informed consent is routinely obtained from patients during clinical examination or when using ICT, before and during Covid-19; c) If electronic resources and digital literacy is still an implication of low use of telemedicine; d) Over and above and as a contribution to new literature, the research study examined a factor that influences the ease of using telemedicine, the method used to store patient information. This factor links to the National Digital Health Strategy for South Africa’s prioritization to establish an integrated information architecture for interoperability and safe sharing of health information across health systems and services. Electronic record keeping, especially for telemedicine consultations, will be beneficial to both medical practitioner and patient because patient information is easier to access and eliminates health risks. 16 Overall, the results obtained in this research will assist stakeholders in the South African healthcare sector in developing frameworks that will make telemedicine more user friendly and strengthen the current South African health ecosystem. 1.2.3 Research questions As per the factors and weaknesses identified in section 1.2.1 that influences a low intake of telemedicine and research objectives stipulated in section 1.2.2, the following section represents the research questions established to research the objectives of this study: 1.2.3.1 Question 1: Does consultation using a telemedicine format provide the same level of engagement and satisfaction as face-to-face consultation? 1.2.3.2 Questions 2: Which method of informed consent is used, written or verbal, for the following activities – namely taking history, examinations, ordering a special investigation such as an X-ray or blood test, referring the patient to a specialist, email patient information, prescriptions and virtual consultation? 1.2.3.3 Question 3: How crucial is access to the right resources, such as electronic devices and skills to navigate the technology, to achieving the desired outcome of telemedicine? 1.2.3.4 Questions 4: Which method, manual or electronic, is used to store patient information? If patient information is stored electronically, is patient information protected through encryption? 1.3 Significance of the research study Telemedicine is an innovative digital solution that has the potential to provide better healthcare services including access to healthcare services in remote areas, thereby minimizing the effect of shortages of healthcare professionals and accessibility. Telemedicine enables a patient to engage with their healthcare providers more frequently, minimising the risk to health deterioration. It may also provide an opportunity to reduce healthcare spending and increase productivity levels of medical practitioners. Determination of the underlying factors and findings from the study will assist in developing a suitable framework to enable telemedicine to be accessible to the broader population and offer a sufficient number of healthcare workers to treat increasing number of patients. Additionally, 17 examining the underlying factors will help address what the resource constraints are and develop a model whereby allocation to required infrastructure and equipment is made easily available and is reliable. The frameworks lead to indirect benefits whereby the need to travel long distances to access adequate healthcare will be eliminated. Hence, in order for telemedicine to gain prominence and integrated into everyday practice, there are key underlying factors that need to be evaluated. South Africans, especially those that depend on state facilities will continue to be underserved (Delobelle, 2013). Moreover, the number of health care workers needed to meet the increasing number of patients will continue to decelerate as more and more health workers will prefer to work for the private sector, patients will have to travel long distances to access adequate healthcare, and there will be restricted resources to treat patients (Delobelle, 2013). 1.4 Delimitations and assumptions of the research study The research focused mainly on the factors that influence the ease of accessing and using telemedicine in the primary healthcare sector. Hence, the following section discusses the delimitations and assumptions of the research study. For the study, family physicians were approached because their specialist is in primary care. The study did not include medical doctors who specialised in other medical specialities because this would impact the objective of the study. Similar to previous research studies, the sample encompassed only medical doctors and not patients, as research factors (namely, seeking verbal verses written consent, regulations imposed by the HPCSA, storage of patient information, resource availability and knowledge to conduct telemedicine consultation and whether telemedicine is an engaging platform for doctors) was appropriate for this study population. Furthermore, the research used a quantitative method of research in order to measure the objectives of this study numerically. The intended study population used was from Johannesburg that represents the population. This region was used for the researcher’s proximity and familiarity of location. 18 1.5 Limitations The following section identifies the limitations experienced during data collection for this research study. Due to the second wave of the Covid-19 pandemic, between December 2020 and January 2021, the acquisition of the targeted number (120) was a challenge to achieve because doctors especially those that focus on primary healthcare under immense pressure. Furthermore, respondents based their answers in the context of Covid-19 whilst the HPCSA was forced to ease guidelines on the use of telemedicine. Also, the aim of the present study was to collect data from a sample based in Johannesburg, however, due to the above-mentioned challenge, the sample was stretched to family physicians practicing in other metropolitan areas in of South Africa. When the study population was approached to participate in the research, it was assumed that they were potential users of telemedicine and could be considered a volunteer sample. 1.6 Preface to the research report To this end, the report has six chapters. Following Chapter 1 on introduction, Chapter 2 provides a literature review covering the underlying issues of telemedicine using past studies, the four attributes the study focused on and the UTAUT framework. Chapter 3 discusses the research strategy, design, procedures used in this study as well as the, validity measures and research limitations. A quantitative, cross-sectional research strategy was used to collect data from 80 medical doctors who specialize as a family practice physician in Johannesburg. Chapter 4 provides a presentation of the findings and Chapter 5 is an interpretation of the findings. Chapter 6 concludes the study with limitations of the present research study and recommendations for future research. 19 2 LITERATURE REVIEW In the following chapter, four themes namely research problem analysis, research knowledge gap analysis, UTAUT model and key attributes that affect telemedicine implementation will be discussed. Chapter 2 forms the foundation used to develop a framework for interpreting the research findings. 2.1 Research problem analysis The root cause of the deteriorating health care system in South Africa is primarily due to its history that can be traced back to the apartheid period between 1948 and 1993 (Maphumulo, 2019). During that period, there was a build-up of economic and social inequality as a result of racial segregation, resulting in a highly fragmented and poorly organized healthcare system (Maphumulo, 2019). Post 1994, even though all citizens became eligible to access quality healthcare services, including private healthcare, a major part of the population is still restricted to access to basic healthcare needs and rely on the state hospitals for treatment. To date, state hospitals rely on government funding, such as tax revenues, to fund for required skills and equipment to treat patients (Delobelle, 2013). Every year, government would allocate a certain percentage of its budget to state healthcare facilities in the nine provinces. (Delobelle, 2013). Even though the government contributes about 40% of all expenditure on health, the public health sector is under pressure to deliver services to about 80% of the population (Mvelase, Dlamini, & Dludla, 2015). On the other hand, the private healthcare sector offered world-class facilities for the smaller population who either had medical insurance or could afford care on an out-of-pocket basis. (Delobelle, 2013). Here, the facilities were independent and did not rely on government for funding (Delobelle, 2013). Post-apartheid, socio-economic inequality still remains a concern in South Africa and the requirement to build a healthcare system for all is far from complete (Delobelle, 2013). Currently, there are several factors causing the unequitable provision of healthcare in the public sector (Delobelle, 2013). These include the public sector being under-resourced and overused and the highly uneven skills distribution due to the migration of healthcare workers from public sector to 20 private sector, loss of staff due to illness, absenteeism, low staff morale, and an increased patient load (Delobelle, 2013). Delobelle (2013) stated in their study that the primary factors, of under-resourced and overused, faced by the public healthcare sector results in additional challenges. These include longer than average waiting period which may increase the health risk of patients, accommodate a certain number of patients for treatment while others in need of medical treatment are either put on a waiting list or turned away and poor hygiene (Delobelle, 2013). Being under-resourced, it further cannot cope with the increased burden of non-communicable & communicable diseases and injuries caused by violence, road traffic and other accidents (Delobelle, 2013). 2.1.1 Introducing telemedicine A potential solution to improve the public healthcare system that is overwhelmed is through telemedicine (Chifamba, 2018). In the HPCSA General Ethical Guidelines for Good Practice in Telemedicine (Booklet 10, page 6), Telemedicine is a digital platform used to provide virtual medical consultation between a healthcare practitioner in one location and patient in a difference location through various modes namely video conference, email, and instant messaging, and SMS (Human Rights, 2014). The HPCSA General Ethical Guidelines for Good Practice in Telemedicine (Booklet 10, page 6) further notes that the practice of telemedicine should involve secure videoconferencing or similar forms of technology to “replicate the interaction of traditional face-to-face consultations between healthcare practitioners and the patient. Telemedicine as defined refers to where information is exchanged electronically either on or off-line, formally, informally or as a need for a second opinion (Human Rights, 2014).” There are two primary methods that can be used namely synchronous and asynchronous, provided the patient has access to an electronic device (Chifamba, 2018). Both methods can be used in numerous settings such as doctor rooms, clinics and hospitals (Chifamba, 2018). Synchronous is when the consultation between the patient and doctor happens in real time while asynchronous is when the exchange of information happens at no given time (Chifamba, 2018). 21 Consultation through telemedicine involves the participation of various stakeholders in order for it to be effective and successful. These stakeholders include healthcare workers, information and communication technologists, economists, managers and policy makers (van Dyk, 2014). 2.1.2 Challenges to implement telemedicine in South Africa Majority of the South African population depend on public healthcare facilities which is currently understaffed, with inadequate resources, especially in rural areas (Delobelle, 2013). Studies have indicated that digital health mediums such as telemedicine can act as a tool to provide improved access to health care to those in need (Adeyelure & Kalema, 2019). However, the implementation of telemedicine has its own unique challenges (Cilliers & Flowerday, 2013). Telemedicine is a listed health solution strategy in the e-health plan (National Digital Health Strategy for South Africa 2019 - 2024, 2019) executed by the Department of Health of South Africa, and is seen as a cost effective and efficient means of delivering health care needs to communities that are under-resourced. The e-health plan constitutes other sectors such as health informatics (Cilliers & Flowerday, 2013). Prior to Covid-19 pandemic, the use of telemedicine in South Africa was low due to regulations imposed by the HPCSA. Regulators consider this innovative digital health solution as unproven and risky, thus the need for set regulation in order to protect both participating parties namely patients and doctors and to establish mutual trust (Jack & Mars, 2013). A regulation posed by the HPCSA General Ethical Guidelines for Good Practice in Telemedicine (Booklet 10, Page 5) is that “all Telemedicine services should involve a healthcare provider where there is an actual face-to-face consultation and physical examination of the patient in a clinical setting. The consulting practitioner will communicate the information to the servicing practitioner, who will then provide the necessary assistance” (Human Rights, 2014). Jack and Mars (2013) reported that this would be an impediment to telemedicine because it is unlikely that patients from underserved areas or rural areas will have had a prior doctor-patient relationship with the distant doctor being consulted. Section 2.1.2.1 to 2.1.2.4 will further discuss the key contributors of accessibility and usefulness of telemedicine. 22 2.1.2.1 Written informed consent An imposing factor to Telemedicine is the requirement of a written informed consent. The HPCSA General Ethical Guidelines for Good Practice in Telemedicine (Booklet 10 page 12) states that “informed consent for the use of telemedicine technologies must be obtained in writing” (Human Rights, 2014). However, this regulation was relaxed during the pandemic to “formal (preferably written) consent for among other things, specific services, including diagnosis and prescriptions and Information Communication Technology equipment to be used must always be secured from the patient” (Guidelines on telemedicine in South Africa, 2020). The HPCSA General Ethical Guidelines for Good Practice in Telemedicine (Booklet 10, page 16) further states that in an emergency situation “The practitioner must provide the patient with emergency instructions when the care provided by telemedicine indicates that a referral to an acute care or emergency facility is necessary for the immediate treatment of the patient. Furthermore, the emergency instructions should be in writing and appropriate to the services being rendered via telemedicine” (Human Rights, 2014). Notably, the requirement of written consent and especially obtaining the signature of patient, the patient’s parent, the patient’s guardian or the patient’s caregiver and signature of the witness will make the implementation of telemedicine more difficult because of inaccessibility and cost to meet this requirement. 2.1.2.2 Engaging platform The HPCSA General Ethical Guidelines for Good Practice in Telemedicine (Booklet 10, page 17) states that during a patient & practitioner consultation, “the quality and quantity of patient information received should be sufficient and relevant for the patient’s clinical condition in order to ensure that accurate medical decisions and recommendations are made for the benefit of the patient” (Human Rights, 2014). Tates et al (2017) identified that the three main pillars of communication namely information exchange, interpersonal relationship building, and shared decision making is more successful and of better quality during offline consultations. 23 Thus, the quality of virtual consultation verses face-to-face consultation poses a challenge to the implementation of telemedicine (Nordal, D, Kvammen, & Løchen, 2001). However, Cilliers and Flowerday (2013) research study presented a different view for the reason that the quality of a physical consultation is similar to a virtual consultation. Although Jack and Mars (2013) found that there are underlying factors that restrain the efficiency of telemedicine, which include poor internet connection that contribute to the low use of telemedicine. Thus, this study considered that further evaluation of effectiveness of telemedicine is needed. 2.1.2.3 Availability of resources and knowledge to implement telemedicine Petersen el al (2019) addressed that lack of awareness about the kinds of technology available, skills needed to apply the technology and its significance are other barriers to telemedicine remaining unsuccessful. Cilliers and Flowerday (2013) indicated in their study that only 54.4% of respondents thought they did have the resources necessary to make use of telemedicine. Prinsloo (2017), research depicted that fifty-six percent of study population agreed that they did have the knowledge needed to operate the system. Similarly, Cilliers and Flowerday (2011) study on whether computer literacy affect telemedicine acceptance among healthcare workers, pointed that less than fifty percent of the participants considered themselves confident about their knowledge of telemedicine. However, is there an opportunity to continue with telemedicine in South Africa and will it be the go-to platform in future. Will digital health be able to fix the challenges the healthcare system faces by narrowing the inequality gap through improving access to quality, accessibility and efficiency of healthcare in South Africa. Public healthcare facilities will become far less under pressured and patients will benefit with accessing a wide range of specialists (Jack & Mars, 2013). There is hope that telemedicine will grant access to healthcare that is more affordable (Chifamba, 2018). Most importantly, patients don’t need to travel long distances for treatment (Jack & Mars, 2013). 2.1.2.4 Patient information Maphumulo and Bhengu (2019) points out that poor record keeping can lead to other problems such as unnecessary delays for patients. The research further points that using the traditional method of storing patient information, patients’ folders can be misplaced. Nevertheless, healthcare 24 workers are then forced to look for these files, causing patients to wait extra longer. In worst scenarios, the medical history of the patient is lost, which can create further complications leading to incorrect diagnosis and in some cases death of the patient (Maphumulo & Bhengu, 2019). In 2019, the Department of Health of South Africa published a white paper, titled National Digital Health Strategy for South Africa2019 – 2024 (page 11), which aims to develop a platform to store patient information electronically that will become easily accessible to stakeholders (such as medical doctors) in the healthcare industry and in preparation for the National Healthcare Insurance (NHI) (National Digital Health Strategy for South Africa 2019 - 2024, 2019). Yet, the HPCSA is concerned about security of patient information and states that “healthcare practitioners using telemedicine should provide safe procedures to avoid any alteration or elimination of patient data; and patient information should only be transmitted from one site to the other and stored, with the full knowledge and approval of the patient, in line with the informed consent guidelines” (Human Rights, 2014). 2.2 Research knowledge gap analysis Section 2.2 discusses the following attributes – namely telemedicine as an engaging platform, seeking written informed consent for clinical procedures and using ICT, availability and knowledge of resources to conduct virtual consultation, and method used to store patient information – and why it contributes towards the accessibility, usefulness and low uptake of telemedicine and an individual’s readiness to opt for telemedicine, taking into consideration the following studies by Prinsloo (2017) and Adeyelure and Kalema (2019). Adeyelure and Kalema (2019) examined factors that influenced a user’s readiness to accept telemedicine in the context of the South African public healthcare sector. Using a quantitative research strategy, Adeyelure and Kalema (2019) examined environmental and socio-economic factors that influence the use of telemedicine namely organization and environment, technology compatibility, perceived usefulness and perceived ease of use. The findings reveal that organization, environment, technology, perceived usefulness and perceived ease of use are factors that influence the implementation of telemedicine in South Africa. The study also revealed that external factors such as government regulations and tariffs will have an influence on user readiness to accept telemedicine (Adeyelure & Kalema, 2019). 25 Prinsloo (2017) also did a study on the factors affecting healthcare workers’ acceptance and use of telemedicine in hospitals in Kwa-Zulu Natal. Here the factors explored were technology acceptance; the influence of socio-demographic factors (age, experience, profession, qualification) and acceptance factors on use and behavioural intention to use telemedicine. There were high levels of agreement amongst respondents regarding the value of telemedicine (performance expectancy, and ease of use (effort expectancy), but less so for the role of social influence on use and the presence of facilitating conditions (such as support and assistance). Examples of factors are trust, demographic features (age and race), personal factors (personal experience), and workplace factors (computer skills, perceived enjoyment, management support). 2.2.1 Performance expectancy: Engaging platform Satisfaction is defined when an individual’s expectations are met and can be achieved through quality, trust and empathy. In telemedicine, satisfaction occurs when there is effective communication between the medical professional and patient and accessibility to the resources needed to implement a virtual consultation and self-care efficacy. This factor links to UTAUT’s performance expectancy because the telemedicine user is confident that the system will enable them to meet their expectations. There have been numerous debates (Nordal, D, Kvammen, & Løchen, 2001) around the satisfaction received when using an online platform. Some of the elements include discomfort, embarrassment, nervousness and difficulty in hearing the doctor. Nevertheless, previous studies have reported that there may be a quality difference between a face-to-face consultation and telemedicine (Nordal, D, Kvammen, & Løchen, 2001). Nordal, et al (2001) evaluated teledermatology in a comparative study of video conferences versus face-to-face consultations and found that there was no significant difference on patients’ satisfaction between teleconsultation and of face-to-face consultation. The only exception was for patients’ feelings of contact with the dermatologist. While Mirzaei and Kashian (2020) did a study to determine if there were any differences in patients' perceptions of communication effectiveness with their physicians through different modes of communication. Here, it was noted that there was no significant difference in satisfaction 26 perceived, information exchange or interpersonal relationship building between face-to-face or digital consultation. Additionally, Tates K, Antheunis ML and et al (2017) study on the “Effect of Screen-to-Screen Versus Face-to-Face Consultation on Doctor-Patient Communication” revealed that there was no significant difference between face-to-face and virtual consultation in terms of information exchange, interpersonal relationship building, and shared decision making. The research design used in their study was experimental, quantitative and the setting was in Netherlands (Tates, Antheunis, Kanters, Nieboer, & Gerritse, 2017). The hypothesis was tested using Preacher and Hayes' procedure (Tates, Antheunis, Kanters, Nieboer, & Gerritse, 2017) and a regression method was used to determine the association between medium of consultation and perceived information exchange, perceived interpersonal relationship building, and perceived shared decision making (Tates, Antheunis, Kanters, Nieboer, & Gerritse, 2017). Taking into consideration the previous studies, this study examined the following factors that contribute to performance expectancy namely satisfaction and confidence, especially in the context of Covid-19. This study also measured the level of communication and information exchange where participants were asked if explanation concerning the disease and treatment was better in telemedicine. 2.2.2 Social influence: Written informed consent An informed consent is providing permission or access to personal information in order to make an informed decision (Townsend, Scott, & Mars, 2019) through methods such as verbal or written. Townsend, Scott, & Mars (2019) explains that an informed consent encompasses detailed information about the treatment process and potential risks and benefits provided by the doctor to the patient in a language that they understand. The study also explains the importance of an explanation of how the virtual consultation will occur such as the role and responsibility of the provider and the patient during the telemedicine interaction. Prior to Covid-19 pandemic, the HPCSA’s regulation on telemedicine was that written informed consent for all aspects of telemedicine is important because it implies that when an individual seeks help from a medical professional knowing that there might be some degree of risk/harm involved, they are not able to bring a claim against the consulting party (Jack & M, 2013). 27 The regulation links to the UTAUT social influence construct because rules and regulations can inhibit one from pursuing telemedicine. In South Africa, there is debate (Jack & Mars, 2013) on whether informed consent for all forms of telemedicine is needed and whether it has to be in writing. According to Jack and Mars (2013) the riskier the medical intervention, the greater the requirement for written informed consent in a traditional consultation. While the World Medical Association (WMA’s) guidelines on the Use of Telemedicine for the Provision of Health Care indicates that that the requirement for written consent is not compulsory, but medical professionals are required to follow relevant protocols for verbal, written or recorded consent and, where appropriate (Jack & Mars, 2013). Using a quantitative, descriptive, cross-sectional survey, Jack and Mars (2013) measured how often doctors (general practitioners and all specialists) and nurses sought informed consent to take patient history; examine patient; order special investigation; refer patient; email patient information, and whether the consent was written or verbal. The overall findings saw that written consent is not regularly obtained from patients during a clinical examination or when using a digital platform. Taking into consideration the Covid-19 pandemic resulting in surge of telemedicine consultations and relaxation of HPCSA regulation on informed consent, this research measured how often family physicians sought written and verbal informed consent for the following procedures: patient history, examine a patient, email patient information, refer patient, prescription, and virtual consultation. 2.2.3 Facilitating condition: Availability & knowledge of resources to conduct telemedicine Venkatesh et al (2003) stated that facilitating conditions play an important role in the acceptance of telemedicine and included the following elements: technical assistance, knowledge of the system and compatibility with other systems already in use. Cilliers and Flowerday (2013) study on the factors that influence the user acceptance of telemedicine found that there was no significant difference for qualification and knowledge of telemedicine. Findings in their research study reported that forty percentage of the study 28 population found that telemedicine is not suitable for the systems in place at their workplace; and fifty percent of the study population assumed that they had the required resources in place to make use of telemedicine (Cilliers & Flowerday, Health information systems to improve health care: A telemedicine case study, 2013). Research by Cilliers and Flowerday (2013) did not make specification on the availability of resources. Whilst in this research, availability of resources namely electronic devices, ease of use and learning to operate the telemedicine system was measured to stipulate their preference of telemedicine. 2.2.4 Effort expectancy: Storing patient information Jack and Mars (2013) indicated that even though electronic record keeping can have a security concern, information on 'who, what, when, and why' is easier to access within a national eHealth ecosystem, less time consuming and treatment can be provided soonest. Storing patient health information manually will make the sharing of this data challenging between medical professionals and between hospitals (Coleman, 2013). In the context of telemedicine, availability of patient information will improve the quality and continuity of care of patients. The 4IR has improved ways to store and safeguard information. Personal data navigates within the health ecosystem leading to extensive sharing and transference and a possible threat to the misuse of such data (Townsend, R E Scott, & Mars, The development of ethical guidelines for telemedicine in South Africa). While, Jack and Mars (2013) pointed that emails and other patient information that is confidential can be encrypted and password protected. In the South African context, record keeping/data protection requires more consideration as the current frameworks is insufficient to determine the reliance of data protection electronically (Townsend, Scott, & Mars, 2019). Townsend, Scott, and Mars (2019) suggest that the guidelines by the HPCSA called ‘Confidentiality: Protecting and providing information’, and ‘General ethical guidelines for good practice in telemedicine’ need to be more aligned with the provisions stipulated in South African legislative law, particularly with regard to data collection, security, storage and transfer, within the national eHealth ecosystem. Since limited research is available on how patient information is stored, this study, measured the procedure used to store patient information and its link to National Digital Health Strategy for 29 South Africa. If patient information was stored electronically, how patient information is protected was also studied to identify its alignment with the HPCSA regulation. In conclusion and from the studies discussed in this section, this research study examined a combination of previously researched and new components in the following categories that maximises the value of telemedicine, especially in the context of the Covid-19 pandemic: quality of information exchange; access and knowledge to the required resources for the consultation to be effective; regulations imposed by HPCSA such as the requirements of written informed consent and method used to store patient information. 2.3 Unified Theory of Acceptance and Use of Technology (UTAUT) Adeyelure and Kalem (2019) stated that ICT can become a vital solution to alleviate some of the challenges faced in public health sectors in South Africa. However, the benefits of this technology can only be seen if healthcare workers have access to the necessary resources, are knowledgeable about navigating required platforms and if they find it useful (Cilliers & Flowerday, 2013). There are various psychological factors, such as understanding and trust, that make the acceptance of technology challenging. Venkatesh et al (2003), deployed eight theories/models of technology to develop the Unified Theory of Acceptance and Use of Technology (UTAUT). The theories used to study the behavioural intention to use a technology included: Theory of Reasoned Action (TRA), the Technology Acceptance Model (TAM), the Motivational Model (MM), the Theory of Planned Behaviour (TPB), a combined Theory of Planned Behaviour/Technology Acceptance Model (C-TPBTAM), the Model of PC Utilization (MPCU), Innovation Diffusion Theory (IDT), and Social Cognitive Theory (SCT) (Venkatesh, Thong, & Xin, 2012). The model consists of four constructs namely performance expectancy, effort expectancy, social influence, and facilitating conditions as well as key moderators including gender, age, voluntariness and experience (Figure 2). This study used the definitions of the four constructs to develop the questions in section 1.2.3. 30 Figure 2: UTAUT model (Prinsloo, 2017) Performance expectancy is defined as the extent to which an employee enjoys certain benefits when using technology for an activity (Venkatesh, Thong, & Xin, 2012). Cohen, Bancilhon, and Jones (2013) explain that performance expectancy is achieved when an individual perceives that a technology works to their advantage, as well as improve productivity and effectiveness. Effort expectancy is the degree of ease associated with consumers’ use of technology (Venkatesh, Thong, & Xin, 2012). In other words, it is about how user friendly a certain technology is and whether it needs a lot of effort to be operational (Cohen, Bancilhon, & Jones, 2013). Petersen et al (2019) observed that if using a technology device is slow and cumbersome, this will likely influence an individual to use it less frequently. Additional studies denote that effort expectancy to be the most important factor influencing health professional satisfaction with an electronic health system (Cohen, Bancilhon, & Jones, 2013). Social influence is defined by the individuals’ decision to use digital health based on the perception of peers on the usefulness of this technology (Venkatesh, Thong, & Xin, 2012). 31 Facilitating conditions encompass the environmental factors, resources and support available to make it easier to perform a task (Venkatesh, Thong, & Xin, 2012). These include financial and technical resources, support and training. Based on the UTAUT model, Venkatesh, Thong, & Xin (2012) research identified that factors that influence behavioural intention to use technology are performance expectancy, effort expectancy, and social influence. Facilitating condition is a factor that influences the determination to use a technology (Venkatesh, Thong, & Xin, 2012). Since the inception of the model, UTAUT serves as a baseline model in technology adoption and has been tested widely including the health sector, both organisational and non-organisational settings (Venkatesh, Thong, & Xin, 2012). The theory was found to explain 70% of the variance in behavioural intention to use a technology and about 50% of the variance in technology use (Cilliers & Flowerday, 2013). The benefit of this model is that it provides for managerial decisions and to determine the effectiveness and value of a technology product is to be implemented. Hennington and Janz (2007) were the first researchers to study the UTAUT model in the acceptance of technology in the health sector. While (Chang, Hwang, Hung, & Li, 2007) recently suggested that UTAUT is a useful framework to use in research related to electronic health adoption; other studies have shown that the UTAUT model can be modified to gain maximum usefulness across different countries. Elsewhere, the UTAUT framework was used to examine physicians’ acceptance of a pharmacokinetics-based clinical decision support system in Taiwan and acceptance of telemedicine in South Africa’s public healthcare system (Cohen, Bancilhon, & Jones, 2013). Some of the limitations of UTAUT Petersen et al (2019) include community, culture, country, organisation, agency, department, person, use of students to explore workplace issues, no use of moderating variables, and lack of exogenous factors. It does not include external factors that can affect the performance of the behaviour. It also does not take into consideration factors such as contexts and situations that may impact usage (Petersen et al, 2019). 32 2.4 Summary and conclusion 2.4.1 Summary of literature reviewed This chapter explored telemedicine and factors that reduce the use of this technology. Telemedicine is a form of information and communication technology used to provide healthcare service virtually and is likely to fix the challenges the healthcare system faces by narrowing the inequality gap through improving access to quality, accessibility and efficiency of healthcare. However, in South Africa economic factors (the digital divide between the rich and poor), educational factors (such as lack of technology literacy) and sociocultural factors (such as lack of support from family and peers) hinder the accessibility and usefulness of telemedicine (Petersen et al, 2019). Hence, the purpose of this study was to use a quantitative, cross-sectional research strategy to evaluate the following factors: written informed consent (social influence), record keeping (effort expectancy), if telemedicine is engaging (performance expectancy) and availability & quality of resources to conduct telemedicine (facilitating condition). 33 3 RESEARCH STRATEGY, DESIGN, PROCEDURE AND METHODS This chapter looks into the research strategy, the research design as well as the procedure and methods undertaken in this study based on the four questions (section 1.2.3) that this research report examined - that is, is telemedicine an engaging platform; method used to seek informed consent to perform a clinical examination or when using ICT; availability and knowledge of resources to implement telemedicine successfully; and the method is used to store patient information. 3.1 Research strategy A research strategy (Research methodology: Research Starters Topic, 2018) is the method undertaken in order to pursue information to meet the objectives of this study. The strategy builds a foundation by developing a consensus on what problems need to be investigated, how will the investigation take place and how will the findings help resolve a current societal issue (Research methodology: Research Starters Topic, 2018). It constitutes of three elements namely qualitative, quantitative, and mixed (Goertzen, 2017). In order to measure the objectives of this study, a quantitative research strategy was implemented. The purpose of using this research strategy, was to study human behaviour by collecting numerical data from the sample representing a population. It involved obtaining numerical data across a certain time (i.e. Covid-19 pandemic) and profession (for the purpose of this study) through statistical inference procedures. Also, questions used in the survey were direct and quantifiable – such as percentage, what proportion, what extent and how much – so that the data could be analysed using statistical models and to gain an understanding about needs of the targeted sample. (Goertzen, 2017). In order to test for reliability, a sample size of 50 or more was needed. Cilliers and Flowerday (2013) used a positivistic, quantitative research methodology to investigate if user acceptance is a factor for the poor uptake of telemedicine in the Eastern Cape using the four major constructs of the UTAUT model as a baseline. The study comprised of hypothesis per construct and required statistical models to prove the hypothesis. Hence, for this study, the UTAUT constructs was used as a foundation to establish the factors that influence the ease of accessing and use of telemedicine leading to the research questions. 34 3.2 Research design A research design is a framework used to collect and analyse data (Goertzen, 2017). The motive of the research design is to find evidence for each research question and for a certain set of criteria namely validity, replication, trustworthiness and authenticity. There is more than one type of research design namely: cross-sectional; longitudinal; quasi-experimental; case study; comparative design. More specifically, a cross-sectional design involves the collection of data at one specific time with one specific sample. It can be used in both qualitative and quantitative research. It is like a status report or even a ‘snapshot’ of reality in a very focused and specified manner (Merriam & Tisdell, 2016). Thus, for this study, the cross-sectional design was leveraged to understand the prevalence of a behaviour in the usage of telemedicine by family physicians during Covid-19 pandemic (single point in time). This research design was highly useful in collecting data in connection with two or more variables. Additionally, the design was used to identify the association of the causes of low use of telemedicine and outcome of interest. Adeyelure and Kalema (2019) research study on examining factors that influence the use of telemedicine used a cross-sectional research design to study the association between dependent and independent variables quantitatively. The motive for the design was to also gather data from participants at one specific point in time (Tope Samuel Adeyelure & Kalema, 2019). 3.3 Research procedure and methods This section documents the actual procedure and methods employed in this research to collect, collate, process, and analyse empirical evidence. 3.3.1 Research data and information collection instrument(s) Research data collection instrument are tools used to collect data from a specified sample to ensure the seamless translation from raw data captured to interpretation (Merriam & Tisdell, 2016). Types of data collection methods include primary data where information is collected by the researcher from scratch and secondary data is using information collected by others such as government publications, websites, ethno-statistics and journal articles. (Merriam & Tisdell, 2016). 35 For this study, a primary data collection method was used so that the findings are most current, taking into consideration preface of Covid-19. According to Merriam and Tisdell (2016), in a primary data collection method, there are various tools used to collect information. Common tools (Merriam & Tisdell, 2016) include questionnaires/surveys (mailed, online, group/collective, individual, panel); interviews (structured, semi-structured, unstructured); observations (participant, non-participant); experiments (experimental, semi-/quasi-experimental). Since this research focused on a quantitative research strategy, the type of instrument structure used was an online questionnaire encompassing close ended questions. The aim of this instrument structure was to minimize differences in responses and to keep it as standard as possible so that true variation is achieved. This research study was associated with the structure of the data collection instrument implemented by Prinsloo (2017) due to its relevance and to generalise findings of the specified sample. Prinsloo (2017) research on factors affecting healthcare workers’ acceptance and use of telehealth in hospitals in KwaZulu-Natal (KZN), used a structured and close ended questionnaire so that the responses can be analysed using relevant statistical models such as t-tests and regression models. The research by Prinsloo (2017) further used a questionnaire that covered demographic variables (gender, age), behavioural intention, performance expectancy, effort expectancy, social influence, and facilitating conditions. Different formats of questions and responses were included in the questionnaire such as yes/no, and degree of agreement or disagreement in Likert scale questions which will be used as a guide for the current research study. Additionally, questions for this study were sourced from previous research studies that investigated the questions stated in Section 1.2.3 to understand the factors affecting low use of telemedicine. The questionnaire was introduced with questions on demographics such as geographic area, gender and age. This was followed by questions based on the four overarching research propositions in section 1.2.3. Each research question comprised sub-questions for the purposes of obtaining more data and statistical models relevant to this research study. While collecting data, each question had a multi-item scale allocated for statistical analysis purposes. Various measurement tools were used to assist with interpreting the data. This included 36 categorial data to interpret gender, geographic area of practice, work place description, telemedicine modalities, profession and how informed consent is taken; ordinal data to interpret age; Likert scale to interpret if telemedicine is engaging, factors that cause low use of telemedicine and how patient information is stored. 3.3.2 Research target population and selection of respondents 3.3.2.1 Research target population When conducting the research, it is important to decide which section of the population is suitable to represent the research proposition and add value to the study. For this study purposes, a population of South African medical doctors was used. Furthermore, the population included medical doctors practicing in state, private or both facilities; male or female; and all age groups. The categories were taken into consideration in order to determine if telemedicine usage is common in males or females, which age category finds it most useful and to ensure that the participants were based in the region and profession as per the objective of this study. Factors such as income level served least importance. The initial sampling strategy was to focus on the Johannesburg region, however, due to a low response rate, the survey was expanded to the rest of South Africa. 3.3.2.2 Sampling or selecting respondents from the target population Sample is usually drawn from a population and data gathered from the sample will lead to assumptions about the population (Bryman & Bell, 2014). A sample of medical doctors specialising as Family Physicians was used in this study. Family Physicians focus on primary healthcare by diagnosing and treating acute or chronic illnesses, hence its relevance to the study. According to Bryman and Bell (2014), there are a number of methods used to formulate a sample such as bias, sampling error, probability sample (simple random sample, systematic sample, stratified random sampling, multi-stage cluster sampling), non-probability sampling (convenience sampling, snowball sampling, quota sampling). 37 In a research conducted by Prinsloo (2017) the sampling method used was convenience sampling where the researcher conducted face to face briefing sessions with staff at seven hospitals in Kwa- Zulu Natal and obtained email addresses. Thereafter, the researcher approached participants individually via email. Each email contained an information sheet explaining the purpose of the study, an invitation to participate and informing the recipient that participation is voluntary and anonymous. Through this method, a large number of people took part in the research. For this study, a non-probability convenience sampling was used so that the accuracy of data obtained is high. Initially, face-to-face briefing sessions with family physician was going to be conducted. However, due to Covid-19 pandemic and restrictions imposed at healthcare facilities, online communication was instrumented. Hence, doctors’ (that specialised as family physicians) names and email address were sourced from the websites of hospital groups namely Netcare, Life Healthcare and Mediclinic. Thereafter, an email containing the research information, link to the online survey and ethics certificate was sent to each medical doctor, inviting them to participate in the research and informing them that participation is voluntary and that all information shared remained confidential by encrypting documents with a password. Contact details was also included in the email. Here, 50 Family Physicians based in Johannesburg was approached. Additionally, an organisation that collects and maintains the contact information of South African healthcare providers and complies with the Protection of Personal Information Act (POPIA), was approached to assist with accessing the contact information of Family Physicians, in both public and private practices. However, due to privacy concerns, the organisation preferred to share the invitation to 580 Family Physicians practicing in various parts of South Africa to participate in the research via email using their resources. Amidst the Covid-19 and medical doctors working longer than normal shifts, a challenge experienced during data collection was doctors’ unavailability to answer the survey shared, hence low response received. To try and alleviate the problem and improve the response rate, telephone conversations with medical doctors was conducted. 38 3.3.3 Ethical considerations when collecting research data The role of ethics is to ensure sustainability and transparency of a study. Ethics or morals is a standard of conduct. When conducting research, participants are likely to share their personal information and experiences and would prefer that the researcher keeps all information shared confidential and not be shared externally by the data collector (Merriam & Tisdell, 2016). A researcher should seek consent from respondents prior to collecting data. Consent is defined as giving permission for something to happen. For research purposes, consent is willingness to participate in the interview/survey/focus group and should include the following characteristics for building trust: honesty, openness, carefulness, accountability and fairness (Merriam & Tisdell, 2016). In this study, an email (appendix 2) containing the following information was sent to each Family Physician - the title and aim of the research undertaken; link to the online survey should the medical doctor be interested to participate with a claim that no personal costs is associated by participating; option to withdraw at any time; noting that information shared will remain confidential; all information that is obtained will be stored and protected; a summary of the research will be sent upon request; and contact details of the researcher, supervisor and University Human Research Ethics Committee. Prior to data collection, approval from School of Graduate School of Business Administration Ethics Committee was obtained. For authenticity, a signed ethics certificate (appendix 3) was shared with medical doctors who wished to participate. 3.3.4 Research data and information collection process When collecting data, various elements should be taken into consideration to gain as much information as possible. These include: what data sources are most valued; what data collection methods is highly effective; what data collection methods will provide valid and reliable results; and any parameters that need to be considered (Merriam & Tisdell, 2016). Family Physicians who agreed to participate had to simply click on a link supplied on the email which took them to the survey. The survey was built on a user-friendly online platform called Survey Monkey. Each survey took an average of 5 minutes to complete. 39 Data collected from participants was then automatically stored in an Excel spreadsheet which could easily be downloaded from Survey Monkey. In order to meet the objectives of the study and present a quality quantitative research, data captured on the Excel spreadsheet was then cleaned whereby respondents who did not specialise as family physicians was deleted and those who did not complete the survey was deleted. 3.3.5 Research data and information processing and analysis 3.3.5.1 Research data and information processing Research data processing is the steps used to filter information acquired to make it more meaningful and comprehensive. The benefits of data processing is that it becomes easy to handle large amounts of data and to reorder great quantities of information quickly. In quantitative research, data processing becomes much simpler and variability is reduced because interviewees are provided with answers to choose from (Merriam & Tisdell, 2016). For the current research, data was coded as per the code number on each completed questionnaire and imported to JMP software for analysis. Data was analysed by the researcher with the assistance of a statistician using both descriptive and analytic statistics. 3.3.5.2 Research data and information analysis Data analysis is defined as interpreting data collected and how it relates to the study. It is about making meaning about the data collected by consolidating, reducing, and interpreting what people have said and what the researcher has seen and read. This is done using statistical models such as descriptive statistics, regressions analysis, cluster analysis, content analysis, Pearson’s correlation coefficient, t-test, statistical significance, univariate analysis and bivariate analysis (Merriam & Tisdell, 2016). The data collected from the study population was analysed using descriptive and analytic statistics. Descriptive statistics was used to summarise data collected from the sample for questions on demographics of the respondents, engaging platform, informed consent, resources and storage of patient information. Measures of central tendency (mean, median or mode) was used to understand common patterns in the use of telemedicine. While, measures of variability, was used to analyse how spread out the distribution of the data was. 40 The types of analytic statistics used was: 1. A Mosaic Plot to summarize the relationship between ease of use and gender (male & female) among several categorical variables 2. A Oneway Analysis of engaging platform & age and engaging platform & gender was done to determine which age group (21-40; 41-60; 61+) and gender (male/female) agrees the most that telemedicine is an engaging platform 3. A Pearson Correlation of family physicians & written informed consent and family physicians & verbal informed consent to determine its association 3.4 Research strengthens—reliability measures applied Reliability Reliability is the degree to which findings are consistent when researchers repeat the process. An assessment cannot be valid unless it is reliable. Factors which stimulate reliability include stability, internal reliability and inter-observer consistency (Bryman, 2014). For this study, a Cronbach's alpha was used to measure internal reliability. The measurement was done using question 1 in appendix 1 on engaging platform as the Likert scale was consistent for all the sub-questions and the score was 0.83. The calculation of the alpha correlation coefficient varies between 0 (no correlation and therefore no internal consistency) and 1 (perfect correlation and therefore complete internal consistency). A result of 0.8 and above usually implies an acceptable level of internal reliability (Bryman, 2014). Validity Validity is an important research criterion that describes the integrity of research findings and consists of four main types namely, internal, external, measurement, and ecological (Bryman, 2014). This section discusses three categories of validity for the current research study. Internal validity refers to whether the current research findings are aligned to the theoretical ideas proposed (Bryman, 2014). Bryman (2014) argued that internal validity tends to strengthen a qualitative research design strategy because of the consistency between concepts and observations through an experimental design. In the current research study, a survey as utilised, making internal reliability difficult to achieve because the research design strategy is quantitative and the outcome variable cannot be controlled by the researcher. 41 External validity is if the research finding can be applied more broadly in external environments such as different location or population (Bryman, 2014). In the current research study, a non- probability convenience sampling was used and respondents were family physicians. Thus, the results may not represent other medical specialities. Ecological validity is described when the research findings can be applied to natural setting such as people’s everyday lives (Bryman, 2014). Hence, the findings in the current research study will help determine the socio-economic conditions influencing daily life conditions and contribute towards future research studies and models to improve the usage of telemedicine in various geographical locations. 3.5 Research weaknesses—technical and administrative limitations Limitations in a research is broad and certain environmental or social factors may hinder achieving the expected results. The targeted number of responses for this study was 120, but 80 responses was acquired. The acquisition of the targeted number was a challenge to achieve because between December 2020 and January 2021, South Africa experienced the second wave of the Covid-19 pandemic, putting doctors especially those that focus on primary healthcare under immense pressure. Also, the aim of the present study was to collect data from a sample based in Johannesburg, however, due to the above mentioned challenge, the sample was stretched to family physicians practicing in other metropolitan areas in of South Africa. A difficulty in the research was extracting a sample that had some sort of telemedicine experience. When the study population was approached to participate in the research, it was assumed that they were potential users of telemedicine and could be considered a volunteer sample. The initial strategy for the research was to use the entire UTAUT model, including the set of questions developed for this specific model. However, the questionnaire and research propositions developed for this research study does not strongly correlate with the requirements of the UTAUT model and statistical analysis required. In a UTAUT model, all questions based on the four constructs has a consistent Likert scale of strongly agree, agree, disagree and strongly disagree. Hence the usage of some parts of the model was omitted in the present study. 42 4 PRESENTATION OF RESEARCH RESULTS The purpose of this study was to determine if the following attributes namely engagement, written and verbal informed consent, method used to store patient information, and availability of resources - make access to telemedicine consultation less feasible. Five questions were drafted and researched to meet the objectives of the study. In this chapter, an analysis was done on the demographics of the sample approached and research questions portrayed as in section 1.2.3. In the study, a Cronbach's alpha was used to measure internal reliability and the score was 0.83. The measurement was done using question 1 on engaging platform as the Likert scale was consistent for all the sub-questions. The calculation of the alpha correlation coefficient varies between 0 (no correlation and therefore no internal consistency) and 1 (perfect correlation and therefore complete internal consistency). A result of 0.8 and above usually implies an acceptable level of internal reliability. 4.1 Socio-Demographic Characteristics of Participants The study population included mixed genders with 65% males and 35% females (Table 1). The age group of participants was ranked into four groups and ranged from 21 – 61+ with majority been between 51+. In terms of practice of work, the study population practiced as family physicians in either a private facility, state facility or both. Table 1: Socio-demographic representation of the study population N Percentage Gender Female 28 35% Male 51 65% Age 21 – 30 4 5% 31 – 40 10 13% 41 – 50 18 22% 51 – 60 30 38% 61+ 18 22% 43 Place of work Private 47 59% State 23 29% Both 8 10% Other 2 2% 4.2 Telemedicine usage Study population reported that 35% used telemedicine daily, 29% weekly, 15% monthly and 21% never used it (Table 2 and Figure 3). For those that made use of telemedicine, 5% used it for diagnosis, 12% used it for treatment, 71% used it for both while 12% reported for other purposes with reasons being following up, consultation and scripts, referral letters and feedback (Table 2) Table 2: Telemedicine usage of the study population N Percentage How often telemedicine is used Daily 28 35% Weekly 23 29% Monthly 12 15% Never 17 21% What purpose is telemedicine used Diagnosis 3 5% Treatment 8 12% Both 46 71% Other 8 12% 44 Figure 3 How often telemedicine is used The study population also reported on the types of telemedicine modalities used from the options that were provided – SMS, telephonic conversation, video call, instant messaging and other with reasons being email. Even though participants could choose from more than one option, telephonic conversation was rated the highest telemedicine modality used by family physicians, followed by instant messaging (Figure 4). Figure 4: Modalities used by study population Daily Weekly Monthly Never 0 10 20 30 40 50 60 70 SMS Telephonic conversation Video call Instant messaging Other Telemedicine modalities 45 More than 40% of the population study find telemedicine to be somewhat useful followed by 30% who find it to be extremely useful (Figure 5). Figure 5: Usefulness of telemedicine as per the study population Nearly half (47.5%) of the participants are somewhat knowledgeable about telemedicine while 37.5% indicated that they are knowledgeable, while only 7.5% very knowledgeable and 7.5% are not knowledgeable about telemedicine (Figure 6). Figure 6: Study population knowledge about telemedicine 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% Extremely useless Somewhat useless Slightly useless Slightly useful Somewhat useful Extremely useful Usefulness of telemedicine 0.00% 5.00% 10.00% 15.00% 20.00% 25.00% 30.00% 35.00% 40.00% 45.00% 50.00% Not knowledgeable Somewhat knowledgeable Knowledgeable Very knowledgeable Knowledge of telemedicine 46 4.3 Telemedicine as an engaging platform Respondents were assigned scores (strongly disagree= 1; strongly agree= 4) and mean scores for the individual indicators and mean total scores as a whole was computed (Table 3). The mean total score was 10.7 out of 16 with individual means ranging from 2.56 to 2.85. Table 3: Telemedicine as an engaging platform Min Max Mean Std Deviation 1. Do you agree that telemedicine enables a doctor to be more productive? 1 4 2.85 0.74 2. Do you agree that telemedicine results in increased confidence in a doctor/patient relationship? 1 4 2.61 0.77 3. Do you agree that as a doctor, satisfaction is usually gained post telemedicine consultation? 1 4 2.56 0.67 4. Are questions answered profusely during a telemedicine consultation? 1 4 2.68 0.65 47 Overall, more than 50% of respondents agreed that telemedicine is an engaging platform with 52.5% agreeing that telemedicine enables a doctor to be more productive (Q1), 55% agreeing that satisfaction is usually gained post telemedicine consultation (Q3); and 55% agreeing that questions answered profusely during a telemedicine consultation (Q4). However, the same percentage of respondents (41.25%) agreed and disagreed that telemedicine results in increased confidence in a doctor/patient relationship (Q2). A relatively low percentage of respondents strongly disagreed that telemedicine is an engaging platform (Figure 7). Figure 7: Percentage of respondents that find telemedicine to be an engaging platform In terms of information exchange during a telemedicine consultation more than half the respondents said that the explanation concerning the disease or treatment is not better (Table 4). Table 4: Information exchange during a telemedicine consultation Yes No Sometimes 1. In your experience, was the explanation concerning the disease better in telemedicine? 2.5% 58.75% 38.75% 2. In your experience, was the explanation concerning the treatment better in telemedicine? 6.25% 50% 43.75% 0.00% 10.00% 20.00% 30.00% 40.00% 50.00% 60.00% Q1 Q2 Q3 Q4 Strongly disagree Disagree Agree Strongly agree 48 4.4 Informed consent preference (written or verbal) to perform a clinical examination or when using ICT For this category, the study population scored according to elements from a list - take patient history, examine a patient, email patient information, refer patient, prescription, virtual consultation, none of the above, and other - for both written informed consent and verbal informed consent. Respondents were allowed to choose more than one option (Figure 8). Overall, more respondents (40%) took verbal consent to conduct virtual consultation compared to 14% that took written consent. Family physicians take an average of 2.61 (Figure 9 and Table 5) procedures relating to clinical examination or when using ICT using verbal format and an average of 1.08 (Figure 10 and Table 6) procedures relating to clinical examination or when using ICT using a written format. Figure 8: Study population preference on informed consent 0 10 20 30 40 50 60 Take patient history Examine a patient Email patient information Refer patient Prescription Virtual consultation None of the above Other Written Verbal 49 Figure 9: Written number of Procedures Table 5: Summary Statistics on written number of procedure Mean 1.0875 Std Dev 1.6396492 Std Err Mean 0.1833183 Upper 95% Mean 1.452386 Lower 95% Mean 0.722614 N 80 Skewness 1.9414634 Kurtosis 3.4516841 Figure 10: Verbal number of Procedures 50 Table 6: Summary Statistics on verbal number of procedure A Pearson Correlation of family physicians and written informed consent was higher at 0.1174 compared to the correlation of 0.0329 of family physicians and verbal informed consent, showing no real evidence of association. Mean 2.6125 Std Dev 1.9967933 Std Err Mean 0.2232483 Upper 95% Mean 3.0568646 Lower 95% Mean 2.1681354 N 80 Skewness 0.3878212 Kurtosis -1.126012 51 4.5 Availability and quality of essential resources to conduct telemedicine As with the facilitating condition, the responses in this category was significant with 72 of the respondents indicating that they do have an electronic device to implement telemedicine consultation followed by 42 who also had access to quality broadband (Figure 11). On average family physicians had access to 1.61 resources to conduct telemedicine consultation. Figure 11: Study population representation on tools used by doctors for telemedicine Of the 17 respondents that use a telemedicine software, 76% said that the software is user friendly. Respondents also reported that 71% of their patients found the telemedicine software to be user friendly (Table 7). Table 7: The efficiency of telemedicine software Yes No 1. If you do use a telemedicine software, is it user friendly for you? 76% 24% 2. If you do use a telemedicine software, is it user friendly for your patient? 71% 29% 0 10 20 30 40 50 60 70 80 Electronic device Telemedicine Software Quality broadband None Tools needed to facilitate a telemedicine consultation 52 The study population also reported that 59 of the medical doctor’s patients do have access to an electronic device, 20 have access to quality network while 18 have access to none (Figure 12). Figure 12: Study population representation on tools used by patients for telemedicine For those respondents that use telemedicine, preference for ease of using telemedicine, learning to operate and usefulness was measured (Table 8). On a scale of 1 = Never and 4= Always, the mean for telemedicine is easy to use was 2.65, which lies between rare and usually. On a scale of 1 = very difficult and 4 = very easy, the mean for learning to operate telemedicine systems was 2.81, leaning more towards easy. On a scale of 1 = never and 4 = always, the mean for telemedicine useful in your job was 2.73, leaning more towards usually. Table 8: Preference for ease of using telemedicine, learning to operate and usefulness Min Max Mean Std Deviation 1. Easy to use (1 = Never; 4= Always) 1 4 2.65 0.70 2. Learning to operate (1= Very difficult; 4= very easy) 1 4 2.81 0.69 3. Useful in your job (1= Never, 4= Always) 1 4 2.73 0.72 0 10 20 30 40 50 60 70 Electronic device Quality network None Tools needed by patients to conduct telemedicine consultation 53 4.6 Modalities used to store patient information The study population reported on how pat