Being Spontaneous: The Future of Telehealth Implementation?

Introduction

Telemedicine is evolving in different ways in health systems with different priorities, funding models, regulations and cultures. Uptake has been generally swifter and more successful in the private sector than in the publicly funded state sector. Early advocates of telemedicine advised that successful telemedicine was dependent on the ease of use of technology with the human elements taken care of by champions who would lead from the front. The days of exploration! There is general agreement that telemedicine is successful when it has reached the point of acceptance and become an integral part of the everyday practice of a discipline replete with clinical guidelines. Radiology is the prime example. We seldom hear of teleradiology except from developing world countries where digital radiology is still a novelty. ¹

What is required to develop a successful telemedicine service? Over many years an almost traditional process has been developed that guides most implementations of telehealth initiatives. In particular, the literature shows that the issues of strategy development, ² needs assessment, ³ readiness assessment, ⁴ business case/plan development, ⁵ change management, ⁶ and evaluation ⁷ have become synonymous with telehealth implementation. Historically, individual approaches have placed differing emphasis on some issues. In the 1970s and 1980s, four basic principles were espoused as principles for success: these were (1) all activities are based on a legitimate need, (2) the simplest least expensive technology is used to meet the need, (3) the network is shared by a variety of users, and (4) users are properly trained and supported ⁸ (Table 1).

In 1997, Yellowlees put forward “seven core principles” for successful development of telemedicine (Table 1). ⁹ This has grown to “eighteen critical success factors” built around strategy and management; organizational implementation and change management; legal, regulatory, and safety issues; and technical infrastructure and market relation ¹⁰ (Table 1).

In the intervening period, various other aspects have been considered. Wootton and Hebert in 2001 spoke of high-quality care at low cost, perspective of cost (to whom the cost or saving is directed), and workload, but also indicated “political imperative” might trump rational descriptions of success. ¹¹ Buck promoted a cognitive-emotional approach, and proposed nine human factors that (if embedded into requirements' engineering, development processes, and product life cycles) would lead to greater success of telehealth initiatives. ¹² Even vendors have shared their opinion: establish a vision, build a long-term financial plan, create a convenient and effective work environment, mainstream telemedicine into the standard care process, plan and assure effective training, have a full-time coordinator(s) and an effective leader and cheerleader, a project plan = manageable milestones = reasonable expectations, horizontal versus vertical implementation, good marketing is critical, and publish or perish. ¹³

A typical methodical approach was exemplified by Burgiss, who recommended seven steps in planning for telehealth success, which were evaluate needs, develop care services plan, develop business plan, plan technology, train personnel, test care and technology plans, and evaluate outcomes. ¹⁴ Such guidance grew from both theory and practical implementation experience to become “best practice” and is still believed to ensure successful implementation of any telehealth initiative.

In contrast to these perspectives, Zanaboni and Wootton recently identified four different perspectives of importance to telehealth implementation. ¹⁵ These were (1) “advantages for users are the crucial determinant of the speed of adoption of technology in healthcare,” (2) “the adoption of telemedicine is similar to that of other health technologies and follows an S-shaped logistic growth curve,” (3) “evidence of cost-effectiveness is a necessary but not sufficient condition for widespread adoption of telemedicine,” and (4) “personal incentives for the health professionals involved in service provision are needed for widespread adoption of telemedicine to occur.”

But what happens if a telemedicine service is unplanned and evolves spontaneously? (Table 1). Is it doomed because the various planning and implementation steps have not been followed or is it an indication of a maturing field with acceptance by clinicians for use of everyday technology, including their mobile phones, to the advantage of both patients and themselves?

This article provides early reports of two telemedicine services that have evolved spontaneously without advanced planning or following established best practice implementation. Anecdotal evidence exists of other spontaneous services arising locally: teleophthalmology, telecardiology, telephysiotherapy, teleradiology, and teleorthopedics. In this context, we apply the Oxford English Dictionary definition of spontaneous: “Performed or occurring as a result of a sudden impulse or inclination and without premeditation or external stimulus.” This lends support to the first and last principles espoused by Zanaboni and Wootton, displaying advantages for users (a crucial determinant of the speed of adoption of technology in healthcare), and personal incentives for the health professionals involved in service provision (necessary for widespread adoption of telemedicine). ¹⁵ The advent of “spontaneous services” flaunts the traditional approach, yet may be most likely to lead to successful integration and sustained application, particularly in the less regulated developing world setting.

Setting

These services began in KwaZulu-Natal Province, South Africa, and have mostly involved staff and postgraduate trainees (residents/registrars) at the local medical school, those working for either government-funded hospitals or the medical school, and some in private practice. KwaZulu-Natal is a province of 94,361 km² with a population of ∼10.5 million people, 53% of whom live in rural areas and ∼85% are dependent on government healthcare. ¹⁶ The medical staff working in government hospitals and the medical school are salaried and not paid per patient on a fee for service basis. Remuneration for the telemedicine service is therefore not an issue.

The hospital system is the traditional model of primary health clinics and district, regional, and central tertiary hospitals with referral following this path. Most district hospitals lack specialists in various disciplines, and specialist staff at regional and tertiary hospitals are expected to participate in outreach programs to district hospitals where there are no specialists. When done, this requires driving to the hospital or flying several hundred kilometers with the air ambulance service. Teleconsultation by videoconferencing has been used to augment outreach in some instances. ¹⁷

There have been several small scale telemedicine initiatives in the province that have been running for up to 12 years. ^17,18 Videoconference-based tele-education is widely used. ¹⁹

More recently, spontaneous telehealth initiatives have arisen.

Case 1: Mobile Teledermatology

The Department of Dermatology at the Nelson R Mandela School of Medicine was an early adopter of synchronous teledermatology in 2003. Four district hospitals regularly conduct scheduled referral sessions with dermatologists at either the medical school or Inkosi Albert Luthuli Central Hospital in Durban. This has saved more than 70% of patients a journey to Durban, a round trip that can take several days. ¹⁷

Store-and-forward (SF) teledermatology has been less successful. In 2001, a group of doctors working in the most rural hospitals were brought to Durban and given a digital camera and training in photography for SF telemedicine. At that time, Internet access in the hospitals was very limited. Physicians would therefore take the photographs and make notes during the consultation with the patient, then download them at home after work. Cases were sent by e-mail to a dermatologist at the medical school from the physicians' own computers and using their own, at that time very expensive, bandwidth. The dedication of referring physicians was great, but attrition and staff turnover led to near termination of the service.

In 2007, as part of an educational exercise, trainee dermatologists were given digital cameras and asked to take photographs of skin lesions when they were uncertain of diagnosis or management and send them to their mentor for advice. This soon failed, despite availability of computers within the hospital, these were not easily accessible and it still required “extra” steps in the workflow.

However, late in 2013, senior dermatologists reported that they began receiving unsolicited SF cases (text messages and images) from their trainees and rural doctors. These cases had been photographed and sent from smartphones or tablets through the WhatsApp application. The dermatologists replied from their mobile device. In some instances, after initial review of the case material, the referring physician was called and the case discussed. Response time varied in relation to urgency of the referral and was sometimes immediate. This service had not been planned, and the dermatologists were happy to respond because dealing with the case online potentially meant one less person in their outpatient clinic. The anecdotal perception of the dermatologists is that 75–80% of patients managed in this way are saved a referral for face-to-face consultation.

Case 2: Teleburns Management

The first dedicated burn unit in KwaZulu-Natal was established in 2000 with 46 beds at Inkosi Albert Luthuli Central Hospital and admits about 150 patients a year. ²⁰ Owing to staffing constraints, only 20 beds are regularly available. There are now three burn units in the province with a theoretical total of 112 burns beds. It is estimated that there are in excess of 7,000 burns patients a year in the province ²¹ and that there should be at least 475 burns beds available. ²⁰ District and Regional Hospitals with surgical units are reluctant to manage burns patients, and inappropriate and delayed referral to burns units of adults (40%) and children (30%) is common. In addition, clinicians who do not regularly manage burns tend to overestimate the area and underestimate the depth of the wound resulting in both over and under triage. The unit has had a policy of not accepting referrals for admission, or to their weekly clinic, without a discussion with the referring doctor to determine whether the patient warrants referral, or when no beds are available, advising on immediate management. A paper record of each telephonic consultation is kept on a standardized preadmission form.

It is not always possible for senior burns staff to be present when dressings are changed in the unit. To assist in making decisions on further management, for example readiness for skin grafting, junior staff began taking photographs of the exposed wounds on their mobile phones and discussing the pictures with their seniors. The images were not transmitted but assessed directly from the stored photographs. This has been a regular practice for more than 3 years and nursing staff also take photographs on their own phones if no doctor is available.

In late 2014, a decision was made to request referring doctors to submit photographs of the burn wounds after the telephonic assessment and completion of a preadmission assessment form, and before making a decision on the patient's acceptance. The images were sent using WhatsApp or occasionally multimedia messaging service. The decision is then sent back to the referring doctor by WhatsApp or short message service (SMS). If the decision is not to accept the patient, advice on further management is given with a request for more photographs after initial treatment. Addition of photographs has altered the management of two-thirds of the 116 patients referred in the first 6 months of its introduction. In November of 2014, a database was developed to store information from the preadmission assessment form and the photographs. The database is currently on a burn specialist's laptop computer and password protected. It is backed up on the hospital's Intranet that is also password protected.

No planning for this service was made, other than a decision to require provision of photographs of the burn wounds before a decision to accept the patient was made. This decision was based on the successful experience of taking photographs during wound dressing for later clinical review, and the desire to improve existing decision making based solely on verbal communication and completion of a preadmission form. The database was seen as a more efficient way of storing data and images, and is the first step in further evolution of the service.

Common Thread

These two examples involve existing services within clinical departments. In each of these cases, what changed? Certainly technology and cultural mindsets—access to and use of smartphones—reduced data costs allowing easy Internet access and a growing culture of sharing images and data through social networks. But doctors had also worked out for themselves that they could—perhaps unwittingly—develop a SF service that met their and their patients' needs. In accordance with the definition of “spontaneous” (performed or occurring as a result of a sudden impulse or inclination and without premeditation or external stimulus), each service was unplanned and is growing.

Many of these referrals occur during the consultant's own busiest onsite consultation period, yet responses are rapid, and have even taken place after hours or during vacation or conference trips overseas. Inherent benefit must be seen for such dedication and sustained activity, as suggested by Zanaboni and Wootton (the service displaying advantages for users, and personal incentives for the health professionals involved in service provision). ¹⁵ As seen in the dermatology example, we anticipate that the next evolution may be to establish “mixed spontaneous services.” That is where the mobile technology will be used to send pictures and a brief patient history (SF or asynchronous), and the specialist will simply call the referring physician immediately to review and make recommendations on the case (real time or synchronous). Such an approach, more compatible with availability of an on-call specialist, would result in very efficient and likely more effective patient care.

Discussion

WhatsApp is a proprietary, free, mobile phone instant messaging client for smartphones using Android, iOS, Windows, Blackberry, or Nokia operating systems. With more than one billion users in February 2016, it is in everyday use. Without the need to access expensive videoconferencing equipment or computer networks, it is not surprising that doctors in a developing world country have found practical uses for it in both clinical and administrative settings. Can such spontaneous and essentially “free” services evolve in the developed world, or do regulations and legislation concerning patient confidentiality and data security preclude its use? Is this just an anomaly of the developing world?

There are more than 30 articles from 10 countries reporting the use of WhatsApp for telemedicine. The WhatsApp chat group feature is most commonly used with text messages, photographs, and or video sent to members of closed groups in clinical teams or departments. Communications range from second opinions, updates on patient admissions, changes in treatment, and sharing of X-rays and clinical photographs to theatre scheduling and sharing of educational material. Six clinical services, other than intradepartmental groups, included a diabetic retinopathy service in India that used a fundal camera attached to a smartphone, ²² a triage service for oral and dental pathology in Italy, ²³ an after-hours second opinion service for emergency maxillofacial injuries in Turkey, ²⁴ an emergency cardiology program in Turkey, ²⁵ and patients in Spain using WhatsApp to contact their allergist. ²⁶ An overview of use of surgical apps in clinical practice mentions the use of WhatsApp by a pediatric surgeon in South Africa. ²⁷ Only the two services in Turkey are similar to the dermatology and burn services reported here.

It should be noted that WhatsApp messages, images, audio, and video have only been end-to-end encrypted across all operating systems since April 2016. Before this only one-to-one text messages sent on Android devices were end-to-end encrypted. ^{28 –30}

A number of practical, legal, and regulatory issues arise when considering the development and practice of spontaneous telehealth services, some of which are outlined hereunder. These shortcomings need to be identified and addressed as any one of these could be leveraged to cease the neophyte spontaneous practices described. However, these spontaneous practices have arisen through an identified health practice need, provide clear benefit to patients and providers alike, improve patient care, and have become embedded into routine practice—each a parameter of success. A much better approach would be to use the shortcomings identified hereunder as examples and develop pragmatic solutions that will educate and guide spontaneous service providers and lead to functional and appropriately delivered services. Conversely, the practical, legal, and regulatory expectations they must achieve have to be appropriate also. Discipline-specific guidelines will be important and have been (or are being) developed in South Africa ³¹ and elsewhere, ^32,33 often based on international guidelines. ^34,35

Furthermore, “social norms” guide what society believes to be acceptable behavior and practice. But social norms change and can change rapidly, especially so in the use of information and communication technologies. Any entity that would strive to set ethical and practice standards for e-health must ensure that its guidelines not merely keep pace with, but perhaps foretell, evolving social norms, and recognize and accommodate them accordingly. To do otherwise would be unethical deprivation of the benefits of e-health to many, including vulnerable populations in most need of these services.

There is currently limited telemedicine or telehealth legislation globally, and none in South Africa. The Health Professions Council of South Africa, a statutory body tasked with regulating the practice of medicine in South Africa, has been working on draft ethical guidelines for the practice of telemedicine in South Africa for more than 8 years. ³⁶ The two services described, which meet the definition of telemedicine, would be in breach of proposed ethical guidelines, which require that telemedicine consultations be restricted to situations in which a previous healthcare provider–patient relationship exists, paper records of written informed consent be kept by the patient, referrer, and specialist, among other issues such as record keeping, data encryption and security, and data quality.

Issues of confidentiality and data security that are well understood in the developed world are less clear cut in the developing world where cultures, beliefs, and mores may not follow Western world norms and expectations. As Pandian noted, “There are no clear guidelines for patient confidentiality from both the Dental Council and Medical Council of India currently and as such is less of an issue in India as compared to the developed countries.” ³⁷ A recent study in KwaZulu-Natal showed that only 35% of patients understood the meaning of the word consent in their mother tongue, isiZulu. ³⁸

However, the current spontaneous services are not without fault and need support and guidance to evolve and mature. In the dermatology service, patient information and responses are stored on the specialists' mobile phones, devices, or computers. There is no central repository of this information. Although the burns service is storing information in its password-protected database, the SMS or WhatsApp messages are stored on the phones. Some are deleted after transmission and entry into the database. The occasional e-mail received by the dermatologists is unencrypted, and no data are available on whether they are encrypting their stored mail. Only 4% of doctors in a previous study in the region encrypted their email. ³⁹ Other issues identified are that clinicians have changed phones (with photographs remaining on the old phone), some practitioners delete cases that are not interesting or educational enough, whereas others simply delete past consultations as their smartphone memory becomes depleted (meaning any record is incomplete), and there is uncertainty about any record of informed consent.

Conclusion

Although specifics have changed over the years in terms of recommended steps for successful telehealth implementation (Table 1), similarities exist. A traditional approach has arisen based on theory and practice and recently included 18 critical success factors that must be addressed. In contrast, a variety of “spontaneous” telehealth implementations have been described. It is anticipated that the same trend is occurring elsewhere in the world, but has not necessarily been recognized as a fundamental shift in approach or not yet published. This article is believed to be the first report of the occurrence of “spontaneous and sustained” telehealth applications, and has provided insights into an alternative approach to implementation and integration of telehealth worldwide. This should stimulate discussion and action around appropriate regulatory guidance, and about how the spontaneous approach can be identified, nurtured, and matured, rather than being prematurely stifled.