All 5 entries tagged Digital Technologies

mhealth innovation and business

Many successful digital platforms are based on facilitated user networks, where the same people buy and sell and deliver and receive things to and from each other e.g. eBay. This is a business model that can deliver quality health care at a low cost for behaviour-dependent long term conditions [1]. Examples include dLife https://dlife.com/, Mumsnet https://www.mumsnet.com/ and Patients Like Me https://www.patientslikeme.com/. These platforms incorporate online social networking but are also integrated with established health care [2]. These platforms also use established business model for sustainability, selling advertising opportunities or the data they collect from users. Venture capital enables many companies with innovative digital platforms to move into the market. Companies may establish service delivery and gain user data through working with employer organisations and run a charitable arm to allow reach into low resource settings. What are the other alternatives for sustainability that keep the ownership of the digital platform with innovators and users? For example, where individuals provide their data can they become part of the business – like a shareholder? Can a user fee be pitched low enough to encourage use by users from low socio-economic backgrounds? Can a payment be crowd-funding rather than a purchase? A small fee may increase user engagement with the platform.

In low and middle income countries private sector initiatives provide low cost, high quality health care where the public sector has failed to deliver[3]. Those which do not rely on donations and grants focus on delivery of specific treatments such as cataract or heart surgery [3] where they can achieve high volume for low cost [1]. What are the business models which would make digital health platforms sustainable in this context? Digital infrastructure is usually in place and platforms in other domains such as money transfers are sustainable. Where potential users of health care in low and middle income countries do not feel empowered to claim any health care that is available and don’t perceive the need for preventive health such as ‘check ups’ the business model needs to change non-users into users [4].

Academic spin-off companies with innovative technology for improving health often fail because of mismatch between technology design and business model[5]. Innovative business models can enable innovative technology to become established [1] but the process of stakeholder interaction about technology design and its value proposition leads to development of a business model that is often not innovative with (often negative) consequences for the spin-off [5]. How are value propositions aligned to allow for successful launch of innovations?

Written by Frances Griffiths 9th December 2018

1. Hwang J, Christensen CMJHA: Disruptive innovation in health care delivery: a framework for business-model innovation. 2008, 27:1329-1335.
2. Griffiths F, Dobermann T, Cave JA, Thorogood M, Johnson S, Salamatian K, Olive G, Francis X, Goudge J: The impact of online social networks on health and health systems: a scoping review and case studies. Policy and Internet 2015.
3. Bhattacharyya O, Khor S, McGahan A, Dunne D, Daar AS, Singer PAJHRP, Systems: Innovative health service delivery models in low and middle income countries-what can we learn from the private sector? 2010, 8:24.
4. Pels J, Kidd TAJIJoP, Marketing H: Business model innovation: Learning from a high-tech-low-fee medical healthcare model for the BOP. 2015, 9:200-218.
5. Lehoux P, Daudelin G, Williams-Jones B, Denis J-L, Longo CJRP: How do business model and health technology design influence each other? Insights from a longitudinal case study of three academic spin-offs. 2014, 43:1025-1038.

Frances Griffiths : 09 Dec 2018 19:44 |  Tags: Business Conditions; Digital Technologies Health; Long Model Term |  Comments (0) |  Close comments |  Report a problem

Social and ethical implications of use of AI in health care

Writing about web page https://warwick.ac.uk/fac/med/research/hscience/sssh/research/lyncs/

In this blog I consider some of the social and ethical implications of the use of AI for algorithm driven clinical decision-making.

Developments in AI for health seem to challenge the role of and even need for health practitioners.

In the medium term, at least in the UK, society seems to require AI and clinicians to work together. There is a growing consensus that for a decision such as a health decision, where the implications for the individual are considerable, the individual should be informed that the decision is being made using AI and have the right to request that the decision is not made by AI^{1 2}. If a decision is made using AI they should have the right to ask for the decision to be reconsidered^{1 2}. It is not yet clear where responsibility lies for the consequences of advice given directly to the patient based on AI³, and there is a deman for the AI algorithms to be explainable⁴.

If part of the work of a clinician undertaken by AI, there is a question of how they maintain their expertise⁵ inorder to take responsibility for decisions and to provide a second opinion when requested. Further, the clinicians need to understand AI in order to be able to explain the decision-making to their patients.

Low and middle income countries (LMIC) do not have the health parctitioner workforce capacity that high income countries enjoy but most of them have good digital infrastructure – far better than water or roads. There is potential for AI to be rapidly taken up in LMIC to keep healthcare cost down whist providing improved services. In this scenario the lack of human capacity in the system might mean the decision-making process is not explained to patients and there is no second opinion available.

1. House of Lords Select Committee on Artificial Intelligence. AI in the UK: ready, willing and able?, 2018.

2. The Alan Turing Institute. The GDPR and Beyond: Elizabeth Denham, UK Information Commissioner, 2018.

3. Floridi L, Taddeo M. What is data ethics?: The Royal Society, 2016.

4. Information Commissioner’s Office. Information Rights Strategic Plan 2017-2021

5. Yang G-Z, Bellingham J, Dupont PE, et al. The grand challenges of Science Robotics. Science Robotics 2018;3(14):eaar7650.

Frances Griffiths : 04 Jul 2018 18:25 |  Tags: And Artificial Capacity; Clinical Communication Countries; Digital Technologies Health Income Intelligence Low Lync Study; Middle Service |  Comments (0) |  Close comments |  Report a problem

Digital communication between patient and clinical team forms basis for Artificial Intelligence

Writing about web page https://warwick.ac.uk/fac/med/research/hscience/sssh/research/lyncs/

People with a health condition have long stretches of time between encounters with their healthcare team when they get on with living with their condition. People experience change in their condition, treatment effects and side effects mostly on their own - without engagement with their healthcare team. They have to interpret what they experience themselves. One of the most memorable comments a patient made to me as a young doctor was:

Going home with a prescription for a new treatment can feel very lonley.

Accessible information and explination makes a difference, but even where we have good evidence about a condition and its treatment we still only know what usuallyy happens most of the time. We cannot predict exactly what will happen to one particular patient¹.

In our study of young people living with longterm conditions (the LYNC study)² it was at times of change - new treatments or worsening symptoms or new life experiences such as going to univeristy - that young people most appreciated and benefited from digital access to their healthcare team. They were able to text, email or phone about what they were experiencing and receive interpretation, reassurance and guidance.

Communication via digital channels is easily recorded and stored. By enabling patients, between routine appointments, to digitally contact their clinical team we can build datasets of their concerns and the team's responses. This can be used as the training dataset for a chatbot based on artificial intelligence (AI), working alongside the clinical team. Initially the chatbot might have sufficient data to learn how to respond to common questions, for example, how to take a medication. As the dataset grows, the AI algorithm can learn how to respond to more complex clinical questions. By linking with patient records, responses can be tailored to individual patients. The AI chatbot can also learn from what happens to patients and does not forget. In theory an AI chatbot could become as good or even better than a clinician, although there would still remain uncertainty.

Clinician and chatbot working together could be quite a team and counter each other's cognitive biases.

This type of development can begin right now but there are ethical and social implications that need our attention.

1. Gorovitz S and McIntyre A. Towards a Theory of Medical Falibility. The Journal of Medicine and Philosophy. 1976:1(1):51-71

2. Griffiths F, Bryce C, Cave J et al. Timely digital patient-clinician communication in specialist clincal services for young people: a mixed-methods study (The LYNC Study). Journal of Medical Internet Research. 2017;19(4):e102

Frances Griffiths : 10 Apr 2018 17:53 |  Tags: Artificial Clinical Communication Digital Technologies Intelligence Lync Study |  Comments (0) |  Close comments |  Report a problem

Where to start with digital communication with patients?

Writing about web page https://warwick.ac.uk/fac/med/research/hscience/sssh/research/lyncs/

Design a bespoke platform for your clinical service or use existing digital systems?

One answer is to do both! Evidence from our empirical study [1] suggests that when introducing digital communication with patients, changing the working patterns of the clinical team is the difficult part. There needs to be adaptation and flexibility, both on the side of the clinical team and on the side of the new intervention - the digital communication system.[2] The ideal might be, for example, a bespoke patient portal. By introducing some elements of the patient portal using existing digital systems gives time for change in work patterns. It also provides insights for the design of the portal. Patients interviewed in the LYNC study continued to use digital communication channels with their clinical team if they were confident the clinical team would respond.[1] So, if we want patients to use a patient portal the clinical team has to be responsive from the moment it is launched.

Consider this example. An oncologist wants to develop a patient portal that can

• Store for patient viewing, the patients individualised treatment plan and any updates
• Provide information for patients to read about their treatment including when to contact the clinical team
• Send reminders for patients receiving chemotherapy for blood tests, chemo sessions and review appointments
• Enable patients to view their blood test results and analysis of trends
• Enable the patient to contact the clinical team when they have symptoms of concern.

At the back end of the patient portal, the clinicians want to:

• View the patients clinical notes while on the move to inform their response to patients contacting them
• Enter information about patient encounters in the clinical notes while on the move.

The oncologist has already identified potential savings from this innovation including reduction in wastage of chemotherapy drugs and reduction in number of A&E attendances.

Oncology patients have a lot to deal with, so a patient portal where everything they need is in one place seems ideal – although we need to check this with patients. Not everyone in the clinical team is as engaged with the idea as our oncologist but there are some potential quick wins that might persuade them. For example, text message reminders improve patient attendance at clinic appointments.[3] The hospital already has a reminder system for standard outpatient appointment. Could the existing system be used to send reminders for blood tests and non-routine appointments?

Testing out the use of video-conferencing using the existing hospital system would allow the enthusiasts in the oncology team to firstly work out how to integrate it into their work pattern and secondly, to assess how much advantage it provides over telephone – is it a necessary part of the patient platform? In the LYNC study we undertook a scoping review of videoconferencing.[1)] Patients mostly like it and sometimes it improves health outcome, but it does not usually save clinician time or save money. However, where there is a specific reason for using it, for example saving travel time for appointments, it can be of benefit. In our example the availability of videoconferencing might provide more information about a patient’s condition than telephone, so the clinician might be able to reduce the number of patients making a trip to the hospital.

Evidence from the LYNC study [4] suggests patients benefit from being able to engage digitally (email, text, mobile phone) with their clinical team when they most need it. This service can be offered using existing systems – hospital email and mobile phones. What is difficult is working out who is going to respond to patient contacts, when and where. Come the launch of the patient portal any new work patterns need to be well established. Our LYNC study Quick Reference Guides [5] suggest what needs to be thought through when offering patients digital engagement with their clinical team.

All innovation needs resource for its implementation. I have suggested digital engagement with patients is introduced in small steps. A danger is that the resources needed for implementation of the small steps is ignored so the innovations fail or the enthusiasts bear the burden. When a new digital system is introduced in one go, the resource needed for implementation is less easy to ignore.

-- Frances Griffiths, Professor of Medicine in Society and Penny Kechagioglou, the Oncologist

References:

1. Griffiths F, Armoiry X, Atherton H, Bryce C, Buckle A, Cave J, et al. The role of digital communication in patient-clinician communication for NHS providers of specialist clinical services for young people (The LYNC study): a mixed methods study. NIHR Journals Library Publications, HS&DR. 2018 in press.

2. May CR, Johnson M, Finch T. Implementation, context and complexity. Implementation Science. 2016;11(1):141.

3. Guy R, Hocking J, Wand H, Stott S, Ali H, Kaldor J. How Effective Are Short Message Service Reminders at Increasing Clinic Attendance? A Meta-Analysis and Systematic Review. Health Services Research. 2012;47(2):614-32.

4. Griffiths F, Bryce C, Cave J, Dritsaki M, Fraser J, Hamilton K, et al. Timely digital patient-clinician communication in specialist clinical services for young people: a mixed-methods study (the LYNC study). Journal of Medical Internet Research. 2017;19(4).

5. LYNC study team. LYNC Study Quick Reference e-book and Topic Guides: University of Warwick; 2017 [10th June 2017]. Available from: https://warwick.ac.uk/fac/med/research/hscience/sssh/research/lyncs/outputs/

Bronwyn Harris : 02 Feb 2018 16:40 |  Tags: Clinical Communication Digital Technologies Lync Study |  Comments (0) |  Close comments |  Report a problem

New Framework to Guide the Evaluation of Technology–Supported Services

Writing about web page https://www.jmir.org/article/viewFile/jmir_v19i11e367/2

Heath and care providers are looking to digital technologies to enhance care provision and fill gaps where resource is limited. There is a very large body of research on their use, brought together in reviews, which among many others, include, establishing effectiveness in behaviour change for smoking cessation and encouraging adherence to ART,[1] demonstrating improved utilisation of maternal and child health services in low- and middle-income countries,[2] and delineating the potential for improvement in access to health care for marginalised groups.[3] Frameworks to guide health and care providers when considering the use of digital technologies are also numerous. Mehl and Labrique’s framework aims to help a low- or middle-income country consider how they can use digital mobile health innovation to help succeed in the ambition to achieving universal health coverage.[4] The framework tells us what is somewhat obvious, but by bringing it together it provides a powerful tool for thinking, planning, and countering pressure from interest groups with other ambitions. The ARCHIE framework developed by Greenhalgh, et al.[5] is a similar tool but for people with the ambition of using telehealth and telecare to improve the daily lives of individuals living with health problems. It sets out principles for people developing, implementing, and supporting telehealth and telecare systems so they are more likely to work. It is a framework that, again, can be used to counter pressure from interest groups more interested in the product than the impact of the product on people and the health and care service. Greenhalgh and team have now produced a further framework that is very timely as it provides us with a tool for thinking through the potential for scale-up and sustainability of health and care technologies.[6]

Greenhalgh, et al. reviewed 28 previously published technology implementation frameworks in order to develop their framework, and use their own studies of digital assistive technologies to test the framework. Like the other frameworks this provides health and care providers with a powerful tool for thinking, planning and resisting. The Domains in the Framework include, among others, the health condition, the technology, the adopter system (staff, patients, carers), the organisation, and the Domain of time – how the technology embeds and is adapted over time. For each Domain in the Framework the question is asked whether it is simple, complicated or complex in relation to scale-up and sustainability of the technology. For example, the nature of the condition: is it well understood and predictable (simple), or poorly understood and unpredictable (complex)? Asking this question for each Domain allows us to avoid the pitfall of thinking something is simple when it is in reality complex. For example, there may be a lot of variability in the health condition between patients, but the technology may have been designed with a simplified textbook notion of the condition in mind. I suggest that even where clinicians are involved in the design of interventions, it is easy for them to forget how often they see patients that are not like the textbook, as they, almost without thinking, deploy their skills to adapt treatment and management to the particular patient. Greenhalgh, et al. cautiously conclude that “it is complexity in multiple domains that poses the greatest challenge to scale-up, spread and sustainability”. They provide examples where unrecognised complexity stops in its tracks the use of a technology.

-- Frances Griffiths, Professor of Medicine in Society

References:

1. Free C, Phillips G, Galli L. The effectiveness of mobile-health technology-based health behaviour change or disease management interventions for health care consumers: a systematic review. PLoS Med. 2013;10:e1001362.
2. Sondaal SFV, Browne JL, Amoakoh-Coleman M, Borgstein A, Miltenburg AS, Verwijs M, et al. Assessing the Effect of mHealth Interventions in Improving Maternal and Neonatal Care in Low- and Middle-Income Countries: A Systematic Review. PLoS One. 2016;11(5):e0154664.
3. Huxley CJ, Atherton H, Watkins JA, Griffiths F. Digital communication between clinician and patient and the impact on marginalised groups: a realist review in general practice. Br J Gen Pract. 2015;65(641):e813-21.
4. Mehl G, Labrique A. Prioritising integrated mHealth strategies for universal health coverage. Science. 2014;345:1284.
5. Greenhalgh T, Procter R, Wherton J, Sugarhood P, Hinder S, Rouncefield M. What is quality in assisted living technology? The ARCHIE framework for effective telehealth and telecare services. BMC Medicine. 2015;13(1):91.
6. Greenhalgh T, Wherton J, Papoutsi C, Lynch J, Hughes G, A’Court C, et al. Beyond Adoption: A New Framework for Theorizing and Evaluating Nonadoption, Abandonment, and Challenges to the Scale-Up, Spread, and Sustainability of Health and Care Technologies. J Med Internet Res. 2017;19(11):e367.

Bronwyn Harris : 15 Jan 2018 11:23 |  Tags: Clinical Communication Digital Technologies Technology Implementation Frameworks |  Comments (0) |  Close comments |  Report a problem