Published in News
On 7 July 2021, GACD hosted our first ever Digital Health Workshop to uncover some of the barriers and lessons from implementing digital health interventions around the world.
Over the past two decades, the use of new technologies has grown exponentially across the world (particularly in high-income countries). Advances in digital technologies have unlocked the potential for a whole new range of approaches to healthcare, including, m-health, e-Health and digital health care, where innovative technologies are used to improve health.
On 7 July 2021, GACD hosted our first ever Digital Health Workshop, featuring expert presentations and discussions to uncover some of the barriers and lessons from implementing digital health interventions particularly as applied to non-communicable diseases (NCDs) in low- and middle- income countries (LMICs).
What lessons have we learned so far?
We invited five researchers to present the lessons they have learned from implementing digital health interventions in different settings around the world.
1. Know the data-sharing legislation
Marieke Hoevenaar-Blom from the University of Amsterdam spoke about her experiences of implementing the PRODEMOS digital health project in China, involving the use a mobile app (in Europe) to improve the self-management of dementia.
All started well with the project until the team learned that Chinese law required that personal data from Chinese participants be hosted in mainland China by a Chinese company. Given the project was a multinational partnership this was further complicated by EU data sharing laws. To comply with Chinese and EU laws, the team found a Chinese company to redevelop the app, and data was then anonymised before sharing.
Learning from this experience, Marieke strongly recommended that all researchers have an informed plan for hosting and data sharing before their project is initiated to avoid these problems.
2. Stand out from the crowd
Abdul Kuddus, from the Diabetic Association of Bangladesh, implemented a mobile-health (mHealth) intervention for type 2 diabetes in rural Bangladesh. The team developed mHealth voice messages, available free of charge, that were sent to anyone with access to a mobile phone in the intervention area twice a week.
The challenges experienced by the team included the high frequency of mobile phone number changes, message fatigue of the participants (who already received a lot of junk mail), and the fact that ingrained habits and behaviours make social change difficult.
To help overcome these issues, Abdul advised sending messages from a trusted source so they aren’t confused with junk mail, to ensure participation of the community in the design, and to use a multi-component, multi-sectoral approach.
3. Consider all aspects of accessibility
Jill Murphy, from the University of British Columbia, investigated the impact of the shift to e-care for mental health conditions in South Asia as part of the EMBED study.
The team found there were barriers to accessing e-care in certain populations, leaving them at risk of poor mental health, including those with pre-existing conditions, racialised and indigenous groups, elderly people, children and youth, people with disabilities and healthcare workers.
These marginalised groups were unsure how to access resources, had concerns about data privacy, could get overwhelmed with the options available, and be reluctant to access services due to lack of privacy in the home. Healthcare workers also identified difficulties in building trust, recognising non-verbal cues and issues around confidentiality.
To make e-care more accessible, Jill advised that the services and resources they be free of cost, advertised widely, involve partnerships with private firms to improve access, be informed by pre-existing guidelines, and involve a diverse range of stakeholders.
4. Collaboration is key
Ray Lam, from the University of British Columbia, outlined his work into implementing measurement-based care for depression using the WeChat platform in China.
The team developed a ‘mini programme’ on the platform, containing a mood-tracking feature, self-management programme, and an online coaching service. It was adapted from a Canadian project called ‘bounce back’, which uses telephone coaching.
Challenges experienced by the researchers included working with developers with limited digital technology knowledge, which made the formatting of lessons and coaching difficult. Ray advised involving collaborators from the very beginning and considering building an intervention from scratch rather than trying to adapt an existing programme.
5. Gain feedback from experts and your target audience
Maike Greve, from the University of Goettinggen, discussed the WHO-PEN at scale project, which is designing an app for community health workers in Eswatini, where deaths from NCDs are high.
The randomised control trial is using an offline app on mobiles and tablets, that supports community health workers to increase awareness and prevention of NCDs amongst local people.
Using an iterative design process with different stakeholders, including experts at the Ministry of Health, the team streamlined the app to ensure it was useful and context appropriate. They also co-designed the logos and visuals in the app with the health workers themselves.
What are the key barriers for implementing digital health projects?
Participants also heard from Josephine van Olmen, from the University of Antwerp, who published a paper as part of the GACD mHealth Working Group about the barriers for implementing mHealth interventions for NCD management in LMICs.
From a scoping review and field-based views from implementers, they found that the impact of mHealth in LMICs is currently modest and variable according to the setting. Key barriers to implementation include:
NCD-related barriers (age, complications, disease progression, NCD prevalence among older people and those with lower socio-economic status associated with lower digital literacy, stigma).
Mobile intervention barriers (translation of personal motivational coaching to a mobile app)
Technical barriers (network provider, operating systems, hardware, software, links between digital systems, separation from existing health management information system)
Health care service barriers Lack of resources, access to care, integration into existing processes)
Contextual barriers (security, gender differences, social, economic, and cultural factors, access to devices, perceptions of tracking software)
Regulatory barriers (lack of clarity on digital health regulations in many countries, reverse billing must allow for special short codes, bulk messaging leading to intervention messages identified as ‘spam’).
Are there other technical considerations?
Dr D Praveen, from the George Institute for Global Health, offered some helpful technical considerations for successful digital interventions in LMICs, drawing from his experience as a contributing author to the WHO’s Be Healthy Be Mobile guidance on implementing mHypertension programmes.
He advised research teams to consider: the type(s) of digital health technology to be used (e.g., SMS, apps, voice, etc.); the available technology options in the context where the research will occur; the process for procurement and adaption of the selected technology; the process of dashboard development to monitor project process and outcomes; procurement of a short code; pricing (which will include negotiation with telecoms regulators, aggregators and operators); data security; and technology piloting and scale-up plans.
What do our experts advise?
We asked each of our expert panellists for one piece of advice they would give for digital health projects:
Sameer Pujari from the World Health Organization, says: “Don’t reinvent the wheel, learn from others.”
Smisha Agarwal from Johns Hopkins Bloomberg School of Public Health says: “Avoid past failures by starting with the WHO digital intervention guide.”
Rick Glazier from the Canadian Institutes of Health Research; Senior Scientist, ICES, says: “Design for impact from the start; plan for scale.”
Maike Greve from Georg-August-Universität, says: “Do it – but do it for the impact not for the proof of concept.”