WHO guideline recommendations on digital health

Speaking at the HETT 2019 conference on healthcare and technology, WHO consultant Susannah Robinson, explained the WHO’s key guideline recommendations for the future of digital health.


In 2019, the World Health Organization (WHO) produced its first comprehensive guideline on digital health interventions. Speaking at the HETT 2019 conference on healthcare and technology, WHO consultant Susannah Robinson explained the WHO’s key guideline recommendations for the future of digital health.


WHO produces guidelines on a number of different health topics for countries around the world: We do this partly to help countries decide what they should be working on within their health systems, what the priorities should be; and also as an investment case [determining] where should they be focusing their resources.


Why digital health?

Until recently, there wasn’t much on digital health; this is a problem for a couple of reasons. Firstly, that makes it difficult for us to give guidance to countries on what works within digital – not necessarily just as a binary question of ‘digital: yes or no’, but [in terms of] what kinds of digital tools might work and how should they be used? What’s the evidence base for it? That is particularly important in contexts where countries might have to make a trade-off between a digital versus a non-digital alternative, so [they must determine] the opportunity cost from digital.


Secondly, along with governments making decisions, we also have global financing mechanisms like the World Bank or the Global Fund, which will look to WHO for guidance on where they should be channelling resources for countries and for themselves. You need that evidence-based guidance to help direct investment into certain areas of digital health.


Thirdly, we’re doing it because we have to: as a Member State-driven organisation, if Member States come to WHO with a request for advice on an area, we have to provide that advice.


In 2016, we had a request from Member States during our executive governing board meeting for guidelines on digital intervention prioritisation for meeting the [United Nations’] Sustainable Development Goals. Then two years later, we had a World Health Assembly resolution on digital health. All our 194 Member States came forward and said: ‘we would like more guidance on this, please do something about it’; so we had to do something about it.


Addressing healthcare challenges

What we tried to focus on is not digital for digital’s sake – it’s about focusing on the challenges that digital is trying to address within a health system, as a means to an end. [In terms of] needs within a health system or challenges that a health system is facing, how does digital, as one of a number of different mediums, help you respond to those needs and challenges?


In this context, it’s [important to consider] the end user – what do individuals need from the health system – and going digital is one way that you might fulfil those needs. The challenges that we typically come across [can pertain to]:

  • Population health;

  • The availability of services themselves;

  • Quality of those health services;

  • Cultural or social acceptability – [whether services are] appropriate for the populations they are trying to reach;

  • Utilisation – are they even being used?

  • Efficiency: are they being delivered in the best way? Can digital health aid that?

  • Cost, obviously of concern and

  • Accountability.


Is [the system] always working in the best way that it could; and if not, how are we improving the accountability of that?


It’s not an objective in and of itself to be digital. It’s [a question of] how do you create a health system that’s as good as it possibly could be, and what [aspects of] that can digital support?


For example, if you know that vaccination for particular groups is a priority, but there’s very low demand for services, could you use digital to improve demand or uptake for vaccination services? [Meanwhile] at the other end of things, at the clinic, is there a way to use digital to help with the commodity stock: to make sure that you have enough vaccines at the right time, and that if you need more you can get hold of them?


Key priorities for digital intervention

WHO breaks this down into three areas: The health content, which is the health information being provided; the health interventions, which is the digital way of doing something, like a messaging service which is delivered on your phone; and then the applications themselves, which [comprises] the software or the digital platform used to deliver that information or service.


The guidelines focus specifically on that second topic, the intervention; so what are the digital functionalities or ways of doing things that should be used within health systems? And does the ICT capability or functionality that the digital tool can offer actually help the health system?


The guidelines themselves talk a lot about accessibility via mobile device. What that doesn’t mean – I’ve been asked to emphasise this – is that everything’s got to be on a phone; but working across the number of countries that we do, being accessible on a phone is the minimum requirement for a lot of content or services. You can [also] have it on a desktop and on a laptop or whatever [device] you want, but at a minimum, these things should be able to work a mobile approach.


Development and delivery

The specific objectives for the guidelines when they came down to it were to review the evidence, to work out what evidence we have, what should be recommended and what shouldn’t; [and to] develop these recommendations based on the evidence for Member States and any associated bodies or organisations.


There is the implementation consideration – it’s not just about saying in a bubble or in a vacuum ‘here are these interventions, you should be using them’; it’s also about context: When you introduce this intervention, what are some other parts of the bigger picture that you need to think about in order for it to be a success? Finally, there was a review in terms of the existing evidence and the lack of evidence in other areas, [identifying] where we have gaps and where we needed to do something about it.


This took over two years, firstly identifying the questions or retrieving the evidence that was done through nine systematic reviews with Cochrane; then a synthesis and assessment process sifting through all that evidence with 500 global experts. Through [a process of] consultations, discussion and formulation they picked out and identified the nine interventions that we now have in the guideline, [which then went through] our own internal process for approval, quality control, submission and publication.


The types of evidence that they reviewed were all standard effectiveness: not necessarily just RCTs, but also this high quality traditional evidence base through these concrete systematic reviews. We had that the implementation side, looking at factors that weren’t necessarily within the control of the intervention itself, like acceptability – again, from a patient or user perspective, that’s a fundamental – [as well as] the feasibility: if you’re looking at introducing something in a given context, is it right for that context; and [determining] what could go wrong, because it probably will, and how to account for that [at the planning stage].


The actual interventions have been focused on sexual, reproductive, maternal, newborn, adolescent and child health; covering [features] like birth notifications, death notifications, stop notifications; remote access for services; and client communication. [They also covered] the digital health record; healthcare provider decision support; communication between healthcare workers either at the same level or potentially between an initial referral and a specialist; and the training of healthcare providers themselves.


Considering context

The digital health guideline recommendations are all within the context of the health system where they’re being used. You could have the best solution in the world and if it’s not appropriate for the reality of the system that it has been put into, it’s not going to have the impact that you’re expecting.


For example, if you have an app or a messaging system [which tells you] that your clinic is about to run out of insulin, but you’ve got no way of ordering or accessing that insulin or getting it delivered in the right way to the right place at the right time, getting a text about it is not going to help your work. The guidelines look at this question around the implementation consideration: what do you need to think about beyond just the intervention, the bigger picture and the strategy?


The key messages from the guidelines for today are that, based on the purely evidential side, digital interventions have modest benefits in some areas. It’s quite limited as to what we can say that they do, but this is not because the evidence says that they don’t do it; there just isn’t the evidence to say that they do do it; so there needs to be a big push for more evidence, [considering both] quality and quantity.


We need to look at the value add of digital within a health system, going back to the investment case for why people should look at using digital versus non-digital alternatives. Thirdly, enabling the environment: what are the other [factors] that need to be there in order for intervention to work in a way that you want it to?


Citation needed

We need to grow the evidence base: a lot of that is for anyone working in digital to continue publishing, collecting and sharing, building that evidence base so that when we do get back to reviewing these guidelines, we’ve got much more information that we can use.


Finally, [we must consider] responsible use of digital health: keeping it tied to the challenges within a health system, not just [implementing digital] for fun. Ultimately [digital must] add value to the user’s experience, because they’re going to make or break whether that intervention is a success or not.


In terms of implications for the UK, a couple come to mind: first, this idea of digital as being a means to an end. It is nice to have digital, but it should always be pegged back into that system need, not just a matter of digital for digital’s sake. There’s also that question around looking at the bigger picture: these things don’t work in isolation, so what is there that can be done to support the development of the ecosystem around it, that enabling environment?


Finally, the biggest key is more evidence. The more that we know about how all of these things work, the better our recommendations to different countries can be, and that includes the UK maybe sharing some of their experiences with other Member States to show what they’ve achieved: those lived stories I think are sometimes as powerful as the client side of things.


SOURCE:Paper.li

Recent Posts

See All

Tuning using HyperOpt in python

HyperOpt provides an optimization interface that accepts an evaluation function and parameter space, and can calculate the loss function value of a point in the parameter space. The user also specifie