HTN Now: Apira on approaches for deploying a mental health EPR

At HTN Now, we were joined by Apira’s founding director Geoff Broome and managing director Alan Brown for a discussion on the deployment of mental health electronic patient records, with particular focus on the choice between two key approaches: ‘get it right first time’ (GIRFT) or ‘land and expand’.

Apira is a digital consultancy firm celebrating its 26th anniversary this year. They work exclusively in health across all care settings, and focus on helping organisations – mainly within the NHS – with “all elements of their system life cycles, from strategy and business cases upfront through to deployment and benefits realisation.”

Geoff and Alan shared how they would each be arguing the pros and cons of the two identified approaches to EPR deployment, although they acknowledged that it is not a binary choice and a hybrid approach is often preferred dependent upon the organisation’s individual needs and circumstances.

GIRFT versus land and expand: definitions

To begin, Geoff and Alan provided definitions for the two approaches.

Operating under the GIRFT approach, they explained, the assumption is that building a mental health EPR that meets all specialist requirements is essential to getting it right first time. This is deemed essential to gaining user acceptance at the outset. If it is not achieved, staff may implement potentially unsafe and hard-to-unpick work-arounds. This approach assumes that migration of data is essential to ensure acceptance, safety for all, and effective service efficiency. “Only in this way,” Alan suggested, “can we achieve a fully functioning system catering for all user needs and complying with regulations and statutes from day one. It’s very much a big bang approach that focuses on taking your time and getting it right first time.”

For the land and expand approach, meanwhile, the assumption is that “it is better to implement a mental health EPR with absolute minimum viable product first, and then optimise the solution over time, based on real-world user experience”. Under this approach, the EPR would only include very basic functionality at the start with standardised assessments, basic care plans and minimum data migration. The aim of this is to avoid deploying things ‘just in case’ and potentially creating something that is unusable in practice.

To gain an understanding of existing views at the start of the webinar, Geoff and Apira ran a poll to identify which of the two approaches the HTN audience preferred. It was a very close call, with 51 percent selecting GIRFT and 49 percent choosing land and expand.

Why land and expand?

Arguing in favour of the land and expand approach, Geoff said: “The clinicians that I’ve worked with for over 30 years tend to be very experiential. They like to have empirical evidence to lean on. I’ve found that once clinicians have got experience of the system, they have a clearer view as to what they want, but also what they are realistically going to use in very busy clinical environments. It’s better suited to them, as opposed to sitting in workshops for weeks with no live system, designing assessments and care plans and approaches to different pathways.”

Another factor to consider, Geoff noted, is that “clinical practice is a moving target. For example, how we assess depression or anxiety in a mental health context and how we measure severity of that condition is something that is under constant debate, with ongoing research. With that in mind, is there really any such thing as getting it right first time?”

Geoff commented that land and expand can offer “a lower project cost overall, because if you’re trying to get something right first time and perfect before you start using it, you can haemorrhage a lot of costs, especially when you think about how you’ll have a large project team that needs to be retained.”

In addition, he commented that EPR projects can often draw attention from people who are keen on technology and data use, “and that’s not a bad thing, but those people might not be representative of all the people working in the clinical fraternity.” Geoff highlighted the risk for the tech-savvy individuals to engage and work on a ‘perfect system’ which can then pose a challenge for everyone else.

Geoff went on to share some more arguments in favour of land and expand, noting that if you select an EPR and get going with it sooner rather than waiting to try and perfect everything, you can maximise the benefits earlier. He highlighted that the danger to the GIRFT approach is that you can “end up replicating everything we had in the old system without challenging whether we really need it”. With land and expand, you “get to be sure that staff want the change, rather than it being something we are pushing at them”. Finally, Geoff stated, “Any use of a system comes at the cost of staff time. With a lot of EPRs, it’s a totally legitimate use of time – we need the systems to be used and used well. But every single decision to use a certain feature comes at a cost, with staff being taken away from patients in order to work on this. We need to be really clear about whether the features involved are wanted on the frontline.”


Moving onto his arguments in favour of the GIRFT approach, Alan emphasised the focus on getting it right rather than getting it done quickly. “It’s not a rush to get the EPR in, it’s a rush to get it in right – doing the right things in the right way and engaging with a lot of people to make sure it’s a big event for the organisation. We don’t want it to be something that trickles in, we want people to be interested in it and focus on change transformation to realise those benefits. I think that’s key with GIRFT – it gives the EPR the profile it needs and deserves.”

Quite often, Alan pointed out, the journey depends on an organisation’s starting point, not just their destination. “If you’ve already got an EPR or a range of systems, you can’t afford to take a step back. What you go live with can’t be less than what you have now. It might do things differently, and you definitely want to improve processes along the way rather than repeating old processes that don’t work. But your minimum viable ladder could be quite far back from where you are at the time that you’re wanting to implement the EPR.”

Next, Alan moved on to the question of financing. Money is committed to the project upfront and can’t be removed, he noted; with land and expand, you risk a “land and strand” approach if later phases end up getting cut. Also, he said, there is an argument for getting the project done before something else comes along and takes funding; Alan used the COVID-19 pandemic as an example of this. He also emphasised: “You want to align your spending with external funding, which at the moment is frontline digitisation. Putting all of the effort upfront certainly allows that alignment.”

With regards to benefits, Alan said, “Although you will start to realise benefits a bit earlier with land and expand, you won’t realise all of them until later on. With GIRFT, although it takes a bit longer to get to the realisation point, overall you’re getting the full range of benefits realised sooner.”

He picked up on the advancements in technology and supplier experience, noting that “EPRs are getting closer to being usable out of the box, which makes it easier to get it right first time with some optimisation and customisation prior to go-live”.

GIRFT can also be cheaper in some ways, Alan added. “You’ve got less resources focused on the governance – you’re focusing very much on the change and the transformation. You don’t want to have to extend your management programme, you want to get it done, optimised, and finished.”

Looking at workforce, Alan commented that the longer it takes to complete the project, the higher the risk of losing team members in later phases.

Alan summarised his other arguments in favour of GIRFT, commenting that as people tend to dislike change, it’s better to change more in one phase than to draw it out. In addition, he said, it’s better to do it once and get it right rather than having to revisit elements, pick things apart and change them. He also noted that integration is more difficult in an environment where things keep changing, and concluded: “The best way to achieve a fully functioning system which caters for all user needs and complies with regulations and statutes from day one is to put effort into making sure you’ve got it right from the start.”

After answering some questions and comments from the HTN audience, Geoff and Alan posed their question again to see if their debate had changed any views. This time, land and expand took the lead with 65 percent of the vote, whilst 35 percent said they preferred GIRFT.

Concluding the session, Alan said: “This is what we do at Apira. We help trusts decide what they want to do, bringing experience from our other customers to the project, and we help them to buy a system and deploy it. As you’ve heard, there isn’t actually a right answer to the question of ‘land and expand or GIRFT’ – we work with you to find the right answer specifically for your organisation. It’s what works best for you that counts.”

Many thanks to Geoff and Alan for joining us.