Maintenance metrics

In December 2023, the Estates Returns Information Collection (ERIC) data showed that the NHS needs to spend £11.6 billion to return its buildings and fixed assets to a suitable condition. David Jones, Director of Estates, Facilities and Capital Development at the University Hospital of Southampton, is studying for a PhD looking into the cause and effect of backlog maintenance and its impact on patient harm

At every conference I go to, everyone talks about the “new” – which is frustrating.

There’s a statistic that says 80% of all clinical knowledge changes over 20 years so, arguably, most of my estate is now defunct. But the rhetoric from all levels of the system is “build new”, so that’s where most of the focus – and money – goes. The majority of CDEL [capital departmental expenditure limit] is calculated using the depreciation of the buildings and IT, etc. so most of the funding that you get through CDEL should go to the maintenance of your existing buildings. Any new buildings should go through an application to NHS England, but that’s not how many Trusts do it.

There are huge pressures on the estates operational team because they’re back of house – they’re not doctors and nurses, so taking money from here won’t hit the headlines – and you can see the budgets eroding over time, revenue-wise. Then we wonder why we’re sitting on £11.6 billion pounds of backlog maintenance (which is a ‘green book’ value, excluding VAT, fees and weighted costs). But nobody is asking the most important question: what is that £11.6 billion impacting?

Cause and effect

When I started my PhD, I reviewed more than 3,000 articles with ‘patient harm’ in them and categorised them into the World Health Organization classifications for patient harm. There were four papers that mentioned patient harm and infrastructure and buildings somewhere but didn’t link cause and effect. It was almost incidental.

I started looking at the ‘Swiss cheese model’ of safety incidents. At the time, the maintenance backlog was about £9.2 billion, so I thought it must be influencing patient harm somewhere. I looked at the public data and, sure enough, 5% of all patient harm incidents are equated to estates, infrastructure and technology. However, that’s what we call ‘active errors’, where, for example, a ceiling tile fell and a patient or staff member got hurt. Then I started looking at the latent impact – for example, if you have a flood and must close beds, therefore a patient doesn’t get their operation etc. This was reinforced in 2023 when the British Medical Association published ‘Building the Future: Brick by brick’, which includes a survey of all their members. They found that 43% of doctors surveyed stated that the condition of their workplace has a negative impact on patient care.

A three-study approach

For my first study, I applied to the NHS National Patient Safety Team for three years’ worth of records. This equated to around 6 million records of patient incidents that have been recorded for all reasons, not just infrastructure. I took out all the non-acute sites and was left with 4.68 million records. Once analysed, I broke that down and I could see that 56,000 of those were down to pure infrastructure incidents, or ‘active’ harm incidents. I was able to analyse those and show which were the main areas of concern, but how is infrastructure affecting each one of those 4.6 million indirectly or as ‘latent’ harm?

I categorised the 56,000 records into 10 areas and I’m working with data analysts at Southampton University to undertake thematic analysis. We can then apply the results to the remaining 4.54 million cases to establish how many of those are infrastructure-related latent harm, which might be further back in the root cause analysis.

The second part of my study is systems dynamics modelling. If I look at an active infrastructure incident that has an impact on staff incidents – which then have an impact on motivation, staff turnover, workload, etc. – you can see all that in the systems modelling.

I’m really interested in how that then impacts finances. The greater the amount of reactive incidents, the greater the impact on Trust revenue funds and that’s a negative spiral on revenue and capital. If you take it to the nth degree, you have a patient incident and you lose patients or healthy life years, therefore the Treasury gets less money via healthy working taxpayers; the DHSC, NHS and ICBs get less; and so on. Even though it’s very marginal, there is a closed loop there. I’m currently conducting Delphi analysis of this model where experts in estates, finance, HR and patient harm will challenge my assumptions and uncover if anything has been missed, added or over-emphasised.

The final bit of my study is around capital funds, where decision-makers are deciding where money goes without having the full picture of where the money needs to go. This is termed ‘bounded rationality’. We don’t capture the full cost of reactive maintenance. For example, say a hospital loses heating in two theatres because it has aged air handling units. They can’t be replaced because the theatre would need to be taken down for a several days to do so but, every time there’s reactive maintenance, a surgery list is cancelled at short notice. There’s an impact on the patients in terms of health and there’s also a financial cost impact of every patient. Even though we’re saying we’ve got £11.6 billion worth of backlog maintenance, the true value is in the cost impact of that £11.6 billion across the NHS.

Clinical services have been using evidence-based medicine for since 1991 very successfully and they’re able to articulate that they’ve got 1,000 patients on their waiting list and they deteriorate by X if they leave them six months or they deteriorate by Y if they leave them 12 months. Within estates and facilities, we can’t do that because we don’t have that level of data and that level of cohesiveness but, for me, that’s where we need to be. We need to be able to say that, if you do not replace an air handling unit at the end of its life, then the likelihood of infection in that ward increases by X% or the impact of an electrical shutdown is Y.

What needs to happen next

I work with the NHS Estates team on backlog maintenance projects and this year the team has requested that Trusts start listing out where major infrastructure failures have impacted clinical services. This will be hugely important data. If we can get to a position where we are changing policy to capture all the cost of reactive maintenance, that will be a big step forward. We also need to be separating funds for the maintenance of the estate and new hospital developments. They can’t compete with each other.

The funds should be based on a need basis, not only for the estate but for the patient. We need to be clear on what that patient pathway is. For example, if we’re no longer offering a specific clinical service at one Trust because a neighbouring Trust have got a Centre of Excellence but they’ve got an issue with their ventilation, we need to be pragmatic from an ICB flow of capital perspective.

In addition, I would like to see a single CAFM [Computer Aided Facility Management] system for the whole of NHS England, so each Trust is submitting data in exactly the same way. We would no longer have to do an ERIC return because the NHS Estates team would have the back door to it and would just pull the data as they need it. If we’re all on the same playing field, every Trust will have to report all the down times that they have, the impact that has on their staffing and labour cost, non-pay costs for repair and replacement and the impact it has on the clinical service.

I don’t think we’ll ever get to the point where we will have the impact on the patient defined, because quite often the impact on the patient is seen days if not weeks after the incident, so trying to triangulate backwards is nearly impossible. But this would be a good start and it is so far away from where we are now.

David Jones is currently Director of Estates, Facilities and Capital Development at the University Hospital of Southampton. He has worked in senior positions across the NHS for 14 years. Prior to working for the NHS, he worked for facilities services contractors in the commercial and public sectors.

David has recently launched a Healthcare Estates Post-Grad Study Group to provide support and guidance to anyone undertaking or thinking of undertaking post-graduate study with the field of healthcare estates.

“We’re trying to get more and more people involved, because the only way we’re going to get the issues resolved is with more eyes on it, more focus and more research,” he says.

Visit David’s LinkedIn profile for updates and information on how to join.

Fortis will be following David’s research and will update again in a future issue.

This article was taken from the May 2024 edition of FORTIS magazine.

FORTIS magazine is a Health Spaces Limited publication. Opinions expressed in FORTIS magazine are not necessarily those of Health Spaces Limited or Dialogue Content Marketing Ltd. Material contained in this publication may not be reproduced, in whole or in part, without prior permission of the publishers. No responsibility can be taken on behalf of advertisements printed in the magazine.

FORTIS magazine – read more

This article was written for the November 2024 edition of FORTIS magazine; a forum for the NHS to share ideas, innovations and case studies. To read the publication in full and access digital copies, visit FORTIS magazine. FORTIS magazine is free for NHS change-makers and leaders and is available as a print or digital copy. FORTIS magazine is managed and owned by Health Spaces Ltd.

Maintenance metrics – Feature: the impact of estates and facilities