The data dilemma: does data help or hinder patient care in mental health?

Matthew Taylor

Hello and welcome to Health on the Line, back after our summer recess. It's great to be talking with you again. Over the coming weeks you'll hear an eclectic mix of NHS leaders and experts on health from outside the system, the kind of people we hope you've come to expect from this podcast. We're going to be hearing from Lord Markham, the health minister, from Chris Ham, who's written a report for us, and the Health Foundation on system improvement. And from the chair of our Primary Care Network, Aruna Garcia. 

But for now, though, it's time to hand over to the chief executive of the Confed’s Mental Health Network, Sean Duggan. And Sean's had a great conversation, which I know you're going to enjoy listening to with Geraldine Strathdee. She's the independent chair of the government's recent Rapid Review into mental health inpatient data.

Sean Duggan

Hello, everybody. I'm Sean Duggan, I'm the chief exec of the mental health network for the NHS Confederation. And a very warm welcome to this latest edition of the podcast, Health on the Line. 

There's a really important area around quality in mental health that has an impact across the system. This edition is about that and I'm really delighted to welcome Dr Geraldine Strathdee, who is a psychiatrist, and she was asked, a while ago, to do a rapid review into data in mental health inpatient settings, and that review has been completed. And Geraldine Strathdee is with us here on the podcast. Hello, Geraldine. 

Geraldine Strathdee

Good morning, Sean. And first of all, can I say thank you so much for inviting me to take part in this podcast and also a good morning to those listening.

The subject I'm going to talk about is of probably equal relevance to people working in acute and primary care as it is to people in mental health trusts, and certainly of really significant relevance to integrated care systems and their future planning.

Sean Duggan

That’s fantastic, Geraldine, and I'm so pleased you mention that because this is a podcast going out to the entire membership, and mental health is a clinical conditions spanning all the aspects of it.

So it has relevance to everyone and those of you working in mental health more particularly more relevance. And I know that a lot of you were involved in the work and the consultation with Geraldine during the preparatory stage of the review. 

Anyway, before we go into the details, Geraldine, tell us just a little bit about yourself.

What took you into psychiatry in the first place? And I know that you've been a real advocate of improving data and mental health for many years, but not everybody knows that. And why this special interest in data? 

Geraldine Strathdee

Okay. Well, first of all, I can start off with why did I decide to specialise in mental health? 

And the story really started with where I grew up. I grew up in Northern Ireland and I grew up in a community where there was a lot of deprivation. There was a lot of difficulties in accessing education and employment and equal access to resources. 

So at the small school I went to, nobody had done science before in my year. For some strange serendipitous reason, seven young women like myself decided that we wanted to do medicine and the school hired in the science teacher for the first time.

And I had seen - my role model really were the local GPs - who I'd seen as highly respected members of the community who probably fulfilled as much of a social stabilising function as a medicine function. So I went to university in Queen’s University Belfast, and was determined to be a GP and then I had the good fortune to be allocated a psychiatric placements as one of my very first medical student placements, and the first person I met was a woman who changed my life.

So I arrived in the large institutions. I'd been told it would be very scary. I mustn't talk to this woman, I mustn't close the door. She was very violent and I sat with her and respectfully closed the door as much as I felt was safe, but also offered her privacy. And she told me about her life, the symptoms she had, the terrifying, distressing paranoia, the hallucinations that she was having.

But she also told me how it started for her. And it started with years and years of systematic physical sexual abuse by her father and other members of her family. And by the end of my hour and a half with her, I realised that the pain, the suffering that there is in mental ill health was as great as and sometimes greater than the pain and the suffering that happens to people who have significant physical ill health.

So that sent my path into believing that I was meant and wanted to become a mental health clinician. So that's the start of my journey. 

Sean Duggan

Thank you, Geraldine. Actually it’s not too dissimilar to the reason why I went to the psychiatric nursing, that's fantastic to hear that. 

Geraldine, so let's get on to the review. And it really started because government had concerns like data - to identifying risks and problems of quality issues, patient safety issues - was not available, particularly around inpatient settings. 

And there've been a number of cases where the quality of inpatient services came to the attention and they asked you to conduct a review. 

As always, with any review, it's about terms of reference and it's about focusing in. And so you look at quality and risks in mental health is very broad.

How did you approach it? How did you define it in the end, and what did you particularly look at? 

Geraldine Strathdee

So I think people can read the report, which sets out it in beautiful, legal and other language exactly what the terms of reference were. But I think just to paraphrase, ministers were concerned by a growing challenge in poor care, in places where they felt care - safety of care and the effectiveness of care - was being compromised. And their questions were, well, is it the problem that there's just no information? No data on mental health? Is that why we can't identify and prevent safety incidents and episodes and challenges in care happening? 

Or is the issue that there is data but it's not actually measuring the focus on the person and their pathway? Is it that you've got data, but it's just not data that can be used as an early alert because obviously the most important thing, the most beneficial thing, is to be able to identify early where safety incidents are going to occur and then step in to prevent them.

The minister are also asking with, I suppose, a focus on thinking about prevention. Can digital, which has been helpful in other areas in healthcare, can digital be helpful in this context? Can it improve both the collection, the analysis and the use of information in order to drive safe care, quality care and patients’ outcomes. 

And of all the data that is gathered, does any of it gather information about the experience and the outcomes of patients and the experience and the involvement of families? 

And how do leaders in the mental health feel? What are they thinking would make a difference? And be able to answer some of those questions. 

So those are the very, I suppose, clear things that ministers pose to us. I think it's important to say that they weren't just asking us to look at mental health NHS trust information. They asked us to review the information that was available about mental health care, particularly on the inpatient pathway that was available in all NHS commissioned care

So in both the NHS provided sector and the independent sector and a few third sector providers. They also in full recognition and probably relevant to your audience. Absolutely aware that good care in mental health inpatient services is very dependent also on what has taken the person into hospital, what has gone before, the contact the person has had with their primary care team, with mental health services themselves, with substance misuse and addiction services, and also very clear that in order to correctly assess the numbers of deaths and the lessons that could be learned for prevention, it was very important to understand the interaction between the information held by the mental health provider and that of the local acute trust - where often in the context of a safety incident - patients of transferred. 

So we set out by doing hopefully the sensible things. We set up a set of guiding principles which were that we would absolutely want to focus on the patient, the information for the patients and for families. That we would do what we described as a systems data value chain methodology, by which we mean that we would take evidence from all the people who contribute to services and safety.

We had 350 stakeholder groups come to us. We set up a very specific group around patients and their families on the information that they wanted and would find helpful and the extent of their involvement in safety. And that was led by Mark Winstanley in Rethink. 

We went out to the fields and we selected organisations in each region. So, in only eight weeks, so of these representative organisations, we chose them for a variety of variables. And we decided that we would do the data value chain method. 

So that meant we went and we asked the patients and the families and then the health care assistants, the primary nurses, the ward managers, the director for that service, right up to board members, the head of the quality committee and integrated care system local partners, as well as chief execs and chairs. 

And the questions we asked about the data on the information was What information do you currently provide to national bodies, regional bodies, local bodies, your own trust?

How much of that information do you get back in a form that means it has real utility to help you manage your caseload and manage patient outcomes in care? How much of it comes back that helps inform your care plan plans? How much of it is gathered using digital means and is that helpful? 

And I think Sean as you say, anyone who knows me knows that my entire focus is to identify challenges, but also to identify solutions or potential solutions.

So the next question we asked people was, are you doing things that you were actually proud of that you want to tell us about? What do you feel you've done that has actually really transformed the information you have, which means you could provide much safer early alert services? 

We also, in the background, had an absolutely superb Department of Health dynamic team that established the major stakeholder groups. We brought together academics, multi-professional experts from all parts of the sector. We brought together the policy makers who have influence of the mental health sector, including, for example, CQC, the regulator, NHS England, etc. And we commissioned a very specific informatics expert, probably, in my view, the best in the country, to bring together that benchmarking of all the different multiple data sources and data organisations that are gathering information from people and to put that into one format.

And we also, in terms of the methodologies we were using, were very, very mindful of some of the solutions that other sectors have had, and put in place to improve safety. So, for example, the airline industry, the nuclear safety industry, where they very much take that pathway value, data value, information value, process value approach to take account of human factors.

In other words, what our staff are being asked to do and is it feasible and how could we change that to improve safety? Thanks to you and others, we had numerous interviews from every part of the system; mental health integrated care system chairs, chief execs, non-executives. We had amazing inputs, both qualitative and quantitative information from the charity, the third sector patient voice organisations and family voice organisations.

So in eight weeks, I think we were probably two days beyond eight weeks, of the time given, we managed to kind of pull together the report, the short report that you've seen and has been published and is just now awaiting government response. 

Sean Duggan

That's great. Thank you very much, Geraldine, for that background. 

And I know, you know, you were saying about staff - explore what staff have been asked to do - one of the things that you were very concerned about is that you saw numerous data streams that people were obliged to complete and to fill in forms or what have you. It was really quite extraordinary. And then when you match that with what people get back out from it in the inpatient units, the managers, the staff and the carers, there was a problem there, and you wanted to address that in the review. 

If we come on to findings and recommendations, if you could feed that into it, Did you do you feel that you did address that particular issue? Because it’s very important for our listeners, particularly in mental health, that are struggling with all the data that they're supposed to fill in and complete. And is it really meaningful at the end? And I know that this will only be addressed with the implementation of your recommendations. 

But Geraldine, tell us about the findings and recommendations.

Geraldine Strathdee

Okay Sean, well, thank you all for that entree. Really important. So if I can just quickly summarise almost key words for each of the sectors, the main sectors that we interviewed, or have got evidence from, what their concerns were. So I think I would like to start as always with the experience and the care as perceived by patients and their families.

So what patients and families said is: “We don't get enough basic accessible information. What is my challenge? How did I get to have this problem? What treatments of all type are going to be available to me? What are the likely outcomes, the best outcomes I could achieve? And how can my family and my support system be involved to help?”

You know, it's a modern world. It's a modern digital world. Lots of organisations make YouTube videos so that it's not so frightening when you have to do something like become an inpatient or attend a kind of a new assessment appointments. So that was one of their key things. 

And just to say that for every single thing that we find, we also find examples of good practice.

So there are organisations that are producing live and YouTube to help reduce the fear and the concern and to make it obvious to people what's going to happen to them. A video saying this is what an acute inpatient ward is like. This is an example of a therapeutic care plan timetable there. Here’s your primary nurse and you have a primary nurse saying: “This is what our time together will be.” A medic saying: “And this is what the ward round will look like, and here are all the kinds of decisions we make.”

And then inputs from people who've been on a ward saying: “And this is what I learned about myself on how to manage my condition, and these are the outcomes I've got.” 

So the big ask was this accessible information much more enabled by having a digital world. Could that be on every single website and for every service that people are likely to access, and particularly for inpatient services?

The second thing, that big key word is: safe care is therapeutic care. So we have a wonderful evidence base in mental health which addresses the fact that outcomes for patients are best if they get assessed needs for their social situation, medication, if that's necessary, psychological therapy, if that's important, physical healthcare, really important. Therapeutic care means not just care and containment, and risk management, but actually receiving the evidence-based effective treatments that truly improve outcomes.

So a huge plea from family members saying: “My daughter was admitted and I can't see that anything is happening except that she's been contained. She seems to be getting medication, but I don't recognise her as the spirit, a dynamic person that went into the service”. And people wanted to have their families more involved, the families and their peer supporters.

So those were kind of the really big headlines on that. And compassion and kindness as part of the culture was felt to be a major safety intervention. So that was patients and families. 

If I then move on to staff, I would be honest. I had expected that people would say they were having to gather a lot of data, but I think we were all absolutely stunned at the volume of the data.

So people fed back from business intelligence teams and others inputting data into 72 different data sets, and that number is growing. One, business intelligence leader described it as the ‘Wild West’. So national bodies asked for data. Regional bodies asked for data. The new and really great provider and specialist collaborative ask for data. And that increasingly ICSs are asking for data, and social care asks for data.

And what people told us, the frontline clinicians and I think this is a really big takeaway message is that they spend an hour doing an assessment with somebody trying to settle the person into the ward, make them feel comfortable and make them feel safe, and then they spend anything from one to two hours entering data about that consultation.

And it's unacceptable. Most workforce said: “I didn't come to work to spend half of my time entering data. But that's what's happening. And what makes it worse is that we don't get it back. We don't get it back in a form that means it helps us to manage our daily work, it helps us to understand more about the patient, it gives us useful clinical information that can then be used to inform the therapies we want to be able to deliver.” 

All members of organisations said; Paraphrase one chief exec: “We know that very many organisations gather data about our organisation, but we don't get almost any of it back in a timely way. Sometimes the data is six months to 24 months old by the time we get it back, and that does not help us understand where we're at. It doesn't help us drive prevention.” 

Non-executive directors saying, “We get board papers, they can be anything from 400 to - some described - 800 pages long. And what we need to have on that is a clear statement of what safety factors we should be looking at, including things like closed culture, acuity factors and others.”

“We get raw data presented to us sometimes, and what we want is answered the following questions as accountable board members. For example, safety incidents, how many are we having? Do we understand the underlying causes of that? Is it workforce? Is it training? Is it closed culture? What is it? What do we need to do about it? How do we benchmark with other organisations? Because we can't get that - including for things like deaths - and what improvement methods will be the most effective? That's what we need and we don't get that in the main with some notable exceptions.” 

Integrated care system mental health leads wonderfully said actually we would really value if there was a bit of a rewrite about data literacy training and programmes.

We know that most organisations - and this was reiterated by chairs and boards – most organisations have really a lot of financial training or financial expertise to help make board-level decisions and commissioning decisions. But we would like to see - we believe that we will only get the safer services when safety and quality and efficiency data - clinical data - is equally available - both in the data and the expertise - as it is for financial data.

So those were many of the kind of common themes that came of a major burden, but were not measuring what matters. And what matters is: Who is the person? Let us help them understand how they get to be here in the condition. Are they getting the therapeutic interventions? And also, many people said the most important measurements are: What is the patient experience and the patient outcomes?

And most people said, “We can't get that.” But again, I can tell you what some of the solutions - really innovative solutions - people were coming up with that we were able to find. 

Sean Duggan

That's great. It's interesting, it's a fairly complex area this, and it's very broad and when it comes down to it, patients and carers saying it's about compassion and kindness, and that has a bearing on the safety within the wards and the community. And we know that, but it's just so important to be reminded of it.

And then the feedback around data literacy, what we need is training around data literacy. So quite often, you know, with these complex, wide-ranging reviews, it does boil down to some getting the basics right as well, doesn't it? And you've made that quite clear. 

So, we've talked about background, process. It's very comprehensive some of the areas you uncovered. 

Just say a bit more about the recommendations and maybe just focus in on a couple of recommendations that you'd say would be a real priority for you. And, bear in mind that we're looking at a mental health issue here, but that spans - as we were saying earlier - ISCs, primary care and acute - just focus on a couple of really important recommendations that we as an audience listening to this need to take away. 

Geraldine Strathdee

So I think the really important recommendations are the one about, I think, a general understanding that the data does not arise de novo. Somebody has to enter the data and the opportunity cost of a high volume of data being entered is less time for therapeutic care by clinical staff.

I was genuinely - I have to say, personally - surprised at how many people said: “But doesn't the data come from the business intelligence teams?” It does. But as the business intelligence teams say, they have to support train-change datasets to get clinicians to enter the data in the main. 

So I think just that pervasive understanding of: Data is not entered free. It comes at a cost. And that cost - especially in a workforce constrained environment - it comes at the cost of spending time with the patient and setting up therapeutic services, often. So that's kind of one key message. I think the second kind of key message going forward really is that for almost every challenge that we identified in the report from the 350 stakeholders, there was an innovative solution. A really good solution being found, because the other pervasive, I suppose, message is there is incredible variation.

So people told us - patients, families, staff, boards - about the real variation - across the 24 hours - in a service. Variation in quality health and safety between day and night, between similar teams, between teams crossing over into the acute, just major variation, that in the main external partners find hard to understand why if you have really good practice, it can't be disseminated across. It's asking people to kind of rethink that. 

The other thing that I believe could actually be transformed and we talked about it in the recommendations is the issue about deaths. People will say we learn lessons. That's how we develop our strategies based on learning lessons. But if you don't know the number and the type and the cause of the deaths in your organisation and you don't have access to national benchmarking information, that makes it much more challenging.

And again, there are many organisations and a number of organisations gathering information about death; death by suicide, death by physical health, death by, I think, what I would call therapeutic negligence. And I say that word advisedly because I think what is often not understood is the major harm that is done. The major lack of safety caused if the full effect of therapeutic interventions are not available, and there are organisations that manage that.

The other important factors are about the staff. We know that where staff don't seem able to speak up and have their voice heard and speak out - if the standards on a unit are not what they would want for themselves or their family - that then leads to compromised safety. And I was inspired by some of the chairs that we spoke to from your NHS Confederation networks, who told us that in their view, they wanted to walk the floor, they wanted to always make themselves available to staff who could speak up.

And I think one of the key messages was: by the time people have to whistle blow, a very great deal of harm has already been done and therefore encouraging boards and leaders to think about can we use the evidence we've got to prevent harm earlier? 

Sean Duggan

Great Geraldine, Thank you so much. And the reviews that you did spans and connects with other reviews and other investigations going on at the moment and it's important they're connected together, isn't it? For the ICSs to have a responsibility to ensure things get done and also for the providers that have to take on board and implement the recommendations.

We've got the Health Service Safety Investigation Branch start their review into inpatient services, mental health inpatient services next month. So that's upon us. We've got NHS England looking at models of care, so to ensure that we improve quality going forward, and that's alongside all the stuff that we're doing about transforming mental health and making it contemporary going forward.

And I know you did connect with all this. Are you confident that these reviews are linked to get together? Because I know that will be a concern for some of our listeners. 

Geraldine Strathdee

Yeah, I suppose what I would say is, just as there are very many diverse information and data collections, I think there are many synergistic, but also at times separate initiatives.

And I think it probably behoves the field to look at are we addressing the fundamental basics of care? In all the initiatives that are happening, are we learning from the good that is happening there at the moment as well as the kind of more challenging? As I've said, we identified good practice or innovative practice for every area for which it was a challenge.

And let me give it to you. I'll talk about integrated care systems. I think there has been some really inspiring work by a couple of integrated care systems with partners in looking at the data that they're collecting, in looking at what constitutes good care, co-designed with patients, and then identifying what data could be reduced. 

So I’m giving you one area, a London ICS, where they managed to reduce from 6,500 items of KPI a year right down to below 3,000 because of the wonderful spirit of collaboration and focus on patient experience and outcomes, and staff being freed up to do quality improvements.

That's what they focused on. And so they managed to make at least a 50 per cent reduction. Very practical, very systematic, done in partnership and collaboration, 50 per cent reduction in the volume of data been asked for. 

The other area is; there are so many different datasets, but they're not connected. So when a board member wants to know what are the causes? Why have we got this level of acuity or safety incidents on a ward? 

There are organisations that have found a way to connect the different data sets that can bring together in one digital platform the level of security on a ward, the level of staffing on that ward, the level of difficult challenges on that ward. They brought together the different data and information. Gathered once, reported in one system in order to be able to set up early alert system. 

You know, we can create organisations, so have 45 wards across the country and on a regular daily basis we can see where exactly we believe the challenges are going to come, and we can therefore put in more support staff, different kind of form of management. 

I’m also struck by whether people like or don't like digital artificial intelligence and some of those more kind of modern aides to support good practice, organisations that have put together in one platform all the information; patient-reported experience, the patient outcome measurement, staff views on outcomes, workforce metrics, mental health metrics, nursing observation metrics. All in the one place, so that actually there is a proper comprehensive picture of what's happening in the organisation.

Inspiring that people in two organisations at least have felt that this was so important that they brought together the expertise to start to progress this much more efficient and direct use of information for the prevention of harm and the improvement of outcomes. 

Sean Duggan

Actually, I identify with that, when you talk about bringing in the common themes. If there are several datasets, try to put it all together, which is what your report and recommendations talk about. 

It's the same when you're leading a provider organisation or an ICSs and you've got all of these reviews going on, we know that they all have common themes. So we know the sort of things that we're going to be seeing from the reviews, which is about better co-production, it's about looking at the culture. You know, we've talked about compassion and kindness earlier, and staff training. 

There are common themes. And I do think when you're faced with a number of these going on, it is about as an organisation, just they're going to happen anyway, so you might as well just pull the common themes together and transform and improve your services to address those common themes rather than get too caught up with the fact there are too many reviews that conflict each other. 

Where there is conflict then obviously we need to try and fathom that out and resolve it. 

Geraldine that’s fantastic. As the last point, and it's the obvious point: What now? I know that you wouldn't conduct a review and then sit back, leave it to others. You'll be involved in it.

What do you want to see? What do the listeners need to do apart from look at the recommendations to do their bit to implement them. But how are you going to be involved and what do you think should be the next steps? 

Geraldine

Well, this is the minister's commissioned review and I was the independent chair that brought together the methodologies and with the help of all the leaders in the field, thought about what we would find and what the solutions would be.

I think what is wonderful is that we made 12 recommendations. The 13th, which I don't regard as a recommendation, is that going forward, the progress and the implementation of the recommendations will be chaired and overseen by a minister. And that shows, I think, how extremely seriously people are taking this. So I think that's terrific. 

The consultation, you know, response from government to the recommendations is being taken very, very seriously, including that whole issue about looking at synergistically what is going on.

I would urge people to think not just about, well, there's all these reviews going on. What is government going to do? Actually many of the solutions can take place within organisations learning from each other. One of the biggest responses that we've had even to the early publication of the report is: “Can you tell us immediately where these other things, these good things are happening? Because we would like to go and visit, we'd like to see it - it may not be the right thing for us, but - we really, really do want to learn.” 

One of the other key messages that came through is there is a lot of good stuff happening out there. And as you say, we know what many of the themes will be. Can we bring the good stuff together to be shared in a very accessible, open access site that can be shared with patients and families as well as with partners across the system? 

There are already, I would say, many initiatives happening. So I'd been to four major national events to circulate - at the request of the organisers - just to circulate what's going on.

This review has influenced and reinforced, I hope, many of the positive initiatives that are already happening, which are perhaps not as widely communicated or circulated as the fact that there's a lot of reviews going on. 

Sean Duggan

Excellent. Yes, I would agree with that. So, Dr Geraldine Strathdee thank you very much for telling us all about the review from start to finish.

And we've got a role in the NHS Confederation to do a lot of what you said really. It’s about sharing good practice, pulling the themes together and we do our bit within the reviews I guess to sort of try and collect all the recommendations together to make it easier for organisations going forward. Yes. So finally, thank you very much, Geraldine.

Geraldine Strathdee

Thank you very much indeed, Sean. And I hope that your members, many of whom have the solutions in their organisations and in their heads, that they can also come together. And maybe you will hold your own review of responses and practical ways in which improvements in safety could take place.