The NHS has seen a real shift in focus towards outcomes. But do we really know what the term means? Dr Rupert Dunbar-Rees, founder and CEO of Outcomes Based Healthcare, talks to Highland Marketing’s Matthew D’Arcy about disentangling outputs to define outcomes.
Look at nearly any NHS strategy or business case and you will almost certainly find the term ‘outcomes’ littered throughout. For Dr Rupert Dunbar-Rees, the need for better outcomes has become widely accepted in the health service, particularly in the last three to four years.
But in the shift away from asking ‘why’ the NHS needs an outcomes focus, the question of ‘how’ it will get there still presents a big problem.
“Outcomes are increasingly accepted as almost the only way forward to creating a sustainable health and care system,” says Dunbar-Rees
“But the very term ‘outcomes’ is one of the most misused terms in the NHS. There is a big danger this has become quite significantly diluted.”
The founder and CEO of Outcomes Based Healthcare, a data specialist organisation focussed on measuring and improving “outcomes which matter to people”, Dunbar-Rees has held a varied career in healthcare. A former GP, he has engaged in everything from delivering frontline care through to academic interests, and has held senior positions, including clinical lead on a £1.25bn NHS procurement programme in the Department of Health (DH).
He sees the currently fashionable eagerness of healthcare organisations to align themselves with outcomes as anything but negative – rather this shows that buy-in now exists for real change. But there is much to do before outcomes which actually matter to individuals sit at the core of people’s experience across health and care.
“I look around at some of the metrics that are out there now, being referred to as ‘outcomes’, and they are pretty much the same key performance indicators (KPIs) that we had when I was working in the DH 10 years ago,” he says.
“The same KPIs we used then have often been recycled, or re-hashed, and called ‘outcomes’. It doesn’t make them outcomes, just because you call them outcomes. There is an outcomes literacy issue to address. We need to move on from liberally scattering the word ‘outcomes’ in any policy document, business case, or funding application. More specifically, we need to disentangle outputs from outcomes.”
Cutting across organisational boundaries
Organisational measures of quality, more traditionally seen in the NHS, may be part of the problem. “We have measured quality for decades by looking at individual care settings or processes in isolation – measuring the so-called ‘outcomes’ of general practice, of A&E, of a department, or of a doctor, a team, or even a speciality. But rarely have outcomes for the individual, people, or populations been the focus,” says Dunbar-Rees. There is a suggestion that all of this equates to measuring outputs rather than real outcomes.
“The only way we can view the real quality of outcomes, is through the individual person receiving care – from their perspective. That cuts across all organisational boundaries. If we want to measure quality of life, there isn’t a single care setting that will deliver that on its own in isolation. We have to push the boundaries of what we have spent decades getting used to, in order to measure and improve ‘true’ outcomes.”
Incentivising best clinical practice and what is best for patients, must now sit at the heart of outcomes, he argues. But too often current contracting and payment models can inhibit this in reality. “A provider gets paid to treat someone with a heart attack or stroke,” he says as an example. “If best clinical practice succeeds, then let’s say you manage to halve the number of cardiac arrests and strokes coming into the hospital. Someone, at some point, will ask ‘what happened to half of our income?’ Providers who really do configure themselves for better outcomes are financially penalised. That cannot be right.”
Political motivation to restructure health and social care for reasons of sustainability could however hold the part of the answer.
“Whatever you call it – new models of care, vanguards, sustainability and transformation plans – everyone is trying to achieve the same ends: a more personalised and responsive healthcare system around groups of people and their needs,” says Dunbar-Rees.
“These programmes specifically have instructions to challenge the organisational status quo, the contractual status quo, and the way that quality is measured – they have been given an explicit purpose to measure outcomes as well as quality processes, and that is incredibly helpful.
“It gives people permission to challenge vested interests in the status quo, and gives a signal from the centre that it is ok for something to cause challenges in the short term for a much bigger prize in the long term.
“Ultimately we need a more sustainable healthcare system with better outcomes and lower cost.”
Identifying entirely new diabetes sub-types
Developing metrics that span across organisational boundaries can be enormously powerful, says Dunbar-Rees. Of the numerous initiatives his organisation is currently engaged in, one is focussed on better understanding the specific needs of people with diabetes, to determine much more accurately the risks faced and treatments needed by individuals within the cohort.
“If 100 people have to take a tablet to avoid one person developing a complication – then that is not good,” he says. “It is not cost effective, and people are unnecessarily exposed to potentially harmful side effects, not knowing if they would be the one to benefit.”
In his days as a GP, Dunbar-Rees would advise his patients with diabetes to carry on taking medicine because it was understood it would do them good.
“Now, if we look at the data, that is not necessarily always true,” he says. “We know you are in a higher risk cohort, but we don’t know if a particular tablet will help you or not.”
Innovate UK-funded research being carried out by Outcomes Based Healthcare, in collaboration with Big Data Partnership, could however soon have huge implications for more targeted and personalised medicine for people with diabetes.
“By using novel data science techniques and looking at cohort characteristics like income and deprivation, and then applying that to health data, and genetic information, we are starting to see there are potential sub-types of type 2 diabetes that are not currently recognised,” says Dunbar-Rees.
“This could provide a much more detailed understanding of who will develop a complication and when, as well as who is not going to develop complications.
“Diabetes, until now, has meant largely uniform treatment. But it isn’t just the two types we thought existed; there are potentially many sub-types. They are not only defined by health characteristics or genetics, but on life characteristics.”
Defining outcomes for different people
The project aims to help predict a whole range of complications for people with diabetes, meaning targeted prevention of heart attacks, strokes, eye disease, kidney disease and limb amputations, based on an understanding of specific individuals at risk.
It is just one example of what might be achieved by examining the experiences of patients far beyond single care settings, and understanding where common issues exist, and where they do not.
“As a sector, we are just starting to get to grips with that now,” says Dunbar-Rees, adding that health and care could learn from other sectors that really understand the similarities and differences between their service users, and where detailed information on customer preferences is providing for powerful targeted marketing.
“Retail in particular has got to a very sophisticated level of functioning in understanding their customer, what their common needs are, and how they differ between particular customer groups. If we apply that to healthcare, we can really start to get under the skin of not providing a uniform service, irrespective of needs. It needs to flex and alter to different groups of patients according to those needs.”
The ultimate aim: for a health and care system organised around populations whose sets of needs are similar, and where the outcomes that matter are similar. Understanding outcomes that matter to individual people is the beginning.