Queen’s Speech

Lord Freyberg Excerpts

Thursday 29th June 2017 Lords Chamber

[Lord Freyberg (CB)]

My Lords, I wish to restrict my remarks on the gracious Speech to health. I welcome the new patient protection Bill and the focus on mental health, but in both the Government are not going far enough in driving a quality agenda into the NHS, and do not seem to understand that care quality and the UK’s attractiveness as a destination to industry for clinical research post Brexit go hand in hand.

Moreover, most of the changes needed to drive quality in the NHS can be achieved without legislation, as they are about how we measure quality and encourage a culture of accountability. Accurate, disease-specific care quality data, sometimes called care outcome data, are the missing link that can drive both these agendas. Given that the life science strategy is in the Minister’s portfolio, and Sir John Bell’s committee has had extensive discussion of what it will take to make the UK attractive to pharma, I would like to draw his attention to the connection.

Care quality data are information on how a particular episode of care turned out for the patient. These outcomes have to be defined and measured in standardised ways, or all manner of games get played by clinicians and hospitals to time and choose the measurement to favour themselves. The not-for-profit International Consortium for Health Outcomes Measurement—ICHOM—has done much to define these standards through international consensus building. It now has standard sets that cover about half of disease by burden. The outcomes are rich in patient-reported outcomes that can now be collected cheaply using smartphone technology. Seven European countries, including Wales, have now committed to using those standards. That allows international comparison, which can only accelerate best practice development. However, England has not made as much progress as we should have in defining quality. What clinical audits we do have are slow to report and miss the high-frequency and important complications that cannot be measured in the electronic health record.

A good example of this is the recent maternity audit, Each Baby Counts, under Professor Lesley Regan, looking at stillbirth and major brain injury in 2015. This affected 1,136 babies that year. What is shocking is that about one-half of those deaths and brain injuries are believed by experts to be preventable. The UK is at the bottom of international league tables for maternity deaths, and has been since the Second World War. I applaud the effort that went into the audit as a first step in lifting the UK from 33rd of 35 countries on severe harm to babies. It is great, too, to see the first audit with 100% coverage of trusts, but we are going too slowly.

What can be done about this? Each Baby Counts made a number of recommendations, but I want to bring to the Minister’s attention the most important, following on from the remarks of the noble Baroness, Lady Cumberlege: the need to change team behaviours to create the right culture of safety. An audit that reports back many months after an event is unlikely to do much good, given the turnover in NHS staff. They can legitimately say, “It wasn’t us”.

Changing culture is hard, but we can learn from other industries with strong safety cultures—in particular, the airline industry, as the noble Lord, Lord Ribeiro, mentioned. Through experience, they have worked out that safety cultures that only look at the severe-impact, low-frequency events, such as a plane crash—or, in the NHS, stillbirth—fail to change the culture. The events are too rare; they can be written off too easily as “chance” by practitioners. So plane safety experts innovated to look at high-frequency risks, such as near misses, and creating a culture of learning through appropriate team-based discussion of those incidents in close to real time. Reducing near misses works—it has led to fewer plane crashes. That is what we need in the NHS, and is missing from the patient safety Bill.

We need to add these higher-frequency, smaller harms to the metrics. Almost always they are, or could be, patient-reportable outcomes. From their red book, most mothers know how blue their baby was. Mothers also know whether they got good care, and are highly engaged. We do not need costly NHS internal data entry to measure this; we can just ask mothers to give us their outcomes via their smartphone. Those data need to be used in weekly or monthly team discussions to improve care, as part of normal team management meetings. There are enough births that it is meaningful to track these higher-frequency statistics at that frequency, unlike stillbirths and brain injury. Those near real-time discussions and continuous small improvements are what changes the culture, not an audit on a major event that reads out 18 months after the event.

If we really want to half the stillbirth and major brain injury rate—or in any other area of care—we need to change the culture, and manage these near misses. To be clear, this is an “and”, not an “or”; we should still audit in depth the severe, low-frequency events such as stillbirth, just as in air safety we investigate crashes in depth. But we need to add the more common harms. To that end, would officials from the Minister’s department meet representatives of the International Consortium for Health Outcomes Measurement to be briefed on work in the UK and elsewhere and to identify areas such as maternity services, mental health and cancer, where that work can make a difference? That difference is critical. Those same disease-specific, patient-reported outcomes are the dynamite of health economics.

If, as the life sciences strategy suggests, we want the UK post Brexit to be the destination of choice for life sciences research, we need to invest in NHS data, especially care outcomes linked to episode-of-care data. Why? First, because the data that pharma are after come in flavours—some rare, some valuable and some ubiquitous. Any care system with an electronic record can supply activity data; almost none can provide reliable care outcomes beyond death and hospital admissions. Scarcity drives value.

Secondly, to pharma, payers like NICE are now as important as, or more important than, decision-makers as regulators in getting a drug to patient. Large-scale outcome data is gold dust for health technology assessment and performance-based pricing. For example, if the NHS is uncertain that, in the real world, a new $100,000-per-patient wonder drug works as well as its makers say, the NHS can demand, “We’ll pay according to real-world performance”, and measure it.

Finally, it takes 15 to 20 years or so to get a drug from bench to market. Less well known is that it then takes another 15 years to get peak clinical uptake, as care systems are so resistant to change, even for effective innovation supported by a strong sales force. Appropriate real-world outcome data will allow truly innovative drugs—

[Noble Lords]

Order.