Expert view: Tim Benson

Transparency in Outcomes’ is the first of several consultation documents to be issued following the publication of the ‘Liberating the NHS’ white paper.

That the Department of Health should choose to make an outcomes framework the subject of its first consultation indicates just how seriously the new government is taking the shift of focus from process targets to outcome measures.

Health secretary Andrew Lansley writes: “we need to recalibrate the whole of the NHS system so it focuses on what really matters to patients and carers and what we know motivates healthcare professionals – the delivery of better health outcomes.”

He goes on to say: “improvement of health outcomes for all will be the primary purpose of the NHS. This means ensuring that the accountabilities running throughout the system are squarely focused on the outcomes achieved for patients – not the processes by which they are achieved.”

The NHS Outcomes Framework will comprise a focussed set of national outcome goals, set by the secretary of state for health, which will provide an indication of the overall performance of the NHS.

It is also intended to act as a catalyst for driving up quality across all NHS services. The outcomes framework will replace the current performance regime based on targets.

The results will be published widely to provide transparency about the quality of local services and are likely to be used to support CQUIN (commissioning for quality and innovation) payments, which are also to become more important.

Principles

The consultation document offers eight principles to guide the development of the outcomes framework:

• Accountability and transparency – results will be published
• Balanced cover of effectiveness, patient experience and safety along the lines set out by Lord Darzi in High Quality Care for All
• Focused on what matters to patients and healthcare professionals
• Promoting excellence and equality
• Focused on outcomes that the NHS can influence but working in partnership with other public services where required
• Internationally comparable (including between UK countries)
• Evolving over time, with annual reviews.

One other principle that might be considered is that outcomes data should be an integral part of day-to-day patient care, not a separate exercise. This is a corollary of the third principle – that indicators should record the effectiveness of treatment from the perspective of both patients and clinicians – but it needs more emphasis.

Outcomes data transparency at the point of care could help patient-clinician communication, enable checks for accuracy and completeness, provide feedback to all stakeholders (patients, clinicians, managers and commissioners) and even update the secretary of state’s dashboard in close to real time.

Almost 30 years ago, Edith Körner advocated that data needed for accountability be based on data that was clinically useful; but her concept was not implemented as she intended and led instead to the separation of management and clinical data flows which has served us ill ever since.

Domains

The proposed Outcomes Framework is organised into five broad domains:

1. Preventing people from dying prematurely
2. Enhancing quality of life for people with long-term conditions
3. Helping people to recover from episodes of ill health or following injury
4. Ensuring people have a positive experience of care
5. Treating and caring for people in a safe environment and protecting them from avoidable harm.

For each domain three levels are proposed:

1. One or more over-arching outcome indicators that frame the NHS Commissioning Board’s broader responsibilities.
2. About five improvement areas, each with their own outcome indicators,
3. About 150 quality standards that will be developed by NICE over the next five years, setting out what high quality care should look like for each of the main pathways of care.

For the first domain, preventable death, the consultation document suggests using the number of preventable deaths in patients under the age of 75 as the overarching indicator.

It recognises that classifying all deaths into just two categories (preventable under 75 and other) is a pretty crude indicator, and for the specific improvement areas it suggests more detailed indicators such as premature mortality due to heart disease and stroke, cancer survival, infant mortality and healthy life expectancy of older people.

The second and third domains, quality of life for people with long-term conditions and recovery from episodes of illness or following injury, focus on the core of what the NHS does and represent the core of the Framework. In these domains we will see increasing use of patient-reported measures alongside clinical outcome measures such as complications and emergency re-admission rates.

The fourth domain, ensuring people have a positive experience of care, is likely to build on current patient experience measures for inpatients, emergency, outpatients and general practice.

The fifth domain, treating and caring for people in a safe environment and protecting them from avoidable harm, has the desired outcome of preventing harmful incidents. Examples of measures include “never events”, hospital-acquired infection, in-patient falls and safety culture.

PROMs

One of the most significant aspects of the proposals is the extension of patient-reported outcome measures (PROMs).

Using generic measures, which focus on how the patient feels and how much they can do, you can make comparisons across all conditions and care settings and use them when you do not know what is the matter with the patient and patients have multiple problems.

The current generation of questionnaire-based outcome measures were developed originally for use in clinical research and population health surveys, not for repeated use in routine care.

The NHS patient-reported outcome measures (PROMs) programme uses generic and condition-specific measures together for all patients having hip and knee replacement, abdominal hernia and varicose vein surgery.

Each patient completes a 16-page questionnaire booklet when they arrive at the hospital for their operation and is sent a second booklet three or six months later (depending on the condition). The scheme started in April 2009 and the first “before and after” results will be published shortly. Initial “before” measures show a completion rate of about 58%.

PROMs can be designed for routine use as part of day-to-day patient care. Such instruments need to be quick and easy to use, with electronic capture and processing. They also need to help patients, clinicians and managers to monitor progress, as well as provide case-mix adjusted analysis for commissioners and accountability.

Case-mix adjustment is essential because what is the matter with the patient is the most important single factor that determines outcome. Some conditions are much easier to treat than others.

The big picture

The ‘Transparency in Outcomes’ consultation document focuses on top-level accountability and everyone must be accountable for spending other peoples’ money.

But in the NHS, what really matters is what you do when face to face with a patient. The real power of outcomes is to focus attention on what helps patients and carers without having the decisions influenced by the way you are paid.

We cannot improve what we cannot measure, so we must measure patient outcomes. This is long overdue. Inevitably we will get what we measure, so we must be careful what we ask for.

Tim Benson is a director of Abies Ltd and Routine Health Outcomes Ltd. He is author of “Principles of Health Interoperability HL7 and SNOMED”, Springer 2010.