Feature / The value of ‘no’

Last month the Audit Commission published a briefing that looked into how much the NHS was spending on treatments that are either clinically ineffective or not cost-effective. As expected, it received much press attention and has stimulated debate about which treatments are low clinical value and, to a lesser extent, whether accountants should be involved in deciding which are provided.

When money is tight it is unavoidable that discussions will take place about what NHS money is spent on and whether it could be spent better. The NHS atlas of variation in healthcare, published in November 2010, was aimed at helping to remove unwarranted variation to increase value and improve quality.

The atlas refers to low clinical value treatments and highlights two issues. First, some patients are receiving treatments that some clinicians would consider unnecessary and of no added value. Second, there is an opportunity cost to providing low clinical value treatments. Commissioners could spend money better, for example, on other types of treatment, either for the people with the same condition or to meet unmet needs in another group of patients.

It is difficult to argue against this, but reducing spending on these treatments is not straightforward. At the moment, there is not a single source of evidence that commissioners can use. Organisations such as the National Institute for Health and Clinical Excellence (NICE) have developed some information, including NHS Evidence and their ‘do not do’ recommendations.

But because there is no national list of low value treatments, commissioners have developed their own approaches locally. This has resulted in duplicated effort and inconsistency across the country. It is hard to explain to patients why the same treatments are appropriate in one part of the country, but not in another.

For some of the low clinical value treatments, there is evidence that they are not clinically effective; for others, there is a lack of evidence of their clinical effectiveness. Certain treatments can be effective for particular patients, but not for others. Therefore some PCTs have grouped their lists of ineffective treatments into the following:

Those they refuse to pay for any longer

Those where patients must satisfy specific thresholds, so only the right patient gets the right treatment.

Approach to the report

We were clear in the briefing that we are not medical experts. The Audit Commission does not employ clinicians and we do not have a view about which treatments are of low clinical value and which are not.

However, we do know that treatments do vary in effectiveness and there is limited money available to the NHS and that money needs to be spent well.

It is appropriate for auditors, and NHS accountants, to consider whether spending money on treatments provides value for money and, as part of that, identify how much is spent on these treatments.

As with commissioners, our first challenge was to identify what should be on the list. We reviewed some of the low value lists prepared by PCTs and chose the list developed by Croydon PCT in 2005. We have no view on the clinical effectiveness of the items on the list, but we selected it because it is used by other PCTs in London and has widespread acceptance among commissioners. Also PCTs elsewhere have used the Croydon list as a starting point and adapted it to suit local circumstances. The 34 treatments on the Croydon list fall into five categories (see box previous page).

Once we had decided on the Croydon list, we calculated the total amount spent on the treatments by looking at activity levels and the average tariff for each one. We estimate that in 2009/10, the total spend in England was £1.9bn. Of course some of these treatments will have been clinically effective in some circumstances, but these vary by PCT, reflecting variations in local clinical practice and referral patterns.

We used national data to identify the 'expected' activity for each PCT. This 'expected' value takes into account the age and sex breakdown for each activity on the list, and the level of deprivation of the PCT.

In other words, the national average level of activity for the type of people that live within a PCT's area. We used this value to estimate the savings, or reductions in spending that are potentially achievable. Hospitals would not see the same reductions in spending, but there would be increased capacity and money available for treatments of higher clinical value.

If PCTs are able to reduce activity to ‘expected’ levels, falls in spending of about £179m are achievable. This is an average of just over £1m for each PCT. If PCTs were able to achieve top-quartile performance (in other words, reduce the number of treatments to the same level as PCTs in the country with the lowest levels), then £441m a year could be available for investment in higher priority treatments.

The figure above shows that, unsurprisingly, the potential for reducing spending varies from one PCT to another. If PCTs are able to achieve 'expected' activity levels, the potential reductions in spending per PCT would range from around £14,000 to £6.4m. If PCTs are able to reduce activity to the levels of the top quarter performers, reductions in spending per PCT would range from £178,000 to £12.5m.

We have developed a tool that commissioners can use to help identify what the local position is. The tool sets out the spending for every PCT, by each of the low clinical value treatments and covers the past five years (see table, above right).

While our analysis has been standardised to account for local variations, it is only an indicator of the likely reductions in spending that are achievable. It does not reflect how easily achievable the reductions in spending are, nor any extra cost that may arise elsewhere.

Cost to the PCT

Also the tool is based on the cost to the PCT and not the cost of the provider in carrying out the treatment. If a PCT successfully manages to reduce the number of treatments carried out, the provider's income will reduce but not necessarily its costs. It will, however, free up provider capacity to deliver more clinically effective care.

For some PCTs the savings are significant, for others less so and the effort required to deliver potentially modest savings may outweigh the savings made, particularly if it could deliver greater savings by focusing efforts elsewhere.

Finance staff have a role in helping to evaluate the options. When researching the briefing we visited some PCTs that had already successfully reduced their spending on low clinical value treatments and they pointed to some key factors. Visible leadership, clinician support, effective communication and accurate data were important, but so was the involvement of finance staff.

We found the PCTs that had participation from finance staff had more success in decommissioning treatments and in particular measuring the reductions in spending. As part of the QIPP (quality, innovation, productivity and prevention) programme, most PCTs are looking at reducing their spending on low clinical value treatments. Using the tool to help identify whether savings targets in cost improvement and QIPP plans look sensible might be a good idea.

The main stumbling block for commissioners is developing the evidence base. Many PCTs hold lists of treatments that they consider to be low value, but there is inconsistency across the country, which is difficult to explain. The briefing has already caused a debate about what are the low value treatments, reflecting different clinical views and the absence of a national approach.

Work is already under way as part of the ‘right care’ strand of QIPP to help clinicians and commissioners work together to develop a single clinical evidence base. It might not be easy, but it would be a big step forward.

The briefing and the tool can be found at www.audit-commission.gov.uk/lowclinicalvalue

Emma Knowles is head of health financial management and efficiency at the Audit Commission