"The NHS will last as long as there are folk left with the faith to fight for it"
Aneurin Bevan

Monday, 25 January 2010

Conservative Green Paper on Health Outcomes: Part 2

The Conservatives policy is to replace "targets" with "outcomes", their Policy Green Paper No 6 explains the rationale behind this policy.

Measurements and Publishing Results

The Conservatives obsession with outcomes is manifest in two ways. First, payments will be based on outcomes. As I mentioned earlier someone has to evaluate the outcome and determine from this how much the hospital will be paid. Second, outcomes have to be collated and then published, and these league tables of outcomes, in and of themselves, will magically improve the hospital.

So what data will be published? Hopefully, the information will be aggregated data and not specific information about an individual's treatment. The policy document says:

"We start from the position that all data should be published unless it can be shown why this will cause harm."

On the surface there is nothing wrong with this policy. But as anyone who has ever tried to make head-nor-tail of a spreadsheet will tell you: if you have too much data you cannot see the overall picture. It is a case of not being able to see the wood for the trees. By all means publish all the data, but experts should also provide an analysis of the data to provide an overall assessment from that data. The problem with this approach is that raw data is rarely instructive and people can make bad choices by not being able to interpret what the data means. Further, some people may appoint themselves 'experts' and interpret the data in a way that fits in with their political views. Such data is best collated, interpreted and published by an independent organisation, like the Office for National Statistics. The Conservatives give no clue about how they think outcomes will be published.

Another comment must be made. Collecting all this data is not simple. It involves administrative staff. The Conservative draft manifesto on health says that they will cut NHS administrative costs (ie staff) by a third. The Conservatives charge that targets require huge numbers of bureaucrats to gather metrics, so will the collation of outcomes be any different? It is unlikely that hospitals will be able to provide all the data the Conservatives say Hospitals will provide if the Conservatives make the administrative cuts that they promise. Indeed, the Conservatives admit this:

"Providers already have capability for the collection, collation and publication of outcome measures. Any transitional costs in areas such as IT and staff training will be minimal, if apparent at all,"

The Conservatives are suggesting that they will not cut the numbers of administrators to gather metrics, so where will their cut of a third come from? Further, this is saying that there will be no cost for NHS providers to change from collection, collation and publication metrics for targets to collection, collation and publication of outcome measures. While it is reassuring that there will be no money spent on the transition, doesn't it make you wonder if the reason why there will be no additional costs is because the NHS administrators will actually be doing exactly the same thing?

The interesting thing is that the same part of the green paper suggests that by scrapping targets the NHS can save money, but the statement above suggests that the same personnel and equipment will be used to collate data for the outcomes. If the savings are not from the actual process of collecting metrics for targets, where will the savings come from? The only conclusion is that the Conservatives expect that simply the change from measuring for targets to measuring outcomes will instantly make the NHS more efficient. This is an unsubstantiated claim.

The same section in the Conservative policy document suggests that targets cause "clinical distortions resulting from activities aimed purely at meeting government targets". Distortions are, of course, changes, so when is a change an improvement, and when is it a distortion? More significantly, the Conservatives make no mention as to how they will prevent the measurement of "outcomes" causing "distortions". (As I have mentioned in a previous blog, since payment will be based on outcomes, there will be a lot of pressure for clinicians to make sure that the outcome that is reported – not necessarily the actual outcome – is as successful as possible. I am not suggesting falsification here, merely that there will be a subconscious suggestion to be optimistic.)

The problem with the policy of publishing data is that the Conservatives do not say how it will improve standards, and they do not say what will happen if a provider has poor outcomes. The only explanation to how a outcomes will improve providers, is that when patients stops using a provider it will improve:

"We believe these results, … should be made publicly available to help patients … choose the right care provider."

It should go without saying that there is no healthcare market, and that a market cannot be created because the size of the investment required centralises treatment in a few providers (principally NHS providers). While patients do have choice, most patients choose a provider that is local to them, above other criteria. Consequently, improvements in healthcare are unlikely to occur due to patients' right to choose.

Outcomes will also, apparently, be the responsibility of the patient:

"developing outcome measures which patients with chronic conditions themselves provide, reflecting their needs and what matters most to them;"

This statement is unclear. We already know that the Conservatives will only pay NHS providers according to outcomes, yet in this statement they say that patients will suggest what the outcomes will be, and consequently, patients will have the power to determine whether providers are paid at all. This is worrying, not least because providers will become reluctant to treat patients if they will not have a guarantee of payment for that treatment, and hence providers are likely to be more reluctant to treat patients who have chronic conditions.

The Conservatives explain further how they will help patients with chronic conditions:

"We would consider using EQ5D – a 2-page patient questionnaire, developed and by EuroQoL. EQ5D been used in population studies over the past decade. The patient conducts a self-assessment that covers five dimensions: mobility; selfcare; usual activities; pain/discomfort and anxiety/depression."

Does this sound like something that most people with chronic conditions will want to participate in? Someone with a long term heart condition will have different needs and priorities than someone with diabetes. Indeed, someone with type 2 diabetes will often have different needs and priorities than someone with type 1 diabetes. Yet the Conservatives are offering a one-size-fits-all questionnaire for all patients with chronic conditions. What is the incentive for the patient to fill in this form? Will treatment be refused if the questionnaire is not returned? Indeed, if there are few patients willing to complete the questionnaire, will this result in an unrepresentative outcome being used?

It is clear that the Conservatives intend to keep a target culture in the NHS except that they will rename targets to outcomes. The Conservatives will maintain the same numbers of administrators to collate outcomes as are being used to collate target data, yet the Conservatives claim that they will cut administration in the NHS by a third. The Conservatives will publish raw outcomes data and hope that patients choosing alternative providers will improve a provider with poor outcomes, something that is unlikely to happen.

No comments:

Post a Comment