No such thing as a justified target

No Justified Targets

Before the election in May, the three health spokesmen for the major parties debated on television. Obviously the question of targets came up and Andrew Lansley said he would scrap “politically motivated” NHS targets but keep those that were “clinically justified”.

The problem is that there is no such thing as a clinically justified target. The reason is that targets are arbitrary and make performance worse. That is never clinically justifiable.

A target to have every suspected cancer patient seen within two weeks has behind it a laudable goal – that patients with serious conditions should get the best treatment possible. There is nothing wrong with that goal but everything wrong with having a target.

Study Demand

To achieve the goal you would start by looking at demand. “How many suspected cancer patients do we get every day/week/month?” You would then look at the current system to see how this demand was dealt with and you would measure it from the patients’ point of view. They want to be seen quickly, they want to get the right treatment and have the best outcome possible. Also they want dignity, respect and to be treated like human beings.

If you are thinking of hitting a target your mind is filled with resources, measures and reporting. I went to Exeter to help run a workshop on how to meet the 18 week target from diagnosis to treatment. One of the questions from the floor was about what monitoring should be put in place to spot when patients are just about to breach the target and so fast-track them so they won’t breach. If you put that monitoring in you are moving resource from the core flow of diagnosis and treatment and over to monitoring. If then you add a fast-track expediting of patients just about to breach you add a further complication to the system where suddenly certain patients jump the queue to make the trust’s figures look better. But remember that patient has jumped over other patients who are now delayed and so more likely to get close to breach. If they get close to breaching they will jump the queue and so the cycle continues. You are in effect increasing the variation in the system while at the same time reducing the resources available to the core flow.

Measures of Purpose

The correct approach is not to have monitoring of near breaches but to measure the end-to-end times of patients from diagnosis to treatment, while also measuring the type of treatment given. You will see that the end-to-end time will vary. This variation is normal in any system. Your first job is to remove any abnormal variation until the system is stable (which is not to say it is yet effective – just stable) then work on the flow of the patient through the system to remove the batching, waiting, errors and rework so as to reduce the variation and improve the system to get the time from diagnosis to right treatment as short as possible.

This method will leave any target setting in the dust. Why 18 weeks? Why not 18 days? Why not 18 hours?

Why two weeks for cancer patients? Why not two days?

Method Not Targets

Targets don’t help you to improve, in fact they distract from improvement. People set targets when they have no method. When they have a effective methods, they get results that would be seen as ludicrous to set as a target.

Best,

Rob