How we measure improvement

We bring together experts and those achieving good results in a given topic area to establish the characteristics of their systems that produce a better outcome.

From this, we identify the key change principles which, if replicated by others, will result in similar good practice. We use these to develop the monthly measures on which participants report.

The next step is to expose people working in the field to the examples of good practice, the change principles and the measures and help them implement and measure change in their own working environment.

We teach a variety of quality improvement techniques, and we roll-out and spread the improvement across the entire country via our 10 area teams.

Example: improving access and responsiveness in primary care

A high quality practice that wants to achieve better health outcomes will, almost automatically, have fast access for patients.

When we began, it was common for a patient to have to wait two weeks to see the doctor, with the average starting point being more than five working days.

The change principles for this piece of work included: measure the demand, shape the demand, match capacity to the demand, and create a contingency plan. Demand included the acute minor ailment and the long-term condition check up. Demand could be shaped by providing a skill mix of health professionals or a variety of different consultation methods, such as telephone or email. Contingency planning covered the predictable (holidays, learning time) and the unpredictable (staff sickness, traffic jams, flu epidemics).

We measured improvements by monitoring practice's third available appointment and the percentage of people seen on their day of choice. The third appointment was selected because a fortuitous cancellation could skew the figures if you used the second available appointment. The day of choice measure was to emphasise the importance of people to be able to book appointments in advance.

The participating practices achieved a reduction in waiting time for an appointment from an average of five days to an average of one day for both GPs and practice nurses, whilst still enabling patients to book in advance.

The Government's 24/48 hour target is different to the measure we used to assess access. As a result, there are undoubtedly some practices that have changed their system to achieve this target without implementing its spirit, and hence prevented booking in advance.

Example: enhancing the management of long-term conditions and health outcomes

We have helped primary care organisations to improve their management of four specific conditions; coronary heart disease (CHD), diabetes, chronic obstructive pulmonary disease (COPD) and common mental illness.

The change principles are the same for all: establish a disease register, systematically deliver the research evidence, involve the patient and adopt a multi-disciplinary, multi-agency approach for those with complex needs.

Our measures were largely proxy outcome measures as in the Quality and Outcomes Framework, for example spirometry in COPD, blood pressure control in diabetes and use of lipid-lowering drugs in heart disease.

This is a valid way of measuring improvement. If the research shows that a particular intervention produces a certain outcome, then measuring the intervention is perfectly justifiable. The numbers involved in an outcome (say premature mortality) are too small at a practice level or naturally vary too much over a period of time to make a rational judgment. For example, a practice will have to intervene with all of its registered patients with heart disease to save a small number of premature deaths.