This site is intended for health professionals only

At the heart of general practice since 1960

How better data will empower our commissioning

Dr Nav Chana explains how marrying PCT public health data with practice-level information can really make a difference to quality of care

Dr Nav Chana explains how marrying PCT public health data with practice-level information can really make a difference to quality of care

Public health data on factors such as deprivation, smoking rates and so on is often presented at a postcode, PCT or regional level and bears little relevance to the reality faced by individual practices.

And yet the great potential of PBC is to manage a community's needs at a locality, cluster or even practice level rather than having services that cover the whole PCT.

It may seem like a simple task to integrate data with disease prevalence, QOF scores and admission rates to get an overall picture of the population a GP practice is working with, but in practice it is extremely difficult.

Yet despite the complexity, a primary care needs assessment project carried out last spring by my PCT (Sutton and Merton) convinced me that this needs to happen far more routinely and frequently than it does at the moment.

For about six months, one of the PCT public health consultants was given the task of analysing routinely collected data at a public health level which was then presented at a practice level with comparison with the PCT average.

The data was presented as a series of bar charts with each practice displayed as one bar against a number of domains, many taking into account the demographics served by that particular practice.

The idea was to get a much better understanding on the unique constraints faced by each practice – and the result was a handful of charts that looked very different from what you might expect.

Some surprises

The exercise threw up several surprises. We were provided with the number of patients over the age of 75 on each practice list, which was incredibly revealing, with some practices having a far larger proportion of elderly patients than others.

Many practices appeared to be dealing with significantly deprived populations and the degree of deprivation varied with practices just a mile or two apart.

Another interesting factor was the number of emergency admissions, for which the data was collected down to practice level but then adjusted for age, sex and variations in deprivation. After that adjustment was done, those practices that on the face of it had the highest number of admissions turned out to have a pattern well within the normal range. Conversely, some of those practices that seemingly did not have a problem with high emergency admission rates became outliers.

COPD example



Figure 1 (left) shows COPD QOF prevalence for a PBC cluster while figure 2 (below) shows respiratory emergency admissions for the same cluster.

It appears that practice 1 has a relatively low COPD prevalence (compared with the PCT average) whereas practice 3 has a much higher prevalence.

Respiratory admissions from practice 1 are much higher than the PCT average while practice 3 admission rates are within the confidence intervals of the PCT average.

Given that the accuracy of any data is debatable, it is important to ensure no judgment is attributed to information such as this, but to use it to understand the differences, and more importantly to consider the practice developmental issues that might arise.

Using the data



As a PBC cluster it is vital to understand what is going on and how to drive up quality. Using these more complex ‘adjusted metrics' highlights where the anomalies that need discussion lie rather than focusing on ‘outliers' from a quick glance at unadjusted figures.

Exercises such as this also show that the PBC referral incentive schemes that have sprung up around the country over recent months may not reflect the true picture if the full demographic profile and population characteristics broken down by practice are not taken into account.

Using the data should form part of the educational and developmental process – understanding the differences between practices and the reasons for them – but the results might also prompt commissioners to think about redesigning some services, for example COPD pathways.

For example, in my practice, where we have a high number of elderly people in a fairly deprived area, we are talking about redesigning services to offer more smoking cessation and to target COPD care at home for domiciliary patients who find access to the practice difficult.

An educational event at which each practice is invited to discuss the data in a constructive non-threatening way is essential.

A matter of routine

The surprising thing to me was not necessarily what the project uncovered – although the final bar charts look very different to the unadjusted versions – but that this type of data analysis is not done as a matter of routine by all PCTs and updated regularly to ensure the information is current.

We need to have this sort of information produced on a regular basis to feed into our commissioning decisions.

Using a public health expert to analyse the figures is the key. He had the understanding to be able to pick apart data on deprivation and social indicators and tally that with disease prevalence and hospital admissions.

Clearly, the vital factor is the ability to present all this down to a practice level.

Looking to the future, I would like to go even further. For example, in a practice of 10,000 patients it might be meaningful to segment the population further and to understand the health needs of the population at the ‘street' level. If most of the heavy smokers live in one or two discrete patches, for example, this is where the smoking cessation health education strategies should be targeted. That is the kind of detail that will really give us a picture of the care we provide and where we can do better.

Dr Nav Chana is vice chair of the NAPC, chair of the Integrated Primary Care Commissioning PBC consortium and a GP in Mitcham, Surrey

Dr Nav Chana: as a PBC cluster it is vital to understand what is going on and how to drive up quality Dr Nav Chana: as a PBC cluster it is vital to understand what is going on and how to drive up quality COPD QOF prevalence Adult respiratory admissions

Rate this article 

Click to rate

  • 1 star out of 5
  • 2 stars out of 5
  • 3 stars out of 5
  • 4 stars out of 5
  • 5 stars out of 5

0 out of 5 stars

Have your say