The challenges facing clinical commissioning groups (CCGs) are numerous, including the need to reduce the number of secondary care referrals and inappropriate hospital admissions that cost the NHS billions of pounds a year.
An article recently highlighted some of the more bizarre examples of A&E attendance including broken false nails, nettle rashes and the removal of permanent ink from someone's forehead after a drinking game went wrong.
This is of little concern to GPs as PCTs pick up the treatment tab - but that is all set to change from 2013.
The first step towards dealing with the problem is getting a clear picture of how, where, when and with what frequency patients are using primary and secondary care services.
Phil Wigglesworth (left) and co-founder Stuart Bond at Downing Street
Inefficient data storage
Previous efforts have been hampered by patient data being kept in different parts of the NHS with no software capable of drawing the information together and analysing it on a large scale.
Some of this information is kept in community care and some in hospitals.
However, the lion's share sits in GP surgeries on a variety of outdated and unconnected computer servers largely unused outside the environs of the surgery.
As a result, it is impossible to see anything but a disjointed picture of healthcare, leaving sizeable gaps in the NHS's understanding of its own business, including patient movement and expenditure.
But what if you could find a way to move across traditional service boundaries and join the pieces of the puzzle together to form a comprehensive picture of the health service and its needs?
This is the challenge we faced in north-east London when the local PCT was looking for an effective way to support GP-led commissioning.
First, we extracted hospital data by analysing the monthly financial recharges made to the trust, which contained all the relevant patient information.
The practice data proved more challenging and we visited dozens of practices teasing relevant information out of practice servers, installing new software and training staff in its use.
The resulting sets of information were processed in a central 'warehouse' which analyses and cross-references millions of records, producing solution-driven algorithms.
These algorithms include the ability to see comparative statistics and performance compared with neighbouring practices, track emerging trends, review financial spend and co-ordinate health initiatives across wider geographical areas, giving CCGs a clear picture of the health landscape and the needs of its patients.
|A GP's VIEW|
'The QOF gives us basic data about numbers of patients but it won't give an indication about the severity of a disease or what treatment patients are on. This system allows comparisons between practices on different indicators for patients. For example, are patients with severe COPD on the right treatment? Are they receiving the right diagnostic investigations? It allows practices and no-one else to identify patients who could be treated better so they don't need to go to hospital. If a practice changes how it records things, the system can quickly see improvements.'
At present, we have a platform of more than 15 software modules in use, including data acquisition, budget management and referral management.
One area where it can make a huge difference is the effective management of long-term conditions through a risk stratification tool that can assess those most prone to hospital admission by analysing their previous two years' medical history.
Risk is calculated through using multiple algorithms and multiple indicators, including the number of inpatient and outpatient appointments made by an individual and the number of visits to their GP, A&E attendances and medication.
The software has produced some surprising results.
Try the patient who went to A&E 279 times in one year; 1% of a borough's population with long-term conditions costing the local PCT £36m a year; and estimated savings of £6.5m over two years through more effective management of patients with heart disease.
It's easy to wail about the waste and paucity that these figures reveal but it also shows there is great capacity for change in the NHS, irrespective of its current budget.
A good example is our work in AF, which is a deciding factor in 14% of all strokes.
The condition affects more than half a million people in the UK with increased vulnerability to stroke caused by blood clots forming in the heart and entering the bloodstream.
Our risk stratification software identified high-risk patients, proving an automatic alert system for prescription of warfarin, an anticoagulant that reduces the threat of clots.
This saves an estimated five patients a year from a stroke, as well as £44,000 in hospital care per patient, showing that savings can equal improved patient care.
Of course, it is important to strike a cautionary note and highlight the fact that technology can only help people if people let it help them.
- Phil Wigglesworth is one of the founders of Health Analytics and has been working with GPs in north-east London for two years developing and implementing a health informatics system now driving commissioning