Going for big gains in the diabetes quality markers
Dr Lorna Gold continues our series on the quality framework
here is considerable variation between GPs in their enthusiasm for, and expertise in, the management of diabetes and the new contract takes this into consideration in the quality markers. There are 99 points available for diabetes mellitus, and even practices that routinely refer every patient with diabetes to hospital and offer no more than the most basic care should be able to achieve many of these if they are prepared to make the effort to enter data from hospital records on computer and to use the exception-reporting system to exclude from their statistics those patients who fail to attend for follow-up, are non-compliant with treatment or whose control is suboptimal on maximal treatment.
The quality markers do not distinguish between type 1 and type 2 diabetes but children aged 16 and under are excluded as diabetes in childhood is normally managed in secondary care.
There is a lot of overlap with coronary heart disease so data that has already been recorded as part of another set of indicators need not be recorded again.
For all indicators, a minimum of 25 per cent coverage is needed to score any points at all, and a sliding scale applies thereafter.
Develop a diabetes register (six points). Do a search on all the oral hypoglycaemic drugs used in your practice, on insulins, and on products such as Hypostop. Such a search may fail to pick up patients whose diabetes is being managed by lifestyle changes alone, so it is worth searching on blood and urine testing strips and lancets.
There are dozens of diabetes codes but your register should require only two: type 1 and 2.
There are several codes for each category and it will simplify data retrieval if everyone who enters information uses the same ones. You may also choose to subdivide patients with type 2 diabetes depending on whether they are on lifestyle modification alone, oral hypoglycaemic agents, insulin or a combination of treatments, but it is not necessary to do so. Although not required by the contract, as a point of good medical practice it would be worth adding a code for gestational diabetes with a prompt to check the fasting blood sugar of these patients annually as they are at very high risk of developing type 2 diabetes.
Ensure you are using current (1999) WHO criteria to diagnose diabetes. It is still surprisingly common to find patients who have been diagnosed as having diabetes on the basis of a single random blood glucose of 8mmol/l or a slightly elevated HbA1C, and almost as common to find a fasting blood sugar of 8mmol/l in the absence of osmotic symptoms being ignored. In asymptomatic patients, a fasting blood glucose above 7.0mmol/l or a random blood glucose above 11.1mmol/l should be repeated on a different day.
If the second result is also high the diagnosis is confirmed. If only one result is abnormal a two-hour glucose tolerance test should be performed and a result above 11.1mmol/l regarded as diagnostic of diabetes. A single abnormal result can be regarded as confirming the diagnosis if the patient also has symptoms suggestive of diabetes. HbA1C should never be performed as a screening test for diabetes, although it is the investigation of choice for monitoring glycaemic control following diagnosis, and in future it may prove to have an independent role in assessment of cardiovascular risk.
It is important you do not systematically over- or under-diagnose diabetes as your register will be compared with the PCO prevalence.
Record the most recent information on the following and arrange for patients who have not had an annual review for over 12 months to be seen. If your practice is heavily paper-based, or if most of your patients with diabetes have their annual reviews in secondary care, you are likely to have to trawl through thick notes to find all the information. Pathology links will make the task simpler.
· Body mass index (three points for 90 per cent). You need only record that it has been measured. We are not yet to be penalised financially if our patients with diabetes are unable or unwilling to lose weight.
· Smoking status (three points for 90 per cent). Patients who have never smoked need to have the fact entered only once. Data on smokers and ex-smokers should be updated annually. Recording that advice to quit (or not to restart) has been given is worth another five points for 90 per cent.
· HbA1C or, in particular circumstances such as patients with haemoglobinopathies, fructosamine (three points for 90 per cent). Text in the numeric value of the most recent HbA1C and, if your practice hopes to earn the maximum points available for diabetes care, list those in whom the result is greater than 7.4 per cent.
· Retinal screening (five points for 90 per cent). This can include ophthalmoscopy by an appropriately trained GP or physician, an examination by an optometrist or attendance at a retinal photography clinic. You need only record it has been performed. This is also a good opportunity to ensure all patients with more than background retinopathy are under the care of a specialist diabetic eye clinic.
· Peripheral pulses (three points for 90 per cent). Record whether each lower limb pulse is present or absent.
· Neuropathy testing (three points for 90 per cent). A monofilament and tuning fork should be used as a minimum standard.
· Blood pressure (three points for 90 per cent). Text in the numeric value of the most recent measurement and, if your practice hopes to earn the maximum points available, list those in whom the reading is 145/85 or greater.
· Serum cholesterol (three points for 90 per cent). Text in the numeric value. If your practice hopes to earn the maximum points, list those in whom the measurement is greater than 5.0mmol/l.
· Serum creatinine (three points for 90 per cent).
· Micro-albuminuria (three points for 90 per cent) except in those patients who have frank proteinuria due to diabetic nephropathy, who should be exception-reported. The contract document does not state what means of testing should be used, suggesting that near-patient testing with dipsticks or a laboratory measurement of albumin/creatinine ratio are both acceptable. A further three points are available for practices that can demonstrate at least 70 per cent of patients with micro-albuminuria or proteinuria are being treated with ACE inhibitors or angiotensin-2 antagonists.
Arrange a system to ensure this information is updated annually. Add those patients who are under 65 to your flu immunisation target group. To earn the maximum three points you need to achieve a very challenging 85 per cent coverage, so be ruthless about recording those who refuse immunisation or in whom it is contraindicated, under exception reporting.
Going for big gains
As with CHD, the biggest winners will be practices treating patients to target. Having told the Government we are clinicians and not statisticians we have a responsibility to demonstrate this by putting as many as possible of these 50 points on to the credit side of our practice accounts.
These are the targets to aim for:
· HbA1C 7.4 per cent or less (16 points for 50 per cent coverage).
· HbA1C 10.0 per cent or less (11 points for 85 per cent coverage).
· Blood pressure 145/85mmHg or less (17 points for 55 per cent coverage). Note that more stringent control is expected in patients with diabetes than in other patients at high risk of cardiovascular events.
· Total cholesterol 5.0mmol/l or less (six points for 60 per cent coverage).