The Quality and Outcomes Framework (QOF; Department of Health [DoH], 2004) has presented an opportunity for the government to introduce performance-related pay in primary care. A total of 99 points are available for diabetes, which equates to £11880 for the average practice. While this may appeal to some clinicians, others have questioned whether offering financial incentives is the appropriate way to fund primary care. After all, is it not part of our professional role to offer high quality clinical care to all of our patients, regardless of any incentives? Like it or not, the QOF is here to stay, at least for the next few years.
General practitioners are a naturally competitive group of professionals, keen to deliver as high (if not higher) a standard of care as their neighbours. The advent of the Freedom of Information Act in 2000 (see http://www.foi.nhs.uk/home.html for more information [accessed 28.09.2005]) has presented an opportunity for practices to directly compare themselves with each other in the first published QOF league table (Health and Social Care Information Centre [HSCIC], 2005).
An analysis of 12 primary care trusts (PCTs) responsible for the care of approximately 60000 patients with diabetes demonstrated that a target HbA1c level of ≤7.4% was achieved in 61.3% of patients, total cholesterol of <5.0mmol/l was achieved in 73.2% of patients and blood pressure of <145/85mmHg was achieved in 71.3% of patients (HSCIC, 2005). These excellent results suggest that ‘financial carrots’ may work, driving up clinical standards of care to levels not usually observed outside clinical trial environments.
Maximising QOF achievement
Type 2 diabetes is increasingly acknowledged in primary care as a metabolic disorder and not just a ‘sugar’ disease. This was highlighted by the United Kingdom Prospective Diabetes Study Group (Manley et al, 1990), which observed that, at the time of diabetes diagnosis, 35% of patients were hypertensive, 18% had evidence of electrocardiogram abnormalities and 1% had suffered a stroke.
Due to its comorbid nature, people with diabetes will probably qualify for three other QOF domains – hypertension, coronary heart disease and stroke/transient ischaemic attack. This accounts for 356 out of a maximum of 550 available clinical points.
Practices achieving high clinical scores last year demonstrated a multidisciplinary approach to structured care, investing their QOF-based income to allow further diversity and change within clinical and administrative teams. The key features of successful practices were as follows.
Recognition of learning needs
Many GPs and practice nurses improved core skills in modern diabetes care by enrolling onto certificate and diploma courses led by teaching centres such as Roehampton University and the University of Warwick. Intensive information technology training also proved necessary for clinical and administrative staff to support an increasing dependence on computers.
Effective use of limited resources encouraged practices to review workforce requirements. Routine procedures such as phlebotomy were devolved to healthcare assistants, allowing practice nurses to concentrate on diabetes patient reviews.
Housebound, residential and care home patients have often received sub-standard diabetes services (Benbow et al, 1997; Fahey et al, 2003). By engaging with and agreeing care protocols with community pharmacists and district and care home nursing teams, patient access to all healthcare professionals has improved. In some areas practice nurses have been given protected time to visit patients in their own homes.
Regular primary healthcare team and clinical audit meetings have encouraged team motivation, highlighted strengths and weaknesses in service delivery and allowed feedback on QOF progress. Communication with local diabetes networks and hospital-based colleagues has been improved by developing integrated care pathways. This has helped patients and secondary care colleagues understand the QOF and encouraged documentation of important clinical information in hospital outpatient letters.
Assigned members of the administrative staff have been given protected time to recall patients who either are not treated to target or are due for routine review. Although this role has often been performed by practice nurses, their own increased clinical workload has made it more sensible to devolve this role to non-clinical staff. Three hours each week has usually been sufficient to allow review of all ten clinical domains for the average practice.
Empowering patients to manage their condition by knowing optimal care standards and being aware of the need for regular review has eased the administrative recall processes.
Accurate disease registers
Most practices have established diabetes registers, but, in order to accurately maintain them, monthly audit reports should be performed. The Primary Care Information Service (home page: http://www.primis.nhs.uk/pages/default.asp [accessed 28.09.2005]) provides a free service to registered practices using MIQUEST integrated software (Miquest Limited, Swindon, Wiltshire).
To identify patients with diabetes who are not on the disease register, searches for the following READ codes are helpful.
- Screening codes: diabetic monitoring, diabetic eye check, diabetic education.
- Laboratory tests: fasting blood glucose >7.0mmol/l, random blood glucose >11.1mmol/l, HbA1c, fructosamine and microalbuminuria.
- Referral codes: to diabetologist, to diabetes specialist nurse, seen in diabetic clinic, diabetic admission.
- Prescriptions: oral hypoglycaemics, insulin, urine glucose strips, blood glucose strips, insulin syringes, insulin needles, insulin pens, lancets.
The Quality Management Analysis System (QMAS) used to interrogate and measure QOF performance relies on information recorded within practice computer systems. Where information is not coded, or coded incorrectly, funding fails to follow the patient. To prevent this, practices should use clinical software templates which code data in a standardised, consistent and reliable format, minimising the need for non-auditable free text. It is also important to ensure mechanisms exist to code hospital outpatient letters when they arrive at the surgery.
Most practices are now linked electronically to their local laboratories by pathology links. Where this facility is unavailable, HbA1c, creatinine and lipid results must be manually recorded on clinical systems.
Regular audit underpins success in achieving QOF targets. Cumbersome clinical software reports have been surpassed by specialist software such as Contract+ (Informatica FrontDesk, Send, Surrey), Contract Manager (MSD Information, Hoddesdon, Hertfordshire) and Apollo (Apollo Medical Systems, Rotherham, South Yorkshire). These software packages allow day-by-day measurement of QOF progress, identifying patients requiring treatment to target, and will usually pay for themselves within a week of purchase.
The appropriate use of exception codes has raised considerable debate. Since none of the diabetes indicators requires 100% achievement, only patients that strictly fulfil DoH guidance on exception reporting should be excluded from the framework (DoH, 2004). Some PCTs are sympathetic to the lack of evidence that all patients with hyperlipidaemia or microalbuminuria should be treated with a statin or angiotensin-converting enzyme-inhibitor. If you have not already done so, you should approach your PCT’s diabetes and clinical governance leads to explore this further.
Our reward for success in delivering last year’s QOF means we should anticipate both bigger and tougher targets when the clinical framework is reviewed in 2006. For altruistic clinicians already practising evidence-based medicine there will be no great surprises. For those of you who prefer ‘carrot dangling’, identify your diabetes at-risk group by recording ethnicity, family history and waist circumference and you will probably keep one step ahead of changes for 2006.