The GP contract of 2004 heralded a new era in evidence-based medicine. Academic modelling suggested a huge potential reduction in deaths from cardiovascular disease by systematically controlling blood sugar, hypertension, and cholesterol.1 So the QOF was born and 25% of our income was linked to its points system. Money is the great motivator, we are intelligent people, it was just a points game and so we got gaming. Establishing disease registers, recall systems, testing and prescribing. Sick patients didn’t get a look in for an appointment, as patients on different registers were receiving endless letters and appointments. Our once-sporty primary care computing systems become beige Austin Allegros; slow, clunky, unreliable, and at risk of crashing at any time. QOF templates also leaked irrelevant data and pop-ups all over the clinical system obscuring relevant information. This protocol-driven care stifled innovation and free thought; both judgement and discretion thrown under the wheels. The production costs were huge, £10 billion in payments directly,2 many billions more on medication and countless lifetimes of everybody’s time. This car crash contract wrecked general practice. But recently the QOF has been recalled in Scotland. Why?
The simple answer is that it didn’t work, there is no evidence it reduced premature death2 or impacted on the trajectory decline in vascular disease.3 The Treatment paradox, means that improvements from managing cholesterol, hypertension, and diabetes are impossible to demonstrate in a single GP practice, given that the absolute benefit to individuals is tiny. The QOF, however was the biggest public health experiment in the world. It should have definitely caused a sea change in vascular disease prevalence at a population level. Now we can shrug our shoulders and move on, pretend this never happened and spend the next two decades attempting to delete all the rubbish on our computer system.
But why didn’t QOF impact on vascular disease? Statin prescriptions more than doubled to 68 million prescriptions a year, diabetic prescription doubled to 47 million prescriptions (new agents lacking mortality data having pushed costs up to a staggering £849 million a year in 2014), and antihypertensives have nearly doubled to 70 million prescriptions a year.4 Biblical scale polypharmacy. The lack of demonstrable impact suggest the current risk factor paradigm might be wrong. If it is wrong, then the current model has systematically stolen wellbeing on a global scale and made the unworried well, into the worried ‘unwell’.
For the epidemiology of vascular disease the modelling algorithms are based on assumptions with wide confidence intervals. The trajectory of decline is absolutely fixed and unchanged since the 1970s. Today the death rate is a mere 20% of what it was in 1970.3 Countries like Spain smoke twice as much as the UK but have lower rates of disease.3,5 Two people who simply live in different postcodes have dramatically different treatment thresholds,6 the overweight live longest,7 saturated fats are not linked to cardiovascular disease8 and a rising cholesterol is not associated with increased vascular disease.9 To name but a few!
Consider also, the treatment benefits of cholesterol and hypertension are relative, with the absolute benefit depending on the background rate of vascular disease. So as the background rate is now a third of what it was in the original statin trials, the numbers needed to treat have tripled. Statins meagre benefits made ever more marginal.
It is apparent there are major confounding factors in play in vascular disease’s terminal decline. We need to explore the known unknowns and discover the unknown unknowns in vascular disease. If we burst this subprime evidence bubble it would crash the medical world, freeing millions of patients from life-long polypharmacy. Now that wouldn’t be just good medicine but brilliant medicine.
- © British Journal of General Practice 2016