GPs in the NHS are the gate keepers of an annual prescription spend of £8.3b in 2007/08.  The paradigm is that GPs decide who should receive medication and more importantly, which type of medication is the most appropriate for a patient. 

As the complexity of prescription item selection increases, the NHS has to ask itself, can it afford to pay for an ever increasing medicines’ budget?

GPs are supported in their decision-making by resources such as the British National Formulary (clinical) and the Drug Tariff (financial) however the available evidence indicates that GP choices (while clinically sound) are often not the most cost-effective.

In 2007 the National Audit Office reported that primary care trusts could save more than £200 million a year without compromising patient care if GPs prescribed cheaper medicines. 

In 2009 Keele University calculated that almost £400 million has been saved by the Department of Health, the NHS and PCTs in England, through more cost effective prescribing as recommended in the NAO report.

 The NHS Information Centre’s Prescribing Support Unit maintains the database of primary care prescribing. They recently published some headline figures for 2007/08.

  • The cost of prescribing in the NHS was £8,325.5m
  • The number of prescriptions in the NHS was 843m
  • Of the 843m prescriptions issued 739.6m were issued at no charge to the patient
  • For the last 10 years there have been larger numbers of prescriptions issued by the NHS year on year. They currently stand at 16.4 prescription items per head of population per annum
  • The introduction of the Category M scheme has created price decreases in many medicines

A more detailed analysis of this prescribing database reveals that there are still many millions that could be saved without compromising patient care or safety. Instead of just switching from expensive brands to cheaper generic medicines, prescribers can also undertake therapeutic class switching.

Scenario One

There are six Selective Serotonin Re-uptake Inhibitors (SSRIs): sertraline, fluvoxamine, citalopram, escitalopram, paroxetine and fluoxetine which all have a similar mechanism of action described in the BNF. They have fewer side-effects than the older Mono Amine Oxidase Inhibitors and Tricyclic antidepressants. 

Of the SSRIs, fluoxetine is the cheapest, has the most evidence of use with the largest number of patients and is the only one recommended by the Committee on Safety of Medicines for use in paediatrics.

On this basis it would be reasonable to think that the vast majority of depression in primary care would be treated with fluoxetine and that the other drugs would only be used where there was strong evidence that fluoxetine was not tolerated or ineffective. 

In fact this was not the case, of £61.6m spent in 07/08 on SSRIs, only £5.8m (9 per cent) was spent on fluoxetine. If all patients prescribed SSRIs had been prescribed fluoxetine, it would have saved the NHS £34m (55 per cent).

So why did GPs prescribe more expensive drugs that are no more effective? 

The answer lies in the analysis of the prescribing database: The SSRI with the highest spend was also the newest and most advertised, escitalopram which cost the NHS £25m (41 per cent). £36.1m was spent on tricyclic antidepressants which comprises 37 per cent by value of anti-depressants.

Scenario Two

The NHS Institute for Innovation and Improvement encourages prescribing productivity improvement through the Better Care, Better Value indicators. As Keel University reported, many GPs have been able to improve the cost effectiveness of statin prescribing in line with NICE recommendations. 

There are five statins: simvastatin, rosuvastatin, pravastatin, atorvastatin and fluvastatin which all have a similar mechanism of action described in the BNF.  They are more effective than other classes of drugs in lowering LDL cholesterol and are used in primary and secondary prevention of cardiovascular disease. 

The cheapest is simvastatin, it has the most evidence of use with the largest number of patients and is recommended by NHS Better Care, Better Value based on NICE guidelines.

Given that there are few alternatives to statins and that they have similar clinical effectiveness, their selection should be based on cost effectiveness with simvastatin used in the first instance for all patients unless it is not tolerated or ineffective. 

Despite clear national guidelines directing prescribers, there is still inefficient, costly prescribing. In 2007/08 there was £450m spent on statins in primary care. Simvastatin was prescribed to 70% of statin patients at a cost of only £53.2m (12 per cent).  If all patients prescribed statins had been prescribed simvastatin, it would have saved the NHS £374m (83 per cent) of the money spent.

So why did GPs prescribe more expensive drugs that are no more effective? The answer lies in the analysis of the prescribing database: atorvastatin was prescribed to 22 per cent of statin patients accounting for the bulk of non-simvastatin spend and cost the NHS £334m (74 per cent) of the total statin budget. Did those patients receiving atorvastatin get another £334m of improved health outcome?

Scenario 3

The prevalence of coeliac disease in the UK population is difficult to establish. There are 62,500 patients with a diagnosis of autoimmune destruction of small bowel villi in response to gluten resulting in impaired absorption of food and recurrent attacks of diarrhoea. 

This diagnosis is based on an IgA tissue transglutaminase test (tTGA) followed by an IgA endomysial antibody test (EMA) if the tTGA is inconclusive.

There are claims that as many as 500,000 people could be affected but there is little evidence to support these statements as the majority of people with gastrointestinal disturbance or diarrhoea are not tested. The disease frequently affects 40-60 year olds however can occur at any age.

In 2007/08 the NHS spent £25.2m on gluten-free food products including bread, flour, biscuits and cakes. At the time that gluten free products were introduced to the NHS they were unavailable through supermarkets, this is no longer the case. 

Not including the cost of the GP’s time, the pharmacist’s time and the dispensing fees; the cost to the NHS is £19.88 for a loaf of gluten free bread which one supermarket sells at £2.28 and £12.63 for a packet of gluten free pasta which the same supermarket sells at £2.12.

Many GPs in the UK are now refusing to prescribe gluten free food products to patients. Nevertheless patients will still request gluten free products on prescription.

The NHS is heading in the right direction by focusing attention on the cost of prescribing in primary care. With industrial scale productivity improvements, it may even reach a billion pounds of prescribing efficiency savings. In order to achieve this goal, GPs, pharmacists and commissioners need to work together with patients to save for the future NHS.