How NICE may be outflankedBMJ 2006; 332 doi: https://doi.org/10.1136/bmj.332.7552.1268 (Published 25 May 2006) Cite this as: BMJ 2006;332:1268
- Correspondence to: R E Ferner
We argued a decade ago that the NHS should not have to pay for new drugs unless they are at least as good as older ones, nor for expensive drugs whose benefits are uncertain.1 Since then, the National Institute for Health and Clinical Excellence (NICE) has been created. NICE appraises technologies that are available to the NHS and recommends whether they should be used unreservedly, with restrictions, or not at all.2 Part of its remit is to ensure equity, but equity is not in everyone's interests. Here, we consider how individuals or groups with specific interests may seek to outflank NICE.
Individual benefit or common good?
When many people share common resources, it is rational for each individual to increase personal use of the resources. But if all individuals do this, the resources are overexploited and eventually everyone will be ruined. This is termed the tragedy of commons.3 The NHS is a common resource. A patient acts rationally in seeking an expensive treatment that produces a benefit (even if small), because the cost falls almost entirely on others. But the NHS cannot support overexploitation indefinitely. It already spends £10.3bn (€15bn; $19bn) a year on drugs, and costs are rising rapidly.w1 One way to avoid overexploitation is to appoint a guardian to administer the commons. NICE plays this role but faces many challenges.
NICE examines the value of drugs only when it is invited to do so by the Department of Health and Welsh Assembly, so it may never review some important aspects of therapeutics. The process is complex, involves many interested parties (stakeholders), and takes around two years to complete.w2 The delays between the initial suggestion and the final determination can allow patterns of treatment of uncertain cost effectiveness to become established. The longer the patterns persist, the harder they become to reverse. The use of verteporfin in treating age related macular degeneration is a case in point.4 Further treatments were licensed during the time NICE took to appraise the evidence and rendered its guidance outdated.w3
NICE clearly appreciates the need to provide thorough analysis in good time. It now aims to appraise a drug for a specific indication in about 40 weeks. Appraisal may begin before regulatory approval has been granted, so that NICE's views are known within a few months of the drug being licensed. Advice is expected soon on the use of trastuzumab in early breast cancer, for example, even though the manufacturer did not apply for a licence until February 2006.
Role of big pharma
Pharmaceutical companies spend over $800m to develop and license a new drug,5 so they are understandably interested in the financial returns. An obvious insurance against loss is to produce a drug that is clearly beneficial and whose costs are proportionate. Even if a new drug offers only slight benefits at an inflated cost, a company may be able to persuade doctors or patients of its value by marketing. It can then hope for sales independent of guidance from NICE or other expert bodies. Companies can also sell drugs by relabelling normal phenomena such as male pattern baldness and social phobia as diseases that then merit drug treatment.6
The promotion of trastuzumab (Herceptin) shows how companies raise awareness of new drugs. Marketing was muted at first, as specialists learnt of pilot studies in patients with advanced disease.w4 A paper reporting an early trial in a wide circulation general medical journal amplified the signal.w5 The paper was accompanied by an enthusiastic press release,w6 encouraging medical correspondents to spread the word to the wider public.w7 These strategies mean people become attuned to the company message long before licensing.
The manufacturer or sponsor of a product has a key role in the appraisal process through the submission of data to NICE's evidence review group. The review group no doubt critically assesses the industry submissions. However, published company data contain inherent biases. Industry funded randomised trials are more likely to favour the treatment under test,7 and authors are more likely to be positive in their conclusions if a randomised trial is funded by a for-profit organisation.8 Published economic evaluations of cancer drugs sponsored by drug companies are also less likely to report unfavourable qualitative conclusions.9
One reason for this bias could be that companies fund the trials that they think are most likely to give positive results, but other explanations are possible. Conscious or unconscious optimism of those most interested in a treatment's success could colour the judgment of study authors. Certainly, expected cost-utility can vary greatly from one study to another. In the case of photodynamic therapy, industry estimated the cost of averting two years of blindness as £70 000, whereas academic analysis suggested costs between £150 000 and £300 000.w8
Companies also harness the media to support their views in battles with NICE. If NICE gives restrictive guidance, drug companies issue press releases decrying the judgment.w9-w11 Pfizer described the recent decision not to recommend inhaled insulin as “perverse,”w10 and Link Pharmaceuticals claimed NICE was denying “potentially life-prolonging treatments” to patients with brain tumours.w9
Patients and patient groups
Patients may wish to have treatments that NICE has yet to pronounce on or has recommended against. Patient groups share a common interest with drug companies in promoting access to specific treatments that others will pay for, and so they provide a route for companies to influence the perceptions of their drugs at a distance.10 As Jo Spink, a public relations professional, explains, “Patients are a powerful force and can highlight the clinical, societal and quality of life benefits of a treatment far more passionately than any press release ever could.”w12
Patient groups have been described as conduits for drug companies to promote their products11 and as “ground troops” to be used to lobby governments for increased access to new drugs.w13 A recent survey found that 76% of patient groups based in the European Union received support from drug companies, although how much they received was unclear.w14 Groups campaigning for NICE to approve specific drugs often have declared corporate relations with pharmaceutical companies.w15-w17 Financial support from a drug company may not compromise a patient group's independence, but companies are not motivated by altruism, and the House of Commons Health Select Committee has advised that measures be taken to limit the influence of industry on patient groups.11
Patient groups often object strongly if NICE's preliminary recommendations are restrictive. Vocal campaigns preceded changes in advice on treatment for Alzheimer's diseasew18-w20 and osteoporosisw21 and patient groups are seeking to overturn NICE's preliminary, unfavourable, view of inhaled insulin.w22
The influence of patients and patient groups can be seen even before a NICE appraisal has been commissioned. Trastuzumab has yet to be licensed for use in early breast cancer, its longer term efficacy is uncertain, and its propensity to cause myocardial damage in patients who have taken anthracyclines is worrying. The cost of treating a patient with trastuzumab in the United Kingdom is about the same as the average annual income. Patients have campaigned successfully to place trastuzumab at the forefront of public and political consciousness.w23 w24 This has led to the feeling that the licensing and NICE approval of trastuzumab for early stage cancer is a foregone conclusion. This is a sentiment expressed by the chief executive officer for Breakthrough Cancer Care: “The fast track appraisal of drugs by NICE is to be welcomed but we need to ensure that once approved, guidance is implemented fully and that cancer patients receive the drugs recommended.”w25
Dying patients denied life saving drugs make compelling copy.w26 Better still, when heartless bureaucrats are persuaded to change their minds under media pressure, the media can take the credit for saving the patient.w27 The Manchester Evening News even received an award for its campaign over trastuzumab.w28 Stories of wonder drugs, carefully encapsulated in press releases, can make useful copy even in the absence of human interest. And careful deployment of celebrities by patient groups can ensure media exposure for both.w29 Newspaper columnists such as Clare Rayner and Thomas Stuttaford have also launched populist attacks on NICE rationing.w30-w33
Many members of parliament are ready to espouse popular causes such as supporting patients with cancer or crippling diseases. Tabling questions in parliament is an effective way to do this. The House of Commons question book lists 84 questions on trastuzumab, 56 on donepezil, and 188 on interferon beta, but only two on ciprofloxacin, and none on penicillin or prednisolone.w34
Ministerial interventions may complicate matters. While he was a health minister, Stephen Ladyman commented on preliminary guidance that recommended the withdrawal of drugs for Alzheimer's disease. He stated that “they [NICE] have to look at the wider impact of this decision. It may well be that once they have looked at the extra evidence, they will come to a different decision.”w35 Even before trastuzumab has been licensed for treatment of early breast cancer, Patricia Hewitt, the secretary of state for health, has stated that primary care trusts “should not refuse to fund Herceptin [trastuzumab] solely on the grounds of its cost.”w36 She has instituted genotyping of all breast cancers, so fostering expectations of trastuzumab treatment directed at the 20% of tumours that are HER-2 positive.12
The work of NICE is critically important to the rational distribution of NHS funds
Patients and drug companies seek to influence the decisions both during and after appraisal
The government risks undermining NICE by making decisions before guidance is published
Appraisal of treatments needs to be insulated from external pressure
Published NICE guidelines may not be implemented as intended.13 Abacus International found that 12 of 28 NICE appraisals were underimplemented and four were overimplemented.14 Since many NICE decisions are based on fine and perhaps generous judgments of cost-utility, overimplementation is likely to be expensive.15
Opportunities exist for distortion long before the NICE appraisal process begins. They continue after the process has begun and persist after preliminary findings are published. Even when the final NICE determination recommends abandonment of or restriction in the use of a treatment, interested parties may make determined efforts to by-pass NICE's ruling. Audits suggest that many of NICE's recommendations are not implemented completely.
It is easy to see why patients, whose illnesses provide strong motivation to obtain a share of NHS resources, wish their voices to be heard. It is also clear that companies, which have to make a profit, will seek what advantage they can. Less easily understood is the way that politicians seem to undermine NICE. The government set up NICE as an independent body, and so ministers should be absolutely scrupulous in allowing it to function independently. By making, or appearing to make, decisions before NICE has pronounced and for reasons of political expediency, some politicians may be subverting the process. It will be a test of NICE's resolve to see that treatments are cost effective when they analyse the data on trastuzumab in early breast cancer.
NICE should ensure that the funding available for drugs in the NHS is spent in a way that best serves patients. But in the ideal world the appraisal process would be insulated, and seen to be insulated, from external financial, political, and emotional pressures.
Competing interests REF was senior registrar to the current chair of NICE.
Funding SEMcD was supported by the Antidote Trust Fund of the Sandwell and West Birmingham Hospitals NHS Trust.
Contributors and sources REF is a consultant physician and clinical pharmacologist and chairs the drug and therapeutics committee of a large NHS Trust. SEMcD is a health scientist interested in prescribing and adverse drug reactions. Both believe that rational therapeutics is important. REF wrote the first draft and will act as guarantor. SEMcD participated in the research for the paper and contributed to the writing.
References w1-w36 are on bmj.com