Introduction

Around the world it is commonly assumed that clinical practice guidelines, systematic reviews and the scientific literature are dependable and credible sources of information about the efficacy and effectiveness of therapeutic products. Health practitioners and consumers expect that these are reliable sources of up-to-date information about treatment options, and policy makers rely on them to guide important healthcare decisions. Yet all these publications are subject to many influences that can threaten their independence¹, including:

• the suitability of clinical research funding
  
• the limitations of conventional clinical trials
  
• the reliability of the evidence base
  
• the competing interests of guideline developers and other experts involved in guideline development.
  

Being situated in the private sector, the pharmaceutical industry is driven by commercial principles. It responds to incentives to develop new drugs, which are often more expensive than established therapies. Because of its size and financial power, the industry has a major influence on the nature of the research agenda. Companies decide which questions will be researched and then they design and fund trials to ensure their new drugs are seen in the best possible light. As a result, the pharmaceutical industry is exerting considerable influence over the development of clinical practice.

In recent years, there has been a trend in the industry to modify society’s perception of disease. It actively campaigns for expanded definitions of diseases, which often results in normal processes being labelled as pathological ones. Despite limited evidence, this is happening in many areas, notably for menopause, blood glucose, blood cholesterol and other lipids, blood pressure and bone density. By expanding indications for treatment, the prescribing rate for the newest drugs increases, as does the likelihood of adverse effects.

Appropriate treatment often involves non-drug therapies or the use of more established drugs. Because there are very few sources of funding for research on either non-drug therapies or therapies that are unlikely to produce an economic benefit for pharmaceutical companies, the evidence base is becoming increasingly biased towards the use of new drug therapies.

Flaws in the design of randomised controlled trials can magnify benefits and obscure the adverse reactions of a drug. Common deficiencies and biases identified in the literature include:

• abuse of placebo – new drugs tend to be compared with placebo rather than an established comparator, which would discern whether the new drug offers an advantage over standard treatment
  
• non-inferiority trials – such trials are designed to show a drug is not worse than standard therapy. There is usually no requirement to demonstrate the superiority of a new drug.
  
• selection of a comparator – ideally a new drug should be tested against the best standard, at the best dose and duration of treatment. There is no requirement to assess the drug's safety and effectiveness against the current gold standard treatment.
  
• surrogate endpoints – it is common for the efficacy of a drug to be measured using an indicator that may (or may not) reflect an advantage, rather than measuring a clinical effect relevant to the patient
  
• composite endpoints – the trend for different endpoints to be grouped and analysed together makes it difficult to determine clinical significance
  
• exclusion of relevant populations – new drugs are frequently tested on men rather than the population that will be using the drug (e.g. women, older people, children, people with comorbid conditions). Consequently, there is often little or no information about which groups of people do better or worse with treatment, and little research on the impact of drugs on patients’ function and lived experiences.
  

Analysis of trial data shows a direct link between the funding source and outcomes of a study, with industry-sponsored trials being approximately two-and-a-half times more likely to favour a drug than publicly funded trials.

Publication bias is another important barrier to independence. Positive results that show an intervention works are more likely to be published than negative results. Trials that fail to show benefit may never be published and seem to be underrepresented in the published literature. In some cases trial results are published selectively, with positive results prioritised, rather than the full set of measured outcomes. These systematic biases are a challenge for all those engaged in health, and in particular to groups producing evidence-based guidelines, evidence reviews and health policy.

In recent years questions have been raised in the international therapeutic community about both the reliability of clinical drug trial data and the management of real or perceived conflicts of interest in guideline development groups. An editorial in the British Medical Journal, for example, indicates that the pharmaceutical industry has repeatedly withheld and misreported data on the safety and efficacy of a range of widely used drugs, limiting treatment benefits, endangering lives and wasting public money.²

Scientific data are always interpreted in the context of existing societal views and influences. People relying on trial results as a basis for developing guidelines need to be aware of the spectrum of influences that can bias the collection, analysis and interpretation of data. Conflicts of interest can occur when a commitment, goal or value that arises from a professional or social role is perceived to unduly influence the independence of data or their interpretation. Such conflicts include:

• pecuniary interests
  
• academic or professional interests that might accrue from publications, awards, media, professional attention or kudos
  
• personal or religious beliefs, or experience of a health-related condition (either personally or a close family member or friend affected).
  

Guideline developers around the world understand the need to create reliable and trustworthy guidelines according to the highest standards and they are becoming increasingly mindful of threats to independence.

Tools have been developed to appraise guidelines and to establish whether the views of the funding body have influenced the content of guidelines and whether the competing interests of guideline development group members have been appropriately managed.³ In the USA, the Institute of Medicine published a report outlining concerns about the quality of the processes used to develop clinical practice guidelines, the limitations in the scientific evidence base on which clinical practice guidelines rely, the lack of transparency in development groups’ methodologies, and the management of conflicts of interest among guideline development group members and funders.¹

In 2012 the Guidelines International Network published recommendations for minimum standards for quality guidelines⁴ and in the same year the National Health and Medical Research Council published a report on managing conflicts of interest in the development of guidelines. Despite the introduction of standards promoting conflict of interest policies, a growing number of reviews of clinical practice guidelines conducted in the last 10 years have revealed that the majority of guideline developers do not in fact publicly disclose any of their authors’ conflicts of interest. In 2009 the Medical Journal of Australia published a review⁵ that analysed 313 Australian clinical practice guidelines. The review found that almost 80% of the guidelines analysed included no information on the conflicts of interest of the members of their development groups. Similar findings have been published in relation to US and European guidelines.

The central aim of Therapeutic Guidelines is to promote the quality use of medicines through the development, publication and distribution of independent therapeutic guidelines. Independence is a critical issue for Therapeutic Guidelines so a number of strategies and policies have been implemented throughout the organisation and its operations to ensure the independence of the guidelines it develops.

The Independence Forum brought together national and international experts, ethicists, guideline developers, health professionals, clinical researchers and medical writers to identify threats to independence and discuss measures that could be taken to identify and counteract them.