News

New way to define disease is needed to reduce overdiagnosis, experts agree

BMJ 2016; 354 doi: https://doi.org/10.1136/bmj.i5082 (Published 21 September 2016) Cite this as: BMJ 2016;354:i5082
  1. Susan Mayor
  1. London

A new process is needed for defining diseases that takes account of the potential risks of overdiagnosis and overtreatment as well as the benefits of appropriate diagnosis and care, researchers, policymakers, and consumer groups agreed at an international congress this week.

“The problem at the moment is that we don’t have an internationally agreed process,” said Paul Glasziou, professor of evidence based practice at Bond University, Robina, Queensland. “Disease definitions are developed on a very ad hoc basis, often by guideline panels making recommendations about tests and treatments who, along the way, incidentally change the definition of a disease,” he told the Preventing Overdiagnosis conference, co-sponsored by The BMJ and held on 20-22 September in Barcelona (www.preventingoverdiagnosis.net).

“We have a clear, internationally agreed process on what evidence is needed to recommend a particular treatment. But there has been a neglect of the fundamental issue that comes before that, which is the definition of disease.”

He explained that the problem with expanding definitions of disease so as to include milder cases was that people might suffer more harms than benefits from treatments. “And people are labelled as having a disease, potentially causing psychological and social problems,” he said, adding that expanding numbers of patients gave health services more people to manage.

“This is a real problem globally,” said Alexandra Barratt, professor of public health at the University of Sydney, who chaired the debate. “It affects countries in Asia, Latin America, and Africa and not just Europe,” she warned, noting that over the past 10 years definitions have widened in a range of diseases, including attention-deficit/hyperactivity disorder, Alzheimer’s disease, hypercholesterolaemia, and myocardial infarction, often driven by the development of new tests or biomarkers.

“This has expanded the pool of people diagnosed with these conditions, but we don’t know if it improves outcomes for patients. And that’s what really matters,” said Barratt.

Mónica Cavagna Guerrero, health expert with the Health and Food Department of the Organisation of Consumers and Users, an independent global consumer group based in Spain, said, “Many of us agree that expanding disease definitions correlates to expansion of drug markets and increase of drug industries’ benefits.

“From a society’s point of view, it’s essential to be certain that widening the limits of a disease is appropriate for the greater good and not to satisfy private or spurious interests.”

DSM is “compromised”

Glasziou proposed that the process for defining disease must consider the balance of harms in relation to benefits. In addition, he proposed that the people defining diseases needed to change. “There is an odd conflict here that is not generally picked up. Most guideline panels are worried about conflicts of interest for particular products, such as drugs or devices. For a disease definition there is a different conflict: the size of your disease territory. Interested specialists get excited about their condition and want to expand their territory—and that aligns with their commercial interest as well.”

He said that the Diagnostic and Statistical Manual of Mental Disorders (DSM), although it uses a systematic process to review evidence to define mental illnesses, was compromised as the panel of psychiatrists on the task force had an intrinsic conflict “because it’s their own disease territory.”

Allen Frances, professor emeritus and former chair of psychiatry at Duke University, Durham, North Carolina, and chair of the DSM-IV task force, told The BMJ, “Disease definition suffers from the ‘fox guarding henhouse syndrome.’ Among the thousands of diagnostic experts I have worked with, not a single one has ever suggested narrowing the definition of his pet disorder. For the vast majority, the bias stems from intellectual, not financial, conflicts of interest, but this is even more difficult to control because it is unconscious, sincere, heartfelt, and not transparent.”

Frances said that Australia had a good system. “Diseases are defined by a group of experts in disease definition that consults with experts in each specialty without being ruled by them.” He added, “Patients should be part of the process but also shouldn’t rule it.”

Cavagna Guerrero suggested that groups that define diseases should follow the Council of Europe’s resolution that excludes people with a conflict of interest from “sensitive decision-making processes.”

“Independent consumers and patients should be part of the panels, as their contribution will surely enrich the discussions—just like in the direct reporting of adverse effects of drugs,” she proposed. “But the information and training for them to participate should be provided by independent sources.”

Glasziou concluded that the process of defining a disease should consider the potential harms of expanding current definitions, including the risks of overtreatment and labelling, and clarify whom to include on panels, taking account of conflicts of interest among professionals working in a disease area and the need to involve well informed consumers. He is a member of the Overdiagnosis Working Group, set up by the Guidelines International Network in 2014, which is working to develop an internationally acceptable set of rules and guidance on the requirements for modifying definitions of disease.

View Abstract

Sign in

Log in through your institution

Free trial

Register for a free trial to thebmj.com to receive unlimited access to all content on thebmj.com for 14 days.
Sign up for a free trial

Subscribe