Psychologist David Elkins had modest ambitions for his petition. He and his colleagues were worried that proposed changes to an influential handbook of mental disorders could classify normal behaviours as psychological conditions, potentially leading to inappropriate treatments. So they laid out their concerns in an open letter, co-sponsored by five divisions of the American Psychological Association in Washington DC. "I thought, 'Well, maybe we'll get a couple or maybe 30 signatures'," says Elkins, an emeritus professor at Pepperdine University in Malibu, California.

But the letter, posted online on 22 October (go.nature.com/uhmvqq), touched a nerve. Within 10 days more than 2,800 people had signed it, many identifying themselves as mental-health professionals.

The petition targets proposed revisions to the Diagnostic and Statistical Manual of Mental Disorders (DSM), a tome used by psychiatrists, psychologists, counsellors and others worldwide to diagnose mental maladies and set research agendas. The American Psychiatric Association, based in Arlington, Virginia, plans to publish a new edition of the manual, DSM-5, in 2013. The association has declined to comment on Elkins's petition.

Psychiatrist Allen Frances, who was the chief architect of DSM-IV and is an outspoken critic of its successor, has dubbed the open letter a "buyer's revolt". "I think the petition is the last best hope to influence the DSM-5 from the outside," says Frances, an emeritus professor at Duke University School of Medicine in Durham, North Carolina.

Elkins's petition is not the first to raise concerns that the DSM-5 proposals could overreach. In June, the British Psychological Society, based in Leicester, issued a critique that highlighted, for example, the proposed addition of 'attenuated psychosis syndrome'. The society argued that this could be used "to stigmatize eccentric people".

There should be a black box warning about how child bipolar disorder is being overdiagnosed. ,

Elkins and his colleagues have complained about other proposals, such as the elimination of a 'bereavement exclusion' in the diagnosis of major depression. The previous edition of the manual recommended that the condition not be diagnosed in people grieving the death of a loved one within the previous two months. The revisions shorten this to two weeks, a change that troubles psychiatrist Ramin Mojtabai of the Johns Hopkins Bloomberg School of Public Health in Baltimore, Maryland. Categorizing these patients as having depression could boost the use of medications when psycho­therapy may be the better treatment, he says.

Efforts to tighten loose definitions of attention deficit and hyperactivity disorder (ADHD) and bipolar disorder in children have also proved controversial. In response to worries that inexact criteria may have contributed to a surge in diagnoses of these conditions since the 1990s, the DSM-5 task force has proposed a syndrome called 'disruptive mood dysregulation disorder', which would provide an alternative to labelling a child as bipolar or having ADHD. But Frances says that is not enough. "There should be a black box warning about how child bipolar disorder is being overdiagnosed," he says. "Instead, they've created a new disorder."

Field trials of the proposed DSM-5 criteria have been completed and investigators plan to publish the results. Helena Kraemer, a statistician and emeritus professor at Stanford University School of Medicine in Palo Alto, California, who is on the DSM-5 committee, says that results from trials of some criteria will indicate whether they generate more frequent diagnoses.

But Mojtabai cautions that trial results may not reflect what will happen when DSM-5 is published. "Any trial is artificial," he says. "The clinicians in these trials have intensive training, but people who will use this manual in clinical practice will not receive that level of instruction."