After more than a decade in the works, the long-awaited fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) arrived last month to little fanfare. Potential users of the DSM had already gotten sneak previews because its publisher, the American Psychiatric Association, for the first time in the manual’s history, had shared draft versions to invite public comment. Much of it, on issues ranging from grief to autism, was bitingly negative. For Alzheimer's and other forms of dementia, the update from the previous version prompted calmer discussions. What are the changes relevant to Alzheimer’s, and how will it affect diagnosis? While the new manual’s section on cognitive disorders differs significantly from the DSM-IV, clinicians may barely change how they diagnose AD and other cognitive disorders, experts suggest. That is the case in no small part because DSM criteria are rarely used outside of psychiatric practices to diagnose Alzheimer's.

A Disorder by Any Other Name
In the DSM-5, what used to be called “Delirium, Dementia, Amnestic, and Other Cognitive Disorders” is now named “Neurocognitive Disorders.” The new classification was introduced to distinguish diseases such as AD from psychiatric disorders that have cognitive impairment as a symptom rather than a defining feature, such as schizophrenia or depression. Many changes to this section stem from psychiatrists’ desire to harmonize the field of cognitive disorders overall and to bring the DSM in line with new diagnostic guidelines for dementia, such as those proposed by the National Institute on Aging and the Alzheimer’s Association in 2011 (see ARF related news story), said Ron Petersen of the Mayo Clinic in Rochester, Minnesota. Petersen was in the working group that revised the Neurocognitive Disorders section.

This section includes three main diagnoses: mild neurocognitive disorder (NCD), major NCD, and delirium. The motivation for condensing all diagnoses into these three categories was to formalize a two-step method of diagnosis that is already commonplace in clinical practice, said Perminder Sachdev at the University of New South Wales, Randwick, Australia, another member of the neurocognitive disorders working group. In that process, clinicians first diagnose cognitive impairment and then look for an underlying etiology that explains the decline, for example, Alzheimer's disease, Parkinson's, or frontotemporal dementia. The DSM-V does not use the term “Alzheimer’s disease” alone as a diagnosis.

The mild NCD diagnosis is new for this revision and possibly the largest departure from the DSM-IV, which made no distinction between mild and full-blown dementia. The criteria for mild NCD are virtually identical to those for "mild cognitive impairment (MCI) due to Alzheimer's disease" as outlined in the NIA/AA diagnostic guidelines, Petersen and Sachdev said (see Albert et al., 2011). In the DSM-5, decline in one or more cognitive domains, while patients maintain their independence, indicates mild NCD. The manual urges clinicians to assess memory loss, attention, executive function, language, and visuospatial, visuoperceptual, or social cognition for this condition.

In the DSM-IV, criteria for dementia of the Alzheimer's type required memory impairment and one or more of the following: aphasia (language problems), apraxia (impaired motor ability), agnosia (failure to recognize known objects), or deterioration in executive function. In contrast, memory loss is not a strict condition of major NCD. This diagnosis is defined as impairment in two or more cognitive domains that is accompanied by a loss of independence. However, to assign a specific etiology of Alzheimer's to either mild or major NCD requires “clear evidence of decline in memory and learning,” according to the manual.

The Neurocognitive Disorders section downplays the term “dementia,” which was a formal diagnosis in the DSM-IV. The APA now replaces that with the term "major NCD," but the manual recognizes dementia as an acceptable alternative. Many psychiatrists are no longer comfortable with the term "dementia," Sachdev said. “The concept of dementia has been Alzheimer-ized. People tend to equate it with Alzheimer's disease, even though we know that AD accounts for less than half of all cases of dementia,” he said. For example, when dealing with cognitive disorders due to HIV, traumatic brain injury, vascular disorders, and frontotemporal disorders, psychiatrists and patients avoid the word dementia.

Some in the Alzheimer’s field find this name change problematic. “The adoption of the politically correct term 'neurocognitive disorder' was a great disservice to patients and to the field in general. It introduces yet another terminology at a time when public understanding and acceptance of the widely used terms have been increasing,” said David Knopman, also of the Mayo Clinic in Rochester. Constantine Lyketsos of Johns Hopkins University in Baltimore, Maryland, thinks the loss of the word dementia could have major ramifications. “If you look around the literature, the dementia diagnosis as defined by the DSM-IV is widespread,” he said. “I think it's going to be hard to stay with the DSM at this point, because it's too much of a discontinuity from what existed before, both clinically and research-wise.”

The DSM is widely used by psychiatrists but sees little use in other clinical fields, suggest some experts. According to Lyketsos, psychiatry in the U.S. is tied to DSM criteria in large part because they map to International Classification of Diseases codes. The ICD is the World Health Organization’s standardized tool for disease diagnoses, which is commonly used by insurance companies. Representatives of the American Neurological Association would not comment on whether their members use the DSM to diagnose AD. Andrew Budson, a neurologist at Boston University, Massachusetts, said that neurologists primarily use NIA/AA criteria and eschew the DSM. Primary care physicians use cognitive tests such as the Mini-Mental Status Examination (MMSE) to assess possible AD, said Christine Laine, an internist at Jefferson Medical College in Philadelphia, Pennsylvania. According to the American Academy of Family Physicians, the DSM-5 revisions are unlikely to affect family medicine practices, as AAFP members rarely use the manual. Like primary care physicians, family practitioners rely on screening tests such as the MMSE to assess cognitive decline.

Like most of the DSM-5, the neurocognitive disorders section requires no biological criteria for diagnosis. It relies exclusively on symptoms. “I think conditions like mild NCD over the years will greatly benefit from the application of biomarkers, but we are not there yet,” Petersen said. “We felt that the field is not quite ready for incorporation of these biomarkers into the diagnostic criteria themselves, but we fully anticipate that will happen.”

The lack of a biological basis for the diagnoses in the DSM-5 led Thomas Insel, director of the National Institute of Mental Health (NIMH), to publicly distance his agency from the manual. Insel's announcement was reported as a snub of the DSM (see articles in ScienceInsider and The New York Times).

Others in the field are moving forward with biomarkers. Tertiary care settings already use varying imaging or biomarkers informally as part of a broad diagnostic assessment. The European Medicines Agency welcomed their use in clinical trials for AD (see ARF related news story), and the FDA approved the first amyloid imaging PET tracer for diagnostic purposes in April 2012 (see ARF related news story). An expert taskforce recommended this technology be used to rule out other causes of dementia in uncertain cases and spelled out general guidelines for its appropriate application in the clinic (see ARF related news story).

Does Mild NCD Blur With Normal Aging?
Allen Frances, professor emeritus at Duke University School of Medicine in Durham, North Carolina, expressed concern that inclusion of mild NCD, like the criteria for MCI due to AD in the NIA/AA guidelines, will lead to false diagnoses of people who are unlikely to progress to AD or other cognitive disorders. “There is no way to accurately identify on clinical grounds who is going to go on to Alzheimer's,” Frances said. “Probably more than half of people who get the [mild NCD] diagnosis are no more likely than the general population to get Alzheimer's.” Frances claims these early diagnoses will worry people unnecessarily while providing no benefit. Frances chaired the DSM-IV taskforce and has led critiques of the DSM-5 with blog posts, e-mail campaigns to reporters, and public commentary.

The debate over false positives when diagnosing early-stage disease is nothing new (see comments in ARF related news story). “Are we creating a new entity now that is essentially just normal aging?” Petersen asked. “The answer is no. Mild NCD is distinct from normal aging.”

“There is some concern as to how this will actually work out in clinical practice,” Sachdev said. Since AD exists along a continuum, the mild NCD diagnosis could “encroach on either side,” he said. That is, some people with normal age-related cognitive loss could be overdiagnosed with mild NCD, while others with early AD could be underdiagnosed as being mild NCD. Because of the uncertainty inherent in diagnosing AD at its very early stages, the DSM-5 guidelines state that mild NCD should only be given a possible, rather than probable, etiology of AD unless the patient has a known genetic predisposition for the disease. In longitudinal AD research, all people with an MCI diagnosis who have episodic memory loss and brain amyloid by PET progress to full AD dementia within three years (see, e.g., ARF related news story).

The DSM working group had to accept that MCI is already being used in the field, and patients are visiting their doctors early enough to warrant the diagnosis, Sachdev said. Lyketsos agrees. “The reality is it's not the DSM leading the field, it's the field leading the DSM. For good or bad, the MCI concept is finding its way into clinical settings,” he said.

As for the DSM's future, there may be no formal compendium called "DSM-6," Petersen said. That's not because the "psychiatrist’s bible" is likely to fade out. Rather, David Kupfer of the University of Pittsburgh, who led the taskforce that oversaw the revision, has promised that the DSM-5 will be a living document, with sections evolving as research into a given disorder has progressed enough to warrant new criteria. For example, validation and incorporation of biomarkers for AD into clinical use could prompt a DSM-5.1 specifically for the neurocognitive disorders section.—Rachel Tompa.

Rachel Tompa is a science writer in Seattle, Washington.


No Available Comments

Make a Comment

To make a comment you must login or register.


News Citations

  1. CTAD: Turning the Ship Around Toward Early Trials
  2. FDA Approves Amyvid for Clinical Use
  3. HAI—Amyloid Imaging in the Clinic: New Guidelines and Data
  4. Noisy Response Greets Revised Diagnostic Criteria for AD
  5. Longitudinal Data Stand Out at AD/PD Imaging Satellite

Paper Citations

  1. . The diagnosis of mild cognitive impairment due to Alzheimer's disease: recommendations from the National Institute on Aging-Alzheimer's Association workgroups on diagnostic guidelines for Alzheimer's disease. Alzheimers Dement. 2011 May;7(3):270-9. Epub 2011 Apr 21 PubMed.

Other Citations

  1. ARF related news story

External Citations

  1. publicly distance
  2. ScienceInsider
  3. The New York Times

Further Reading