The latest version of the American Psychiatric Association’s (APA) controversial diagnostic code – “the DSM-5” – continues the check-list approach used in previous editions. To receive a specific diagnosis, a patient must exhibit a minimum number of symptoms in different categories. One problem – this implies someone either has a mental illness or they don’t.
To avoid missing people who ought to be diagnosed, over time the criteria for many conditions have expanded, and nowhere is this more apparent than in the case of post traumatic stress disorder (PTSD). Indeed, in their new analysis of the latest expanded diagnostic criteria for PTSD, Isaac Galatzer-Levy and Richard Bryant calculate that there are now 636,120 ways to be diagnosed with PTSD based on all the possible combinations of symptoms that would fulfil a diagnosis for this condition.
First defined as a distinct disorder in 1980, for many years PTSD was diagnosed based on a patient exhibiting a sufficient number of various symptoms in three categories: reexperiencing symptoms (e.g. flashbacks); avoidance and numbing symptoms (e.g. diminished interest in activities); and arousal symptoms (e.g. insomnia). For the latest version of the DSM, a new symptom category was introduced: alterations in mood and cognition (e.g. increased shame). This means a diagnosis of PTSD is now met according to the patient having a minimum of 8 of 19 possible symptoms across four categories (or criteria), so long as these appear after they witnessed or experienced an event involving actual or threatened harm.
Putting these various diagnostic permutations into the statistical grinder, Galatzer-Levy and Bryant arrive at their figure of 636,120 ways to be diagnosed with PTSD. This compares to 79,794 ways based on DSM-IV – the previous version of the APA’s diagnostic code. The net has not widened in this fashion for all conditions – for example the criteria for panic disorder have tightened (there were 54,698 “ways” to be diagnosed with panic disorder in DSM-IV, compared with 23,442 ways in DSM-5).
Galatzer-Levy and Bryant believe the PTSD scenario exemplifies the problem with using a set of pre-defined criteria to identify whether a person has a mental health problem or not. In the pursuit of increasing diagnostic reliability, the code loses its meaning in a fog of heterogeneity. The authors fear that despite the increasing diagnostic complexity, people who need help are still missed, while others continue to be misdiagnosed. They believe this could be the reason why the research into risk factors for PTSD, and into the effectiveness of interventions for the condition, tends to produce such highly varied results.
The ideal situation, according to Galatzer-Levy and Bryant, is for our understanding and description of mental health problems to be based on empirical data – in this case about how people respond to stress and trauma. They say a useful approach is to use statistical techniques that reveal the varieties of ways that people are affected over time – a complexity that is missed by simple symptom check-lists. For instance, Galatzer-Levy and Bryant say there are at least three patterns in the way people respond to stressful events – some cope well and show only short-lived symptoms; others struggle at first but recover with time; while a third group continue struggling with chronic symptoms.
“Such an empirical approach for identifying behavioural patterns both in clinical and nonclinical contexts is nascent,” the authors conclude. “A great deal of work is necessary to identify and understand common outcomes of disparate, potentially traumatic, and common stressful life events.”
Isaac R. Galatzer-Levy and Richard A. Bryant (2013). 636,120 Ways to Have Posttraumatic Stress Disorder. Perspectives on Psychological Science