|Home | About | Journals | Submit | Contact Us | Français|
Accurate assessment of the individual risk for psychotic disorders has great value. We need to determine the reliability and rate of conversion to a psychotic disorder in clinical samples before we can recommend a risk syndrome diagnosis for general practice. The assessment of risk carries its own risks, including stigmatization and inappropriate treatment, potentially leading to adverse outcomes.
The burden of psychotic disorders is considerable. They are associated with significant disability, morbidity, and mortality, resulting in drastically reduced life expectancy.1 While these facts are sobering, there is hope that psychotic disorders can be prevented.2 Most psychotic disorders are now viewed as an evolving pathology of the human brain, shaped by genetic risk factors and environmental stressors. The long trajectory of brain maturation and the remarkable plasticity in response to injury have created reasonable hope that we can intervene and prevent mental illness. To achieve such an ambitious goal, we need reliable methods of early detection and accurate risk assessment.
How can we get there? First, we need to define the timeline of a psychotic disorder. This requires diagnostic criteria to separate a risk stage from the disorder. Second, we need to identify which risk factors can predict the subsequent emergence of a psychotic disorder. Finally, we need to assess risk factors reliably and in time for intervention and prevention.
Let us assume that we have the ability to accurately record all signs and symptoms of psychosis throughout the life of an individual. We will define psychosis as (a) reality distortion (delusion and hallucination), (b) disorganized thought/language, and (c) markedly abnormal psychomotor behavior (such as stupor and mutism). We also have a dimensional rating scale that allows us to reliably assess the degree of psychosis. Each observation of psychosis can then be recorded in a graph, such as the one shown in figure 1A. On the y-axis we rate the degree of psychosis (on a scale from 0 to 100, with 20 being defined as the diagnostic threshold) and on the x-axis we map out the development of the illness. Then we can distinguish at least 4 stages.
With the luxury of continuous observation, the graph in figure 1A can be recorded online, as events happen. Alternatively, it can be constructed retrospectively, assuming total recall for all periods of life. Neither is realistic and we have to rely on the retrospective epidemiological study of cohorts diagnosed with psychotic disorders. Such studies have provided us with promising leads for a prospective assessment of risk for a psychotic disorder.
The lifetime prevalence of psychotic disorders is 3%.6 Various genetic and environmental factors contribute to this baseline risk in the population and can accumulate, thereby significantly elevating the risk for a given individual. Table 1 lists some of the risk factors for psychosis and distinguishes (1) risks that may exert their effect at any time, ie, before or after the onset of the illness, (2) risks that occur early on in life, ie, during the asymptomatic risk stage, and (3) risks that develop in close proximity to the transition into disease, ie, during the symptomatic risk stage.
Emerging technologies such as structural and functional neuroimaging are being explored to delineate a temporal profile of brain changes during the symptomatic (or even asymptomatic) risk state. For example, subtle deficits in cortical volume10 and an abnormally elevated release of dopamine11 have been associated with the later emergence of a psychotic disorder.
The period of subthreshold psychosis before the first episode of a psychotic disorder is often referred to as the prodrome. The prodrome is, by definition, the nascent stage of a disorder (hence the term latent schizophrenia).
In contrast, a risk syndrome is not necessarily linked to a disorder. The value of a risk syndrome increases with the accuracy in predicting future outcomes (eg, metabolic syndrome leading to diabetes or cardiovascular disease; mild cognitive impairment leading to dementia), but the conversion to disease is, by definition, less than 100%.
In short, a prodrome is part and parcel of the disorder (see figure 1), a risk syndrome is not. A prodrome can be diagnosed only after the disorder has declared itself. In contrast, a risk syndrome can be diagnosed prospectively.
How can clinicians and researchers currently assess the risk for a psychotic disorder? First, they can review demographic and historical information to establish baseline and proximal risk factors. Examples include a detailed family history, genetic testing for mutations and risk alleles, and a thorough developmental history, starting with events in utero. Second, they can assess the cumulative risk of several factors. A striking example is the combination of early cannabis use and certain risk alleles, leading to a significantly increased risk for schizophrenia.12 Third, they can look for changes of sensation, thinking, affect, and volition as proximal risk factors. These mental status changes are typically minor variants of fully developed psychotic symptoms.
Several research groups have developed global risk factors, relying primarily on proximal clinical risk syndromes.13 For example, the Structured Interview for Prodromal Syndromes includes a family history questionnaire and rates 4 domains (positive, negative, disorganization, and general symptoms) of psychopathology with the 19-item Scale of Prodromal Symptoms.14 Others have focused on basic symptoms, defined as the earliest subjectively experienced symptoms of psychosis, including changes in volition and affect, peculiar changes of thinking, speech, and perception, and genuine psychotic symptoms.9
Is the current state of risk assessment for psychotic disorders mature enough to warrant inclusion in diagnostic systems such as the Diagnostic and Statistical Manual of Mental Disorders and International Classification of Diseases? What evidence do we want to see?
First, the assessment has to be feasible in clinical practice. For example, we cannot screen the general population and need to define a target population, such as help-seeking individuals. Second, the diagnosis has to be reliable, not only in the research setting but also in clinical practice. This will require education about the subtle differences between subthreshold and diagnostic psychotic symptoms. Third, the diagnosis has to be valid. For example, does the risk syndrome predict course and outcome, response to treatment, and biomarkers such as cortical volume or dopamine release? Fourth, the social, legal, and medical consequences of diagnosing a risk syndrome need to be evaluated. How do we protect individuals, who will not progress to a psychotic disorder, from bias and stigma? Will the diagnosis of a risk syndrome for psychotic disorders be sufficient for the Federal Drug Administration (FDA) to approve therapeutic interventions (similar to the attempts to gain FDA approval for conditions such as mild cognitive impairment and metabolic syndrome)? Finally, there are several ethical implications: What is the value of risk assessment without proper options for intervention? What to do when a person does not want to know the individual risk?
Risk assessment has the potential to lift the burden of psychotic disorders but we need to be careful not to make matters worse.
The author is a member of the DSM-V Work Group on Psychotic Disorders. The opinions expressed here do not necessarily reflect the consensus of the DSM-V Work Group or Task Force.