Multimodal Approach May Improve Ability to Predict Transition to Psychosis
Abstract
Thirty percent of people classified at ultra-high risk of psychosis experience a transition, but simple clinical assessments can raise the predictive value to 70 percent.
Despite best efforts to identify patients at greatest risk of developing psychosis, studies suggest that some 70 percent of people identified as ultra-high risk (UHR) for psychosis never develop a psychotic disorder.
A study published in September in Translational Psychiatry suggests that using a multimodal diagnostic approach that combines a person’s historical risk factors (such as family history of psychosis), clinical assessments (such as Positive and Negative Syndrome Scale), and blood biomarkers (including measures of oxidative stress and fatty acids) may be able to better predict UHR patients most likely to develop psychosis.
Researchers have been searching for ways to improve their ability to identify individuals at risk of developing psychosis since UHR criteria were first introduced in the 1990s. A patient is considered to be UHR if he or she experiences one or more of the following: attenuated positive symptoms (APS); positive symptoms at full psychotic intensity for brief, limited periods known as brief intermittent psychotic symptoms (BLIPS); and a combination of genetic risk accompanied by functional decline, known as genetic risk and deterioration (GRD).
“Currently, all patients in the ultra-high-risk group are considered to have a similar chance of a future psychotic episode,” said Scott Clark, Ph.D., M.B.B.S., a clinical investigator of psychiatry at the University of Adelaide in Australia.
But as a meta-analysis published earlier this year in JAMA Psychiatry suggests, current UHR criteria may fail to fully capture the heterogeneity of the group and each subgroup may carry different levels of risk. The meta-analysis, which compiled 33 studies including over 4,000 individuals, found that the presence of BLIPS conferred a higher psychosis risk than the other groups, whereas GRD by itself conferred no greater risk than someone not designated UHR.
“What we need is a more reliable—and flexible—method of prediction to tailor care appropriately for the people who need it most,” Clark said.
Clark led a multinational team that developed a staged diagnostic tool that combines basic bedside clinical assessments with a person’s risk history and their plasma concentrations of the fatty acids omega-3 and nervonic acid (both critical for healthy brain function). In a retrospective assessment of 40 UHR individuals (who were enrolled in the placebo group of a year-long trial examining of the effect of omega-3 fatty acid supplementation on psychosis risk), this new approach demonstrated greatly improved accuracy.
Of the 40 patients tested, the multimodal diagnosis identified eight of the 11 people who transitioned to psychosis within the year while also finding one false positive among the 29 patients who did not transition. This yielded a sensitivity of 73 percent, a specificity of 96 percent, and overall accuracy of about 70 percent.
The model was also able to classify 77 percent of patients as low or high risk using only the historical and clinical information.
“A staged approach to risk assessment would then be the most efficient, using fatty-acid markers only when the probability following history and clinical assessment is between 0.1 and 0.9, that is, 23 percent of participants in the current study,” the authors wrote. “Extending this staged approach, resource-intensive neuroimaging or electrophysiology could be reserved for cases that remain at intermediate risk based on clinical and blood biomarker assessments.”
In addition, because the model was designed to be flexible, other diagnostic tools in development, such as visual tests or language processing software, could one day be incorporated to provide a more comprehensive patient stratification and increase the accuracy of the model even further.
Clark said his group now plans to team up with Orygen Youth Health in Melbourne, Australia, to test this model in an independent population of at-risk individuals, and if those results are positive, he is hopeful that he and colleagues will be able to initiate a large, prospective clinical trial.
Clark’s study was supported by the James and Diana Ramsay Foundation. ■
“Prediction of Transition From Ultra-High Risk to First-Episode Psychosis Using a Probabilistic Model Combining History, ClinicalAssess-ment, and Fatty-Acid Biomarker” can be accessed here.
