Article Review: Cannon, T. D., Yu, C., Addington, J., Bearden, C. E., Cadenhead, K. S., Cornblatt, B. A.,…Kattan, M. W. (2016). An individualized risk calculator for research in prodromal psychosis.

 

Psychosis has been described as a terrifying experience that has been associated with shame, guilt, and humiliation (National Alliance on Mental Illness [NAMI], 2011).  As indicated by NAMI (2011) delay in assessment, identification, diagnosis, and treatment for psychosis is a public health crisis, for which efforts of prevention and early intervention are now being emphasized throughout communities. Therefore, understanding the onset of psychosis is necessary.

 

For the majority of individuals there is a period prior to the onset of psychosis during which individuals begin to exhibit changes in beliefs, thoughts, and perceptions (Cannon et al., 2016). Though not a diagnosis, according to the Center for the Assessment and Prevention of Prodromal States (CAPPS; 2011) this period of time is the prodromal period and could last from a couple of days to years. It is during this time that the subtle changes are said to represent “attenuated forms of delusions, formal thought disorder and hallucinations” (Cannon et al., 2016, p. 1).  Individuals with such an onset or prodromal psychosis are designated high-risk and over a 2-year period, about 20% to 35% of these individuals go on to develop full psychotic symptoms (Cannon et al., 2016, pg. 1). As a result, Cannon et al. (2016) created a risk calculator to calculate the probability of conversion to psychosis among individuals identified with prodromal psychosis.

 

Cannon et al. (2016) emphasized that past research has investigated risk factors for conversion (e.g., demographic factors, symptoms), yielding high predictability and specificity, yet low sensitivity for the identification of conversion.  Their current research focused on the ability to scale risk during initial patient contact by using easily accessible clinical, cognitive, and demographic variables.  The study utilized data from the second phase of the North American Prodrome Longitudinal Study from 2008 to 2013.  Participants in this study participated in the Structured Interview for Prodromal Syndromes (SIPS) and the Structured Clinical Interview (Diagnostic and Statistical Manual- [DSM] IV).  Individuals with substance dependence, neurological disorders, an estimated IQ below 70, or past diagnosis of a psychotic disorder were excluded from the study. Follow-up evaluations were schedule for every 6-months for 2 years. Participants were identified as having high-risk syndromes (attenuated psychotic symptoms syndrome, brief intermittent psychotic symptom syndrome, and familial risk and deterioration syndrome).  The final cohort consisted of 596 participants, who were followed up to the point of conversion to psychosis or up to 2 years.

 

By assessing the importance of each predictor variable, Cannon et al. (2016) created a risk calculator by “using time-to-event proportional hazards regression” (p. 2). The authors identified eight predictor variables apriori:  age; SIPS items P1 and P2; Brief Assessment of Cognition in Schizophrenia (BACS), symbol coding raw score; Hopkins Verbal Learning Test-Revised, scores summed; stressful life events; family history of psychosis; a decline in functioning as shown on the Global Functioning Social Scale; and trauma history.  More specifically, the SIPS items P1 and P2 assess unusual thought content and suspiciousness, which, per Cannon et al. (2016), for high-risk individuals have shown to be strongly predictive of psychosis.  Additionally, the literature has demonstrated that slower processing speed, lower verbal learning, and memory functioning are predictive of conversion (Cannon et al., 2016).  A decline in social functioning prior to conversion, childhood traumas, and stressful life events have also been shown to be predictive of psychosis in high-risk individuals. The Research Interview Life Events Scale and the Childhood Trauma and Abuse Scale were used to assess traumatic experiences. Finally, family history of psychosis was included though the authors indicated the literature does not support this factor as a “robust predictor” of conversion (p. 3). Regardless it was included due to the elevated risk compared to individuals with no familial history.

 

The results indicated that within the 2-year period, 84 individuals in the sample converted to psychosis. The mean time to conversion was 7.3 months. A 16% probability of conversion was reported. The overall model’s C-index was 0.71.  Overall, the authors concluded that high levels of suspiciousness and unusual thought, decline in social functioning, lower verbal learning and memory performance, slower speeds of processing, and a younger age at baseline created a higher risk for conversion to psychosis. The variables of stressful life events, trauma, and family history were not predictive of conversion.

 

Given that this study used an established database, generalizability could be a concern. The authors argue for “predictive inference” (p. 4), due to the community based service centers used in the establishment of the database. Still, for clinical utility, each respective client should be assessed regarding relative fit to the sample.  Additionally, the authors report that the output for the risk calculator is without a confidence interval.  Thus, individuals are provided a percentage for conversion risk, which is taken without consideration of error or a range of values. As such, there is concern of patient distress if the risk calculator yields a relatively high conversion probability. The authors note the benefit of utilizing the risk calculator for identifying research participants (e.g., meeting a particular threshold), utilizing this risk calculator to communicate risk relative to treatment, and to identify the cost-benefit ratio related to treatment options.  The risk-calculator did not include biological factors, and with future research may need to be amended to accommodate other factors that are predictive of conversion.

 

Overall this tool is related to early identification of risk, yet it appears to be more so applicable to research studies.  The authors further note that a decision tree has been installed to ensure that individuals who use the risk calculator are in fact professionals that have conducted a SIPS interview and the client has a diagnosis of a prodromal risk syndrome.  This risk calculator appears to be a practical tool, but clinical utility may be equivocal due to concerns of reporting risk of conversion to clients and not providing a probability of remission.  The risk calculator is aligned with early intervention. Knowing the probability of conversion may help encourage clients to engage in treatment and help clinicians or researchers recommend treatment options best aligned to meet the client’s needs.

 

Dannie S. Harris
WKPIC Doctoral Intern

 

References
Cannon, T. D., Yu, C., Addington, J., Bearden, C. E., Cadenhead, K. S., Cornblatt, B. A.,…Kattan, M. W. (2016). An individualized risk calculator for research in prodromal psychosis. American Journal of Psychiatry. Advance online publication. http://dx.doi.org/10.1176/appi.ajp.2016.15070890

 

Center for the Assessment and Prevention of Prodromal States. (2011). What is the Prodrome? Retrieved from https://www.semel.ucla.edu/capps/what-prodrome

 

National Alliance on Mental Illness. (2011). First episode: Psychosis, results from a 2011 NAMI survey. Retrieved from http://www.nami.org/psychosis/report

 

This entry was posted in Blog, Continuing Education, Current Interns, Mental Health and Wellness and tagged , , . Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *

Before you post, please prove you are sentient.

What is 7 times 3?