The onset of psychosis can be devastating and typically occurs in late adolescence or early adulthood (1). This is a critical developmental stage for relationships, education and employment, and illness onset can threaten the potential for a productive, inclusive, and fulfilling adult life. Early detection programs such as the Scandinavian early treatment and intervention in psychosis (TIPS) study have endeavored to improve long term outcomes associated with psychosis by detecting and intervening early in the course of illness (2, 3). The TIPS early detection program showed significantly improved functional outcomes 10 years post diagnosis (4). The success of such programs has fueled interest in examining the feasibility of detecting psychosis at an even earlier stage, thereby further reducing disability (secondary prevention), or even averting manifestation of psychosis itself (primary prevention) (5, 6).

Prior to a first episode of psychosis (FEP), most individuals experience an extended period of sub-clinical psychotic-like symptoms, known as the ultra high risk (UHR) state (7), clinical high risk for psychosis (CHR-P) (8), or at risk mental state (ARMS) for psychosis (9). When assessed, a UHR state is usually defined as meeting criteria for one or more of three syndromes: the attenuated positive symptoms (APS) syndrome, the brief limited intermittent psychotic symptoms (BLIPS) syndrome, or the genetic risk and/or the deterioration (GRD) syndrome (10). These syndromes are assessed by validated measures, such as the Comprehensive Assessment of At-Risk Mental States (CAARMS) (11) or the Structured Interview for Prodromal Syndromes (SIPS) (7). Moreover, in the fifth version of the Diagnostic and Statistical Manual of Mental Disorders (DSM 5), APS was included as a condition for further study (12).

The UHR state is associated with high rates of experienced trauma, victimization such as bullying, substance use, comorbid mental disorders, suicidal ideation, and self-harm (1, 8, 13). While UHR individuals are more likely to be male (58%) (1), female gender is a predictor of poorer outcomes in terms of general psychiatric symptoms and functioning at long term follow up (14). Further, UHR is associated with a 22% risk of developing a psychotic disorder within 3 years (1). It should also be noted that the BLIPS syndrome overlaps diagnostically with the DSM-5 category of brief psychotic episodes (15). Up to 60% of UHR individuals who do not develop psychosis continue to display subthreshold psychotic symptoms, or meet criteria for other mental health disorders (16–18), and impairments in social functioning are as common as in other mental disorders (1).

Each of the three UHR syndromes includes declining psychosocial functioning, and often leads to the person seeking help (19). However, help seeking appears to be delayed with a recent comprehensive review of 42 meta-analyses reporting a mean duration of nearly 2 years of APS prior to accessing specialized services (1). The vast majority of UHR individuals (95%) are neither referred to UHR services nor assessed during the UHR phase (20, 21). It is likely that the overall burden of psychosis risk in secondary mental health care is mostly unknown. Clinician referrals of UHR individuals to specialized mental health services depend largely on subjective judgments of psychosis risk (22). While semi-structured interviews for psychosis prediction such as the SIPS (7) have an excellent overall prognostic performance, and the sensitivity is high (95%), the specificity is low (47%) (23). Thus, UHR criteria are still unable to correctly predict transition in all individuals who will later develop psychosis (22). There will always be challenges when attempting to detect cases of a low incidence illness stemming from populations with low or moderate risk, or, in this case, probably with no or few UHR symptoms (24, 25). This limits the predictive utility of the UHR concept outside clinical samples that have undergone risk enrichment (1).

The Prevention of Psychosis study (POP) (26) was launched in 2012 and builds on the 23 years of experience of the TIPS study (3, 27) in which intensive awareness and information campaigns were combined with low threshold detection and assessment. The awareness and information campaigns were key to the success of TIPS (2). In POP, the original psychosis specific information campaigns were adapted to encompass a broader range of symptoms and psychological distress consistent with the UHR criteria.

The aim of this study was to investigate whether a systematic early detection program (POP) modeled on the successful early detection of psychosis program of TIPS (27), would improve the detection rate of persons at UHR for psychosis. We formulated the following research questions:

The overall study design was a prospective 2 year follow-up study of UHR individuals. UHR individuals in the catchment area of Stavanger University Hospital, Norway were recruited through intensified case detection within secondary mental health clinical services and in the general population. Consenting participants received tailored individualized treatment within the secondary mental health services and structured follow-up assessments.

The main finding of this study was that we were able to identify 141 UHR help-seeking individuals, indicating a yearly incidence of seven per 100,000 people, compared to our prediction of 90 per 100,000 people. It is important to note that despite the implementation of information campaigns for detection, our study included help seeking individuals rather than being restricted to a broad population level campaign. Thus we detected around 8% of the expected cases compared with 4–5% of cases reported in a recent meta review (1). It is expected that the incidence of UHR would be higher in the risk enriched environment of mental health care compared to the general community. Even though one of the targeted information strategies was aimed at providing education for clinicians in the specialized mental health services on signs and symptoms of the UHR phase, the study was not able to reach the expected UHR incidence level. Study findings reflect previous UHR research, showing a low referral rate of potential UHR individuals to UHR detection centers, since many individuals are already being treated in other secondary mental health care services (20). There is a need for improved tools for case detection such as the proposed individualized transdiagnostic risk calculator (21, 22).

Where reducing DUP is achieved through targeted information about specific symptoms such as hallucinations or delusions, this is currently not possible in UHR due to the non-specificity of early emerging symptoms. However, the treatment provided in our study was not aimed solely at preventing conversion to psychosis but also at improving psychosocial functioning and family relations through CBT, psychotherapy and family work. Medication (anti-anxiety, antidepressants and antipsychotic medication) was available according to the treatment algorithm if needed and the clinicians recommended such use with 47% of the sample reported using such medication during the 2 year follow up.

Our sample was young, mostly adolescents, with a mean age of 16.9 years whereas a comprehensive review of 42 studies showed a mean age of UHR/CHR participants of 20.6 years (1). While we had equal representation of males and females, the review showed a predominance of males. Depression and anxiety were the most common mental health problems for our sample as indicated by diagnostic category and the Calgary Depression Scale for Schizophrenia (CDSS) scores (40), averaging above cutoff at baseline.

The 2 year conversion rate of 20% is consistent with other studies (1, 46). It should be noted that individuals who developed a FEP were immediately offered treatment for psychotic symptoms and thus had a DUP of 0 weeks. The SIPS negative symptoms subscale emerged as the symptom dimension significantly associated with conversion to psychosis in all analyses. This warrants some discussion, as conversion is defined by positive, not negative, symptoms. However, social withdrawal is a central feature in negative symptoms. In youth, healthy psychosocial development commonly occurs in social arenas, such as school or college, sports clubs, friend groups, and romantic involvements (47). Emerging psychological problems of a severe nature hamper the social functioning essential for relationships with peers (48). Hence, we argue that when mental health problems reach a level at which the interaction with peers becomes compromised, this may be indicative of more severe psychological distress, and in some cases, psychosis.

Finally we are not able to state that the POP early detection program resulted in an increase in the number of detected and successfully treated UHR individuals to a level needed to influence the incidence of FEP, as the treated incidence of FEP was stable in catchment area across the time-period. This will be a focus for further studies.

Our study capitalized on the strengths of the TIPS longitudinal cohort study to implement one of the first large scale information campaigns targeting young people at UHR for psychosis. Contrary to many other UHR studies, we targeted both community and clinical populations. Our study provides important learning for all UHR researchers regarding the process of recruitment and retention of this young and vulnerable population. The primary limitations of our study concern generalisability and representativeness. A small sample size and a relatively high attrition rate limit generalisability. Attrition does however appear to be at random, and the sample can be assumed to be representative with regard to baseline characteristics.

In conclusion, the early detection strategies employed were not able to detect UHR cases to the extent predicted by the population statistics and further case enrichment appears to be needed to improve this. Enhanced screening focusing on subclinical psychotic experiences and social anhedonia in the adolescent general population might bring more UHR individuals into clinical services (49, 50). As negative symptoms were significant predictors of transition to psychosis, a greater focus on negative or negative-like symptoms and social withdrawal seems warranted. The use of social media may to an extent mask social withdrawal, and we thus propose to expand the use of these platforms in UHR detection work. Further, we propose combining this with the use of enhanced screening tools like the individualized transdiagnostic risk calculator, possibly with an addition of social withdrawal (21, 22). Finally, we propose that social media and other awareness campaigns be focused on the general population, whereas enhanced tools for detection be used in risk enriched environments, predominantly specialized adolescent mental health services. This could improve UHR detection rates and enhance help-seeking behavior in the target population.

The datasets presented in this article are not readily available because according to Norwegian law, data sharing requires approvals from the Regional Committees for Medical and Health Research Ethics, and from the Data Protection Officer at Stavanger University Hospital, on the basis of specific research proposals. Requests to access the datasets should be directed to inge.joa@sus.no.

The studies involving human participants were reviewed and approved by National Committee for Medical and Health Research Ethics (2009/949). Participation in the study was based on written informed consent. Parents or legal guardians gave informed consent for patients younger than 16 years of age.

IJ, WH, and JJ contributed to conception and design of the study as well as data collection. JL and MW also contributed to the data collection. WH conducted the statistical analysis. IJ wrote the first draft of the manuscript. All authors contributed to manuscript revision, read, and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.