What Is an Attenuated Psychotic Symptom? On the Importance of the Context

Abstract

Attenuated psychotic symptoms (APS) are the key criteria to identify the individuals at enhanced risk of developing psychotic disorders. Competing clinicians-rated or self-rated psychometric instruments can also be used to detect APS, which makes it difficult to interpret their actual clinical significance. This article summarizes the empirical differences between the clinicians-rated and self-rated interviews and explores the impact of the context (referral pathways, settings, and assessment procedures) on the clinical significance of the APS.

The Ultra High-Risk State for Psychosis

Over the past two decades, clinical criteria have been developed to identify the individuals at ultra high-risk (UHR) of developing the first episode of psychosis within a relatively short period of time (eg, 2–3 years¹). UHR individuals have become a major focus of academic research, and the detection of UHR individuals is increasingly being recognized as an important component of clinical services for early psychosis intervention² (eg, NICE guidelines³, recent NHS England AWT EIP guidelines,⁴ and DSM-5 diagnostic manual). Research in this field has the potential to shed light on the development of major psychotic disorders and to alter their course.⁵

Recognition of Attenuated Psychotic Symptoms: At the Core of Psychosis Prevention

The diagnosis of the UHR state requires the presence of one or more of the following clinical features: attenuated psychotic symptoms (APS) and/or brief limited and intermittent psychotic symptoms (BLIPS); and/or genetic risk and deterioration (GRD) syndrome (for details see⁶). However, recent evidence indicated that the GRD subgroup is not actually associated with an increased risk for psychosis,⁶ while the BLIPS subgroup also includes individuals who are already presenting symptoms of psychotic disorders.^7,8 Therefore, the APS subgroup, which represents the vast majority of UHR cases (85%),⁶ can be considered as the core high-risk subgroup.

Psychometric Assessment of Attenuated Psychotic Symptoms

Clinicians-Rated Psychometric Instruments

The APS features can be assessed trough semi-structured psychometric interviews.⁹ The original instrument, the Comprehensive Assessment of At-Risk Mental State (CAARMS 12/2006¹⁰) is a detailed psychopathological interview (7 subscales with 177 questions, footnote to table 1) addressing a wide array of signs and symptoms of the potential risk of psychosis, among which APS are of primary importance. Clinicians are required to investigate and rate the severity, frequency, onset time, and overall distress of each component of the APS. Since the CAARMS is semi-structured, CAARMS interviews are conducted with a fairly open framework, which allows focused, conversational, two-way communication. The interviewer follows a specific guideline but is able to follow topical trajectories in the conversation or that may stray from the guide when it seems appropriate or to ask additional questions to better elucidate the clinical meaning of each answer. Such diagnostic assessment is time consuming, as it may encompass multiple sessions (up to 2 hours for the diagnostic part²) and several post hoc clinicians-based and team-based discussions to ensure clinical validity.² More importantly, it also requires specific psychopathological training¹¹ in order to ensure an adequate inter-rater reliability. The CAARMS assessment is one of the most complex psychopathological investigations in clinical psychiatry. Recognizing this, the NICE clinical guidelines CG178 1.2.2.1 have recommended that the CAARMS assessment should be carried out by “a consultant psychiatrist or a trained specialist with experience”¹² in the field.

Table 1.

Qualitative Comparison of Item Similarities Between the Semi-Structured Clinicians-Rated Psychometric Instruments (CAARMS 12/2006) and the Structured Self-Reported Psychometric Instruments for the assessment of Attenuated Psychotic Symptoms

CAARMS Version 12/200610	CAPE-P1515
[Positive symptoms: Unusual thought content] Have you felt that people were trying to give you messages? What is it like? How did it start?	Do you ever feel as if people seem to drop hints about you or say things with a double meaning?
[Positive symptoms: Unusual thought content] Do you feel that others, or the world, have changed in some way?	Do you ever feel as if some people are not what they seem to be?
[Positive symptoms: Non-bizarre ideas] Has anybody been giving you a hard time or trying to hurt you?	Do you ever feel as if you are being persecuted in some way?
[Positive symptoms: Non-bizarre ideas] Has anybody been giving you a hard time or trying to hurt you?	Do you ever feel as if there is a conspiracy against you?
[Positive symptoms: Unusual thought content] Have you felt that someone, or something, outside yourself has been controlling your thoughts, feelings, actions or urges? Have you had feelings or impulses that don’t seem to come from yourself?	Do you ever feel as if electrical devices such as computers can influence the way you think?
[Positive symptoms: Non-bizarre ideas] Do you feel like people have been talking about you, laughing at you, or watching you? What is it like? How do you know this?	Do you ever feel that people look at you oddly because of your appearance?
[Positive symptoms: Unusual thought content] Have you felt that ideas or thoughts are being taken out of your head? How does that happen?	Do you ever feel as if the thoughts in your head are being taken away from you?
[Positive symptoms: Unusual thought content] Have you felt that ideas or thoughts that are not your own have been put into your head? How do you know they are not your own? Where do they come from?	Do you ever feel as if the thoughts in your head are not your own?
[Positive symptoms: Unusual thought content] Are your thoughts broadcast so that other people know what you are thinking?	Have your thoughts ever been so vivid that you were worried other people would hear them?
[Positive symptoms: Perceptual abnormalities] Do you ever hear things that other people seem not to (such as sounds or voices)?	Do you ever hear your own thoughts being echoed back to you?
[Positive symptoms: Unusual thought content] Have you felt that someone, or something, outside yourself has been controlling your thoughts, feelings, actions or urges? Have you had feelings or impulses that don’t seem to come from yourself?	Do you ever feel as if you are under the control of some force or power other than yourself?
[Positive symptoms: Perceptual abnormalities] Do you ever hear things that may not really be there? Do you ever hear things that other people seem not to (such as sounds or voices)? What do you hear? At the time you hear these things, how real do they seem? Do you realise they are not real at the time, or only later?	Do you ever hear voices when you are alone?
	Do you ever hear voices talking to each other when you are alone?
[Positive symptoms: Unusual thought content] Do you feel puzzled by anything? Do familiar surroundings feel strange?	Do you ever feel as if a double has taken the place of a family member, friend or acquaintance?
[Positive symptoms: Perceptual abnormalities] Do you have visions, or see things that may not really be there? Do you ever see things that others can’t, or don’t seem to? What do you see? At the time that you see these things, how real do they seem? Do you realise they are not real at the time, or only later?	Do you ever see objects, people or animals that other people cannot see?

Note: CAARMS 12/2006 subscales:

• Positive symptoms: Unusual thought content, Non-bizarre ideas, Perceptual abnormalities, Disorganized speech

• Cognitive change attention/concentration: Subjective experience, Observed cognitive change

• Emotional disturbance: Subjective emotional disturbance, Observed blunted affect, Observed inappropriate affect

• Negative symptoms: Alogia, Avolition/apathy, Anhedonia

• Behavioural change: Social isolation, Impaired role function, Disorganising/odd/stigmatizing behavior, Aggression/dangerous behavior

• Motor/physical changes: Subjective complaints of impaired motor functioning, Informant reported or observed changes in motor functioning, Subjective complaints of impaired bodily sensation, Subjective complaints of impaired autonomic functioning

• General psychopathology: Mania, Depression, Suicidality and self harm, Mood swings/lability, Anxiety, Obsessive compulsive disorder symptoms, Dissociative symptoms, Impaired tolerance to normal stress

Total number of items/questions in the CAARMS 12/2006: 176 (follow-up questions for severity, frequency/duration, distress not accounted). A subset of the items that are matching the CAPE-15 items is reported here.

Self-Reported Psychometric Instruments

As conducting semi-structured clinical interviews, such as the CAARMS, is highly demanding, the researchers have developed several structured self-reported screening tools that are aimed at quick identification of APS. Among the others (eg, the PRIME Screening [PS], the Prodromal Questionnaire [PQ], and the Psychosis-Like Symptom Interview [PLSI], for a systematic review of all available screening instruments see¹³), the Community Assessment of Psychic Experiences [CAPE] is one of the most widely used tools and also the one with the most robust prospective research. The original 42-item version takes 10–12 minutes to complete and includes items for self-reported APS (self-reported APS are also termed “psychotic-like experiences,” PLE¹⁴), items for self-reported negative symptoms, and items on depressive experiences.¹³ Based on the direct questions, the person is required to self-evaluate the presence of APS (ranging from “never” to “nearly always”), as well as the extent of associated distress (ranging from “not distressed” to “very distressed”). Recently, the CAPE-42 has been further reduced to a 15-item screener (CAPE-15),¹⁵ which takes only 5 minutes to complete,¹³ and selectively focuses on the persecutory ideation, perceptual abnormalities, and bizarre experiences. Other self-reported instruments can take even less time to be completed (eg, 1 minute and 40 seconds on average for the PS¹⁶).

2 Hours vs 5 Minutes Assessment for Attenuated Psychotic Symptoms: Does it Make any Difference?

Despite the different target populations, for which they were originally developed, clinicians-rated and self-reported instruments for the assessment of APS have been interchangeably and inconsistently^17,18 used to identify the individuals at risk of developing psychosis. However, not all the APS are the same, and some important differences should be taken into account.¹⁷

Psychometric Validity

The clinical validity of semi-structured clinician-rated interviews has been confirmed at the meta-analytical levels, including the large samples collected worldwide. These meta-analyses have indicated a good ability to rule out psychosis and a modest ability to rule it in, with an overall excellent area under the curve of 0.90 (95% confidence interval: 0.87–0.93), which is comparable to the other tests in preventive medicine.¹⁹ Conversely, the majority of self-reported structured instruments are poorly validated (ie, sensitivity, specificity, positive and negative predictive values, and inter-rater reliability are unknown¹³) and lack the appropriate assessment of the relevant psychometric properties and acceptance by patients and clinicians²⁰ in suitable clinical samples.

Prognostic Validity

APS detected by trained clinicians are associated with up to 29% risk of developing into psychosis over the following 2 years,^1,21 while self-reported APS (measured with the Psychiatric Epidemiology Research Interview) are associated with just a 0.5% risk of developing psychosis over the following 24 years.²² There is less data and quality is poor regarding prognostic accuracy of other self-reported APS instruments for the longitudinal prediction of psychosis. The available small studies employing the PS and PQ lacked the external validation and were biased toward overoptimistic estimates¹⁶ like those produced by post hoc manipulation of cutoffs.²³ Furthermore, some of these studies have been conducted in samples that had already been filtered through semi-structured interviews.¹⁶ Indeed, larger studies employing the PQ revealed poor positive predictive values of 5% at more than 3-year follow-up, with a rate of false positives to true positives that was approximately 20-fold.²⁴ A large study employing the PLSI concluded that positive predictive values for predicting psychotic disorders were too low to offer real potential for clinical use.²⁵

How the Context Impacts the Clinical Significance of Attenuated Psychotic Symptoms

Sampling Biases

The profound difference between the risk of psychosis yielded by clinicians-rated and self-reported interviews of APS is partially due to the fact that the former are usually employed in selected individuals who are referred to high-risk services on the suspicion of psychosis risk,^26,27 while the latter are usually used in unselected individuals of the general population.²⁸ This phenomenon has been described as risk enrichment in help-seeking populations. A recent study has confirmed that the individuals referred to high-risk services have a pretest (ie, before the UHR assessment is performed) 15% risk (at 38 months) of developing psychosis, a value which is significantly higher than the 0.1% 38-month risk expected in the general population,²⁷ which confirms substantial risk enrichment.

The Setting and the Clinical Meaning of Attenuated Psychotic Symptoms

Earlier studies showed that prevalence of psychosis in community samples is strongly influenced by the methods of assessments and diagnosis.²⁹ A more recent study has directly challenged the assumption that self-reported APS are a reliable estimate of clinicians-rated APS, within the same pool of individuals accessing high-risk services (n = 123), and thus controlling for sampling biases and differential risk enrichment.¹⁸ The authors found that level of agreement between the two instruments was poor and insufficient, even if additional qualifiers, such as level of distress, preoccupation, and conviction, were considered.¹⁸ In particular, self-rated APS overestimated the prevalence of clinician-rated psychotic symptoms several fold.¹⁸ The disagreement was evident for delusion-like experiences and was maximum for hallucination-like experiences.¹⁸ The study concluded that self-reported APS cannot be used to assess the pathological meaningfulness of these experiences as evaluated by clinicians.¹⁸

Recruitment Strategies Impact Clinical Significance of APS

Likewise, there is evidence that the type of recruitment strategies adopted to select individuals for UHR assessment can enrich the clinical significance of APS.²⁷ The highest pretest risk of psychosis is observed when the individuals are exclusively referred from mental health professionals or services, “because of difficulties that had arisen in the diagnostic procedure.”³⁰ Such samples have become increasingly enriched with psychosis-prone individuals, who were passing through several mental health service filters and talking with psychiatrists about their problem. Conversely, when the individuals were recruited from the general population or were self-referred, the clinical significance of their APS and the associated risk of developing psychosis was the lowest.³¹ These data confirm that the pathological meaningfulness of APS accumulates if the clinicians filter them. A recent study confirmed these hypotheses.²⁶

A Practical Example

The above arguments indicate that the context (both in terms of filtered/unfiltered referral pathways, setting and assessment procedures) impacts the clinical significance of APS. Despite that the CAARMS and the CAPE share some qualitative item similarities (see table 1 for a summary), this does not exclude the important differences. For instance, self-answering to the questions like (table 1)

• “Do you ever hear voices when you are alone?”

• “Do you ever feel as if you are being persecuted in some way?”

• “Do you ever feel that people look at you oddly because of your appearance?”

at the beginning of class during the first week of the college course³² does not have the same clinical significance, as answering to these same questions when they are put forward by a psychiatrist in a clinical service. In other words, the clinical significance of the responses to these questions cannot be “predefined”³³ by the virtue of their similar verbal formulation. APS cannot be considered as atomic features that become individuated “in themselves,”, independent of their developmental³⁴ and experiential context.³³

Contextual Recognition of Attenuated Psychosis Symptoms

What does it mean that the attenuated psychotic symptom changes its clinical significance depending on the contextual circumstances in which it occurs? How is it possible? Before answering that question, we need to address the issue and understand what is the meaning of the term “symptom” in psychiatry.³⁵ The notion of symptom was imported to psychiatry after the medicalization of madness in the beginning of 19th century. However, whereas the clinical validity of a symptom in somatic medicine is strongly associated with its referential function toward its root cause in a malfunctioning biological substrate (eg, coughing may indicate a lung disease, an acute abdominal pain may indicate appendicitis), this is not the case in psychiatry. In contemporary psychiatry, first there is no robust knowledge on the specific pathophysiology of the vast majority of symptoms (eg, lack of clinically valid biomarkers³⁶) and, second, there is even a lack of theoretical account on the nature of symptoms and on the processes of their formation. Thus, the symptom in psychiatry is considered at the implicit common sense level as an unproblematic thing-like object, independent of other symptoms and easily described by the so-called operational definition. The symptom is envisaged in analogy with a ripe fruit hanging in the patient’s consciousness and only waiting for a gentle push from the structured interview questions in order to fall down and come into complete view to trickle down to its corresponding diagnostic criterion.³⁷ This is a fundamentally mistaken epistemological position. In psychiatry, the operations do not comprise autonomous things-like semi-permanent objects, which can be defined in themselves without any contextual dimension. Mental phenomena are essentially different from physical objects. As insightfully observed by Descartes, mental phenomena have no spatial extension (a perception of a tall building is not itself tall). Much later, Henri Bergson drove the attention to the fact that the use of the terms, such as intensity and magnitude applied to mental phenomena, is, in fact, practical, but an abusive transfer of spatial metaphors, valid only for the external world. Psychiatric symptoms are not semi-permanent well-delimited objects but consist of certain aspects of the stream of consciousness that is reported by the patient. Indeed, they contain affective and cognitive elements that are always relative and dependent on other experiences and manifest the underlying changes in the structure of consciousness. Finally, they are subjected to the processes of conceptualization and verbalization, both embedded in the interpersonal context of the psychiatric interview. To conclude, the behavioristic ideal of the impersonal access to a preexisting “objective mental reality” is a dangerous illusion.

In other words, the articulation of a symptom is not entirely independent of the way in which it is elicited, eg, through the introspective attention required in self-assessment questionnaires or through the dialogic framework of the semi-structured psychiatric interview. In the latter case, the validity of symptom’s description and the nature of the symptom are objectively arrived at by the intersubjective agreement with a trained and experienced psychiatrist, who recognizes psychologically meaningful interrelations among the patient’s descriptions and thereby discerns their clinical value. Further to this point, it is also possible that patients themselves may talk to doctors differently in a clinical semi-structured context than they do to self-reports. To take an instance from the table 1 concerning the experience of self-reference (eg, “Do you ever feel that people look at you oddly because of your appearance?), there are countless possibilities of false positive and negative responses. A patient may misunderstand the question or think that the formulation does not apply to him because his own experience of other’s exaggerated interest is not really bound to his appearance. He may believe that he is actually the center of the universe and therefore it is not surprising or odd to be looked at, or he may simply feel self-reference for no reason (which is typically characteristic of incipient psychosis) or because others consider him as being outside of human kind. Finally, the patient may answer negatively because he thinks that others avoid looking at him because they are polite and will not acknowledge his inhuman presentation. No wonder that the same interview question elicits responses with diverse clinical importance. Clearly, while the potential for tangential answers or misunderstandings is relatively uncontrolled in self-assessment structured tools, clinician-based semi-structured interviews enable a more articulated investigation and understanding of the meaning and context of the answers, with optimal tuning between the interviewer and the patient.¹⁸ Such context-sensitive understanding is what guarantees a more precise recognition of prototypical APS presentations, as they emerge in the clinical dialogue.

Conclusions

The clinical significance of APS depends on the way (and the context) in which it is assessed. Discerning the validity and nature of APS presupposes the availability of extensive contextual information that only trained and experienced clinicians can elicit through semi-structured interviews. Self-reported APS detected with structured instruments, not filtered by contextual information, have minimal direct clinical significance for the prediction of psychosis onset. On the contrary, self-reported instruments may have some clinical impact, when used in sequential testing to enrich samples undergoing subsequent semi-structured interviews.^38,39 These issues should be carefully considered by researchers and clinicians, who are working in the field of psychosis prevention.

Funding

The study was funded in part by a 2014 NARSAD Young Investigator Award to P.F.-P.

Conflict of Interest

None.