Stigma
Stigma inflicted by syndrome labels
Yung et al. (Reference Yung, Wood, Malla, Nelson, McGorry and Shah2020) disagree with our argument that stigma may be imposed by labels such as ultra high risk (UHR). Referring to the investigation by Kim et al. (Reference Kim, Polari, Melville, Moller, Kim, Amminger and Nelson2017), Yung et al. write that ‘a recent study found that the terms ARMS, UHR or ‘Attenuated Psychotic Symptoms’ were not thought to be stigmatizing by ARMS individuals and only a minority thought these names should be changed.’ If these labels are not stigmatizing to individuals in at risk mental state services (ARMS), this would indeed be a strong argument against our claim. However, this is not the case. Kim et al. report that between 36.7% and 42% of ARMS patients found these labels stigmatizing. Although it is true that a minority thought that these names needed to be changed, that minority was substantial (e.g. ‘ultra high risk’: 42%). Moreover, Kim et al. (Reference Kim, Polari, Melville, Moller, Kim, Amminger and Nelson2017) report that the majority of patients with a family history of psychosis perceived that the ‘ultra high risk’ label carried a strong stigma.
Yung et al. posit that stigma occurs earlier in association with the distressing symptoms themselves rather than the risk-label. For this, they refer to a study by Yang et al. (Reference Yang, Link, Ben-David, Gill, Girgis, Brucato and Corcoran2015) that in our opinion is not fully relevant as the risk label was explicit neither in the name of the program [Center of Prevention and Evaluation (COPE)] nor in the items (e.g. ‘About coming to this program, I have felt ashamed’). In fact, Yang et al. point out that certain stereotypes gain salience upon clinical high risk identification and that simply being aware of them is related to negative outcomes due to anticipated rejection by others. They also found that labeling-related shame was associated with increased anxiety.
Stigma inflicted by institutional labels
Yung et al. recommend that services for ARMS individuals should carry nonstigmatizing names. We agree with them on this. They then list services with acronyms that do not have the terms prodromal, schizophrenia, or psychosis in their names that might stigmatize ARMS individuals as ‘psychotics in spe.’ A simple Internet search and a review of online directories of early detection facilities (e.g. http://www.easacommunity.org/files/Early_Psychosis_Intervention_Program_Directory.pdf; https://iepa.org.au) show that many of these institutions do still carry names emphasizing the risk for psychosis [some even have the term prodromal in their name, such as the Staglin Music Festival Center for the Assessment and Prevention of Prodromal States (CAPPS)].
Yung et al. expressed surprise that we cited a study (Yang et al., Reference Yang, Anglin, Wonpat-Borja, Opler, Greenspoon and Corcoran2013) involving college students showing that the ‘psychosis risk designation’ was associated with significantly more stigma in terms of status loss and discrimination than labels for nonpsychotic disorders. We find this study very relevant, because the attitudes of a nonclinical group may in fact coincide with the attitudes of at-risk individuals contemplating whether to enter a specialized service. Further, the study shows that the students considered the at-risk label as stigmatizing as the schizophrenia label (p. 284). Yung et al. argue that providing information about the meaning of at-risk states ameliorated these attitudes. In fact, no significant stigma reduction for the psychosis risk label was found when such information was provided in 5 out of 7 dimensions, and stigma remained high (e.g. violence toward others likely: 36%).
Yung et al. criticized that we did not cite studies showing that individuals feel valued for the opportunity to speak about their experiences in ARMS services (e.g. Morrison et al., Reference Morrison, Birchwood, Pyle, Flach, Stewart, Byrne and French2013). This is an important point and we regret that this was not explicitly stated in our article.
Yung et al. also object that we did not cite a recent paper by Corcoran (Reference Corcoran2016) that, according to their reading, would refute our position. We disagree. We concur with many statements in that article, for example, that ‘the emotional risks of stigma and discrimination associated with the label of psychosis risk are real, especially if the label occurs without information about what it means’ (p. 638). We remain skeptical as to whether the measures suggested to prevent (revert) stigma – once the possible outcome of psychosis is stated (see Kim et al., Reference Kim, Polari, Melville, Moller, Kim, Amminger and Nelson2017 on the high frequency with which the label psychosis is used in an early interventional service) – are sufficient.
Labels, stigma, and antipsychotic medication
We appreciate that Yung et al. endorse the practice of not telling participants they might suffer from a diagnosis of ‘Attenuated Psychosis Syndrome.’ According to Yang et al. the recommended approach includes ‘discussing current symptoms and risk, what being ‘at risk’ of psychosis means.’ We do not agree with this recommendation because psychosis is not the most likely outcome. Further, contemplating a possible diagnosis of psychosis in the future enhances the risk that physicians will consider antipsychotic medication to offer some relief from distress to individuals who expect treatment because they have received a diagnostic label (for a critical discussion of off-label prescription of antipsychotics see Rani, Murray, Byrne, & Wong, Reference Rani, Murray, Byrne and Wong2008).
Yung et al. write that antipsychotics ‘are not the usual treatment for ARMS individuals’ but later acknowledge that services in the United States seem to prescribe antipsychotics more readily than in other countries. In an editorial, Yung (Reference Yung2010) even writes: ‘High numbers of UHR patients are receiving antipsychotic medication’ (p. 198). The authors find that antipsychotics are not justified in most cases. We disagree and think that the off-label use of antipsychotics in ARMS is not justified at all given the serious adverse side effects and the lack of evidence from meta-analyses (e.g. Devoe, Farris, Townes, & Addington, Reference Devoe, Farris, Townes and Addington2019). In addition, the relatively high prevalence of off-label prescription of antipsychotics in possibly ‘prodromal’ individuals is by no means confined to the United States (see Nieman et al., Reference Nieman, Rike, Becker, Dingemans, van Amelsvoort, de Haan and Linszen2009; Rüsch et al., Reference Rüsch, Heekeren, Theodoridou, Müller, Corrigan, Mayer and Rössler2015).
Can treatment reliably prevent transition to psychosis?
Yung et al. criticize us for claiming that there is little evidence that specific interventions can prevent transition to psychosis. They (see also Nelson, Amminger, & McGorry, Reference Nelson, Amminger and McGorry2018) refute the inferences from the meta-analysis by Davies et al. (Reference Davies, Cipriani, Ioannidis, Radua, Stahl, Provenzani and Fusar-Poli2018). To save space, we would like to refer readers to the upcoming reply letter by Fusar-Poli's group in response to this criticism. Moreover, this meta-analysis is not the only source that tempers enthusiasm for the efficacy of interventions targeted at individuals at putative risk for psychosis (see Devoe et al., Reference Devoe, Farris, Townes and Addington2019; Bosnjak Kuharic et al., Reference Bosnjak Kuharic, Kekin, Hew, Rojnic Kuzman and Puljak2019).
Other points of criticism
Yung et al. also criticize us for incorrectly citing work by Birchwood and Macmillan (Reference Birchwood and Macmillan1993) because their work would refer to early intervention for psychosis and not ARMS services. In fact, we cited this study when describing the beginnings of early detection (Birchwood and Macmillan write: ‘The notion of early treatment at the first episode has raised the intriguing possibility of preventing the first occurrence of frank psychotic symptoms,’ p. 374).
Yung et al. also refer to a paper by our group (Miegel, Jelinek, & Moritz, Reference Miegel, Jelinek and Moritz2019) demonstrating that patients with depression and obsessive-compulsive disorder often report the fear of becoming ‘insane’ or psychotic. We should have added that lifetime psychotic symptoms and bipolar disorder led to exclusion from this study; in addition, this sample was far older than the age of usual exacerbation. So, the particular fear of these participants was unfounded but still prevalent and significantly correlated with suicidality. Citing this study was meant to emphasize our point that the possibility of later psychosis or ‘insanity’ should be de-emphasized due to the stigma associated with it.
In the section titled ‘ARMS services are stigmatising and cause fear of psychosis,’ it appears as if we provided only three references to support our statement, all of which Yung et al. deem irrelevant or misleading. In fact, our article contained many more citations on the topic. Although the paper by Corcoran et al. (Reference Corcoran, Malaspina and Hercher2005) indeed did not report any empirical data, its inclusion is justified in our view as it presents an expert's opinion on inherent risks in prodromal research (reflection on the risk of medication exposure, stigma among false-positive cases, and limits of confidentiality). We were also criticized for citing a study by Rüsch et al. (Reference Rüsch, Heekeren, Theodoridou, Müller, Corrigan, Mayer and Rössler2015) because it investigated a mixed sample with participants at risk for bipolar disorder or at risk for psychosis. This is misleading. A total of 87% of the sample fulfilled criteria for (ultra) high risk for psychosis, and in a subsidiary analysis with this subsample the predictive relationship between perceived harm due to stigma with transition was confirmed. Yung et al. also seem to criticize that the Schizophrenia Proneness Interview was used, but this interview is administered in many early detection centers.
Yung et al. also questioned the following statement we made: ‘We regard it as a great step forward that assessments in this area are increasingly incorporating interviews.’ They encouraged us to read the relevant literature demonstrating that structured interviews have always been central to identifying ARMS individuals and are the most widely accepted instruments for the identification of ARMS. This statement was taken out of context. We had highlighted that contemporary high risk research has evolved from research on schizotypy/schizotypal personality, which has been predominantly conducted using questionnaires such as the Schizotypal Scale. Items from these scales form the backbone of many instruments used today in early detection.
We appreciate that guidelines recommend interviews, but it is well known from many other disorders, such as attention deficit disorder (Schneider, Schöttle, Hottenrott, Gallinat, & Moritz, Reference Schneider, Schöttle, Hottenrott, Gallinat and Moritz2020), that guidelines and diagnostic recommendations are often ignored by clinicians. As stated in the original article, even with interviews the problem persists that an individual with (alleged) ARMS is often the only source the interviewer has, and response biases such as underreporting (e.g. due to low insight) or overreporting (e.g. to receive as much help as possible) are not readily addressed.
The authors summarize statements made by our group and Ajnakina et al. (Reference Ajnakina, David and Murray2019) as follows: ‘People meeting ARMS criteria have already started the process of developing psychotic disorder and therefore intervention at this stage is too late.’ This summary is not correct. In fact, we emphasized that so far no treatment can reliably prevent a later manifestation of psychosis. According to Yung et al. ‘prodromal’ suggests ‘inevitable progression to psychotic disorder.’ We used the term in the sense it is understood by many experts. Our understanding of the label prodrome is also consistent with the DSM-5 (APA, 2013), where it is mentioned that prodromal symptoms may not turn into a psychotic disorder (p. 361).
Yung et al. judged our appeal for hope-oriented services ‘ironic’ because they felt we believe that no treatment can fully justify hope. We kindly ask readers to refer back to our original paper. Based on the most recent meta-analyses (see above), we remain skeptical that current treatment programs can prevent psychosis. We also criticize many clinics for still having the terms psychosis or prodromal in their name, which may foster pathological self-labeling and is not hope-oriented. Yung et al. themselves acknowledge in their commentary that psychosis is not the most prevalent outcome in at-risk individuals.
Yung et al. posit that our group and Ajnakina et al. (Reference Ajnakina, David and Murray2019) conflate the first episode of psychosis with schizophrenia. Psychosis is an umbrella term, and clinicians often use it as shorthand for the schizophrenia spectrum. Clinicians seem to prefer the label psychosis because it is allegedly less stigmatizing than schizophrenia. The term first episode of psychosis is a descriptive rather than a diagnostic term. As an aside, according to our reading of the DSM-5, first episode of psychosis is only used in the context of a substance- or medication-induced psychotic disorder. To conclude, we would like to repeat our contention that some ARMS services put ‘a strong emphasis on the (relatively low) possibility of schizophrenia’ as if ‘schizophrenia is looming over’ the patients.
Yung et al. write the following: ‘But just because other poor outcomes can occur in ARMS individuals, this does not mean that the development of psychotic disorder is any less important.’ Yes, clinicians should have in mind the possible development of psychosis. But in view of the low (and declining) transition rates – Yung et al. themselves mention transient or benign experiences – and in view of the stigma associated with the labels psychosis or schizophrenia, we want to caution early detection centers against overemphasizing the relatively remote possibility of psychosis. We urge clinicians to strengthen their focus on current symptoms and not to speculate about a possible future psychosis as implied by the at-risk for psychosis concept.
Duration of untreated prodrome
Yung et al. argue that we presented a poorly articulated explanation of why the weak correlation between a long DUP and poor outcome in schizophrenia patients is relevant to our criticism of ARMS services. We tried to stress that it is implicit in some articles that the duration of untreated prodrome is correlated with the risk of manifest psychosis or a poor outcome. Specific studies are lacking, however, and this assumption seems to be deduced from an alleged relationship between DUP and outcome, for which evidence is on the decline (see Moritz et al., Reference Moritz, Gawęda, Heinz and Gallinat2019). The argument of Yung et al. that a ‘very short DUP may have a marked effect on symptom reduction, highlighting this role of ARMS services for those who develop psychotic disorder,’ is thus speculative given current information.
You have
Access
Introduction
We appreciate the opportunity to reply to the article by Yung et al. (Reference Yung, Wood, Malla, Nelson, McGorry and Shah2020) that critiques arguments made by our group (Moritz, Gawęda, Heinz, & Gallinat, Reference Moritz, Gawęda, Heinz and Gallinat2019) as well as Ajnakina, David, and Murray (Reference Ajnakina, David and Murray2019) concerning the early detection of psychosis paradigm.
As stated in our original article (Moritz et al., Reference Moritz, Gawęda, Heinz and Gallinat2019), we deem the idea of preventing the exacerbation of severe mental illnesses laudable and respect the important work of McGorry and other pioneers in this area. Nonetheless, we want to refute many of the arguments raised in the comments by Yung et al. and highlight some incorrect inferences derived from our statements that in some cases were taken out of context. The reply by Yung et al. often addresses single aspects of particular studies. We kindly ask readers to refer to our original article (Moritz et al., Reference Moritz, Gawęda, Heinz and Gallinat2019) to ensure they see the bigger picture.
Stigma
Stigma inflicted by syndrome labels
Yung et al. (Reference Yung, Wood, Malla, Nelson, McGorry and Shah2020) disagree with our argument that stigma may be imposed by labels such as ultra high risk (UHR). Referring to the investigation by Kim et al. (Reference Kim, Polari, Melville, Moller, Kim, Amminger and Nelson2017), Yung et al. write that ‘a recent study found that the terms ARMS, UHR or ‘Attenuated Psychotic Symptoms’ were not thought to be stigmatizing by ARMS individuals and only a minority thought these names should be changed.’ If these labels are not stigmatizing to individuals in at risk mental state services (ARMS), this would indeed be a strong argument against our claim. However, this is not the case. Kim et al. report that between 36.7% and 42% of ARMS patients found these labels stigmatizing. Although it is true that a minority thought that these names needed to be changed, that minority was substantial (e.g. ‘ultra high risk’: 42%). Moreover, Kim et al. (Reference Kim, Polari, Melville, Moller, Kim, Amminger and Nelson2017) report that the majority of patients with a family history of psychosis perceived that the ‘ultra high risk’ label carried a strong stigma.
Yung et al. posit that stigma occurs earlier in association with the distressing symptoms themselves rather than the risk-label. For this, they refer to a study by Yang et al. (Reference Yang, Link, Ben-David, Gill, Girgis, Brucato and Corcoran2015) that in our opinion is not fully relevant as the risk label was explicit neither in the name of the program [Center of Prevention and Evaluation (COPE)] nor in the items (e.g. ‘About coming to this program, I have felt ashamed’). In fact, Yang et al. point out that certain stereotypes gain salience upon clinical high risk identification and that simply being aware of them is related to negative outcomes due to anticipated rejection by others. They also found that labeling-related shame was associated with increased anxiety.
Stigma inflicted by institutional labels
Yung et al. recommend that services for ARMS individuals should carry nonstigmatizing names. We agree with them on this. They then list services with acronyms that do not have the terms prodromal, schizophrenia, or psychosis in their names that might stigmatize ARMS individuals as ‘psychotics in spe.’ A simple Internet search and a review of online directories of early detection facilities (e.g. http://www.easacommunity.org/files/Early_Psychosis_Intervention_Program_Directory.pdf; https://iepa.org.au) show that many of these institutions do still carry names emphasizing the risk for psychosis [some even have the term prodromal in their name, such as the Staglin Music Festival Center for the Assessment and Prevention of Prodromal States (CAPPS)].
Yung et al. expressed surprise that we cited a study (Yang et al., Reference Yang, Anglin, Wonpat-Borja, Opler, Greenspoon and Corcoran2013) involving college students showing that the ‘psychosis risk designation’ was associated with significantly more stigma in terms of status loss and discrimination than labels for nonpsychotic disorders. We find this study very relevant, because the attitudes of a nonclinical group may in fact coincide with the attitudes of at-risk individuals contemplating whether to enter a specialized service. Further, the study shows that the students considered the at-risk label as stigmatizing as the schizophrenia label (p. 284). Yung et al. argue that providing information about the meaning of at-risk states ameliorated these attitudes. In fact, no significant stigma reduction for the psychosis risk label was found when such information was provided in 5 out of 7 dimensions, and stigma remained high (e.g. violence toward others likely: 36%).
Yung et al. criticized that we did not cite studies showing that individuals feel valued for the opportunity to speak about their experiences in ARMS services (e.g. Morrison et al., Reference Morrison, Birchwood, Pyle, Flach, Stewart, Byrne and French2013). This is an important point and we regret that this was not explicitly stated in our article.
Yung et al. also object that we did not cite a recent paper by Corcoran (Reference Corcoran2016) that, according to their reading, would refute our position. We disagree. We concur with many statements in that article, for example, that ‘the emotional risks of stigma and discrimination associated with the label of psychosis risk are real, especially if the label occurs without information about what it means’ (p. 638). We remain skeptical as to whether the measures suggested to prevent (revert) stigma – once the possible outcome of psychosis is stated (see Kim et al., Reference Kim, Polari, Melville, Moller, Kim, Amminger and Nelson2017 on the high frequency with which the label psychosis is used in an early interventional service) – are sufficient.
Labels, stigma, and antipsychotic medication
We appreciate that Yung et al. endorse the practice of not telling participants they might suffer from a diagnosis of ‘Attenuated Psychosis Syndrome.’ According to Yang et al. the recommended approach includes ‘discussing current symptoms and risk, what being ‘at risk’ of psychosis means.’ We do not agree with this recommendation because psychosis is not the most likely outcome. Further, contemplating a possible diagnosis of psychosis in the future enhances the risk that physicians will consider antipsychotic medication to offer some relief from distress to individuals who expect treatment because they have received a diagnostic label (for a critical discussion of off-label prescription of antipsychotics see Rani, Murray, Byrne, & Wong, Reference Rani, Murray, Byrne and Wong2008).
Yung et al. write that antipsychotics ‘are not the usual treatment for ARMS individuals’ but later acknowledge that services in the United States seem to prescribe antipsychotics more readily than in other countries. In an editorial, Yung (Reference Yung2010) even writes: ‘High numbers of UHR patients are receiving antipsychotic medication’ (p. 198). The authors find that antipsychotics are not justified in most cases. We disagree and think that the off-label use of antipsychotics in ARMS is not justified at all given the serious adverse side effects and the lack of evidence from meta-analyses (e.g. Devoe, Farris, Townes, & Addington, Reference Devoe, Farris, Townes and Addington2019). In addition, the relatively high prevalence of off-label prescription of antipsychotics in possibly ‘prodromal’ individuals is by no means confined to the United States (see Nieman et al., Reference Nieman, Rike, Becker, Dingemans, van Amelsvoort, de Haan and Linszen2009; Rüsch et al., Reference Rüsch, Heekeren, Theodoridou, Müller, Corrigan, Mayer and Rössler2015).
Can treatment reliably prevent transition to psychosis?
Yung et al. criticize us for claiming that there is little evidence that specific interventions can prevent transition to psychosis. They (see also Nelson, Amminger, & McGorry, Reference Nelson, Amminger and McGorry2018) refute the inferences from the meta-analysis by Davies et al. (Reference Davies, Cipriani, Ioannidis, Radua, Stahl, Provenzani and Fusar-Poli2018). To save space, we would like to refer readers to the upcoming reply letter by Fusar-Poli's group in response to this criticism. Moreover, this meta-analysis is not the only source that tempers enthusiasm for the efficacy of interventions targeted at individuals at putative risk for psychosis (see Devoe et al., Reference Devoe, Farris, Townes and Addington2019; Bosnjak Kuharic et al., Reference Bosnjak Kuharic, Kekin, Hew, Rojnic Kuzman and Puljak2019).
Other points of criticism
Yung et al. also criticize us for incorrectly citing work by Birchwood and Macmillan (Reference Birchwood and Macmillan1993) because their work would refer to early intervention for psychosis and not ARMS services. In fact, we cited this study when describing the beginnings of early detection (Birchwood and Macmillan write: ‘The notion of early treatment at the first episode has raised the intriguing possibility of preventing the first occurrence of frank psychotic symptoms,’ p. 374).
Yung et al. also refer to a paper by our group (Miegel, Jelinek, & Moritz, Reference Miegel, Jelinek and Moritz2019) demonstrating that patients with depression and obsessive-compulsive disorder often report the fear of becoming ‘insane’ or psychotic. We should have added that lifetime psychotic symptoms and bipolar disorder led to exclusion from this study; in addition, this sample was far older than the age of usual exacerbation. So, the particular fear of these participants was unfounded but still prevalent and significantly correlated with suicidality. Citing this study was meant to emphasize our point that the possibility of later psychosis or ‘insanity’ should be de-emphasized due to the stigma associated with it.
In the section titled ‘ARMS services are stigmatising and cause fear of psychosis,’ it appears as if we provided only three references to support our statement, all of which Yung et al. deem irrelevant or misleading. In fact, our article contained many more citations on the topic. Although the paper by Corcoran et al. (Reference Corcoran, Malaspina and Hercher2005) indeed did not report any empirical data, its inclusion is justified in our view as it presents an expert's opinion on inherent risks in prodromal research (reflection on the risk of medication exposure, stigma among false-positive cases, and limits of confidentiality). We were also criticized for citing a study by Rüsch et al. (Reference Rüsch, Heekeren, Theodoridou, Müller, Corrigan, Mayer and Rössler2015) because it investigated a mixed sample with participants at risk for bipolar disorder or at risk for psychosis. This is misleading. A total of 87% of the sample fulfilled criteria for (ultra) high risk for psychosis, and in a subsidiary analysis with this subsample the predictive relationship between perceived harm due to stigma with transition was confirmed. Yung et al. also seem to criticize that the Schizophrenia Proneness Interview was used, but this interview is administered in many early detection centers.
Yung et al. also questioned the following statement we made: ‘We regard it as a great step forward that assessments in this area are increasingly incorporating interviews.’ They encouraged us to read the relevant literature demonstrating that structured interviews have always been central to identifying ARMS individuals and are the most widely accepted instruments for the identification of ARMS. This statement was taken out of context. We had highlighted that contemporary high risk research has evolved from research on schizotypy/schizotypal personality, which has been predominantly conducted using questionnaires such as the Schizotypal Scale. Items from these scales form the backbone of many instruments used today in early detection.
We appreciate that guidelines recommend interviews, but it is well known from many other disorders, such as attention deficit disorder (Schneider, Schöttle, Hottenrott, Gallinat, & Moritz, Reference Schneider, Schöttle, Hottenrott, Gallinat and Moritz2020), that guidelines and diagnostic recommendations are often ignored by clinicians. As stated in the original article, even with interviews the problem persists that an individual with (alleged) ARMS is often the only source the interviewer has, and response biases such as underreporting (e.g. due to low insight) or overreporting (e.g. to receive as much help as possible) are not readily addressed.
The authors summarize statements made by our group and Ajnakina et al. (Reference Ajnakina, David and Murray2019) as follows: ‘People meeting ARMS criteria have already started the process of developing psychotic disorder and therefore intervention at this stage is too late.’ This summary is not correct. In fact, we emphasized that so far no treatment can reliably prevent a later manifestation of psychosis. According to Yung et al. ‘prodromal’ suggests ‘inevitable progression to psychotic disorder.’ We used the term in the sense it is understood by many experts. Our understanding of the label prodrome is also consistent with the DSM-5 (APA, 2013), where it is mentioned that prodromal symptoms may not turn into a psychotic disorder (p. 361).
Yung et al. judged our appeal for hope-oriented services ‘ironic’ because they felt we believe that no treatment can fully justify hope. We kindly ask readers to refer back to our original paper. Based on the most recent meta-analyses (see above), we remain skeptical that current treatment programs can prevent psychosis. We also criticize many clinics for still having the terms psychosis or prodromal in their name, which may foster pathological self-labeling and is not hope-oriented. Yung et al. themselves acknowledge in their commentary that psychosis is not the most prevalent outcome in at-risk individuals.
Yung et al. posit that our group and Ajnakina et al. (Reference Ajnakina, David and Murray2019) conflate the first episode of psychosis with schizophrenia. Psychosis is an umbrella term, and clinicians often use it as shorthand for the schizophrenia spectrum. Clinicians seem to prefer the label psychosis because it is allegedly less stigmatizing than schizophrenia. The term first episode of psychosis is a descriptive rather than a diagnostic term. As an aside, according to our reading of the DSM-5, first episode of psychosis is only used in the context of a substance- or medication-induced psychotic disorder. To conclude, we would like to repeat our contention that some ARMS services put ‘a strong emphasis on the (relatively low) possibility of schizophrenia’ as if ‘schizophrenia is looming over’ the patients.
Yung et al. write the following: ‘But just because other poor outcomes can occur in ARMS individuals, this does not mean that the development of psychotic disorder is any less important.’ Yes, clinicians should have in mind the possible development of psychosis. But in view of the low (and declining) transition rates – Yung et al. themselves mention transient or benign experiences – and in view of the stigma associated with the labels psychosis or schizophrenia, we want to caution early detection centers against overemphasizing the relatively remote possibility of psychosis. We urge clinicians to strengthen their focus on current symptoms and not to speculate about a possible future psychosis as implied by the at-risk for psychosis concept.
Duration of untreated prodrome
Yung et al. argue that we presented a poorly articulated explanation of why the weak correlation between a long DUP and poor outcome in schizophrenia patients is relevant to our criticism of ARMS services. We tried to stress that it is implicit in some articles that the duration of untreated prodrome is correlated with the risk of manifest psychosis or a poor outcome. Specific studies are lacking, however, and this assumption seems to be deduced from an alleged relationship between DUP and outcome, for which evidence is on the decline (see Moritz et al., Reference Moritz, Gawęda, Heinz and Gallinat2019). The argument of Yung et al. that a ‘very short DUP may have a marked effect on symptom reduction, highlighting this role of ARMS services for those who develop psychotic disorder,’ is thus speculative given current information.
Conclusions
We have responded to several of the criticisms made by Yung et al. We stand by our position that early detection centers for psychosis should be renamed and restructured. At the same time, we want to emphasize our opinion that if all early detection services were named and run like the centers founded by some of the authors of the critique (e.g. headspace, soulspace), we would not have written our original article. We also would like to express our belief that Yung et al. and our group may be more in agreement than it appears at first sight from this debate. Many of their recommendations are congruent with our criticism of early intervention services.