Towards the DSM-6: Results of a Survey of Experts on the Reintroduction of First-Rank Symptoms as Core Criteria of Schizophrenia and on Redefining Hallucinations Free

Abstract

Introduction

The Diagnostic and Statistical Manual of Mental Disorders is now in its fifth edition. First published in 1952, it aimed to overcome the “polyglot of diagnostic labels and systems, effectively blocking communication and the collection of medical statistics” (DSM-I, p. v). The DSM covers the entire range of psychiatric problems, relying on the latest evidence, and consensus in the scientific community.^1,2 Accordingly, diagnostic criteria in psychiatry/clinical psychology are subject to change.

A disorder with a complex diagnostic past is schizophrenia/psychosis. The term “schizophrenia” was used in the DSM-I together with the label “dementia praecox.” In subsequent editions, the DSM—with widely varying weights of diagnostic importance—reflected the ideas of Kraepelin (especially on mental deterioration), Bleuler (especially on disorganization), Schneider (see below), and more recently advocates of the positive–negative concept.^3,4 The present article focuses on two prominent positive symptoms: first-rank symptoms [FRS; often referred to as a loss of ego boundaries such as thought broadcasting (FRS are sometimes also characterized as disorders of the personal self or as disorders of self-activity and coherence)] and hallucinations.

Last year, our group published an article on Kurt Schneider’s impact on the classification and diagnostic criteria of schizophrenia.⁵ The article sheds light on how Schneider’s work was reflected in the different editions of the DSM and the empirical status of FRS. Since Schneider did not present any studies and never rigorously verified the allegedly prominent role of FRS in schizophrenia,^6,7 his contribution regarding FRS was eminence rather than evidence-based. The pathognomonic status of FRS, based on the claim that they are unique to schizophrenia, has been broadly refuted.^8,9 At the same time, there is evidence that these types of symptoms continue to have special importance in treatment decisions and can be used in differential diagnosis,¹⁰ particularly in resource-limited settings. Soares-Weiser and colleagues¹¹ similarly argue for the utility of FRS as a diagnostic tool, despite the known limitations. Heinz and colleagues¹² propose that the significance of FRS might be reduced to the extent that a solitary FRS occurrence alone may not be adequate for diagnosing schizophrenia but that the absence of FRS should serve as a cautionary indication, suggesting the need for thorough examination to exclude other potential causes before confirming a diagnosis of schizophrenia or schizoaffective disorder. Moreover, FRS has been shown to be more common in schizophrenia than in other disorders,¹³ even in the decisive study by Carpenter and colleagues.⁸ In line with this, meta-analytical evidence¹⁴ suggests a high prevalence of self-disorders in schizophrenia-spectrum disorders compared to other mental illnesses.¹⁵ Given the close relationship between FRS and self-disorders,¹⁶ these findings lend some support to the call to give FRS greater weight.

Our review also proposes a redefinition of hallucinations in light of newer research refuting a claim made in the DSM-5 that hallucinations are “vivid and clear, with the full force and impact of normal perceptions, and not under voluntary control” (p. 87). In fact, hallucinations in schizophrenia often lack central aspects of a true perception such as normal volume and clarity as well as authenticity.^17–21 This view broadly coincides with Schneider’s.²²

Our review ends with a plea to seek experts’ opinions on the future diagnostic importance of FRS and the definition of hallucinations. For the present study, we therefore conducted a survey asking clinicians and researchers for their opinions regarding the diagnostic status of FRS and the definition of hallucinations. While we had no strong hypotheses on experts’ views on the diagnostic weight of Schneiderian symptoms, we expected that most experts would advocate a redefinition of hallucinations to capture two kinds of symptoms: internally generated phenomena that cannot be distinguished from real perceptions irrespective of insight and perception-like phenomena that show differences from a real perception but that the holder is convinced are non-self-generated.

Methods

From November 6, 2023, to January 6, 2024, we invited experts on schizophrenia/psychosis via several sources to take part in our survey. First, we emailed all editorial members of Schizophrenia Research and Schizophrenia Bulletin, the two largest journals that specialize in research on psychosis spectrum disorders. The International Consortium for Hallucination Research also advertised the survey to its members. Further, we communicated the results of our earlier article via LinkedIn and invited experts on schizophrenia to take part in the survey. As an incentive, each completer was offered free access to a 12 h e-learning curriculum on metacognitive training for psychosis developed by our group (www.uke.de/e-mct).

The landing page of the survey was accessed by 196 individuals (this may include multiple web accesses by the same person). A total of 172 individuals proceeded to the first page, and 148 completed at least one question on the core items of interest. Blind to the results, we discarded 12 individuals for the following reasons (multiple criteria could apply): less than 3 years of research or treatment experience (n = 9), bachelor’s degree only (n = 2), no knowledge of the DSM-5 criteria (n = 3). Thus, the final sample encompassed 136 experts. We also formed a subsample of 53 verified experts who reported having at least 6 years of experience with research and/or treatment of schizophrenia and provided their email addresses (see below). The study was approved by the local ethics committee of psychologists at the University Medical Center Hamburg (Germany; LPEK-0715).

Survey

This survey was composed using unipark. Participants completed the questionnaire in a median time of 5 min and 45 s. The introduction to the survey emphasized its short length and indicated that it was directed at mental health professionals with clinical and/or research experience with schizophrenia/psychosis. It stated that we would pose questions on “current diagnostic criteria for schizophrenia in the DSM-5 as well as the DSM-5 definition of hallucinations.” We noted that we might share the results of the survey with the American Psychiatric Association, particularly the editors of the DSM-6, to contribute to a discussion on future diagnostic criteria for the condition.

The survey first asked about the profession, the highest academic achievement, clinical as well as research experience (see table 1), and knowledge of the diagnostic criteria for schizophrenia. Further, we asked whether the participant was a member of an organization related to schizophrenia (eg, Schizophrenia International Research Society) or of an editorial board (optional response). We then posed the questions shown in tables 2 and 3 and asked whether they had read our article in Schizophrenia Bulletin. Participants could leave their name and email address at the end of the survey, and they were sent a confirmation email (holders of the email account were instructed to contact us if they had not entered the email address themselves). Upon finishing the questionnaire, the participants received a copy of our review article⁵ and were allowed to take the survey a second time after reading the article (only 6 individuals chose to do so; these data are not presented).

Results

Sample characteristics are displayed in table 1. Approximately half of the sample held the title of professor (50.7%), with full professors representing the largest subgroup (25%). A slight majority were psychiatrists (53.7%), followed by psychologists (38.2%).

Approximately half of the sample (49.3%) agreed that the future DSM-6 should give FRS the special weight they were given in the DSM-IV (disagreement: 34.6%). Almost half of the sample agreed that FRS have a special relevance in the prognosis and treatment of psychosis (54.4%; disagreement: 34.6%). Most experts disagreed with the statement that FRS are nonspecific (59.6%; agreement: 28.7%). The complementary question about the specificity of FRS was affirmed by 69.1% (disagreement: 22.8%), with a lower number of verified experts sharing this opinion (60.0%).

The results on hallucinations were more consistent. Approximately 4 out of 5 participants (79.7%) agreed that hallucinations may not always have the “full force and impact of normal perceptions” as required in the DSM-5. Three out of 4 participants (76.3%; the agreement among verified experts was even higher: 85.7%) confirmed the following statement: “Voices that are not similar to perceptions but which the individual is convinced are real and external are also hallucinations.” In line with this, the majority (83.9%) agreed that the definition of hallucinations should also include internal events that, while not phenomenologically fully resembling a perception (ie, they may only carry some [attenuated] features of perceptions), are believed to be true. Still, only a minority (28.0%) endorsed the response that only the latter definition was valid. Approximately two-thirds of the sample (66.1%) disagreed that the DSM-6 definition of hallucinations should stay as before.

Discussion

Experts’ opinions and diagnostic criteria for schizophrenia/psychosis have varied greatly over the years. Kraepelin regarded intellectual deterioration as a core feature (although he described cases with full remission not resulting in dementia),²³ Bleuler emphasized the “splitting of mental functioning” due to disorders of association, and the current definition in the DSM-5 emphasizes positive symptoms.^3,4

The present study focused on two of the core symptoms of schizophrenia: (a) FRS (often described as a disturbance in the ego boundaries, usually accompanied by a delusional interpretation), and (b) hallucinations. Whereas the DSM-IV gives special weight to FRS, the DSM-5 does not. Regarding hallucinations, the different editions of the DSM have been mostly consistent.⁵

The present results are more straightforward for hallucinations than for the status of FRS. A vast majority of the experts who participated in our survey responded in favor of broadening the current definition of hallucination to include perception-laden phenomena that the individual believes to be externally generated but that may nevertheless largely differ from real perceptions on important characteristics. As for the Schneiderian FRS, a slim majority (54.4%) of experts indicated they are of special relevance in the prognosis and treatment of psychosis. While a majority agreed with the suggestion to change the current DSM criteria to give FRS more weight, as in the DSM-IV, the endorsement is not very strong. In our opinion, the absence of FRS should at a minimum prompt clinicians to thoroughly consider diagnoses other than schizophrenia.

Our results have some limitations. We did our best to contact all members of the editorial boards of Schizophrenia Bulletin and Schizophrenia Research via email, but not all responded. We also asked members of the International Consortium for Hallucination Research to disseminate our invitation to participate in the survey. While we designed this approach to gather a large number of true experts on schizophrenia/psychosis, we cannot claim that the participants are fully representative. In fact, there is (a) no clear criterion or definition of what constitutes a true expert sample,²⁴ and (b) those from Western countries were overrepresented. Another trend in research, the involvement of patients,²⁵ was also not fulfilled. Since the survey could be taken anonymously, there are some concerns pertaining to validity. Participants had the option to give their email address; those who did were then sent a confirmation (see above). Participants who chose to disclose their identity and who had at least 6 years of research and clinical experience formed a special subgroup; this group, however, showed few deviations from the results of the entire group. Finally, further research is needed to delineate the boundaries between perceptions, illusions, hallucinations, and FRS.

To conclude, the majority of the experts agreed with the suggestion to change the current definition of hallucinations to include cases where cognitions have perception-like features without meeting the full force of perception but where the individual is convinced the inner speech (inner image, etc.) is real and is occurring due to non-self-generated (ie, external to the self) forces. The aspect of conviction is of special importance to distinguish a hallucination from an intrusive thought, which can also carry some perceptual features as in flashbacks or obsessive thoughts^26–29 although the individual knows the image has been created in their own mind. A too-liberal definition would otherwise unduly pathologize common phenomena such as catchy tunes or inner speech.

The results are more ambiguous for FRS. As we have argued previously, their diagnostic status requires more than consensus among experts; we may need more studies to pinpoint the relevance of FRS. Yet, as we have shown before, there is indeed evidence that while FRS are not fully pathognomonic to schizophrenia (nor is any other symptom) they are far more common than in other disorders (eg, depression), especially when multiple symptoms are present.⁸ As noted,⁵ we propose that studies should ask not only for the presence of these symptoms but also for the degree of conviction. While “as if” feelings may be common beyond schizophrenia, we presume that FRS held with conviction will be rather specific for schizophrenia. We hope that this article will stimulate discussion among experts about the definition of core psychotic symptoms and the composition of diagnostic criteria for the forthcoming edition of the DSM.

Conflict of Interest

The authors have declared that there are no conflicts of interest in relation to the subject of this study.

References