The British Journal of Psychiatry (2023)

223, 271–272. doi: 10.1192/bjp.2023.68

Editorial

Negative symptoms in the clinic: we

treat what we can describe
Brian Kirkpatrick, Lauren Luther and Gregory P. Strauss

Summary Keywords
Recent research has led to important changes in the concepts Negative symptoms; schizophrenia; psychometrics; assess-
and assessment of negative symptoms in schizophrenia. We ment; factor analysis.
review current negative symptom concepts and their clinical
implications, as well as new methods of assessing these symp-
toms. These changes hold promise for improving our under- Copyright and usage
standing and treatment of negative symptoms. © The Author(s), 2023. Published by Cambridge University Press
on behalf of the Royal College of Psychiatrists.

This evidence, which crosses cultures, languages and scales,

Brian Kirkpatrick (pictured) is Professor of Psychiatry at the University of Arkansas for
raises the possibility that these five factors reflect brain function,
Medical Sciences, Little Rock, USA. His research focuses on the conceptualisation,
measurement and treatment of negative symptoms in schizophrenia, and the metabolic
and possibly discrete functional circuits.2 It also raises the possibility
problems found in schizophrenia. Lauren Luther is an assistant research scientist at the that there are five separate treatment targets.2 That is, there may be
University of Georgia, Athens, USA. Her research focuses on identifying and treating biomedical or psychosocial treatments that are effective for one or
novel negative symptoms mechanisms in schizophrenia. Gregory P. Strauss is an
Associate Professor of Psychology and Neuroscience at the University of Georgia,
more of the negative symptoms but not for others. In that case, a
Athens, USA. His research focuses on the phenomenology, aetiology, assessment and treatment study that does not improve the total score on a rating
treatment of negative symptoms in groups across the psychosis spectrum. scale – the usual outcome measure for studies of negative symptoms
– may be falsely negative, with an effect in one or two domains
buried by the lack of response in others. On the other hand, there
are network analysis studies – yet more psychometrics! – that
Recent findings have led to important changes in the concept of suggest effective treatment of avolition may lead to improvement
negative symptoms, including what should be considered a negative in the other negative symptoms.3
symptom, the relationships among these symptoms and how to
measure them. These issues are often discussed in articles using
advanced statistical methods, but the issues in such articles are Transdiagnostic study
key for clinicians and their patients, with implications for clinical
evaluation and treatments. The factor analysis studies imply that it may be useful to study
the five negative symptoms separately. Study of individual nega-
tive symptoms fits with an important research approach, the
‘transdiagnostic’ study of areas of psychopathology. To give
The factor structure of negative symptoms one example, psychotic symptoms are transdiagnostic as they
are found in a variety of illnesses, including schizophrenia,
There is a growing consensus that there are five types or domains of bipolar disorder and some forms of dementia. Pharmacological
negative symptoms: alogia (poverty of speech), blunted affect, avoli- treatment of psychosis also has some efficacy across these disor-
tion, asociality and anhedonia (decreased experience of pleasure).1 ders. Does psychosis also have biological underpinnings that are
New negative symptom rating scales, the Brief Negative Symptom transdiagnostic? Negative symptoms also occur in disorders
Scale (BNSS) and the Clinical Assessment Interview for Negative other than schizophrenia, for instance anhedonia is found in
Symptoms (CAINS), were developed in response to this consensus. depression. Researchers are investigating whether negative symp-
However, the relationships among these five domains are under toms share common correlates such as genetics, treatment
debate. This issue has been explored using factor analysis, a statis- response and functional circuits.
tical method that defines groups or ‘factors’ of item scores that The current issue of the BJPsych has an example of a transdiag-
tend to correlate with each other. Initial studies using exploratory nostic study4 of a single domain of negative symptoms, anhedonia.
factor analysis found two factors, one comprising expressivity The authors found evidence for a transdiagnostic risk factor: people
(alogia and blunted affect) and the second comprising the other who reported childhood trauma had an increased risk of anhedonia
three negative symptoms. However, confirmatory factor analysis in adulthood whether they were depressed or – a separate group –
(CFA), which permits the testing of alternative hypotheses about were at clinical high risk of psychosis.
the factors within a scale, has most frequently found factors that
reflect the five domains listed above. These five factors were
found across the interview-based negative symptom rating scales, The limitations of rating scales
comprehensive self-report questionnaires, diverse cultures/lan-
guages (Eastern and Western), both genders, multiple phases of Negative symptom rating scales have an inherent limitation that can
illness (clinical high risk, first episode, chronic) and different statis- lead to ambiguity in the interpretation of study results. Consider the
tical techniques.2 There is some covariation across these factors, but example of asociality. A person may not socialise because he or she
it is limited and people usually do not have significant impairment is depressed, paranoid, anxious, disorganised, etc., and so has a ‘sec-
in all five factors. ondary’ negative symptom. Alternatively, the person may simply

271
https://doi.org/10.1192/bjp.2023.68 Published online by Cambridge University Press
 Kirkpatrick et al

have a lack of interest in social relationships that cannot be attribu-

ted to these other problems, and so has a ‘primary’ negative Brian Kirkpatrick , MD, MSPH, Psychiatric Research Institute, University of Arkansas
for Medical Sciences, Little Rock, Arizona, USA; Lauren Luther, PhD, Department of
symptom. A change in asociality on a negative symptom rating Psychology, University of Georgia, Athens, Georgia, USA; Gregory P. Strauss , PhD,
Department of Psychology, University of Georgia, Athens, Georgia, USA
scale score is therefore ambiguous: is the change due to increased
interest in others or due to an improvement in one of these other Correspondence: Brian Kirkpatrick. Email: [email protected]
problems, such as anxiety? The same ambiguity arises with all of First received 6 Mar 2023, final revision 9 Apr 2023, accepted 10 May 2023
the negative symptoms. The clinician who encounters evidence of
improvement in negative symptoms, or reads about improvement
in those symptoms in a treatment trial, should consider whether sec-
ondary symptoms also improved. Unfortunately, to date there is Data availability
little evidence for an effective pharmacological treatment of Data availability is not applicable to this article as no new data were created or analysed in this
primary negative symptoms. study.

Clinical rating scales have other problems as well. Their validity

depends on a patient’s memory of symptoms over the previous days
Author contributions
to weeks, awareness of impairment, willingness to report symptoms
and behaviours, and other factors. Moreover, all raters have their B.K. led the writing and conceptual framework of the manuscript. L.L. and G.P.S. contributed to
drafting and revising the manuscript and provided conceptual guidance.
biases and difficulties in making ratings, which can decrease the reli-
ability and validity of ratings. A method that overcomes some of the
limitations inherent to symptom rating scales is digital phenotyping, Funding
which uses mobile technology such as smartphones and wearable
This work was supported by National Institute of Mental Health (NIMH) grants R21-MH122863,
devices to collect data during everyday life. Digital phenotyping R01-MH116039, R01-MH120092 and R61-MH121560 to G.P.S.
data can be grouped into ‘active’ and ‘passive’ data collection
methods.5 Active data collection requires users to complete a task
such as a survey or a video ‘selfie’, whereas passive approaches Declaration of interest
involve unobtrusive data collection that occurs automatically (e.g. G.P.S. and B.K. are original developers of the Brief Negative Symptom Scale (BNSS) and receive
via sensors in wearables). Several active and passive digital pheno- royalties and consultation fees from Medavante-ProPhase in connection with commercial use
of the BNSS and other professional activities; these fees are donated to the Brain and Behavior
typing measures have shown promise as measures of negative symp- Research Foundation. G.P.S. and B.K. are also part owners and co-founders of Quantic
toms, including geolocation (GPS coordinate data that show Innovations, which provides digital phenotyping data collection, analysis and interpretation
and has contracts with Karuna and Sunovion. B.K. has received honoraria and travel support
location and location changes), accelerometry (measures of move- from ProPhase LLC for training pharmaceutical company raters on the BNSS, consulting fees
ments in three dimensions) and – using audio and video recordings and travel support from Genentech/Roche, Minerva Neurosciences, Lundbeck, Acadia,
Karuna, Otsuka and Medavante-ProPhase, consulting fees from anonymised pharmaceutical
– natural language processing and automated analysis of facial companies and investors through Decision Resources, Inc., and Guideposts, and from
expressions and vocal characteristics.5 There is preliminary evi- Wockhardt Bio AG for consulting on a legal issue. G.P.S. has consulted for Minerva
Neurosciences, Acadia, Otsuka, Sunovion, Boeringher-Ingelheim, Karuna and Lundbeck.
dence that the five negative symptoms can be distinguished by
digital measures.
References

1 Kirkpatrick B, Fenton WS, Carpenter WT, Marder SR. The NIMH-MATRICS con-
Conclusions sensus statement on negative symptoms. Schizophr Bull 2006; 32: 214–9.
2 Strauss GP, Ahmed AO, Young JW, Kirkpatrick B. Reconsidering the latent
In the past 20 years, the concepts and assessments of negative symp- structure of negative symptoms in schizophrenia: a review of evidence sup-
porting the 5 consensus domains. Schizophr Bull 2019; 45: 725–9.
toms have changed substantially: a new consensus on which features
3 Strauss GP, Zamani Esfahlani F, Sayama H, Kirkpatrick B, Opler MG, Saoud JB,
should be considered negative symptoms; ratings scales based on
et al. Network analysis indicates that avolition is the most central domain for
this consensus; recognition of the factor structure of negative symp- the successful treatment of negative symptoms: evidence from the roluperi-
toms, and the implications of these factors for research and possibly done randomized clinical trial. Schizophr Bull 2020; 46: 964–70.
treatment; wider recognition of the distinction between primary and 4 O’Brien KJ, Ered A, Korenic SA, Olino TM, Schiffman J, Mittal VA, et al. Childhood
secondary negative symptoms; and the development of digital mea- trauma, perceived stress and anhedonia in individuals at clinical high risk for
psychosis. Br J Psychiatry this issue [Epub ahead of print] 5 Jan 2023. Available
sures. As concepts and assessment tools in part determine the treat- from: https://doi.org/10.1192/bjp.2022.185.
ments patients receive, these changes in concepts and measurement
5 Harvey PD, Depp CA, Rizzo AA, Strauss GP, Spelber D, Carpenter LL, et al.
hold promise for improving the assessment and treatment of nega- Technology and mental health: state of the art for assessment and treatment.
tive symptoms. Am J Psychiatry 2022; 179: 897–914.

272
https://doi.org/10.1192/bjp.2023.68 Published online by Cambridge University Press