Schizophrenia Research Forum - A Catalyst for Creative Thinking

ICOSR 2009—Social Outcome and Schizophrenia: Back to the Future

Editor's Note: Over the next weeks, we will be bringing you reports from the International Congress on Schizophrenia Research (ICOSR), 29 March to 1 April 2009, and the satellite Mt. Sinai Conference on Cognition in Schizophrenia, 27-28 March, in San Diego, California. In the case of the ICOSR, we are very grateful to organizers Carol Tamminga and Chuck Schulz, who helped us recruit some roving reporters from among the Young Investigator awardees. A special thanks also to Laura Rowland and Scott Sponheim, who directed the Young Investigator program. Our first news story is from Amy Pinkham of the University of Pennsylvania.

1 April 2009. On 30 March 2009, Kristin Cadenhead of the University of California, San Diego, chaired a morning session featuring informative presentations from several members of the North American Prodrome Longitudinal Study (NAPLS) consortium, which is under the direction of Robert Heinssen at NIMH. Cadenhead and the other speakers—Jean Addington (University of Toronto), Barbara Cornblatt (Zucker Hillside Hospital), and Elaine Walker (Emory University)—emphasized three primary conclusions:

1. Impairments in social functioning are prominent in individuals who are at risk of developing schizophrenia and are even more pronounced in those individuals who do convert to psychosis.

2. These impairments are stable over time and do not worsen after formal illness onset.

3. The identification of these impairments highlights the necessity of targeted interventions that specifically focus on social and role functioning and that can be implemented in high-risk groups during the prodromal phase.

Jean Addington presented first and began by explaining the concept of the prodromal period in the development of schizophrenia, which is marked by functional decline and in some cases the appearance of attenuated or brief psychotic symptoms. With proper identification of individuals who are at risk of developing schizophrenia, this period allows a valuable opportunity for early intervention and ameliorative change. Addington reported results from three different investigations focusing on social functioning during the prodromal period. The first investigation utilized data from the NAPLS consortium and cluster analyses to identify four patterns of social functioning in individuals who were identified as being at risk based on clinical presentation. Identified patterns included stable-good, stable-intermediate, deteriorating, and poor-deteriorating. Addington noted that individuals in the deteriorating and poor-deteriorating groups were more likely to have a more severe clinical presentation but that membership in either of these groups was not associated with an increased rate of conversion to psychosis. In the second investigation, Addington used data from the PREDICT (Prodromal Research for Early Detection in a Collaborative Team) study to compare social functioning in individuals at clinical high risk (CHR), individuals with a first episode of psychosis, individuals with chronic psychosis, and non-clinical controls. All three clinical groups showed significantly lower levels of social functioning than controls, and more strikingly, CHR individuals showed comparable levels of functioning relative to first episode and chronic individuals. The last investigation highlighted the potential role of cognitive and environmental factors in social impairment by demonstrating that CHR individuals show increased levels of social defeat and negative self schema. Addington summarized by noting that impaired social functioning is the most devastating feature of the prodromal period, that social functioning impairments are prominent in CHR regardless of eventual conversion, and that understanding social functioning impairments will be useful for investigating the underlying pathophysiology of psychosis and for developing effective interventions.

Barbara Cornblatt presented next and first discussed the neurodevelopmental model upon which prodromal work is based. This model postulates that genetic vulnerabilities and in utero insults are reflected in every stage of development and that social functioning abnormalities are the downstream results of abnormal neural development. Cornblatt then presented findings from a subset of the NAPLS consortium data in which social and role functioning were assessed in individuals who met criteria for risk based on the presence of attenuated positive symptoms and who had follow-up data over a two and a half year period. In this study, individuals who converted to psychosis at any point during the follow-up period were identified and then matched to at-risk individuals who did not convert to psychosis during the follow-up period. A total of 50 converter-non-converter pairs was identified, and social and role functioning at baseline was compared between these groups. For social functioning, converters showed lower levels of functioning than non-converters at baseline and all follow-up time points. Notably, in both groups, social functioning ability was stable over time, and for converters, did not decline further after conversion occurred. Results were similar for role functioning, indicating that converters showed impaired role functioning relative to non-converters and that these impairments were stable and not impacted by conversion. In her closing remarks, Cornblatt emphasized the finding that both social and role functioning appear to be independent from the onset of psychosis and that these impairments therefore require early and specific intervention.

Kristin Cadenhead then reviewed previous work demonstrating that in addition to social functioning impairments, CHR individuals show impairments in neurocognition relative to healthy individuals. Directly following from these findings, she sought to determine whether baseline neurocognitive abilities, symptomatology, and social functioning were predictive of social functioning outcome at one-year follow-up. Participants were individuals from the Cognitive Assessment and Risk Evaluation (CARE) program at UCSD. Consistent with data from other presenters, CHR individuals failed to show a significant improvement in social functioning from baseline to follow-up, and no changes in social functioning were associated with conversion to psychosis. Multiple regression analyses also demonstrated that baseline executive functioning abilities and symptoms of disorganization were significant predictors of social functioning ability at follow-up. Cadenhead concluded that these associations between baseline neurocognitive abilities and disorganized symptoms and later functional outcome provide some insight into potential targets for early interventions that may positively alter the course of illness.

Elaine Walker concluded the symposium by presenting data from an investigation of the relationship between cortisol levels and role functioning in CHR individuals. Participants were individuals from the Emory Prodrome Project, and social and role functioning as well as cortisol levels were assessed at baseline, 7-10-month follow-up, and one-year follow-up. Of the 43 participants, 13 converted to psychosis after the follow-up period, and as reported in the other presentations, role functioning at baseline and one-year follow-up was worse for the individuals who would convert and stable over time in both converters and non-converters. Changes in cortisol levels were assessed by calculating area under the curve (AUC) indices, which allowed for comparisons of absolute levels of cortisol as well as increases in cortisol levels relative to baseline. Analyses of these indices identified higher levels of cortisol in individuals who would convert relative to non-converters at both 7-10-month and one-year follow-up. Additionally, increases in cortisol levels from baseline were significantly and positively correlated to role function deficits at baseline and one-year follow-up. Finally, Walker also presented an additional analysis that investigated whether increased cortisol levels were related to conversion. This analysis revealed that cortisol levels mediated the relationship between role functioning and conversion, suggesting that stress-induced increases in HPA activity may be superimposed on normative developmental increases and may hasten or trigger conversion to psychosis.—Amy Pinkham.

Comments on News and Primary Papers
Comment by:  Kathryn Abel
Submitted 22 April 2009
Posted 5 May 2009

It is impossible to understand the meaning of cortisol correlations in the summary from the final presentation without knowing more about those data. It looks as though they were cross-sectional measures in women and men at high risk, but the summary does not comment on their ages or menstrual phases or other data on socioeconomic status, smoking, drug, alcohol use or time of day of measures (although I'm sure the investigators know all this information).

Place in social hierarchy is an interesting confound of HPA responses, as well as baseline measures, in human and other animal populations and possibly important/relevant to schizophrenia.

View all comments by Kathryn AbelComment by:  Amresh Shrivastava
Submitted 16 July 2009
Posted 16 July 2009

Perspectives in social outcome of schizophrenia
The ongoing discussion about social outcome is exciting and timely. Repeatedly it has been demonstrated that only clinical remission does not reflect “real-life” situation (Eberhard et al., 2009; Cassidy et al., 2009). The decline in social functions not only forms a pathognomonic symptom of schizophrenia but also serves as a marker and predictor for outcome status. Further, social outcome in early psychosis also serves as an indicator, which helps differentiate various syndromes amongst the broad classification of “non-affective psychosis” or “schizophrenia spectrum disorder” (Horan et al., 2009). The early intervention research for at-risk individuals is faced with three main difficult objectives:

1. To identify whether the individual will transition to florid psychotic state, no matter of what diagnostic category.

2. To find measures which can contain the transition and restore the psychosocial functions.

3. To develop clinical markers which can predict with high reliability and specificity chances of developing hardcore schizophrenia.

The appropriateness of multidimensional measurement of outcome needs to be investigated. We suggest that it needs to measure at least on two different parameters: clinical and social.

Research in the early phase of psychosis has developed evidence for early intervention, particularly “critical period” (Birchwood et al., 1998; Cassidy et al., 2009). It provides scientific and biological arguments for early identification and treatment. Furthermore, it has been suggested that the critical period should be extended to include prodromal as well as early phase of psychosis. There is more to “early intervention” in psychosis than merely intervening early.

The long-term outcome of schizophrenia should answer these two questions:

1. What do we do to maximize outcome and achieve complete social integration, and when?

2. How can we identify subjects who may not respond favorably to treatments and facilitate planning for impending disability right from the beginning?

Social outcome
In the literature, social recovery (overall good outcome) is measured by different social and occupational parameters and is found to be around 40-60 percent (Priebe, 2007). One of our long-term follow-up studies (Srivastava, A, Thakar, M, unpublished) attempted to identify the effects of multiple outcome criteria in 10 years’ long-term follow-up of first-episode hospitalized patients. The social parameters included quality of life, level of functioning, independent living, interpersonal social functioning, work and employment, presence of family burden, and social burden (Meltzer et al., 1997). We observed that outcome status declines if both clinical and social parameters are included: 61 percent of patients showed clinical remission, 32 percent showed good social remission, and only 25 percent improved on both clinical and social parameters. Just 10 percent of the patients fulfilled a maximum of four parameters. The largest group of patients (25.7 percent) recovered on only two parameters. Social parameters appear to be inter-dependent, yet recovery on any one parameter is not associated with recovery on any other parameter. Therefore, some patients are employed but unable to live independently and others show good quality of life but are still unable to be gainfully employed. In this study it is observed that only 23-25 percent of patients show social recovery on two to three different parameters.

Conceptual framework
The outcome criteria have been ill defined for long-term studies. Commonly used outcome terminology has been “poor outcome," "good outcome," "favorable outcome," "unfavorable outcome," "hospitalization," "repeated relapse rate," "living with family," "return to education," "employment," and "marital status." It is difficult to arrive at a conclusion for minimum criteria for defining “good outcome” in the long term. Recently there have been a few attempts to look at this aspect scientifically (Addington et al., 2004; Crumlish et al., 2009), particularly from the working group of the American Psychiatric Association on defining criteria for remission, which has developed a consensus statement about “remission” in schizophrenia. (Andreasen and Olsen, 1982) Based on an examination of popular rating scales (SAPS and SANS, PANSS, and BPRS), the working group identified appropriate criteria to serve as the basis for defining “symptomatic remission” in schizophrenia. In addition, the working group proposed that remission criteria might be described separately for positive and negative symptoms, to allow primary consideration of these symptom groups independently in the assessment of symptomatic remission. Specific items selected for consideration as criteria for remission in schizophrenia were chosen to map the three dimensions of psychopathology identified by factor analyses: 1) psychoticism; 2) disorganization; and 3) negative symptoms. This represents the conceptualization and quantification of remission. It is noteworthy that remission as an outcome measure is far less than what is required to capture real-life outcome. There are expectations from the families and the public at large to define the outcome in a real-life situation.

We therefore propose that outcome needs to be measured on a minimum of two groups of parameters, each having three or more parameters in the least: 1) clinical outcome that measures psychopathology, side effects, and hospitalization. The psychopathology includes aggression, suicidality, and depression besides parameters of “psychoticism”; 2) social outcome, which measures quality of life, global assessment of functioning, and social cognition. The parameter of social cognition includes ability to learn and reproduce, coping, independent living, and interpersonal skills.

We need to investigate dimensions of social decline and social improvements that are objective, definable, and quantifiable. Thus, the ongoing work of social outcome is very significant.


Addington J, Van Mastrigt S, Addington D. Duration of untreated psychosis: impact on 2-year outcome. Psychol Med. 2004 Feb 1;34(2):277-84. Abstract

Andreasen NC, Olsen S. Negative v positive schizophrenia. Definition and validation. Arch Gen Psychiatry. 1982 Jul 1;39(7):789-94. Abstract

Birchwood M, Todd P, Jackson C. Early intervention in psychosis. The critical period hypothesis. Br J Psychiatry Suppl. 1998 Jan 1;172(33):53-9. Abstract

Cassidy CM, Norman R, Manchanda R, Schmitz N, Malla A. Testing Definitions of Symptom Remission in First-Episode Psychosis for Prediction of Functional Outcome at 2 Years. Schizophr Bull. 2009 Mar 25; Abstract

Crumlish N, Whitty P, Clarke M, Browne S, Kamali M, Gervin M, McTigue O, Kinsella A, Waddington JL, Larkin C, O”Callaghan E. Beyond the critical period: longitudinal study of 8-year outcome in first-episode non-affective psychosis. Br J Psychiatry. 2009 Jan 1;194(1):18-24. Abstract

Eberhard J, Levander S, Lindström E. Remission in schizophrenia: analysis in a naturalistic setting. Compr Psychiatry. 2009 May-Jun ;50(3):200-8. Abstract

Horan WP, Kern RS, Shokat-Fadai K, Sergi MJ, Wynn JK, Green MF. Social cognitive skills training in schizophrenia: an initial efficacy study of stabilized outpatients. Schizophr Res. 2009 Jan 1;107(1):47-54. Abstract

Meltzer HY, Rabinowitz J, Lee MA, Cola PA, Ranjan R, Findling RL, Thompson PA. Age at onset and gender of schizophrenic patients in relation to neuroleptic resistance. Am J Psychiatry. 1997 Apr 1;154(4):475-82. Abstract

Priebe S. Social outcomes in schizophrenia. Br J Psychiatry Suppl. 2007 Aug 1;50():s15-20. Abstract

Srivastava, Amresh and Thakar, Meghan, Outcome in Schizophrenia: The Long-Term Good Outcome in Schizophrenia Is Not Yet Good Enough (2008). Psychiatry Presentations. Paper 7.

View all comments by Amresh Shrivastava