Welcome to the Schizophrenia Resource Centre

Welcome, this website is intended for healthcare professionals in EMEA with an interest in the treatment of schizophrenia. By clicking the link below you are declaring and confirming that you are a healthcare professional

You are here

Duration of untreated psychosis and negative symptoms — A systematic review and meta-analysis of individual patient data

Schizophrenia Research, 1-3, 142, pages 12 - 19



Longer duration of untreated psychosis (DUP) is associated with poorer outcome in terms of positive symptoms, relapse rate, and time to remission. In contrast, the association with negative symptoms is less consistent.


The study had three aims. First, to arrive at a more precise estimate of the correlation between DUP and negative symptoms than previous reviews, by substantially increasing the amount of available data. Second, to see whether the strength of this correlation attenuated over longer follow-up intervals. Third, to determine whether there is a relationship between DUP and changes in negative symptoms.


Relevant databases were searched for studies published between December 1992 and March 2009 that reported data on DUP and negative symptoms. We obtained individual patient data where possible and calculated summary correlations between DUP and negative symptoms for each study at baseline, short and long-term follow-up. We used multilevel regression analysis to examine whether the effect of DUP on negative symptoms was the greatest in the early stages of illness.


We included 28 non-overlapping studies from the 402 papers detected by the search strategy. After contacting the authors we obtained individual patient data from 16 of these studies involving 3339 participants. The mean DUP was 61.4 weeks (SD = 132.7, median DUP = 12.0). Shorter DUP was significantly associated with less severe negative symptoms at baseline and also at short (1–2 years) and longer term follow-up (5–8 years) (r = 0.117, 0.180 and 0.202 respectively, p < 0.001). The relationship between improvement in negative symptoms and DUP was found to be non-linear: people with a DUP shorter than 9 months showed substantially greater negative symptom reduction than those with a DUP of greater than 9 months.


Shorter DUP is associated with less severe negative symptoms at short and long-term follow up, especially when the DUP is less than 9 months. Since there is no effective treatment for negative symptoms, reducing DUP to less than 9 months may be the best way to ameliorate them.

Keywords: First episode psychosis, Schizophrenia, Treatment delay, Negative symptoms.

1. Introduction

Negative symptoms are a core component of the schizophrenia syndrome and are commonly described in terms of five dimensions: blunted affect, alogia, anhedonia, avolition and asociality (Blanchard et al, 2005, Kirkpatrick et al, 2006, and Makinen et al, 2008). Negative symptoms are associated with poor functional outcome ( Malla et al., 2004 ), cognitive deficits ( Heydebrand et al., 2004 ), social dysfunction and poor quality of life (Addington and Addington, 1993, Schmitz et al, 2007, and Petersen et al, 2008). Negative symptoms are common: the prevalence in short-term follow-up studies (up to 2.5 years) is about 45% (Pogue-Geile and Harrow, 1985, Malla et al, 2002, and Malla et al, 2004), and in longer term studies (7.5–10 years) 20–30% ( Herbener and Harrow, 2001 ).

There is no established treatment for primary negative symptoms (Kirkpatrick et al, 2006 and Buckley and Stahl, 2007). Pharmacological treatments, such as antipsychotics have only a marginal impact on negative symptom severity ( Leucht et al., 2011 ). Some psychosocial treatment trials have shown an effect on negative symptoms. Cognitive behavioral therapy showed a positive effect on negative symptoms up to 24 months of follow-up ( NICE, 2010 ). In addition there have been three small but promising trials on peer support groups ( Castelein et al., 2008 ), music therapy ( Gold et al., 2009 ) and body oriented psychosocial therapy ( Rohricht and Priebe, 2006 ).

Pending replication of these studies, it may be that the best way to deal with negative symptoms apart from cognitive therapy is prevention. Many countries have already adopted an early intervention approach to the treatment of psychosis on the basis that there is a robust association between duration of untreated psychosis and the long-term severity of positive symptoms. A similar association between negative symptoms and outcome would support early intervention for negative symptoms ( Melle et al., 2008 ).

So far two meta-analyses have examined the correlation between DUP and negative symptoms (Marshall et al, 2005 and Melle et al, 2008). The first by Marshall et al. reported an association at 6 and 12 months but not at baseline and 24 months ( Marshall et al., 2005 ). The second by Perkins et al. claimed that patients with shorter DUP experienced less negative symptoms at baseline and follow-up ( Perkins, 2006 ). While making a valuable contribution, both reviews share three limitations. First, neither study provided a precise analysis of the effect of DUP at long term follow-up, thus Marshall et al. did not consider data on negative symptoms beyond 24 months and Perkins et al. summarized data of all follow-up assessments varying from 3 months to 15 years into one combined effect size. Second, both studies relied exclusively on published data, which meant that they excluded a substantial number of studies that collected pertinent data but did not publish it in a format that could be used in their analyses. Third, both studies used combined correlation coefficients calculated by different techniques, for example parametric in some studies and non-parametric in others.

As a consequence of these limitations, neither study was able to consider the relationship over time i.e. the linearity of the relationship between DUP and negative symptoms. So far most studies did focus on the relationship between DUP and positive symptoms ( Drake et al., 2000 ). Not only a positive association was found in most studies but also a non-linearity of this relationship between DUP and positive symptoms. This means that in a patient with a DUP of less than 12 months, the effect of a DUP reduction of for example 2 months on positive symptoms would be greater than a two month reduction in someone with a DUP of more than 12 months. It is possible therefore that a similar non-linear relationship might exist with negative symptoms, with important clinical implications.

The present study therefore had three aims: (a) to provide a precise estimate of the correlation between DUP and negative symptoms by substantially increasing the amount of available data contributing to the analysis; (b) to see if the strength of any correlation attenuates at longer follow up intervals; (c) to determine whether the relationship between DUP and negative symptoms was non-linear.

2. Methods

2.1. Search strategy

The search aimed to detect all cohort studies that had examined DUP and negative symptoms in first episode psychotic patients, and were available for review up to March 2009. The search included the databases MEDLINE and PUBMED using the keywords “DUP” AND “psychosis” OR “schizophrenia” and the combination “duration” AND “untreated” AND “psychosis” OR “schizophrenia”. In addition references cited in these papers were examined.

2.2. Inclusion criteria and data extraction

Studies were included if they met two criteria. First, participants had experienced a first episode of psychosis defined as: schizophrenia, schizoaffective disorder, schizophreniform disorder, delusional disorder, brief psychotic disorder or psychosis NOS, according to either the Diagnostic Statistical Manual (DSM-IV) or the International Classification of Diseases (ICD-10) classification systems. Second, DUP and negative symptoms had been measured using a standardized method (which for negative symptoms was one of the three commonly used scales in this area), i.e.: the Positive and Negative Symptom Scale (PANSS) ( Kay et al., 1987 ), the Scale for the Assessment of Negative Symptoms (SANS) ( Andreasen, 1989 ), or the Brief Psychiatric Rating Scale (BPRS) ( Overall and Gorham, 1962 ). We included all follow-up studies regardless of research design.

Abstracts were screened independently by two reviewers (N.B. and R.K.), and copies were obtained of any papers describing potentially eligible studies. Each paper was assessed independently by two reviewers and any disagreements resolved by discussion with a third reviewer (L.W.) (for table of included studies see Appendix 1 ). To overcome the limitations of previous meta-analyses it was necessary to obtain a substantial dataset of individual patient data (IPD) from first-episode studies (Stewart and Parmar, 1993 and Stewart and Tierney, 2002). The authors of included studies were contacted and asked to provide anonymized individual patient data on: gender, age at onset, DUP, and negative symptom scores at baseline and all available follow-up points.

2.3. Analysis

We calculated the correlation between DUP and negative symptoms for short and long term follow up. We defined short-term follow up as between 12 and 24 months and long term follow up as 60 to 96 months. Studies of which a Spearman correlation was published were used for sensitivity analysis.

In cases where more than one assessment was available for short or long term follow up, we used the first assessment. The data were analyzed using a two-step approach. First, individual patient data from each study were analyzed using the non-parametric Spearman's Rank Correlation. We choose a non-parametric test because of the significant positive skew in the distribution of DUP. In the second step, a meta-analysis of the aggregated data produced for each study was performed using Comprehensive Meta‐Analysis in order to summarize the correlations (CMA, version 2005, Biostat, Inc., Englewood, NJ, 2005) ( Borenstein et al., 2009 ). We used random effect models because of great differences between research designs. Correlation data were synthesized with Fisher's z transformation into a single correlation coefficient (r) with 95% confidence intervals (CI).

CMA also tests for the heterogeneity of the sample populations using I2, a test parameter, which evaluates the null hypothesis that all studies are assessing the same effect size. I2 indicates the percentage of total variation across studies due to heterogeneity rather than chance, and ranges from 0% (no heterogeneity) to 100% (high heterogeneity). Values for I2 of 25%, 50% and 75% are considered to represent low, moderate and high heterogeneity respectively ( Higgins et al., 2003 ).

A major concern when conducting a meta-analysis is publication bias: studies that report relatively large effect sizes are more likely to be published than studies with smaller effect sizes. If the included studies are a biased sample of all relevant studies, consequently the bias will be reflected in the mean effect computed by the meta-analysis. A funnel plot can be made to examine the publication bias using Comprehensive Meta-Analysis (CMA, version 2005, Biostat, Inc., Englewood, NJ, 2005).

We examined the linearity of the relationship between DUP and negative symptom change using the method developed by Drake ( Drake et al., 2000 ). We used a multilevel regression analysis (XTREG in STATA release 11.0) (Snijders and BRJ, 1999 and StataCorp, 2009) to estimate the magnitude of the effect of DUP on negative symptom severity reduction after various periods of follow-up with the individuals nested within each study. Subsequently, explanatory variables (gender and age at onset) were added, as potential confounders. The skewed distribution of DUP was corrected by converting DUP to its logarithm (logDUP) thus allowing the use of parametric statistics. The change of negative symptoms from baseline to follow-up was calculated as a percentage of baseline negative symptom severity and added in the multilevel regression analysis as the dependent variable. We conducted the analysis for negative symptom change at short and long-term follow up, as defined above.

3. Results

3.1. Studies and patients

A total of 402 papers were identified by the search strategy, from which we identified 28 non-overlapping first episode studies that met inclusion criteria (see Appendix 1 ) (Montague et al, 1989, Barnes et al, 2000, Craig et al, 2000, de Haan et al, 2000, Drake et al, 2000, Ho et al, 2000, Larsen et al, 2000, Black et al, 2001, Verdoux et al, 2001, Cullberg et al, 2002, Malla et al, 2003, Addington et al, 2004, Sim et al, 2004, Chen et al, 2005, Harris et al, 2005, Manchanda et al, 2005, Melle et al, 2005, Oosthuizen et al, 2005, Petersen et al, 2005, Wade et al, 2005, Clarke et al, 2006, Ucok et al, 2006, Wunderink et al, 2006, Crespo-Facorro et al, 2007, Malla et al, 2007, Vyas et al, 2007, Gorna et al, 2008, and Yamazawa et al, 2008). The selected studies included a total of 3998 participants. Following written requests, authors of 16 studies submitted their data sets that covered a total of 3339 participants. From the 12 studies of which individual patient data were not available, 2 studies reported a Spearman correlation at short term follow-up (1–2 years). Table 1 shows the distribution of DUP, gender and age at onset of participants of whom individual patient data were available at baseline. The follow-up data of 2 studies were incomplete and therefore only baseline data were used in our analysis (Montague et al, 1989, Malla et al, 2003, Melle et al, 2004, Manchanda et al, 2005, and Malla et al, 2007).

Table 1 Summary of the 16 included first episode studies.

Study inclusion period N

Gender Age at onset (SD) Mean DUP in weeks

Median DUP

in weeks
N at short term follow-up

(12–24 months)
N at long term follow-up

(60–96 months)
Addington ( Addington et al., 2004 )

(Jan. 1997–Dec. 2000)
240 161 79 23.4 (6.7) 26.4 (10.3) 77.1 (134.0) 26.0 199 50
Barnes ( Barnes et al., 2000 )

(Feb 1995–Oct 2001)
98 75 23 23.5 (6.6) 27. 2 (8.3) 54.0 (88.1) 20.0 96
Chen ( Chen et al., 2005 )

(Sept. 1997–March 2000)
93 42 51 29.0 (9.2) 33.0 (9.7) 44.4 (135.7) 25.7 93
Clarke ( Clarke et al., 2006 )

(Febr.1995–Febr. 1999)
152 96 56 27.8 (6.4) 26.6 (8.3) 71.0 (115.9) 23.6 94
Craig ( Craig et al., 2000 )

354 194 160 28.6 (15.8) 30.3 (11.9) 66.4 (186.2) 3.6 258
Cullberg ( Cullberg et al., 2002 )

(Jan. 1996–Dec. 1997)
175 97 78 25.3 (6.4) 29.4 (7.1) 66.4 (163.2) 3.6 122

Drake ( Drake et al., 2000 )

(July 1996–aug. 1998)
255 176 79 28.0 (9.9) 32.2 (10.7) 38.7 (82.2) 12.0 182
Gorna ( Gorna et al., 2008 )

86 34 52 26.6 (7.0) 23.6 (4.6) 40.4 (57.0) 12.9 83 74
Harris ( Harris et al., 2005 )

(March 1989–July 1997)
318 216 102 21.7 (4.1) 23.9 (5.2) 24.1 (50.1) 5.8 295 135
Malla ( Malla et al., 2007 )

(January 1997–July 2003)
153 116 37 22.8 (7.2) 28.9 (12.0) 70.9 (110.3) 23.9
Manchanda ( Manchanda et al., 2005 )

(May 1996–Dec. 2001)
122 93 29 22.1 (6.3) 28.8 (11.2) 70.7 (120.6) 21.6 122
Melle ( Melle et al., 2004 )

301 176 125 25.1 (7.9) 29.4 (11.2) 47.4 (117.7) 9.0 271
Montague ( Montague et al., 1989 )

109 51 58 26.7 (6.8) 28.1 (8.3) 28.9 (88.1) 20.0
Petersen ( Petersen et al., 2005 )

(Jan. 1998–Dec. 2000)
578 235 343 24.3 (6.6) 24.5 (6.1) 106.0 (177.2) 34.4 394 247
Ucok ( Ucok et al., 2006 )

(> 1996)
148 84 64 23.3 (6.2) 25.6 (9.3) 32.8 (39.5) 17.1 54
Wunderink ( Wunderink et al., 2006 )

(Okt. 2001–Dec. 2002)
157 113 44 25.6 (6.0) 28.8 (7.2) 46.0 (111.6) 4.4 120 105
Overall 3339 1959 1380 25.0 (8.6) 27.3 (9.2) M 61.0 (125.5)

F 61.9 (142.4)
13.3 1268



M = male.

F = female.

The mean age at onset of participants was 26.0 years (SD = 8.9), and the mean DUP was 61.4 weeks (SD = 132.7 weeks, median DUP = 12.0 weeks). 59% of the participants at baseline were males.

4. Association between duration of untreated psychosis and negative symptoms

Fig. 1 shows a summary of the correlations between DUP and negative symptoms at baseline and short (1–2 years) and long-term follow-up (5–8 years). The data show a statistically significant positive correlation between DUP and negative symptoms at baseline (Fisher's z = 0.117, 95%CI 0.064–0.17), at short-term follow-up (Fisher's z = 0.18, 95%CI 0.086–0.274) and at long-term follow-up (Fisher's z = 0.202, 95%CI 0.137 − 0.267). There is no evidence for attenuation in the strength of the association with longer follow up. Tests for heterogeneity showed that the studies were low to moderately heterogeneous (I2 = 48% at baseline, 75.9% at short term, 27.3% at long term follow-up). A sensitivity analysis was done by adding the reported Spearman correlations to the meta-analysis at short term follow-up from the two studies of which individual patient data were not obtained (Sim et al, 2004, Oosthuizen et al, 2005, and Subramaniam et al, 2007). These two additional studies did not substantially alter the reported correlations of heterogeneity statistics (I2 = 71.8%).


Fig. 1 Correlations between DUP and negative symptoms at baseline, after 1-2 years follow-up and after 5-8 years follow-up.

5. Predicted negative symptom change

2150 patients nested in 12 studies were included in the multilevel analysis to explore the effect of DUP on negative symptom reduction at short-term follow-up. The mean follow-up assessment was 15.5 months (95%CI 15.3–15.7). In the multilevel analysis for long-term follow-up, 795 individuals nested in 7 studies were included and the mean follow-up assessment was 71.4 months (95%CI 70.2–72.7).

Fig 2 and Fig 3 show the percentage change in negative symptoms predicted by a given DUP at short and long-term follow-up respectively.


Fig. 2 Predicted change in negative symptoms after short term, follow-up (1–2 years) against duration of untreated psychosis in weeks (with 95% CIs).


Fig. 3 Predicted change in negative symptoms after longer term, follow-up (5–8 years) against duration of untreated psychosis in weeks (with 95% CIs).

6. Discussion

The analysis showed that there is a significant positive association between DUP and negative symptoms at: baseline, short (1–2 years) and long-term (5–8 years) follow-up. The effect is non-linear: negative symptoms and DUP are positively associated if DUP was less than 9 months. For those patients with a DUP longer than 9 months negative symptoms are not associated linearly. A reduction in DUP in someone with a shorter DUP (i.e. less than 9 months) might therefore have a greater impact on negative symptoms than the same reduction in someone with a DUP greater than nine months. These findings are in accordance with the findings of the TIPS study and the study of (Malla et al, 2002) and (Malla et al, 2004), both of which suggested that reduction of DUP may be as important for improving the severity of negative symptoms as it is for positive symptoms ( Melle et al., 2008 ). Our finding of a non-linear association between DUP and negative symptoms is similar to that reported by Drake et al. for total PANSS scores ( Drake et al., 2000 ) albeit in a small sample with a short follow-up.

The association between a longer DUP and persistence of negative symptoms (after 1–2 years) is consistent with the hypothesis that in many cases psychosis is a clinical manifestation of a progressive pathological process in which early detection and intervention could be effective in ameliorating the course of the disorder. Fig 2 and Fig 3 suggest the existence of a critical period of DUP of about 9 months in which the association with negative symptoms is particularly strong. This finding supports arguments for early detection and intervention programs, since apart from prevention, there is no evidence-based treatment for negative symptoms ( Melle et al., 2008 ). However, it is important to emphasize that this analysis does not prove that there is a causal association between DUP and negative symptoms.

An important strength of this study is the use of individual patient data (IPD). The use of IPD in meta-analyses (Stewart and Parmar, 1993 and Stewart and Tierney, 2002) offers a greater resolution of effect size than meta-analyses based on study level data. Individual patient data permitted us to reanalyze all the data using the same method of correlation and to calculate correlations that had not been previously published. We were also able to combine data across studies to explore the relationship between DUP and negative symptoms.

As a result of obtaining individual patient data, this review has substantially increased the amount of information available for analysis. Consequently, it has extended the findings of previous meta-analyses by providing more precise estimates of the correlation between duration of untreated psychosis and negative symptoms, examining the correlation at both short and long-term follow up and exploring the linearity of the association. This is of clinical importance because the window of opportunity to intervene to improve the prognosis of psychosis is considered to last 3–5 years ( Birchwood, 2000 ).

The main limitation of the review is that data were obtained on only 16 of 28 eligible studies. The 12 studies not obtained included 659 patients. Three studies reported correlations between DUP and negative symptoms at 12 to 24 months of follow-up of which 2 were Spearman correlations. Adding data of these two studies in the meta-analysis did not substantially alter the reported correlations of heterogeneity statistics. We were not able to correct for possible confounding variables on the correlation between DUP and negative symptoms, such as drug use or premorbid adjustment because we did not have access to these data.

Previous studies have noted considerable variation in first episode psychosis cohorts ( Schmitz et al., 2007 ), for example in: participant characteristics; instruments to assess symptoms; measures of DUP; and definitions of DUP. In this meta-analysis we included observational studies as well as interventional studies. No interventional studies targeting negative symptoms were included. Some studies rely entirely on the patients' own reports and thus reflect the onset of the subjective experiences of psychosis while most studies have validated ratings of DUP by including interviews with family members, hospital records and reviews after one year of follow-up. Despite these problems, this analysis detected only a mild to moderate degree of statistical heterogeneity between studies, at a level that was not sufficient to undermine the main findings. We conducted a funnel plot in order to test publication bias. The plot shows the presence of symmetry for all analysis. Since the sampling error is random, this underpins the idea that there is no substantial publication bias.

In summary, the relationship between DUP and negative symptoms has been underestimated, and is in fact strong and persistent. A DUP of less than 9 months appears to be a strong predictor of improvement of negative symptoms, while most patients with a DUP longer than 9 months show persistent negative symptoms. It is unknown to what extent reducing DUP will improve outcome. However efforts to shorten DUP might decrease negative symptom severity as has been done in the Scandinavian TIPS study ( Melle et al., 2008 ). The absence of substantially effective treatments for negative symptoms supports preventive interventions with the potential to ameliorate negative symptoms, e.g. by reducing DUP.

Role of funding source

There was no funding for this study.


Nynke Boonstra designed the study and wrote the protocol together with Rianne Klaassen and Lex Wunderink. Together they performed the literature searches and analysis. Max Marshall supported the analysis and interpretation, and Sjoerd Sytema, Lieuwe de Haan and Durk Wiersma read and revised the manuscript. All authors contributed to and have approved the final manuscript.

Conflict of interest

All authors declare that they have no conflicts of interest.


The authors wish to thank the research teams who were so kind to send us their raw data. In particular we want to thank: Jean Addington, Thomas Barnes, Evelyn Bromet, Eric Chen, Eadbhard O'Callaghan, John Stirling, Richard Drake, Lena Flyckt, Krystyna Gorna, Krystyna Karacz, Lone Petersen, Meredith Harris, Rahul Manchanda, Ross Norman, Ingrid Melle, Alp Ucok and Lex Wunderink.

Appendix 1. Description of included cohorts


Cohort Cohort size Country Age of population in years Length of follow up (months) Symptom scale DUP definition Instrument used to determine DUP
Addington et al. (2004) 240 Canada 36 PANSS From the time the individual first described the onset of any positive sx that could be rated as 4 or more on the PANSS until the first effective treatment was initiated. IRAOS
Barnes et al. (2000) and Subramaniam et al. (2007) 98 United Kingdom 16–55 12 SANS Onset of psychotic sx to first treatment with antipsychotic medication. NS
Black et al. (2001) 21 Canada 6 PANSS Onset of positive psychotic sx (score of ≥ 4 on any of the positive subscale items in the PANSS, these sx must have lasted throughout the day for several days or appeared several times a week) until the beginning of treatment with antipsychotic medication. IRAOS
Chen et al. (2005) 93 Hong Kong 18–55 36 PANSS Onset date for the earliest psychotic sx. IRAOS
Clarke et al. (2006) 152 Ireland 12 > 96 PANSS First noted psychotic symptoms to presentation to the psychiatric services for initiation of adequate treatment of a psychotic illness. SCID
Craig et al. (2000) 354 United States 15–60 24 SANS Occurrence of the first clear psychotic symptom to first psychiatric hospitalization. SCID
Crespo-Facorro et al. (2007) 61 Spain 15–60 36 SANS First continuous (present most of the time) psychotic symptoms to initiation of adequate antipsychotic drug treatment. SCID
Birchwood (2000) and Cullberg et al. (2002) 175 Sweden 18–45 60 BPRS First psychotic sx until the first contact with psychiatric services. SCID
de Haan et al. (2000) (61) 88 The Netherlands Adolescents 72 PANSS First onset of psychotic sx until start of antipsychotic medication (for a minimum of 6 weeks). NS
Drake et al. (2000) 255 United Kingdom 16–64 18 PANSS Onset of delusions and hallucinations. NS
Gorna et al. (2008) 86 Poland Adults 60 PANSS Onset of first psychotic symptoms to the initiation of antipsychotic treatment ICD 10
Harris et al. (2005) 318 Australia 16–45 96 SANS Onset of psychosis to initiation of treatment. RPMIP
Ho et al. (2000) 74 United States 6 SANS Onset of first sx to the initiation of neuroleptic treatment CASH + PSYCH-base
Larsen et al. (2000) 43 Norway 15–55 12 PANSS Onset of psychotic sx to hospitalization for psychosis or initiation of adequate treatment. SCID
Malla et al. (2007) 172 Canada 16–50 12 PANSS Onset of psychotic sx to the time of initiation of continuous antipsychotic treatment, plus any periods of psychosis previously experienced and spontaneously remitted. CORS
Malla et al. (2003) 153 Canada 16–50 36 SANS, Initial onset of psychosis to treatment
(antipsychotic medication of a dosage that should actually lead to a significant response in most patients for a period of time (4 weeks)).
Manchanda et al. (2005) 122 Canada 16–50 24 SANS Onset of clear symptoms of psychosis to the initiation of antipsychotic medications. CORS
Melle et al. (2005) 301 Scandinavia 18–65 24 PANSS Onset of psychosis (first week with sx corresponding to a PANSS score of ≥ 4 or on positive subscale items 1, 3, 5 or 6 or on the general subscale item 9) until start of adequate treatment (structured treatment with antipsychotic medications or hospitalization in a highly staffed psychiatric ward to manage psychotic sx,). SCID
Montague et al. (1989) 109 United Kingdom 16–50 120 SANS First onset of positive symptoms to index admission PSE-9
Oosthuizen et al. (2005) 57 South Africa 16–55 24 PANSS Onset of overt hallucinations or delusions, up to the initiation of treatment with antipsychotic medication. SCID
Petersen et al. (2005) 578 Scandinavia 18–45 24 SANS Appearance of at least one psychotic sx until initiation of adequate treatment SCAN & IRAOS.
Sim et al. (2004) 278 Singapore 18–40 24 PANSS Onset of psychotic sx and the time that treatment was initiated. SCID-P
Ucok et al. (2006) 148 Turkey 15–45 12 SANS Onset of the first positive sx to the first hospitalization. SCID e.
Verdoux et al. (2001) 65 France < 60 24 PANSS Onset of positive symptoms to first admission. MINI
Vyas et al. (2007) 40 United Kingdom Adolescents 48 PANSS Age of onset of psychosis tp age at initiation of antipsychotic treatment. SCID/KID-SCID
Wade et al. (2005) 126 Australia 15–30 15 SANS Onset of psychotic symptoms to service entry. RPMIP
Wunderink et al. (2006) 157 The Netherlands 18–45 24 PANSS First manifestation of any positive psychotic sx to the start of antipsychotic treatment. SCAN
Yamazawa et al. (2008) 34 Japan 16–44 12 PANSS Onset of psychotic symptoms to the first prescription of neuroleptics for psychotic symptoms and the first prescription of neuroleptics for psychosis NS

CASH; Comprehensive Assessment of Psychiatric History.

CORS; Course of Onset and Relapse Schedule.

IRAOS; Interview for the Retrospective Assessment of the Onset of Schizophrenia.

MINI; Mini International Neuropsychiatric Interview.

PANSS; Positive and Negative Syndrome Scale.

PSE; Present State Examination.

RPMIP; Royal Park Multidiagnostic Instrument for Psychosis.

SANS; Structured Assessment of Negative Symptoms.

SAPS; Structured Assessment of Positive Symptoms.

SCID; Structured Clinical Interview for DSM disorders.

SCAN; Schedules for Clinical Assessment in Neuropsychiatry.

ICD-10; International Statistical Classification of Diseases.

NS; not specified.


  • Addington and Addington, 1993 J. Addington, D. Addington. Premorbid functioning, cognitive functioning, symptoms and outcome in schizophrenia. J. Psychiatry Neurosci.. 1993;18(1):18-23
  • Addington et al., 2004 J. Addington, S. Van Mastrigt, D. Addington. Duration of untreated psychosis: impact on 2-year outcome. Psychol. Med.. 2004;34(2):277-284
  • Andreasen, 1989 N.C. Andreasen. The Scale for the Assessment of Negative Symptoms (SANS): conceptual and theoretical foundations. Br. J. Psychiatry Suppl.. 1989;7:49-58
  • Barnes et al., 2000 T.R. Barnes, S.B. Hutton, M.J. Chapman, S. Mutsatsa, B.K. Puri, E.M. Joyce. West London first-episode study of schizophrenia. Clinical correlates of duration of untreated psychosis. Br. J. Psychiatry. 2000;177:207-211
  • Birchwood, 2000 M. Birchwood. Early intervention and sustaining the management of vulnerability. Aust. N. Z. J. Psychiatry. 2000;34(Suppl.):S181-S184
  • Black et al., 2001 K. Black, L. Peters, Q. Rui, H. Milliken, D. Whitehorn, L.C. Kopala. Duration of untreated psychosis predicts treatment outcome in an early psychosis program. Schizophr. Res.. 2001;47(2–3):215-222
  • Blanchard et al., 2005 J.J. Blanchard, W.P. Horan, L.M. Collins. Examining the latent structure of negative symptoms: is there a distinct subtype of negative symptom schizophrenia?. Schizophr. Res.. 2005;77(2–3):151-165
  • Borenstein et al., 2009 M. Borenstein, L.V. Hedges, J.P.T. Higgins, R.P. Rowlands. Introduction to Meta-Analysis. (John Whiley & Sons Ltd., The Atrium, Southern Gate, Chisterer, West Sussex, United Kingdom, 2009)
  • Buckley and Stahl, 2007 P.F. Buckley, S.M. Stahl. Pharmacological treatment of negative symptoms of schizophrenia: therapeutic opportunity or cul-de-sac?. Acta Psychiatr. Scand.. 2007;115(2):93-100
  • Castelein et al., 2008 S. Castelein, R. Bruggeman, J.T. van Busschbach, G.M. van Der, A.D. Stant, H. Knegtering, et al. The effectiveness of peer support groups in psychosis: a randomized controlled trial. Acta Psychiatr. Scand.. 2008;118(1):64-72
  • Chen et al., 2005 E.Y. Chen, C.L. Hui, R.C. Chan, E.L. Dunn, M.Y. Miao, W.S. Yeung, et al. A 3-year prospective study of neurological soft signs in first-episode schizophrenia. Schizophr. Res.. 2005;75(1):45-54
  • Clarke et al., 2006 M. Clarke, P. Whitty, S. Browne, O. McTigue, M. Kamali, M. Gervin, et al. Untreated illness and outcome of psychosis. Br. J. Psychiatry. 2006;189:235-240
  • Craig et al., 2000 T.J. Craig, E.J. Bromet, S. Fennig, M. Tanenberg-Karant, J. Lavelle, N. Galambos. Is there an association between duration of untreated psychosis and 24-month clinical outcome in a first-admission series?. Am. J. Psychiatry. 2000;157(1):60-66
  • Crespo-Facorro et al., 2007 B. Crespo-Facorro, R. Roiz-Santianez, J.M. Pelayo-Teran, C. Gonzalez-Blanch, R. Perez-Iglesias, A. Gutierrez, et al. Caudate nucleus volume and its clinical and cognitive correlations in first episode schizophrenia. Schizophr. Res.. 2007;91(1–3):87-96
  • Cullberg et al., 2002 J. Cullberg, S. Levander, R. Holmqvist, M. Mattsson, I.M. Wieselgren. One-year outcome in first episode psychosis patients in the Swedish Parachute project. Acta Psychiatr. Scand.. 2002;106(4):276-285
  • de Haan et al., 2000 L. de Haan, G.M. van Der, J. Wolthaus. Duration of untreated psychosis and the long-term course of schizophrenia. Eur. Psychiatry. 2000;15(4):264-267
  • Drake et al., 2000 R.J. Drake, C.J. Haley, S. Akhtar, S.W. Lewis. Causes and consequences of duration of untreated psychosis in schizophrenia. Br. J. Psychiatry. 2000;177:511-515
  • Gold et al., 2009 C. Gold, H.P. Solli, V. Kruger, S.A. Lie. Dose–response relationship in music therapy for people with serious mental disorders: systematic review and meta-analysis. Clin. Psychol. Rev.. 2009;29(3):193-207
  • Gorna et al., 2008 K. Gorna, K. Jaracz, F. Rybakowski, J. Rybakowski. Determinants of objective and subjective quality of life in first-time-admission schizophrenic patients in Poland: a longitudinal study. Qual. Life Res.. 2008;17(2):237-247
  • Harris et al., 2005 M.G. Harris, L.P. Henry, S.M. Harrigan, R. Purcell, O.S. Schwartz, S.E. Farrelly, et al. The relationship between duration of untreated psychosis and outcome: an eight-year prospective study. Schizophr. Res.. 2005;79(1):85-93
  • Herbener and Harrow, 2001 E.S. Herbener, M. Harrow. Longitudinal assessment of negative symptoms in schizophrenia/schizoaffective patients, other psychotic patients, and depressed patients. Schizophr. Bull.. 2001;27(3):527-537
  • Heydebrand et al., 2004 G. Heydebrand, M. Weiser, J. Rabinowitz, A.L. Hoff, L.E. DeLisi, J.G. Csernansky. Correlates of cognitive deficits in first episode schizophrenia. Schizophr. Res.. 2004;68(1):1-9
  • Higgins et al., 2003 J.P. Higgins, S.G. Thompson, J.J. Deeks, D.G. Altman. Measuring inconsistency in meta-analyses. BMJ. 2003;327(7414):557-560
  • Ho et al., 2000 B.C. Ho, N.C. Andreasen, M. Flaum, P. Nopoulos, D. Miller. Untreated initial psychosis: its relation to quality of life and symptom remission in first-episode schizophrenia. Am. J. Psychiatry. 2000;157(5):808-815
  • Kay et al., 1987 S.R. Kay, A. Fiszbein, L.A. Opler. The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophr. Bull.. 1987;13(2):261-276
  • Kirkpatrick et al., 2006 B. Kirkpatrick, W.S. Fenton, W.T. Carpenter Jr., S.R. Marder. The NIMH-MATRICS consensus statement on negative symptoms. Schizophr. Bull.. 2006;32(2):214-219
  • Larsen et al., 2000 T.K. Larsen, L.C. Moe, L. Vibe-Hansen, J.O. Johannessen. Premorbid functioning versus duration of untreated psychosis in 1 year outcome in first-episode psychosis. Schizophr. Res.. 2000;45(1–2):1-9
  • Leucht et al., 2011 S. Leucht, S. Heres, W. Kissling, J.M. Davis. Evidence-based pharmacotherapy of schizophrenia. Int. J. Neuropsychopharmacol.. 2011;:1-16
  • Makinen et al., 2008 J. Makinen, J. Miettunen, M. Isohanni, H. Koponen. Negative symptoms in schizophrenia: a review. Nord. J. Psychiatry. 2008;62(5):334-341
  • Malla et al., 2002 A.K. Malla, J.J. Takhar, R.M. Norman, R. Manchanda, L. Cortese, R. Haricharan, et al. Negative symptoms in first episode non-affective psychosis. Acta Psychiatr. Scand.. 2002;105(6):431-439
  • Malla et al., 2003 A. Malla, R. Norman, T. McLean, D. Scholten, L. Townsend. A Canadian programme for early intervention in non-affective psychotic disorders. Aust. N. Z. J. Psychiatry. 2003;37(4):407-413
  • Malla et al., 2004 A.K. Malla, R.M. Norman, J. Takhar, R. Manchanda, L. Townsend, D. Scholten, et al. Can patients at risk for persistent negative symptoms be identified during their first episode of psychosis?. J. Nerv. Ment. Dis.. 2004;192(7):455-463
  • Malla et al., 2007 A. Malla, N. Schmitz, R. Norman, S. Archie, D. Windell, P. Roy, et al. A multisite Canadian study of outcome of first-episode psychosis treated in publicly funded early intervention services. Can. J. Psychiatry. 2007;52(9):563-571
  • Manchanda et al., 2005 R. Manchanda, R. Norman, A. Malla, R. Harricharan, J. Takhar, S. Northcott. EEG abnormalities and two year outcome in first episode psychosis. Acta Psychiatr. Scand.. 2005;111(3):208-213
  • Marshall et al., 2005 M. Marshall, S. Lewis, A. Lockwood, R. Drake, P. Jones, T. Croudace. Association between duration of untreated psychosis and outcome in cohorts of first-episode patients: a systematic review. Arch. Gen. Psychiatry. 2005;62(9):975-983
  • Melle et al., 2004 I. Melle, T.K. Larsen, U. Haahr, S. Friis, J.O. Johannessen, S. Opjordsmoen, et al. Reducing the duration of untreated first-episode psychosis: effects on clinical presentation. Arch. Gen. Psychiatry. 2004;61(2):143-150
  • Melle et al., 2005 I. Melle, U. Haahr, S. Friis, K. Hustoft, J.O. Johannessen, T.K. Larsen, et al. Reducing the duration of untreated first-episode psychosis — effects on baseline social functioning and quality of life. Acta Psychiatr. Scand.. 2005;112(6):469-473
  • Melle et al., 2008 I. Melle, T.K. Larsen, U. Haahr, S. Friis, J.O. Johannesen, S. Opjordsmoen, et al. Prevention of negative symptom psychopathologies in first-episode schizophrenia: two-year effects of reducing the duration of untreated psychosis. Arch. Gen. Psychiatry. 2008;65(6):634-640
  • Montague et al., 1989 L.R. Montague, D. Tantam, D. Newby, P. Thomas, N. Ring. The incidence of negative symptoms in early schizophrenia, mania and other psychoses. Acta Psychiatr. Scand.. 1989;79(6):613-618
  • NICE, 2010 NICE. The Nice Guideline on Core Interventions in the Treatment and Management of Schizoprenia in Adults in Primary and Secondary Care, updated edition. The British Psychological Society and The Royal College of Psychiatrists. 2010;
  • Oosthuizen et al., 2005 P. Oosthuizen, R.A. Emsley, N. Keyter, D.J. Niehaus, L. Koen. Duration of untreated psychosis and outcome in first-episode psychosis. Perspective from a developing country. Acta Psychiatr. Scand.. 2005;111(3):214-219
  • Overall and Gorham, 1962 J.E. Overall, D.R. Gorham. The brief psychiatric rating scale. Psychol. Rep.. 1962;10:799-812
  • Perkins, 2006 D.O. Perkins. Review: longer duration of untreated psychosis is associated with worse outcome in people with first episode psychosis. Evid. Based Ment. Health. 2006;9(2):36
  • Petersen et al., 2005 L. Petersen, P. Jeppesen, A. Thorup, M.B. Abel, J. Ohlenschlaeger, T.O. Christensen, et al. A randomised multicentre trial of integrated versus standard treatment for patients with a first episode of psychotic illness. BMJ. 2005;331(7517):602
  • Petersen et al., 2008 L. Petersen, A. Thorup, J. Oqhlenschlaeger, T.O. Christensen, P. Jeppesen, G. Krarup, et al. Predictors of remission and recovery in a first-episode schizophrenia spectrum disorder sample: 2-year follow-up of the OPUS trial. Can. J. Psychiatry. 2008;53(10):660-670
  • Pogue-Geile and Harrow, 1985 M.F. Pogue-Geile, M. Harrow. Negative symptoms in schizophrenia: their longitudinal course and prognostic importance. Schizophr. Bull.. 1985;11(3):427-439
  • Rohricht and Priebe, 2006 F. Rohricht, S. Priebe. Effect of body-oriented psychological therapy on negative symptoms in schizophrenia: a randomized controlled trial. Psychol. Med.. 2006;36(5):669-678
  • Schmitz et al., 2007 N. Schmitz, A. Malla, R. Norman, S. Archie, R. Zipursky. Inconsistency in the relationship between duration of untreated psychosis (DUP) and negative symptoms: sorting out the problem of heterogeneity. Schizophr. Res.. 2007;93(1–3):152-159
  • Sim et al., 2004 K. Sim, V. Swapna, S. Mythily, R. Mahendran, E.H. Kua, P. McGorry, et al. Psychiatric comorbidity in first episode psychosis: the Early Psychosis Intervention Program (EPIP) experience. Acta Psychiatr. Scand.. 2004;109(1):23-29
  • Snijders and BRJ, 1999 T.A.B. Snijders, BRJ. Multi-Level Analysis: An Introduction to Basis and Advances Multilevel Modeling (SAGE Publication, London, 1999)
  • StataCorp., 2009 StataCorp. Stata Statistical Software: Release 11. (College Station, StataCorp LP, Texas, 2009)
  • Stewart and Parmar, 1993 L.A. Stewart, M.K. Parmar. Meta-analysis of the literature or of individual patient data: is there a difference?. Lancet. 1993;341(8842):418-422
  • Stewart and Tierney, 2002 L.A. Stewart, J.F. Tierney. To IPD or not to IPD? Advantages and disadvantages of systematic reviews using individual patient data. Eval. Health Prof.. 2002;25(1):76-97
  • Subramaniam et al., 2007 M. Subramaniam, E. Pek, S. Verma, Y.H. Chan, S.A. Chong. Diagnostic stability 2 years after treatment initiation in the early psychosis intervention programme in Singapore. Aust. N. Z. J. Psychiatry. 2007;41(6):495-500
  • Ucok et al., 2006 A. Ucok, A. Polat, S. Cakir, A. Genc. One year outcome in first episode schizophrenia. Predictors of relapse. Eur. Arch. Psychiatry Clin. Neurosci.. 2006;256(1):37-43
  • Verdoux et al., 2001 H. Verdoux, F. Liraud, C. Bergey, F. Assens, F. Abalan, J. van Os. Is the association between duration of untreated psychosis and outcome confounded? A two year follow-up study of first-admitted patients. Schizophr. Res.. 2001;49(3):231-241
  • Vyas et al., 2007 N.S. Vyas, M. Hadjulis, A. Vourdas, P. Byrne, S. Frangou. The Maudsley early onset schizophrenia study. Predictors of psychosocial outcome at 4-year follow-up. Eur. Child Adolesc. Psychiatry. 2007;16(7):465-470
  • Wade et al., 2005 D. Wade, S. Harrigan, J. Edwards, P.M. Burgess, G. Whelan, P.D. McGorry. Patterns and predictors of substance use disorders and daily tobacco use in first-episode psychosis. Aust. N. Z. J. Psychiatry. 2005;39(10):892-898
  • Wunderink et al., 2006 A. Wunderink, F.J. Nienhuis, S. Sytema, D. Wiersma. Treatment delay and response rate in first episode psychosis. Acta Psychiatr. Scand.. 2006;113(4):332-339
  • Yamazawa et al., 2008 R. Yamazawa, T. Nemoto, H. Kobayashi, B. Chino, H. Kashima, M. Mizuno. Association between duration of untreated psychosis, premorbid functioning, and cognitive performance and the outcome of first-episode schizophrenia in Japanese patients: prospective study. Aust. N. Z. J. Psychiatry. 2008;42(2):159-165


a Friesland Mental Health Care Service, Department of Education and Research, Leeuwarden, The Netherlands

b University Medical Center Groningen, Department of Psychiatry, University of Groningen, Groningen, The Netherlands

c NHL University of Applied Science, Institute of Healthcare and Welfare Studies, Leeuwarden, The Netherlands

d Rivierduinen, Mental Health Care Service, Leiden, The Netherlands

e University of Manchester, Manchester, UK

f Academic Medical Center, Amsterdam, Department of Psychiatry, The Netherlands

lowast Corresponding author at: Friesland Mental Health Care Service, PO Box 932, 8901 BS Leeuwarden, The Netherlands. Tel.: + 599 3158 2848989; fax: + 599 3158 2848040.