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 (DUP) and the course of schizophrenia in a 20-year follow-up study

Psychiatry Research, 3, 219, pages 420 - 425


Our aims were to analyze the relationship between the duration of psychosis (DUP) and short- and long-term outcomes of treatment, and to determine the cut-off point between short and long DUP at which differences in treatment outcomes are most significant. We assessed 80 participants with schizophrenia at the point of their first hospitalization. Fifty participated in all four follow-ups over 20 years. DUP was divided into short (up to 6 months) and long (over 6 months). ANOVA and Chi-square tests were employed to identify significant differences in both clinical and social indicators of functioning. ROC curves were used to estimate the best DUP division point. Significant differences favoring the short-DUP group were found for: GAF, total severity of symptoms and severity of positive symptoms, social functioning measured according to DSM-III criteria, employment, and social contacts. The optimal cut-off point for DUP was the 23rd week. We concluded that: (1) the relationship between longer DUP and worse overall treatment outcomes was sustained throughout the 20 years, (2) a positive correlation between DUP and the severity of psychopathological symptoms was observed over the first 12 years of illness, (3) the results indicate the efficacy of early therapeutic interventions in psychosis.

Keywords: Predictors, Psychosis, Long-term study.

1. Introduction

In recent years, the criterion of duration of untreated psychosis (DUP) has risen to prominence as one of the fundamental factors determining prognosis in the course of schizophrenia-related psychotic disorders. Two meta-analyses (Marshall et al, 2005 and Perkins et al, 2005) describing the results of research into the correlation between DUP and the course of illness and treatment outcomes unequivocally indicate the existence of a positive correlation, confirmed primarily by the aggregate results of numerous short-term studies (involving up to 5 years of observation). The first of these meta-analyses ( Marshall et al., 2005 ) indicates a positive correlation between longer DUP and worse treatment outcomes in terms of general functioning, positive symptoms, and quality of life, but not between longer DUP and negative symptoms or social functioning. The second meta-analysis ( Perkins et al., 2005 ) demonstrates a positive correlation between shorter DUP and better response to treatment—lesser severity of general psychopathology, positive and negative symptoms, and better general functioning. A significant meta-analysis concerning the relationship between DUP and negative symptoms in the short and medium term was conducted by Boonstra et al. (2012) Contrary to previous reports, it shows that there is an association between those areas. Based on a review of studies from the years 1992–2009, Boonstra found that a shorter DUP is associated with lesser severity of negative symptoms in both short-term (1–2 years) and medium-term (5–8 years) studies. The decrease in the severity of negative symptoms was non-linear. A DUP shorter than 9 months markedly increased the chances of less severe negative symptoms in follow-ups.

Prospective studies which begin with the first episode of psychosis (FEP) and assess the effect of DUP on remote treatment outcomes (over 10 years) are rarer. Our study falls into this category. It is a prospective study beginning at the point of first psychiatric hospitalization (real time), and it is long-term in duration (as per McGlashan (1988) ): running over 10 years). We shall make reference to other studies fulfilling the same criteria—by Helgason (1990) , Huber (1997) , Wiersma et al. (2000) , Kua et al. (2003) , Bottlender et al. (2003) , Röpcke and Eggers (2005) , Kinoshita et al. (2005) , White et al. (2009) , Shirvastava et al. (2010) , Ichinose et al. (2010) , and Cechnicki et al. (2010) . In these studies, the period of observation ranges from 10 to 28 years. The results of most of these studies indicate the existence of a correlation between DUP and remote treatment outcomes, although several of them present disparate results.

Diverse criteria have been adopted to divide DUP into short- and long-term, and various definitions of the end of DUP (e.g. psychiatric hospitalization, making contact with a therapist, starting to take neuroleptics, etc.) have been offered. Some of the studies (Helgason, 1990 and Huber, 1997; Bottlender et al., 2003 ) did not include any intermediate assessment points, which makes assessment of the dynamics of potential changes over time difficult. In the studies (Huber, 1997, Wiersma et al, 1998, Wiersma et al, 2000, and Kua et al, 2003), DUP analysis is included as one of many aspects of long-term catamnestic study, and the assessment is usually restricted to the influence of DUP on clinical outcomes. In the wake of ambiguous results, answers to three questions remain unsettled. First, is there a correlation between DUP and long-term outcomes in both clinical and social functioning over the course of many years of illness (in our case, 20)? Second, does this relationship remain constant over the years, or do the differences between the short- and long-DUP subgroups tend to fluctuate over time? Third, is there a cut-off point enabling us to define a period of untreated psychosis beyond which the treatment outcomes for schizophrenic psychoses (schizophrenia) tend to deteriorate markedly, and if so, what is it?

2. Methods

2.1. Participants

Eighty patients hospitalized for the first time for schizophrenia agreed to participate. These participants were diagnosed according to the DSM III criteria and rediagnosed in accordance with DSM IV at their 12-year follow-up. The study began with two assessments, one at admission and the other at discharge from the index hospitalization. The subsequent assessments were conducted at the 3- (K3), 7- (K7), 12- (K12), and 20-year (K20) points following the index hospitalization. All of the participants were residents of Cracow and none of them had any history of organic brain disorder or addiction to psychoactive substances. The patients in our study were admitted in the years 1985–1988. All the subjects gave their informed consent to participation in the study. The follow-ups were conducted on the basis of a clinical assessment by two experienced clinicians. Of the 80 original subjects, 50 participated in all the follow-ups over the 20 years of the study ( Table 1 ). There are no statistically significant differences between these groups in any of the areas described.

Table 1 Demographic and social descriptive statistics at the time of index hospitalization.

Demographic and social factors N=80 N=50
No. of cases Percentage No. of cases Percentage
 Female 46 57 30 60
 Male 34 43 20 40
 Higher ed. 23 29 13 26
 Uncompleted higher ed./studying 7 9 4 8
 Secondary ed. 31 39 20 40
 Vocational ed. 14 17 11 22
 Primary ed. 5 6 2 4
Marital status        
 Married 27 34 17 34
 Single 53 66 33 66
DSMIII social functioning before first admission        
 Best 0 0 0 0
 Very good 14 18 10 20
 Good 21 26 12 24
 Fair 25 31 17 34
 Poor 16 20 10 20
 Very poor 3 4 1 2
 Declined 1 1 0 0
 Job/studies 70 88 46 92
 Sheltered job 0 0 0 0
 Odd jobs 0 0 0 0
 unemployed 10 12 4 8
Social contacts        
 A deep and satisfying relationship 12 15 8 16
 Numerous superficial and satisfying relationships 31 39 21 42
 One superficial and satisfying relationship 15 19 10 20
 Unsatisfying or frustrating relationships 12 15 6 12
 Lack of relationships 10 12 5 10
Duration of untreated psychosis (weeks) Mean Median Mean Median
(Comparison of means. t-student P=0.983) 46.53 13 46.28 12.5

2.2. Procedures and tools

DUP was estimated during a clinical interview with each patient and his or her family at index admission. The beginning of the DUP period was defined as the first occurrence of positive symptoms, with the end marked by the date of admission at the index hospitalization. In order to be able to relate our results to those obtained by other researchers, we dichotomized DUP, with a cut-off point between the short- and long-term groups set at the end of the 6th month. On the basis of this division, we examined the correlations between DUP and treatment outcomes using analysis of variance. Then, based on the treatment outcomes, we attempted to verify the division point for short and long DUP, and using ROC curves we established the cut-off point which would maximize the differences in outcomes between these two groups.

The Brief Psychiatric Rating Scale (BPRS)—a tool constructed in 1960 by Overall and Gorham (1962) —was used to determine the level of psychopathological symptoms. The original version described 16 symptoms, the 1972 modification assessed 18 symptoms, but in our study we used the extended version by Lukoff and Nuechterlein (1986) , which covers 24 symptoms, each assessed on a scale of 1–7.

The Global Assessment of Functioning (GAF) is Axis V of the DSM “multi-axial” system for assessment, which is intended as a reflection of the evaluating clinician׳s judgment of a patient׳s ability to function in daily life. This 100-point scale measures psychological, social, and occupational functioning. The other clinical outcomes assessed at every follow-up include number of relapses, number of readmissions, and duration of readmissions. Social contacts were assessed on a 4-degree scale: 1: numerous contacts beyond family and fellow patients, 2: sporadic contact beyond family and other patients, 3: only with other patients and family, and 4: only within family. Employment was assessed on a 4-degree scale: 1: open-market job/studying, 2: sheltered job, 3: odd jobs, and 4: unemployed.

2.3. Statistical analysis

The following statistical methods were used:

  • ANOVA, to compare the temporal dynamics of the GAF and BPRS results across 20 years between the groups with shorter and longer DUP; two-factor ANOVA, with one between-subjects factor (shorter vs. longer DUP) and one repeated-measures factor (subsequent GAF and BPRS results),
  • chi-square test, to compare assessments of social contacts and employment,
  • Receiver Operating Characteristics (ROC) curves were used to determine the cut-off point for the DUP division, plotting “sensitivity” against “1—specificity”; the statistical significance of the area under the curve was determined by means of asymptotic tests.

The ROC curves were computed separately for every follow-up, and the data of all subjects participating in a given follow-up were included (n=72 after 3 years,n=71 after 7 years,n=72 after 12 years,n=64 after 20 years).

All the remaining analyses were performed using the data of 50 patients assessed at all follow-ups. All statistical analyses were performed using IBM SPSS Statistics 20 software.

3. Results

In the first step, we analyzed the relationship between DUP and the dynamics of psychopathological symptoms as measured on the BPRS using ANOVA, and found no interaction effect, as the dynamics of the changes in both groups were quite similar. There was however one significant main effect of DUP: the subgroup with short DUP had a significantly lower level of psychopathological symptoms over the 20 years than the long-DUP subgroup. Results favoring the short DUP subgroup were obtained for general severity of symptoms after first hospitalization (P=0.019) and after 7 (P=0.045) and 12 (P=0.046) years; and in the severity of positive symptoms after first hospitalization (P=0.002), and after 7 (P=0.023) and 12 (P=0.009) years. In terms of general functioning measured using the GAF scale, the short-DUP subgroup obtained significantly better results than the long-DUP subgroup after 3 (P=0.017), 7 (P=0.001), 12 (P=0.001), and 20 (P=0.016) years from first hospitalization. General social functioning (axis V of DSM-III classification) was better in the short-DUP group at all assessments. In terms of employment, at 7 and 12 years from first hospitalization the short-DUP group had statistically significantly better statistics than the other group. As for social contacts, at the time of the 3-year follow-up following first hospitalization, short-DUP subjects had significantly more interaction outside of their families and peers in the patient community. The results are presented in Table 2 .

Table 2 Differences between subgroups with short and long DUP.

Treatment outcome ANOVA—main effect of DUP Pairwise comparisons between DUP groups at follow-ups (P-levels)
1st admission 1st discharge K3 K7 K12 K20
BPRS total F=6.433 P=0.015 (38.3 vs. 43.9) 0.974 0.019 0.108 0.045 0.046 0.068
BPRS positive symptoms F=6.366 P=0.015 (8.4 vs. 10.2) 0.666 0.002 0.063 0.023 0.009 0.244
BPRS negative symptoms F=1.402 P=0.242 (8.5 vs. 9.5) 0.718 0.600 0.272 0.519 0.804 0.063
No. of inpatient rehospitalizations F=0.040 P=0.842 (2.1 vs. 2.0) nd nd 0.351 0.896 0.994 0.792
Total time of inpatient rehospitalizations F=0.005 P=0.941 (18.7 vs. 18.3) nd nd 0.373 0.849 0.677 0.804
No. of relapses F=0.022 P=0.833 (3.4 vs. 3.3) nd nd 0.688 0.780 0.721 0.602
GAF F=13.818 P=0.001 (64.5 vs. 51.8) nd nd 0.017 <0.001 <0.001 0.016
Social functioning DSM-III axis 5 F=14.805 P=<0.001 (3.7 vs. 4.7) nd nd 0.014 0.001 0.015 0.001
Treatment outcome Pairwise χ2 comparisons across follow-ups between short and long DUP (P-levels)
Employment   nd nd 0.596 0.108 0.016 0.304
Social contacts   nd nd 0.034 0.130 0.272 0.427

The other indicators of the course of the illness, including number of relapses, number of rehospitalizations, duration of rehospitalizations, and severity of negative symptoms, were all found not to differ between the subgroups with short and long DUP.

3.1. The cut-off point

In order to assess the most prominent point of division between short and long DUP, we used ROC curves. The results are shown in Fig. 1 .


Fig. 1 The results of ROC analyses.

On most of the graphs above there is a relatively clear cut-off point which divides the group in terms of the duration of untreated psychosis in relation to various treatment outcomes. With respect to clinical outcomes as measured using the BPRS, chances of a successful remission defined according to Andreasen׳s criteria—BPRS results for seven items (grandiosity, suspiciousness, unusual thought content, hallucinatory behavior, conceptual disorganization, mannerisms/posturing, and blunted affect) not exceeding three points ( Andreasen et al., 2005 )—were highest and statistically significant at 7 years from index hospitalization, provided the treatment had begun within 53 weeks of the onset of psychotic symptoms. After 20 years, the cut-off point fell to 23 weeks from the onset of psychotic symptoms. As for general functioning assessed using the GAF, in three of the four follow-ups (K3, K12, and K20) the cut-off point fell on the 23rd week; the exception was K7, when the watershed was placed at 12.5 weeks.

4. Discussion

In this discussion we will address the correlation between DUP and patients׳ clinical status and functioning, the dynamics of the observed phenomena, and use of intermediate assessment points, and assess the cut-off point for division of DUP into short- and long-term. The conclusions arising from our study are largely concordant with the results of the abovementioned meta-analyses (Marshall et al, 2005 and Perkins et al, 2005). We confirmed the existence of a positive correlation between short DUP and better general functioning, level of general psychopathology, and positive symptoms. As in the study by Marshall et al. (2005) (though different than Perkins et al. (2005) ), we found no correlation between DUP and negative symptoms. Unlike the first of the meta-analyses ( Marshall et al., 2005 ), we did observe a correlation between DUP and social functioning.

In our previous study ( Cechnicki et al., 2010 ), we analyzed a period of 12 years from first hospitalization and examined the correlation between DUP and four clinical indicators of treatment outcomes: severity of symptoms, duration and number of re-hospitalizations, and number of relapses. We found a correlation in which a shorter DUP was associated with a lesser severity of positive and general symptoms and a significantly smaller number of relapses. As the period of observation grew longer and a larger number of treatment-outcome indicators were included, we obtained an increasingly complete picture of the impact of DUP on the patient׳s overall functioning.

4.1. Correlation with general functioning

Our Cracow study, like other studies (Kua et al, 2003, Bottlender et al, 2003, White et al, 2009, and Ichinose et al, 2010), confirms the positive correlation between a shorter DUP and favorable short- and long-term outcomes in terms of general functioning, including general social functioning. In three studies (Kua et al, 2003, Bottlender et al, 2003, and Ichinose et al, 2010), the assessment was made using GAS, in one ( White et al., 2009 ) it was made using GAF, and in the paper by Ichinose et al. (2010) the Disability Assessment Schedule (DAS) was also used. An interesting result obtained in our study was the persistence of the correlation between DUP and the overall assessment of functioning over 20 years, despite the eventual disappearance of that between DUP and the severity of general and positive clinical symptoms.

4.2. Correlation with severity of psychopathological symptoms

The results of our study support those obtained in the studies (Helgason, 1990, Huber, 1997, Wiersma et al, 1998, Kua et al, 2003, Bottlender et al, 2003, Kinoshita et al, 2005, White et al, 2009, Ichinose et al, 2010, and Cechnicki et al, 2010), which describe the existence of a positive correlation between the duration of untreated psychosis and long-term clinical treatment outcomes. However, our results contradict those obtained in the studies (Wiersma et al, 2000, Röpcke and Eggers, 2005, and Shirvastava et al, 2010). The longest study—begun by Kinoshita et al. (2005) and continued by Ichinose et al. (2010) —indicated that the correlation between psychopathological status and DUP was preserved for 10 years after the first psychotic episode, then disappeared in the 15-year assessment, and reappeared at 28 years from the first psychosis. With regard to the papers by Wiersma et al. (2000) , Röpcke and Eggers (2005) , and Shirvastava et al. (2010) , which indicated the lack of correlation between DUP and long-term treatment outcomes, we may seek an explanation for the lack of such associations. In the multicentre study by Wiersma et al. (2000) , there is a lack of a clear definition of the beginning and the end of DUP, and data concerning DUP were available for only 56% of the study group, which is why they were not included in more detailed analyses. In that paper, and that paper only, DUP was used as a continuous variable. All these factors may have influenced the results obtained. (In our Cracow study, such correlations with psychopathological status also tend to disappear when DUP is used as a continuous variable.)

Röpcke and Eggers (2005) obtained their results from a relatively small group of 39 young people (the mean age of participants at the commencement of the study being 16 years), who were observed over the next 15 years. In that group the prognostic value of DUP was not confirmed. This may be due to the fact that, owing to the early age of disease onset, this group was not representative of the general population of schizophrenia patients, a notion that is corroborated by the poor treatment outcomes after 15 years (only 8% had favorable courses of illness). Shirvastava et al. (2010) showed that there is no correlation between DUP and various indicators of treatment outcomes (PANSS, HDRS, social functionality, suicidality, and recovery). Their negative results could, in turn, have been influenced by the following factors: a significant decline in the number of participants over the course of the study (from 200 to 101 people); the fact that 73% of the group had a DUP of 6 to 24 months, i.e. the distribution was significantly biased towards a long DUP; and, lastly, that the recovery criteria, GAF of over 80 points, seem too stringent.

4.3. Lack of correlation with the negative syndrome

Schmitz et al. (2007) noted that the relationship between DUP and symptoms for negative symptoms was interpreted in terms of heterogeneity among patients diagnosed with schizophrenia. Similar results were observed in one of the meta-analyses ( Marshall et al., 2005 ), and a similar phenomenon was seen in the group we assessed. Our results thus differ from those obtained by Boonstra et al. (2012) in this area. The difference in these results may be due to various causes. Firstly, the period of observation differed. In our study it was 20 years, while the longest study among those included in the meta-analysis lasted only 8 years. The predictive value of DUP in studies lasting longer than 10 years thus remains undetermined. Other methodological differences, such as the character of the study group, the cut-off point for DUP, definitions of negative symptoms and type of statistical analyses employed, may also have contributed to the differences between the results. For example, the Cracow study included a diagnostically homogenous group of first-episode patients, whereas 12 of the total of 28 studies cited by Boonstra et al. were not first-episode studies.

4.4. Correlation with number of re-hospitalizations

Helgason (1990) conducted a 20-year follow-up of 107 schizophrenia patients from Iceland undergoing first-time treatment. One-third of the sample was admitted within 1 year of onset (early admission) and the rest after 1 year (late admission). For 18 of the 20 years of catamnesis, those admitted later (long-DUP patients) had a higher number of readmissions. We did not find an association between the number of re-hospitalizations and DUP, which may be due to a different DUP divide or to the different therapeutic context, which in Cracow is oriented toward reduction of inpatient hospitalizations.

4.5. Correlation with social indicators

Kua et al. (2003) included the employment status of their subjects in their analyses. Unfortunately, this aspect of patients׳ progress was treated as one element of an overall end result which also included the treatment mode of a given patient. This situation renders impossible the assessment of the correlation between DUP and employment status alone. Our results indicate the existence of such a relationship after 12 years from index hospitalization.

4.6. The problem of stability of relationships over time—intermediate assessment points

In our Cracow study, the assessments performed at intermediate follow-ups make it possible to observe the dynamics of potential changes over time. Similar work has been done in a different cultural context by Kinoshita et al. (2005) . They showed a correlation between DUP and treatment results (specifically, remission in accordance with PSE-9 criteria) at 1, 2, 5 and 10 years after first psychotic episode, but found no relationship after 15 years. That study was continued by Ichinose et al. (2010) . At 28 years from the onset of illness, the researchers found a relationship between DUP and both severity of clinical symptoms and level of social functioning. The results of our study are largely concordant with those of Kinoshita et al. and Ichinose et al. Similarly, we observed a temporary evaporation of correlations between DUP and aspects including the severity of positive symptoms after 20 years, but it would be worthwhile to assess them again at, say, a 30-year follow-up.

In the studies by(Wiersma et al, 1998) and (Wiersma et al, 2000)and Kua et al. (2003) intermediate catamneses were not used to assess the dynamics of the course of the illness. Wiersma et al. did not analyze the relationship between DUP and the patients׳ status at 1, 2 and 15 years after onset of illness. Neither did he assess the dynamics of changes over those years. In the study by Kua et al. (2003) , patients were assessed at 5, 10, 15, and 20 years from first hospitalization. That study was conducted by medical students and based on an abbreviated version of GAS. During the follow-up period of the study, the number of subjects fell dramatically from 402 in 1975 to 216 in 1995, which amounts to a loss of 46.3% of the original participants.

4.7. The point of increased risk

One of the objectives of this paper was to assess the point in time within the period of untreated psychosis beyond which the risk of an unfavorable course of illness increases significantly. To date, the division into short and long DUP has usually been set arbitrarily, at differing points in time. Our analyses, based on our own treatment results, indicate that the 23rd week marks the final point by which treatment should be initiated in order to reduce the risk of long-term unfavorable outcomes, both in terms of clinical symptoms and in the area of social functioning. These results corroborate the observation of a need for special therapeutic programmes for early detection in the treatment of schizophrenic psychoses. Such special therapeutic programmes could take a range of forms. For instance, the report by Shrivastava et al. (2012) on the efficacy of helpline interventions in reducing DUP and decreasing the risk of suicide among people experiencing their first episode of psychosis is as compelling as that on early therapeutic intervention programmes by Cullberg et al. (2006) and as the idea of open dialog mooted by(Seikkula et al, 2006) and (Seikkula et al, 2011).

4.8. Conclusions

Summing up the results of our study, we can confirm the existence of a positive correlation between DUP and the treatment outcomes of people suffering from schizophrenia observed over a period of 20 years. At the same time, we can also confirm its variability, depending on the period of observation and the outcome indicators employed. A shorter DUP remains significantly, stably correlated with better overall treatment results assessed using the GAF scale, and with better general social functioning assessed according to the criteria stated in the 5th dimension of DSM III at all assessment points over the 20-year period of observation. The correlation with severity of positive and general psychopathological symptoms remained stable over 12 years after first hospitalization. We did not observe any correlation between DUP and severity of negative symptoms, or with the duration or number of re-hospitalizations or relapses. Among social indicators, we observed only isolated correlations between shorter DUP and greater number of social contacts after 3 years, and better employment situation after 12 years. It is possible that the complex interplay between these dependencies is affected by a range of mediating indicators. The analyses based on our own treatment outcomes indicate that the 23rd week of DUP is the time by which we should begin treatment if we want to decrease the risk of long-term unfavorable outcomes, with respect to both clinical symptoms and social functioning.

One weakness of this study was the rather limited number of participants and the lack of structured tools for assessment of DUP, e.g. the Royal Park Multi-diagnostic Instrument for Psychosis (RPMIP) ( McGorry et al., 1990 ) or the Interview for the Retrospective Assessment of the Onset of Schizophrenia (IRAOS) ( Häfner et al., 1992 ). These tools were published in 1990 and 1992 respectively, which was several years after our study began. This same limitation pertains to all the studies referenced in this paper. Moreover, while the researchers involved in our study had a significant level of clinical experience, no formal inter-rater reliability tests were conducted.


We wish to thank Mr. Konrad Wroński for his significant assistance with the preparation of this paper. This study was financed from Grant no. K/ZDS/001469 to Collegium Medicum, Jagiellonian University.


  • Andreasen et al., 2005 N.C. Andreasen, W.T. Carpenter, J.M. Kane, R.A. Lasser, S.R. Marder, D.R. Weinberger. Remission in schizophrenia: proposed criteria and rationale for consensus. American Journal of Psychiatry. 2005;162:441-449 Crossref
  • Boonstra et al., 2012 N. Boonstra, R. Klaassen, S. Sytema, M. Marshall, L. De Haan, L. Wunderink, D. Wiersma. Duration of untreated psychosis and negative symptoms—a systematic review and meta-analysis of individual patient data. Schizophrenia Research. 2012;142:12-19 Crossref
  • Bottlender et al., 2003 R. Bottlender, T. Sato, M. Jaeger, U. Wegener, J. Wittmann, A. Strauss, H.J. Möller. The impact of the duration of untreated psychosis prior to first psychiatric admission on the 15-year outcome in schizophrenia. Schizophrenia Research. 2003;62:37-44 Crossref
  • Cechnicki et al., 2010 A. Cechnicki, I. Hanuszkiewicz, R. Polczyk, A. Bielaoska. Prospektywna ocena wpływu czasu nieleczonej psychozy na przebieg schizofrenii (a prospective assessment the influence of duration of untreated psychosis on the course of schizophrenia). Psychiatria Polska. 2010;3:381-394
  • Cullberg et al., 2006 J. Cullberg, M. Mattsson, S. Levander, R. Holmqvist, L. Tomsmark, C. Elington, I.M. Wieselgren. Treatment costs and clinical outcome for first episode schizophrenia patients: a 3-year follow-up of the Swedish “Parachute Project” and two comparison groups. Acta Psychiatrica Scandinavica. 2006;114(4):274-281 Crossref
  • Häfner et al., 1992 H. Häfner, A. Riecher-Rössler, M. Hambrecht, K. Mauer, S. Meissner, A. Schmidtke, B. Fätkenheuer, W. Löffler, W. van der Heiden. IRAOS: an instrument for the assessment of onset and early course of schizophrenia. Schizophrenia Research. 1992;6(3):209-223
  • Helgason, 1990 L. Helgason. Twenty years׳ follow-up of first psychiatric presentation for schizophrenia: what could have been prevented?. Acta Psychiatrica Scandinavica. 1990;81(3):231-235 Crossref
  • Huber, 1997 G. Huber. The heterogenous course of schizophrenia. Schizophrenia Research. 1997;28:177-185 Crossref
  • Ichinose et al., 2010 H. Ichinose, Y. Nakane, H. Nakane, H. Kinoshita, Y. Ohta, S. Honda, H. Ozawa. Nagasaki Schizophrenia Study: relationship between ultra long-term outcome (after 28 years) and duration of untreated psychosis. Acta Medica Nagasakiensia. 2010;54(3):59-66
  • Kinoshita et al., 2005 H. Kinoshita, Y. Nakane, H. Nakane, Y. Ishizaki, Y. Ohta, S. Honda, H. Ozawa. Nagasaki schizophrenia study: influence of the duration of untreated psychosis on long-term outcome. Acta Medica Nagasakiensia. 2005;50:17-22
  • Kua et al., 2003 J. Kua, K.E. Wong, E.H. Kua, W.F. Tsoi. A 20-year follow-up study on schizophrenia in Singapore. Acta Psychiatrica Scandinavica. 2003;108:118-125 Crossref
  • Lukoff et al., 1986 D. Lukoff, K.H. Nuechterlein, J. Ventura. Manual for the expanded brief psychiatric rating scale. Schizophrenia Bulletin. 1986;13:261-276
  • 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. Archives of General Psychiatry. 2005;62:975-983 Crossref
  • McGlashan, 1988 T.H. McGlashan. A selective review of recent north American long-term follow-up studies of schizophrenia. Schizophrenia Bulletin. 1988;14(4):515-542 Crossref
  • McGorry et al., 1990 P.D. McGorry, D.L. Copolov, B.S. Singh. Royal park multidiagnostic instrument for psychosis: part I. Rationale and review. Schizophrenia Bulletin. 1990;16(3):501-515 Crossref
  • Overall and Gorham, 1962 J.E. Overall, D.R. Gorham. The brief psychiatric rating scale. Psychological Reports. 1962;10:799-812
  • Perkins et al., 2005 D.O. Perkins, H. Gu, K. Boteva, J.A. Lieberman. Relationship between duration of untreated psychosis and outcome in first-episode schizophrenia: a critical review and meta-analysis. American Journal of Psychiatry. 2005;162:1785-1804 Crossref
  • Röpcke and Eggers, 2005 B. Röpcke, C.h. Eggers. Early-onset schizophrenia. A 15-year follow-up. European Child and Adolescent Psychiatry. 2005;14:341-350
  • 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. Schizophrenia Research. 2007;93:152-159 Crossref
  • Seikkula et al., 2006 J. Seikkula, J. Aaltonen, B. Alakare, K. Haarakangas, Y. Keranen, K. Lehtinen. Five-year experience of first-episode nonaffective psychosis in open-dialog model. Psychotherapy Research. 2006;16(2):214-228 Crossref
  • Seikkula et al., 2011 J. Seikkula, B. Alakare, J. Aaltonen. The comprehensive open-dialog approach in Western Lapland:II. Long-term stability of acute psychosis outcomes in advanced community care: the Western Lapland project. Psychosis: Psycholgical, Social and Integrative Approaches. 2011;3(3):192-204 Crossref
  • Shirvastava et al., 2010 A. Shirvastava, N. Shah, M. Johnston, L. Stitt, M. Thakar, G. Chinnasamy. Effects of duration of untreated psychosis on long -term outcome of people hospitalized with first episode schizophrenia. Indian Journal of Psychiatry. 2010;52:164-167
  • Shrivastava et al., 2012 A.K. Shrivastava, M.E. Johnston, L. Stitt, M. thakar, G. Sakel, S. Iyer, N. Shah, Y. Bureau. Reducing treatment delay for early intervention: evaluation of a community based crisis helpline. Annals of General Psychiatry. 2012;11:20 Crossref
  • White et al., 2009 C. White, J. Stirling, R. Hopkins, J. Morris, L. Montague, D. Tantam, S. Lewis. Predictors of 10-year outcome of first-episode psychosis. in: Psychological Medicine. 39 (, 2009) 1447-1456 Crossref
  • Wiersma et al., 1998 D. Wiersma, F.J. Nienhuis, C.J. Slooff, R. Giel. Natural course of schizophrenic disorders: a 15-year follow-up of a Dutch incidence cohort. Schizophrenia Bulletin. 1998;24:75-85 Crossref
  • Wiersma et al., 2000 D. Wiersma, J. Wanderling, E. Dragomirecka, K. Ganev, G. Harrison, W. An Der Heiden, F.J. Nienhuis, D. Walsh. Social disability in schizophrenia: its development and prediction over 15 years in incidence cohorts in six European centres. Psychological Medicine. 2000;30:1155-1167 Crossref


Department of Community Psychiatry, Chair of Psychiatry, Collegium Medicum, Jagiellonian University in Cracow, Pl. Sikorskiego 2/8, 31-115 Krakow, Poland

lowast Corresponding author. Tel.: +48 12 421 51 17; fax: +48 12 422 56 74.

This study was conducted by the Cracow Schizophrenia Research Group.