versão impressa ISSN 0185-3325
Salud Ment vol.33 no.6 México nov./dez. 2010
Predictors of shortterm course in Mexican firstepisode psychosis patients
Predictores de curso temprano en pacientes mexicanos con primer episodio psicótico
Lizzette GómezdeRegil,1 Thomas R. Kwapil,2 Arsenio RosadoFranco,3 Neus BarrantesVidal1,4,5
1 Departament de Psicologia Clínica i de la Salut. Facultat de Psicologia. Universitat Autónoma de Barcelona.
2 University of North Carolina at Greensboro. Department of Psychology.
3 Hospital Psiquiátrico Yucatán, México.
4 CIBER Salud Mental, Instituto de Salud Carlos III, Spain.
5 Sant Pere ClaverFundació Sanitaria. Departament de Salut Mental.
Departament de Psicologia Clínica i de la Salut.
Universitat Autónoma de Barcelona.
Bellaterra (Barcelona), Spain, 08193.
Tel. +34 93 581 3864.
Fax. +34 93 581 2125.
Recibido primera versión: 9 de febrero de 2010.
Segunda versión: 17 de mayo de 2010.
Aceptado: 16 de junio de 2010.
Background and objectives
The identification of prognostic factors in patients with schizophrenia and related psychotic disorders should enhance our understanding of the aetiology of these disorders and improve their treatment. The first years following an initial episode of psychosis are a <<critical period>> for biological and psychosocial influences that affect future outcome. Both, shortterm outcome and baseline predictors, have been defined by different measures, making the comparison among studies difficult. Studies of the predictive value of baseline demographic and clinical characteristics in the Mexican population are still limited. Hence, the present study aims to: 1. replicate the prognostic value of selected patient characteristics previously related to the shortterm course of psychosis in Mexican firstepisode psychosis patients, and 2. retrospectively assess their prognostic value in the prediction of diagnosis, presence of psychotic residual symptoms, and number of psychotic episodes at least threeyears later.
Information on baseline predictor variables (sociodemographic, premorbid phase, context of the first episode, dimensions of psychopathology) and clinical outcome (diagnosis, residual symptomatology, psychotic episodes) was obtained from the clinical records of 51 patients with a shortterm course of psychosis and whose available followup period was at least three years long (mean = 5.8,SD = 2.1).
Poor premorbid adjustment and hospitalization at first psychotic episode were significant predictors of a schizophrenia diagnosis. Lower educational level and an insidious type of onset significantly predicted the presence of residual symptoms. Hospitalization at first psychotic episode and higher scores on the psychotic dimension at onset significantly predicted subsequent psychotic episodes.
Low educational level increased the risk of residual symptoms, possibly because it hinders treatment continuity. Poor premorbid adjustment was related to a schizophrenia diagnosis at the followup assessment, supporting previous findings of their high ratings for premorbid impairment, including social withdrawal and dysfunctional peer relationship. Insidious onset was predictive of persistent residual symptoms; an association possibly mediated by the duration of untreated psychosis (DUP). Being hospitalized at first episode was a significant prognostic factor for schizophrenia diagnosis and multiple psychotic episodes; the severity and nature of symptoms at first episode that require hospitalization might account for these associations. Replicating previous findings, multipleepisode patients scored significantly higher than the singleepisode patients on the psychoticism dimension. Most baseline factors did not predict diagnosis. This seems congruent with a dimensional view of psychosis suggesting that even though schizophrenic and nonschizophrenic psychoses are classified as separate families of disorders, they exist along a continuum of psychosis that crosses diagnostic boundaries, sharing aetiological and risk factors. Currently, both the amelioration of severe psychotic symptoms and the improvement of psychosocial functioning and quality of life are feasible aims. Symptom exacerbation and hospitalizations might cause cumulative deterioration and impair the patient's social reintegration. Thus, relapse prevention is an important objective in treatment. The identification of reliable predictors of illness course has significant implications for treatment and service planning.
The predictive value of several factors was replicated in this sample of patients with psychotic illnesses, although predictors seem to relate differently to the three shortterm course measures. Comprehensively mapping the development and outcome of the first episode of psychosis requires the use of standardized measurement tools and the longitudinal assessment of multiple outcome measures.
Key words: Firstepisode psychosis, course predictors, outcome criteria, illness course, schizophrenia.
Antecedentes y objetivos
La identificación de factores pronósticos en pacientes con esquizofrenia y otros trastornos psicóticos relacionados podría facilitar la comprensión de la etiología de estos trastornos así como mejorar los tratamientos existentes. Los primeros años a partir del primer episodio psicótico son un <<período crítico>> en que factores biológicos y psicosociales influyen en el pronóstico futuro del trastorno. El hecho de que tanto el curso temprano como los predictores de línea base hayan sido definidos según diferentes medidas ha dificultado la comparación entre estudios. Los estudios sobre el valor predictivo de características clínicas y demográficas de línea base en población mexicana son aún escasos. En este sentido, el presente estudio tiene como objetivos: 1. replicar el valor pronóstico de algunas características del paciente previamente relacionadas con el curso temprano de la psicosis en una muestra de pacientes mexicanos con un primer episodio psicótico, y 2. evaluar retrospectivamente su valor como predictores del diagnóstico final, la presencia de síntomas psicóticos residuales y el número de episodios psicóticos ocurridos al menos tres años más tarde.
Se recabó información acerca de variables predictoras de línea base (sociodemográficas, de fase premórbida, contexto del primer episodio, y dimensiones de la psicopatología) y sobre la evolución clínica (diagnóstico, síntomas psicóticos residuales, episodios psicóticos) de los expedientes de 51 pacientes con psicosis de curso temprano y con un tiempo de seguimiento disponible de al menos tres años (media = 5.8, SD = 2.1).
Un ajuste premórbido pobre y haber sido hospitalizado en el primer episodio psicótico fueron predictores significativos de un diagnóstico de esquizofrenia. Un nivel educativo más bajo y un inicio insidioso de la enfermedad fueron predictores significativos de la presencia de síntomas residuales. La hospitalización durante el primer episodio psicótico y puntuaciones altas en la dimensión psicótica al inicio de la enfermedad fueron predictores significativos de episodios psicóticos posteriores.
Se observó que un nivel educativo bajo incrementa el riesgo de síntomas residuales, posiblemente al dificultar la continuidad en el tratamiento. Se encontró que el ajuste premórbido pobre está relacionado con pacientes con esquizofrenia, corroborando así hallazgos previos sobre las altas puntuaciones de éstos en discapacidad premórbida, incluidos retraimiento social y relaciones disfuncionales con los pares. Un inicio insidioso predijo la presencia de síntomas residuales persistentes. La Duración de la Psicosis No Tratada (DPNT) puede actuar como mediador en tal asociación. La hospitalización en el primer episodio fue un factor pronóstico de diagnóstico de esquizofrenia y de múltiples episodios psicóticos. La gravedad y naturaleza de los síntomas en un primer episodio que requieran hospitalización es un factor a tomar en cuenta para explicar estas asociaciones. Los pacientes con episodios múltiples puntuaron más alto que los pacientes con un único episodio en la dimensión de psicoticismo. En concordancia con estudios previos, los pacientes con episodios psicóticos múltiples tuvieron puntuaciones más altas en la dimensión de psicoticismo que los pacientes con un episodio único de psicosis. La mayoría de los factores de línea base no predijeron el diagnóstico. Esto coincide con un enfoque dimensional de la psicosis, el cual sugiere que, aunque las psicosis esquizofrénicas y las no esquizofrénicas se clasifiquen como trastornos independientes, ambas existirían a lo largo de un continuo de psicosis, cruzando límites diagnósticos y compartiendo factores etiológicos y de riesgo en común. Actualmente, no sólo la mejoría de los síntomas psicóticos graves, sino también el buen funcionamiento psicosocial y la calidad de vida son metas viables. La exacerbación de síntomas y las hospitalizaciones pueden causar un deterioro acumulativo y afectar la reintegración social del paciente. Por ello, la prevención de recaídas es también un objetivo importante del tratamiento. La identificación de predictores confiables del curso de la enfermedad tiene importantes implicaciones para la planeación de tratamientos y servicios.
Se replicó el valor predictivo de varios factores en esta muestra de pacientes con psicosis, aunque los predictores parecen relacionarse de manera diferente con cada una de las tres medidas de curso temprano. Por ello, se ha de poner atención en la medición y seguimiento estandarizados de diversas medidas del curso de la enfermedad a fin de trazar un mapa completo de la evolución y el pronóstico de un primer episodio de psicosis.
Palabras clave: Primer episodio psicótico, predictores de curso, criterios de pronóstico, curso de la enfermedad, esquizofrenia.
Schizophrenia is one of the most disabling mental disorders; however, it can no longer be conceived as a hopeless and inevitable pathway to deterioration.1 The course following a first psychotic episode is clearly heterogeneous.2 Although schizophrenia is typically viewed as a chronic and episodic disorder, between 1222% of patients never relapse or experience residual symptoms after their first episode of psychosis.3'4
Although the course of psychosis is heterogeneous (whether treated or untreated), its presentation seems most severe and disturbing during the onset and the first years of illness.5 Eventually, between two and five years after the first episode, psychotic disorders appear to plateau and follow a more stable course.3 These first years following the initial episode of psychosis (the socalled <<critical period>>) are viewed as a crucial time during which biological and psychosocial changes have decisive effects on the patient. Characteristics assessed during the critical period provide promising predictors of patients' longterm outcome.5 Moreover, evidence6 indicates that the course and the severity of psychotic illnesses are predictable by year three (including, on average, 12 months of untreated psychosis).5
Current early intervention programs and research are based on the premises that this <<critical period>> influences the longterm course of psychosis and that the critical period is particularly malleable to intervention.5 Early intervention efforts aim at reducing suicide and relapse rates, preventing social and cognitive deterioration, and ameliorating persisting symptoms.7'8 These programs have a greater impact on illness course and outcome when applied in the early phase of the disorder.9 The identification of characteristics that predict clinical and functional outcomes in newly diagnosed psychosis patients should enhance our understanding of such disorders and provide guidance for treatment.
The complexity and heterogeneity of schizophrenia and related psychoses require reliable and valid measures of outcome to capture patients' functioning and impairment over time.3 Schizophrenic psychoses show, compared to schizoaffective and affective psychoses, a poorer global outcome, more deteriorating course, greater presence of negative symptoms, and more persistent impairments in several aspects of social life, such as communication and cognitive functions.10'11 A variety of clinical, functional and quality of life measures have been used to assess outcome,7 but this diversity makes the comparison among studies difficult.8
The most widely used outcome is diagnosis, which can be reliably established after approximately six months of onset of psychosis.12 Illness course is also extensively reported as an outcome measure, varying from a full recovery to a chronic deteriorating course.12 Some studies, simplifying the course of psychosis as <<poor>> or <<good>>, have defined course by relying either on the presence of residual symptoms4'13 or on the occurrence of subsequent relapses into acute psychosis.14 However, there is a shortage of studies comparing the impact of using either one or the other, particularly on their ability to evaluate the utility of putative prognostic indicators.
Studies have also differed in the premorbid and firstepisode factors analysed as possible predictors of outcome. Sociodemographic variables, clinical features, premorbid characteristics, context of presentation of the first episode of psychosis, and type of treatment have been the most common factors related to short and longterm prognosis. Literature on this topic is abundant, suggesting that various factors, such as early age at onset, male gender, single status, poor premorbid adjustment, lack of insight, and symptom severity at onset are highly related to poor outcome,3'15 although not all findings concur.7'8
Studies with Mexican first episodepsychosis patients indicate that this population does not differ significantly in its baseline demographic and clinical characteristics when compared to populations from developed countries.16 Although the predictive value of the DUP has been replicated in the Mexican population in a oneyear followup study,17 the predictive value of other first episode psychosis characteristics after the critical period needs study.
Research has identified so far important predictors of outcome. However, there is a shortage of studies analysing the association of premorbid and first episode variables with different outcome definitions. Furthermore, studies of shortterm course predictors in Mexican firstepisode psychosis patients are also limited. Therefore, this study aims to: 1. replicate the prognostic value of factors (sociodemographic, premorbid, context of the first psychotic episode, and psychopathology dimensions) previously related to the shortterm course of psychosis in retrospectively assessed Mexican firstepisode psychosis patients, and 2. assess their prognostic value in the prediction of final diagnosis, presence of psychotic residual symptoms, and number of psychotic episodes.
This is a retrospective case series study focusing on the shortterm course of psychosis in a cohort of patients who have received mental health care in the adult service of the Hospital Psiquiátrico Yucatán (HPY). The HPY is a public institution located in the city of Merida, Mexico, that offers inpatient and outpatient care to all patients in need. The HPY has a broad catchment area that includes patients from neighbouring states (e.g. Campeche, Quintana Roo); however, for this study, sampling was restricted to the inhabitants of the city of Merida. Data were collected through the review of all clinical files after obtaining formal authorization and ethical approval from the Hospital Committee. Additional inclusion criteria were: 1. occurrence of a first episode of psychosis between 1999 and 2005; 2. age at onset 1645 years; and 3. a primary current DSMIVTR12 diagnosis of schizophrenia, schizophreniform, schizoaffective disorder, delusional disorder, brief psychotic disorder, or psychosis not otherwise specified. Exclusion criteria were: 1. psychoses of affective, organic, or toxic type, 2. an evident intellectual disorder, and 3. no followup information available.
An initial random sample of 111 cases was selected. LG was responsible for the examination of the clinical histories and the review of current diagnoses according to DSMIVTR criteria12 as some might have changed since onset. Nine cases were excluded: 3 affective psychoses, 2 organic psychoses, 2 toxic psychoses, 1 missing file, and 1 case with a duplicated file. Furthermore, 51 cases with a followup time period shorter than three years were omitted. The final sample of 51 shortterm course psychosis patients included 23 men and 28 women, with an average age at first episode of 28.1 (SD=7.6). All cases in the sample were followed for at least three years (mean=5.8, SD=2.1) and had received antipsychotic medication.
The predictors identified at the first episode included: 1. sociodemographic data (gender, marital status, educational level, occupational status), 2. premorbid phase characteristics (premorbid adjustment, identified trigger, type of onset), 3. features of the context of the first episode (hospitalization, substance abuse, level of insight), and 4. dimensions of psychotic psychopathology. Classification of premorbid adjustment was based on the medical record information about possible i) learning, ii) behavioural, iii) emotional or iv) social difficulties present at any time before the first psychotic episode. Based on the available information from clinical files, premorbid adjustment was categorized as poor or good. For psychopathology, the recorded presence of symptoms corresponding to each of its three dimensions was rated by translating the clinical records information into the most representative Positive and Negative Syndrome Scale (PANSS)18 items based on the criteria provided by Andreasen et al.19 The psychoticism dimension included recorded symptoms of delusions, unusual thought content and hallucinatory behaviour; the disorganization dimension included symptoms of conceptual disorganization, mannerism or posturing; the dimension of negative symptoms included blunted affect, social withdrawal and lack of spontaneity.
Outcome was classified according to three criteria. First, DSMIVTR12 last available diagnoses were dichotomized into: 1. schizophrenia, and 2. other psychoses. A second criterion grouped cases as: 1. with residual symptoms, or 2. with no residual symptoms, at the time of the outcome assessment. A third criterion considered the number of psychotic episodes recorded during the followup period (including the initial episode), classifying cases as: 1. single episode, or 2. multiple episodes.
First, Pearson correlations were run to explore possible associations among the three outcome criteria. Next, separate regressions were computed for sociodemographic, premorbid phase, context of first psychotic episode, and psychopathology variables with the predictors for each analysis entered simultaneously. Statistical analyses were computed with SPSS, version 15.20
Table 1 presents the results of the binary logistic regressions predicting the three outcome measures.
Current diagnosis and number of psychotic episodes were significantly correlated (r=+0.32, p=0.02); that is, patients with schizophrenia were likely to have experienced more than one psychotic episode. The presence/absence of residual symptoms was not significantly associated with either current diagnosis (r=0.06) or to the number of psychotic episodes (r=+0.05).
Sociodemographic factors at onset were not associated with a diagnosis of schizophrenia or reports of relapse at the followup. However, lower educational level at onset was associated with heightened risk of residual symptoms at the followup.
Premorbid phase variables
Poor premorbid adjustment was significantly associated with a subsequent diagnosis of schizophrenia. An insidious onset was associated with residual symptoms, and oddly with those who did not relapse (single episode). Nevertheless, it must be pointed out that 11 of the 19 patients (57.9%) with a single episode presented residual symptoms; furthermore, all 11 had an insidious onset.
Context of first psychotic episode
Patients who were hospitalized during their first psychotic episode were more likely to have a diagnosis of schizophrenia and to relapse (multiple episodes) by the followup assessment. Substance abuse did not appear as a significant predictor of outcome, although only a small percentage of patients reported this abuse (n=6, 11.8%). Surprisingly, poor insight at the firstepisode identified patients who subsequently were diagnosed with psychoses other than schizophrenia and who did not relapse.
Dimensions of psychopathology at onset
Psychotic symptoms present at onset significantly related to multiple episodes, whereas disorganization related to absence of residual symptoms. Negative symptoms did not predict any of the three outcome measures.
Baseline characteristics are useful predictors of shortterm outcome in psychosis, yet they relate differently to particular outcome measures: schizophrenia was predicted by poor premorbid adjustment and hospitalization, residual symptoms by lower educational level and an insidious onset, whereas multiple psychotic episodes were related to hospitalization and psychoticism.
Predictors of outcome
The mean age at psychosis onset of this sample is higher than that of some other firstepisode studies, although it is consistent with other previous findings obtained in firstepisode Mexican patients.21 Thus, there seems to be a significant range in terms of age at onset, possibly due to sociological differences between regions of the country or to differences in the access to mental health care. In any case, further research should examine how differences in age at onset might impact the relative importance of different predictors of later outcome.
An earlier age at onset, male gender, single marital status, lower educational level, and no daily occupation, among other sociodemographic factors, have been associated with a poorer outcome in firstepisode psychosis patients;4'15 nevertheless, in this study they were not significantly relations to diagnosis or relapses. A low educational level at onset did increase the risk of residual symptoms, possibly because it hinders treatment continuity. A review found that higher education and good social functioning of patients with psychosis were associated with good adherence to treatment.22
Premorbid phase variables
Poor premorbid adjustment has been associated with more negative symptoms in the shortterm course of illness, less improvement in negative symptoms, and overall poorer clinical and social functioning. On the other hand, good premorbid adjustment has been related to better clinical outcome, not only in chronic schizophrenia, but also in affective psychoses,23 and in psychotic disorders that are substance induced.24 Premorbid adjustment appears to be an important predictor of diagnosis. Firstepisode psychotic patients who later develop schizophrenia compared to those who develop bipolar disorder have not only shown more persistent positive and negative symptoms at followup, but also higher ratings of premorbid impairment, including social withdrawal and dysfunctional peer relationships.25'26
In the present study, an insidious onset significantly predicted residual symptoms. This association might be mediated by DUP. An acute onset relates to shorter DUP in patients,27 possibly because the sudden changes and appearance of psychotic symptoms might well be more noticeable to patients and relatives, prompting treatment seeking. On the other hand, an insidious type of onset has been found predictive of longer DUP27 and poorer global psychopathological and psychosocial outcome.28 Although an insidious type of onset usually relates to a poor outcome, in our study it was surprisingly related to singleepisode outcome. However, this may simply reflect the relatively short followup period. Previous research has shown that relapses in the shortterm course of illness are not related to the type of onset1528 though they seemed related to DUP and to the delay in intensive psychosocial treatment.28
Context of first psychotic episode
The study of early psychosis has used hospitalization as an important outcome measure analysing predictive factors of rehospitalization,29'30 time spent in hospital,31 and time between hospitalizations.30 Here, we considered being hospitalized at first episode as a prognostic factor, resulting in significant risk for schizophrenia diagnosis and presence of multiple episodes at the reassessment, but not for residual symptoms. The severity and nature of symptoms at first episode that require hospitalization might account for the association with a later diagnosis of schizophrenia. It has been suggested that lacking objective measures of symptoms, hospitalization can be used as a <<proxy>> measure of a psychotic episode32 usually characterised by a significant deterioration due to positive symptoms. For this study, we considered hospitalization as a sign of severity and it was a significant prognostic factor of shortterm course. Multiple episodes were also related to hospitalization at first episode. In a retrospective study, Rosen and Garety4 found that hospitalization at first episode turned out to be a significant predictor only when the outcome definition took into account relapses and not only residual symptoms. Being hospitalized and under supervised treatment might have a more counteracting effect on residual symptoms, but not on the likelihood of relapse.
Poor insight has been associated with poorer cognitive functioning,33 increased risk of relapse,15;34 and readmission.34 On the other hand, good insight of illness has been related to higher levels of depression.33'35. Furthermore, most evidence supports an association throughout the first years after an initial episode between poor insight and increased symptoms,33'35 though not all findings concur.34 Various studies support the assumption of a causal chain connecting poor insight with poor treatment adherence and thus with impaired outcome and functioning; although this seems apparent during the treatment phase, the association with longterm adherence remains unclear.36 In our study, insight was not considered at present but at the time of the first episode. Contrary to expectations, poor insight was significantly associated to other nonaffective psychoses and singleepisode outcome. These results are not easy to explain based on the information available and important factors that might mediate this effect (e.g. severity and nature of symptoms at onset, perception of condition as a mental disorder) must be considered on standardized prospective assessments.
Dimensions of psychopathology at onset
None of the three dimensions of psychopathology were associated with subsequent diagnoses. Although the result could be due to insufficient statistical power, an alternative possibility is that the nature of psychotic psychopathology at onset, though more severe in schizophrenic psychoses, is not specifically associated with later diagnoses. This would be consistent with research indicating that schizophrenia, schizoaffective disorder, and affective illness share common features and a general set of aetiological and risk factors.37/38
Psychotic (positive) symptoms were more common than disorganized and negative symptoms at the onset for the whole sample and for all groups. This was not surprising, given that positive symptoms typically herald the onset of a first, acute episode, and because for many patients negative symptoms develop as part of a chronic course of the disorder. The psychotic dimension was only significant when predicting multiple vs. singleepisode patients, with the former group scoring higher. A 7year followup of schizophrenic outpatients showed that lower positive symptoms were characteristic of those patients who did not relapse.39
Only a few patients displayed disorganized symptoms in their clinical histories, although surprisingly those with higher scores were more likely to be part of the nonresidual symptom group. Unfortunately, the retrospective nature of the study restricts information to explore further the prognostic nature of these results.
For our sample, negative symptoms were not significant predictors of outcome. Negative symptoms at onset tend to be associated with residual symptoms more than other dimensions of psychosis or premorbid personality.40 Moreover, in a retrospective study comparing groups of patients with single or multiple psychotic episodes, negative symptoms at first contact was the only dimension of psychopathology that stood out as a significant prognostic factor.4 However, we did not replicate this finding. However, this may reflect that information on negative symptoms was not as readily noted and recorded as were the more striking positive and disorganized symptoms that typically signal the onset of a psychotic episode. Furthermore, negative symptoms at the time of first episode might be masked by those symptoms that cause severe behavioural disturbances, or they might evolve later in the course of illness.
Analysis of outcome criteria
Among the three selected shortterm course measures, only diagnosis and number of episodes were significantly related, as most patients presented schizophrenia and had suffered multiple psychotic episodes. However, their only common prognostic factor of <<poor outcome>> was hospitalization. Hence the importance of using different measures to map with completeness the course and outcome of firstepisode psychosis.
Poor premorbid adjustment and hospitalization at first episode were the only significant predictors of schizophrenia; hence, at first episode, clinical assessment must place particular attention to patients who require hospitalization and who have presented previous difficulties, as they are in higher risk to develop schizophrenia. Most baseline factors did not predict diagnosis. Even though schizophrenia implies a general poorer outcome than other psychoses, both affective and nonaffective,10'11 whether they differ etiologically is an issue still debated.37'41 Some results suggest that even early in its course schizophrenia is distinguishable not only from affective psychoses,42 but also from schizoaffective disorders.43 Nevertheless, other findings suggest that schizophrenia has some overlapping features with schizoaffective disorder (e.g. cognitive performance)41 and even with bipolar disorder.37 A dimensional view of psychosis suggests that even though schizophrenic and nonschizophrenic psychoses are seen as distinct entities, they would exist along a continuum of psychosis that crosses diagnostic boundaries and would have in common aetiological and risk factors.38 Moreover, whether schizophrenia can be predicted at onset has also important clinical implications, as it involves that it might not be appropriate to make predictions at firstepisode regarding diagnosis. As it is wellknown, stating a premature diagnosis of schizophrenia can have adverse consequences for clinicians (e.g., therapeutic nihilism) and patients (e.g., hopelessness, stigma, demoralisation and depression).
A lower educational level and insidious onset were significantly related to patients with residual symptoms. These factors stand out as robust baseline predictors that hold predictive value overandabove methodological similarities and differences among studies. The amelioration of core signs and symptoms is indispensable but not enough for recovery because persistent symptomatology, even if at a low level of severity, can interfere with behaviour and functioning, hindering patients' social, educational, and occupational development, and their chances of social reintegration.44 Thus, the possibility of identifying patients at firstepisode likely to suffer residual symptomatology has significant implications for treatment and service planning.
Higher functioning, lower positive symptoms, higher ability in selfcare, and higher IQ relate to single episode patients,39 whereas poor insight,15 poorer premorbid childhood functionality, and noncompliance to the treatment highly contribute to relapse risk.45 In the present study, hospitalization at first episode and psychotic symptoms significantly predicted multiple psychotic episodes. Relapses may have an important effect not only on the clinical, but also on the social functioning of patients. Exacerbation of symptoms and hospitalizations might cause cumulative deterioration in functioning and a diminished ability to maintain employment and relationships.44 Thus, early intervention as well as standard treatment programs in psychosis must work to prevent relapses and to promote the maintenance of a stable clinical status.46
Though the study of the course of psychosis should ideally rely on a prospective design, a retrospective study provides valuable information over a period of time and is recommended as a sensible starting point when research on this topic is developing at a new site. The number of baseline measures had to be restrained depending on the availability from casenotes; this might be useful to draw attention on what factors clinicians pay attention to in daily practice, as well as on how they record information.
Subsequent research may include patients who are inhabitants of other communities, which could yield interesting data on the search of mental health care, availability of services, and awareness of illness. Broadening the inclusion criteria to other types of psychosis such as affective, toxic, and organic, might also provide useful information of the vast psychosis spectrum. In a prospective study, a thorough exploration of the premorbid phase, the onset characteristics, and clinical family background at firstpsychotic episode would certainly enrich the possibility of significant and generalizable findings.
Historically, schizophrenia and related psychoses have been characterized erroneously as necessarily having a deteriorating course. However, the course of these disorders is heterogeneous with many patients showing good recovery. Three alternative definitions of shortterm course were retrospectively analyzed in a Mexican sample of firstepisode psychosis patients: final diagnosis, presence of residual symptoms, and number of psychotic episodes. Findings indicate that some baseline variables are useful predictors for this particular population, and they appear to relate differently to particular outcome measures. Given that not all predictors relate similarly to different outcome measures, attention must be placed on the standardized and discreet assessment of varied predictors and outcome indexes.
LGR thanks the Consejo Nacional de Ciencia y Tecnología (CONACyT), Mexico, for funding number 187498.
NBV and TRK are thankful to Generalitat de Catalunya for their support (2009SGR672).
Authors thank all clinicians and patients from Hospital Psiquiátrico Yucatán for making this research possible, and Hospital Committee for facilities and support.
1. Penn DL, Waldheter EJ, Perkins DO, Mueser KT et al. Psychosocial treatment for firstepisode psychosis: A research update. Am J Psychiatry 2005;162(12):22202232. [ Links ]
2. Shepherd M, Watt D, Falloon I, Smeeton N. The natural history of schizophrenia: A fiveyear followup study of outcome and prediction in a representative sample of schizophrenics. Psychol Med 1989;15:146. [ Links ]
3. Barnes TRE, Pant A. Longterm course and outcome of schizophrenia. Psychiatry 2005;4(10):2932. [ Links ]
4. Rosen K, Garety P. Predicting recovery from schizophrenia: A retrospective comparison of characteristics at onset of people with singe and multiple episodes. Schizophr Bull 2005;31(3):7357750. [ Links ]
5. Birchwood M. The critical period for early intervention. In: Birchwood M, Fowler D, Jackson C (eds.). Early intervention in psychosis: A guide to concepts, evidence and interventions. Chichester, UK: John Wiley & Sons Ltd; 2000; pp.2863. [ Links ]
6. Wiersma D, Nienhuis FJ, Slooff CJ, Giel R. Natural course of schizophrenic disorders: A 15year followup of a Dutch incidence cohort. Schizophr Bull 1998;24(1):7585. [ Links ]
7. Malla AK, Norman RMG, Joober R. Firstepisode psychosis, early intervention, and outcome: What have we learned? Can J Psychiatry 2005;50(14):881891. [ Links ]
8. Menezes NM, Arenovich T, Zipursky RB. A systematic review of longitudinal outcome studies of firstepisode psychosis. Psychol Med 2006;36(10):13491362. [ Links ]
9. McGorry PD, Killackey E, Yung A. Early intervention in psychosis: Concepts, evidence and future directions. World Psychiatry 2008;7:148156. [ Links ]
10. Marneros A, Deister A, Rohde A. Psychopathological and social status of patients with affective, schizophrenic and schizoaffective disorders after longterm course. Acta Psychiatr Scand. 1990;82(5):352358. [ Links ]
11. Móller H, Bottlender R, Wegner U, Wittmann J et al. Longterm course of schizophrenic, affective and schizoaffective psychosis: Focus on negative symptoms and their impact on global indicators of outcome. Acta Psychiatr Scand 2000;102:5457. [ Links ]
12. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders (4th ed., text revision). Washington, DC; 2000. [ Links ]
13. VázquezBarquero JL, CuestaNuñez MJ, De la Varga M, HerreraCastanedo S et al. The Cantabria first episode schizophrenia study: A summary of general findings. Acta Psychiatr Scand 1995;91(3):156162. [ Links ]
14. Altamura AC, Bassetti R, Sassella F, Salvadori D et al. Duration of untreated psychosis as a predictor of outcome in firstepisode schizophrenia: A retrospective study. Schizophr Res 2001;52(1):2936. [ Links ]
15. GómezdeRegil L, Kwapil TR, Blanqué JM, Vainer E et al. Predictors of outcome in the early course of firstepisode psychosis. Eur J Psychiat 2010 (in press). [ Links ]
16. Apiquian R, Ulloa RE, Páez F, Nicolini H. The Mexican firstepisode psychotic study: Clinical characteristics and premorbid adjustment. Schizophr Res 2002;53(1):161163. [ Links ]
17. ApiquiánGuitart R, FresánOrellana A, GarciaAnaya M, LóyzagaMendoza C et al. Impacto de la duración de la psicosis no tratada en pacientes con primer episodio psicótico. estudio de seguimiento a un año. Gac Méd Méx 2006;142(2):113120. [ Links ]
18. Kay SR, Fiszbein A, Opler LA. The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophr Bull 1987;13:261276. [ Links ]
19. Andreasen NC, Carpenter WTJ, Kane JM, Lasser RA et al. Remission in schizophrenia: Proposed criteria and rationale for consensus. Am J Psychiatry 2005;162(3):441449. [ Links ]
20. SPSS Inc. SPSS base 15.0 Users's guide. USA: PrenticeHall; 2006. [ Links ]
21. Apiquián R, Páez F, Loyzaga C, Cruz E et al. Estudio mexicano sobre el primer episodio psicótico: Resultados preliminares, características sociodemográficas y clínicas. Salud Mental 1997 10;20:17. [ Links ]
22. Nosé M, Barbui C, Tansella M. How often do patients with psychosis fail to adhere to treatment programmes? A systematic review. Psychol Med 2003;33(7):11491160. [ Links ]
23. Haim R, Rabinowitz J, Bromet E. The relationship of premorbid functioning to illness course in schizophrenia and psychotic mood disorders during two years following first hospitalization. J Nerv Ment Dis 2006;194(10):791795. [ Links ]
24. Caton CLM, Hasin DS, Shrout PE, Drake RE et al. Predictors of psychosis remission in psychotic disorders that cooccur with substance use. Schizophr Bull 2006;32(4):618625. [ Links ]
25. Werry JS, McClellan JM, Chard L. Childhood and adolescent schizophrenic, bipolar, and schizoaffective disorders: A clinical and outcome study. J Am Acad Child Adolesc Psychiatry 1991;30(3):457465. [ Links ]
26. McClellan J, McCurry C Early onset psychotic disorders: Diagnostic stability and clinical characteristics. Eur Child Adolesc Psychiatry 1999;8(1): 1319. [ Links ]
27. Morgan C, AbdulAl R, Lappin JM, Jones P et al. Clinical and social determinants of duration of untreated psychosis in the ÆSOP firstepisode psychosis study. Br J Psychiatry 2006;189:446452. [ Links ]
28. Bromet EJ, Naz B, Fochtmann LJ, Carlson GA et al. Longterm diagnostic stability and outcome in recent firstepisode cohort studies of schizophrenia. Schizophr Bull 2005;31(3):639649. [ Links ]
29. De Haan L, Linszen DH, Lenior ME, Doderlein de Win E et al. Duration of untreated psychosis and outcome of schizophrenia: Delay in intensive psychosocial treatment versus delay in treatment with antipsychotic medication. Schizophr Bull 2003;29(2):341348. [ Links ]
30. Miettunen J, Lauronen E, Veijola J, Koponen H et al. Patterns of psychiatric hospitalizations in schizophrenic psychoses within the Northern Finland 1966 birth cohort. Nord J Psychiatry 2006;60(4):286293. [ Links ]
31. Jablensky A, Sartorius N. What did the WHO studies really find? Schizophr Bull 2008;34(2):253255. [ Links ]
32. Braw Y, Bloch Y, Mendelovich S, Ratzoni G et al. Cognition in young schizophrenia outpatients: Comparison of firstepisode with multiepisode patients. Schizophr Bull 2008;34(3) :544554. [ Links ]
33. Saeedi H, Addington J, Addington D. The association of insight with psychotic symptoms, depression, and cognition in early psychosis: A 3year followup. Schizophr Res 2007;89(1):123128. [ Links ]
34. Drake RJ, Dunn G, Tarrier N, Bentall RP et al. Insight as a predictor of the outcome of firstepisode nonaffective psychosis in a prospective cohort study in England. J Clin Psychiatry 2007;68(1):8186. [ Links ]
35. McEvoy JP, Johnson J, Perkins D, Lieberman JA et al. Insight in firstepisode psychosis. Psychol Med 2006;36(10):13851393. [ Links ]
36. Lincoln TM, Lüllmann E, Rief W. Correlates and longterm consequences of poor insight in patients with schizophrenia. A systematic review. Schizophr Bull 2007;33(6):13241342. [ Links ]
37. Murray RM, Sham P, Van Os J, Zanelli J et al. A developmental model for similarities and dissimilarities between schizophrenia and bipolar disorder. Schizophr Res 2004;71(2):405416. [ Links ]
38. Tsuang MT, Stone WS, Faraone SV. Toward reformulating the diagnosis of schizophrenia. Am J Psychiatry 2000;157(7):10411050. [ Links ]
39. Di Michele V, Bolino F, Mazza M, Roncone R et al. Relapsing versus non relapsing course of schizophrenia: A cohort study in a community based mental health service. Epidemiol Psichiatr Soc 2007;16(1):5058. [ Links ]
40. Cuesta MJ, Peralta V, Gil P, Artamendi M. Premorbid negative symptoms in firstepisode psychosis. Eur J Psychiat 2007;21(3):220229. [ Links ]
41. Heinrichs RW, Ammari N, Vaz SM, Miles AA. Are schizophrenia and schizoaffective disorder neuropsychologically distinguishable? Schizophr Res 2008;99(1):149154. [ Links ]
42. Mojtabai R, Bromet EJ, Harvey PD, Carlson GA et al. Neurological differences between firstadmission schizophrenia and psychotic affective disorders. Am J Psychiatry 2000;157(9):14531460. [ Links ]
43. Jager M, Bottlender R, Strauss A, Moller H. Fifteenyear followup of ICD10 schizoaffective disorders compared with schizophrenia and affective disorders. Acta Psychiatr Scand 2004;109(1):3037. [ Links ]
44. Nasrallah HA, Lasser R. Improving patient outcomes in schizophrenia: Achieving remission. J Psychopharmacol 2006; 11;20(6):5761. [ Links ]
45. Ücok A, Polat A, Cakir S, Gene A. One year outcome in first episode schizophrenia: Predictors of relapse. Eur Arch Psychiatry Clin Neurosci 2006;256(1):3743. [ Links ]
46. Taylor M, Chaudhry I, Cross M, McDonald E et al. Towards consensus in the longterm management of relapse prevention in schizophrenia. Hum Psychopharmacol Clin Exp 2005;20(3):175181. [ Links ]