Risk factors for relapse following treatment for first episode psychosis: A systematic review and meta-analysis of longitudinal studies

Centre for Youth Mental Health, The University of Melbourne, Australia
Schizophrenia Research (Impact Factor: 3.92). 06/2012; 139(1-3):116-28. DOI: 10.1016/j.schres.2012.05.007
Source: PubMed


Preventing relapse is an essential element of early intervention in psychosis, but relevant risk factors and precise relapse rates remain to be clarified. The aim of this study was to systematically compile and analyse risk factors for and rates of relapse in the early course of psychosis.
Systematic review and meta-analysis of English and non-English language, peer-reviewed, longitudinal studies, with a minimum 12-month follow-up and at least 80% of participants diagnosed with a first episode of psychosis (FEP) that reported risk factors for relapse.
Of 153 potentially relevant articles, 29 were included in the study. Pooled prevalence of relapse of positive symptoms was 28% (range=12-47%), 43% (35-54%), 54% (40-63%) at 1, 1.5-2, and 3 years follow-up, in that order. A total of 109 predictors were analysed, with 24 being assessed in at least 3 studies. Of those, 20 predictors could be extracted for meta-analysis. Medication non-adherence, persistent substance use disorder, carers' critical comments (but not overall expressed emotion) and poorer premorbid adjustment, increased the risk for relapse 4-fold, 3-fold, 2.3-fold and 2.2-fold, respectively.
Clinical variables and general demographic variables have little impact on relapse rates. Conversely, non-adherence with medication, persistent substance use disorder, carers' criticism and poorer premorbid adjustment significantly increase the risk for relapse in FEP. Future studies need to address the methodological limitations of the extant research (e.g. definition of relapse), focus on the identification of protective factors and evaluate theoretically derived models of relapse.

97 Reads
  • Source
    • "High levels of carer burden have been found to be present in various chronic illnesses such as dementia [8-10], bipolar disorder [11] and psychosis [3]. Carer burden is particularly high during first episode psychosis (FEP) [5], and there is evidence to suggest that this is a predictor of poor long-term outcome for the patient [12]. However our understanding of this association is poor. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Background Carer burden is high during First Episode Psychosis (FEP) and evidence suggests that this is a predictor of poor long-term outcome. However our understanding of factors associated with higher burden is poor. We propose that carers’ cultural backgrounds and health belief models will influence their perceived burden of care, over and above that explained by severity of illness. Methods Patients with FEP and their primary Carers were recruited from the Early Intervention Service. Patients and Carers completed a range of validated measures, self-report ethnicity and demographic information together with the Multidimensional Health Locus of Control and Caregiver Burden Inventory. Results Significant correlations were found between carer burden and health beliefs, which differed by ethnicity and gender. High physical burden was experienced by Black carers with an external locus of control; time restrictions and emotional burden correlated with an external locus of control in Asian carers. For White carers, external locus of control correlated with time dependence burden. In all ethnic groups female carers experienced more time dependency, physical and developmental burden. No significant correlations were found between patient measures of severity or duration of illness and carer burden. Conclusions The type of burden experienced by carers differed between gender and ethnicity and was related to their health belief models. Thus the explanation and understanding of illness appears to be more salient than simply a patient’s severity of illness when considering the development of carer burden. Interventions to tackle high carer burden, and thus expressed emotion to improve outcome in patients, may need increasing focus here.
    BMC Psychiatry 06/2014; 14(1):171. DOI:10.1186/1471-244X-14-171 · 2.21 Impact Factor
  • Source
    • "First, it has been argued that the components of EE are related to different variables and are best studied separately. Alvarez-Jimenez and colleagues (Alvarez-Jimenez et al., 2012 ; Álvarez-Jiménez et al., 2008 ) found that only criticism seemed to predict relapse , while emotional over- involvement was more strongly related to caregiver dis - tress . Second , the view of EE as a marker for a ' dysfunc tional family ' has been criticized ( Jansen et al. , 2013 ; Van Os , Marcelis , Germeys , Graven & Delespaul , 2001 ) , suggesting that especially over - involvement is a natural and understandable reaction to a crisis and is associated with caregivers who are involved in treatment and care . "
    [Show abstract] [Hide abstract]
    ABSTRACT: In first-episode psychosis, the family is considered an important part in the recovery process. This is often accompanied by significant distress, which is acknowledged in numerous studies. However, little is known about the psychological factors involved. One hundred and twenty-seven caregivers of persons with first-episode psychosis completed a series of questionnaires aimed at investigating the contribution of expressed emotion and metacognitions to caregiver distress. Linear mixed model analysis found that emotional over-involvement and metacognitions independently predicted caregiver distress. Mediation analysis using bootstrapping showed that emotional over-involvement could be seen as mediating the effect of metacognitions on distress. The current study is a first step towards understanding the role of metacognitions in caregiver distress, thus opening up for the possibility of using interventions from ‘contextual behaviour therapies’. Implications and future studies are discussed. Copyright
    Clinical Psychology & Psychotherapy 05/2014; DOI:10.1002/cpp.1907 · 1.66 Impact Factor
  • Source
    • "Patients showed lower IQ in childhood and adolescence (Woodberry et al. 2008; Dickson et al. 2012) Unclear effect sizes a Worldwide prevalence is around 0.3% for 1-year prevalence, 0.5% for lifetime prevalence and 0.72% for lifetime morbid risk prevalence (Saha et al. 2005) Worldwide incidence is between 11 and 15.2 per 100 000 with a 5.6-fold variance across regions (McGrath et al. 2004) Moderate-quality evidence Large effect sizes a Patients have experienced increased exposure to Toxoplasma gondii antibodies in utero (Fekadu et al. 2010), maternal diabetes, low birth weight (<2000 g) (Cannon et al. 2002) and childhood adversities (Matheson et al. 2012) There are increased rates of schizophrenia in black Caribbean and black African immigrant groups and in their descendants, particularly those living in white communities (Cantor-Graae & Selten, 2005; Bourque et al. 2011; Kirkbride et al. 2012) Patients have increased mortality due to natural causes, particularly from cardiovascular, coronary, digestive, endocrine, infectious, genito-urinary, neoplastic, neurological and respiratory diseases (Saha et al. 2007) Patients have an increased risk of visual impairment (Oud & Meyboom-De Jong, 2009) Patients show higher rates of smoking (de Leon & Diaz, 2005) Medium effect sizes a Patients have increased rates of exposure to urbanicity (Vassos et al. 2012), emergency caesarean section, congenital malformations and uterine atony (Cannon et al. 2002) Patients have increased reporting of childhood central nervous system viral infections (Khandaker et al. 2012) There are increased rates of schizophrenia in first-and second-generation immigrants, particularly immigrants from developing countries (McGrath et al. 2004; Cantor-Graae & Selten, 2005; Saha et al. 2005; Bourque et al. 2011) Increased internalized stigma is related to reduced hope, self-esteem, empowerment, self-efficacy, quality of life, social support and treatment adherence (Livingston & Boyd, 2010) Lifetime risk of suicide in patients is around 1.8%, with risk in the earlier stages of the illness being around 5.6% (Palmer et al. 2005). Factors significantly associated with suicide include a history of deliberate self-harm, hopelessness, feelings of guilt or inadequacy, depressed mood, suicidal ideas and a family history of suicide (Large et al. 2011b) Increased positive symptoms, negative symptoms and general psychopathology are related to decreased subjective and objective quality of life and general well-being (Eack & Newhill, 2007) Longer duration of untreated psychosis is related to poorer clinical and social outcomes, and poorer response to treatment (Marshall et al. 2005; Large & Nielssen, 2008; Farooq et al. 2009; Boonstra et al. 2012) Rates of positive symptom relapse are around 28% at 1 year post-treatment and up to 54% at 3 years post-treatment and are associated with substance use, poor treatment adherence, high levels of critical family comments and poor pre-morbid adjustment (Alvarez-Jimenez et al. 2012) Physicians are the most likely first point of contact and the most common referral source to mental health care is emergency services (Anderson et al. 2010) Small effect sizes a Patients have increased rates of pre-morbid traumatic brain injury (Molloy et al. 2011) There is an increased risk of schizophrenia with winter/spring births, increased latitude and decreased annual mean daily temperature in the northern hemisphere (Kinney et al. 2009) Patients have increased childhood and early adolescent social withdrawal, anxiety, depression, social maladjustment, deviant behaviour, aggression, disruptiveness, delusions, hallucinations and general psychopathology. They also have delays in onset and development of talking, poor receptive and expressive language, and poor oral and reading skills in childhood (Tarbox & Pogue-Geile, 2008; Welham et al. 2009; Rubio et al. 2012) Patients have older fathers at birth (>50 years) (Miller et al. 2011) "
    [Show abstract] [Hide abstract]
    ABSTRACT: Background. True findings about schizophrenia remain elusive; many findings are not replicated and conflicting results are common. Well-conducted systematic reviews have the ability to make robust, generalizable conclusions, with good meta-analyses potentially providing the closest estimate of the true effect size. In this paper, we undertake a systematic approach to synthesising the available evidence from well-conducted systematic reviews on schizophrenia. Method. Reviews were identified by searching Medline, EMBASE, CINAHL, Current Contents and PsycINFO. The decision to include or exclude reviews, data extraction and quality assessments were conducted in duplicate. Evidence was graded as high quality if reviews contained large samples and robust results; and as moderate quality if reviews contained imprecision, inconsistency, smaller samples or study designs that may be prone to bias. Results. High- and moderate-quality evidence shows that numerous psychosocial and biomedical treatments are effective. Patients have relatively poor cognitive functioning, and subtle, but diverse, structural brain alterations, altered electrophysiological functioning and sleep patterns, minor physical anomalies, neurological soft signs, and sensory alterations. There are markers of infection, inflammation or altered immunological parameters; and there is increased mortality from a range of causes. Risk for schizophrenia is increased with cannabis use, pregnancy and birth complications, prenatal exposure to Toxoplasma gondii, childhood central nervous system viral infections, childhood adversities, urbanicity and immigration (first and second generation), particularly in certain ethnic groups. Developmental motor delays and lower intelligence quotient in childhood and adolescence are apparent. Conclusions. We conclude that while our knowledge of schizophrenia is very substantial, our understanding of it remains limited.
    Psychological Medicine 02/2014; 44(16). DOI:10.1017/S0033291714000166 · 5.94 Impact Factor
Show more