- Email: [email protected]
Risk factors in early and late onset schizophrenia
Risk Factors in Early and Late Onset Schizophrenia Laura Chen, Ajit Selvendra, Anne Stewart, David Castle PII: DOI: Reference:
S0010-440X(17)30221-3 doi: 10.1016/j.comppsych.2017.09.009 YCOMP 51896
To appear in:
Comprehensive Psychiatry
Please cite this article as: Chen Laura, Selvendra Ajit, Stewart Anne, Castle David, Risk Factors in Early and Late Onset Schizophrenia, Comprehensive Psychiatry (2017), doi: 10.1016/j.comppsych.2017.09.009
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Title:
RI P
T
Risk Factors in Early and Late Onset Schizophrenia
Authors:
Melbourne Medical School, The University of Melbourne, Parkville, Victoria, Australia
b
MA NU
a
SC
Laura Chena, Ajit Selvendraa,b, Anne Stewarta, David Castlea,b
Mental Health Service, St. Vincent’s Hospital, Fitzroy, Victoria, Australia
CE
PT
Dr. Ajit Selvendra Consultant Psychiatrist St. Vincent’s Mental Health, Hawthorn CMHS, 642 Burwood Rd, Hawthorn East, 3123 Victoria, Australia
ED
Corresponding Author:
Phone: +613-9231-5900
AC
Email: [email protected]
ACCEPTED MANUSCRIPT
T
Abstract
RI P
Background: This study looks at key risk factors in patients with schizophrenia to identify trends according to age of onset, comparing presentations prior to 26 years (youth onset), between 26 and 40 years (middle onset), and after 40 years of age (late onset).
SC
Methods: The early psychosis program at St Vincent’s Hospital Melbourne treats patients presenting in the early stages of psychosis between 16 and 65 years of age. A database was
MA NU
developed to capture key risk factors in all patients with an eventual diagnosis of schizophrenia (n=225). Risk factor profiles were then generated and compared for patients based on age of onset.
Results: Older age of onset was associated with weaker family history of schizophrenia, lower rates of substance use, better early psychosocial functioning and higher educational
ED
achievement. Female preponderance and comorbid physical health problems were particularly notable in the late onset cohort. Later life schizophrenia also showed a relatively
PT
greater association with psychosocial factors proximal to psychosis onset, such as unemployment.
CE
Discussion: Clear trends are noticeable with age. Older patients have characteristic differences in their background risk factors compared to youth onset patients, including less hereditary influence and relatively more emphasis on later life risk factors. Identifying the
AC
roles of specific risk factors in these distinct age-onset groups can enhance our understanding of underlying aetiology and facilitate service development to meet the needs of each specific age group.
Keywords Schizophrenia Late onset schizophrenia Risk factors in schizophrenia Early psychosis
1
ACCEPTED MANUSCRIPT
T
1. Introduction
RI P
Schizophrenia is a heterogeneous condition, capturing psychoses presentations of variable symptom profiles and severity (1). While the majority of cases occur before middle age, a
SC
significant portion of new onset cases occur in the later decades of life (2). Presentations of
MA NU
new-onset schizophrenia at different life stages can vary considerably, and has led to international consensus in classifying schizophrenia with onset prior to 40 years of age(yoa) as Early-Onset Schizophrenia (EOS1), those with onset at and after 40yoa as Late-Onset Schizophrenia (LOS2), and those with onset after 60yoa as Very-Late-Onset Schizophrenia-
ED
like psychosis (vLOS3) (3). In countries like Australia, there is an additional focus on specifically youth-onset schizophrenia, with the cut-off age for most early-psychosis
PT
programmes being 26 years of age. This age-based approach may facilitate a better
CE
understanding and management of schizophrenia at different life-stages.
AC
The aetiopathology of schizophrenia remains poorly understood, with many potential models being proposed (4, 5). Research has focused on identification of associated risk factors to elucidate the underlying mechanisms of the disorder, and has helped yield greater understanding of disease development and progression. Due to the youth preponderance in schizophrenia and tendency of past research to look at schizophrenia patients as a homogenous group, there is a much poorer understanding of age-related differences in schizophrenia, and the specific features and risk factors relevant to late onset cases.
1
EOS – Early Onset Schizophrenia. Schizophrenia with onset prior to 40 years of age LOS – Late Onset Schizophrenia. Schizophrenia with onset after 40 years of age 3 vLOS – Very Late Onset Schizophrenia-like psychosis. Schizophrenia with onset after 60 years of age 2
2
ACCEPTED MANUSCRIPT Early onset schizophrenia presents most commonly during early adulthood (late teens to early twenties) and is associated with predominant male gender, a positive family history of
T
schizophrenia and other psychiatric conditions, particularly affective disorders (2, 6-9). Early
RI P
life socioenvironmental risk factors, particularly if present during key developmental periods (early childhood and adolescence), are also implicated. These include factors such as urban
SC
rearing environment, childhood trauma, migration and minority status, traumatic brain injury and substance use (10-15). Comorbid psychiatric illness, particularly depression, anxiety and
MA NU
substance use disorders are frequently associated with and precede psychosis onset (16, 17).
Comparatively few studies have looked specifically at risk factors in late onset schizophrenia.
ED
Existing literature suggest possible differences in the background risk factors of older individuals including greater female preponderance, associated family history of affective
PT
disorders, and cluster A personality traits (6, 18, 19). A few focal studies also suggest an important role for socioenvironmental factors such as migration and trauma exposure (20,
CE
21). Later life stressors and the notion of accumulated loss appear to be important in the
AC
precipitation of psychosis in older patients (3), but the specific influence of many other schizophrenia-associated risk factors remain unclear.
1.1 Aim
This study aims to review the prevalence of risk factors among patients treated for schizophrenia in the St Vincent’s Hospital early psychosis program between mid-2006 to mid-2015, and to compare these among different age-onset groups.
3
ACCEPTED MANUSCRIPT 2. Methods St. Vincent’s Hospital covers inner city and inner urban municipalities in Melbourne,
T
Australia, and provides comprehensive care for psychosis. The Early Psychosis program
RI P
captures all people within the St Vincent’s catchment area aged 16-65 presenting with their
MA NU
months of treatment for a first episode of psychosis.
SC
first episode of untreated psychosis (of any type), or those who are within their first 18
A file review of potential risk factors for psychosis was performed using patient medical records. Subjects included all patients from the St Vincent’s Mental Health Service Early Psychosis Program with an eventual diagnosis of schizophrenia (n=225). Diagnoses of
ED
schizophrenia were made using a DSM IV format at 3 months after all treating clinicians, including the regular treating psychiatrist, had seen the case and come to a consensus. If there
PT
was any lack of diagnostic clarity, the patient’s history was reviewed by and discussed with
CE
the early psychosis consultant to clarify the final diagnosis.
AC
A range of literature-based risk factors for schizophrenia were examined and a subset of the better captured and clinically relevant factors were further analysed for trends and associations. These factors are defined in Table A.1 of Appendix A. Trends were then reviewed in respect to 3 subgroups, defined by onset before 26 (youth onset), between 26 and 40 (middle onset), and after 40 years of age (late onset). Age cut-offs were based on current Australian youth-psychosis program definitions (22) (under 26 versus 26 and older), and international consensus (3) (40 and under versus over 40). Results were graphed to identify patterns related to age and gender (Figure 1), and chi-square testing conducted to assess significance of trends and differences between the youth onset subgroup and each older onset subgroup (Table 1). Risk factor definitions were strictly operationalised to ensure accuracy of
4
ACCEPTED MANUSCRIPT identification, and inter-rater reliability testing conducted by comparing file reviews of the same patients conducted independently by different reviewers. The results compared
SC
RI P
T
demonstrated good inter-rater consistency in risk factors identified.
MA NU
3. Results 3.1 Figures & Tables
ED
Figure 1. Age of onset, based on gender
CE
PT
Table 1. Comparison of risk factor frequencies in different age onset groups
225 cases of diagnosed schizophrenia from the St Vincent’s early psychosis program were
AC
identified and included in the study. This included 104 (46%) cases with youth onset, 81 (36%) cases with middle onset, and 40 (18%) cases with late onset. Of the late onset cohort, 2 individuals had disease onset after 60yoa, and for the purposes of this study were included within the group of ‘late onset schizophrenia’. The peak age of onset in the study sample was between 17 and 22 years, with median age of 27. A similar peak in early life was observed in both sexes though significantly higher amongst males (Figure 1). Male and female prevalence rates approximated each other in later life, with a slight female preponderance.
A marked reduction in the rate of positive family history of schizophrenia was seen with increased age of onset (Table 1). Statistically significant age-associated trends were also seen 5
ACCEPTED MANUSCRIPT in early psychosocial impairment (lesser with increased age of onset), levels of educational completion (higher with increased age of onset), and rates of both recent and lifetime
T
substance use (statistically significant and lower in the late onset cohort). An increased
RI P
frequency of unemployment and comorbid physical health problems was identified in the late onset group specifically. Similar rates of reported childhood trauma, and previous psychiatric
SC
conditions were found in the different age groups. Rates of family history of overall mental illness trended downward with increased age of onset, however the numbers involved were
MA NU
too small to reach accepted levels of significance. Of these findings, trends in family history of schizophrenia specifically, early psychosocial functioning, educational completion and
ED
some substance use categories were found to be statistically significant on chi-square testing.
On examining risk factors proximal to the onset of psychosis, late onset schizophrenia was
PT
associated with significantly higher rates of unemployment, but otherwise similar rates of social isolation and unstable accommodation to other age groups. Trends in proximal
AC
CE
substance use (within 3 months of onset) were similar to that of lifetime substance use.
4. Discussion
4.1 Age and Gender The peak age of onset (17-22 years), and relative proportions of different age-onset patients were similar to those seen in past age-based studies on schizophrenia (2, 3). The distinct early life male predominance, and slight female preponderance in the late onset subgroup also parallels past research on gender differences in schizophrenia, and supports the role of female gender as a potential risk factor for schizophrenia in later life (23) (6, 24).
6
ACCEPTED MANUSCRIPT
4.2 Family History
T
A graded reduction in family history of both schizophrenia and non-specific psychiatric
RI P
illness was seen with increased age, though statistically significant only for schizophrenia specifically. High rates of both specific and non-specific psychiatric family history in youth
SC
onset patients is consistent with past family studies (8, 9) and suggests a polygenic contribution to youth onset schizophrenia with schizophrenia-specific factors. Lower rates of
MA NU
family history seen in older onset groups indicate an overall lower genetic burden of disease compared to youth onset patients, a finding also consistent with past research (18) (19). Indeed, Howard et al’s (18) study of vLOS patients suggests that late onset disease may have
ED
no genetic association with schizophrenia specifically, but is instead associated with depressive family history. The trends in this study are unclear in relation to non-specific
PT
family history, but support the theory that genetic factors have a relatively smaller role to
CE
play in middle and late onset schizophrenia.
AC
4.3 Early developmental risk factors Significantly better early psychosocial functioning was noted in both the middle and late onset subgroups compared to the youth onset subgroup. Poor early psychosocial functioning is a recognised antecedent factor for schizophrenia, suggestive of underlying pathology present during the premorbid stages of disease development (25). These findings suggest the presence of less early life pathology in late onset patients, and is consistent with past research showing better premorbid functioning in LOS patients compared to those with earlier onset disease (3, 21, 26). Whether early psychosocial functioning in the late onset cohort is the same as that of the general population or there is some degree of impairment is unclear, but it is evident that the focus of pathology in late onset disease occurs in later life.
7
ACCEPTED MANUSCRIPT
The middle onset age group was found to have significantly higher proportions of first
T
generation migrants. Higher rates were also seen in late onset age group, however the
RI P
numbers involved were too small to reach accepted levels of significance. Bourque et al (12, 27) found migration to be a distinct risk factor for schizophrenia overall, with greatest risk
SC
among first-generation migrants and those that migrated during early childhood. This association has also been found, specifically in relation to late onset disease, in a London
MA NU
study of vLOS patients (20), suggesting an important role for migration in later life schizophrenia. While the relationship between migration and psychosis is complex, the increased risk is thought to be mediated by social factors including social adversity,
ED
marginalisation and discrimination associated with migratory status (12, 28). Older susceptible individuals may have a poorer ability to adapt to these and other migration-related
PT
changes, or suffer the effect of cumulative exposure to such stresses over time.
CE
4.4 Psychiatric and physical health factors
AC
Similar rates of psychiatric comorbidity were seen in all 3 subgroups. This captured any formal mental health diagnosis received prior to the index psychotic episode, excluding substance use disorder. Reported rates fell within the expected range found in past psychosis studies (between 23-60% lifetime prevalence of a psychiatric comorbidity) and are higher than would be expected of normal population (16, 29). These findings indicate that a background history of any psychiatric morbidity is a risk factor in both early and late onset schizophrenia. While the relationship between schizophrenia and specific psychiatric conditions is beyond the scope of this study, these findings are suggestive of shared pathophysiology or vulnerabilities between schizophrenia and other psychiatric conditions.
8
ACCEPTED MANUSCRIPT In regard to physical health, high rates of health problems were reported in all onset age categories, consistent with past research showing above average rates of comorbid health
T
problems in schizophrenia patients (30). Late onset patients had a statistically significant
RI P
greater proportion (42.5%) of active, pre-existing medical conditions compared to the youth onset group (22.1%), indicating a general state of poorer health among older onset patients.
SC
Rates were similar to that seen in past LOS studies (31). This was not unexpected as older patients, by virtue of their age, have a greater likelihood of physical illness. However, this
MA NU
does not discount medical illness as a risk factor for LOS, which may contribute to the manifestation of psychosis through increased stress, social isolation and physiological change. Studies suggest that shared familial factors may contribute to both schizophrenia risk
ED
and health conditions like type 2 diabetes (32), and have shown that the number of medical comorbidities independently contribute to the severity of psychosis experienced in
PT
schizophrenia patients (33). It is conceivable that an increased burden of comorbid physical illness may contribute to the manifestation of an underlying vulnerability to psychosis,
CE
through both reduced reserve and potentially shared familial factors. In either case, it is
AC
evident that these non-specific physical and hereditary factors play a larger role in LOS compared to EOS.
4.5 Substance Use This study looked specifically at cannabis and amphetamine use, though many other substances are also implicated in schizophrenia risk (16). Our subject cohort displayed markedly higher rates of both cannabis and amphetamine use compared to Australian national average; 55.6% and 36.9% for lifetime cannabis and amphetamine use respectively compared to national average of 34.8% and 7.0% (34). The youth onset subgroup displayed the highest rates of drug use in all categories, and a graded reduction in rates of use was seen with
9
ACCEPTED MANUSCRIPT increased age of onset. Robust studies have found both substances, particularly cannabis, to be implicated in the development of psychosis and schizophrenia (14, 15). Our findings
T
support this association, but suggest that substance use may play a relatively small role in late
RI P
onset disease, where rates of use close approach the Australian national average (34). While this may suggest lesser contribution of substance use to late onset disease, lower rates of
MA NU
4.6 Proximal Psychosocial Factors
SC
substance use may also reflect generational age-related and cultural practises in drug use.
High rates of unemployment, social isolation and unstable accommodation were seen across all subgroups prior to onset of their first psychotic episode. This suggests that these factors
ED
are common to schizophrenia onset at all ages, and is concordant with past studies on sociooccupational functioning in schizophrenia patients (35). Of the proximal risk factors, only
PT
unemployment was found at statistically significant higher rates in the late onset subgroup. Past studies have highlighted social isolation in association with late onset psychosis (19), but
CE
other factors like proximal unemployment and unstable housing remain little understood.
AC
These factors constitute a common presenting picture of socio-occupational decline in psychosis patients (36, 37), but it is conceivable that they may also contribute towards psychosis progression and precipitation. Proximal factors may be key stressors amid other contributing factors, tipping vulnerable individuals over the threshold to psychosis.
4.7 Overview Existing risk factor research on EOS cohorts support a predominantly neurodevelopmental model of schizophrenia pathology, emphasizing early life risk factors such as positive family history, maternal factors and childhood trauma (38, 39). This is highlighted in the two-hit model of schizophrenia aetiology, which focuses on genetic vulnerability as the vital ‘first
10
ACCEPTED MANUSCRIPT hit’ in the pathological pathway, while other risk factors (biological, environmental, psychological, and other) provide the subsequent ‘second-hit’ in the development and
T
manifestation of symptomatic schizophrenia (40). These models view schizophrenia chiefly
RI P
as a disorder of brain development and focus on risk factors that disrupt this process (5, 38,
SC
39).
Later research with increased focus on older onset groups highlight the additional role of later
MA NU
life neuropathology and neurodegenerative factors in schizophrenia, in which later life insults and disease-related pathology contribute to onset and progression of disease (5, 41-44). Our findings were consistent with this view, showing that middle and late onset schizophrenia
ED
involve a different interplay of risk factors, with relatively less early life morbidity and emphasis on heredity, and greater emphasis on later life stressors, such as comorbid physical
PT
and psychiatric illness, migration and loss of employment. These factors are thought to mediate pathology through increased social stress, which has been widely proposed to have a
CE
causal role in the development of schizophrenia (28, 45). Previous neuroimaging studies have
AC
demonstrated the ability of stress to produce altered neural processing and enduring grey matter change (46-48). This is captured in the ‘three-hit hypothesis’ and progressive neurodevelopmental model of schizophrenia aetiology, which emphasize the role neurodegenerative factors which may be induced or accelerated by later life insults and disease onset itself (41-43). The cumulative effects of later life stressors, age- and diseaserelated neurodegeneration, may in combination be significant enough to tip vulnerable older individuals over the ‘threshold’ to psychosis (44).
11
ACCEPTED MANUSCRIPT While imprecise, these increasingly integrative models highlight the heterogeneous and multifactorial nature of schizophrenia, and better represent the broad spectrum of this
RI P
T
disorder.
4.8 Study strengths and limitations
SC
This study was limited by the lack of a normal control sample. Data was collected retrospectively from existing medical records and, there are potential gaps in the information
MA NU
in the history presented in the records. To enhance the consistency of data collected, simple binary measures were used for most parameters, and inter-rater reliability was assessed to
ED
ensure that data collection was concordant between different reviewers.
This study has several benefits in utilizing an early psychosis program population. Most
PT
current service models for early psychosis intervention in Australia and internationally are restricted to the youth cohort. Our sample captures all cases of schizophrenia with onset
CE
between 16 and 65 that present to the public mental health system in the St Vincent’s mental
AC
health catchment area. Early psychosis studies are more likely to capture cases that would be later lost to follow-up, including cases of milder illness and individuals who make a full recovery after a single psychotic episode, allowing a better approximation of the true population incidence.
8. Conclusion Schizophrenia onset at different ages appear to be associated with a different balance of risk factors. The findings in this study are in line with existing literature in regards to the predominant youth-onset schizophrenia, supporting the association with risk factors such as
12
ACCEPTED MANUSCRIPT substance use, positive family history and male gender. However, the risk factor profiles identified support a relatively lesser contribution of hereditary factors in middle and late
T
onset schizophrenia, and suggest that later life changes and stressors such as medical
RI P
problems and unemployment may play a more significant role in disease progression in these groups. Further research focusing on identified areas may help to elucidate underlying
SC
mechanisms and facilitate tailoring of clinical management and service development to the
MA NU
needs of different age groups.
Acknowledgments
This study was conducted in the St Vincent’s Mental Health Service with support of
PT
Ethical Statement
ED
dedicated staff and senior clinicians at the Hawthorn CMHS and Clarendon CMHS.
CE
Permission by the St Vincent’s Hospital (Melbourne) Human Research Ethics Committee-A
AC
was granted to conduct the study.
Disclosure of Interest The authors declare that they have no competing interests.
13
ACCEPTED MANUSCRIPT Appendix A
T
Table A.1. Operational definitions used to assign risk factor presence Risk factor
Operational definition
Age
Age of onset
The age at which the index psychosis episode first
RI P
Risk factor class
occurred. The closest accurate date was sought: From medical records, where the date of onset was
SC
known and recorded, OR
The date of first presentation to a medical service
MA NU
minus the duration of untreated psychosis (DUP), where the DUP was known, OR
Where DUP was unknown, the date of first presentation to a medical service
Family history of any
History
mental illness
PT
ED
Family Psychiatric
Developmental
and 2nd degree relatives only (Relative’s diagnosis had to be clearly stated in the file) Where there is a known diagnosis of Schizophrenia in any
Schizophrenia
1st and 2nd degree relatives only
1st generation migrant
Patient was born in a country other than Australia, and
status
Migrated to Australia prior to the onset of psychosis
Childhood trauma
Any of the following experienced at ≤16 years of age:
AC
Risk Factors
Includes any formal psychiatric diagnoses in any 1 st
Family history of
CE
Early
Childhood abuse (physical, sexual, emotional)
Neglect
Other childhood events cited as being traumatic in the patient’s file
o
Death of a parent or other close relative
o
Parental separation
Severe bullying
Poor early psychosocial
Specific mention of poor social adjustment in childhood
functioning
(<16yoa) or where inferred from case manager reports:
Weak or absent interpersonal connections in childhood
Psychiatric and
Significant sociobehavioural issues in childhood
Educational completion
Patient has completed a tertiary level education
Previous psychiatric
Any formal mental health diagnosis received prior to
14
ACCEPTED MANUSCRIPT Physical Health
conditions
the onset of the index psychotic episode, excluding
Factors
substance use disorder Physical health
Any medical issue(s) still active at the time of psychosis
problems
onset, which were considered by the researchers to
T
represent a potential source of psychological stress. These
RI P
included:
Malignancy
Chronic illnesses – asthma, diabetes mellitus, hepatitis
Substance Use
Lifetime substance use
SC
C, chronic pain, epilepsy
Cardiovascular disease
Lifetime history of any use of Cannabis or
Recent substance use
MA NU
Amphetamines.
Use of Cannabis or Amphetamines during the 3 months preceding the onset of First Episode Psychosis
Proximal
Unemployment
Psychosocial
Patient unemployed at the time of psychosis onset due to:
o
PT
Unstable
ED
Factors
CE
accommodation
AC
Low social
connectedness
Loss of stable employment during the 2 years preceding psychosis onset, OR
o
Background of long-term unemployment
Predominant accommodation status during the 2 years preceding psychosis onset:
Homeless
Itinerant
Living in support housing
Inferred from case manager reports:
Weak or absent connections to family, friends and broader community at the time of psychosis onset
15
ACCEPTED MANUSCRIPT References
AC
CE
PT
ED
MA NU
SC
RI P
T
1. Buchanan RW, Carpenter WT. Domains of psychopathology: an approach to the reduction of heterogeneity in schizophrenia. The Journal Of Nervous And Mental Disease. 1994;182(4):193-204. 2. Harris MJ, Jeste DV. Late-Onset Schizophrenia: An Overview. Schizophrenia Bulletin. 1988;14(1):39-55. 3. Howard R, Rabins PV, Seeman MV, Jeste DV. Late-onset schizophrenia and very-late-onset schizophrenia-like psychosis: an international consensus. The International Late-Onset Schizophrenia Group. The American journal of psychiatry. 2000;157(2):172-8. 4. Gaebel W, Zielasek J. Integrative etiopathogenetic models of psychotic disorders: methods, evidence and concepts. Schizophrenia Bulletin. 2011;37 Suppl 2:S5-S12. 5. Gupta S, Kulhara P. What is schizophrenia: A neurodevelopmental or neurodegenerative disorder or a combination of both? A critical analysis. Indian Journal of Psychiatry. 2010;52(1):21-7. 6. Susana O, Judith U, Jesús C, Xavier L, Jayashri K. Gender Differences in Schizophrenia and First-Episode Psychosis: A Comprehensive Literature Review. Schizophrenia Research and Treatment. 2012:1-9. 7. Schultz SH, North SW, Shields CG. Schizophrenia: a review. American family physician. 2007;75(12):1821-9. 8. Maier W, Lichtermann D, Franke P, Heun R, Falkai P, Rietschel M. The dichotomy of schizophrenia and affective disorders in extended pedigrees. Schizophrenia research. 2002;57(2– 3):259-66. 9. Faridi K, Pawliuk N, King S, Joober R, Malla AK. Prevalence of psychotic and non-psychotic disorders in relatives of patients with a first episode psychosis. Schizophrenia research. 2009;114(13):57-63. 10. van Os J, Kenis G, Rutten BPF. The environment and schizophrenia. Nature. 2010(7321):203. 11. Bendall S, Jackson HJ, Hulbert CA, McGorry PD. Childhood trauma and psychotic disorders: A systematic, critical review of the evidence. Schizophrenia Bulletin. 2008;34:568-79. 12. Bourque F, van der Ven E, Malla A. A meta-analysis of the risk for psychotic disorders among first- and second-generation immigrants. Psychological Medicine. 2011;41(5):897-910. 13. Orlovska S, Pedersen MS, Benros ME, Mortensen PB, Agerbo E, Nordentoft M. Head injury as risk factor for psychiatric disorders: a nationwide register-based follow-up study of 113,906 persons with head injury. The American journal of psychiatry. 2014;171(4):463-9. 14. Moore THM, Zammit S, Lingford-Hughes A, Barnes TRE, Jones PB, Burke M, et al. Articles: Cannabis use and risk of psychotic or affective mental health outcomes: a systematic review. The Lancet. 2007;370:319-28. 15. Barkus E, Murray RM. Substance Use in Adolescence and Psychosis: Clarifying the Relationship. Annual review of clinical psychology. 2010;6:365-90. 16. Buckley PF, Miller BJ, Lehrer DS, Castle DJ. Psychiatric Comorbidities and Schizophrenia. Schizophrenia Bulletin. 2009;35(2):383-402. 17. Achim AM, Maziade M, Raymond É, Olivier D, Mérette C, Roy M-A. How Prevalent Are Anxiety Disorders in Schizophrenia? A Meta-Analysis and Critical Review on a Significant Association. Schizophrenia Bulletin. 2011;37(4):811-21. 18. Howard RJ, Graham C, Sham P, Dennehey J, Castle DJ, Levy R, et al. A controlled family study of late-onset non-affective psychosis (late paraphrenia). The British Journal of Psychiatry. 1997;170:511-4. 19. Wynn Owen PA, Castle DJ. Late-Onset Schizophrenia: Epidemiology, Diagnosis, Management and Outcomes. Drugs & Aging. 1999;15(2):81-9.
16
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
MA NU
SC
RI P
T
20. Reeves SJ, Sauer J, Stewart R, Granger A, Howard RJ. Increased first-contact rates for verylate-onset schizophrenia-like psychosis in African- and Caribbean-born elders The British Journal of Psychiatry. 2001(179 (2) ):172-4. 21. Golay P, Alameda L, Mebdouhi N, Baumann P, Ferrari C, Solida A, et al. Age at the time of onset of psychosis: A marker of specific needs rather than a determinant of outcome? European Psychiatry. 2017;45:20-6. 22. Selvendra A, Baetens D, Trauer T, Petrakis M, Castle D. First episode psychosis in an adult area mental health service-A closer look at early and late-Onset first episode psychosis Australasian Psychiatry, 2014; 22(3):235-41. 23. Castle D, Sham P, Murray R. Differences in distribution of ages of onset in males and females with schizophrenia. Schizophrenia research. 1998;33(3):179-83. 24. Thorup A, Waltoft BL, Pedersen CB, Mortensen PB, Nordentoft M. Young males have a higher risk of developing schizophrenia: a Danish register study. Psychological Medicine. 2007;37(4):479-84. 25. Welham J, Isohanni M, Jones P, McGrath J. The Antecedents of Schizophrenia: A Review of Birth Cohort Studies. Schizophrenia Bulletin. 2009;35(3):603-23. 26. Jeste DV, Harris MJ, Krull A, Kuck J, McAdams LA, Heaton R. Clinical and neuropsychological characteristics of patients with late-onset schizophrenia. The American journal of psychiatry. 1995;152(5):722-30. 27. Bourque F, van der Ven E, Fusar-Poli P, Malla A. Immigration, social environment and onset of psychotic disorders. Current Pharmaceutical Design. 2012;18(4):518-26. 28. Selten J-P, van der Ven E, Rutten BPF, Cantor-Graae E. The Social Defeat Hypothesis of Schizophrenia: An Update. Schizophrenia Bulletin. 2013;39(6):1180-6. 29. Cassano GB, Pini S, Mastrocinque T, Saettoni M, Papasogli A, Dell'Osso L. II. Phenomenology: Prevalence and clinical correlates of psychiatric comorbidity in patients with psychotic disorders. Schizophrenia research. 1998;29:33. 30. Carney CP, Jones L, Woolson RF. Medical Comorbidity in Women and Men with Schizophrenia: A Population-Based Controlled Study. Journal of General Internal Medicine. 2006(11):1133. 31. Greenfield P, Joshi S, Christian S, Lekkos P, Gregorowicz A, Fisher HL, et al. First episode psychosis in the over 35 s: is there a role for early intervention? Early Intervention In Psychiatry doi: 101111/eip12322 [Epub ahead of print]. 2016. 32. Foley DL, Mackinnon A, Morgan VA, Watts GF, Castle DJ, Waterreus A, et al. Common familial risk factors for schizophrenia and diabetes mellitus. Australian & New Zealand Journal of Psychiatry. 2016;50(5):488-94. 33. Dixon L, Postrado L, Delahanty J, Fischer PJ, Lehman A. The Association of Medical Comorbidity in Schizophrenia with Poor Physical and Mental Health. The Journal of Nervous and Mental Disease. 1999;187(8):496-502. 34. National Drug Strategy Household Survey detailed report: 2013. Canberra: Australian Institute of Health and Welfare; 2014 Nov 25. 35. Melle I, Friis S, Hauff E, Vaglum P. Social functioning of patients with schizophrenia in highincome welfare societies. Psychiatric Services. 2000;51(2):223-8. 36. Vargas G, Strassnig M, Sabbag S, Gould F, Durand D, Stone L, et al. The course of vocational functioning in patients with schizophrenia: Re-examining social drift. Schizophrenia Research: Cognition. 2014;1(1):41-6. 37. Cullen K, Guimaraes A, Wozniak J, Anjum A, Schulz SC, White T. Trajectories of social withdrawal and cognitive decline in the schizophrenia prodrome. Clinical Schizophrenia & Related Psychoses. 2011;4(4):229-38. 38. Rapoport JL, Giedd JN, Gogtay N. Neurodevelopmental model of schizophrenia: update 2012. Molecular Psychiatry. 2012;17(12):1228-38.
17
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
MA NU
SC
RI P
T
39. Rapoport JL, Addington A, Frangou S. The neurodevelopmental model of schizophrenia: What can very early onset cases tell us? Current Psychiatry Reports. 2005;7(2):81-2. 40. Bayer TA, Falkai P, Maier W. Genetic and non-genetic vulnerability factors in schizophrenia: the basis of the "Two hit hypothesis". Journal Of Psychiatric Research. 1999;33(6):543-8. 41. Keshavan MS. Development, disease and degeneration in schizophrenia: a unitary pathophysiological model. Journal of Psychiatric Research. 1999;33(6):513-21. 42. Wyatt RJ. Neuroleptics and the natural course of schizophrenia. Schizophr Bull. 1991;17(2):325-51. 43. Aricioglu F, Ozkartal CS, Unal G, Dursun S, Cetin M, Müller N. Neuroinflammation in Schizophrenia: A Critical Review and The Future. Klinik Psikofarmakoloji Bulteni. 2016;26(4):429-37. 44. Castle DJ, Howard R. What do we know about the aetiology of late-onset schizophrenia? European Psychiatry. 1992;7(3):99-108. 45. Mondelli V, Dazzan P, Hepgul N, Di Forti M, Aas M, D'Albenzio A, et al. Abnormal cortisol levels during the day and cortisol awakening response in first-episode psychosis: the role of stress and of antipsychotic treatment. Schizophrenia research. 2010;116(2):234-42. 46. Akdeniz C, Schäfer A, Streit F, Haller L, Wüst S, Kirsch P, et al. Sex-Dependent Association of Perigenual Anterior Cingulate Cortex Volume and Migration Background, an Environmental Risk Factor for Schizophrenia. Schizophrenia Bulletin. 2017;43(4):925-34. 47. Meyer-Lindenberg A, Tost H. Neural mechanisms of social risk for psychiatric disorders. Nature neuroscience. 2012;15(5):663-8. 48. Chen Y, Baram TZ. Toward Understanding How Early-Life Stress Reprograms Cognitive and Emotional Brain Networks. Neuropsychopharmacology. 2016;41(1):197-206.
18
ACCEPTED MANUSCRIPT
RI P
T
Figure
Schizophrenia frequency by gender and age (5yr intervals)
SC
45
40
MA NU
35
Frequency
30
25 20
15
5
0 5
10
15
20
25
30
35
40
45
Age of onset
Female
Male
CE
PT
0
ED
10
AC
Figure 1. Age of onset, based on gender
19
50
55
60
65
70
75
ACCEPTED MANUSCRIPT Table 1
Proportion of
Youth onset
Middle onset
Late onset
total sample
<26 years
26-40 years
>40 years
100%
n=104
n=81
n=40
64.9%
72.1%
40.0%
47.1%
RI P
Diagnosis of Schizophrenia
67.9% (P=0.534)
40.0% (P<0.01)
35.8% (P=0.122)
30% (P=0.063)
24.0%
4.9% (P<0.01)
2.5% (P<0.01)
28.9%
44.4% (P=0.028)
42.5%
illness Family history of
13.3%
Schizophrenia Early Developmental Risk Factors 1st generation migrant status
ED
36.9%
Childhood trauma
42.2%
functioning
Tertiary completed)
38.3% (P=0.228)
37.5% (P=0.298)
27.9%
9.9% (P<0.01)
5.0% (P<0.01)
20.0%
8.7%
28.4% (P<0.01)
32.5% (P<0.01)
31.6%
34.6%
28.4% (P=0.519)
30.0% (P=0.2)
28.9%
22.1%
30.9% (P=0.178)
42.5%
CE
Educational completion (% of
47.1%
(P=0.839)
17.3%
PT
Poor early psychosocial
MA NU
Family Psychiatric History Family history of any mental
SC
(n=225) Gender (% of males)
T
Table 1. Comparison of risk factor frequencies in different age onset groups
Psychiatric and Physical Health Factors
conditions
AC
Previous psychiatric
Physical health problems
(P=0.014) Substance Use Lifetime Cannabis use
55.6%
65.4%
53.1% (P=0.09)
35.0% (P<0.01)
Recent Cannabis use (within 3
34.2%
46.2%
28.4% (P=0.014)
15.0% (P<0.01)
36.9%
41.4%
39.5% (P=0.8)
20.0%
months preceding psychosis onset) Lifetime Amphetamine use
(P=0.016) Recent amphetamine use
22.7%
27.9%
(within 3 months preceding psychosis onset) Proximal Psychosocial Factors
20
23.5% (P=0.495)
7.5% (P<0.01)
ACCEPTED MANUSCRIPT Low social connectedness
58.7%
59.6%
54.3% (P=0.555)
72.5% (P=0.124)
Unstable accommodation
26.9%
28.4%
27.2% (P=0.971)
35.0% (P=0.340)
53.9%
60.5% (P=0.365)
RI P
60.0%
T
Unemployment
75.0% (P=0.016)
Note: P-values generated represent the comparisons with the under 26 age onset group. Comparisons with
AC
CE
PT
ED
MA NU
SC
P<0.05 are denoted in red.
21
ACCEPTED MANUSCRIPT
T
Highlights
Schizophrenia onset at different ages is associated with differing risk factors.
Male gender and heredity are significant risk factors in early onset schizophrenia.
Female gender and health problems are more common in older onset schizophrenia.
Recent and lifetime drug use are more common in early onset schizophrenia.
Graded trends are observed with age in several risk factors.
AC
CE
PT
ED
MA NU
SC
RI P
22
S0010-440X(17)30221-3 doi: 10.1016/j.comppsych.2017.09.009 YCOMP 51896
To appear in:
Comprehensive Psychiatry
Please cite this article as: Chen Laura, Selvendra Ajit, Stewart Anne, Castle David, Risk Factors in Early and Late Onset Schizophrenia, Comprehensive Psychiatry (2017), doi: 10.1016/j.comppsych.2017.09.009
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Title:
RI P
T
Risk Factors in Early and Late Onset Schizophrenia
Authors:
Melbourne Medical School, The University of Melbourne, Parkville, Victoria, Australia
b
MA NU
a
SC
Laura Chena, Ajit Selvendraa,b, Anne Stewarta, David Castlea,b
Mental Health Service, St. Vincent’s Hospital, Fitzroy, Victoria, Australia
CE
PT
Dr. Ajit Selvendra Consultant Psychiatrist St. Vincent’s Mental Health, Hawthorn CMHS, 642 Burwood Rd, Hawthorn East, 3123 Victoria, Australia
ED
Corresponding Author:
Phone: +613-9231-5900
AC
Email: [email protected]
ACCEPTED MANUSCRIPT
T
Abstract
RI P
Background: This study looks at key risk factors in patients with schizophrenia to identify trends according to age of onset, comparing presentations prior to 26 years (youth onset), between 26 and 40 years (middle onset), and after 40 years of age (late onset).
SC
Methods: The early psychosis program at St Vincent’s Hospital Melbourne treats patients presenting in the early stages of psychosis between 16 and 65 years of age. A database was
MA NU
developed to capture key risk factors in all patients with an eventual diagnosis of schizophrenia (n=225). Risk factor profiles were then generated and compared for patients based on age of onset.
Results: Older age of onset was associated with weaker family history of schizophrenia, lower rates of substance use, better early psychosocial functioning and higher educational
ED
achievement. Female preponderance and comorbid physical health problems were particularly notable in the late onset cohort. Later life schizophrenia also showed a relatively
PT
greater association with psychosocial factors proximal to psychosis onset, such as unemployment.
CE
Discussion: Clear trends are noticeable with age. Older patients have characteristic differences in their background risk factors compared to youth onset patients, including less hereditary influence and relatively more emphasis on later life risk factors. Identifying the
AC
roles of specific risk factors in these distinct age-onset groups can enhance our understanding of underlying aetiology and facilitate service development to meet the needs of each specific age group.
Keywords Schizophrenia Late onset schizophrenia Risk factors in schizophrenia Early psychosis
1
ACCEPTED MANUSCRIPT
T
1. Introduction
RI P
Schizophrenia is a heterogeneous condition, capturing psychoses presentations of variable symptom profiles and severity (1). While the majority of cases occur before middle age, a
SC
significant portion of new onset cases occur in the later decades of life (2). Presentations of
MA NU
new-onset schizophrenia at different life stages can vary considerably, and has led to international consensus in classifying schizophrenia with onset prior to 40 years of age(yoa) as Early-Onset Schizophrenia (EOS1), those with onset at and after 40yoa as Late-Onset Schizophrenia (LOS2), and those with onset after 60yoa as Very-Late-Onset Schizophrenia-
ED
like psychosis (vLOS3) (3). In countries like Australia, there is an additional focus on specifically youth-onset schizophrenia, with the cut-off age for most early-psychosis
PT
programmes being 26 years of age. This age-based approach may facilitate a better
CE
understanding and management of schizophrenia at different life-stages.
AC
The aetiopathology of schizophrenia remains poorly understood, with many potential models being proposed (4, 5). Research has focused on identification of associated risk factors to elucidate the underlying mechanisms of the disorder, and has helped yield greater understanding of disease development and progression. Due to the youth preponderance in schizophrenia and tendency of past research to look at schizophrenia patients as a homogenous group, there is a much poorer understanding of age-related differences in schizophrenia, and the specific features and risk factors relevant to late onset cases.
1
EOS – Early Onset Schizophrenia. Schizophrenia with onset prior to 40 years of age LOS – Late Onset Schizophrenia. Schizophrenia with onset after 40 years of age 3 vLOS – Very Late Onset Schizophrenia-like psychosis. Schizophrenia with onset after 60 years of age 2
2
ACCEPTED MANUSCRIPT Early onset schizophrenia presents most commonly during early adulthood (late teens to early twenties) and is associated with predominant male gender, a positive family history of
T
schizophrenia and other psychiatric conditions, particularly affective disorders (2, 6-9). Early
RI P
life socioenvironmental risk factors, particularly if present during key developmental periods (early childhood and adolescence), are also implicated. These include factors such as urban
SC
rearing environment, childhood trauma, migration and minority status, traumatic brain injury and substance use (10-15). Comorbid psychiatric illness, particularly depression, anxiety and
MA NU
substance use disorders are frequently associated with and precede psychosis onset (16, 17).
Comparatively few studies have looked specifically at risk factors in late onset schizophrenia.
ED
Existing literature suggest possible differences in the background risk factors of older individuals including greater female preponderance, associated family history of affective
PT
disorders, and cluster A personality traits (6, 18, 19). A few focal studies also suggest an important role for socioenvironmental factors such as migration and trauma exposure (20,
CE
21). Later life stressors and the notion of accumulated loss appear to be important in the
AC
precipitation of psychosis in older patients (3), but the specific influence of many other schizophrenia-associated risk factors remain unclear.
1.1 Aim
This study aims to review the prevalence of risk factors among patients treated for schizophrenia in the St Vincent’s Hospital early psychosis program between mid-2006 to mid-2015, and to compare these among different age-onset groups.
3
ACCEPTED MANUSCRIPT 2. Methods St. Vincent’s Hospital covers inner city and inner urban municipalities in Melbourne,
T
Australia, and provides comprehensive care for psychosis. The Early Psychosis program
RI P
captures all people within the St Vincent’s catchment area aged 16-65 presenting with their
MA NU
months of treatment for a first episode of psychosis.
SC
first episode of untreated psychosis (of any type), or those who are within their first 18
A file review of potential risk factors for psychosis was performed using patient medical records. Subjects included all patients from the St Vincent’s Mental Health Service Early Psychosis Program with an eventual diagnosis of schizophrenia (n=225). Diagnoses of
ED
schizophrenia were made using a DSM IV format at 3 months after all treating clinicians, including the regular treating psychiatrist, had seen the case and come to a consensus. If there
PT
was any lack of diagnostic clarity, the patient’s history was reviewed by and discussed with
CE
the early psychosis consultant to clarify the final diagnosis.
AC
A range of literature-based risk factors for schizophrenia were examined and a subset of the better captured and clinically relevant factors were further analysed for trends and associations. These factors are defined in Table A.1 of Appendix A. Trends were then reviewed in respect to 3 subgroups, defined by onset before 26 (youth onset), between 26 and 40 (middle onset), and after 40 years of age (late onset). Age cut-offs were based on current Australian youth-psychosis program definitions (22) (under 26 versus 26 and older), and international consensus (3) (40 and under versus over 40). Results were graphed to identify patterns related to age and gender (Figure 1), and chi-square testing conducted to assess significance of trends and differences between the youth onset subgroup and each older onset subgroup (Table 1). Risk factor definitions were strictly operationalised to ensure accuracy of
4
ACCEPTED MANUSCRIPT identification, and inter-rater reliability testing conducted by comparing file reviews of the same patients conducted independently by different reviewers. The results compared
SC
RI P
T
demonstrated good inter-rater consistency in risk factors identified.
MA NU
3. Results 3.1 Figures & Tables
ED
Figure 1. Age of onset, based on gender
CE
PT
Table 1. Comparison of risk factor frequencies in different age onset groups
225 cases of diagnosed schizophrenia from the St Vincent’s early psychosis program were
AC
identified and included in the study. This included 104 (46%) cases with youth onset, 81 (36%) cases with middle onset, and 40 (18%) cases with late onset. Of the late onset cohort, 2 individuals had disease onset after 60yoa, and for the purposes of this study were included within the group of ‘late onset schizophrenia’. The peak age of onset in the study sample was between 17 and 22 years, with median age of 27. A similar peak in early life was observed in both sexes though significantly higher amongst males (Figure 1). Male and female prevalence rates approximated each other in later life, with a slight female preponderance.
A marked reduction in the rate of positive family history of schizophrenia was seen with increased age of onset (Table 1). Statistically significant age-associated trends were also seen 5
ACCEPTED MANUSCRIPT in early psychosocial impairment (lesser with increased age of onset), levels of educational completion (higher with increased age of onset), and rates of both recent and lifetime
T
substance use (statistically significant and lower in the late onset cohort). An increased
RI P
frequency of unemployment and comorbid physical health problems was identified in the late onset group specifically. Similar rates of reported childhood trauma, and previous psychiatric
SC
conditions were found in the different age groups. Rates of family history of overall mental illness trended downward with increased age of onset, however the numbers involved were
MA NU
too small to reach accepted levels of significance. Of these findings, trends in family history of schizophrenia specifically, early psychosocial functioning, educational completion and
ED
some substance use categories were found to be statistically significant on chi-square testing.
On examining risk factors proximal to the onset of psychosis, late onset schizophrenia was
PT
associated with significantly higher rates of unemployment, but otherwise similar rates of social isolation and unstable accommodation to other age groups. Trends in proximal
AC
CE
substance use (within 3 months of onset) were similar to that of lifetime substance use.
4. Discussion
4.1 Age and Gender The peak age of onset (17-22 years), and relative proportions of different age-onset patients were similar to those seen in past age-based studies on schizophrenia (2, 3). The distinct early life male predominance, and slight female preponderance in the late onset subgroup also parallels past research on gender differences in schizophrenia, and supports the role of female gender as a potential risk factor for schizophrenia in later life (23) (6, 24).
6
ACCEPTED MANUSCRIPT
4.2 Family History
T
A graded reduction in family history of both schizophrenia and non-specific psychiatric
RI P
illness was seen with increased age, though statistically significant only for schizophrenia specifically. High rates of both specific and non-specific psychiatric family history in youth
SC
onset patients is consistent with past family studies (8, 9) and suggests a polygenic contribution to youth onset schizophrenia with schizophrenia-specific factors. Lower rates of
MA NU
family history seen in older onset groups indicate an overall lower genetic burden of disease compared to youth onset patients, a finding also consistent with past research (18) (19). Indeed, Howard et al’s (18) study of vLOS patients suggests that late onset disease may have
ED
no genetic association with schizophrenia specifically, but is instead associated with depressive family history. The trends in this study are unclear in relation to non-specific
PT
family history, but support the theory that genetic factors have a relatively smaller role to
CE
play in middle and late onset schizophrenia.
AC
4.3 Early developmental risk factors Significantly better early psychosocial functioning was noted in both the middle and late onset subgroups compared to the youth onset subgroup. Poor early psychosocial functioning is a recognised antecedent factor for schizophrenia, suggestive of underlying pathology present during the premorbid stages of disease development (25). These findings suggest the presence of less early life pathology in late onset patients, and is consistent with past research showing better premorbid functioning in LOS patients compared to those with earlier onset disease (3, 21, 26). Whether early psychosocial functioning in the late onset cohort is the same as that of the general population or there is some degree of impairment is unclear, but it is evident that the focus of pathology in late onset disease occurs in later life.
7
ACCEPTED MANUSCRIPT
The middle onset age group was found to have significantly higher proportions of first
T
generation migrants. Higher rates were also seen in late onset age group, however the
RI P
numbers involved were too small to reach accepted levels of significance. Bourque et al (12, 27) found migration to be a distinct risk factor for schizophrenia overall, with greatest risk
SC
among first-generation migrants and those that migrated during early childhood. This association has also been found, specifically in relation to late onset disease, in a London
MA NU
study of vLOS patients (20), suggesting an important role for migration in later life schizophrenia. While the relationship between migration and psychosis is complex, the increased risk is thought to be mediated by social factors including social adversity,
ED
marginalisation and discrimination associated with migratory status (12, 28). Older susceptible individuals may have a poorer ability to adapt to these and other migration-related
PT
changes, or suffer the effect of cumulative exposure to such stresses over time.
CE
4.4 Psychiatric and physical health factors
AC
Similar rates of psychiatric comorbidity were seen in all 3 subgroups. This captured any formal mental health diagnosis received prior to the index psychotic episode, excluding substance use disorder. Reported rates fell within the expected range found in past psychosis studies (between 23-60% lifetime prevalence of a psychiatric comorbidity) and are higher than would be expected of normal population (16, 29). These findings indicate that a background history of any psychiatric morbidity is a risk factor in both early and late onset schizophrenia. While the relationship between schizophrenia and specific psychiatric conditions is beyond the scope of this study, these findings are suggestive of shared pathophysiology or vulnerabilities between schizophrenia and other psychiatric conditions.
8
ACCEPTED MANUSCRIPT In regard to physical health, high rates of health problems were reported in all onset age categories, consistent with past research showing above average rates of comorbid health
T
problems in schizophrenia patients (30). Late onset patients had a statistically significant
RI P
greater proportion (42.5%) of active, pre-existing medical conditions compared to the youth onset group (22.1%), indicating a general state of poorer health among older onset patients.
SC
Rates were similar to that seen in past LOS studies (31). This was not unexpected as older patients, by virtue of their age, have a greater likelihood of physical illness. However, this
MA NU
does not discount medical illness as a risk factor for LOS, which may contribute to the manifestation of psychosis through increased stress, social isolation and physiological change. Studies suggest that shared familial factors may contribute to both schizophrenia risk
ED
and health conditions like type 2 diabetes (32), and have shown that the number of medical comorbidities independently contribute to the severity of psychosis experienced in
PT
schizophrenia patients (33). It is conceivable that an increased burden of comorbid physical illness may contribute to the manifestation of an underlying vulnerability to psychosis,
CE
through both reduced reserve and potentially shared familial factors. In either case, it is
AC
evident that these non-specific physical and hereditary factors play a larger role in LOS compared to EOS.
4.5 Substance Use This study looked specifically at cannabis and amphetamine use, though many other substances are also implicated in schizophrenia risk (16). Our subject cohort displayed markedly higher rates of both cannabis and amphetamine use compared to Australian national average; 55.6% and 36.9% for lifetime cannabis and amphetamine use respectively compared to national average of 34.8% and 7.0% (34). The youth onset subgroup displayed the highest rates of drug use in all categories, and a graded reduction in rates of use was seen with
9
ACCEPTED MANUSCRIPT increased age of onset. Robust studies have found both substances, particularly cannabis, to be implicated in the development of psychosis and schizophrenia (14, 15). Our findings
T
support this association, but suggest that substance use may play a relatively small role in late
RI P
onset disease, where rates of use close approach the Australian national average (34). While this may suggest lesser contribution of substance use to late onset disease, lower rates of
MA NU
4.6 Proximal Psychosocial Factors
SC
substance use may also reflect generational age-related and cultural practises in drug use.
High rates of unemployment, social isolation and unstable accommodation were seen across all subgroups prior to onset of their first psychotic episode. This suggests that these factors
ED
are common to schizophrenia onset at all ages, and is concordant with past studies on sociooccupational functioning in schizophrenia patients (35). Of the proximal risk factors, only
PT
unemployment was found at statistically significant higher rates in the late onset subgroup. Past studies have highlighted social isolation in association with late onset psychosis (19), but
CE
other factors like proximal unemployment and unstable housing remain little understood.
AC
These factors constitute a common presenting picture of socio-occupational decline in psychosis patients (36, 37), but it is conceivable that they may also contribute towards psychosis progression and precipitation. Proximal factors may be key stressors amid other contributing factors, tipping vulnerable individuals over the threshold to psychosis.
4.7 Overview Existing risk factor research on EOS cohorts support a predominantly neurodevelopmental model of schizophrenia pathology, emphasizing early life risk factors such as positive family history, maternal factors and childhood trauma (38, 39). This is highlighted in the two-hit model of schizophrenia aetiology, which focuses on genetic vulnerability as the vital ‘first
10
ACCEPTED MANUSCRIPT hit’ in the pathological pathway, while other risk factors (biological, environmental, psychological, and other) provide the subsequent ‘second-hit’ in the development and
T
manifestation of symptomatic schizophrenia (40). These models view schizophrenia chiefly
RI P
as a disorder of brain development and focus on risk factors that disrupt this process (5, 38,
SC
39).
Later research with increased focus on older onset groups highlight the additional role of later
MA NU
life neuropathology and neurodegenerative factors in schizophrenia, in which later life insults and disease-related pathology contribute to onset and progression of disease (5, 41-44). Our findings were consistent with this view, showing that middle and late onset schizophrenia
ED
involve a different interplay of risk factors, with relatively less early life morbidity and emphasis on heredity, and greater emphasis on later life stressors, such as comorbid physical
PT
and psychiatric illness, migration and loss of employment. These factors are thought to mediate pathology through increased social stress, which has been widely proposed to have a
CE
causal role in the development of schizophrenia (28, 45). Previous neuroimaging studies have
AC
demonstrated the ability of stress to produce altered neural processing and enduring grey matter change (46-48). This is captured in the ‘three-hit hypothesis’ and progressive neurodevelopmental model of schizophrenia aetiology, which emphasize the role neurodegenerative factors which may be induced or accelerated by later life insults and disease onset itself (41-43). The cumulative effects of later life stressors, age- and diseaserelated neurodegeneration, may in combination be significant enough to tip vulnerable older individuals over the ‘threshold’ to psychosis (44).
11
ACCEPTED MANUSCRIPT While imprecise, these increasingly integrative models highlight the heterogeneous and multifactorial nature of schizophrenia, and better represent the broad spectrum of this
RI P
T
disorder.
4.8 Study strengths and limitations
SC
This study was limited by the lack of a normal control sample. Data was collected retrospectively from existing medical records and, there are potential gaps in the information
MA NU
in the history presented in the records. To enhance the consistency of data collected, simple binary measures were used for most parameters, and inter-rater reliability was assessed to
ED
ensure that data collection was concordant between different reviewers.
This study has several benefits in utilizing an early psychosis program population. Most
PT
current service models for early psychosis intervention in Australia and internationally are restricted to the youth cohort. Our sample captures all cases of schizophrenia with onset
CE
between 16 and 65 that present to the public mental health system in the St Vincent’s mental
AC
health catchment area. Early psychosis studies are more likely to capture cases that would be later lost to follow-up, including cases of milder illness and individuals who make a full recovery after a single psychotic episode, allowing a better approximation of the true population incidence.
8. Conclusion Schizophrenia onset at different ages appear to be associated with a different balance of risk factors. The findings in this study are in line with existing literature in regards to the predominant youth-onset schizophrenia, supporting the association with risk factors such as
12
ACCEPTED MANUSCRIPT substance use, positive family history and male gender. However, the risk factor profiles identified support a relatively lesser contribution of hereditary factors in middle and late
T
onset schizophrenia, and suggest that later life changes and stressors such as medical
RI P
problems and unemployment may play a more significant role in disease progression in these groups. Further research focusing on identified areas may help to elucidate underlying
SC
mechanisms and facilitate tailoring of clinical management and service development to the
MA NU
needs of different age groups.
Acknowledgments
This study was conducted in the St Vincent’s Mental Health Service with support of
PT
Ethical Statement
ED
dedicated staff and senior clinicians at the Hawthorn CMHS and Clarendon CMHS.
CE
Permission by the St Vincent’s Hospital (Melbourne) Human Research Ethics Committee-A
AC
was granted to conduct the study.
Disclosure of Interest The authors declare that they have no competing interests.
13
ACCEPTED MANUSCRIPT Appendix A
T
Table A.1. Operational definitions used to assign risk factor presence Risk factor
Operational definition
Age
Age of onset
The age at which the index psychosis episode first
RI P
Risk factor class
occurred. The closest accurate date was sought: From medical records, where the date of onset was
SC
known and recorded, OR
The date of first presentation to a medical service
MA NU
minus the duration of untreated psychosis (DUP), where the DUP was known, OR
Where DUP was unknown, the date of first presentation to a medical service
Family history of any
History
mental illness
PT
ED
Family Psychiatric
Developmental
and 2nd degree relatives only (Relative’s diagnosis had to be clearly stated in the file) Where there is a known diagnosis of Schizophrenia in any
Schizophrenia
1st and 2nd degree relatives only
1st generation migrant
Patient was born in a country other than Australia, and
status
Migrated to Australia prior to the onset of psychosis
Childhood trauma
Any of the following experienced at ≤16 years of age:
AC
Risk Factors
Includes any formal psychiatric diagnoses in any 1 st
Family history of
CE
Early
Childhood abuse (physical, sexual, emotional)
Neglect
Other childhood events cited as being traumatic in the patient’s file
o
Death of a parent or other close relative
o
Parental separation
Severe bullying
Poor early psychosocial
Specific mention of poor social adjustment in childhood
functioning
(<16yoa) or where inferred from case manager reports:
Weak or absent interpersonal connections in childhood
Psychiatric and
Significant sociobehavioural issues in childhood
Educational completion
Patient has completed a tertiary level education
Previous psychiatric
Any formal mental health diagnosis received prior to
14
ACCEPTED MANUSCRIPT Physical Health
conditions
the onset of the index psychotic episode, excluding
Factors
substance use disorder Physical health
Any medical issue(s) still active at the time of psychosis
problems
onset, which were considered by the researchers to
T
represent a potential source of psychological stress. These
RI P
included:
Malignancy
Chronic illnesses – asthma, diabetes mellitus, hepatitis
Substance Use
Lifetime substance use
SC
C, chronic pain, epilepsy
Cardiovascular disease
Lifetime history of any use of Cannabis or
Recent substance use
MA NU
Amphetamines.
Use of Cannabis or Amphetamines during the 3 months preceding the onset of First Episode Psychosis
Proximal
Unemployment
Psychosocial
Patient unemployed at the time of psychosis onset due to:
o
PT
Unstable
ED
Factors
CE
accommodation
AC
Low social
connectedness
Loss of stable employment during the 2 years preceding psychosis onset, OR
o
Background of long-term unemployment
Predominant accommodation status during the 2 years preceding psychosis onset:
Homeless
Itinerant
Living in support housing
Inferred from case manager reports:
Weak or absent connections to family, friends and broader community at the time of psychosis onset
15
ACCEPTED MANUSCRIPT References
AC
CE
PT
ED
MA NU
SC
RI P
T
1. Buchanan RW, Carpenter WT. Domains of psychopathology: an approach to the reduction of heterogeneity in schizophrenia. The Journal Of Nervous And Mental Disease. 1994;182(4):193-204. 2. Harris MJ, Jeste DV. Late-Onset Schizophrenia: An Overview. Schizophrenia Bulletin. 1988;14(1):39-55. 3. Howard R, Rabins PV, Seeman MV, Jeste DV. Late-onset schizophrenia and very-late-onset schizophrenia-like psychosis: an international consensus. The International Late-Onset Schizophrenia Group. The American journal of psychiatry. 2000;157(2):172-8. 4. Gaebel W, Zielasek J. Integrative etiopathogenetic models of psychotic disorders: methods, evidence and concepts. Schizophrenia Bulletin. 2011;37 Suppl 2:S5-S12. 5. Gupta S, Kulhara P. What is schizophrenia: A neurodevelopmental or neurodegenerative disorder or a combination of both? A critical analysis. Indian Journal of Psychiatry. 2010;52(1):21-7. 6. Susana O, Judith U, Jesús C, Xavier L, Jayashri K. Gender Differences in Schizophrenia and First-Episode Psychosis: A Comprehensive Literature Review. Schizophrenia Research and Treatment. 2012:1-9. 7. Schultz SH, North SW, Shields CG. Schizophrenia: a review. American family physician. 2007;75(12):1821-9. 8. Maier W, Lichtermann D, Franke P, Heun R, Falkai P, Rietschel M. The dichotomy of schizophrenia and affective disorders in extended pedigrees. Schizophrenia research. 2002;57(2– 3):259-66. 9. Faridi K, Pawliuk N, King S, Joober R, Malla AK. Prevalence of psychotic and non-psychotic disorders in relatives of patients with a first episode psychosis. Schizophrenia research. 2009;114(13):57-63. 10. van Os J, Kenis G, Rutten BPF. The environment and schizophrenia. Nature. 2010(7321):203. 11. Bendall S, Jackson HJ, Hulbert CA, McGorry PD. Childhood trauma and psychotic disorders: A systematic, critical review of the evidence. Schizophrenia Bulletin. 2008;34:568-79. 12. Bourque F, van der Ven E, Malla A. A meta-analysis of the risk for psychotic disorders among first- and second-generation immigrants. Psychological Medicine. 2011;41(5):897-910. 13. Orlovska S, Pedersen MS, Benros ME, Mortensen PB, Agerbo E, Nordentoft M. Head injury as risk factor for psychiatric disorders: a nationwide register-based follow-up study of 113,906 persons with head injury. The American journal of psychiatry. 2014;171(4):463-9. 14. Moore THM, Zammit S, Lingford-Hughes A, Barnes TRE, Jones PB, Burke M, et al. Articles: Cannabis use and risk of psychotic or affective mental health outcomes: a systematic review. The Lancet. 2007;370:319-28. 15. Barkus E, Murray RM. Substance Use in Adolescence and Psychosis: Clarifying the Relationship. Annual review of clinical psychology. 2010;6:365-90. 16. Buckley PF, Miller BJ, Lehrer DS, Castle DJ. Psychiatric Comorbidities and Schizophrenia. Schizophrenia Bulletin. 2009;35(2):383-402. 17. Achim AM, Maziade M, Raymond É, Olivier D, Mérette C, Roy M-A. How Prevalent Are Anxiety Disorders in Schizophrenia? A Meta-Analysis and Critical Review on a Significant Association. Schizophrenia Bulletin. 2011;37(4):811-21. 18. Howard RJ, Graham C, Sham P, Dennehey J, Castle DJ, Levy R, et al. A controlled family study of late-onset non-affective psychosis (late paraphrenia). The British Journal of Psychiatry. 1997;170:511-4. 19. Wynn Owen PA, Castle DJ. Late-Onset Schizophrenia: Epidemiology, Diagnosis, Management and Outcomes. Drugs & Aging. 1999;15(2):81-9.
16
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
MA NU
SC
RI P
T
20. Reeves SJ, Sauer J, Stewart R, Granger A, Howard RJ. Increased first-contact rates for verylate-onset schizophrenia-like psychosis in African- and Caribbean-born elders The British Journal of Psychiatry. 2001(179 (2) ):172-4. 21. Golay P, Alameda L, Mebdouhi N, Baumann P, Ferrari C, Solida A, et al. Age at the time of onset of psychosis: A marker of specific needs rather than a determinant of outcome? European Psychiatry. 2017;45:20-6. 22. Selvendra A, Baetens D, Trauer T, Petrakis M, Castle D. First episode psychosis in an adult area mental health service-A closer look at early and late-Onset first episode psychosis Australasian Psychiatry, 2014; 22(3):235-41. 23. Castle D, Sham P, Murray R. Differences in distribution of ages of onset in males and females with schizophrenia. Schizophrenia research. 1998;33(3):179-83. 24. Thorup A, Waltoft BL, Pedersen CB, Mortensen PB, Nordentoft M. Young males have a higher risk of developing schizophrenia: a Danish register study. Psychological Medicine. 2007;37(4):479-84. 25. Welham J, Isohanni M, Jones P, McGrath J. The Antecedents of Schizophrenia: A Review of Birth Cohort Studies. Schizophrenia Bulletin. 2009;35(3):603-23. 26. Jeste DV, Harris MJ, Krull A, Kuck J, McAdams LA, Heaton R. Clinical and neuropsychological characteristics of patients with late-onset schizophrenia. The American journal of psychiatry. 1995;152(5):722-30. 27. Bourque F, van der Ven E, Fusar-Poli P, Malla A. Immigration, social environment and onset of psychotic disorders. Current Pharmaceutical Design. 2012;18(4):518-26. 28. Selten J-P, van der Ven E, Rutten BPF, Cantor-Graae E. The Social Defeat Hypothesis of Schizophrenia: An Update. Schizophrenia Bulletin. 2013;39(6):1180-6. 29. Cassano GB, Pini S, Mastrocinque T, Saettoni M, Papasogli A, Dell'Osso L. II. Phenomenology: Prevalence and clinical correlates of psychiatric comorbidity in patients with psychotic disorders. Schizophrenia research. 1998;29:33. 30. Carney CP, Jones L, Woolson RF. Medical Comorbidity in Women and Men with Schizophrenia: A Population-Based Controlled Study. Journal of General Internal Medicine. 2006(11):1133. 31. Greenfield P, Joshi S, Christian S, Lekkos P, Gregorowicz A, Fisher HL, et al. First episode psychosis in the over 35 s: is there a role for early intervention? Early Intervention In Psychiatry doi: 101111/eip12322 [Epub ahead of print]. 2016. 32. Foley DL, Mackinnon A, Morgan VA, Watts GF, Castle DJ, Waterreus A, et al. Common familial risk factors for schizophrenia and diabetes mellitus. Australian & New Zealand Journal of Psychiatry. 2016;50(5):488-94. 33. Dixon L, Postrado L, Delahanty J, Fischer PJ, Lehman A. The Association of Medical Comorbidity in Schizophrenia with Poor Physical and Mental Health. The Journal of Nervous and Mental Disease. 1999;187(8):496-502. 34. National Drug Strategy Household Survey detailed report: 2013. Canberra: Australian Institute of Health and Welfare; 2014 Nov 25. 35. Melle I, Friis S, Hauff E, Vaglum P. Social functioning of patients with schizophrenia in highincome welfare societies. Psychiatric Services. 2000;51(2):223-8. 36. Vargas G, Strassnig M, Sabbag S, Gould F, Durand D, Stone L, et al. The course of vocational functioning in patients with schizophrenia: Re-examining social drift. Schizophrenia Research: Cognition. 2014;1(1):41-6. 37. Cullen K, Guimaraes A, Wozniak J, Anjum A, Schulz SC, White T. Trajectories of social withdrawal and cognitive decline in the schizophrenia prodrome. Clinical Schizophrenia & Related Psychoses. 2011;4(4):229-38. 38. Rapoport JL, Giedd JN, Gogtay N. Neurodevelopmental model of schizophrenia: update 2012. Molecular Psychiatry. 2012;17(12):1228-38.
17
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
MA NU
SC
RI P
T
39. Rapoport JL, Addington A, Frangou S. The neurodevelopmental model of schizophrenia: What can very early onset cases tell us? Current Psychiatry Reports. 2005;7(2):81-2. 40. Bayer TA, Falkai P, Maier W. Genetic and non-genetic vulnerability factors in schizophrenia: the basis of the "Two hit hypothesis". Journal Of Psychiatric Research. 1999;33(6):543-8. 41. Keshavan MS. Development, disease and degeneration in schizophrenia: a unitary pathophysiological model. Journal of Psychiatric Research. 1999;33(6):513-21. 42. Wyatt RJ. Neuroleptics and the natural course of schizophrenia. Schizophr Bull. 1991;17(2):325-51. 43. Aricioglu F, Ozkartal CS, Unal G, Dursun S, Cetin M, Müller N. Neuroinflammation in Schizophrenia: A Critical Review and The Future. Klinik Psikofarmakoloji Bulteni. 2016;26(4):429-37. 44. Castle DJ, Howard R. What do we know about the aetiology of late-onset schizophrenia? European Psychiatry. 1992;7(3):99-108. 45. Mondelli V, Dazzan P, Hepgul N, Di Forti M, Aas M, D'Albenzio A, et al. Abnormal cortisol levels during the day and cortisol awakening response in first-episode psychosis: the role of stress and of antipsychotic treatment. Schizophrenia research. 2010;116(2):234-42. 46. Akdeniz C, Schäfer A, Streit F, Haller L, Wüst S, Kirsch P, et al. Sex-Dependent Association of Perigenual Anterior Cingulate Cortex Volume and Migration Background, an Environmental Risk Factor for Schizophrenia. Schizophrenia Bulletin. 2017;43(4):925-34. 47. Meyer-Lindenberg A, Tost H. Neural mechanisms of social risk for psychiatric disorders. Nature neuroscience. 2012;15(5):663-8. 48. Chen Y, Baram TZ. Toward Understanding How Early-Life Stress Reprograms Cognitive and Emotional Brain Networks. Neuropsychopharmacology. 2016;41(1):197-206.
18
ACCEPTED MANUSCRIPT
RI P
T
Figure
Schizophrenia frequency by gender and age (5yr intervals)
SC
45
40
MA NU
35
Frequency
30
25 20
15
5
0 5
10
15
20
25
30
35
40
45
Age of onset
Female
Male
CE
PT
0
ED
10
AC
Figure 1. Age of onset, based on gender
19
50
55
60
65
70
75
ACCEPTED MANUSCRIPT Table 1
Proportion of
Youth onset
Middle onset
Late onset
total sample
<26 years
26-40 years
>40 years
100%
n=104
n=81
n=40
64.9%
72.1%
40.0%
47.1%
RI P
Diagnosis of Schizophrenia
67.9% (P=0.534)
40.0% (P<0.01)
35.8% (P=0.122)
30% (P=0.063)
24.0%
4.9% (P<0.01)
2.5% (P<0.01)
28.9%
44.4% (P=0.028)
42.5%
illness Family history of
13.3%
Schizophrenia Early Developmental Risk Factors 1st generation migrant status
ED
36.9%
Childhood trauma
42.2%
functioning
Tertiary completed)
38.3% (P=0.228)
37.5% (P=0.298)
27.9%
9.9% (P<0.01)
5.0% (P<0.01)
20.0%
8.7%
28.4% (P<0.01)
32.5% (P<0.01)
31.6%
34.6%
28.4% (P=0.519)
30.0% (P=0.2)
28.9%
22.1%
30.9% (P=0.178)
42.5%
CE
Educational completion (% of
47.1%
(P=0.839)
17.3%
PT
Poor early psychosocial
MA NU
Family Psychiatric History Family history of any mental
SC
(n=225) Gender (% of males)
T
Table 1. Comparison of risk factor frequencies in different age onset groups
Psychiatric and Physical Health Factors
conditions
AC
Previous psychiatric
Physical health problems
(P=0.014) Substance Use Lifetime Cannabis use
55.6%
65.4%
53.1% (P=0.09)
35.0% (P<0.01)
Recent Cannabis use (within 3
34.2%
46.2%
28.4% (P=0.014)
15.0% (P<0.01)
36.9%
41.4%
39.5% (P=0.8)
20.0%
months preceding psychosis onset) Lifetime Amphetamine use
(P=0.016) Recent amphetamine use
22.7%
27.9%
(within 3 months preceding psychosis onset) Proximal Psychosocial Factors
20
23.5% (P=0.495)
7.5% (P<0.01)
ACCEPTED MANUSCRIPT Low social connectedness
58.7%
59.6%
54.3% (P=0.555)
72.5% (P=0.124)
Unstable accommodation
26.9%
28.4%
27.2% (P=0.971)
35.0% (P=0.340)
53.9%
60.5% (P=0.365)
RI P
60.0%
T
Unemployment
75.0% (P=0.016)
Note: P-values generated represent the comparisons with the under 26 age onset group. Comparisons with
AC
CE
PT
ED
MA NU
SC
P<0.05 are denoted in red.
21
ACCEPTED MANUSCRIPT
T
Highlights
Schizophrenia onset at different ages is associated with differing risk factors.
Male gender and heredity are significant risk factors in early onset schizophrenia.
Female gender and health problems are more common in older onset schizophrenia.
Recent and lifetime drug use are more common in early onset schizophrenia.
Graded trends are observed with age in several risk factors.
AC
CE
PT
ED
MA NU
SC
RI P
22