Transcript Document

Epidemiology of
Schizophrenia
Dr Golam Khandaker
Wellcome Trust Clinical Research Fellow in Psychiatry
[email protected]
Contents
• Why epidemiology?
• Schizophrenia
– Disease burden
– Demographic correlates
– Social and biological risk factors
Why epidemiology?
•
•
•
•
Describes a condition
Clues to it’s origin, outcome, etc.
Generates hypotheses for experiments
Examines experimental findings in human
population
Schizophrenia
• Disease burden
– Prevalence & incidence
• Demographic correlates
– Age, gender and social class
• Social risk factors
– Ethnicity & migration
– Urban birth
• Biological risk factors and
antecedents
–
–
–
–
–
–
–
–
–
–
Gene
Paternal age
Season of birth
Pregnancy and birth
complications
Maternal infection in
pregnancy
Childhood infections
Childhood development: motor
Childhood development:
cognitive
Cannabis use
Adult infection and immune
dysfunction
Disease burden
Prevalence
• Point Prevalence: 4.6 (1.9- 10.0) per 1000
• One year prevalence: 3.3 (1.3- 8.2) per
1000
• Life-time prevalence: 4.0 (1.6- 12.1) per
1000
• Issues of case definition, denominator, etc
Saha et al (2005) PLoS Medicine, 2, 413-433
McGrath et al (2008) Epidemiologic Reviews, 30, 67- 76
Incidence
• Is about 20 cases per 100,000 per year
• WHO study
– Similar incidence across the globe
– ?reality or bias in methods
• Incidence may be declining
• Schizophrenia point prevalence is much
higher than it’s annual incidence… why?
Textbook of Psychiatric Epidemiology. 3rd edition, page 265
Demographic correlates
Age and Gender
Patients (%)
30
Female
Male
20
10
0
12-14 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59
Age group (y)
Hafner et al 1993
Socioeconomic status
• Strong correlation with lower social class
• Effects of insidious onset
• May be unemployed many years before,
and many years after first diagnosis
• Some indication of a social drift1
1. Byrne et al (2005), Social Psychiatry and Psychiatric Epidemiology, 39, 87- 96
Social risk factors
Ethnicity & Migration
• AESOP study, UK1
– 9.1 times ↑ risk in African-Caribbean
– 5.8 times ↑ risk in Black African
– No higher risk in Asians
• Higher risk among Surinamese immigrants
to Netherlands
• Overall estimated migrant/ native born rate
ratio is 4 to 5
1. Kirkbride et al (2008), British Journal of Psychiatry, 193, 18- 24
Ethnicity & Migration: mechanism
• ?Genes: rate is lower at the country of
origin
• ?Stress of immigration: high risk only in
Black immigrants, and also in the 2nd
generation
• Obstetric complication
• Decreased production of vitamin D
• Social and psychological factors
Textbook of Psychiatric Epidemiology. 3rd edition, page 268
Urban birth
• Relative risk 2 to 4 for urban birth
• ?Interaction with genetic risk
• What about cities?
– Environmental pollution
– Crowding and infection
– Stress and differences in cultural environment
Textbook of Psychiatric Epidemiology. 3rd edition, page 268
Biological risk factors and
antecedents
Genes
• Heritability is above 50%
• ‘Recurrence’ risk
– Both parents 89%
– Monozygotic twin 50%
– Father or mother only 10%
– Full sibling 8%
– Half sibling 2.5%
– First cousin 2.3%
Which genes?
• Some candidate genes
– Dopamine receptor D1 (DRD1)
– Dysbindin aka. Dystrobrevin binding protein 1
(DTNBP1)
– Methylenetetrahydrofolate reductase
(MTHFR) gene
– Tryptophan hydroxylase 1 (TPH1) gene
– COMT
– Zinc Finger Protein 804A gene (ZNF804A)
Paternal Age
• Paternal age >50 vs. 25- 29 years, relative
risk 1.66 (1.46- 1.89)1
• Continuing germ line cell division in male
throughout reproductive period leading to
accumulating genetic abnormalities from
de novo mutations
1. Miller et al (2010). Schizophrenia Bulletin
Season of birth & Latitude
• Winter birth:
– small risk (10% increase) but reliably
replicated
– both hemispheres
• Cooler latitude:
– 40 to 50 degrees from equator
• Increased infection as a possible cause
50th parallel north
40th parallel north
Line of cancer 23 N
Equator
Pregnancy and birth complications
Risk factor
Odds ratio (95% confidence interval)
Number of subjects
Maternal diabetes
7.76 (1.37 to 3.90)
2,146
Birth weight < 2000g
3.89 (1.40 to 0.84)
11,430
Emergency caesarean
3.24 (1.40 to 7.50)
508,680
Congenital
malformations
2.35 (1.21 to 4.57)
509,518
Uterine atony
2.29 (1.51 to 3.50)
508,362
Rhesus factors
2.00 (1.01 to 3.96)
18,296
Asphyxia
1.74 (1.15 to 2.62)
3,406
Bleeding in pregnancy
1.69 (1.14 to 2.52)
526,195
Birth weight < 2500g
1.67 (1.22 to 2.29)
537,339
Preeclampsia
1.36 (0.99 to 1.85)
11,987
Cannon et al (2002), American Journal of Psychiatry, 159, 1080- 1092
Maternal Infection in Pregnancy
• Serologically confirmed influenza
– First trimester, odds ratio 7
– First half of pregnancy, odds ratio 3
• Serologically confirmed HSV-2 and Toxoplasma
Gondii
• Respiratory infection in 2nd trimester, genital and
reproductive infection during periconceptional
periods
• Increased maternal proinflammatory cytokines,
e.g. IL-8 (2nd trimester) and TNF-α (3rd trimester)
Brown et al (2010), American Journal of Psychiatry
Childhood infections
• CNS infection, particularly viral infections
most consistently associated with risk
• Risk estimate varies from 1.3 to 4.5
Childhood development: motor
1.8
12 months n=1556
11 months n=1295
10 months n=1144
09 months n=1002
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
0
15
16
17
18
19
20
21
22
23
24
25
26
27
Figure: Cumulative incidence of schizophrenia by age
of learning to stand without support within males
Isohanni et al 2003
Childhood development: cognitive
Group by
Age at testing
Early
Early
Early
Early
Early
Early
Early
Early
Early (pooled)
Late
Late
Late
Late
Late
Late
(pooled)
Study name
Statistics for each study
Std diff
in means
Jones
Crow (Male)
Crow (Female)
T Cannon
Walker
Dunedin
Seidman
Osler
Gunnell
Caspi
Zammit
Reichenberg
Urf
er-Parnes
Urfer-Parnas
-0.274
-0.709
-0.518
-0.529
-0.150
-0.504
-0.648
-0.144
-0.392
-0.260
-0.600
-0.527
-0.565
-0.372
-0.464
Standard
error
0.205
0.260
0.269
0.133
0.178
0.175
0.226
0.108
0.082
0.129
0.218
0.053
0.023
0.025
0.063
Std diff in means and 95% CI
p-Value
0.181
0.006
0.054
0.000
0.398
0.004
0.004
0.181
0.000
0.044
0.006
0.000
0.000
0.000
0.000
-1.00
-0.50
Fav ours A
Cases low IQ
Meta Analysis
0.00
0.50
1.00
Fav ours B
Cases high IQ
SMD (standardized mean difference) is 0.44, i.e. mean IQ in future cases is about 7 IQ
points lower than controls
Figure: Premorbid IQ and schizophrenia
10 20
5
2
1
.5
.2
60
80
100
Estimated mean IQ
Reichenberg
Gunnell
Jones
120
Zammit
Tiihonen
Pooled
Figure: A linear association between premorbid IQ deficit
and risk of schizophrenia
140
Figure: Premorbid IQ and age of onset of schizophrenia
Cannabis use
• Review of 35 longitudinal populationbased studies of psychosis,1 odds ratio
– ever use of cannabis 1.4
– frequent use 2.1
• Possible interaction with gene (COMT)
• Possible grey matter volume loss and
ventricular enlargement in first episode
1. Moore T et al (2007), Lancet, 370, 319- 328
Adult Immune dysfunction
• Immune imbalance or Proinflammatory
state
– Increased cytokines, e.g. IL-6
• Auto-antibodies
– NMDA-R or VGKC auto-antibodies in first
episode cases
• Increased infection
– e.g. Toxoplasma gondii
Summary
• 20 new cases of schizophrenia per 100,000
population per year
• Prevalence is about 3 to 4 per 1000
• Strong genetic component with several
candidate genes
• Associations exist with male gender, lower social
class, African ethnic origin, birth and pregnancy
complications, prenatal and childhood infection
• Future case show delays in motor milestone,
and deficit in cognitive performance
• Indication of immune dysfunction in adult cases
Further reading
• Textbook of Psychiatric Epidemiology, 3rd edition.
Edited by Tsuang, Tohen & Jones. Willey-Blackwell
Publishing 2011
• The Epidemiology of Schizophrenia, edited by Murray,
Jones, Susser, van Os & Cannon. Cambridge University
Press 2009