Social class at birth and risk of psychosis

Abstract

Background:

Individuals with psychotic disorders are represented more in the lower social classes, yet there is conflicting evidence to whether these individuals drift into the lower social classes or whether lower social class is a risk factor for developing psychosis. The aim of this study was to examine whether the social class at birth is a risk factor for developing psychosis.

Methods:

We included individuals with a first episode of psychosis (FEP) whose social class at birth was determined from birth records. We employed a case-control study design and also compared the distribution of the social classes at birth of the cases to that of the general population.

Results:

A total of 380 individuals with an FEP and 760 controls were included in the case-control study. The odds ratio for developing an FEP associated with social class (low vs high) was .62 (95% confidence interval (CI): .46–.85, p < .001), indicating that individuals from a lower social class at birth have a reduced risk of psychosis. Individuals born between 1961 and 1980 with an FEP were more likely to be from a higher social class at birth compared to the general population (60.8% vs 36.7%, χ² = 60.85, df = 1, p < .001). However, this association was not observed for those born between 1981 and 1990.

Conclusion:

A higher social class at birth is associated with a greater risk for developing a psychotic disorder; however, this effect may show temporal variation.

Keywords

Psychosis social class schizophrenia

Introduction

An association between social class and psychiatric illness has long been recognised. In 1939, Faris and Dunham observed that there was a higher rate of individuals with schizophrenia in the lower social classes and higher rates of schizophrenia in the more socially disadvantaged areas (Faris & Dunham, 1939). This introduced the pertinent question, namely, was this observation a result of downward drift associated with schizophrenia, the ‘social drift’ theory, or was the disorder caused by the environment, the ‘social causation’ theory (Cantor-Graae, 2007).

The social drift theory, which states that the social class of the families of individuals with schizophrenia is representative of the distribution of the social classes in the general population, was supported by the initial findings of Goldberg and Morrison (Goldberg & Morrison, 1963), and these findings were replicated in the United Kingdom (Hare, Price, & Slater, 1972; Jones et al., 1993). However, two large birth cohorts studies in Sweden (Wicks, Hjern, Gunnell, Lewis, & Dalman, 2005) and in Israel (Corcoran et al., 2009; Werner, Malaspina, & Rabinowitz, 2007) and matched case-control studies from the United Kingdom (Castle, Scott, Wessely, & Murray, 1993; Harrison, Gunnell, Glazebrook, Page, & Kwiecinski, 2001) have demonstrated that a lower social class at birth is a risk factor for psychosis. Furthermore, cohort studies from the United Kingdom (Done, Crow, Johnstone, & Sacker, 1994; Jones, Rodgers, Murray, & Marmot, 1994), Holland (Wiersma, Giel, De Jong, & Slooff, 1983) and Finland (Makikyro et al., 1997) and a case-control study from Ireland (Mulvany et al., 2001) have produced conflicting findings and demonstrated that coming from a higher social class is a risk factor for developing psychosis. The case-control study conducted in Ireland, in addition to finding a higher risk of schizophrenia among people from a higher social class at birth, found that individuals from a higher social class at birth presented over 8 years earlier with a first episode of psychosis (FEP) than those from a lower social class at birth (Mulvany et al., 2001).

A recent systematic review on the relationship between social class at birth and risk of psychosis identified a possible temporal trend, as studies conducted post 2001 found an association between lower social class at birth and a risk for psychotic disorder (Kwok, 2014). Therefore, in light of this contradictory evidence, we aimed to determine whether the social class at birth is a risk factor for developing psychosis, utilising two methodologies, a case-control study and a comparison of the distribution of the social classes in the general population according to census periods.

Methodology

Participants and setting

We included two cohorts in this study, a cohort of individuals with an FEP who presented to an urban catchment area community mental health service (population 165,000) between 1995 and 1999 and a second cohort, referred to as the Early Intervention cohort, who presented to this catchment and two adjacent catchment area mental health services (total population 360,000) between February 2006 and August 2011. We defined an FEP as having no previous history of a psychotic episode and not having received antipsychotic medication for more than 30 days prior to the referral. We included individuals with co-morbid substance misuse, substance dependence or co-morbid personality disorder. We only included individuals who were born in Ireland.

Instruments

All individuals underwent a Structured Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders IV (SCID) assessment, an instrument that has shown good reliability (Kappa: .93) and validity in the assessment of psychotic disorders (Fennig, Craig, Lavelle, Kovasznay, & Bromet, 1994; First, Spitzer, Gibbon, & Williams, 2002; Skre, Onstad, Torgersen, & Kringlen, 1991). We defined a schizophrenia-spectrum disorder as including schizophrenia and schizophreniform disorder. The Lewis and Murray Obstetric Complications Scale (MLOCS) (Lewis, Owen, & Murray, 1989) was used to detect the presence of any obstetrical complications, and this instrument was used in the interview with the subject’s mother. The MLOCS was only conducted in the early intervention cohort.

Cases and controls

We obtained the birth records from the Register of Births, and in Ireland, births were recorded in a register by hand in each maternity hospital. There are usually 8–10 births recorded on each page. Controls were selected from the first eligible birth before and after the case on the birth register. Each case was matched for gender, date of birth (within 1 week) and hospital of birth with two controls.

Sources of information

Social class at birth was determined from the father’s occupation as recorded on the subject’s birth certificate, and each occupation was coded according to the Republic of Ireland Census Classification (Central Statistics Office, 2006). Two researchers recorded the occupation of the father that was stated on the birth record, and one researcher coded it according to the above classification system. There are six social classes that are broadly defined as social class I – professional workers, social class II – managerial and technical, social class III – non-manual, social class IV – skilled manual, social class V – semi-skilled and social class VI – unskilled. We dichotomised the grouping of social class into higher social class (social classes I–III) and lower social class (social classes IV–VI); this is system of grouping that has been utilised in previous studies (Castle et al., 1993; Harrison et al., 2001; Mulvany et al., 2001; Werner et al., 2007). Information pertaining to family history of psychotic disorders and paternal age was obtained from interviews with the individual or their family.

Ethical approval

Ethical approval to obtain information from the cases’ birth records was obtained from the St John of God Hospitaller Order Ethics committee.

Statistical analysis

Data were entered into MS Access and then exported to SPSS version 20 for analysis. Odds ratios were calculated for the case-control study (Schlesselman, 1982), and chi-square tests were used to determine whether there were significant differences between two categorical groups.

The distribution of the social classes in previous years was obtained from the censuses of Ireland. Censuses in Ireland were carried out in 1961, 1966, 1971, 1979, 1981, 1986 and 1991. Individuals were compared to the census that preceded their birth, for example, an individual who was born in 1963 would be compared to the 1961 census. However, there was insufficient information available in the 1979 Census to determine the distribution of the social classes. The distribution of the social classes in the different censuses were compared to determine whether they were similar, so as to group them together. In the 1961, 1966 and 1971 Census, the distribution of the higher and lower social classes did not differ by more than 1%, and therefore, they were grouped together.

Power

To detect a difference of 10% in the proportion of social classes (lower vs higher) with a power of 90% at a two-sided 5% significance level, 403 cases and 806 controls were required.

Results

Participants

A total of 608 individuals were diagnosed with an FEP, and 84.2% (n = 512) were born in Ireland. The social class at birth could not be determined for 82 individuals, and the reasons for this are presented in Figure 1. Therefore, the social class at birth was determined for 84.0% (n = 430) of those born in Ireland. A flow diagram of the inclusion of participants is displayed in Figure 1. Of those included in the study, 63.7% (n = 274) were male, 80.2% (n = 345) were single and 48.3% (n = 208) had a diagnosis of schizophrenia or schizophreniform disorder. A total of 57.4% (n = 247) were from a higher social class at birth and 42.6% (n = 183) were from a lower social class at birth. The demographic and clinical characteristics of the cohort, including the full distribution of the social classes, are presented in Table 1.

Figure 1.

Flow diagram of participants involved in the case-control study.

Table 1.

Demographic and clinical characteristics of participants.

	Full cohort (N = 430)			Case-control subsample (N = 380)
	%	n	Mean ± SD	%	n	Mean ± SD
Gender
Male	63.7	274		64.5	245
Female	36.3	156		35.5	135
Mean age			30.9 ± 10.9			30.1 ± 10.4
Marital status, %
Married or living with partner	13.0	56		10.5	40
Single	80.2	345		82.9	315
Separated/divorced	6.5	28		6.3	24
Widowed	0.2	1		0.3	1
Social class at birth
Social class I – professional	13.7	59		12.6	48
Social class II – managerial and technical	25.3	109		26.1	99
Social class III – non-manual	18.4	79		18.9	72
Social class IV – skilled manual	24.0	103		23.9	91
Social class V – semi-skilled	15.3	66		15.8	60
Social class VI – unskilled	3.3	14		2.6	10
Family history of psychosis
First-degree relative	10.0	43		9.7	37
Second-degree relative	12.1	52		12.4	47
Diagnosis
Schizophrenia and schizophreniform	48.3	208		48.1	183
Bipolar affective disorder	13.5	58		14.7	56
Major depressive disorder	7.4	32		7.1	27
Schizoaffective disorder	0.9	4		1.1	4
Delusional disorder	9.5	41		9.2	35
Substance-induced psychotic disorder	10.2	44		10.3	39
Brief psychotic disorder	4.7	20		4.2	16
Psychotic disorder NOS	3.3	14		3.2	12
Psychotic disorder due to general medical condition	2.1	9		2.1	8
Co-morbid substance abuse
Alcohol abuse	20.0	86		20.0	76
Cannabis abuse	20.2	87		20.8	79
Clinical characteristics
Global functioning, mean (SD)			33.5 ± 14.4			33.6 ± 14.6
Insight, mean (SD)			7.5 ± 3.1			7.5 ± 3.1

SD: standard deviation; NOS: not otherwise specified.

Social class as a risk factor for psychosis – case-control study

Controls could not be obtained for a further 50 cases, resulting in a total of 380 individuals with an FEP and 187 in the sub-group with a diagnosis of a schizophrenia or schizophreniform disorder who were included in the case-control study.

There were 380 cases and 760 controls included in the case-control study. A total of 37.3% (n = 142) of cases and controls were from the same social class, 24.5% (n = 93) of both controls were from a different social class to that of the case and the remainder had one control from a higher and lower social class. The outcomes of the case-controls studies are presented in Table 2. The odds ratio for developing an FEP associated with social class (low vs high) was .62 (95% confidence interval: .46–.85, p < .001). This result indicates that individuals from a lower social class at birth have a reduced risk of psychosis compared with those from higher social classes.

Table 2.

Outcomes of case-control studies.

Outcome for all first episodes of psychosis			n	%
Case	Control 1	Control 2
Lower	Lower	Lower	67	17.6
Lower	Lower	Higher	26	6.8
Lower	Higher	Lower	35	9.2
Lower	Higher	Higher	33	8.7
Higher	Lower	Lower	60	15.8
Higher	Lower	Higher	36	9.5
Higher	Higher	Lower	48	12.6
Higher	Higher	Higher	75	19.7
Outcome for schizophrenia diagnoses			n	%
Case	Control 1	Control 2
Lower	Lower	Lower	34	18.2
Lower	Lower	Higher	7	3.7
Lower	Higher	Lower	12	6.4
Lower	Higher	Higher	20	10.7
Higher	Lower	Lower	29	15.5
Higher	Lower	Higher	20	10.7
Higher	Higher	Lower	23	12.3
Higher	Higher	Higher	42	22.5

In the sub-group of 187 individuals who presented with a diagnosis of a schizophrenia-spectrum disorder, the odds ratio for developing schizophrenia associated with social class (low vs high) was .58 (95% CI: .55–.83, p = .007). This result indicates that individuals from a lower social class at birth have a reduced risk of schizophrenia compared with those from higher social classes.

Comparison of the social class at birth of cases to the distribution of the social class of the general population

Information was available from the 1961 census and onwards regarding the distribution of the social classes in the general population, except for the 1979 census, and this information is presented in Table 3. Of the 430 individuals with an FEP, 83.7% (n = 360) were born in 1960 or after, and therefore, their social class at birth could be compared to the distribution of the social classes in the general population. There were only five individuals born in 1991 or later, and the distribution of the social classes of this group was not compared to that of the general population because of the small numbers involved. The distribution of the social classes of the cases compared to the general population from their corresponding censuses are presented in Table 3.

Table 3.

Comparison of the distribution of the social class at birth of cases to the social classes of the general population.

	Cases 1961–1980		General pop 1971		Cases 1981–1985		General pop 1981		Cases 1986–1990		General pop 1986
	n	%	n	%	n	%	n	%	n	%	n	%
All first-episode psychoses diagnoses
Social class I	38	15.5	53,065	4.8	4	7.5	70,173	5.7	8	14.0	123,918	9.5
Social class II	61	24.9	149,847	13.6	14	26.4	204,957	16.6	20	35.1	217,952	16.7
Social class III	50	20.4	202,229	18.4	7	13.2	248,195	20.2	6	10.5	290,742	22.3
Social class IV	52	21.2	274,364	24.9	15	28.3	331,247	26.9	19	33.3	281,263	21.6
Social class V	37	15.1	382,519	34.7	12	22.6	356,403	28.9	4	7.0	256,458	19.7
Social class VI	7	2.9	39,439	3.6	1	1.9	20,999	1.7	0	0	132,910	10.2
Total	245	100	1,101,463	100	53	100	1,231,974	100	57	100	1,303,243	100
	χ² = 110.49, df = 5, p < .001				χ² = 5.41, df = 5, p = .39				χ² = 30.37, df = 5, p < .001
Higher social class	149	60.8	405,141	36.8	25	47.2	523,325	42.5	34	59.6	632,612	48.5
Lower social class	96	39.2	696,322	63.2	28	52.8	708,649	57.5	23	40.4	670,631	51.5
	χ² = 60.85, df = 1, p < .001				χ² = 0.48, df = 1, p = .49				χ² = 2.82, df = 1, p = .09
Schizophrenia-spectrum disorder diagnoses
Social class I	20	14.7	53,065	4.8	1	4.2	70,173	5.7	5	21.7	123,918	9.5
Social class II	40	29.4	149,847	13.6	8	33.3	204,957	16.6	6	26.1	217,952	16.7
Social class III	25	18.4	202,229	18.4	5	20.8	248,195	20.2	2	8.7	290,742	22.3
Social class IV	27	19.9	274,364	24.9	7	29.2	331,247	26.9	9	39.1	281,263	21.6
Social class V	21	15.4	382,519	34.7	3	12.5	356,403	28.9	1	4.3	256,458	19.7
Social class VI	3	2.2	39,439	3.6	0	0	20,999	1.7	0	0	132,910	10.2
Total	136	100	1,101,463	100	24	100	1,231,974	100	23	100	1,303,243	100
	χ² = 69.24, df = 5, p < .001				χ² = 6.82, df = 5, p = .23				χ² = 15.11, df = 5, p = .01
Higher social class	85	62.5	405,141	36.8	14	58.3	523,325	42.5	13	56.5	632,612	48.5
Lower social class	51	37.5	696,322	63.2	10	41.7	708,649	57.5	10	43.5	670,631	51.5
	χ² = 38.68, df = 1, p < .001				χ² = 2.47, df = 1, p = .12				χ² = 7628.9, df = 1, p < .001

Distribution of the social class at birth in FEP cases born 1961–1980 compared to the distribution of the social class in the general population in 1971

A total of 245 individuals with an FEP were born between 1961 and 1980, and 60.8% (n = 149) were from a higher social class at birth and 39.2% (n = 96) were from a lower social class at birth. From the 1971 census, 36.7% of the population were in the higher social classes and 63.2% were in the lower social classes. At birth, individuals who subsequently developed an FEP were represented more in the higher social classes when compared to the general population (60.8% vs 36.7%, χ² = 60.85, df = 1, p < .001). This finding was consistent in the sub-group of individuals with a schizophrenia-spectrum disorder (62.5% vs 36.7%, χ² = 38.68, df = 1, p < .001).

Distribution of the social class at birth of FEP cases born 1981–1985 compared to the distribution of the social class in the general population in 1981

A total of 53 individuals with an FEP were born between 1981 and 1985, and 47.2% (n = 25) of were from a higher social class at birth and 52.8% (n = 28) were from a lower social class at birth. From the 1981 census, 42.5% of the population were in the higher social classes and 57.5% were in the lower social classes. At birth, individuals who subsequently developed an FEP were not represented more in the higher social classes when compared to the general population (47.2% vs 42.5%, χ² = 0.48, df = 1, p = .49). This finding was consistent in the sub-group of individuals with a schizophrenia-spectrum disorder (58.3% vs 42.5%, χ² = 2.47, df = 1, p = .12).

Distribution of the social class at birth of those born 1986–1990 compared to the distribution of the social class in the general population in 1986

A total of 57 individuals with an FEP were born between 1986 and 1990, and 59.6% (n = 34) were from a higher social class at birth and 40.4% (n = 23) were from a lower social class at birth. From the 1986 census, 48.5% of the population were in the higher social classes and 51.5% were in the lower social classes. There were no differences in the distribution of the social classes in individuals who developed an FEP and the distribution of the social classes in the general population (59.6% vs 48.5%, χ² = 2.82, df = 1, p = .09). This finding was consistent in the sub-group of individuals with a schizophrenia-spectrum disorder (56.5% vs 48.5%, χ² = 0.59, df = 1, p = .44).

Post hoc analysis – possible confounders in the relationship between social class at birth and the risk for psychosis

A number of established risk factors for psychosis, specifically family history, substance abuse, place of birth, paternal age and obstetric complications, could be more prevalent in different social classes. Therefore, post hoc analysis was performed to determine whether these factors were more prevalent in the cases from a higher social class at birth compared to the cases from a lower social class at birth. This information was not available on controls.

There was no difference in the proportion of individuals with an FEP with a family history of psychosis in a first (7.7% vs 11.7%, χ = 1.95, df = 1, p = .16) or second (13.7% vs 10.9%, χ² = 0.74, df = 1, p = .39) degree relative in the lower or higher social classes. There was no difference in the prevalence of alcohol abuse (24.0% vs 17.0%, χ² = 3.26, df = 1, p = .07) or cannabis abuse (20.5% vs 20.5%, χ² = 0.0, df = 1, p = .96) in individuals with an FEP from the lower social classes at birth compared to those from the higher social classes. Similarly, there was no difference in whether an individual was born in an urban or rural area according to their social class at birth (χ² = 0.97, df = 1, p = .33). The father’s age at the time of the subject’s birth was only recorded in the early intervention cohort, and it was only recorded for 47.7% (n = 135) of cases. The mean age of fathers in the higher social class group was 32.6 (standard deviation (SD): 7.3) years and it was 30.7 (SD: 12.1) years in the lower social class group, and this difference was not statistically significant (t = 1.12, df = 134, p = .16). As the MLOCS was conducted in the interview with the subject’s mother, it was only completed for 45% (n = 128). There was no difference in the prevalence of any obstetrical complications in individuals with an FEP according to the social classes at birth (46.4% vs 37.5%, χ² = 1.04, df = 1, p = .31).

Discussion

Summary of findings

The case-control study demonstrated that individuals from a higher social class at birth had a higher risk of developing psychosis or schizophrenia, and this was replicated in the cohort born between 1961 and 1980. However, no association between the social class at birth and risk of psychosis or schizophrenia was seen in those born from 1981 to 1990.

Strengths and limitations

The strengths of this study are that we employed two methodologies to test our hypotheses and we included a clinically representative cohort from geographically defined catchment areas. A further strength of the study is that we obtained objective data on the social class at birth that was not subject to recall bias. Additionally, we obtained controls from the general population rather than cohorts with diagnoses of other mental disorders.

There are a number of limitations to this study that should be considered when interpreting these results. While the total number of participants could be considered large, the final number included in the case-control was marginally short of what was estimated to sufficiently power the study; furthermore, only 62.5% of all those who presented with an FEP were included in the case-control study. We must also consider the possibility of a selection bias, as it is possible that certain individuals experiencing an FEP may not have presented or have been detected by the mental health services. The early intervention for psychosis service undertook extensive public and professional educational campaigns that would have facilitated the detection of cases of psychosis (O’Donoghue et al., 2012); however, the other cohort included in this study had no such educational campaign, therefore increasing the possibility of a selection bias. Second, we did not know the maiden name of a small number of females and were therefore unable to locate their birth certificates. Finally, we did not have information about the controls at follow-up and whether they developed psychosis, and with a lifetime risk of psychosis approximately 3% in the general population, it is likely that approximately 23 of the 760 controls would have developed psychosis (Van Os & Kapur, 2009).

Comparisons with previous literature

This study replicates the findings of the previous study undertaken in Ireland that used a similar methodology of a case-control study and found that the risk of developing schizophrenia was also associated with a higher social class at birth (Mulvany et al., 2001). This study also replicates the findings from the Northern Finland 1966 Birth Cohort (Makikyro et al., 1997), the British 1946 Birth Cohort (Jones et al., 1994) and the Dutch cohort study (Wiersma et al., 1983). Despite being replicated now a number of times, this finding has not drawn much discussion or further research exploring the possible factors more prevalent in a family of higher social class that may have an aetiological role in the development of psychosis and schizophrenia. There have been a number of limitations throughout all of the studies conducted to date on this topic. The only study to have included a power analysis was the case-control study by Mulvany et al. (2001); however, the large cohort study conducted in Israel had a large number of cases (n = 520) and was likely to have been sufficiently powered. Furthermore, Mulvany et al. (2001) only included inpatients in their study, a limitation shared by Goldberg and Morrison (1963), Done et al. (1994) and Jones et al. (1993). Interestingly, the studies that only included inpatients tended to find that there was either no difference in the social class at birth or that a higher social class at birth was a risk factor. By including only inpatients from the public system, it is possible that a selection bias may have been introduced, as individuals with a first presentation of a psychotic disorder from higher social classes are more likely to be admitted to private psychiatric hospitals (Muntaner, Wolyniec, McGrath, & Pulver, 1994).

Implications of findings

One possible explanation for the inconsistent findings relating to whether the social class at birth is a risk factor for psychosis in the literature to date is that the social structure and stratification of social classes may vary across countries. This theory is supported by the similarity of the findings between this study and those of Mulvany et al. who undertook their study in a similar setting in Ireland. However, studies undertaken in the United Kingdom yielded contrasting results (Castle et al., 1993; Jones et al., 1994), as did studies undertaken in Sweden and Finland (Makikyro et al., 1997; Wicks et al., 2005).

It is possible that the social class at birth is a confounder in the relationship between either known or unknown risk factors for psychosis. There are a number of established factors that are associated with an increased risk of psychosis, namely, obstetric complications, cannabis, prenatal infection, prenatal stress, childhood trauma and increased paternal age (Clarke, Kelleher, Clancy, & Cannon, 2012; Torrey et al., 2009). These risk factors may be distributed unequally among the social classes, for example, exposure to infections, and may be more common in individuals in the lower social classes (Eaton & Harrison, 2001). Increased paternal age, especially in fathers greater than 55 years of age, is a risk factor for the offspring developing schizophrenia (Torrey et al., 2009). It is possible that these older fathers are represented more in the higher social classes, as they would have had longer in their respective careers to advance and achieve promotions. Therefore, it is plausible that the studies examining social class as a risk factor for psychosis actually report the net result of the interaction between all of these effects. While we examined for any difference in the distribution of these factors among cases in post hoc analysis, we did not have this information for controls. Further research controlling for all these factors could help disentangle the possible confounding.

Another potential explanation for the study findings is that children from families of higher social class may have higher expectations placed upon themselves, both by themselves and by their families. In addition to the decline in social, educational and occupational functioning that can occur in the prodrome of the disorder (Jones et al., 1993), the potential stress associated with not achieving these expectations could further increase the risk of transition to a psychotic disorder. The finding that individuals from a higher social class at birth have a younger age of presentation with an FEP lends support to this theory (Mulvany et al., 2001). To our knowledge, this hypothesis has not been tested; however, it is one possible explanation for the over-representation of individuals from a higher social class at birth in the cohort with an FEP.

Even within this study there was some variation in the findings, as no difference in the social class at birth was found between cases and the general population born between 1981 and 1990. There are a number of possible explanations for these findings; first, it could reflect the true relationship between social class at birth and the risk of psychosis, in that for those born in this decade, there was no increased risk of psychosis according to an individual’s social class at birth. Second, the lack of an association could be a result of insufficient power when the cohort was divided into groups according to the census period in which they were born. Alternatively, the influence of social class on the risk of psychosis could have changed over time. A possible temporal trend in the association between social class and psychotic disorders was suggested in a recent systematic review, as studies conducted post 2001 found an association between a lower social class at birth and a higher risk of psychosis (Kwok, 2014). Such a temporal trend is not inconceivable, as an association between social class and atopic medical disorders such as asthma and allergic rhinitis has already been observed (Braback, Hjern, & Rasmussen, 2005).

Finally, it has been established that there is a higher incidence of psychosis in neighbourhoods that have higher levels of social deprivation (Kirkbride, Jones, Ullrich, & Coid, 2012), and this finding is consistent with the social drift that is established with the psychotic disorders (Turner & Wagenfeld, 1967; Wiersma et al., 1983). However, less is known about when these individuals actually drift into these neighbourhoods and at what stage of the psychotic disorder do the environmental factors have the strongest influence. These environmental risk factors, such as social class and social deprivation, may seem non-specific, and at this stage, it may be difficult to envisage how this research may result in either treatment or preventative strategies. However, a good example of an environmental risk factor that may lead to preventative strategies is the established finding that individuals born in winter and spring have a higher risk of developing schizophrenia, and this has led to the exploration of vitamin D deficiency as a possible modifiable risk factor (McGrath, Brown, & St Claire, 2011; Takei, Sham, O’Callaghan, Glover, & Murray, 1995).

This study has demonstrated that overall individuals with a psychotic disorder are over-represented in families of higher social class at birth; however, this effect may fluctuate temporarily or even be in decline. Given the temporal trends observed in other disorders, further longitudinal evaluation of the effect of social class of origin may open new avenues of exploration in the aetiology of psychosis.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Braback

Hjern

Rasmussen

(2005). Social class in asthma and allergic rhinitis: A national cohort study over three decades. European Respiratory Journal, 26, 1064–1068.

Cantor-Graae

(2007). The contribution of social factors to the development of schizophrenia: A review of recent findings. Canadian Journal of Psychiatry, 52, 277–286.

Castle

D. J.

Scott

Wessely

Murray

R. M.

(1993). Does social deprivation during gestation and early life predispose to later schizophrenia? Social Psychiatry and Psychiatric Epidemiology, 28, 1–4.

Central Statistics Office. (2006). Census 2006 results: Appendix 7: Census of Ireland classification of occupations. Retrieved from http://www.cso.ie/en/media/csoie/census/census2006results/volume8/appendix_2-7_definitions.pdf

Clarke

M. C.

Kelleher

Clancy

Cannon

(2012). Predicting risk and the emergence of schizophrenia. Psychiatric Clinics of North America, 35, 585–612.

Corcoran

Perrin

Harlap

Deutsch

Fennig

Manor

... Susser

(2009). Effect of socioeconomic status and parents’ education at birth on risk of schizophrenia in offspring. Social Psychiatry and Psychiatric Epidemiology, 44, 265–271.

Done

D. J.

Crow

T. J.

Johnstone

E. C.

Sacker

(1994). Childhood antecedents of schizophrenia and affective illness: Social adjustment at ages 7 and 11. British Medical Journal, 309, 699–703.

Eaton

Harrison

(2001). Life chances, life planning and schizophrenia. International Journal of Mental Health, 30, 58–81.

Faris

R. E.

Dunham

H. W.

(1939). Mental disorders in urban areas. Chicago, IL: University of Chicago Press.

10.

Fennig

Craig

Lavelle

Kovasznay

Bromet

E. J.

(1994). Best-estimate versus structured interview-based diagnosis in first-admission psychosis. Comprehensive Psychiatry, 35, 341–348.

11.

First

M. B.

Spitzer

R. L.

Gibbon

Williams

J. B.

(2002). Structured clinical interview for DSM-IV axis 1 disorders, Patient Edition. New York: New York State Psychiatric Institute.

12.

Goldberg

Morrison

(1963). Schizophrenia and social class. British Journal of Psychiatry, 109, 785–802.

13.

Hare

E. H.

Price

J. S.

Slater

(1972). Parental social class in psychiatric patients. British Journal of Psychiatry, 121, 515–534.

14.

Harrison

Gunnell

Glazebrook

Page

Kwiecinski

(2001). Association between schizophrenia and social inequality at birth: Case-control study. British Journal of Psychiatry, 179, 346–350.

15.

Jones

P. B.

Bebbington

Foerster

Lewis

S. W.

Murray

R. M.

Russell

... Wilkins

(1993). Premorbid social underachievement in schizophrenia. Results from the Camberwell Collaborative Psychosis Study. British Journal of Psychiatry, 162, 65–71.

16.

Jones

Rodgers

Murray

Marmot

(1994). Child development risk factors for adult schizophrenia in the British 1946 birth cohort. The Lancet, 344, 1398–1402.

17.

Kirkbride

J. B.

Jones

P. B.

Ullrich

Coid

J. W.

(2012). Social deprivation, inequality, and the neighborhood-level incidence of psychotic syndromes in East London. Schizophrenia Bulletin, 37, 1413–1425.

18.

Kwok

(2014). Is there evidence that social class at birth increases risk of psychosis? A systematic review. International Journal of Social Psychiatry, 60, 801–808.

19.

Lewis

S. W.

Owen

M. J.

Murray

R. M.

(1989). Obstetric complications and schizophrenia: Methodology and mechanisms. In Schulz

Tamming

(Eds.), Schizophrenia: Scientific progress (pp. 56–68). New York, NY: Oxford University Press.

20.

Makikyro

Isohanni

Moring

Oja

Hakko

Jones

Rantakallio

(1997). Is a child’s risk of early onset schizophrenia increased in the highest social class? Schizophrenia Research, 23, 245–252.

21.

McGrath

Brown

St Claire

(2011). Prevention and schizophrenia – The role of dietary factors. Schizophrenia Bulletin, 37, 272–283.

22.

Mulvany

O’Callaghan

Takei

Byrne

Fearon

Larkin

(2001). Effect of social class at birth on risk and presentation of schizophrenia: Case-control study. British Medical Journal, 323, 1398–1401.

23.

Muntaner

Wolyniec

McGrath

Pulver

A. E.

(1994). Psychotic inpatients’ social class and their first admission to state or private psychiatric Baltimore hospitals. American Journal of Public Health, 84, 287–289.

24.

O’Donoghue

Lyne

Renwick

Madigan

Kinsella

Clarke

... O’Callaghan

(2012). A descriptive study of ‘non-cases’ and referral rates to an early intervention for psychosis service. Early Intervention Psychiatry, 6, 276–282.

25.

Schlesselman

(1982). Case-control studies: Design, conduct, analysis. New York, NY: Oxford University Press.

26.

Skre

Onstad

Torgersen

Kringlen

(1991). High interrater reliability for the Structured Clinical Interview for DSM-III-R Axis I (SCID-I). Acta Psychiatrica Scandinavica, 84, 167–173.

27.

Takei

Sham

P. C.

O’Callaghan

Glover

Murray

R. M.

(1995). Schizophrenia: Increased risk associated with winter and city birth – A case-control study in 12 regions within England and Wales. Journal of Epidemiology and Community Health, 49, 106–107.

28.

Torrey

E. F.

Buka

Cannon

T. D.

Goldstein

J. M.

Seidman

L. J.

Liu

... Yolken

R. H.

(2009). Paternal age as a risk factor for schizophrenia: How important is it? Schizophrenia Research, 114, 1–5.

29.

Turner

R. J.

Wagenfeld

M. O.

(1967). Occupational mobility and schizophrenia: An assessment of the social causation and social selection hypotheses. American Sociological Review, 32, 104–113.

30.

Van Os

Kapur

(2009). Schizophrenia. The Lancet, 374, 635–645.

31.

Werner

Malaspina

Rabinowitz

(2007). Socioeconomic status at birth is associated with risk of schizophrenia: Population-based multilevel study. Schizophrenia Bulletin, 33, 1373–1378.

32.

Wicks

Hjern

Gunnell

Lewis

Dalman

(2005). Social adversity in childhood and the risk of developing psychosis: A national cohort study. American Journal of Psychiatry, 162, 1652–1657.

33.

Wiersma

Giel

De Jong

Slooff

C. J.

(1983). Social class and schizophrenia in a Dutch cohort. Psychological Medicine, 13, 141–150.