Disentangling genetic and social confounding in schizophrenia

Disentangling genetic and social confounding in schizophrenia | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Disentangling genetic and social confounding in schizophrenia Laurie Haig, Isabella Badini, Joseph Hayes, Jennifer Dykxhoorn, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9633381/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Schizophrenia is a severe psychiatric disorder with known genetic and socioenvironmental antecedents. We tested whether genetic liability to schizophrenia is associated with socioenvironmental exposures for schizophrenia at birth which could indicate confounding. Data were drawn from the first wave of the 1958 British Birth Cohort, the National Child Development Study (NCDS). We selected 13 socioenvironmental exposures which have been linked to increased schizophrenia risk, including measures of social class, parental demographics, and pre- and perinatal exposures. We applied mutually adjusted regression to test the association between these socioenvironmental exposures and child’s polygenic susceptibility to schizophrenia measured via polygenic index (PGI). The final sample included 6,274 participants. Genetic liability to schizophrenia was not randomly distributed and associated with some socioenvironmental exposures at birth. Children whose mothers smoked more prior to pregnancy had a 0.008 standard deviation (SD) higher schizophrenia PGI per unit increase in cigarettes smoked (95%CI:0.002 to 0.014). Children born to older mothers had lower PGI, with a 0.008SD decrease per additional year of maternal age (95%CI:0.000 to 0.016). Compared with children born in the southeast of England, those born in the north, south, southwest, and Scotland had lower PGI. Children with higher values for the genetic principal component capturing north versus south geographical clustering had higher PGI (0.106 SD per unit; 95%CI:0.076 to 0.136). Children did not differ meaningfully by other socioenvironmental exposures. This has implications for causal inference and Mendelian randomisation studies of schizophrenia, as they may be affected by confounding. Biological sciences/Genetics Health sciences/Diseases/Psychiatric disorders/Schizophrenia Figures Figure 1 Figure 2 Figure 3 Introduction Schizophrenia is a severe psychiatric disorder affecting over 23 million people worldwide [ 1 ]. It is characterised by positive symptoms, including delusions and hallucinations, as well as negative symptoms, such as apathy, avolition and social withdrawal. Schizophrenia is associated with increased psychiatric and physical comorbidity, including cardiovascular disease and other lifestyle risks, contributing to high disease burden and premature mortality [ 1 – 3 ]. Schizophrenia has a substantial genetic component. Twin and family studies have demonstrated higher concordance in monozygotic twins than in other first-degree relatives, with an estimated heritability of 80% [ 4 ]. More recently, genome-wide association studies (GWAS) have found 287 distinct loci associated with schizophrenia across 120 genes [ 5 ]. Several social and environmental exposures have been associated with schizophrenia onset, including familial socioeconomic position (SEP), parental education, maternal age, region of birth, neighbourhood deprivation, and perinatal factors such as maternal infection during pregnancy, parity (number of previous births), and maternal smoking and alcohol consumption during pregnancy [ 6 – 11 ]. However, whether these associations are causal remains unclear. For example, exposure to environmental stressors like high population density and neighbourhood deprivation may cause schizophrenia (social causation), or individuals who are at higher risk for developing schizophrenia may be more likely to move to or remain in such areas (social drift/social selection hypothesis) [ 11 ]. Mendelian randomisation studies use genetic variants identified in GWAS to explore causal relationships between exposures and outcomes [ 12 , 13 ]. Several Mendelian randomisation studies have estimated the effects of genetic liability to schizophrenia and other outcomes, including smoking [ 14 ] cannabis use [ 15 , 16 ], cardiovascular risk [ 17 ], inflammation [ 18 ], and population density [ 19 ]. However, all these studies have used samples of unrelated individuals, making it challenging to determine whether these effects are due to genetic liability to schizophrenia or are confounded by population stratification, dynastic effects, or assortative mating [ 20 ]. Mendelian randomisation is an application of instrumental variable analysis, and thus depends on the three core instrumental variable assumptions [ 21 ]. These include the relevance assumption: that genetic variants are robustly associated with the exposure of interest [ 21 ]; exclusion assumption: that the genetic variants are not related to the outcome of interest other than via the exposure [ 21 ]; and independence assumption: that genetic instruments are independent of other factors that confound the exposure-outcome relationship [ 22 ]. If these three assumptions hold, then the estimates can provide reliable evidence of the causal effect of the exposure on the outcome (Fig. 1 A). However, population-based estimates of the associations between genetic variants and the outcome can also arise because they share common causes that confound the exposure-outcome relationship (Figs. 1 B and 1 C). If genetic variants are not randomly distributed across the population, they may become correlated with environmental, social, or demographic factors that influence both the genetic variants and the outcome, thereby violating the independence assumption (Fig. 1 D). For example, if schizophrenia-associated variants correlate with a third variable (e.g. social class) (arrow 1 in Fig. 1 D), and are also associated with the outcome (e.g. smoking) (arrow 2), this creates a non-causal pathway between the genetic instrument and the outcome, making it impossible to obtain an unbiased estimate of the association between the exposure and outcome (e.g. genetic liability to schizophrenia and genetic liability to smoking). [Figure 1 here] Previous studies have attempted to disentangle mechanisms of social causation versus social selection using genetically informed designs. A UK study using ALSPAC data found that schizophrenia polygenic index (PGI) did not predict birth into more densely populated neighbourhoods, although they found that children born into more deprived and socially fragmented neighbourhoods had higher PGI for schizophrenia [ 23 ]. Similarly, a Danish study found that individuals with higher schizophrenia PGI were more likely to reside in the capital at age 15, but not at birth, providing further support for the social drift hypothesis [ 24 ]. An analysis of 4 cohorts in the UK, Australia, and the Netherlands, found that individuals with high genetic risk for schizophrenia lived in more densely populated areas at a greater than chance level and lived in postcodes with higher population density even after controlling for SEP [ 19 ]. Their analysis also included a two-sample Mendelian randomisation, which suggested that genetic risk for schizophrenia had a causal effect on likelihood of living in urban areas [ 19 ]. However, the authors note that these results could be due to confounding in previous generations, which may make individuals both more likely to have a high genetic risk for schizophrenia and more likely to live in urban areas [ 19 ]. It is plausible that individuals with a higher genetic risk for schizophrenia, such as the parents and ancestors of individuals recruited in the studies mentioned above, may have certain traits that make them more likely to move to more urban and more deprived areas [ 25 ]. If parents with higher genetic liability for schizophrenia also tend, on average, to experience greater socioeconomic disadvantage, their children may be more likely to be exposed to social and environmental conditions that are associated with increased schizophrenia risk. This is in line with intergenerational (social) selection, whereby families with higher genetic liability to schizophrenia are more likely to drift to or remain in more socioeconomically disadvantaged neighbourhoods over time [ 11 ]. The high heritability of schizophrenia, combined with the increased likelihood of individuals with high genetic risk for schizophrenia to live in more urban and deprived areas, makes it difficult to determine true causal pathways generating these associations. For instance, any association found between child PGI for schizophrenia and an exposure of interest (e.g. smoking) could result from the association between the parental genotype (or related traits) and the parental environment, resulting in an increased likelihood of the child being born into an environment with more adverse socioenvironmental exposures, rather than as a direct result of the child’s genetics (Fig. 1 B). Consequently, relationships between socioenvironmental exposures and schizophrenia risk, which may have been previously interpreted as causal in both Mendelian randomisation and social determinants of health research, may in fact be the result of confounding. Here, we assess the correlation between genetic markers and socioenvironmental exposures around the time of birth to quantify the degree of confounding attributable to the familial and social factors to provide further insight into the aetiology of these associations. Our primary hypothesis is that genetic liability to schizophrenia is associated with socioenvironmental exposures at birth. Materials and methods Data Following a pre-registered analysis plan ( 10.17605/OSF.IO/NJZAU ; see Supplementary Table 1 for deviations [ 26 ]), we used data from the 1958 British birth cohort, the National Child Development Study (NCDS). NCDS is a nationally representative birth cohort that follows a sample of 17 416 people born in the UK in a single week in 1958 (98% of all births in Great Britain in that week) [ 27 , 28 ]. This cohort has been followed up across multiple time points to monitor their educational, physical, and social development, health, and inequalities throughout childhood, adolescence, and adulthood. To date, 12 sweeps have been completed, the most recent in 2000 when the cohort was age 62 [ 27 ]. The NCDS cohort was genotyped using whole blood samples collected at age 44 [ 29 ]. Genotyping and quality control procedures have been described elsewhere [ 29 ]. Genetic data were available for 6324 participants (36%). After removing multiple births and participants without exposure data, the final sample was 6274 participants (Fig. 2 ). [Figure 2 here] Exposures We selected socioenvironmental exposures for schizophrenia based on the theoretical knowledge of important pre- and perinatal socioeconomic exposures previously linked to schizophrenia[ 6 ] and data availability within our sample. Socioenvironmental exposures were measured using maternal interviews during the first wave of NCDS. We included three measures of social class: social class of mother’s father (social class I [highest] – V [lowest], other, no father), social class of mother’s husband (social class I-V, other, no husband) and mother’s paid job during pregnancy (social class I-V, no job during pregnancy; Supplementary Tables 2–3) [ 30 ]. We measured parental demographics at the time of the child’s birth (maternal age, father’s age, child region of birth [defined by 11 NCDS regions, Supplementary Fig. 1], mother education [did mother stay past school leaving age: yes or no]), and pre- and perinatal exposures (parity, antenatal care [midpoint of category for week of first visit, for example 1st to 3rd week was recoded as 2, and midpoint of category for total number of visits, for example 5–9 total antenatal visits was recoded to 7], infection during pregnancy [yes or no infection – influenza or German measles], smoking prior to and during pregnancy [midpoint of category for number of cigarettes smoked, for example, smoking 1 to 4 cigarettes daily was recoded to 2.5]). When markers were coded categorically or as binary, the highest or most advantaged category was used as the reference (e.g. social class I was used as the reference for all social class variables). We included offspring sex and principal components of ancestry as covariates in all analyses to control for confounding by population stratification. Outcome We assessed genetic liability to schizophrenia using the PGI based on the 2022 schizophrenia genome-wide association study (GWAS) [ 5 ]. Schizophrenia PGI were calculated using LDpred-2 [ 31 ]. The PGIs were standardised (mean = 0, standard deviation = 1). A subset of the NCDS genotyped sample (n = 3000) was included as controls in this GWAS, which introduced the risk of inflated PGI-trait associations due to sample overlap, however the low population prevalence of schizophrenia minimises the risk of inflation [ 32 ]. We filtered variants with an imputation INFO score > 95% and an allele frequency > 1% to limit analyses to very well imputed variants. Analysis We estimated the joint associations of PGI with the exposures at birth using a mutually adjusted robust linear regression in R (version 4.2.3) [ 33 ]. Bivariate models of PGI and each socioenvironmental exposure were run as a sensitivity analysis to investigate the independent association of the schizophrenia PGI with each exposure. Following an exploration of missingness patterns, we used mice (version 3.18.0) in R to impute missing data [ 34 ]. We calculated the fraction of missing data for each parameter to inform the number of imputations required. Code files have been uploaded to Github: https://github.com/LaurieHaig/NCDS_paper Results The final sample included 6 274 participants with genetic data. The average maternal age was 27.41 years, while fathers were slightly older at an average of 30.38 years. Participants were spread across regions (17% from the southeast of England, including London) and social classes (59% of husbands and 47% of fathers were in social class III, while 62% of mothers had no job during pregnancy) (Table 1 ; for the non-imputed data, see Supplementary Table 4). Table 1 Sample characteristics and bivariate associations with schizophrenia PGI Predictor n (%) Bivariate β (95% CI), p-value Schizophrenia PGI (mean, SD) 4.62 (1.01) Child sex Male 49.8% Female 50.2% 0.033 (-0.017, 0.083), p = 0.194 Social Class mother’s husband Social class I (ref) 4.6% Social class II 13.8% -0.088 (-0.225, 0.049), p = 0.208 Social class III 59.5% -0.034 (-0.157, 0.089), p = 0.584 Social class IV 11.9% 0.006 (-0.135, 0.146), p = 0.937 Social class V 7.7% 0.022 (-0.128, 0.172), p = 0.773 Other 0.2% 0.390 (-0.285, 1.065), p = 0.257 No husband 2.3% 0.095 (-0.112, 0.302), p = 0.369 Social Class mother’s father Social class I (ref) 2.7% Social class II 14.1% -0.094 (-0.267, 0.079), p = 0.286 Social class III 47.2% -0.099 (-0.262, 0.063), p = 0.230 Social class IV 13.6% -0.079 (-0.252, 0.093), p = 0.366 Social class V 12.2% -0.100 (-0.274, 0.074), p = 0.259 Other 2.1% -0.017 (-0.256, 0.222), p = 0.890 No father 8.0% -0.104 (-0.284, 0.076), p = 0.258 Mum’s paid job during pregnancy Social class I (ref) 1.7% Social class II 0.7% 0.353 (-0.018, 0.724), p = 0.062 Social class III 6.3% -0.027 (-0.244, 0.190), p = 0.806 Social class IV 22.2% 0.055 (-0.145, 0.255), p = 0.591 Social class V 7.1% 0.051 (-0.163, 0.265), p = 0.639 No job 62.0% 0.032 (-0.162, 0.226), p = 0.749 Maternal smoking prior to pregnancy 5.78 (7.81) 0.006 (0.003, 0.009), p = 0.001 Maternal smoking during pregnancy 4.31 (7.09) 0.005 (0.001, 0.008), p = 0.010 Maternal age at last birthday 27.42 (5.55) -0.004 (-0.009, 0.001), p = 0.094 Was mum at school past leaving age? Stayed past leaving age (ref) 26.4% Did not stay past leaving age 73.6% -0.016 (-0.073, 0.041), p = 0.586 Parity 1.23 (1.45) 0.009 (-0.009, 0.026), p = 0.340 Week of first antenatal visit 16.40 (6.51) 0.005 (0.001, 0.009), p = 0.012 Total antenatal visits 12.28 (5.45) -0.006 (-0.010, -0.001), p = 0.018 Maternal infection during pregnancy No infection during pregnancy (ref) 87.6% Yes infection during pregnancy 12.4% 0.037 (-0.041, 0.115), p = 0.349 Husband’s age in years 30.34 (6.30) -0.001 (-0.005, 0.003), p = 0.505 Region of birth Southeast including London (ref) 16.8% East & West Riding 8.4% -0.087 (-0.192, 0.018), p = 0.106 East 7.4% -0.084 (-0.194, 0.025), p = 0.132 Midlands 9.9% -0.006 (-0.106, 0.093), p = 0.898 North 7.6% -0.166 (-0.274, -0.057), p = 0.003 North Midlands 7.9% -0.058 (-0.166, 0.049), p = 0.287 North West 11.7% 0.034 (-0.061, 0.129), p = 0.482 Scotland 13.9% 0.000 (-0.090, 0.091), p = 0.995 South 5.7% -0.222 (-0.342, -0.101), p = 0.000 South West 5.6% -0.128 (-0.249, -0.007), p = 0.038 Wales 5.1% 0.150 (0.024, 0.277), p = 0.020 Maternal smoking, younger maternal age and being born in the southeast of England were associated with higher schizophrenia PGI. Participants of mothers who smoked more prior to pregnancy had, on average, a 0.008 (95% CI: 0.002 to 0.014) standard deviation higher PGI per additional cigarette smoked (Table 2 , Fig. 3 ). On average, a child of a 30-year-old mother had a PGI 0.08 standard deviations lower (95% CI: 0.00 to 0.16) than a child of a 20-year-old mother. Children born in the north (β= -0.214 [95% CI: -0.324, -0.103]), south (β= -0.198 [95% CI -0.318, -0.078]), southwest (β= -0.125 [95% CI: -0.246, -0.005]) and Scotland (β= -0.148 [95% CI: -0.249, -0.048]) all had significantly lower PGI compared to children born in the southeast of England (including London). Furthermore, the PGIs varied across major geographic gradients. A one-standard-deviation increase in the first genetic principal component of ancestry was associated with a 0.106 (95% CI: 0.076, 0.136) standard deviation increase in schizophrenia PGI. We found little evidence of differences in PGI by social class, father’s age, maternal education, parity, antenatal care, or maternal infection. Table 2 Association between socioenvironmental exposures and schizophrenia PGI outcome Predictor β SE 95% CI p-value Intercept 0.237 0.164 [-0.084, 0.558] 0.148 Social class mother’s husband Social class II -0.056 0.070 [-0.193, 0.081] 0.422 Social class III -0.018 0.066 [-0.147, 0.112] 0.791 Social class IV 0.028 0.076 [-0.12, 0.177] 0.710 Social class V -0.011 0.081 [-0.169, 0.148] 0.894 Other 0.439 0.340 [-0.228, 1.107] 0.197 No husband 0.041 0.110 [-0.175, 0.257] 0.707 Social class mother’s father Social class II -0.092 0.088 [-0.265, 0.081] 0.297 Social class III -0.100 0.086 [-0.269, 0.068] 0.243 Social class IV -0.097 0.092 [-0.277, 0.083] 0.290 Social class V -0.133 0.093 [-0.316, 0.05] 0.155 Other -0.037 0.123 [-0.279, 0.205] 0.763 No father -0.133 0.094 [-0.318, 0.051] 0.157 Mum’s paid job during pregnancy Social class II 0.345 0.188 [-0.025, 0.714] 0.067 Social class III -0.016 0.112 [-0.236, 0.203] 0.883 Social class IV 0.068 0.103 [-0.135, 0.27] 0.512 Social class V 0.074 0.111 [-0.143, 0.291] 0.506 No job 0.061 0.100 [-0.136, 0.257] 0.547 Maternal smoking prior to pregnancy 0.008 0.003 [0.002, 0.014] 0.010 Maternal smoking during pregnancy -0.004 0.003 [-0.011, 0.002] 0.207 Maternal age at last birthday -0.008 0.004 [-0.016, 0] 0.040 Was mum at school past leaving age? Did not stay past leaving age -0.025 0.033 [-0.09, 0.041] 0.458 Parity 0.013 0.012 [-0.01, 0.036] 0.267 Week of mother’s first antenatal visit 0.002 0.002 [-0.002, 0.006] 0.372 Total number of antenatal visits -0.003 0.002 [-0.008, 0.002] 0.233 Yes infection during pregnancy 0.047 0.039 [-0.031, 0.124] 0.238 Husband’s age in years 0.002 0.003 [-0.004, 0.009] 0.506 Region of birth East & West Riding -0.062 0.054 [-0.169, 0.045] 0.254 East -0.045 0.056 [-0.155, 0.064] 0.418 Midlands 0.001 0.051 [-0.1, 0.101] 0.989 North -0.214 0.056 [-0.324, -0.103] 0.000 North Midlands -0.027 0.055 [-0.135, 0.081] 0.629 North West -0.014 0.050 [-0.112, 0.084] 0.781 Scotland -0.148 0.051 [-0.249, -0.048] 0.004 South -0.198 0.061 [-0.318, -0.078] 0.001 South West -0.125 0.061 [-0.246, -0.005] 0.041 Wales 0.011 0.074 [-0.135, 0.156] 0.887 Child sex 0.028 0.025 [-0.021, 0.078] 0.261 PC1 0.106 0.015 [0.076, 0.136] 0.000 PC2 0.003 0.013 [-0.022, 0.028] 0.801 PC3 -0.022 0.015 [-0.052, 0.008] 0.148 PC4 -0.005 0.013 [-0.031, 0.02] 0.688 PC5 0.010 0.013 [-0.015, 0.035] 0.441 PC6 -0.032 0.013 [-0.057, -0.008] 0.010 PC7 -0.005 0.013 [-0.03, 0.02] 0.688 PC8 -0.030 0.013 [-0.055, -0.005] 0.017 PC9 0.007 0.013 [-0.017, 0.032] 0.559 PC10 0.012 0.013 [-0.013, 0.037] 0.348 PC11 0.007 0.013 [-0.018, 0.032] 0.580 PC12 0.003 0.013 [-0.022, 0.027] 0.834 PC13 -0.020 0.013 [-0.045, 0.004] 0.107 PC14 0.006 0.013 [-0.019, 0.03] 0.655 PC15 0.016 0.013 [-0.008, 0.041] 0.191 PC16 0.033 0.013 [0.008, 0.058] 0.009 PC17 0.010 0.013 [-0.015, 0.035] 0.426 PC18 0.021 0.013 [-0.004, 0.046] 0.096 PC19 0.021 0.013 [-0.003, 0.046] 0.091 PC20 0.009 0.013 [-0.015, 0.034] 0.466 Values are pooled estimates from multiple imputation; CI = confidence interval. [Table 2 here; Fig. 3 ] In a sensitivity analysis, we estimated bivariate associations between the schizophrenia PGI and each exposure to assess the independent association of the schizophrenia PGI with each socioenvironmental exposure (Table 1 ). Overall, the bivariate associations were consistent with the mutually adjusted model (Table 2 ). Children whose mothers smoked prior to (β = 0.006 [95% CI: 0.003, 0.009]) and during pregnancy (β = 0.005 [95% CI: 0.001, 0.008]) had higher schizophrenia PGI, while this association was only seen for smoking prior to pregnancy in the mutually adjusted model. Children born in the south, north, and southwest of England had lower PGI than those born in the southeast, in both the bivariate correlations and the mutually adjusted model. Children born in Scotland had lower PGI in the mutually adjusted model (β= -0.148 [95% CI: -0.249, -0.048]) but not the bivariate model [β = 0.000 (95%CI: -0.090, 0.091)], while children born in Wales had higher PGI in the bivariate model [β = 0.150 (95%CI: 0.024, 0.277)] but not the mutually adjusted model [β = 0.011 (95%CI: -0.135, 0.156)]. Children born to younger mothers had higher PGI in both the mutually adjusted and bivariate models. In the bivariate model, children whose mothers’ first antenatal visit was later in their pregnancy had higher PGI, with each one-week delay associated with a 0.005 SD higher PGI (95% CI: 0.001, 0.009) and children whose mothers attended more antenatal visits had on average 0.006 SD lower PGI (95% CI -0.010, -0.001) per additional visit, however these associations were not detected in the mutually adjusted model. Discussion We found that the schizophrenia PGI was associated with several socioenvironmental exposures at birth. This suggests that children born in specific regions, whose mothers smoked more heavily before pregnancy or were younger, were also likely to have a higher genetic liability to schizophrenia. We found clear regional differences in the distribution of offspring schizophrenia PGI at birth, suggesting that ancestors with higher schizophrenia PGI were more likely to move to or remain in certain regions within the UK. The reference category for region of birth was the southeast of England (including London), and our results suggest that schizophrenia PGI was, on average, higher here than in nearly all other regions. This aligns with previous studies, which have reported genetic risk for schizophrenia to be higher in more urban and densely populated areas [ 19 , 23 , 24 ]. This also aligns with previously reported regional differences in schizophrenia incidence across the UK, as incidence was generally higher in more urban areas [ 35 , 36 ]. Similarly, urban versus rural birth or birth into more deprived neighbourhoods has been associated with a higher incidence of schizophrenia [ 23 , 37 – 41 ]. The first genetic principal component of ancestry, which generally captures north versus south geographical clustering in ancestral gradients [ 42 ] was also associated with PGI at birth, providing further evidence for distinct regional differences. Additionally, we found that children of mothers who smoked more prior to pregnancy had a higher schizophrenia PGI compared to children of mothers who smoked less. Recent studies have found smoking during pregnancy to be more common among mothers in routine and manual occupations, and smoking cessation to be lower among more disadvantaged pregnant women [ 43 ]. Smoking rates today tend to be higher among those with lower socioeconomic position [ 43 ], however, in the 1950s and 60s, smoking was more common among UK women of all social classes [ 44 , 45 ], and in our sample, around 30–50% of women smoked within each social class (Supplementary Table 5). Furthermore, we found little evidence of differences in PGI by social class, or of any other socially patterned variables such as education, parity, or maternal infection, in either the mutually adjusted or bivariate models, though there was some evidence for differences in PGI by antenatal care in the bivariate models only. We did, however, find that children born to younger mothers had a higher PGI compared to those born to older mothers. Overall, these results provide evidence from a nationally representative sample that schizophrenia PGI was not randomly distributed in the population but rather varied by UK region of birth, principal components related to major ancestral geographic gradients, and maternal smoking. This suggests that associations between the schizophrenia PGI reported in Mendelian randomization studies may reflect these confounding factors (Figs. 1 B and 1 D). Our results may also have implications for studies investigating the social determinants of schizophrenia, as they suggest differences in incidence of schizophrenia by region and maternal smoking, which may be explained by differences in genetic liability. This also has consequences for molecular genetic research using population-based samples, as it highlights potential for confounding pathways if the outcome of interest differs by region or parental smoking patterns. Even a small bias in the association of the schizophrenia PGI and smoking could have a large bias on the estimate in Mendelian randomisation studies [ 20 ], suggesting previously reported bidirectional effects of genetic liability to schizophrenia on smoking may be explained by maternal smoking or regional differences [ 14 ]. However, future studies would benefit from family-based designs that include genetic information on both biological parents and their children or siblings to confirm this and more robustly test the causal relationships between genetic liability and socioenvironmental exposures by controlling for parental genotype [ 20 ]. Limitations NCDS is a population-representative birth-cohort dataset that enables us to investigate many socioenvironmental exposures at birth. However, because the first wave was conducted in 1958, data availability and the nature of relationships between socioenvironmental exposures may differ from those observed in a more modern context. As noted earlier, smoking patterns were different among women in the UK at this time, making our findings of the significance of smoking prior to pregnancy more difficult to interpret. Other exposures, such as antenatal care, level of education, and occupations (particularly for women and pregnant women), differ as well. Future studies should investigate these relationships in more recent population cohorts. Additionally, although the NCDS included nearly all births across Great Britain in a single week, the genetic data sample was limited to individuals of European ancestry and may have been affected by differential attrition, as samples were taken at age 44. Therefore, it is important that these associations be tested in more diverse ancestry cohorts with appropriate PGI. Furthermore, Choi et al. highlighted the risk of inflated PGI-trait associations due to sample overlap, which was present in our study. However, their simulations showed a reduced risk when overlap was limited to controls and to traits with lower population prevalence [ 46 ]. As schizophrenia is a binary trait with a low population prevalence (~ 1%) and the sample overlap with NCDS was only among GWAS controls (degree of overlap = 0.0784), the risk of inflation is low [ 46 ]. Conclusions Schizophrenia PGI was associated with pre-natal exposures in a population-based study. This demonstrates that genetic liability for schizophrenia is unlikely to be randomly distributed across the population and indicates a potential for confounding in studies using Mendelian randomisation and those investigating the social determinants of schizophrenia. Future research would benefit from family-based designs with genetic information from both biological parents to explore these associations after accounting for parental genetic liability. Declarations Acknowledgements: We thank the participants in the 1958 British Birth Cohort and their families for taking part in this study. Author contributions: LH, JD and NMD contributed to the conception of the work. All authors contributed to the analysis and interpretation of data for the work. All authors contributed to the drafting of the work, critical review and final approval of the version to be published. Conflict of Interest: JFH has received consultancy fees from Wellcome Trust, juli inc and Swiss Re. No other authors declare conflicts of interest. Funding support: LH is supported by the UCL-Birbeck MRC Doctoral Training Programme (MR/W006774/1). JFH is funded by UKRI grant MR/V023373/1, the University College London Hospitals NIHR Biomedical Research Centre and the NIHR North Thames Applied Research Collaboration. Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. Data Sharing Statement: Data from the 1958 British Birth Cohort is freely available to researchers in the UK and around the world. Deidentified data is available through public repositories (primarily the UK data service), and researchers can apply to access other data, including genetic data, directly from CLS (https://cls.ucl.ac.uk/data-access-training/data-access/). For the purpose of Open Access, the author has applied a CC BY licence to any Author Accepted Manuscript version arising from this submission. Supplementary information is available at Molecular Psychiatry’s website. References Solmi M, Seitidis G, Mavridis D, Correll CU, Dragioti E, Guimond S, et al. Incidence, prevalence, and global burden of schizophrenia - data, with critical appraisal, from the Global Burden of Disease (GBD) 2019. Molecular Psychiatry 2023 28:12. 2023;28:5319–5327. Correll CU, Solmi M, Croatto G, Schneider LK, Rohani-Montez SC, Fairley L, et al. Mortality in people with schizophrenia: a systematic review and meta-analysis of relative risk and aggravating or attenuating factors. World Psychiatry. 2022;21:248–271. Wahbeh MH, Avramopoulos D. Gene-Environment Interactions in Schizophrenia: A Literature Review. Genes 2021, Vol 12,. 2021;12. Hilker R, Helenius D, Fagerlund B, Skytthe A, Christensen K, Werge TM, et al. Heritability of Schizophrenia and Schizophrenia Spectrum Based on the Nationwide Danish Twin Register. Biol Psychiatry. 2018;83:492–498. Trubetskoy V, Pardiñas AF, Qi T, Panagiotaropoulou G, Awasthi S, Bigdeli TB, et al. Mapping genomic loci implicates genes and synaptic biology in schizophrenia. Nature. 2022;604:502–508. Davies C, Segre G, Estradé A, Radua J, De Micheli A, Provenzani U, et al. Prenatal and perinatal risk and protective factors for psychosis: a systematic review and meta-analysis. Lancet Psychiatry. 2020;7:399–410. Hakulinen C, Webb RT, Pedersen CB, Agerbo E, Mok PLH. Association Between Parental Income During Childhood and Risk of Schizophrenia Later in Life. JAMA Psychiatry. 2020;77:17. Schneider M, Müller CP, Knies AK. Low income and schizophrenia risk: A narrative review. Behavioural Brain Research. 2022;435:114047. Sariaslan A, Fazel S, D’Onofrio BM, Långström N, Larsson H, Bergen SE, et al. Schizophrenia and subsequent neighborhood deprivation: revisiting the social drift hypothesis using population, twin and molecular genetic data. Transl Psychiatry. 2016;6:e796–e796. Dohrenwend BP, Levav I, Shrout PE, Schwartz S, Naveh G, Link BG, et al. Socioeconomic Status and Psychiatric Disorders: The Causation-Selection Issue. Science (1979). 1992;255:946–952. Kirkbride JB, Anglin DM, Colman I, Dykxhoorn J, Jones PB, Patalay P, et al. The social determinants of mental health and disorder: evidence, prevention and recommendations. World Psychiatry. 2024;23:58–90. Smith GD, Ebrahim S. ‘Mendelian randomization’: can genetic epidemiology contribute to understanding environmental determinants of disease? Int J Epidemiol. 2003;32:1–22. Sanderson E, Levin MG, Walker V, Yuan S, Badini I, Dolce J, et al. Challenges and future directions for Mendelian randomization [Submitted for publication]. Nat Genet. Wootton RE, Richmond RC, Stuijfzand BG, Lawn RB, Sallis HM, Taylor GMJ, et al. Evidence for causal effects of lifetime smoking on risk for depression and schizophrenia: a Mendelian randomisation study. Psychol Med. 2020;50:2435–2443. Vaucher J, Keating BJ, Lasserre AM, Gan W, Lyall DM, Ward J, et al. Cannabis use and risk of schizophrenia: a Mendelian randomization study. Mol Psychiatry. 2017;23:1287. Gage SH, Jones HJ, Burgess S, Bowden J, Davey Smith G, Zammit S, et al. Assessing causality in associations between cannabis use and schizophrenia risk: a two-sample Mendelian randomization study. Psychol Med. 2017;47:971–980. Veeneman RR, Vermeulen JM, Abdellaoui A, Sanderson E, Wootton RE, Tadros R, et al. Exploring the Relationship Between Schizophrenia and Cardiovascular Disease: A Genetic Correlation and Multivariable Mendelian Randomization Study. Schizophr Bull. 2022;48:463–473. Hartwig FP, Borges MC, Horta BL, Bowden J, Davey Smith G. Inflammatory Biomarkers and Risk of Schizophrenia: A 2-Sample Mendelian Randomization Study. JAMA Psychiatry. 2017;74:1226–1233. Colodro-Conde L, Couvy-Duchesne B, Whitfield JB, Streit F, Gordon S, Kemper KE, et al. Association Between Population Density and Genetic Risk for Schizophrenia. JAMA Psychiatry. 2018;75:901. Davies NM, Hemani G, Neiderhiser JM, Martin HC, Mills MC, Visscher PM, et al. The importance of family-based sampling for biobanks. Nature. 2024;634:795–803. Walker V, Sanderson E, Levin MG, Damraurer SM, Feeney T, Davies NM. Reading and conducting instrumental variable studies: guide, glossary, and checklist. BMJ. 2024;387:78093. De Leeuw C, Savage J, Bucur IG, Heskes T, Posthuma D. Understanding the assumptions underlying Mendelian randomization. European Journal of Human Genetics. 2022;30:653–660. Solmi F, Lewis G, Zammit S, Kirkbride JB. Neighborhood Characteristics at Birth and Positive and Negative Psychotic Symptoms in Adolescence: Findings From the ALSPAC Birth Cohort. Schizophr Bull. 2020;46:581–591. Paksarian D, Trabjerg BB, Merikangas KR, Mors O, Børglum AD, Hougaard DM, et al. The role of genetic liability in the association of urbanicity at birth and during upbringing with schizophrenia in Denmark. Psychol Med. 2018;48:305–314. Gage SH, Davey Smith G, Munafò MR. Schizophrenia and neighbourhood deprivation. Transl Psychiatry. 2016;6:e979–e979. Willroth EC, Atherton OE. Best Laid Plans: A Guide to Reporting Preregistration Deviations. Adv Methods Pract Psychol Sci. 2024;7. Power C, Elliott J. Cohort profile: 1958 British birth cohort (National Child Development Study). Int J Epidemiol. 2006;35:34–41. CLS | 1958 National Child Development Study. https://cls.ucl.ac.uk/cls-studies/1958-national-child-development-study/ . Accessed 19 February 2026. Shireby G, Morris TT, Wong A, Chaturvedi N, Ploubidis GB, Fitzsimmons E, et al. Data Resource Profile: Genomic data in multiple British birth cohorts (1946–2001)—linkage with health, social, and environmental data from birth to old age. Int J Epidemiol. 2025;54:dyaf141. SOC 2000 - Office for National Statistics. https://www.ons.gov.uk/methodology/classificationsandstandards/standardoccupationalclassificationsoc/socarchive . Accessed 19 February 2026. LDpred2/README.md at main · AndreAllegrini/LDpred2 · GitHub. https://github.com/AndreAllegrini/LDpred2/blob/main/README.md . Accessed 29 January 2026. Wellcome Trust Case Control Consortium 2 - WTCCC. https://www.wtccc.org.uk/ccc2/index.html . Accessed 19 February 2026. R Core Team. R: A Language and Environment for Statistical Computing. 2023. van Buuren S, Groothuis-Oudshoorn K. mice: Multivariate imputation by chained equations in R. J Stat Softw. 2011;45:1–67. Kirkbride JB, Fearon P, Morgan C, Dazzan P, Morgan K, Tarrant J, et al. Heterogeneity in Incidence Rates of Schizophrenia and Other Psychotic Syndromes. Arch Gen Psychiatry. 2006;63:250. Grigoroglou C, Munford L, Webb RT, Kapur N, Ashcroft DM, Kontopantelis E. Prevalence of mental illness in primary care and its association with deprivation and social fragmentation at the small-area level in England. Psychol Med. 2020;50:293–302. Lewis G, Dykxhoorn J, Karlsson H, Khandaker GM, Lewis G, Dalman C, et al. Assessment of the Role of IQ in Associations Between Population Density and Deprivation and Nonaffective Psychosis. JAMA Psychiatry. 2020;77:729. Werner S, Malaspina D, Rabinowitz J. Socioeconomic Status at Birth Is Associated With Risk of Schizophrenia: Population-Based Multilevel Study. Schizophr Bull. 2006;33:1373–1378. March D, Hatch SL, Morgan C, Kirkbride JB, Bresnahan M, Fearon P, et al. Psychosis and Place. Epidemiol Rev. 2008;30:84–100. Mortensen PB, Pedersen CB, Westergaard T, Wohlfahrt J, Ewald H, Mors O, et al. Effects of Family History and Place and Season of Birth on the Risk of Schizophrenia. New England Journal of Medicine. 1999;340:603–608. Lewis G, David A, Andréasson S, Allebeck P. Schizophrenia and city life. Lancet. 1992;340:137–140. Novembre J, Johnson T, Bryc K, Kutalik Z, Boyko AR, Auton A, et al. Genes mirror geography within Europe. Nature 2008 456:7218. 2008;456:98–101. Hiscock R, Bauld L, Amos A, Fidler JA, Munafò M. Socioeconomic status and smoking: A review. Ann N Y Acad Sci. 2012;1248:107–123. Berridge V. Constructing women and smoking as a public health problem in Britain 1950-1990s. Gend Hist. 2001;13:328–348. Graham H. Smoking prevalence among women in the European Community 1950–1990. Soc Sci Med. 1996;43:243–254. Choi SW, Mak TSH, Hoggart CJ, O’Reilly PF. EraSOR: a software tool to eliminate inflation caused by sample overlap in polygenic score analyses. Gigascience. 2022;12. Additional Declarations The authors have declared there is NO conflict of interest to disclose Supplementary Files GeneticandsocialconfoundingschizophreniaSITable1.docx Supplementary Table 1: Preregistration Deviations Table GeneticandsocialconfoundingschizophreniaSITable2.docx Supplementary Table 2: Socioenvironmental exposure variables included in analysis GeneticandsocialconfoundingschizophreniaSITable3.docx Supplementary Table 3: Recoding of n540 (Mum's paid job during pregnancy) GeneticandsocialconfoundingschizophreniaSITable4.docx Supplementary Table 4: Sample characteristics of the non-imputed sample GeneticandsocialconfoundingschizophreniaSITable5.docx Supplementary Table 5. Proportion of mothers who smoked by mother's social class GeneticandsocialconfoundingschizophreniaSIFigure1.jpg Supplementary Figure 1: Map of NCDS regions Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9633381","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":636351981,"identity":"d8817e39-b343-41cc-9574-7b20e037d36d","order_by":0,"name":"Laurie Haig","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABEElEQVRIie2RMUvDQBiG3yPQLB9mPVH/Q0qgUAz+lhwFXYIIXTo4HBTaJdA1k/6Fin8g8oEuwToKXSqCk0PGgFLMmUWHazsK3sPdccd9D997HOBw/EH2zJKY4TWbimLAb29Cm9L5qYj88BTf7laltZpiinkHxS971WqE82DqP4HShZqNIaoaHFkVSqM8KTGUTBeQ5VLlDG8/A/fswdIIagKlmRJ0s6WaN8EOAI6tSvDeKGuoa6Oo9aNRvI+NijRdNNSc/QIFFUbpmC72YPJtiOReqhsm3GkaRDmLST8Lz6zPD4LBragvY3W1eHh5/aSTo9l0zM/16LirbU6LbCaFRXsQesOv/MJf7VTmcDgc/48vGLZOQX6E00cAAAAASUVORK5CYII=","orcid":"https://orcid.org/0009-0005-1834-6573","institution":"University College London","correspondingAuthor":true,"prefix":"","firstName":"Laurie","middleName":"","lastName":"Haig","suffix":""},{"id":636351982,"identity":"927c385c-e075-4977-9cfe-84a9b20018a3","order_by":1,"name":"Isabella Badini","email":"","orcid":"https://orcid.org/0000-0002-3959-5908","institution":"UCL","correspondingAuthor":false,"prefix":"","firstName":"Isabella","middleName":"","lastName":"Badini","suffix":""},{"id":636351983,"identity":"5e3b9667-223e-42ed-98a7-e274af53da49","order_by":2,"name":"Joseph Hayes","email":"","orcid":"https://orcid.org/0000-0003-2286-3862","institution":"UCL","correspondingAuthor":false,"prefix":"","firstName":"Joseph","middleName":"","lastName":"Hayes","suffix":""},{"id":636351984,"identity":"42b594ac-b36e-4c33-b80d-a654fc95e8bc","order_by":3,"name":"Jennifer Dykxhoorn","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Jennifer","middleName":"","lastName":"Dykxhoorn","suffix":""},{"id":636351985,"identity":"757b4f8f-ceb8-4eb1-8a88-13d9cb8eca8b","order_by":4,"name":"Neil Davies","email":"","orcid":"https://orcid.org/0000-0002-2460-0508","institution":"University College London","correspondingAuthor":false,"prefix":"","firstName":"Neil","middleName":"","lastName":"Davies","suffix":""}],"badges":[],"createdAt":"2026-05-06 16:35:33","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9633381/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9633381/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":109338159,"identity":"909695e7-1704-4d04-9095-9a91016b583d","added_by":"auto","created_at":"2026-05-15 17:50:15","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":954448,"visible":true,"origin":"","legend":"\u003cp\u003eDirected Acyclic Graphs\u003c/p\u003e","description":"","filename":"Figure1DAGsGeneticsocialconfounding.png","url":"https://assets-eu.researchsquare.com/files/rs-9633381/v1/741067d16c930acc6ccb8ce6.png"},{"id":109405652,"identity":"5df766d9-3c7f-4534-a4d3-9c515c9343d6","added_by":"auto","created_at":"2026-05-17 13:19:34","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":269645,"visible":true,"origin":"","legend":"\u003cp\u003eSTROBE Diagram\u003c/p\u003e","description":"","filename":"Figure2StrobeDiagramGeneticsocialconfounding.png","url":"https://assets-eu.researchsquare.com/files/rs-9633381/v1/2dd2a1829344accdd6ce2b05.png"},{"id":109405649,"identity":"c524b19a-e4bd-4224-93b9-9626407b30e4","added_by":"auto","created_at":"2026-05-17 13:19:33","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":651973,"visible":true,"origin":"","legend":"\u003cp\u003eCoefficient plot of imputed regression model result\u003c/p\u003e","description":"","filename":"OnlineFigure3coefficientplotGeneticandsocialconfounding.png","url":"https://assets-eu.researchsquare.com/files/rs-9633381/v1/dae204330096876d59eac210.png"},{"id":109406161,"identity":"426ea7e0-d69f-4808-b139-555beb34b44a","added_by":"auto","created_at":"2026-05-17 13:26:02","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2597092,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9633381/v1/88018a31-7f47-44dc-8339-d51879f85341.pdf"},{"id":109405852,"identity":"5d52e8e8-e57b-4e90-a265-6a11bdeb7fe9","added_by":"auto","created_at":"2026-05-17 13:20:40","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":27860,"visible":true,"origin":"","legend":"Supplementary Table 1: Preregistration Deviations Table","description":"","filename":"GeneticandsocialconfoundingschizophreniaSITable1.docx","url":"https://assets-eu.researchsquare.com/files/rs-9633381/v1/7c883695d10bab5656125fc6.docx"},{"id":109338161,"identity":"8035adf3-dd46-4068-979c-1d0197454283","added_by":"auto","created_at":"2026-05-15 17:50:15","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":15126,"visible":true,"origin":"","legend":"Supplementary Table 2: Socioenvironmental exposure variables included in analysis","description":"","filename":"GeneticandsocialconfoundingschizophreniaSITable2.docx","url":"https://assets-eu.researchsquare.com/files/rs-9633381/v1/925db2fef7b97392a2a4432d.docx"},{"id":109405622,"identity":"326bd207-288e-45a1-ac2d-8b502fa54c85","added_by":"auto","created_at":"2026-05-17 13:19:25","extension":"docx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":16353,"visible":true,"origin":"","legend":"Supplementary Table 3: Recoding of n540 (Mum's paid job during pregnancy)","description":"","filename":"GeneticandsocialconfoundingschizophreniaSITable3.docx","url":"https://assets-eu.researchsquare.com/files/rs-9633381/v1/52dcc0b35e86ec74f43bebc7.docx"},{"id":109338167,"identity":"7e266484-6b45-481a-89ce-8f3ce58fa590","added_by":"auto","created_at":"2026-05-15 17:50:15","extension":"docx","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":19849,"visible":true,"origin":"","legend":"Supplementary Table 4: Sample characteristics of the non-imputed sample","description":"","filename":"GeneticandsocialconfoundingschizophreniaSITable4.docx","url":"https://assets-eu.researchsquare.com/files/rs-9633381/v1/d0ec40eb39aee8bc6d435d4d.docx"},{"id":109338165,"identity":"a431934e-ad5a-4d2f-a575-f259d56b24ba","added_by":"auto","created_at":"2026-05-15 17:50:15","extension":"docx","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":15511,"visible":true,"origin":"","legend":"Supplementary Table 5. Proportion of mothers who smoked by mother's social class","description":"","filename":"GeneticandsocialconfoundingschizophreniaSITable5.docx","url":"https://assets-eu.researchsquare.com/files/rs-9633381/v1/526683a441cbc6ffa4d2c70a.docx"},{"id":109405599,"identity":"d2c81f25-77f4-4036-a08d-5231f22063b0","added_by":"auto","created_at":"2026-05-17 13:19:18","extension":"jpg","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":76899,"visible":true,"origin":"","legend":"Supplementary Figure 1: Map of NCDS regions","description":"","filename":"GeneticandsocialconfoundingschizophreniaSIFigure1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9633381/v1/385421b4d8b127b754dc3537.jpg"}],"financialInterests":"The authors have declared there is \u003cb\u003eNO\u003c/b\u003e conflict of interest to disclose","formattedTitle":"Disentangling genetic and social confounding in schizophrenia","fulltext":[{"header":"Introduction","content":"\u003cp\u003eSchizophrenia is a severe psychiatric disorder affecting over 23\u0026nbsp;million people worldwide [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. It is characterised by positive symptoms, including delusions and hallucinations, as well as negative symptoms, such as apathy, avolition and social withdrawal. Schizophrenia is associated with increased psychiatric and physical comorbidity, including cardiovascular disease and other lifestyle risks, contributing to high disease burden and premature mortality [\u003cspan additionalcitationids=\"CR2\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eSchizophrenia has a substantial genetic component. Twin and family studies have demonstrated higher concordance in monozygotic twins than in other first-degree relatives, with an estimated heritability of 80% [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. More recently, genome-wide association studies (GWAS) have found 287 distinct loci associated with schizophrenia across 120 genes [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eSeveral social and environmental exposures have been associated with schizophrenia onset, including familial socioeconomic position (SEP), parental education, maternal age, region of birth, neighbourhood deprivation, and perinatal factors such as maternal infection during pregnancy, parity (number of previous births), and maternal smoking and alcohol consumption during pregnancy [\u003cspan additionalcitationids=\"CR7 CR8 CR9 CR10\" citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. However, whether these associations are causal remains unclear. For example, exposure to environmental stressors like high population density and neighbourhood deprivation may cause schizophrenia (social causation), or individuals who are at higher risk for developing schizophrenia may be more likely to move to or remain in such areas (social drift/social selection hypothesis) [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eMendelian randomisation studies use genetic variants identified in GWAS to explore causal relationships between exposures and outcomes [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. Several Mendelian randomisation studies have estimated the effects of genetic liability to schizophrenia and other outcomes, including smoking [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e] cannabis use [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e], cardiovascular risk [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e], inflammation [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e], and population density [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. However, all these studies have used samples of unrelated individuals, making it challenging to determine whether these effects are due to genetic liability to schizophrenia or are confounded by population stratification, dynastic effects, or assortative mating [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eMendelian randomisation is an application of instrumental variable analysis, and thus depends on the three core instrumental variable assumptions [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. These include the relevance assumption: that genetic variants are robustly associated with the exposure of interest [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]; exclusion assumption: that the genetic variants are not related to the outcome of interest other than via the exposure [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]; and independence assumption: that genetic instruments are independent of other factors that confound the exposure-outcome relationship [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. If these three assumptions hold, then the estimates can provide reliable evidence of the causal effect of the exposure on the outcome (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA). However, population-based estimates of the associations between genetic variants and the outcome can also arise because they share common causes that confound the exposure-outcome relationship (Figs.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB and \u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eC). If genetic variants are not randomly distributed across the population, they may become correlated with environmental, social, or demographic factors that influence both the genetic variants and the outcome, thereby violating the independence assumption (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eD). For example, if schizophrenia-associated variants correlate with a third variable (e.g. social class) (arrow 1 in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eD), and are also associated with the outcome (e.g. smoking) (arrow 2), this creates a non-causal pathway between the genetic instrument and the outcome, making it impossible to obtain an unbiased estimate of the association between the exposure and outcome (e.g. genetic liability to schizophrenia and genetic liability to smoking).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e[Figure \u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e here]\u003c/p\u003e \u003cp\u003ePrevious studies have attempted to disentangle mechanisms of social causation versus social selection using genetically informed designs. A UK study using ALSPAC data found that schizophrenia polygenic index (PGI) did not predict birth into more densely populated neighbourhoods, although they found that children born into more deprived and socially fragmented neighbourhoods had higher PGI for schizophrenia [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. Similarly, a Danish study found that individuals with higher schizophrenia PGI were more likely to reside in the capital at age 15, but not at birth, providing further support for the social drift hypothesis [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. An analysis of 4 cohorts in the UK, Australia, and the Netherlands, found that individuals with high genetic risk for schizophrenia lived in more densely populated areas at a greater than chance level and lived in postcodes with higher population density even after controlling for SEP [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. Their analysis also included a two-sample Mendelian randomisation, which suggested that genetic risk for schizophrenia had a causal effect on likelihood of living in urban areas [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. However, the authors note that these results could be due to confounding in previous generations, which may make individuals both more likely to have a high genetic risk for schizophrenia and more likely to live in urban areas [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. It is plausible that individuals with a higher genetic risk for schizophrenia, such as the parents and ancestors of individuals recruited in the studies mentioned above, may have certain traits that make them more likely to move to more urban and more deprived areas [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. If parents with higher genetic liability for schizophrenia also tend, on average, to experience greater socioeconomic disadvantage, their children may be more likely to be exposed to social and environmental conditions that are associated with increased schizophrenia risk. This is in line with intergenerational (social) selection, whereby families with higher genetic liability to schizophrenia are more likely to drift to or remain in more socioeconomically disadvantaged neighbourhoods over time [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThe high heritability of schizophrenia, combined with the increased likelihood of individuals with high genetic risk for schizophrenia to live in more urban and deprived areas, makes it difficult to determine true causal pathways generating these associations. For instance, any association found between child PGI for schizophrenia and an exposure of interest (e.g. smoking) could result from the association between the parental genotype (or related traits) and the parental environment, resulting in an increased likelihood of the child being born into an environment with more adverse socioenvironmental exposures, rather than as a direct result of the child\u0026rsquo;s genetics (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB). Consequently, relationships between socioenvironmental exposures and schizophrenia risk, which may have been previously interpreted as causal in both Mendelian randomisation and social determinants of health research, may in fact be the result of confounding.\u003c/p\u003e \u003cp\u003eHere, we assess the correlation between genetic markers and socioenvironmental exposures around the time of birth to quantify the degree of confounding attributable to the familial and social factors to provide further insight into the aetiology of these associations. Our primary hypothesis is that genetic liability to schizophrenia is associated with socioenvironmental exposures at birth.\u003c/p\u003e"},{"header":"Materials and methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eData\u003c/h2\u003e \u003cp\u003eFollowing a pre-registered analysis plan (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.17605/OSF.IO/NJZAU\u003c/span\u003e\u003cspan class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e; see Supplementary Table\u0026nbsp;1 for deviations [\u003cspan class=\"CitationRef\"\u003e26\u003c/span\u003e]), we used data from the 1958 British birth cohort, the National Child Development Study (NCDS). NCDS is a nationally representative birth cohort that follows a sample of 17 416 people born in the UK in a single week in 1958 (98% of all births in Great Britain in that week) [\u003cspan class=\"CitationRef\"\u003e27\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e28\u003c/span\u003e]. This cohort has been followed up across multiple time points to monitor their educational, physical, and social development, health, and inequalities throughout childhood, adolescence, and adulthood. To date, 12 sweeps have been completed, the most recent in 2000 when the cohort was age 62 [\u003cspan class=\"CitationRef\"\u003e27\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThe NCDS cohort was genotyped using whole blood samples collected at age 44 [\u003cspan class=\"CitationRef\"\u003e29\u003c/span\u003e]. Genotyping and quality control procedures have been described elsewhere [\u003cspan class=\"CitationRef\"\u003e29\u003c/span\u003e]. Genetic data were available for 6324 participants (36%). After removing multiple births and participants without exposure data, the final sample was 6274 participants (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e[Figure \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e here]\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eExposures\u003c/h3\u003e\n\u003cp\u003eWe selected socioenvironmental exposures for schizophrenia based on the theoretical knowledge of important pre- and perinatal socioeconomic exposures previously linked to schizophrenia[\u003cspan class=\"CitationRef\"\u003e6\u003c/span\u003e] and data availability within our sample. Socioenvironmental exposures were measured using maternal interviews during the first wave of NCDS.\u003c/p\u003e \u003cp\u003eWe included three measures of social class: social class of mother’s father (social class I [highest] – V [lowest], other, no father), social class of mother’s husband (social class I-V, other, no husband) and mother’s paid job during pregnancy (social class I-V, no job during pregnancy; Supplementary Tables\u0026nbsp;2–3) [\u003cspan class=\"CitationRef\"\u003e30\u003c/span\u003e]. We measured parental demographics at the time of the child’s birth (maternal age, father’s age, child region of birth [defined by 11 NCDS regions, Supplementary Fig.\u0026nbsp;1], mother education [did mother stay past school leaving age: yes or no]), and pre- and perinatal exposures (parity, antenatal care [midpoint of category for week of first visit, for example 1st to 3rd week was recoded as 2, and midpoint of category for total number of visits, for example 5–9 total antenatal visits was recoded to 7], infection during pregnancy [yes or no infection – influenza or German measles], smoking prior to and during pregnancy [midpoint of category for number of cigarettes smoked, for example, smoking 1 to 4 cigarettes daily was recoded to 2.5]). When markers were coded categorically or as binary, the highest or most advantaged category was used as the reference (e.g. social class I was used as the reference for all social class variables).\u003c/p\u003e \u003cp\u003eWe included offspring sex and principal components of ancestry as covariates in all analyses to control for confounding by population stratification.\u003c/p\u003e\n\u003ch3\u003eOutcome\u003c/h3\u003e\n\u003cp\u003eWe assessed genetic liability to schizophrenia using the PGI based on the 2022 schizophrenia genome-wide association study (GWAS) [\u003cspan class=\"CitationRef\"\u003e5\u003c/span\u003e]. Schizophrenia PGI were calculated using LDpred-2 [\u003cspan class=\"CitationRef\"\u003e31\u003c/span\u003e]. The PGIs were standardised (mean = 0, standard deviation = 1). A subset of the NCDS genotyped sample (n = 3000) was included as controls in this GWAS, which introduced the risk of inflated PGI-trait associations due to sample overlap, however the low population prevalence of schizophrenia minimises the risk of inflation [\u003cspan class=\"CitationRef\"\u003e32\u003c/span\u003e]. We filtered variants with an imputation INFO score \u0026gt; 95% and an allele frequency \u0026gt; 1% to limit analyses to very well imputed variants.\u003c/p\u003e\n\u003ch3\u003eAnalysis\u003c/h3\u003e\n\u003cp\u003eWe estimated the joint associations of PGI with the exposures at birth using a mutually adjusted robust linear regression in R (version 4.2.3) [\u003cspan class=\"CitationRef\"\u003e33\u003c/span\u003e]. Bivariate models of PGI and each socioenvironmental exposure were run as a sensitivity analysis to investigate the independent association of the schizophrenia PGI with each exposure. Following an exploration of missingness patterns, we used mice (version 3.18.0) in R to impute missing data [\u003cspan class=\"CitationRef\"\u003e34\u003c/span\u003e]. We calculated the fraction of missing data for each parameter to inform the number of imputations required. Code files have been uploaded to Github: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/LaurieHaig/NCDS_paper\u003c/span\u003e"},{"header":"Results","content":"\u003cp\u003eThe final sample included 6 274 participants with genetic data. The average maternal age was 27.41 years, while fathers were slightly older at an average of 30.38 years. Participants were spread across regions (17% from the southeast of England, including London) and social classes (59% of husbands and 47% of fathers were in social class III, while 62% of mothers had no job during pregnancy) (Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e; for the non-imputed data, see Supplementary Table\u0026nbsp;4).\u003c/p\u003e\u003cp\u003e \u003c/p\u003e\u003cdiv class=\"gridtable\"\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003ctable id=\"Tab1\" border=\"1\"\u003e \u003ccaption\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eSample characteristics and bivariate associations with schizophrenia PGI\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003c/colgroup\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\"\u003e \u003cp\u003ePredictor\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\"\u003e \u003cp\u003en (%)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\"\u003e \u003cp\u003eBivariate β (95% CI), p-value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eSchizophrenia PGI (mean, SD)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e4.62 (1.01)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eChild sex\u003c/p\u003e \u003cp\u003eMale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e49.8%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eFemale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e50.2%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.033 (-0.017, 0.083), p = 0.194\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eSocial Class mother’s husband\u003c/p\u003e \u003cp\u003eSocial class I (ref)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e4.6%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eSocial class II\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e13.8%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e-0.088 (-0.225, 0.049), p = 0.208\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eSocial class III\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e59.5%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e-0.034 (-0.157, 0.089), p = 0.584\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eSocial class IV\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e11.9%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.006 (-0.135, 0.146), p = 0.937\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eSocial class V\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e7.7%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.022 (-0.128, 0.172), p = 0.773\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eOther\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.2%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.390 (-0.285, 1.065), p = 0.257\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eNo husband\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e2.3%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.095 (-0.112, 0.302), p = 0.369\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eSocial Class mother’s father\u003c/p\u003e \u003cp\u003eSocial class I (ref)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e2.7%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eSocial class II\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e14.1%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e-0.094 (-0.267, 0.079), p = 0.286\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eSocial class III\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e47.2%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e-0.099 (-0.262, 0.063), p = 0.230\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eSocial class IV\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e13.6%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e-0.079 (-0.252, 0.093), p = 0.366\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eSocial class V\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e12.2%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e-0.100 (-0.274, 0.074), p = 0.259\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eOther\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e2.1%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e-0.017 (-0.256, 0.222), p = 0.890\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eNo father\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e8.0%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e-0.104 (-0.284, 0.076), p = 0.258\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eMum’s paid job during pregnancy\u003c/p\u003e \u003cp\u003eSocial class I (ref)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e1.7%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eSocial class II\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.7%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.353 (-0.018, 0.724), p = 0.062\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eSocial class III\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e6.3%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e-0.027 (-0.244, 0.190), p = 0.806\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eSocial class IV\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e22.2%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.055 (-0.145, 0.255), p = 0.591\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eSocial class V\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e7.1%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.051 (-0.163, 0.265), p = 0.639\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eNo job\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e62.0%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.032 (-0.162, 0.226), p = 0.749\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eMaternal smoking prior to pregnancy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e5.78 (7.81)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.006 (0.003, 0.009), p = 0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eMaternal smoking during pregnancy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e4.31 (7.09)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.005 (0.001, 0.008), p = 0.010\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eMaternal age at last birthday\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e27.42 (5.55)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e-0.004 (-0.009, 0.001), p = 0.094\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eWas mum at school past leaving age?\u003c/p\u003e \u003cp\u003eStayed past leaving age (ref)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e26.4%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eDid not stay past leaving age\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e73.6%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e-0.016 (-0.073, 0.041), p = 0.586\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eParity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e1.23 (1.45)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.009 (-0.009, 0.026), p = 0.340\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eWeek of first antenatal visit\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e16.40 (6.51)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.005 (0.001, 0.009), p = 0.012\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eTotal antenatal visits\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e12.28 (5.45)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e-0.006 (-0.010, -0.001), p = 0.018\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eMaternal infection during pregnancy\u003c/p\u003e \u003cp\u003eNo infection during pregnancy (ref)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e87.6%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eYes infection during pregnancy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e12.4%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.037 (-0.041, 0.115), p = 0.349\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eHusband’s age in years\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e30.34 (6.30)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e-0.001 (-0.005, 0.003), p = 0.505\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eRegion of birth\u003c/p\u003e \u003cp\u003eSoutheast including London (ref)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e16.8%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eEast \u0026amp; West Riding\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e8.4%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e-0.087 (-0.192, 0.018), p = 0.106\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eEast\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e7.4%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e-0.084 (-0.194, 0.025), p = 0.132\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eMidlands\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e9.9%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e-0.006 (-0.106, 0.093), p = 0.898\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eNorth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e7.6%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e-0.166 (-0.274, -0.057), p = 0.003\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eNorth Midlands\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e7.9%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e-0.058 (-0.166, 0.049), p = 0.287\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eNorth West\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e11.7%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.034 (-0.061, 0.129), p = 0.482\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eScotland\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e13.9%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.000 (-0.090, 0.091), p = 0.995\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eSouth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e5.7%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e-0.222 (-0.342, -0.101), p = 0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eSouth West\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e5.6%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e-0.128 (-0.249, -0.007), p = 0.038\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eWales\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e5.1%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e0.150 (0.024, 0.277), p = 0.020\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/table\u003e\u003c/div\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eMaternal smoking, younger maternal age and being born in the southeast of England were associated with higher schizophrenia PGI. Participants of mothers who smoked more prior to pregnancy had, on average, a 0.008 (95% CI: 0.002 to 0.014) standard deviation higher PGI per additional cigarette smoked (Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e, Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e). On average, a child of a 30-year-old mother had a PGI 0.08 standard deviations lower (95% CI: 0.00 to 0.16) than a child of a 20-year-old mother. Children born in the north (β= -0.214 [95% CI: -0.324, -0.103]), south (β= -0.198 [95% CI -0.318, -0.078]), southwest (β= -0.125 [95% CI: -0.246, -0.005]) and Scotland (β= -0.148 [95% CI: -0.249, -0.048]) all had significantly lower PGI compared to children born in the southeast of England (including London). Furthermore, the PGIs varied across major geographic gradients. A one-standard-deviation increase in the first genetic principal component of ancestry was associated with a 0.106 (95% CI: 0.076, 0.136) standard deviation increase in schizophrenia PGI. We found little evidence of differences in PGI by social class, father’s age, maternal education, parity, antenatal care, or maternal infection.\u003c/p\u003e\u003cp\u003e \u003c/p\u003e\u003cdiv class=\"gridtable\"\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" class=\"colspec\"\u003e\u003c/div\u003e\u003ctable id=\"Tab2\" border=\"1\"\u003e \u003ccaption\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eAssociation between socioenvironmental exposures and schizophrenia PGI outcome\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003c/colgroup\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\"\u003e \u003cp\u003ePredictor\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\"\u003e \u003cp\u003eβ\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\"\u003e \u003cp\u003eSE\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\"\u003e \u003cp\u003e95% CI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\"\u003e \u003cp\u003ep-value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eIntercept\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.237\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.164\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.084, 0.558]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.148\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eSocial class mother’s husband\u003c/p\u003e \u003cp\u003eSocial class II\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e-0.056\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.070\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.193, 0.081]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.422\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eSocial class III\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e-0.018\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.066\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.147, 0.112]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.791\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eSocial class IV\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.028\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.076\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.12, 0.177]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.710\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eSocial class V\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e-0.011\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.081\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.169, 0.148]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.894\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eOther\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.439\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.340\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.228, 1.107]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.197\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eNo husband\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.041\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.110\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.175, 0.257]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.707\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eSocial class mother’s father\u003c/p\u003e \u003cp\u003eSocial class II\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e-0.092\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.088\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.265, 0.081]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.297\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eSocial class III\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e-0.100\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.086\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.269, 0.068]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.243\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eSocial class IV\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e-0.097\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.092\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.277, 0.083]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.290\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eSocial class V\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e-0.133\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.093\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.316, 0.05]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.155\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eOther\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e-0.037\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.123\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.279, 0.205]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.763\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eNo father\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e-0.133\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.094\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.318, 0.051]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.157\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eMum’s paid job during pregnancy\u003c/p\u003e \u003cp\u003eSocial class II\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.345\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.188\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.025, 0.714]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.067\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eSocial class III\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e-0.016\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.112\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.236, 0.203]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.883\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eSocial class IV\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.068\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.103\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.135, 0.27]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.512\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eSocial class V\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.074\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.111\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.143, 0.291]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.506\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eNo job\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.061\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.100\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.136, 0.257]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.547\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eMaternal smoking prior to pregnancy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.008\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.003\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[0.002, 0.014]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.010\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eMaternal smoking during pregnancy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e-0.004\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.003\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.011, 0.002]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.207\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eMaternal age at last birthday\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e-0.008\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.004\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.016, 0]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.040\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eWas mum at school past leaving age?\u003c/p\u003e \u003cp\u003eDid not stay past leaving age\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e-0.025\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.033\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.09, 0.041]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.458\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eParity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.013\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.012\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.01, 0.036]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.267\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eWeek of mother’s first antenatal visit\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.002, 0.006]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.372\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eTotal number of antenatal visits\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e-0.003\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.008, 0.002]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.233\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eYes infection during pregnancy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.047\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.039\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.031, 0.124]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.238\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eHusband’s age in years\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.003\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.004, 0.009]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.506\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eRegion of birth\u003c/p\u003e \u003cp\u003eEast \u0026amp; West Riding\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e-0.062\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.054\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.169, 0.045]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.254\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eEast\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e-0.045\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.056\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.155, 0.064]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.418\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eMidlands\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.051\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.1, 0.101]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.989\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eNorth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e-0.214\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.056\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.324, -0.103]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eNorth Midlands\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e-0.027\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.055\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.135, 0.081]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.629\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eNorth West\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e-0.014\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.050\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.112, 0.084]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.781\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eScotland\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e-0.148\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.051\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.249, -0.048]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.004\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eSouth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e-0.198\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.061\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.318, -0.078]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eSouth West\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e-0.125\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.061\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.246, -0.005]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.041\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eWales\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.011\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.074\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.135, 0.156]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.887\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003eChild sex\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.028\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.025\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.021, 0.078]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.261\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003ePC1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.106\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.015\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[0.076, 0.136]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003ePC2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.003\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.013\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.022, 0.028]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.801\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003ePC3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e-0.022\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.015\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.052, 0.008]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.148\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003ePC4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e-0.005\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.013\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.031, 0.02]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.688\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003ePC5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.010\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.013\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.015, 0.035]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.441\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003ePC6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e-0.032\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.013\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.057, -0.008]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.010\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003ePC7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e-0.005\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.013\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.03, 0.02]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.688\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003ePC8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e-0.030\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.013\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.055, -0.005]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.017\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003ePC9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.007\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.013\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.017, 0.032]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.559\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003ePC10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.012\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.013\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.013, 0.037]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.348\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003ePC11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.007\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.013\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.018, 0.032]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.580\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003ePC12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.003\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.013\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.022, 0.027]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.834\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003ePC13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e-0.020\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.013\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.045, 0.004]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.107\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003ePC14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.006\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.013\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.019, 0.03]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.655\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003ePC15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.016\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.013\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.008, 0.041]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.191\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003ePC16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.033\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.013\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[0.008, 0.058]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.009\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003ePC17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.010\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.013\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.015, 0.035]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.426\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003ePC18\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.021\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.013\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.004, 0.046]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.096\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003ePC19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.021\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.013\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.003, 0.046]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.091\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\"\u003e \u003cp\u003ePC20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.009\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.013\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\"\u003e \u003cp\u003e[-0.015, 0.034]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\"\u003e \u003cp\u003e0.466\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"5\"\u003eValues are pooled estimates from multiple imputation; CI = confidence interval.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e[Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e here; Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e]\u003c/p\u003e\u003cp\u003eIn a sensitivity analysis, we estimated bivariate associations between the schizophrenia PGI and each exposure to assess the independent association of the schizophrenia PGI with each socioenvironmental exposure (Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e). Overall, the bivariate associations were consistent with the mutually adjusted model (Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e). Children whose mothers smoked prior to (β = 0.006 [95% CI: 0.003, 0.009]) and during pregnancy (β = 0.005 [95% CI: 0.001, 0.008]) had higher schizophrenia PGI, while this association was only seen for smoking prior to pregnancy in the mutually adjusted model. Children born in the south, north, and southwest of England had lower PGI than those born in the southeast, in both the bivariate correlations and the mutually adjusted model. Children born in Scotland had lower PGI in the mutually adjusted model (β= -0.148 [95% CI: -0.249, -0.048]) but not the bivariate model [β = 0.000 (95%CI: -0.090, 0.091)], while children born in Wales had higher PGI in the bivariate model [β = 0.150 (95%CI: 0.024, 0.277)] but not the mutually adjusted model [β = 0.011 (95%CI: -0.135, 0.156)]. Children born to younger mothers had higher PGI in both the mutually adjusted and bivariate models. In the bivariate model, children whose mothers’ first antenatal visit was later in their pregnancy had higher PGI, with each one-week delay associated with a 0.005 SD higher PGI (95% CI: 0.001, 0.009) and children whose mothers attended more antenatal visits had on average 0.006 SD lower PGI (95% CI -0.010, -0.001) per additional visit, however these associations were not detected in the mutually adjusted model.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eWe found that the schizophrenia PGI was associated with several socioenvironmental exposures at birth. This suggests that children born in specific regions, whose mothers smoked more heavily before pregnancy or were younger, were also likely to have a higher genetic liability to schizophrenia.\u003c/p\u003e \u003cp\u003e We found clear regional differences in the distribution of offspring schizophrenia PGI at birth, suggesting that ancestors with higher schizophrenia PGI were more likely to move to or remain in certain regions within the UK. The reference category for region of birth was the southeast of England (including London), and our results suggest that schizophrenia PGI was, on average, higher here than in nearly all other regions. This aligns with previous studies, which have reported genetic risk for schizophrenia to be higher in more urban and densely populated areas [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. This also aligns with previously reported regional differences in schizophrenia incidence across the UK, as incidence was generally higher in more urban areas [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]. Similarly, urban versus rural birth or birth into more deprived neighbourhoods has been associated with a higher incidence of schizophrenia [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan additionalcitationids=\"CR38 CR39 CR40\" citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e]. The first genetic principal component of ancestry, which generally captures north versus south geographical clustering in ancestral gradients [\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e] was also associated with PGI at birth, providing further evidence for distinct regional differences.\u003c/p\u003e \u003cp\u003eAdditionally, we found that children of mothers who smoked more prior to pregnancy had a higher schizophrenia PGI compared to children of mothers who smoked less. Recent studies have found smoking during pregnancy to be more common among mothers in routine and manual occupations, and smoking cessation to be lower among more disadvantaged pregnant women [\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e]. Smoking rates today tend to be higher among those with lower socioeconomic position [\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e], however, in the 1950s and 60s, smoking was more common among UK women of all social classes [\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e], and in our sample, around 30\u0026ndash;50% of women smoked within each social class (Supplementary Table\u0026nbsp;5). Furthermore, we found little evidence of differences in PGI by social class, or of any other socially patterned variables such as education, parity, or maternal infection, in either the mutually adjusted or bivariate models, though there was some evidence for differences in PGI by antenatal care in the bivariate models only. We did, however, find that children born to younger mothers had a higher PGI compared to those born to older mothers.\u003c/p\u003e \u003cp\u003eOverall, these results provide evidence from a nationally representative sample that schizophrenia PGI was not randomly distributed in the population but rather varied by UK region of birth, principal components related to major ancestral geographic gradients, and maternal smoking. This suggests that associations between the schizophrenia PGI reported in Mendelian randomization studies may reflect these confounding factors (Figs.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB and \u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eD). Our results may also have implications for studies investigating the social determinants of schizophrenia, as they suggest differences in incidence of schizophrenia by region and maternal smoking, which may be explained by differences in genetic liability. This also has consequences for molecular genetic research using population-based samples, as it highlights potential for confounding pathways if the outcome of interest differs by region or parental smoking patterns. Even a small bias in the association of the schizophrenia PGI and smoking could have a large bias on the estimate in Mendelian randomisation studies [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e], suggesting previously reported bidirectional effects of genetic liability to schizophrenia on smoking may be explained by maternal smoking or regional differences [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. However, future studies would benefit from family-based designs that include genetic information on both biological parents and their children or siblings to confirm this and more robustly test the causal relationships between genetic liability and socioenvironmental exposures by controlling for parental genotype [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e].\u003c/p\u003e\n\u003ch3\u003eLimitations\u003c/h3\u003e\n\u003cp\u003eNCDS is a population-representative birth-cohort dataset that enables us to investigate many socioenvironmental exposures at birth. However, because the first wave was conducted in 1958, data availability and the nature of relationships between socioenvironmental exposures may differ from those observed in a more modern context. As noted earlier, smoking patterns were different among women in the UK at this time, making our findings of the significance of smoking prior to pregnancy more difficult to interpret. Other exposures, such as antenatal care, level of education, and occupations (particularly for women and pregnant women), differ as well. Future studies should investigate these relationships in more recent population cohorts. Additionally, although the NCDS included nearly all births across Great Britain in a single week, the genetic data sample was limited to individuals of European ancestry and may have been affected by differential attrition, as samples were taken at age 44. Therefore, it is important that these associations be tested in more diverse ancestry cohorts with appropriate PGI. Furthermore, Choi et al. highlighted the risk of inflated PGI-trait associations due to sample overlap, which was present in our study. However, their simulations showed a reduced risk when overlap was limited to controls and to traits with lower population prevalence [\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e]. As schizophrenia is a binary trait with a low population prevalence (~\u0026thinsp;1%) and the sample overlap with NCDS was only among GWAS controls (degree of overlap\u0026thinsp;=\u0026thinsp;0.0784), the risk of inflation is low [\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e].\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eSchizophrenia PGI was associated with pre-natal exposures in a population-based study. This demonstrates that genetic liability for schizophrenia is unlikely to be randomly distributed across the population and indicates a potential for confounding in studies using Mendelian randomisation and those investigating the social determinants of schizophrenia. Future research would benefit from family-based designs with genetic information from both biological parents to explore these associations after accounting for parental genetic liability.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements:\u0026nbsp;\u003c/strong\u003eWe thank the participants in the 1958 British Birth Cohort and their families for taking part in this study.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions:\u0026nbsp;\u003c/strong\u003eLH, JD and NMD contributed to the conception of the work. All authors contributed to the analysis and interpretation of data for the work. All authors contributed to the drafting of the work, critical review and final approval of the version to be published.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of Interest:\u0026nbsp;\u003c/strong\u003eJFH has received consultancy fees from Wellcome Trust, juli inc and Swiss Re. No other authors declare conflicts of interest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding support:\u0026nbsp;\u003c/strong\u003eLH is supported by the UCL-Birbeck MRC Doctoral Training Programme (MR/W006774/1). JFH is funded by UKRI grant MR/V023373/1, the University College London Hospitals NIHR Biomedical Research Centre and the NIHR North Thames Applied Research Collaboration.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRole of the Funder/Sponsor:\u0026nbsp;\u003c/strong\u003eThe funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Sharing Statement:\u0026nbsp;\u003c/strong\u003eData from the 1958 British Birth Cohort is freely available to researchers in the UK and around the world. Deidentified data is available through public repositories (primarily the UK data service), and researchers can apply to access other data, including genetic data, directly from CLS (https://cls.ucl.ac.uk/data-access-training/data-access/). For the purpose of Open Access, the author has applied a CC BY licence to any Author Accepted Manuscript version arising from this submission. Supplementary information is available at Molecular Psychiatry’s website.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eSolmi M, Seitidis G, Mavridis D, Correll CU, Dragioti E, Guimond S, et al. Incidence, prevalence, and global burden of schizophrenia - data, with critical appraisal, from the Global Burden of Disease (GBD) 2019. Molecular Psychiatry 2023 28:12. 2023;28:5319\u0026ndash;5327.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCorrell CU, Solmi M, Croatto G, Schneider LK, Rohani-Montez SC, Fairley L, et al. Mortality in people with schizophrenia: a systematic review and meta-analysis of relative risk and aggravating or attenuating factors. World Psychiatry. 2022;21:248\u0026ndash;271.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWahbeh MH, Avramopoulos D. Gene-Environment Interactions in Schizophrenia: A Literature Review. Genes 2021, Vol 12,. 2021;12.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHilker R, Helenius D, Fagerlund B, Skytthe A, Christensen K, Werge TM, et al. Heritability of Schizophrenia and Schizophrenia Spectrum Based on the Nationwide Danish Twin Register. Biol Psychiatry. 2018;83:492\u0026ndash;498.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTrubetskoy V, Pardi\u0026ntilde;as AF, Qi T, Panagiotaropoulou G, Awasthi S, Bigdeli TB, et al. Mapping genomic loci implicates genes and synaptic biology in schizophrenia. Nature. 2022;604:502\u0026ndash;508.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDavies C, Segre G, Estrad\u0026eacute; A, Radua J, De Micheli A, Provenzani U, et al. Prenatal and perinatal risk and protective factors for psychosis: a systematic review and meta-analysis. Lancet Psychiatry. 2020;7:399\u0026ndash;410.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHakulinen C, Webb RT, Pedersen CB, Agerbo E, Mok PLH. Association Between Parental Income During Childhood and Risk of Schizophrenia Later in Life. JAMA Psychiatry. 2020;77:17.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchneider M, M\u0026uuml;ller CP, Knies AK. Low income and schizophrenia risk: A narrative review. Behavioural Brain Research. 2022;435:114047.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSariaslan A, Fazel S, D\u0026rsquo;Onofrio BM, L\u0026aring;ngstr\u0026ouml;m N, Larsson H, Bergen SE, et al. Schizophrenia and subsequent neighborhood deprivation: revisiting the social drift hypothesis using population, twin and molecular genetic data. Transl Psychiatry. 2016;6:e796\u0026ndash;e796.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDohrenwend BP, Levav I, Shrout PE, Schwartz S, Naveh G, Link BG, et al. Socioeconomic Status and Psychiatric Disorders: The Causation-Selection Issue. Science (1979). 1992;255:946\u0026ndash;952.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKirkbride JB, Anglin DM, Colman I, Dykxhoorn J, Jones PB, Patalay P, et al. The social determinants of mental health and disorder: evidence, prevention and recommendations. World Psychiatry. 2024;23:58\u0026ndash;90.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSmith GD, Ebrahim S. \u0026lsquo;Mendelian randomization\u0026rsquo;: can genetic epidemiology contribute to understanding environmental determinants of disease? Int J Epidemiol. 2003;32:1\u0026ndash;22.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSanderson E, Levin MG, Walker V, Yuan S, Badini I, Dolce J, et al. Challenges and future directions for Mendelian randomization [Submitted for publication]. Nat Genet.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWootton RE, Richmond RC, Stuijfzand BG, Lawn RB, Sallis HM, Taylor GMJ, et al. Evidence for causal effects of lifetime smoking on risk for depression and schizophrenia: a Mendelian randomisation study. Psychol Med. 2020;50:2435\u0026ndash;2443.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVaucher J, Keating BJ, Lasserre AM, Gan W, Lyall DM, Ward J, et al. Cannabis use and risk of schizophrenia: a Mendelian randomization study. Mol Psychiatry. 2017;23:1287.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGage SH, Jones HJ, Burgess S, Bowden J, Davey Smith G, Zammit S, et al. Assessing causality in associations between cannabis use and schizophrenia risk: a two-sample Mendelian randomization study. Psychol Med. 2017;47:971\u0026ndash;980.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVeeneman RR, Vermeulen JM, Abdellaoui A, Sanderson E, Wootton RE, Tadros R, et al. Exploring the Relationship Between Schizophrenia and Cardiovascular Disease: A Genetic Correlation and Multivariable Mendelian Randomization Study. Schizophr Bull. 2022;48:463\u0026ndash;473.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHartwig FP, Borges MC, Horta BL, Bowden J, Davey Smith G. Inflammatory Biomarkers and Risk of Schizophrenia: A 2-Sample Mendelian Randomization Study. JAMA Psychiatry. 2017;74:1226\u0026ndash;1233.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eColodro-Conde L, Couvy-Duchesne B, Whitfield JB, Streit F, Gordon S, Kemper KE, et al. Association Between Population Density and Genetic Risk for Schizophrenia. JAMA Psychiatry. 2018;75:901.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDavies NM, Hemani G, Neiderhiser JM, Martin HC, Mills MC, Visscher PM, et al. The importance of family-based sampling for biobanks. Nature. 2024;634:795\u0026ndash;803.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWalker V, Sanderson E, Levin MG, Damraurer SM, Feeney T, Davies NM. Reading and conducting instrumental variable studies: guide, glossary, and checklist. BMJ. 2024;387:78093.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDe Leeuw C, Savage J, Bucur IG, Heskes T, Posthuma D. Understanding the assumptions underlying Mendelian randomization. European Journal of Human Genetics. 2022;30:653\u0026ndash;660.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSolmi F, Lewis G, Zammit S, Kirkbride JB. Neighborhood Characteristics at Birth and Positive and Negative Psychotic Symptoms in Adolescence: Findings From the ALSPAC Birth Cohort. Schizophr Bull. 2020;46:581\u0026ndash;591.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePaksarian D, Trabjerg BB, Merikangas KR, Mors O, B\u0026oslash;rglum AD, Hougaard DM, et al. The role of genetic liability in the association of urbanicity at birth and during upbringing with schizophrenia in Denmark. Psychol Med. 2018;48:305\u0026ndash;314.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGage SH, Davey Smith G, Munaf\u0026ograve; MR. Schizophrenia and neighbourhood deprivation. Transl Psychiatry. 2016;6:e979\u0026ndash;e979.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWillroth EC, Atherton OE. Best Laid Plans: A Guide to Reporting Preregistration Deviations. Adv Methods Pract Psychol Sci. 2024;7.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePower C, Elliott J. Cohort profile: 1958 British birth cohort (National Child Development Study). Int J Epidemiol. 2006;35:34\u0026ndash;41.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCLS | 1958 National Child Development Study. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://cls.ucl.ac.uk/cls-studies/1958-national-child-development-study/\u003c/span\u003e\u003cspan address=\"https://cls.ucl.ac.uk/cls-studies/1958-national-child-development-study/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Accessed 19 February 2026.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShireby G, Morris TT, Wong A, Chaturvedi N, Ploubidis GB, Fitzsimmons E, et al. Data Resource Profile: Genomic data in multiple British birth cohorts (1946\u0026ndash;2001)\u0026mdash;linkage with health, social, and environmental data from birth to old age. Int J Epidemiol. 2025;54:dyaf141.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSOC 2000 - Office for National Statistics. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.ons.gov.uk/methodology/classificationsandstandards/standardoccupationalclassificationsoc/socarchive\u003c/span\u003e\u003cspan address=\"https://www.ons.gov.uk/methodology/classificationsandstandards/standardoccupationalclassificationsoc/socarchive\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Accessed 19 February 2026.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLDpred2/README.md at main \u0026middot; AndreAllegrini/LDpred2 \u0026middot; GitHub. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/AndreAllegrini/LDpred2/blob/main/README.md\u003c/span\u003e\u003cspan address=\"https://github.com/AndreAllegrini/LDpred2/blob/main/README.md\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Accessed 29 January 2026.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWellcome Trust Case Control Consortium 2 - WTCCC. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.wtccc.org.uk/ccc2/index.html\u003c/span\u003e\u003cspan address=\"https://www.wtccc.org.uk/ccc2/index.html\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Accessed 19 February 2026.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eR Core Team. R: A Language and Environment for Statistical Computing. 2023.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evan Buuren S, Groothuis-Oudshoorn K. mice: Multivariate imputation by chained equations in R. J Stat Softw. 2011;45:1\u0026ndash;67.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKirkbride JB, Fearon P, Morgan C, Dazzan P, Morgan K, Tarrant J, et al. Heterogeneity in Incidence Rates of Schizophrenia and Other Psychotic Syndromes. Arch Gen Psychiatry. 2006;63:250.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrigoroglou C, Munford L, Webb RT, Kapur N, Ashcroft DM, Kontopantelis E. Prevalence of mental illness in primary care and its association with deprivation and social fragmentation at the small-area level in England. Psychol Med. 2020;50:293\u0026ndash;302.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLewis G, Dykxhoorn J, Karlsson H, Khandaker GM, Lewis G, Dalman C, et al. Assessment of the Role of IQ in Associations Between Population Density and Deprivation and Nonaffective Psychosis. JAMA Psychiatry. 2020;77:729.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWerner S, Malaspina D, Rabinowitz J. Socioeconomic Status at Birth Is Associated With Risk of Schizophrenia: Population-Based Multilevel Study. Schizophr Bull. 2006;33:1373\u0026ndash;1378.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMarch D, Hatch SL, Morgan C, Kirkbride JB, Bresnahan M, Fearon P, et al. Psychosis and Place. Epidemiol Rev. 2008;30:84\u0026ndash;100.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMortensen PB, Pedersen CB, Westergaard T, Wohlfahrt J, Ewald H, Mors O, et al. Effects of Family History and Place and Season of Birth on the Risk of Schizophrenia. New England Journal of Medicine. 1999;340:603\u0026ndash;608.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLewis G, David A, Andr\u0026eacute;asson S, Allebeck P. Schizophrenia and city life. Lancet. 1992;340:137\u0026ndash;140.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNovembre J, Johnson T, Bryc K, Kutalik Z, Boyko AR, Auton A, et al. Genes mirror geography within Europe. Nature 2008 456:7218. 2008;456:98\u0026ndash;101.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHiscock R, Bauld L, Amos A, Fidler JA, Munaf\u0026ograve; M. Socioeconomic status and smoking: A review. Ann N Y Acad Sci. 2012;1248:107\u0026ndash;123.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBerridge V. Constructing women and smoking as a public health problem in Britain 1950-1990s. Gend Hist. 2001;13:328\u0026ndash;348.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGraham H. Smoking prevalence among women in the European Community 1950\u0026ndash;1990. Soc Sci Med. 1996;43:243\u0026ndash;254.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChoi SW, Mak TSH, Hoggart CJ, O\u0026rsquo;Reilly PF. EraSOR: a software tool to eliminate inflation caused by sample overlap in polygenic score analyses. Gigascience. 2022;12.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-9633381/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9633381/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eSchizophrenia is a severe psychiatric disorder with known genetic and socioenvironmental antecedents. We tested whether genetic liability to schizophrenia is associated with socioenvironmental exposures for schizophrenia at birth which could indicate confounding. Data were drawn from the first wave of the 1958 British Birth Cohort, the National Child Development Study (NCDS). We selected 13 socioenvironmental exposures which have been linked to increased schizophrenia risk, including measures of social class, parental demographics, and pre- and perinatal exposures. We applied mutually adjusted regression to test the association between these socioenvironmental exposures and child’s polygenic susceptibility to schizophrenia measured via polygenic index (PGI). The final sample included 6,274 participants. Genetic liability to schizophrenia was not randomly distributed and associated with some socioenvironmental exposures at birth. Children whose mothers smoked more prior to pregnancy had a 0.008 standard deviation (SD) higher schizophrenia PGI per unit increase in cigarettes smoked (95%CI:0.002 to 0.014). Children born to older mothers had lower PGI, with a 0.008SD decrease per additional year of maternal age (95%CI:0.000 to 0.016). Compared with children born in the southeast of England, those born in the north, south, southwest, and Scotland had lower PGI. Children with higher values for the genetic principal component capturing north versus south geographical clustering had higher PGI (0.106 SD per unit; 95%CI:0.076 to 0.136). Children did not differ meaningfully by other socioenvironmental exposures. This has implications for causal inference and Mendelian randomisation studies of schizophrenia, as they may be affected by confounding.\u003c/p\u003e","manuscriptTitle":"Disentangling genetic and social confounding in schizophrenia","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-15 17:50:10","doi":"10.21203/rs.3.rs-9633381/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"21b17f98-8189-4422-8e1b-e59fa3df32d4","owner":[],"postedDate":"May 15th, 2026","published":true,"recentEditorialEvents":[{"type":"editorAssigned","content":"","date":"2026-05-07T16:05:08+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-05-07T14:43:08+00:00","index":"","fulltext":""},{"type":"submitted","content":"Molecular Psychiatry","date":"2026-05-06T16:31:58+00:00","index":"","fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":67732539,"name":"Biological sciences/Genetics"},{"id":67732540,"name":"Health sciences/Diseases/Psychiatric disorders/Schizophrenia"}],"tags":[],"updatedAt":"2026-05-15T17:50:11+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-15 17:50:10","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9633381","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9633381","identity":"rs-9633381","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}