京都大学 / 新制・論文博士 / 博士(医学) / 乙第13438号 / 論医博第2237号 / 新制||医||1054(附属図書館) / 京都大学大学院医学研究科医学専攻 / (主査)教授 川上 浩司, 教授 滝田 順子, 教授 万代 昌紀 / 学位規則第4条第2項該当 / Doctor of Medical Science / Kyoto University / DFAM

Yes / Born in Bradford (BiB) is a prospective multi-ethnic pregnancy and birth cohort study that was established to examine determinants of health and development during childhood and, subsequently, adult life in a deprived multi-ethnic population in the north of England. Between 2007 and 2010, the BiB cohort recruited 12,453 women who experienced 13,776 pregnancies and 13,858 births, along with 3353 of their partners. Forty five percent of the cohort are of Pakistani origin. Now that children are at primary school, the first full follow-up of the cohort is taking place. The aims of the follow-up are to investigate the determinants of children's pre-pubertal health and development, including through understanding parents' health and wellbeing, and to obtain data on exposures in childhood that might influence future health. Methods: We are employing a multi-method approach across three data collection arms (community-based family visits, school based physical assessment, and whole classroom cognitive, motor function and wellbeing measures) to follow-up over 9000 BiB children aged 7-11 years and their families between 2017 and 2021. We are collecting detailed parent and child questionnaires, cognitive and sensorimotor assessments, blood pressure, anthropometry and blood samples from parents and children. Dual x-ray absorptiometry body scans, accelerometry and urine samples are collected on subsamples. Informed consent is collected for continued routine data linkage to health, social care and education records. A range of engagement activities are being used to raise the profile of BiB and to disseminate findings. Discussion: Our multi-method approach to recruitment and assessment provides an efficient method of collecting rich data on all family members. Data collected will enhance BiB as a resource for the international research community to study the interplay between ethnicity, socioeconomic circumstances and biology in relation to cardiometabolic health, mental health, education, cognitive and sensorimotor development and wellbeing. / BiB receives core infrastructure funding from the Wellcome Trust (WT101597MA) and the National Institute for Health Research (NIHR) under its Collaboration for Applied Health Research and Care (CLAHRC) for Yorkshire and Humber and Clinical Research Network (CRN) research delivery support. Further support for genome-wide and multiple ‘omics measurements is from the UK Medical Research Council (G0600705), National Institute of Health Research (NF-SI-0611-10196), US National Institute of Health (R01 DK10324), and the European Research Council under the European Union’s Seventh Framework Programme (FP7/2007–2013) / ERC grant agreement no 669545. The follow-up of BiB participants, which is the focus of this paper, is funded by a joint grant from the UK Medical Research Council and UK Economic and Social Science Research Council (MR/N024397/1) and a grant from the British Heart Foundation (CS/16/4/32482.) D.A.L. works in a unit that receives UK Medical Research Council funding (MC_UU_00011/6) and is a UK National Institute of Health Research senior investigator (NF-SI-0611-10196).

Abstract Temporomandibular disorders (TMD) are described as a variety of dysfunctions and pains related to the masticatory system. Clicking in the temporomandibular joints (TMJs) and pain in the masticatory muscles are the most common symptoms and signs of TMD. Other frequently reported symptoms are headache, facial pain, and symptoms in the ears. The prevalence of TMD signs in the adult population has been shown to vary between 33% and 86%, being more common in women than in men. TMD has been found to be most prevalent at 20–50 years of age. The aim of this cross-sectional study was to investigate the prevalence of TMD and associated risk factors in the Finnish adult population based on the Northern Finland Birth Cohort (NFBC) 1966. Altogether 1 962 cohort subjects at the age of 46 years (1 050 women, 912 men) responded to questionnaires and participated in a clinical medical and dental examination in 2012–2013. The stomatognathic examination was performed according to a modified protocol of diagnostic criteria for TMD (DC/TMD). The prevalence of TMD signs among the examined cohort subjects was 34.2%; women had clinical signs of TMD more often than men. Clinical signs were registered more often than self-reported TMD symptoms. The most common clinical signs were clicking in the TMJs (26.2%) and palpation pain in the masticatory muscles (11.2%). The most common diagnosis was disc displacement with reduction (7.0%). TMD signs were associated with unstable occlusion. Occlusal disturbances were associated with palpation pain in the masticatory muscles, as well as myalgia and arthralgia diagnoses. Pain-related TMD symptoms and clinical signs correlated strongly with a poor self-reported health condition as well as with painful general health problems. / Tiivistelmä Purentaelimistön toimintahäiriöt (engl. temporomandibular disorders, TMD) on yhteisnimitys leukanivelten, puremalihasten, hampaiston ja niihin liittyvien kudosten sairaus- ja kiputiloille. Leukanivelen naksuminen ja kipu purentaelimistön alueella ovat tavallisimmat oireet, mutta myös päänsärky, kasvokipu ja korvaoireet ovat yleisiä TMD-potilailla. Väestötutkimuksissa on havaittu suurta vaihtelua TMD:n esiintyvyydessä (33–86 %). TMD:n oireita sekä löydöksiä todetaan tavallisimmin 20–50-vuotiailla ja naisilla enemmän kuin miehillä. Tutkimuksen tarkoituksena oli selvittää TMD:n esiintyvyyttä ja riskitekijöitä suomalaisessa aikuisväestössä Pohjois-Suomen syntymäkohortti 1966:n (Northern Finland Birth Cohort, NFBC) aineistossa. Vuosina 2012–2013 tutkittiin 1 962 kohortin henkilöä 46 vuoden ikäisinä (1 050 naista, 912 miestä). Tutkimus koostui kyselyistä sekä lääketieteellisestä ja suun terveydentilan kliinisestä tutkimuksesta. Purentaelimistön tutkimuksessa käytettiin kansainvälistä TMD:n diagnostista kriteeristöä (engl. diagnostic criteria for TMD, DC/TMD). TMD-löydösten esiintyvyys tutkitussa kohorttiaineistossa oli 34,2 %. Kliiniset löydökset tutkituilla henkilöillä olivat yleisempiä kuin heidän raportoimansa oireet. Yleisimmät löydökset olivat leukanivelen naksuminen (26,2 %) ja palpaatiokipu puremalihaksissa (11,2 %). Yleisin diagnoosi oli leukanivelen välilevyn palautuva virheasento (7,0 %). TMD-löydösten ja epästabiilin purennan välillä havaittiin yhteys. Purennassa olevien häiriöiden ja puremalihasten palpaatiokivun sekä leukanivelkivun välillä todettiin selvä yhteys. TMD-löydöksillä ja niihin liittyvillä kivuilla havaittiin tilastollisesti merkitsevä yhteys koettuun heikentyneeseen terveydentilaan sekä kivuliaisiin yleissairauksiin.

Abstract Study samples in medical research are selected according to the objectives of the studies. Researchers seek to collect data as extensively and reliably as possible. In practice, however, data are often missing or may be incorrect. This thesis covers some of the problems concerning missing data and data collection in psychiatric research. Methods for adjusting for missing data and for evaluating the reliability of data are presented. The data originate from the Northern Finland 1966 Birth Cohort (N = 12058). This study explored how participation in an epidemiologic study that includes questionnaires and a clinical examination is affected by mental health (N = 11540), and whether non-participants experience more severe clinical symptoms than participants in a psychiatric field study (N = 145) among subjects with a psychosis. Inverse probability weighting (IPW) was used to adjust for non-participation in comparisons of brain volumes between schizophrenia and control groups. The precision of self-reported medication use was also explored (N = 7625). In an epidemiologic study of all cohort members, subjects with a psychiatric disorder participated less actively than those without one. In the psychiatric field study, non participants were more often patients with schizophrenia than other psychoses. The psychiatric symptoms of non-participants were more severe and they needed more hospital care than participants. The use of IPW led to higher estimates of cerebrospinal fluid volume and lower estimates of grey and white matter volumes in schizophrenia patients, and increased the statistical significance of the differences in brain volume estimates between the schizophrenia and control groups. The precision of self-reported data on psychoactive medication use was substantial. Due to non-participation, the true prevalence of psychiatric disorders is probably higher than the prevalence estimates from field studies that are based on data provided by participants only. In order to reflect the true differences in the target population, weighting methods can be used to improve estimates affected by non-participation. Regarding psychoactive medication use, data collected by postal questionnaire can be assumed accurate enough for study purposes. However, it may underestimate the prevalence of medication use due to non-participation. / Abstract Tutkimusaineisto valitaan tutkimuksen tavoitteiden perusteella. Tavoitteena on kerätä kattava ja virheetön aineisto. Käytännössä kuitenkin osa tiedoista voi puuttua tai olla virheellistä. Tässä väitöskirjassa esitellään yleisesti menetelmiä huomioida puuttuva tieto analyyseissä ja arvioida aineistojen luotettavuutta psykiatrisessa tutkimuksessa. Aineisto perustuu Pohjois-Suomen vuoden 1966 syntymäkohorttiin (N = 12058). Väitöskirjassa tutkittiin, miten psykiatrinen sairastavuus vaikuttaa osallistumiseen epidemiologisessa tutkimuksessa, joka sisälsi kyselyitä ja terveystutkimuksen (N = 11540), sekä erosiko psykiatriseen kenttätutkimukseen osallistuneiden ja osallistumattomien psykoosipotilaiden kliininen taudinkuva toisistaan (N = 145). Käänteisen todennäköisyyden painotusmenetelmää käytettiin korjaamaan puuttuvan tiedon aiheuttamaa virhettä aivovolyymien estimaateissa skitsofreniapotilailla. Lisäksi arvioitiin itse ilmoitetun lääkekäyttötiedon luotettavuutta (N = 7625). Epidemiologisessa tutkimuksessa ne kohortin jäsenet, joilla oli jokin psykiatrinen sairaus, osallistuivat passiivisemmin kuin ne, joilla ei ollut psykiatrista sairautta. Psykoosipotilaat, jotka eivät osallistuneet psykiatriseen kenttätutkimukseen, sairastivat tutkimukseen osallistuneita useammin skitsofreniaa kuin muita psykooseja ja heidän taudinkuvansa oli vakavampi. Painottaminen kasvatti aivonesteen ja alensi harmaan ja valkean aineen tilavuuksien estimaatteja skitsofreniapotilailla, ja lisäsi aivovolyymien erojen tilastollista merkitsevyyttä skitsofreniapotilaiden ja vertailuhenkilöiden välillä. Itse ilmoitetun psykoaktiivisten lääkkeiden käyttötiedon luotettavuus oli merkittävä. Kadosta johtuen psykiatristen sairauksien todellinen vallitsevuus on todennäköisesti korkeampi kuin vallitsevuuden estimaatit, jotka on laskettu tutkimukseen osallistuneiden tiedoista. Painotusmenetelmiä voidaan käyttää parantamaan puuttuvan tiedon vääristämiä estimaatteja, koska painottamalla huomioidaan todellisia eroja kohdeväestössä. Tutkittaessa lääkekäyttötietoa postikyselyillä kerätyn aineiston voidaan olettaa olevan laadultaan riittävä tutkimustarpeisiin.

Abstract The duration of untreated psychosis (DUP) and its association with the course of illness in schizophrenia has been widely addressed, but the conclusions still remain essentially unclear. The studies with unselected samples and longitudinal data are few and needed. The aim of this dissertation was to study DUP and the long-term course of illness in schizophrenia in a meta-analysis and in the Northern Finland 1966 Birth Cohort (NFBC 1966). A meta-analysis of 31 original studies on the long-term association between DUP and clinical and social outcomes showed that long DUP had a small but statistically significant association with poor symptomatic, social, and global outcomes. However, there was no significant correlation between DUP and use of treatment, employment, and quality of life. In the NFBC 1966, the associations between DUP and the outcomes (n=89) were assessed with several measurements of mental and psychical state, treatment factors, education, and ability to work. In the short-term outcome, long DUP was associated with longer first hospitalization and increased risk of re-hospitalization. In the long-term, long DUP was associated only with a decreased rate of disability pension. When the association between DUP and brain structures using magnetic resonance imaging (MRI) was analyzed (n=46), long DUP correlated with decreased gray-matter density in the right hippocampus. DUP may be a modest marker of a more severe clinical phenotype regarding early outcome, but in the NFBC 1966, longer DUP seems to associate with better clinical and occupational outcomes. The relatively modest effect of DUP on outcomes indicates that shortening DUP may have positive effects on the long-term clinical course. However, it is unlikely that early intervention alone could significantly improve the overall long-term outcome of schizophrenia. Early detection of psychosis is one of the possibilities to achieve the improvement, especially in the short-term outcome of schizophrenia. / Tiivistelmä Hoitamattoman psykoosin kestoa ja sen yhteyttä skitsofrenian taudinkulkuun on tutkittu laajasti, mutta johtopäätökset asiasta ovat jääneet olennaisilta osilta epävarmoiksi. Aihetta koskevia valikoitumattomien aineistojen pitkittäisseurantoja on olemassa vain vähän. Tämän osajulkaisuväitöskirjan tavoitteena oli tutkia hoitamattoman psykoosin keston ja skitsofrenian taudinkulun pitkäaikaista yhteyttä meta-analyysin sekä Pohjois-Suomen vuoden 1966 syntymäkohorttiaineiston avulla. Meta-analyysi hoitamattoman psykoosin keston ja kliinisen ja sosiaalisen ennusteen välillä osoitti, että pidentynyt hoitamattomuus on tilastollisesti merkitsevästi yhteydessä vakavampiin oireisiin ja huonompaan sosiaaliseen toimintakykyyn sekä kokonaisennusteeseen pitkäaikaisseurannoissa. Merkitsevää yhteyttä ei ollut hoitamattoman psykoosin keston ja sairaalahoitojen määrän, työllistymisen tai elämänlaadun välillä. Pohjois-Suomen vuoden 1966 syntymäkohortissa tutkittiin hoitamattoman psykoosin yhteyttä taudinkulkuun ja ennusteeseen (n=89) ja aivojen rakenteeseen (n=46). Kahden ensimmäisen vuoden seurannassa pitkä hoitamattomuus oli yhteydessä pidempään ensimmäiseen sairaalahoitoon ja uuden sairaalahoidon kohonneeseen riskiin. Pitkäaikaisseurannassa pidentynyt hoitamattomuus oli merkitsevästi yhteydessä vähäisempään työkyvyttömyyseläkkeellä olemiseen. Kun aivojen rakennetta tutkittiin magneettikuvantamisen (MRI) avulla, löytyi yhteys pidentyneen hoitamattomuuden ja harmaan aineen pienemmän tiheyden välillä oikean hippokampuksen alueella. Hoitamattoman psykoosin keston tilastollisesti merkitsevä yhteys taudin eri ennusteisiin viittaa siihen, että psykoosin varhaisella tunnistamisella voi olla myönteisiä vaikutuksia. Muut mahdollisuudet parantaa skitsofrenian pitkäaikaisennustetta lienevät keskeisiä ja voivat tehostaa varhaisen tunnistamisen vaikutuksia.

Abstract The Northern Finland 1966 Birth Cohort is an unselected, population-based sample of 12,058 live born children. The present study is based on 10,934 individuals living in Finland at the age of 16 years. Ninety-six research diagnoses fulfilling operational DSM-III-R criteria for schizophrenia by age 34 years were found in the reassessment of clinical diagnoses. Of these 96 cases, 55 (57%) had concordant diagnoses (both the clinical and research diagnosis was schizophrenia) and 41 (43%) had discordant diagnoses (the clinical diagnosis was other than schizophrenia). Diagnostic discordance was associated with low parental social class in 1980, later age at onset, comorbid diagnosis of mental retardation, shorter treatment periods and lower number of treatment episodes. Unwanted pregnancy and parental history of psychosis increased the risk for schizophrenia. The combination of unwantedness of pregnancy and parental history of psychosis elevated the risk of schizophrenia over 8-fold in offspring compared to those without either risk factor. Both low and high birth weight increased the risk of later schizophrenia. Both short and tall babies also had elevated risk. A reverse J-shape curve described the associations between birth weight, length and schizophrenia. The ages when cohort members learned to stand, walk and became potty-trained were related to subsequent incidence of schizophrenia and other psychoses. Earlier milestones reduced and later milestones increased the risk in a linear manner. In conclusion, these results indicate that schizophrenia has complex phenomenology and developmental pathways. Its multiple symptomatology with no single specific defining feature and no absolute validation criteria makes the phenomenological-based diagnosis of schizophrenia challenging. Unwanted pregnancy may act as an additive factor for subjects already vulnerable to schizophrenia and psychoses may have a developmental dimension expressed as deviant foetal development and delayed milestones. These results support the neurodevelopmental theory suggesting that schizophrenia results from small developmental deviances with origins very early in development. / Tiivistelmä Tämän tutkimuksen tavoitteena oli tutkia skitsofrenian diagnostiikan osuvuuteen liittyviä tekijöitä sekä raskauden ei-toivottavuuden, lapsen syntymäpainon ja -pituuden ja skitsofrenian sekä varhaislapsuuden kehityksen ja skitsofrenian ja muiden psykoosien välistä yhteyttä. Pohjois-Suomen vuoden 1966 syntymäkohortti on prospektiivinen, epidemiologinen seurantatutkimus, johon kuuluu 12,058 vuonna1966 syntynyttä lasta. Prospektiivista tietoa on kerätty raskaudesta alkaen elinolosuhteista, elintavoista ja perheestä ja perimästä. Tämä tutkimus perustuu 10,934 henkilöön, jotka ovat asuneet 16-vuotiaana Suomessa eivätkä ole kieltäneet tietojensa käyttämistä tutkimukseen. Kliinisten diagnoosien uudelleen arvioinnissa DSM-III-R:n kriteerit täyttäviä skitsofreniatapauksia 34 ikävuoteen mennessä todettiin 96. Näistä 96 tapauksesta myös 55:n (57 %) kliininen diagnoosi oli skitsofrenia ja 41:n (43 %) kliininen diagnoosi oli muu kuin skitsofrenia. Diagnostiikan epätarkkuus liittyi matalaan sosiaaliluokkaan vuonna 1980, myöhäiseen sairastumisikään, samanaikaiseen kehitysvamma-diagnoosiin, lyhyisiin hoitoaikoihin ja vähäiseen hoitojaksojen määrään. Ei-toivottu raskaus yhdessä vanhemman psykoosisairauden kanssa lisäsi lapsen skitsofrenian riskiä yli 8-kertaiseksi. Matala ja korkea syntymäpaino lisäsivät skitsofrenian riskiä kaksinkertaiseksi ja samansuuruinen riskin kasvu havaittiin myös lapsilla, jotka olivat syntyessään lyhyitä tai pitkiä. J-muotoinen yhteys havaittiin syntymäpainon ja -pituuden sekä skitsofrenian riskin välillä. Ikä, jolloin lapsi oppi seisomaan, kävelemään ja kuivaksi, liittyi myöhempään skitsofreniaan ja sairastumiseen muihin psykooseihin. Varhainen oppiminen laski ja myöhäinen kehitys kohotti riskiä sairastua lineaarisesti. Tutkimuksen tulokset osoittivat, että skitsofrenia on monimuotoinen kliininen oireyhtymä, mikä tekee diagnostiikan haastavaksi. Raskauden ei-toivottavuus liittyneenä vanhemman psykoosisairauteen lisää lapsen skitsofrenian riskiä. Myös poikkeava syntymäpaino ja -pituus lisäävät skitsofrenian riskiä. Psykooseihin voi liittyä kehityksellinen dimensio, joka ilmenee viivästyneenä kehityksenä.

Abstract The aim of this dissertation was to study outcomes in schizophrenia and their predictors in a meta-analysis and in the Northern Finland Birth Cohort 1966 (NFBC 1966). The NFBC 1966 is an unselected, population-based cohort consisting of 12,068 pregnant women and their 12,058 live-born children. This dissertation utilises data that has been collected from medical records, national registers and from two extensive psychiatric studies conducted when the cohort members were 34 and 43 years old, including interview, neurocognitive and brain magnetic resonance imaging data, and questionnaires. Depending on the topic investigated, the sample size ranges between 43 and 103 individuals with schizophrenia. The meta-analysis found that approximately 13.5% of subjects with schizophrenia recovered both clinically and socially, and the recovery rate has not increased in recent decades. Studies from countries with poorer economic indices had higher recovery estimates. In the NFBC 1966, individuals with schizophrenia who were young and single at illness onset, who experienced an insidious onset, and who had more hospital treatment days early on, were at greater risk of a poor outcome in terms of later psychiatric hospitalisations and lack of remission. A novel finding was an association between suicidal ideations at onset and higher number of later psychiatric hospitalisations. Associations were detected between decreased gray matter density in the left frontal and limbic areas and decreased total white matter volume, and concurrent poor outcomes at 34 years. Concerning neurocognitive functioning at 34 years, better long-term verbal memory predicted a better global outcome (symptoms, hospital treatments, social relationships and working combined) and better visual memory predicted a better vocational outcome nine years later. The results of this study show that recovery is possible, but not very common in schizophrenia. Though outcomes are relatively difficult to predict, many clinically relevant predictors were observed that can be used in predicting outcome in a nearly 20-year follow-up. However, more research is needed in order to explore predictors that could possibly be modified via early interventions so as to enhance outcomes. / Tiivistelmä Tämän väitöstutkimuksen tarkoituksena oli tutkia skitsofrenian ennustetta ja ennustetekijöitä meta-analyysin ja Pohjois-Suomen vuoden 1966 syntymäkohortin avulla. Pohjois-Suomen vuoden 1966 syntymäkohortti on valikoitumaton, yleisväestöpohjainen kohortti, johon kuuluu 12 068 raskaana olevaa naista ja heidän 12 058 elävänä syntynyttä lastaan. Tässä väitöstutkimuksessa hyödynnetiin sairauskertomuksia, kansallisia rekistereitä sekä kahdessa laajassa kenttätutkimuksessa (34- ja 43-vuotistutkimukset) kerättyjä tietoja, jotka koostuvat haastatteluista, useista kyselyistä, neuropsykologisesta tutkimuksesta sekä aivojen magneettikuvauksesta. Tutkimuksen aiheesta riippuen aineiston koko eri osajulkaisuissa vaihteli 43:n ja 103:n välillä. Meta-analyysin perusteella 13,5 % skitsofreniaa sairastavista toipuu sekä kliinisesti että sosiaalisesti, eikä toipuminen ole viime vuosikymmeninä yleistynyt. Toipuneiden osuus oli suurempi köyhissä maissa. Pohjois-Suomen vuoden 1966 syntymäkohorttitutkimuksissa todettiin, että huonompi ennuste myöhempien sairaalahoitojen ja remission suhteen oli niillä, jotka olivat sairastuessaan nuoria ja naimattomia, joiden psykoosisairaus alkoi hitaasti ja joilla oli sairauden alkuvaiheissa enemmän sairaalahoitoja. Uusi löydös oli yhteys itsetuhoisten ajatusten ja myöhempien sairaalahoitojen välillä. Tiettyjen aivoalueiden tilavuuden ja rakenteen muutokset liittyivät monella tavoin samanhetkiseen taudinkuvaan 34-vuotiaana. Neurokognitiivisessa testauksessa parempi viivästetty kielellinen muisti 34-vuotiaana ennusti parempaa kokonaisvaltaista vointia (oireet, sairaalahoidot, sosiaaliset suhteet ja työssäkäynti yhdistettynä) ja parempi näönvarainen muisti ennusti työssäoloa 9 vuoden seurannassa. Tämän väitöstutkimuksen tulokset osoittavat, että skitsofreniasta toipuminen on mahdollista, vaikkakaan ei kovin yleistä. Vaikka taudinkulun ennustaminen on haastavaa, tutkimuksessa havaittiin useita kliinisesti merkittäviä tekijöitä, joilla on ennustearvoa jopa 20 vuoden seurannassa. Lisätutkimuksia kuitenkin tarvitaan, jotta löydettäisiin sellaisia ennustetekijöitä, joihin kohdistuvalla varhaisella interventiolla voitaisiin parantaa skitsofrenian ennustetta.

Background: Preeclampsia, small for gestational age (SGA), placental abruption, and fetal death are pregnancy complications linked to the utero-placental vasculature with serious consequences for maternal and infant well-being. Elevated homocysteine, a marker of cardiovascular disease risk, is postulated to play a role in placenta-mediated complications, but epidemiologic studies have reported inconsistent findings. The two primary objectives of this thesis were to 1: comprehensively investigate the association of homocysteine with placenta-mediated complications and examine modifying effects of pre-specified factors on this association, and 2: comprehensively investigate determinants of maternal homocysteine during pregnancy. Methods: A systematic review and meta-analysis of prospective studies was conducted to address thesis objective 1. The Ottawa and Kingston (OaK) Birth Cohort, a prospective cohort study that recruited pregnant women between 2002 and 2009, was used to address thesis objectives 1 and 2. Homocysteine concentration was measured between 12 and 20 weeks gestation. Analyses based on the OaK Birth Cohort consisted of multivariable regressions using restricted cubic splines to model associations with continuously distributed variables. Results: Objective 1: In an analysis of 7587 participants, a significant association between homocysteine concentration and a composite outcome of any placenta-mediated complication was observed (odds ratio (OR) for a 5 µmol/L increase: 1.63, 95% Confidence Interval (CI) 1.23-2.16) and SGA (OR 1.76, 95% CI 1.25-2.46), with potential modifying effects of the methylene tetrahydrofolate reductase (MTHFR) 677C>T variant (SGA) and high-risk pregnancy (preeclampsia). In the systematic review identifying 30 prospective cohort or nested case-control studies, a random effects meta-analysis of pooled mean differences in homocysteine between cases and controls in 28 studies revealed significantly higher means for SGA: 0.35 µmol/L (95% CI 0.19 to 0.51, I2=33%); and preeclampsia: 0.87 µmol/L (95% CI 0.52 to 1.21, I2=92%). Significant sources of heterogeneity were study region (SGA and preeclampsia), adjusting for covariates (preeclampsia), folate status (preeclampsia), and severity (preeclampsia). Objective 2: In 7587 OaK participants, factors related to favourable health status were associated with lower maternal homocysteine concentrations. Folic acid supplementation during pregnancy of >1 mg/day did not substantially increase serum folate concentration. Conclusion: This thesis suggests an independent effect of slightly higher homocysteine concentration in the early to mid-second trimester on the risk of any placenta-mediated complication, SGA, and preeclampsia. Modifying effects explain some of the variability in previous studies. Favourable preconception health status was associated with lower maternal homocysteine.

Mental health reflects an array of causal influences that span biological, psychological, and social circumstances, with resultant underlying causal pathways to poor mental health outcomes in childhood that are complex. Key features of this complexity are reciprocal interactions between person and environment that take place over time. The core of this thesis seeks to attend to the complexity of development to move the field of developmental health forward toward greater explanation, and more successful prediction and prevention. The focal point of the thesis is the psychosocial determinants of childhood mental health, the resource domain of the developing child, and the interplay between characteristics of the individual child, the family, and the community. The eventual goal is to better understand why and how socioeconomic circumstances impact on developmental health. One component of this thesis focuses on the expansion of extant developmental theory. The other component focuses on the development of an analytic strategy that more appropriately reflects the intricacies of this theoretical expansion. In the process, data are analysed, principally as a heuristic strategy, to illustrate the analytical approach needed to support the theoretical framework. The specification of a bioecological conceptual framework suitable to guide research and policy in developmental health is the first principal objective of the thesis. A critical examination of the resource framework proposed by Brooks-Gunn, Brown, Duncan, and Anderson Moore (1995) reveals it to be centred on family and community resources, but otherwise silent with respect to the physical and psychological resources of the child. The quintessential point of this thesis is that theory in developmental health must be able to account for the contribution individuals make to their own development. A modified resource framework is proposed that acknowledges financial, physical, human, and social capital, within the domains of the individual child, the family, and the community. The second principal objective of the thesis, the development of analytical methods that focus on the individual child and the complexity of data generated by this theoretical approach, is then introduced. Theory and method are thus integrated when comprehensive measures of characteristics in multiple domains across developmental periods are modeled using longitudinal data from the Western Australian Pregnancy Cohort (Raine) Study (Newnham, Evans, Michael, Stanley, & Landau, 1993). The mothers of 2,860 children were enrolled at 18 weeks in pregnancy and the children have been followed at birth, one, two, three, five, and eight years of age. Eighty-nine per cent (2,537 /2,860) of families were available for follow-up at eight and 74 per cent (2,126/2,860) of families responded. Extensive demographic, psychological, and developmental data were available for the children and their families and a limited amount of data were available for the communities in which they reside. A measure of mental health morbidity, the Child Behaviour Checklist (Achenbach, 1991), was available for the children at two, five, and eight years of age. In the first instance, dichotomous summary variables are derived for the demographic, psychological, and developmental variables of interest. Variables are then selected for inclusion in one of several explanatory models. To create a mathematical representation of resource characteristics, the information for each child is concatenated as a series of binary strings. Frequency tabulation is then used to aggregate the data and odds ratios are calculated to determine the degree of risk associated with each string of code, or pattern of factors relative to a nominated mental health outcome. The results provided a scaffold from which this theoretical and analytical approach is compared and contrasted with the reviewed literature. Two principal themes of investigation are pursued. The first theme to be examined is the interplay between characteristics of the child, family, and community and the contribution children make to their own development. The specific approach models the interaction between selected characteristics of the child, family and community in each of four developmentally significant time periods. The theoretical position adopted in the present study suggests that the effect of any personal or contextual factor on later development, if a relationship does truly exist, is most likely to be differential. That is, it is a combination of influences that determines developmental outcomes for children, not any single factor acting independently. The modelling process demonstrates that, for the children involved, personal and contextual factors impact mental health differentially depending on various other individual, family and/or community characteristics. The modelling process identifies patterns of factors that impact relatively small, but significant, numbers of children because the models focus on the effect for individual children rather than the effect for the group. For example, one model suggests that the effect of intra-uterine growth restriction for the group as a whole may be minimal, but the impact for some children could be critical depending on the combination of family and community influences, such as the mothers level of education, the family’s experience of significant life stress, and residence in a relatively disadvantaged community. The second theme to be examined is the possibility that the accumulation of resource deficits or risk characteristics, over time, amplifies the likelihood of mental health problems in childhood. The approach models selected characteristics of the child in each of the four periods of development collectively, and it also models selected characteristics spanning each of the four time periods discretely. The results suggest that latency, pathway, and recency effects may operate simultaneously, and that timing and accumulated burden may both be important determinants of risk. For example, with regard to children whose family experienced life stress, these three effects operated in a systematic way to increase the degree of risk of a mental health problem. In summary, the aggregation of data at the individual level is a productive approach in seeking to explain population level social phenomena. While seemingly paradoxical, the identification of the joint, interactive effects between individual, family, and community characteristics, better allows for the quantification of family and community characteristics operating through multiple causal pathways.

Abstract Schizophrenia and other psychotic disorders are severe and disabling mental disorders that break out during early adulthood, often when a person is in his/her early 20s. Furthermore, functional decline in many cognitive areas, including the ability to communicate in social interactions and impaired facial expression recognition, is typical to patients with schizophrenia. Understanding the risk factors of psychosis is essential as these disorders may be more amenable to treatment in their early stages. However, recognition of those at the highest risk of psychosis is challenging as no definitive biomarkers are available. Functional MRI is a promising tool that can potentially identify neural signals relating to the individual’s risk of psychosis onset. Psychotic disorders are etiologically heterogeneous disorders — both environmental and genetic factors have been linked to the onset of psychotic disorders. The most influential risk factor for a psychotic disorder is familial risk with genetic loading. The present study examines whether familial risk of psychosis (FR), the polygenic risk score for schizophrenia (PRS) and early adversity associate with brain response to faces. We used fMRI to measure blood oxygen level dependent (BOLD) response to faces. Our study showed that FR associated with deviant prefrontal cortex BOLD responses. In addition, we detected that interregional BOLD signal and grey matter volume varied as a function of PRS; the lowest functional and structural covariance was detected in individuals with high PRS. We also detected that early adversities associated with brain response to faces and that this association varied as a function of glucocorticoid receptor gene expression. Our findings indicate that the above risk factors of psychosis associate with brain response to faces. / Tiivistelmä Skitsofrenia ja muut psykoosisairaudet ovat vakavia mielenterveyden häiriöitä, jotka puhkeavat usein nuorella aikuisiällä. Eräs tyypillinen piirre psykoosisairauksille on vaikeus tunnistaa muiden ihmisten kasvonilmeitä. Psykoosisairauksien riskitekijöiden ymmärtäminen on tärkeää, sillä hoito tehoaa parhaiten sairastumisen alkuvaiheessa. Suurimmassa psykoosivaarassa olevien henkilöiden tunnistaminen on kuitenkin haastavaa, sillä luotettavia tautiin liittyviä biomarkkereita ei ole saatavilla. Toiminnallinen magneettikuvaus (fMRI) on lupaava työkalu, jolla saattaa olla tulevaisuudessa käyttöarvoa psykoosivaaraan liittyvien aivomuutosten tunnistamisessa. Etiologialtaan psyykoosisairaudet ovat heterogeenisiä: sekä ympäristö että perinnölliset tekijät vaikuttavat yksilön sairastumisriskiin. Voimakkain riskitekijä on suvullinen psykoosialttius. Tässä osajulkaisuväitöskirjassa tutkitaan suvullisen psykoosialttiuden, skitsofrenian polygeenisen riskipisteen (PRS) sekä varhaisten vastoinkäymisten yhteyttä aivojen kasvonilmeitä tulkitsevaan järjestelmään. Tutkimuksessa on hyödynnetty fMRI-kuvausta kasvonilmestimuluksen aikana. Tutkimuksessamme suvullinen psykoosialttius oli yhteydessä etuotsalohkon fMRI-signaalimuutoksiin. Tämän lisäksi havaitsimme, että kasvonilmejärjestelmän fMRI-signaalin ja harmaan aineen kovarianssi oli yhteydessä PRS:ään: matalin aivoalueiden välinen korrelaatio havaittiin henkilöillä, joiden PRS oli korkea. Havaitsimme myös, että varhaiset vastoinkäymiset ovat yhteydessä kasvonilmeiden aikaansaamiin aivovasteisiin. Tämä assosiaatio oli myös yhteydessä glukokortikoidireseptorin geenin ilmentymiseen. Väitöskirjan löydökset viittaavat siihen, että edellä mainitut psykoosin riskitekijät ovat yhteydessä kasvonilmeitä tulkitsevaan järjestelmään.