Hypertensive disorders in pregnancy and risk of severe mental disorders in the offspring in adulthood: The Helsinki Birth Cohort Study

Journal of Psychiatric Research 46 (2012) 303e310

Contents lists available at SciVerse ScienceDirect

Journal of Psychiatric Research journal homepage: www.elsevier.com/locate/psychires

Hypertensive disorders in pregnancy and risk of severe mental disorders in the offspring in adulthood: The Helsinki Birth Cohort Study Soile Tuovinen a, *, Katri Räikkönen a, Anu-Katriina Pesonen a, c, Marius Lahti a, Kati Heinonen a, Kristian Wahlbeck b, f, Eero Kajantie b, c, Clive Osmond d, David J.P. Barker d, Johan G. Eriksson b, e, f, g, h a

Department of Behavioural Sciences, University of Helsinki, Helsinki, Finland National Institute for Health and Welfare, Helsinki, Finland Hospital for Children and Adolescents, Institute of Clinical Medicine, University of Helsinki, Helsinki, Finland d MRC Epidemiology Resource Centre, University of Southampton, Southampton, UK e Department of General Practice and Primary Health Care, Institute of Clinical Medicine, University of Helsinki, Helsinki, Finland f Vasa Central Hospital, Vasa, Finland g Folkhälsan Research Centre, Helsinki, Finland h Helsinki University Central Hospital, Unit of General Practice, Helsinki, Finland b c

a r t i c l e i n f o

a b s t r a c t

Article history: Received 29 August 2011 Received in revised form 8 November 2011 Accepted 23 November 2011

Background: Hypertensive disorders may affect the fetal developmental milieu and thus hint at mechanisms by which prenatal adversity associates with mental disorders in later life. We examined if hypertension without proteinuria and preeclampsia in pregnancy predict serious mental disorders in the offspring, and if sex, childhood socioeconomic status, length of gestation and parity modify these associations. Methods: We included 5970 women and men born after a normotensive, hypertensive or preeclamptic pregnancy defined by using mother’s blood pressure and urinary protein measurements at maternity clinics and birth hospitals. Mental disorders requiring hospitalization or contributing to death were identified from the Finnish Hospital Discharge and Causes of Death Registers between years 1969 and 2004. Results: In comparison to the offspring born after normotensive pregnancies, offspring born after pregnancies complicated by hypertension without proteinuria were at 1.19-fold (CI: 1.01e1.41, P-value ¼ 0.04) higher risk of any mental disorder and 1.44- (CI: 1.11e1.88, P-value < 0.01) and 1.39-fold (CI: 0.99e1.93, P-value ¼ 0.05) higher risk of mood and anxiety disorder, respectively. In contrast, preeclampsia was associated, with a lower risk of any mental disorder in the male offspring (P-value ¼ 0.02; P-value ¼ 0.04 for interaction ‘sex  normotension/preeclampsia’). Conclusions: Hypertension without proteinuria in pregnancy was associated with a higher risk of serious mental disorders in the offspring in adulthood. Preeclampsia seems to, in turn, associate with lower risk of severe mental disorders in male offspring. Ó 2011 Elsevier Ltd. All rights reserved.

Keywords: Hypertension without proteinuria Preeclampsia Mental disorders Offspring Lifespan Prospective

1. Introduction Growing empirical evidence suggests that a suboptimal prenatal environment, reflected in smaller body size at birth and/or shorter length of gestation, associates with the risk of any mental disorder (Abel et al., 2010), and specific disorders including schizophrenia (Abel et al., 2010; Wahlbeck et al., 2001; Cannon et al., 2002; Nilsson et al., 2005; St Clair et al., 2005; Byrne et al., 2007), personality disorder (Hoek et al., 1996; Neugebauer et al., 1999; * Corresponding author. Tel.: þ358 9 191 29 524; fax: þ358 9 191 29 52. E-mail address: soile.tuovinen@helsinki.fi (S. Tuovinen). 0022-3956/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.jpsychires.2011.11.015

Lahti et al., 2010), mood disorder (Abel et al., 2010; Patton et al., 2004; Costello et al., 2007; Raikkonen et al., 2008) and substance use disorder (Abel et al., 2010) in later life. These findings lend credence to the Developmental Origins of Health and Disease (DOHaD) -hypothesis (Barker et al., 2005). According to this hypothesis, a suboptimal prenatal environment may permanently alter organ structure and function of the body’s biological feedback systems and render an individual susceptible to diseases later in life. Yet, the mechanisms through which these prenatal influences manoeuvre are not well understood. Hypertensive disorders in pregnancy, including chronic hypertension, gestational hypertension, and (pre)eclampsia, complicate

304

S. Tuovinen et al. / Journal of Psychiatric Research 46 (2012) 303e310

approximately 10e16% of all pregnancies (Allen et al., 2004; Roberts et al., 2005). These disorders threaten the health and wellbeing of both the mother and the fetus, and are among the key underlying causes of prematurity and intrauterine growth restriction (IUGR) (Himmelmann et al., 1996; Gray et al., 1999). Thus, by compromising the prenatal developmental environment, hypertensive disorders may provide an experimental platform to study and inform why smaller body size at birth and shorter length of gestation may be linked with mental disorders. Yet, studies testing if hypertensive disorders in pregnancy are associated with mental disorders in the offspring are scarce and the findings are controversial. Preeclampsia has been shown to be associated with a higher risk of schizophrenia in offspring in adolescence (Kendell et al., 1996; Doherty et al., 2006) and adulthood (O’Dwyer, 1997), and our recent study demonstrated that preeclampsia was associated with depressive symptoms in the offspring at age 60 (Tuovinen et al., 2010). According to another recent study, gestational hypertension associated with a higher level of internalizing behaviours, including withdrawal and anxious/depressed behaviour and somatic complaints, in the offspring at age 14 (Robinson et al., 2009). In contrast to expected, however, preeclampsia was associated with a lower level of internalizing behaviours in the offspring at the ages of 5 and 8 years (Robinson et al., 2009). Another study reported that preeclampsia was not associated with anxiety disorder and/or depression in the offspring at 20e30 years of age (Berle et al., 2006). Accordingly, our large, longitudinal epidemiological study examined if different hypertensive disorders in pregnancy associate in the offspring with an increased risk of mental disorders from early to late adulthood severe enough to warrant or contribute to hospitalisation or be the underlying, intermediate or contributing cause of death in men and women who participated in the Helsinki Birth Cohort Study (HBCS). Additionally, we tested if any potential associations between hypertensive disorders in pregnancy and mental disorders differed according to sex, childhood socioeconomic status, length of gestation and parity. The rationale for testing modulation by these factors was motivated by studies showing that the occurrence of mental disorders may vary by sex (Kessler et al., 1994; Bijl et al., 1998; Leray et al., 2010) and by social class (Fryers et al., 2005), and the occurrence of hypertensive disorders in pregnancy may vary according to social class (Gudmundsson et al., 1997; Silva et al., 2008). Further, hypertensive disorders, especially preeclampsia, that occur in preterm pregnancy are on average more severe and may be different in etiology than disorders occurring at term (Skjaerven et al., 2005; Luo et al., 2007). Finally, hypertensive disorders appear to occur more frequently and may be qualitatively different in primi- than in multiparous pregnancies (Allen et al., 2004; Luo et al., 2007). 2. Materials and methods 2.1. Participants The Helsinki Birth Cohort Study (HBCS) comprises 13,345 men and women who were born as singletons between 1934 and 1944 in one of the two maternity hospitals in Helsinki. These men and women attended child welfare clinics during childhood, and were still living in Finland in 1971, by which time a unique personal identification number had been assigned to each resident of the country. The HBCS has been described in detail elsewhere (Eriksson et al., 2001), and has been approved by the Ethics Committee of the National Public Health Institute. Maternal blood pressure and urinary protein tests during pregnancy were available from 6410 participants (48% of the initial population). These data were used to diagnose hypertensive

disorders (see below the definition of Hypertensive disorders). Of those individuals with available data, 233 participants were excluded due to missing information on gestational age; 2 participants were excluded due to imprecise data on hospitalization for mental disorder; 32 participants were excluded because they had died and the year of death was not recorded; 173 participants were excluded due to missing data on socioeconomic status in childhood. Hence, the analytic sample comprised 5970 participants (3131 men and 2839 women). The included and excluded participants had a similar sex distribution, birth weight, maternal height and weight at delivery and they had equally often obtained a diagnosis of any mental disorder over the follow-up (P-values  0.22). The included participants had 0.3 years younger mothers (P-value < 0.01), 50.8% of the included participants were primiparous vs. 46.8% of the excluded participants (P-value < 0.001), 64.1% of the included participants were born to lower social classes vs. 54% of the excluded participants (P-value < 0.001), and the included participants were more often born between years 1937e40 (the percentages of subjects who were included were 44.4%, 52.3% and 40.5% among participants who were born between 1934e36, 1937e40 and 1941e44, respectively; P-value < 0.001). 2.2. Measures 2.2.1. Hypertensive disorders For identifying pregnancy hypertensive disorders, we used mothers’ blood pressure and urinary protein measurements recorded at antenatal clinics or at birth hospital. These data and definitions of pregnancy hypertensive disorders have been described in detail (Kajantie et al., 2009). Briefly, all pregnant women were encouraged to attend the antenatal clinics, which were introduced in Helsinki from 1928 onwards (Pelkonen, 1940). Of the 13,345 subjects in the original cohort, these maternal data were available for 6410 (48.0%). For these mothers, there were on average 2.0 blood pressure and 2.5 urinary protein measurements recorded in each pregnancy. Based on this information, we defined three groups of mothers: (1) mothers with preeclampsia, with proteinuria ‘þ’ (approximates to 1 mg/mL of albumin) (Pelkonen, 1940) and systolic blood pressure 140 mmHg or diastolic pressure 90 mmHg (includes also superimposed preeclampsia); (2) mothers with hypertension without proteinuria, with systolic pressure 140 mmHg or diastolic pressure 90 mmHg (includes gestational and chronic hypertension); (3) normotensive mothers with neither systolic pressure attaining 140 mmHg nor diastolic pressure 90 mmHg. In this study sample there were 108 participants born to normotensive mothers with proteinuria. These definitions are consistent with the National High Blood Pressure Education Program Working Group on High Blood Pressure in Pregnancy (NHBPEP) 2000 criteria (Kajantie et al., 2009; National High Blood Pressure Education Program Working Group on High Blood Pressure in Pregnancy, 2000), with two exceptions: first, we considered one high blood pressure measurement to be sufficient for diagnosis because our data did not allow us to require two separate measurements as the NHBPEP criteria; second, our data included only a qualitative measurement of proteinuria (Pelkonen, 1940; National High Blood Pressure Education Program Working Group on High Blood Pressure in Pregnancy, 2000). 2.2.2. Modifying and confounding variables In the analyses, we considered several factors that may modify/ confound the associations between hypertensive disorders in pregnancy and mental disorders (Raikkonen et al., 2008; Himmelmann et al., 1996; Gudmundsson et al., 1997; Silva et al., 2008; Roberts et al., 2003; Eskenazi et al., 1991). Sex, childhood socioeconomic status (lower, middle), based on father’s highest

S. Tuovinen et al. / Journal of Psychiatric Research 46 (2012) 303e310

achieved occupation extracted from birth, child welfare clinic, and school records, length of gestation (preterm/term) and parity (1st, 2nd) were tested both as modifying and confounding factors. Year of birth, birth weight (g), mother’s height (cm), weight (kg) and age, extracted from hospital birth records, were treated as confounding factors. Mother’s body mass index (BMI) was calculated as weight measured before delivery divided by height2 (kg/m2). 2.2.3. Serious mental disorders Diagnoses on mental disorders severe enough to warrant or contribute to hospital treatment were identified from the Finnish Hospital Discharge Register (HDR), and severe enough to be the underlying, intermediate, or contributing cause of death from the Finnish Causes of Death Register (CDR). The HDR and the CDR include nationwide data on all inpatient episodes and deaths of residents in Finland at an individual level between 1969 and 2004 using International Statistical Classification of Diseases (ICD) coding system. Table 1 presents the diagnostic codes. Cases with serious mental disorders were defined as having a primary or subsidiary diagnosis for mental disorders in the HDR or an underlying, intermediate and contributing mental disorder as cause of death in the CDR. The psychiatric diagnostic classification used in Finland was ICD-8 until 1986; from 1987 to 1995, the classification was ICD-9 based on DSM-III-R criteria, and from 1996, it was ICD-10. For this study, we converted ICD-8 and ICD-9 diagnoses to current ICD-10 codes. The principal outcome was first-ever diagnosis of a serious mental disorder, defined as a HDR or CDR diagnosis convertible to an ICD-10 code for a mental or behavioral disorder (ICD-10 codes

305

F10eF69; for ICD-8 and ICD-9-codes, see Table 1). Our definition of mental disorders excluded organic disorders (ICD-10: F0) and learning disabilities (ICD-10: F7). Due to the age range of participants, disorders of development and disorders typically occurring in childhood and adolescence (ICD-10: F8eF9) were also not recorded. Sub-analyses were per protocol carried out for the following diagnostic groups: mental and behavioural disorders due to psychoactive substance use (ICD-codes F10e19); schizophrenia and other non-affective psychotic disorders (ICD-codes F20eF29); mood disorders (ICD-codes F30eF39); anxiety disorders (ICDcodes F40eF48); and personality disorders (F60eF61). However, in the case of acute intoxication due to substance abuse (F1x.0 in ICD-10 and 305 in ICD-9), only the primary diagnosis was used from both registers because alcohol intoxication is a frequent subsidiary diagnosis in Finnish medical practice and does not necessarily indicate a substance use disorder. The HDR (Keskimäki and Aro, 1991) and the CDR (Lahti and Penttila, 2001) have been found to be valid and reliable tools for epidemiological research. The HDR has been validated with regard to the diagnoses of bipolar disorder (Kieseppä et al., 2000), any psychotic disorder (Perala et al., 2007), and schizophrenia (Makikyro et al., 1998; Pihlajamaa et al., 2008). All these diagnoses show high levels of specificity in the register against medical records (Kieseppä et al., 2000; Perala et al., 2007; Makikyro et al., 1998). The validity of the alcohol psychosis- and alcoholism -diagnoses in HDR during the usage of ICD-8 has also found support in one earlier study (Poikolainen, 1983). We are not aware of studies where the validity of the other substance use disorder, or depressive, anxiety, and personality disorder diagnoses in the HDR would have been formally examined against medical records. 2.3. Statistical analyses

Table 1 Mental disorders according to International Statistical Classification of Diseases (ICD) codes used in the Finnish Hospital Discharge Register and the Finnish Causes of Death Register. Mental disorders:

International Classification of Diseases 10th (ICD-10; in use 1996-present), 9th (ICD-9; in use between 1987e1995), and 8th (ICD-8; in use between 1969e1986) diagnostic codes:

Anya

ICD-10: F1eF6 ICD-9: 291e292, 295e298, 300e304, 305, 3071A, 3074, 3075Ae3075B, 3078A, 3079X, 3090A, 3092Ce3099X, 312 ICD-8: 291, 295, 296e305, 306.4e306.5, 306.8, 306.98, 307 ICD-10: F10eF19 ICD-9: 291e292, 303e305 ICD-8: 291, 303e304 ICD-10: F20eF29 ICD-9: 295, 297e298 ICD-8: 295, 297, 298.10e299.99 ICD-10: F30eF39 ICD-9: 296, 3004A, 3011D ICD-8: 296, 298.00, 300.40, 300.41, 301.10 ICD-10: F40eF48 ICD-9: 3000Ae3003A, 3006Ae3009X, 3078A, 309 ICD-8: 300.00e300.30, 300.50e300.99, 305, 306.80, 307.99 ICD-10: F60eF61 ICD-9: 3010A, 3012Ae3015A, 3016Ae3018X ICD-8: 301.00, 301.20e301.99

Substance Use (ICD-10: Mental and behavioral disorders due to psychoactive substance use) Psychotic (ICD-10: Schizophrenia, schizotypal, and delusional disorders) Mood (ICD-10: Mood (affective))

Anxiety (ICD-10: Neurotic, stress-related and somatoform disorders)

Personality (ICD-10: Specific and mixed personality disorders)

Note. a A variety of behavioural syndromes (corresponding to codes F50eF59 or F62eF69 in ICD-10) were included in the Any mental disorder ecategory, but due to heterogeneity and small sample size, we were not able to treat it as a separate category.

We used Cox proportional hazard models as the primary data analytical tool. The participants were followed up from year 1969 to their death, migration, onset of mental disorders or December, 31, 2004. Participants whose mental disorder diagnose changed over time or with co-morbid disorders were included in each specific mental disorder category and in any mental disorder category. Thus, participants with specific mental disorders do not sum up to participants having any mental disorder. All hazard ratios were stratified according to year of birth and sex, and adjusted for gestational age, birth weight, father’s occupational status in childhood, parity, mother’s age and BMI at delivery. Because the unadjusted and adjusted associations were virtually identical, we present the adjusted associations only. Cumulative incidences of mental disorders adjusted for sex, year of birth and childhood socioeconomic background were calculated separately for offspring born after normotensive/hypertensive/preeclamptic pregnancies. We included interactions by ‘sex, father’s occupational status, preterm/term length of gestation, parity  hypertensive disorders in pregnancy’ (each interaction tested separately) followed by their main effects in hazard models testing if sex, father’s occupational status, preterm/term length of gestation and parity modified the associations. All neonatal characteristics were converted to z-scores by sex (Royston, 1991). In supplementary analyses, we tested if maternal proteinuria was associated with mental disorders. We also tested if exclusion of those participants born to normotensive mothers with proteinuria affected the results. 3. Results Table 2 shows the characteristics of the participants according to mothers’ hypertensive disorder during pregnancy. Hypertension

306

S. Tuovinen et al. / Journal of Psychiatric Research 46 (2012) 303e310

Table 2 Characteristics of the participants. Offspring born to pregnant mothers with Normotension N ¼ 3963

Hypertension N ¼ 1733

Mean (SD)/N (%)

Mean (SD)/N (%)

Gender Men 2089 Women 1874 Length of gestation (days) 279.3 Birth weight (g) 3436.3 Mother’s age 27.9 Mother’s BMI 26.0 Year of birth 1934 101 1935 171 1936 232 1937 240 1938 268 1939 337 1940 377 1941 472 1942 378 1943 789 1944 598 Parity Primiparous 1912 Multiparous 2051 Father’s occupational status in childhood Lower class 2528 Middle class 1435

(53) (47) (12.7) (464.8) (5.2) (2.8)

897 836 279.2 3392.5 29.0 26.8

Preeclampsia N ¼ 274

P

P

Mean (SD)/N (%)

(52) (48) (12.6) (497.7) (5.7) (3.2)

0.51

145 129 274.9 3037.8 28.5 26.9

0.88 0.001 <0.001 <0.001 <0.001

(1.0) (2.6) (5.5) (7.0) (8.0) (10.4) (9.9) (14.3) (5.1) (20.7) (15.6)

0.95 <0.001 <0.001 0.08 <0.001 <0.001

(2.5) (4.3) (5.9) (6.1) (6.8) (8.5) (9.5) (11.9) (9.5) (19.9) (15.1)

17 45 95 122 139 180 171 247 89 358 270

(48) (52)

929 (54) 803 (46)

<0.001

191 (70) 83 (30)

<0.001

(64) (36)

1139 (66) 594 (34)

0.16

159 (58) 115 (42)

0.06

without proteinuria or preeclampsia was not associated with sex of the offspring or with father’s occupational status in childhood, hypertension without proteinuria was not associated with length of gestation and preeclampsia was not associated with mother’s age at delivery. However, participants whose mothers had hypertension without proteinuria in pregnancy had lower birth weight and were more often born in later years, to older or to primiparous mothers and to mothers with a higher BMI at delivery. Participants whose mothers had preeclampsia in pregnancy had shorter length of gestation and lower birth weight, were more often born in earlier years and to primiparous mothers and to mothers with higher BMI. Table 3 shows the number of participants with mental disorders according to maternal hypertensive disorders. Before proceeding to analyses targeting the major study questions, we tested if the risk of mental disorders varied by sex, year of birth, father’s occupational status in childhood, length of gestation, parity, birth weight and mother’s age or BMI at delivery. The risk of any mental (1.70, Confidence Interval (CI): 1.46e1.99, P-value < 0.001), substance use

3 19 22 28 25 37 33 35 19 33 20

(53) (47) (16.0) (580.7) (5.6) (3.1) (1.0) (6.9) (8.0) (10.2) (9.1) (13.5) (12.0) (12.8) (6.9) (12.0) (7.3)

(4.73, CI: 3.60e6.21, P-value < 0.001) and personality disorders (1.60, CI: 1.05e2.43, P-value ¼ 0.03) were higher for men than for women. The risk of any mental (1.18, CI: 1.00e1.38, P-value ¼ 0.04) and substance use disorders (1.25, CI: 1.00e1.55, P-value ¼ 0.05) were higher for subjects who were born to fathers with a lower compared to a middle occupational status. Sex and father’s occupational status were not associated with other mental disorders among this sample (P-values > 0.07). Year of birth, preterm/term length of gestation, parity, birth weight, mother’s age or BMI at delivery were not associated with any or the more specific mental disorders (P-values > 0.15). 3.1. Maternal hypertension without proteinuria/preeclampsia in pregnancy and mental disorders Table 3 shows that in comparison to the offspring born after normotensive pregnancies, offspring born after pregnancies complicated by hypertension without proteinuria were at 1.19-fold

Table 3 Risk of severe mental disorder of offspring born after normotensive vs. hypertensive/preeclamptic pregnancies. Disorders

No disorders vs. Any Substance use Psychotic Mood Anxiety Personality

N

5264 706d 379 142 266 166 94

Normotension %a

11.4 6.6 2.4 4.4 2.8 1.7

Hypertension %a

13.2 7.4 3.2 5.9 3.8 1.8

Preeclampsia %a

8.8 4.7 2.5 4.1 2.2 1.7

Normotension vs. hypertension

Normotension vs. preeclampsia

Hazard ratiob (95% CI)c

P

Hazard ratiob (95% CI)

1.19 1.11 1.38 1.44 1.39 1.09

0.04 0.39 0.08 0.007 0.05 0.73

0.75 0.61 1.15 0.96 0.86 1.12

(1.01, (0.88, (0.96, (1.11, (0.99, (0.68,

1.41) 1.40) 1.99) 1.88) 1.93) 1.73)

(0.48, (0.32, (0.49, (0.48, (0.35, (0.39,

1.17) 1.17) 2.69) 1.91) 2.16) 3.17)

c

P 0.20 0.14 0.75 0.90 0.75 0.84

Referent: Normotension (1.00). a % refers to the cumulative incidence of severe mental disorders according to mothers’ hypertensive disorder during pregnancy derived from models adjusting for sex, year of birth and father’s occupational status in childhood. b Cox regression models are stratified for sex and year of birth and adjusted for gestational age, weight at birth, father’s occupational status in childhood, parity, mother’s age and BMI at delivery. c 95% CI refers to 95% confidence interval. d The N of specific severe mental disorders does not sum up to having any disorder, because of change in diagnostic symptoms over time and/or co-morbidity.

S. Tuovinen et al. / Journal of Psychiatric Research 46 (2012) 303e310

(CI: 1.01e1.41, P-value ¼ 0.04) increased risk of any mental disorder and at 1.44-fold (CI: 1.11e1.88, P-value ¼ 0.007) increased risk of mood disorder. They were also at 1.39-fold (CI: 0.99e1.93, P-value ¼ 0.05) increased risk of anxiety disorder which approached the level of significance. Maternal hypertension without proteinuria was not associated with other mental disorders (P-values > 0.08). Maternal preeclampsia was not associated with any or the more specific mental disorders (P-values > 0.14). 3.1.1. Modulation by sex, father’s occupational status in childhood, preterm/term length of gestation and parity Men born after pregnancies complicated by hypertension without proteinuria, compared to men born after normotensive pregnancies, were at 1.50- (CI: 1.05e2.16, P-value ¼ 0.03) and 1.80fold (CI: 1.14e2.84, P-value ¼ 0.01) increased risks of mood and anxiety disorder, respectively. In women, maternal hypertension without proteinuria was not associated with mental disorders (Pvalues > 0.11). The interaction ‘normotension vs. hypertension without proteinuria  sex’ was not significant for mood disorder (Pvalue ¼ 0.43), but approached significance for anxiety disorder (Pvalue ¼ 0.07). Men born after pregnancies complicated by maternal preeclampsia, compared to men born after normotensive pregnancies had, in contrast, a lower, a 0.45-fold (CI: 0.23e0.88, P-value ¼ 0.02) risk of any mental disorder and also had a lower, 0.44-fold (CI: 0.19e0.99, P-value ¼ 0.05) risk of substance use disorder. In women, maternal preeclampsia was not associated with mental disorders (Pvalues > 0.24). The interaction ‘normotension vs. preeclampsia  sex’ was significant for any mental disorder (P-value ¼ 0.04) (Fig. 1, Panel A) and approached significance in the analyses of substance use disorder (P-value ¼ 0.06) (Fig. 1, Panel B). The interactions ‘normotension vs. hypertension without proteinuria/preeclampsia  lower vs. middle father’s occupational status in childhood preterm/term length of gestation/parity’ were not significant in tests of mental disorders (P-values > 0.14). Maternal proteinuria was not associated with mental disorders (P-values > 0.21). Excluding participants born to normotensive mothers with proteinuria from the analyses had little effect on any results. 4. Discussion We found that individuals who were born after pregnancies complicated by hypertension without proteinuria were at a higher risk of any mental disorder and of mood disorder, and tended to have a higher risk also for anxiety disorder. These disorders were serious and were identified from the Finnish nationwide registers

307

of hospitalizations and deaths from early to late adulthood across over three decades. This risk was particularly emphasized in men born after hypertensive pregnancies without proteinuria who showed a significantly elevated risk of mood and anxiety disorders. While only significant in men, the sex difference in the associations ought to be interpreted with caution, since the interaction ‘normotension vs. hypertension  sex’ in the analyses of mental disorders only approached statistical significance. These findings are compatible with the DOHaD-hypothesis (Barker et al., 2005) and suggest that hypertension without proteinuria in pregnancy, by compromising the fetal developmental milieu, may pose a risk of serious mental disorders in subsequent life. The slightly higher risk observed in men than in women is also compatible with the proposition (Barker, 1998) and some empirical findings (Brown et al., 1995, 2000; Thompson et al., 2001) suggesting that men might be overall more vulnerable than women to the ill health consequences of early environmental adversities. While our findings may hint at mechanisms by which prenatal adversity associates with mental disorders in subsequent life, the exact mechanisms linking hypertension without proteinuria in pregnancy with higher risk of mental disorders in subsequent life remain to be determined. This is an obvious and major study limitation. Hypertension in pregnancy may induce fetal malnutrition and oxidative distress (Roberts and Redman, 1993; Sibai et al., 2005) and activate similar sympathetic nervous system responses in the mother that are often seen under psychosocial stress (Schobel and Grassi, 1998; Rang et al., 2002). Genetic, epigenetic, and other physiological and psychosocial mechanisms may also be involved, but we cannot address any of these mechanisms in the current study. In contrast to the study hypotheses, our findings revealed that maternal preeclampsia was associated with a lower risk of any mental disorder and of substance use disorder in the offspring. These associations were seen in men only, and interaction analyses by sex confirmed that the association of preeclampsia with any mental disorder was significantly different between men and women and with substance use disorder it approached significance. These associations are counterintuitive and we cannot exclude the possibility of a chance finding. It is of note, however, that hypertension and preeclampsia in pregnancy may operate and exert effects on offspring development through different mechanisms (Doherty et al., 2006; Robinson et al., 2009; Roberts and Redman, 1993; Sibai et al., 2005; Kurki et al., 2000). Not all effects need to be adverse: It is plausible that the seriousness of preeclampsia associates with more effective and sensitive parenting explaining the counterintuitive associations of preeclampsia in men (Newnham et al., 2004). In addition, it is of

Fig. 1. Panel A: The interaction ‘normotension vs. preeclampsia  sex’ for any mental disorder (P-value ¼ 0.04). Panel B: The interaction ‘normotension vs. preeclampsia  sex’ for substance use disorder (P-value ¼ 0.06). Analyses were stratified for sex and year of birth and adjusted for gestational age, weight at birth, father’s occupational status in childhood, parity, mother’s age and BMI at delivery. % refers to the cumulative incidence of any mental disorder and substance use disorder according to mothers’ hypertensive disorder (normotension vs. preeclampsia) during pregnancy derived from models adjusting for year of birth and father’s occupational status in childhood.

308

S. Tuovinen et al. / Journal of Psychiatric Research 46 (2012) 303e310

note, that authoritative parenting (compared with uninvolved) is more strongly associated with men’s psychological well-being than women’s (Rothrauff et al., 2009). It may well be that such a parenting styles may buffer from any ill effects of the preeclampsia-associated early environmental adversity. This assumption remains, however, untested in the present study. In general, our findings both agree and disagree with the findings of previous studies. In agreement with our findings, one study reported that gestational hypertension was associated in the adolescent offspring with higher internalizing behavior, including withdrawal and anxious/depressed behaviour and somatic complaints, and also in partial agreement with our findings, preeclampsia was, in turn, associated with lower internalizing behavior in childhood (Robinson et al., 2009). Our findings, however, disagree with the previous studies showing that preeclampsia is associated in the adolescent (Kendell et al., 1996; Dalman et al., 1999) and adult offspring (O’Dwyer, 1997) with a higher risk of schizophrenia, with our own previous finding showing an association between preeclampsia and higher depressive symptoms in older adults (Tuovinen et al., 2010), and with the findings showing no associations between preeclampsia and depression and/or anxiety in offspring (Berle et al., 2006). In interpreting the correspondence between the current and the previous findings the following differences should be emphasized. We examined serious mental disorders that were identified from the Finnish nationwide registers. The previous studies have used maternal (Robinson et al., 2009) and self-ratings (Tuovinen et al., 2010; Berle et al., 2006) and psychiatric interviews (O’Dwyer, 1997) as tools to measure sub-clinical symptoms and mental disorders, and are, hence, also addressing symptoms and disorders that do not require hospital treatment or contribute to death. Further, we identified serious disorders over three decades from early to late adulthood. The previous studies have examined symptoms and disorders in samples that are much younger in age than our sample and have used a single time-point in assessing the presence of mental disorders and symptoms (O’Dwyer, 1997; Tuovinen et al., 2010; Robinson et al., 2009; Berle et al., 2006), and hence do not offer a life-course perspective. Strength of our study lies in the large and well-characterized birth cohort. Yet, as expected, our study did not include a large number of offspring born after preeclamptic pregnancies, while a larger number of offspring born after pregnancies with hypertension without protenuria allowed for more statistical power. Only half of the cohort had antenatal records of hypertension and proteinuria, and were therefore eligible for the present study. Their mothers tended to be slightly shorter and younger, more of them were primiparous, and more were married to manual workers. These differences would, however, introduce a bias only if the association between hypertensive disorders in pregnancy and offspring mental disorders were different in people who have antenatal records as compared to those who do not. This seems unlikely but cannot be excluded. Moreover, our data did not allow us to require two elevated blood pressure measurements to establish diagnosis of hypertension (National High Blood Pressure Education Program Working Group on High Blood Pressure in Pregnancy, 2000). As a possible consequence of this, the cumulative incidence of hypertension disorder in pregnancy was higher than that reported in the literature. However, the overall prevalence rate of 4.6% for preeclampsia is consistent with the commonly reported rates of 3e5% (Roberts and Cooper, 2001). We have no data on the complications of preeclampsia, including eclampsia, which at that time complicated 0.6% of all pregnancies at Helsinki University Central Hospital (Parviainen, 1946) and may be associated with more severe consequences for the offspring.

Further limitations relate to the generalizability from the findings. We focused on mental disorders that required hospitalization or contributed to death, and therefore our findings may not generalize to less severe mental health disorders and sub-clinical symptoms. Further, while the HDR and the CDR have been found to be valid and reliable tools for epidemiological research, this may not pertain to all the mental disorder diagnoses. In addition, it is possible that obstetric complications predispose the mother to postnatal depression, which may in turn pose a risk for mental disorders in the offspring by genetic mechanisms or by compromising the early attachment relationship. We do not have data on maternal and not on paternal psychopathology either and therefore cannot rule out this possibility. We do not either have data available on maternal stress, smoking or alcohol consumption during pregnancy. Psychological stress during pregnancy may favor the development of hypertensive disorders in pregnancy (Paarlberg et al., 1995) while smoking during pregnancy may protect against hypertensive disorders in pregnancy (Zhang et al., 1997). Furthermore, stress (Watson et al., 1999; Pesonen et al., 2005; Van den Bergh et al., 2008), smoking (Heinonen et al., 2010), and alcohol consumption (Kuczkowski, 2007; Pesonen et al., 2009) during pregnancy are associated with less optimal developmental outcomes in the offspring. Hence, we cannot rule out that psychological stress, smoking and alcohol consumption during pregnancy may to some extent explain our findings. Although we are unaware of exact data, data on the incidence of lung cancer in women suggest that smoking during pregnancy between the years of 1934e44 was uncommon in Finland (Cancer in Finland, 2009). Finally, the HDR and CDR carry diagnoses from 1969 onwards; therefore, it is important to keep in mind that individuals hospitalized for mental causes in childhood and youth only, remain undetected. 5. Conclusion We found that hypertension without proteinuria in pregnancy was associated with a higher risk of any mental disorder and of mood disorders requiring hospitalization or contributing to death in the offspring in adulthood. Preeclampsia was, in turn, associated with a lower risk of any mental disorder in the male offspring. Role of funding source This study was supported by grants from the Academy of Finland, European Science Foundation (EuroSTRESS), University of Helsinki, the British Heart Foundation, the Finnish Foundation of Cardiovascular Research, the Finnish Diabetes Research Foundation, the Finnish Medical Society (Duodecim), Finska Läkaresällskapet, the Finnish Graduate School of Psychology, the Päivikki and Sakari Sohlberg Foundation, the Juho Vainio Foundation, the Yrjö Jahnsson Foundation, the Signe and Ane Gyllenberg Foundation, the Jalmari and Rauha Ahokas foundation, the Emil Aaltonen Foundation, the Finnish Ministry of Education and the Finnish Foundation for Paediatric Research. Contributors Author contributions: 1. Conception and design, or acquisition of data, or analysis and interpretation of data (Ms. Tuovinen, Lahti, Drs. Räikkönen, Pesonen, Heinonen, Wahlbeck, Kajantie, Osmond, Barker, Eriksson); 2. Drafting the article or revising it critically for important intellectual content (Ms. Tuovinen, Lahti, Drs. Räikkönen, Pesonen, Heinonen, Wahlbeck, Kajantie, Osmond, Barker, Eriksson); 3. Final approval of the version to be published

S. Tuovinen et al. / Journal of Psychiatric Research 46 (2012) 303e310

(Ms. Tuovinen, Lahti, Drs. Räikkönen, Pesonen, Heinonen, Wahlbeck, Kajantie, Osmond, Barker, Eriksson). Conflict of interest The authors declare no conflict of interest. Acknowledgements The authors would like to thank the staff at the National Institute for Health and Welfare and at the University of Helsinki, Department of Behavioural Sciences for their assistance with the manuscript. References Abel KM, Wicks S, Susser ES, Dalman C, Pedersen MG, Mortensen PB, et al. Birth weight, schizophrenia, and adult mental disorder: is risk confined to the smallest babies? Archives of General Psychiatry 2010;67:923e30. Allen VM, Joseph K, Murphy KE, Magee LA, Ohlsson A. The effect of hypertensive disorders in pregnancy on small for gestational age and stillbirth: a population based study. BMC Pregnancy and Childbirth 2004;4:17. Barker DJP. Mothers, babies and health in later life; 1998. Barker DJP, Osmond C, Forsen TJ, Kajantie E, Eriksson JG. Trajectories of growth among children who have coronary events as adults. The New England Journal of Medicine 2005;353:1802e9. Berle JO, Mykletun A, Daltveit AK, Rasmussen S, Dahl AA. Outcomes in adulthood for children with foetal growth retardation. A linkage study from the NordTrondelag Health Study (HUNT) and the Medical Birth Registry of Norway. Acta Psychiatrica Scandinavica 2006;113:501e9. Bijl RV, Ravelli A, van Zessen G. Prevalence of psychiatric disorder in the general population: results of The Netherlands Mental Health Survey and Incidence Study (NEMESIS). Social Psychiatry and Psychiatric Epidemiology 1998;33: 587e95. Brown AS, Susser ES, Lin SP, Neugebauer R, Gorman JM. Increased risk of affective disorders in males after second trimester prenatal exposure to the Dutch hunger winter of 1944e45. The British Journal of Psychiatry: The Journal of Mental Science 1995;166:601e6. Brown AS, van Os J, Driessens C, Hoek HW, Susser ES. Further evidence of relation between prenatal famine and major affective disorder. The American Journal of Psychiatry 2000;157:190e5. Byrne M, Agerbo E, Bennedsen B, Eaton WW, Mortensen PB. Obstetric conditions and risk of first admission with schizophrenia: a Danish national register based study. Schizophrenia Research 2007;97:51e9. Cancer in Finland. Cancer statistics of the National Institute for Health and Welfare (THL). Cancer Society of Finland; 2009. Publication No. 76 Helsinki ISSN 0585-9603. Cannon M, Jones PB, Murray RM. Obstetric complications and schizophrenia: historical and meta-analytic review. The American Journal of Psychiatry 2002; 159:1080e92. Costello EJ, Worthman C, Erkanli A, Angold A. Prediction from low birth weight to female adolescent depression: a test of competing hypotheses. Archives of General Psychiatry 2007;64:338e44. Dalman C, Allebeck P, Cullberg J, Grunewald C, Koster M. Obstetric complications and the risk of schizophrenia: a longitudinal study of a national birth cohort. Archives of General Psychiatry 1999;56:234e40. Doherty DA, Magann EF, Francis J, Morrison JC, Newnham JP. Pre-pregnancy body mass index and pregnancy outcomes. International Journal of Gynaecology and Obstetrics: The Official Organ of the International Federation of Gynaecology and Obstetrics 2006;95:242e7. Eriksson JG, Forsen T, Tuomilehto J, Osmond C, Barker DJ. Early growth and coronary heart disease in later life: longitudinal study. BMJ (Clinical Research Ed.) 2001; 322:949e53. Eskenazi B, Fenster L, Sidney S. A multivariate analysis of risk factors for preeclampsia. JAMA: The Journal of the American Medical Association 1991; 266:237e41. Fryers T, Melzer D, Jenkins R, Brugha T. The distribution of the common mental disorders: social inequalities in Europe. Clinical Practice and Epidemiology in Mental Health: CP & EMH 2005;1:14. Gray PH, O’Callaghan MJ, Harvey JM, Burke CJ, Payton DJ. Placental pathology and neurodevelopment of the infant with intrauterine growth restriction. Developmental Medicine and Child Neurology 1999;41:16e20. Gudmundsson S, Bjorgvinsdottir L, Molin J, Gunnarsson G, Marsal K. Socioeconomic status and perinatal outcome according to residence area in the city of Malmo. Acta Obstetricia et Gynecologica Scandinavica 1997;76:318e23. Heinonen K, Raikkonen K, Pesonen AK, Andersson S, Kajantie E, Eriksson JG, et al. Behavioural symptoms of attention deficit/hyperactivity disorder in preterm and term children born small and appropriate for gestational age: a longitudinal study. BMC Pediatrics 2010;10:91. Himmelmann K, Himmelmann A, Niklasson A, Svensson A. Hypertension in pregnancy and size at birth. Blood Pressure 1996;5:278e84.

309

Hoek HW, Susser E, Buck KA, Lumey LH, Lin SP, Gorman JM. Schizoid personality disorder after prenatal exposure to famine. The American Journal of Psychiatry 1996;153:1637e9. Kajantie E, Eriksson JG, Osmond C, Thornburg K, Barker DJ. Pre-eclampsia is associated with increased risk of stroke in the adult offspring: the Helsinki birth cohort study. Stroke; A Journal of Cerebral Circulation 2009;40: 1176e80. Kendell RE, Juszczak E, Cole SK. Obstetric complications and schizophrenia: a case control study based on standardised obstetric records. The British Journal of Psychiatry: The Journal of Mental Science 1996;168:556e61. Keskimäki I, Aro S. Accuracy of data on diagnoses, procedures and accidents in the Finnish Hospital Discharge Register. International Journal of Health Services 1991;2:15e21. Kessler RC, McGonagle KA, Zhao S, Nelson CB, Hughes M, Eshleman S, et al. Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States. Results from the National Comorbidity Survey. Archives of General Psychiatry 1994;51:8e19. Kieseppä T, Partonen T, Kaprio J, Lönnqvist J. Accuracy of register- and record-based bipolar I diagnoses in Finlandda study of twins. Acta Neuropsychiatrica 2000; 12:106e9. Kuczkowski KM. The effects of drug abuse on pregnancy. Current Opinion in Obstetrics & Gynecology 2007;19:578e85. Kurki T, Hiilesmaa V, Raitasalo R, Mattila H, Ylikorkala O. Depression and anxiety in early pregnancy and risk for preeclampsia. Obstetrics and Gynecology 2000;95: 487e90. Lahti M, Raikkonen K, Wahlbeck K, Heinonen K, Forsen T, Kajantie E, et al. Prenatal origins of hospitalization for personality disorders: the Helsinki birth cohort study. Psychiatry Research 2010;179:226e30. Lahti RA, Penttila A. The validity of death certificates: routine validation of death certification and its effects on mortality statistics. Forensic Science International 2001;115:15e32. Leray E, Camara A, Drapier D, Riou F, Bougeant N, Pelissolo A, et al. Prevalence, characteristics and comorbidities of anxiety disorders in France: results from the “Mental Health in General Population” Survey (MHGP). European Psychiatry: The Journal of the Association of European Psychiatrists; 2010. Luo ZC, An N, Xu HR, Larante A, Audibert F, Fraser WD. The effects and mechanisms of primiparity on the risk of pre-eclampsia: a systematic review. Paediatric and Perinatal Epidemiology 2007;21:36e45. Makikyro T, Isohanni M, Moring J, Hakko H, Hovatta I, Lonnqvist J. Accuracy of register-based schizophrenia diagnoses in a genetic study. European Psychiatry: The Journal of the Association of European Psychiatrists 1998;13:57e62. National High Blood Pressure Education Program Working Group on High Blood Pressure in Pregnancy. Report of the national high blood pressure education program working group on high blood pressure in pregnancy. American Journal of Obstetrics and Gynecology 2000;183:S1e22. Neugebauer R, Hoek HW, Susser E. Prenatal exposure to wartime famine and development of antisocial personality disorder in early adulthood. JAMA: The Journal of the American Medical Association 1999;282:455e62. Newnham JP, Doherty DA, Kendall GE, Zubrick SR, Landau LL, Stanley FJ. Effects of repeated prenatal ultrasound examinations on childhood outcome up to 8 years of age: follow-up of a randomised controlled trial. Lancet 2004;364:2038e44. Nilsson E, Stalberg G, Lichtenstein P, Cnattingius S, Olausson PO, Hultman CM. Fetal growth restriction and schizophrenia: a Swedish twin study. Twin Research and Human Genetics: The Official Journal of the International Society for Twin Studies 2005;8:402e8. O’Dwyer JM. Schizophrenia in people with intellectual disability: the role of pregnancy and birth complications. Journal of Intellectual Disability Research: JIDR 1997;41:238e51. Paarlberg KM, Vingerhoets AJ, Passchier J, Dekker GA, Van Geijn HP. Psychosocial factors and pregnancy outcome: a review with emphasis on methodological issues. Journal of Psychosomatic Research 1995;39:563e95. Parviainen S. Fréquence de la néphrogestose pendant les années de guerre [The incidence of nephrogestosis during the years of the war; in French]. Acta Obstetricia et Gynecologica Scandinavica 1946;26:171e92. Patton GC, Coffey C, Carlin JB, Olsson CA, Morley R. Prematurity at birth and adolescent depressive disorder. The British Journal of Psychiatry: The Journal of Mental Science 2004;184:446e7. Pelkonen E. deUntersuchungen über die Frequenzschwankungen der Nierengestosen in Finnland und die auf sie einwirkenden Faktoren. Acta Societatis Medicorum Duodecim 1940; Ser B29:1e139. Perala J, Suvisaari J, Saarni SI, Kuoppasalmi K, Isometsa E, Pirkola S, et al. Lifetime prevalence of psychotic and bipolar I disorders in a general population. Archives of General Psychiatry 2007;64:19e28. Pesonen AK, Räikkönen K, Matthews K, Heinonen K, Paavonen JE, Lahti J, et al. Prenatal origins of poor sleep in children. Sleep 2009;32:1086e92. Pesonen A-, Räikkönen K, Strandberg TE, Järvenpää A-. Continuity of maternal stress from the pre- to the postnatal period: associations with infant’s positive, negative and overall temperamental reactivity. Infant Behavior & Development 2005;28:36e47. Pihlajamaa J, Suvisaari J, Henriksson M, Heila H, Karjalainen E, Koskela J, et al. The validity of schizophrenia diagnosis in the Finnish Hospital Discharge Register: findings from a 10-year birth cohort sample. Nordic Journal of Psychiatry 2008; 62:198e203. Poikolainen K. Accuracy of hospital discharge data: five alcohol-related diseases. Drug and Alcohol Dependence 1983;12:315e22.

310

S. Tuovinen et al. / Journal of Psychiatric Research 46 (2012) 303e310

Raikkonen K, Pesonen AK, Heinonen K, Kajantie E, Hovi P, Jarvenpaa AL, et al. Depression in young adults with very low birth weight: the Helsinki study of very low-birth-weight adults. Archives of General Psychiatry 2008;65:290e6. Rang S, Wolf H, Montfrans GA, Karemaker JM. Non-invasive assessment of autonomic cardiovascular control in normal human pregnancy and pregnancyassociated hypertensive disorders: a review. Journal of Hypertension 2002;20: 2111e9. Roberts CL, Algert CS, Morris JM, Ford JB, Henderson-Smart DJ. Hypertensive disorders in pregnancy: a population-based study. The Medical Journal of Australia 2005;182:332e5. Roberts JM, Cooper DW. Pathogenesis and genetics of pre-eclampsia. Lancet 2001; 357:53e6. Roberts JM, Pearson G, Cutler J, Lindheimer M. NHLBI Working Group on research on hypertension during pregnancy. Summary of the NHLBI Working Group on research on hypertension during pregnancy. Hypertension 2003;41:437e45. Roberts JM, Redman CW. Pre-eclampsia: more than pregnancy-induced hypertension. Lancet 1993;341:1447e51. Robinson M, Mattes E, Oddy WH, de Klerk NH, Li J, McLean NJ, et al. Hypertensive diseases of pregnancy and the development of behavioral problems in childhood and adolescence: the Western Australian Pregnancy Cohort Study. The Journal of Pediatrics 2009;154:218e24. Rothrauff TC, Cooney TM, An JS. Remembered parenting styles and adjustment in middle and late adulthood. Journal of Gerontology B Psychological Science and Social Sciences 2009;64:137e46. Royston P. Constructing time-specific reference ranges. Statistics in Medicine 1991; 10:675e90. Schobel HP, Grassi G. Hypertensive disorders of pregnancy: a dysregulation of the sympathetic nervous system? Journal of Hypertension 1998;16:569e70. Sibai B, Dekker G, Kupferminc M. Pre-eclampsia. Lancet 2005;365:785e99.

Silva LM, Coolman M, Steegers EA, Jaddoe VW, Moll HA, Hofman A, et al. Low socioeconomic status is a risk factor for preeclampsia: the Generation R Study. Journal of Hypertension 2008;26:1200e8. Skjaerven R, Vatten LJ, Wilcox AJ, Ronning T, Irgens LM, Lie RT. Recurrence of pre-eclampsia across generations: exploring fetal and maternal genetic components in a population based cohort. BMJ (Clinical Research Ed.) 2005; 331:877. St Clair D, Xu M, Wang P, Yu Y, Fang Y, Zhang F, et al. Rates of adult schizophrenia following prenatal exposure to the Chinese famine of 1959e1961. JAMA: The Journal of the American Medical Association 2005;294:557e62. Thompson C, Syddall H, Rodin I, Osmond C, Barker DJ. Birth weight and the risk of depressive disorder in late life. The British Journal of Psychiatry: The Journal of Mental Science 2001;179:450e5. Tuovinen S, Raikkonen K, Kajantie E, Pesonen AK, Heinonen K, Osmond C, et al. Depressive symptoms in adulthood and intrauterine exposure to preeclampsia: the Helsinki Birth Cohort Study. BJOG: An International Journal of Obstetrics and Gynaecology 2010;117:1236e42. Van den Bergh BR, Van Calster B, Smits T, Huffel SV, Lagae L. Antenatal maternal anxiety is related to HPA-axis dysregulation and self-reported depressive symptoms in adolescence: a prospective study on the fetal origins of depressed mood. Neuropsychopharmacology 2008;33:536e45. Wahlbeck K, Forsen T, Osmond C, Barker DJ, Eriksson JG. Association of schizophrenia with low maternal body mass index, small size at birth, and thinness during childhood. Archives of General Psychiatry 2001;58:48e52. Watson JB, Mednick SA, Huttunen M, Wang X. Prenatal teratogens and the development of adult mental illness. Development and Psychopathology 1999;11: 457e66. Zhang J, Zeisler J, Hatch MC, Berkowitz G. Epidemiology of pregnancy-induced hypertension. Epidemiologic Reviews 1997;19:218e32.