Maternal Bipolar Disorder Increased Low Birthweight and Preterm Births: A Nationwide Population-based Study

Research report

Maternal bipolar disorder increased low birthweight and preterm births:

A nationwide population-based study

Hsin-Chien Lee

, Herng-Ching Lin

⁎

Taipei Medical University-Shuang Ho Hospital, Department of Psychiatry & Sleep Center, Taipei, Taiwan b

Taipei Medical University, School of Medicine, Department of Psychiatry, Taipei, Taiwan c

School of Health Care Administration, Taipei Medical University, 250 Wu-Hsing St., Taipei 110, Taiwan

a r t i c l e i n f o

a b s t r a c t

Article history: Received 7 January 2009

Received in revised form 21 May 2009 Accepted 21 May 2009

Available online 6 June 2009

Objective:To investigate pregnancy outcomes, including low birthweight, preterm births, and small-for-gestational-age (SGA) among women with bipolar disorder, schizophrenia compared with women with no history of mental illness using nationwide population-based data. Methods: This study linked the Taiwan National Health Insurance Research Dataset with the national birth certificate registry. A total of 528,398 singleton births between 2001 and 2003 were included; 337 were diagnosed with bipolar disorder. Multivariate logistic regression analyses were carried out to examine the relationship between maternal bipolar disorder, schizophrenia and the odds of low birthweight, preterm births, and SGA, after adjusting for characteristics of infant, mother and father.

Results: It shows that pregnant women with bipolar disorder were more likely to have LBW infants (9.8% vs. 5.7%), preterm births (14.2% vs. 6.9%) and SGA (22.3% vs. 15.7%) than pregnant women with no history of mental illness. The adjusted odds of low birthweight for women with bipolar disorder was 1.66 times (95% CI, 1.16–2.38) that of women with no history of mental illness. In terms of preterm births and SGA, the adjusted odds ratios were 2.08 (95% CI, 1.53– 2.83) and 1.47 (95% CI, 1.14–1.91) respectively, for women with bipolar disorder, compared to their counterparts with no history of mental illness.

Conclusions: We conclude that women with bipolar disorder had increased risk of low birthweight, preterm births, and SGA than women without a history of mental illness. More active monitoring and early intervention to counter potential adverse pregnancy outcomes for pregnant women with bipolar disorder should be initiated.

© 2009 Elsevier B.V. All rights reserved. Keywords:

Bipolar disorder Pregnancy outcome Low birthweight

1. Introduction

Mothers suffering from severe mental disorders are more likely to have adverse pregnancy outcomes. For example, Howard et al. reported that women with a history of psychotic disorders had a higher proportion of stillbirths and neonatal deaths compared with other women (Howard et al., 2004). Previous studies also found that schizophrenia women have increased risk of preterm delivery, stillbirth and low birth-weight (LBW) (Nilsson et al., 2002; Rifkin et al., 1994;

Bennedsen et al., 1999; Sacker et al., 1996). Besides schizo-phrenia,MacCabe et al. (2007)observed that mothers with affective disorders had elevated risk for preterm birth and small or growth-retarded babies. Webb et al. likewise reported higher risk of fatal birth defects associated with maternal affective disorder (Webb et al., 2007). However, few studies to date have focused specifically on women with bipolar disorder.

Jablensky et al. (2005)have examined the characteristics of infants born to women with schizophrenia, bipolar disorder, or major depression in Western Australia. They found an increased risk of obstetric complications among women with schizophrenia and women with bipolar disorder. However, no increased risk of LBW and preterm birth was

⁎ Corresponding author. Tel.: +886 2 2736 1661x3613; fax: +886 2 2378 9788.

E-mail address:[email protected](H.-C. Lin).

0165-0327/$– see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.jad.2009.05.019

Contents lists available atScienceDirect

Journal of Affective Disorders

observed among the women with bipolar disorder. It is still not clear whether the increased risk of adverse pregnancy outcomes associated with other severe mental disorders is also characteristic of bipolar disorder.

Therefore, this study aims to investigate pregnancy out-comes among women with bipolar disorder compared with women with no history of mental illness using nationwide population-based data. Understanding the association between maternal bipolar disorder and the risk of adverse pregnancy outcomes might help researchers understand possible genetic or environmental mechanisms linking maternal mental illness and pregnancy outcomes, as well as generate opportunities for policy makers and clinicians to provide optimal prenatal care. 2. Methods

2.1. Database

Two databases were used in this study. Thefirst database was the National Health Insurance Research Dataset (NHIRD), published by the National Health Research Institute in Taiwan. The NHIRD includes inpatient and ambulatory care claims under Taiwan National Health Insurance (NHI) with over 21 million enrollees, representing around 96% of the island's population.

The second database used in this study is the birth certificate registry published by the Ministry of the Interior in Taiwan. The data on birth certificates includes birthdates of infants and their parents, gestational week at birth, infant birthweight, gender, parity, place of birth, parental educational level, and maternal marital status. With assistance from the Bureau of the NHI in Taiwan, mother's and infant's unique personal identification numbers provided links between the NHIRD and birth certificate data. Overall, 98.8% of fathers and 100.0% of mothers were identified on the birth certificate registry. All personal identifiers were encrypted by the Bureau of NHI before release to the researchers. Confidentiality assurances were addressed by abiding the data regulations of the Bureau of NHI.

2.2. Study sample

The sample for this research initially comprised all women with singleton births in Taiwan between January 1, 2001, and December 31, 2003 (n = 593,205). We excluded women who had received treatment for any type of mental disorder (except schizophrenia and bipolar disorder: ICD-9-CM codes 295.XX, 296.0X, 296.1X, 296.4X, 296.5X, 296.6X, 296.7X, 296.80 or 296.89) withinfive years prior to the index delivery (n = 165). Similarly, we excluded women whose husbands or partners received treatment for any type of mental disorder in the previousfive years (n=1093) since prior studies have documented a significant association between paternal mental disorder and pregnancy outcome. Finally, we selected only thefirst delivery if a woman had more than one delivery during the study period. Ultimately, 528,891 women fulfilled our criteria and were included in our study.

2.3. Variables of interest

The outcome variables were preterm birth, small-for-gestational-age (SGA) and LBW. According the World Health Organization, the standard cut-off point for LBW in infants is

2500 g (b2500 g, ≥2500 g), and preterm birth was defined as birth occurring at a gestational ageb37 weeks (World Health Organization, 1992). SGA is defined as birth weight below the tenth percentile for gestational age (Hsieh et al., 1991).

The independent variable of interest was whether or not a woman had bipolar disorder as identified from the catastrophic illness card issued by the Bureau of National Health Insurance. Women with schizophrenia were also selected as a comparison group. In Taiwan, once their diagnoses have been verified, patients with severe mental disorders (ICD-9-CM codes 290 and 293 through 297) may be issued a catastrophic illness card in order to reduce theirfinancial burden. Since copayments for psychiatric care are waived for catastrophic illness cardholders, out of self interest, the majority of patients with serious mental disorders are likely to have applied for the card and to be recorded in the registry. Applications for catastrophic illness cards must be signed by a board-certified psychiatrist after diagnosis is veri_{fied through a series of visits.}

We also adjusted for possible factors associated with preterm birth, SGA and LBW in the regression models. These include characteristics of the infant (gender and parity), mother (age, the highest educational level, marital status and whether gestation was complicated by hypertension or diabetes mellitus), father (age and the highest educational level) and family monthly income. Maternal ages were classified as b20, 20–34 and N34. Parity was grouped into the following categories: 1, 2, ≥3. Maternal and paternal education levels were categorized into four levels: elementary school or lower, junior high school, senior high school, college or above. Family monthly income was categorized according to four groups:bNT$15,000, NT15,000– NT30,000, NT30,001–NT50,000, ≥NT50,001.

2.4. Statistical analysis

The SAS statistical package (SAS System for Windows, Version 8.2) was used to conduct the analyses in this study. Pearsonχ2

tests were used to explore differences between women with bipolar disorder and women with no history of mental illness, in terms of characteristics of infant, mother and father. Multivariate logistic regression analyses were carried out to examine the relationships between maternal bipolar disorder and the odds of LBW, preterm birth and SGA, after adjusting for the characteristics of infant, mother and father. The relationships between maternal schizophrenia and these adverse pregnancy outcomes were also examined for comparison. Although maternal and paternal education levels are typically highly correlated with each other, we found their collinearity (correlation coefficient=0.367) was tolerable and therefore both were kept in the regression model. In addition, we adjusted for age differences between parents in the regression model. A two-sided p-value ofb0.05 was considered statistically significant for this study. 3. Results

Of the total sample of 528,891 pregnant women, 337 (0.064%) were diagnosed with bipolar disorder.Table 1shows the details of the distribution of the characteristics of infants, fathers and parturients, separated into those with bipolar disorder, those with schizophrenia and those with no history of mental illness. It shows that pregnant women with bipolar disorder were more

likely to have LBW infants (9.8% vs. 5.7%), preterm births (14.2% vs. 6.9%) and SGA (22.3% vs.15.7%) than pregnant women with no history of mental illness. Pearson_χ2_{also shows that there were}

significant differences betweenwomenwith bipolar disorder and women with no history of mental illness in terms of maternal age (p = 0.049), highest maternal educational level (p_b0.001), marital status (pb0.001), gestational hypertension (p=0.016), family monthly income (p=0.018) and highest paternal educa-tional level (pb0.001) (not shown onTable 1).

Table 2describes the distribution and crude odds ratios of LBW, preterm birth and SGA of the maternal bipolar disorder and non-mental disorder groups. The regression analysis shows that women with bipolar disorder were more likely to have LBW infants, preterm births and SGA babies than women with no history of mental illness.

Details of the adjusted odds ratios for the risk of LBW, preterm birth and SGA by maternal bipolar disorder are provided inTable 3. As the table shows, following adjustment

Table 1

Comparisons of women with bipolar disorder and women with no history of mental illness in relation to sociodemographic characteristics and gestational comorbid medical disorders in Taiwan, 2001–2003 (n=528,891).

Variable Women with bipolar disorder Women with schizophrenia Women with no history of mental illness

Total no. % Total no. % Total no. %

Infant characteristics Birthweight (g) b2500 33 9.8 49 9.9 30,055 5.7 ≥2500 304 90.2 444 90.1 498,006 94.3 Gender Male 180 53.4 275 55.8 276,041 52.3 Female 157 46.6 218 44.2 252,020 47.7 Parity 1 160 47.5 253 51.3 277,765 52.6 2 126 37.4 151 30.6 175,897 33.3 3 or more 51 15.1 89 18.1 74,399 14.1

Gestational age (week)

b37 48 14.2 49 9.9 36,245 6.9

≥37 289 85.8 444 96.1 491,816 93.1

Small for gestational age

Yes 75 22.3 115 23.3 82,699 15.7 No 262 77.7 378 76.7 445,362 84.3 Maternal characteristics Age b20 11 3.3 9 1.8 19,893 3.8 20–34 282 83.7 400 81.1 456,654 86.5 N34 44 13.1 84 17.0 51,514 9.8 Education level

Elementary school or lower 7 2.1 18 3.7 10,284 2.0

Junior high school 89 26.4 152 30.8 83,669 15.8

Senior high school 207 61.4 291 59.0 358,640 67.9

College or above 34 10.1 32 6.5 75,468 14.3 Marital status Married 316 93.8 452 91.7 512,776 97.1 Other 21 6.2 41 8.3 15,285 2.9 Gestational diabetes Yes 0 0 3 0.6 940 0.2 No 337 100 490 99.4 527,121 99.8 Gestational hypertension Yes 5 1.5 1 0.2 2,800 0.5 No 332 98.5 492 99.8 525,261 99.5

Family monthly income

bNT$15,000 107 31.8 129 26.2 163,219 30.9 NT$15,000–30,000 87 25.8 173 35.1 112,603 21.3 NT$30,001–50,000 100 29.7 157 31.9 152,268 28.8 NNT$50,000 43 12.8 34 6.9 99,971 18.9 Paternal characteristics Age b30 124 36.8 125 25.4 204,382 38.7 30–34 122 36.2 170 34.5 196,646 37.2 N34 91 27.0 198 40.2 127,033 24.1 Education level

Elementary school or lower 8 2.4 37 7.5 8130 1.5

Junior high school 88 26.1 161 32.7 96,681 18.3

Senior high school 203 60.2 267 54.2 326,915 61.9

for the infant's gender, parity, maternal age, highest paternal and maternal educational level (separately), mothers' marital status and family monthly income, the odds of LBW for women who had bipolar disorder was 1.66 times (95% CI = 1.16_{–2.38) that of women without bipolar disorder. In} terms of preterm births and SGA, the adjusted odds ratios were 2.08 (95% CI, 1.53–2.83) and 1.47 (95% CI, 1.14–1.91) respectively, for women with bipolar disorder compared to their counterparts with no history of mental illness.

In the comparison group of women with schizophrenia, the odds ratios were only significant for LBW and SGA, but not for preterm births.

4. Discussion

This nationwide population-based study found that women with bipolar disorder had increased risk of adverse pregnancy outcomes: the odds of LBW, preterm birth and SGA for these women were 1.66, 2.08 and 1.47 times, respectively, greater than for women with no history of mental illness, after maternal, paternal and infant characteristics were taken into account. Ourfindings accord with previous studies on women with schizophrenia that indicated a higher risk of premature birth or small or growth-retarded babies (Howard et al., 2004; Nilsson et al., 2002; Rifkin et al., 1994; Bennedsen et al., 1999; Sacker et al., 1996).

A study byMacCabe et al. (2007)previously reported that mothers with affective disorder had elevated risk of giving birth to preterm (OR = 2.67) and LBW babies (OR = 2.22) using the Swedish Medical Birth Register between 1983 and 1997. Despite limiting our study sample specifically to bipolar disorder, our findings remained consistent with theirs; only the risk magnitude of the OR differed. Conversely, a similar study by Jablensky et al. (2005) reported no differences in LBW and preterm births comparing women with bipolar disorder and a comparison group of women who gave birth in Western Australia from 1980 to 1992. They even found that women who were not in the psychiatric case register had higher preterm birth rates than women with bipolar disorder (7.6% vs. 6.2%), although this relationship did not reach a significant level. The differentfindings from these two studies may be difficult to explain due to different study populations, regions, and eras. However, since all explanations for LBW among mothers with schizophrenia offered by Jablensky et al. could also apply to mothers with bipolar disorder, our results may be closer to the facts. As they suggested, more women with severe mental disorders are now capable of living in the community and having children, thus the growing number of mothers with bipolar disorder may drive up the possibility of having LBW infants, accelerating the trend to a visible level in our updated study.

Mechanisms underlying the increased risk of adverse pregnancy outcomes for women with bipolar disorder remain unclear. However, a recent literature review by Walker and colleagues concluded that a number of susceptibility genes are shared between schizophrenia and bipolar disorder (Walker et al., 2002; Laursen et al., 2007). Growing evidence of an etiologic overlap between schizophrenia and bipolar disorder suggest that the possible underlying mechanisms for adverse pregnancy outcomes may be the same for both disorders. One study byBennedsen (1998)summarized possible risk factors for

T able 2 Distributions of LB W , pre term birth and SG A among w omen with bipolar disor der , schizophrenia, and no histo ry of mental illness, 20 0 1– 20 03 (n = 528,89 1). Variable Lo w birthwe ight Pre term birth Small for gesta tional ag e W omen with Y es, N o OR, 95% CI Y e s N o OR, 95% CI Y e s N o OR, 95% CI n (r o w %) n (r o w %) n (r o w %) n (r o w %) n (r o w %) n (r o w %) B ip o la r d is o rd er 3 3 (9.8) 304 (90.4) 1.80 ⁎⁎ (1 .26 – 2.58) 48 (1 4.2) 289 (85.8) 2.26 ⁎⁎⁎ (1 .66 – 3.06) 75 (22.3) 262 (77 .7) 1.54 ⁎⁎⁎ (1 .1 9– 2.0 0) Schizophre nia 49 (9.9) 4 4 4 (90. 1) 1.83 ⁎⁎⁎ (1 .36 – 2.46) 49 (9.9) 4 4 4 (90. 1) 1.50 ⁎ (1 .1 2– 2.0 1) 11 5 (23.3) 3 7 8 (7 6.7) 1.64 ⁎⁎⁎ (1 .3 3– 2.02) N o histo ry of mental illness 30,055 (5.7) 498,0 06 (94.3) 1. 0 0 36,2 45 (6.9) 49 1,81 6 (93. 1) 1. 0 0 82,699 (1 5.7) 4 45,362 (84.3) 1. 0 0 Note : ⁎ indicat es p b 0.05; ⁎⁎ indicat es p b 0.0 1; ⁎⁎⁎ indicates p b 0.0 0 1. Odds ra tios (OR) ar e unadjuste d.

low birth weight and preterm birth among schizophrenia women. She has categorized potential risk factors as follows: smoking, substance abuse, including alcohol, cannabinoids and other illicit drugs, caffeine consumption, socioeconomic factors, parity and maternal age, nutritional factors, maternal physical illness and antenatal care. The higher risks of LBW and SGA babies among schizophrenic women observed in this study may lend support to a common underlying mechanism.

In addition,Jablensky et al. (2005)have proposed that the increased risk of adverse pregnancy outcomes among women with affective disorder can be explained by the clustering of adverse maternal characteristics. Women with affective dis-orders are more likely to have unhealthy lifestyles, including poor diet, a lack of exercise and obesity, largely attributable to

their socioeconomic disadvantages and to lack of adequate social support. These reproductive hazards among women with bipolar disorder could increase their risk of adverse pregnancy outcomes.

However, in the current study, even after adjusting for socioeconomic factors, parity, maternal age, and maternal physical illness, the relationship between increased risk of adverse pregnancy outcomes and bipolar disorder remained. It seems that these factors may not be the major determinants of adverse pregnancy outcomes among women with bipolar disorder.

Smoking is likely to be one explanation for the relationship. One study by Wilens et al. found that bipolar disorder was associated with a significant age-adjusted risk for cigarette

Table 3

Adjusted Odds ratios of LBW, preterm birth and SGA for women with bipolar disorder or schizophrenia compared to women with no history of mental illness in Taiwan (n = 528,891).

Variable Low birthweight Preterm birth Small for gestational age

Adjusted OR, 95% CI Adjusted OR, 95% CI Adjusted OR, 95% CI Women with

Bipolar disorder 1.66⁎⁎ (1.16–2.38) 2.08⁎⁎⁎ (1.53–2.83) 1.47⁎⁎ (1.14–1.91)

Schizophrenia 1.62⁎⁎ (1.20–2.18) 1.33 (0.99–1.79) 1.52⁎⁎⁎ (1.23–1.87)

No history of mental illness 1.00 1.00 1.00

Maternal characteristics Age (years) b20 1.00 1.00 1.00 20–34 0.75⁎⁎⁎ (0.72–0.80) 0.81⁎⁎⁎ (0.77–0.85) 0.73⁎⁎⁎ (0.71–0.76) N34 0.92⁎ (0.86–0.99) 1.01⁎⁎ (1.03–1.17) 0.74⁎⁎⁎ (0.71–0.79) Education level

Elementary school or lower 1.37⁎⁎⁎ (1.28–1.48) 1.25⁎⁎⁎ (1.1701.34) 1.23⁎⁎⁎ (1.17–1.29)

Junior high school 1.22⁎⁎⁎ (1.18–1.26) 1.18⁎⁎⁎ (1.14–1.21) 1.15⁎⁎⁎ (1.12–1.17)

Senior high school 1.00 1.00 1.00

College or above 0.92⁎⁎⁎ (0.88–0.96) 0.95⁎⁎ (0.91–0.98) 0.90⁎⁎⁎ (0.87–0.92) Marital status Married 0.61⁎⁎⁎ (0.58–0.64) 0.66⁎⁎⁎ (0.63–0.70) 0.66⁎⁎⁎ (0.63–0.68) Others 1.00 1.00 1.00 Gestational hypertension Yes 3.33⁎⁎⁎ (3.12–3.80) 2.81⁎⁎⁎ (2.55–3.10) 2.15⁎⁎⁎ (1.98–2.34) No 1.00 1.00 1.00 Infant characteristics Gender Male 0.78⁎⁎⁎ (0.77–0.80) 1.21⁎⁎⁎ (1.19–1.24) 0.71⁎⁎⁎ (0.70–0.72) Female 1.00 1.00 1.00 Parity 1 1.00 1.00 1.00 2 0.79⁎⁎⁎ (0.77–0.81) 1.05⁎⁎⁎ (1.03–1.08) 0.82⁎⁎⁎ (0.80–0.83) 3 or more 0.80⁎⁎⁎ (0.77–0.83) 1.19⁎⁎⁎ (1.15–1.23) 0.78⁎⁎⁎ (0.77–0.80)

Family monthly income

bNT$15,000 1.00 1.00 1.00

NT$15,000–30,000 0.97 (0.94–1.01) 0.99 (0.96–1.02) 0.94⁎⁎⁎(0.92–0.96)

NT$30,001–50,000 0.91⁎⁎⁎ (0.88–0.94) 0.96⁎ (0.94–0.99) 0.89⁎⁎⁎ (0.88–0.91)

NNT$50,000 0.88⁎⁎⁎ (0.84–0.91) 0.94⁎⁎⁎ (0.91–0.97) 0.85⁎⁎⁎ (0.83–0.87)

Age difference (paternal age–maternal age) 1.00 (0.99–1.01) 1.00 (0.99–1.01) 1.00 (0.99–1.01) Paternal characteristics

Education level

Elementary school or lower 1.31⁎⁎⁎ (1.21–1.43) 1.19⁎⁎⁎ (1.10–1.28) 1.28⁎⁎⁎ (1.21–1.35)

Junior high school 1.17⁎⁎⁎ (1.13–1.20) 1.10⁎⁎⁎ (1.07–1.13) 1.14⁎⁎⁎ (1.11–1.16)

Senior high school 1.00 1.00 1.00

College or above 0.88⁎⁎⁎ (0.85–0.92) 0.93⁎⁎⁎ (0.90–0.97) 0.90⁎⁎⁎ (0.88–0.93)

Note:⁎ indicates pb0.05; ⁎⁎ indicates pb0.01; ⁎⁎⁎ indicates pb0.001. Note:⁎ indicates pb0.05; ⁎⁎ indicates pb0.01; ⁎⁎⁎ indicates pb0.001.

In the fully adjusted model, maternal age, education level, marital status, and gestational hypertension, and infant's gender and parity, plus family monthly income, parental age difference, and paternal education level were included.

smoking (hazard ratio=12.3) (Wilens et al., 2008). Another study by Gonzalez-Pinto et al. even indicated that bipolar disorder (in both genders) was significantly associated (OR=4.4) with heavy smoking (more than 1 pack per day) (Gonzalez-Pinto et al.,1998). However, data on cigarette smoking was not available in our dataset. But the study byMacCabe et al. (2007)did show that the relationship between mothers with affective disorder and elevated risk of adverse pregnancy outcomes persists even after adjusting for smoking. Therefore, smoking may not be the sole cause of this relationship.

The treatment for bipolar disorder itself, whether persis-tent or discontinued, during pregnancy may be another potential factor contributing to this relationship. At present, it is still difficult to estimate the risks and benefits of taking such medication during pregnancy, due to inadequate data (Wisner et al., 2000).

The strengths of this study include its use of nationwide population-based datasets linking the NHIRD with birth certificates and the adjusting for important characteristics of mother, father and infant. However, thefindings of this study should be interpreted within the context of three limitations. First, we identified mothers diagnosed with bipolar disorder from the registry of catastrophic illness released by the Bureau of the NHI, potential sampling bias existed, such as afinancial incentive for bipolar patients with low socioeconomic status to apply for the catastrophic illness card. Secondly, with respect to validity and reliability, there was no standardized diagnostic algorithm of bipolar disorder in service claims data. In addition, this dataset did not allow us to account for differences in the severity of bipolar disorder among patients. Thirdly, informa-tion on mothers' smoking history, substance abuse, alcohol consumption, nutrition and body mass index are not available through our datasets.

Despite these limitations, this study found that women with bipolar disorder had increased risk of LBW and preterm births than women without bipolar disorder, after adjusting for potential confounders. Clinicians should be aware of the increased risk of adverse pregnancy outcomes among these women. More active monitoring and early intervention to counter potential LBW and preterm births should be initiated for women with bipolar disorder. In addition, further studies are needed to elucidate the underlying mechanisms linking bipolar disorder among mothers and fetal development.

Role of funding source Nothing declared. Conflict of interest

No conflict declared.

Acknowledgement

This study was supported by grant 98TMU-SHH-05 from the Taipei Medical University-Shuang Ho Hospital.

References

Bennedsen, B.E., 1998. Adverse pregnancy outcome in schizophrenic women: occurrence and risk factors. Schizophr. Res. 33, 1–26.

Bennedsen, B.E., Mortensen, P.B., Olesen, A.V., Henriksen, T.B., 1999. Preterm birth and intra-uterine growth retardation among children of women with schizophrenia. Br. J. Psychiatry 175, 239–245.

Gonzalez-Pinto, A., Gutierrez, M., Ezcurra, J., Aizpuru, F., Mosquera, F., Lopez, P., de Leon, J., 1998. Tobacco smoking and bipolar disorder. J. Clin. Psychiatry 59, 225–228.

Howard, L.M., Goss, C., Leese, M., Appleby, L., Thornicroft, G., 2004. The psychosocial outcome of pregnancy in women with psychotic disorders. Schizophr. Res. 71, 49–60.

Hsieh, T.T., Hsu, J.J., Chen, C.C., Chiu, T.H., Liou, J.D., Hsieh, C.C.,1991. An analysis of birth weight by gestational age in Taiwan. J. Formosan Med. Assoc. 90, 382–387.

Jablensky, A.V., Morgan, V., Zubrick, S.R., Bower, C., Yellachich, L.A., 2005. Pregnancy, delivery, and neonatal complications in a population cohort of women with schizophrenia and major affective disorders. Am. J. Psychiatry 162, 79–91.

Laursen, T.M., Munk-Olsen, T., Nordentoft, M., Bo Mortensen, P., 2007. A comparison of selected risk factors for unipolar depressive disorder, bipolar affective disorder, schizoaffective disorder, and schizophrenia from a Danish population-based cohort. J. Clin. Psychiatry 68, 1673–1681. MacCabe, J.H., Martinsson, L., Lichtenstein, P., Nilsson, E., Cnattingius, S.,

Murray, R.M., Hultman, C.M., 2007. Adverse pregnancy outcomes in mothers with affective psychosis. Bipolar Disord. 9, 305–309. Nilsson, E., Lichtenstein, P., Cnattingius, S., Murray, R.M., Hultman, C.M., 2002.

Women with schizophrenia: pregnancy outcome and infant death among their offspring. Schizophr. Res. 58, 221–229.

Rifkin, L., Lewis, S., Jones, P., Toone, B., Murray, R., 1994. Low birth weight and schizophrenia. Br. J. Psychiatry 165, 357–362.

Sacker, A., Done, D.J., Crow, T.J., 1996. Obstetric complications in children born to parents with schizophrenia: a meta-analysis of case–control studies. Psychol. Med. 26, 279–287.

Walker, J., Curtis, V., Murray, R.M., 2002. Schizophrenia and bipolar disorder: similarities in pathogenic mechanisms but differences in neurodevelop-ment. Int. Clin. Psychopharmacol. 3, S11–S19 Suppl.

Webb, R.T., Pickles, A.R., King-Hele, S.A., Appleby, L., Mortensen, P.B., Abel, K.M., 2007. Parental mental illness and fatal birth defects in a national birth cohort. Psychol. Med. 38, 1495–1503.

Wilens, T.E., Biederman, J., Adamson, J.J., Henin, A., Sgambati, S., Gignac, M., Sawtelle, R., Santry, A., Monuteaux, M.C., 2008. Further evidence of an association between adolescent bipolar disorder with smoking and substance use disorders: a controlled study. Drug Alcohol Depend. 95, 188–198. Wisner, K.L., Zarin, D.A., Holmboe, E.S., Appelbaum, P.S., Gelenberg, A.J.,

Leonard, H.L., Frank, E., 2000. Risk-benefit decision making for treatment of depression during pregnancy. Am. J. Psychiatry 157, 1933–1940. World Health Organization, 1992. International Statistical Classification of

Diseases and Related Health Problems, 10th Revision. World Health Organization, Geneva.