From the *Department of Epidemiology, Mailman School of Public Health, Columbia University, New York, New York; †New York State Psychiatric Institute, New York, New York; and ‡College of Physicians and Surgeons, Columbia University, New York, New York.
Correspondence: Michaeline Bresnahan, Department of Epidemiology, Mailman School of Public Health, 722 W. 168th Street, NY, NY 10032. E-mail: firstname.lastname@example.org.
In this issue of Epidemiology, Stalberg and colleagues report the lack of an association between prenatal ultrasound and risk of schizophrenia in adulthood. These findings contribute to the search for unintended effects of an intervention designed to improve prenatal care. Although no adverse effect of ultrasound was reported, other gestational exposures have been associated with increased risk of schizophrenia. By extending the causal time frame to include long-term latent effects we are confronted with a potential paradox: interventions beneficial in early life may have undetected adverse consequences in adulthood.
Since the alarm was first sounded for the teratogenic consequences of diethylstilbestrol (DES),1 the potential for latent effects of prenatal exposure on adult health has been well recognized.2 DES became the archetypal cautionary tale. The adverse effects of a drug taken by an expectant mother might not be manifest in offspring during infancy or even childhood, but could appear in adulthood. Later, investigators became concerned that affected offspring might go on to have children with related health problems even though these children were never exposed to the drug.3,4 DES underscored the need for a revised time frame for “do no harm.” The article in this issue examining the association of ultrasound exposure in gestation to risk of schizophrenia in adulthood reflects these concerns.5
The study of prenatal ultrasound and schizophrenia also falls squarely within the approach of lifecourse epidemiology.6 The importance of early life exposures and the lengthened causal time frame are fundamental to lifecourse epidemiology, and have been an important focus in schizophrenia research. In our own work, prenatal nutritional,7 infectious,8 and chemical9 exposures have been tied to schizophrenia in adulthood. In other work, a variety of prenatal and obstetric complications have been associated with later risk of schizophrenia (reviewed by Cannon et al).10,11 In several areas the findings are strong but not definitive; among the best established is nutritional deficiency. Maternal starvation during pregnancy was first identified as posing increased risk of schizophrenia in the context of the Dutch Hunger Winter, a famine induced by a Nazi blockade during WWII. Offspring conceived at the height of the famine were found to be at a 2-fold increased risk of schizophrenia in adulthood. Recent research on the Chinese famine in 1959–1960 replicated this finding in a vastly different context.12 The Chinese famine was brought about during the Great Leap Forward, an attempt at rapid industrialization in China. As in the Dutch example, there was a 2-fold excess risk of schizophrenia among individuals conceived during the height of the famine in Annhui Province. A third study from another region in China also provides concordant results.13
The notion of latent effects is not intuitive. The insight that comes with the lengthened causal time frame extends our notion of “harm” to include the wellbeing of the infant, child and adult. By extending our consideration of time we are confronted with the paradox that interventions known to be beneficial for infants and children may have undetected adverse consequences in adulthood and vice versa,14 as well as with the complexities that accompany consideration of more than one disease outcome.
The outcomes we should be investigating are not always obvious across large spans of time, or easily ascertained. The article in this issue by Stalberg and colleagues illustrates a valuable strategy to establish hypothesized exposure-latent outcome associations. Linking birth and psychiatric registries to hospital ultrasound records, the authors were able to explore the association between prenatal ultrasound exposures and risk of schizophrenia decades later. In this instance, the strategy was applicable because the exposure and outcome were available in treatment records and registries. This study also illustrates the power that national registries in Scandinavia and Israel have brought to the study of schizophrenia.15,16
Investigating latent effects of gestational exposures can also begin in older cohorts, such as the National Collaborative Perinatal Study and the Child Health and Development Study.17,18 These pregnancy/birth cohorts were established in the mid-twentieth century. The initial intent of these projects included assessing the impact of infection, medications, cigarettes, alcohol, and other exposures during gestation on obstetric, infant and childhood outcomes. Some of the participants in this research have been followed into middle age for a variety of outcomes, notably schizophrenia.19,20 The utility of these cohorts, however, is limited to examining those exposures in existence at the time the cohorts were assembled, and identifiable through information previously collected or detectable in stored biologic samples.
The scope of relevant early life experience is increasingly appreciated, and this research challenge is being taken up by investigators in many areas, including psychiatry.21 Thinking broadly, in modern societies millions of children are prenatally exposed to antiretroviral cocktails, maternal antidepressants, ultrasound, and in vitro fertilization procedures, and virtually all children are prenatally exposed to toxins (potential teratogens) in our environment.22 Increased understanding of the health impact of these intentional and unintentional exposures during gestation and childhood is a major motivation for establishing a large pregnancy/birth cohort of approximately 100,000 in the United States (The National Children's Study),23 following the precedents of Norway (Norwegian Mother and Child Cohort)24 and Denmark (Danish National Birth Cohort).25 Although these longitudinal studies represent a significant step in tracking the effects of environment on early growth and development, their contributions to our understanding of latent effects expressed in mid- and late-life are a distant return on this investment. In the meantime substantial smaller cohorts that are more advanced in age will be contributing to our understanding of adult outcomes.26
No single design will circumvent the need for foresight, ingenuity of design, passage of time, and good luck required to capture latent effects. The lifecourse framework, however, provides a broad context and appropriate mindset for designing effective strategies to examine these questions. Having accepted the notion that early events—such as prenatal exposures—can have significant latent effects, we must also move beyond the exposure-outcome association to articulate mechanisms. Understanding these will increase our capacity to project possible effects at different points in the life course and begin to realistically appreciate benefits and costs of a given exposure or intervention. Acting from the lifecourse perspective, we should aim to exceed the mandate of “do no harm,” and promote lifetime health.
ABOUT THE AUTHORS
MICHAELINE BRESNAHAN is Assistant Professor in the Department of Epidemiology at Mailman School of Public Health, Columbia University. Dr. Bresnahan's work has focused on neurodevelopmental schizophrenia and autism, and pregnancy/birth cohorts research. EZRA SUSSER is the Anna Cheskis Gelman and Murray Charles Gelman Professor and Chair of the Department of Epidemiology at Mailman School of Public Health, and Professor of Psychiatry at Columbia University. Dr. Susser's primary research has been on the epidemiology of mental disorders, and on the role of early life experience in health and disease throughout the lifecourse. He and colleagues recently completed a textbook Psychiatric Epidemiology, Searching for the Causes of Mental Disorders.
1. Herbst AL, Ulfelder H, Poskanzer DC. Adenocarcinoma of the vagina. Association of maternal stilboestrol therapy with tumor appearance in young women. N Engl J Med. 1971;284:878–81.
2. Giusti RM, Iwamoto K, Hatch EE. Diethylstilbestrol revisited: a review of the long-term health effects. Ann Intern Med. 1995;122:778–788.
3. Klip H, Verloop J, van Gool JD, et al, for the OMEGA Project Group. Hypospadias in sons of women exposed to diethylstilbestrol in utero: a cohort study. Lancet. 2002;359:1102–7.
4. Titus-Ernstoff L, Troisi R, Hatch EE, et al. Menstrual and reproductive characteristics of women whose mothers were exposed in utero to diethylstilbestrol (DES). Int J Epidemiol. 2006;35:862–68.
5. Stalberg K, Haglund B, Axelsson O, et al. Prenatal ultrasound scanning and the risk of schizophrenia and other psychoses. Epidemiology. 2007;18:577–582.
6. Kuh D, Ben-Shlomo Y. A Life Course Approach to Chronic Disease Epidemiology. 2nd ed. Oxford: Oxford University Press; 2004.
7. Susser E, Neugebauer R, Hoek HW, et al. Schizophrenia after prenatal famine. Further evidence. Arch Gen Psychiatry. 1996;53:25–31.
8. Brown AS, Begg MD, Gravenstein S, et al. Serologic evidence for prenatal influenza in the etiology of schizophrenia. Arch Gen Psychiatry. 2004;61:774–780.
9. Opler MG, Brown AS, Graziano J, et al. Prenatal lead exposure, delta-aminolevulini acid, and schizophrenia. Environ Health Perspect. 2004;112:548–552.
10. Cannon M, Jones PB, Murray RM. Obstetric complications and schizophrenia: historical and meta-analytic review. Am J Psychiatry. 2002;159:1080–1092.
11. Clarke MC, Harley M, Cannon M. The role of obstetric events in schizophrenia. Shizophr Bull. 2006;32:3–8.
12. St Clair D, Xu M, Wang P, et al. Rates of adult schizophrenia following prenatal exposure to the Chinese famine of 1959–1961. JAMA. 2005;294:557–562.
13. Xu Mingxing, Susser ES, St Clair D (manuscript). Prenatal malnutrition and adult schizophrenia: Further evidence from 1959–61 Chinese famine.
14. Lucock M, Yates Z. Folic acid—vitamin and panacea or genetic time bomb? Nat Rev Genet. 2005;6:235–240.
15. Mortensen PB, Pedersen C, Westergaard T, et al. Family history and place and season of birth on the risk of schizophrenia. N Engl J Med. 1999;340:603–608.
16. Malaspina D, Harlap S, Fenning S, et al. Advancing paternal age and the risk of schizophrenia. Arch Gen Psychiatry. 2001;58:361–367.
17. Broman S. The Collaborative Perinatal Project: an Overview. In: Mednick SA, Harway M, and Finello KM, eds. Handbook of Longitudinal Research Vol I. New York: Praeger, 1984:185–227.
18. van den Berg BJ, Christianson RE, Oechsli FW. The California Child Health and Development Studies of the School of Public Health, University of California at Berkeley. Paediatr Perinat Epidemiol. 1988;2:265–282.
19. Susser E, Schaefer C, Brown A, Begg M, Wyatt RJ. The design of the prenatal determinants of schizophrenia (PDS). Schizophr Bull 26. 2000;2:257–273.
20. Buka SL, Tsuang MT, Torrey EF, Klebanoff MA, Bernstein D, Yolken RH. Maternal infections and subsequent psychosis among offspring. Arch Gen Psychiatry. 2001;58:1032–1037.
21. Susser E, Schwartz S, Morabia A, et al. Psychiatric Epidemiology: Searching for the Causes of Mental Disorders. New York: Oxford University Press; 2006.
22. Landrigan PJ, Trasande L, Thorpe L, et al. The National Children's Study: a 21-year prospective study of 100,000 American children. Pediatrics. 2006;118:2173–2186.
24. Magnus P, Inrgens LM, Haug K, et al, and the MoBa Study Group. Cohort profile: the Norwegian Mother and Child Cohort Study (MoBa). Int J Epidemiol. 2006;35:1146–1150.
25. Olsen J, Melbye M, Olsen SF, et al. The Danish National Birth Cohort–its background, structure and aim. Scand J Public Health. 2001;29:300–307.
26. Golding J. Children of the Nineties: A Longitudinal Study of Pregnancy and Childhood based on the population of Avon (ALSPAC). West Engl Med J. 1990;105:80–82.
© 2007 Lippincott Williams & Wilkins, Inc.