Adult Schizophrenia Following Prenatal Exposure to the Chinese Famine of 1959-1961

Abstract & Commentary

By Sarah L. Berga, MD, James Robert McCord Professor and Chair, Department of Gynecology and Obstetrics, Emory University School of Medicine, Atlanta, GA, is Associate Editor for OB/GYN Clinical Alert

Dr. Berga is a consultant for Pfizer, Organon, and Women First, Inc., and is involved in research for Berlex and Health Decisions, Inc.

Synopsis: Prenatal exposure to famine doubles the risk of schizophrenia in later life.

Source: St Clair D, et al. Rates of adult schizophrenia following prenatal exposure to the Chinese famine of 1959-1961. JAMA. 2005;294:557-562.

Schizophrenia is a common form of severe mental illness characterized by disordered thoughts processes, hallucinations, delusions, and social withdrawal. The cause has been attributed to many factors, including obstetrical complications, season of birth, maternal stress, prenatal viral exposure, and poor maternal nutrition. However, schizophrenia is increasingly viewed as a neurodevelopmental disorder, with environmental influences during early brain development modifying the risk. The lifetime risk is roughly 1%. It is universally distributed and rates are similar in men and women. In the 1990s, 2 studies found that prenatal exposure to famine during the Dutch Hunger Winter of 1944-1945 roughly doubled the risk of schizophrenia.1,2 The aim of the present study was to see if this association held in another population exposed to famine.

The current study was done in China. The famine occurred from 1959 to 1961 and was due to immense social and economic upheaval coupled with bad weather. The study was made possible by referral patterns in the region. The Fourth People’s Hospital in the only psychiatric hospital in the region where the famine was greatest and records are complete from 1971 to the present. The risk of schizophrenia for each year of birth was measured by the cumulative incidence of outpatient consultation and inpatient admission. More than 97% of the cases with schizophrenia from 1955 to 1965 were born in the region. During the famine, birth rates decreased by 80% and cumulative mortality rates for children during the famine years roughly doubled to 40%. The odds ratio for schizophrenia was 0.89 in 1959, 2.30 in 1960 (P < 0.001), 1.93 in 1961 (P < 0.001), and 0.95 in 1962.


This study may not seem immediately relevant to our typical population of pregnant women. However, its lessons are worth considering. It has long been assumed that undernutrition in pregnancy carries few fetal consequences, as the fetus is spared at maternal expense. Emerging evidence on several fronts appears to directly challenge this assumption. Most of us have heard of the Barker hypothesis, which posits that maternal undernutrition sufficient to produce infants that are small for gestational age predisposes those same offspring to excess and premature cardiovascular disease in later life. Hansen et al have shown that significant maternal grief in early gestation markedly increases the risk of congenital malformations.3 Haddow et al have shown that even subclinical hypothyroidism lowers overall intelligence and increases the risk of poor neurodevelopment.4 Folate deficiency leads to neural tube defects, particularly in predisposed populations. Further, there is a link between severe undernutrition and psychopathology, including schizophrenia, antisocial personality, and schizoid personality disorder. Additionally, the culture conditions for embryos during assisted reproductive techniques may increase the risk for certain congenital disorders attributable to altered fetal genomic imprinting, ie, some aspect of the embryo’s culture conditions (nutrition) changes the way that the maternal and paternal alleles are methylated. Methylation patterns, in turn, alter the timing and dosage of gene expression, thus influencing embryo development.

In an accompanying editorial, Neugebauer points out that the most pressing question from a public health and interventionalist perspective is whether the relevant restriction of interest constitutes a global nutritional deficiency or a specific micronutrient deficiency.5 Our own research on stress-induced anovulation suggests that combined metabolic and psychosocial stress is far more potent in eliciting maternal biochemical perturbations than either nutritional or psychosocial stress alone. Certainly famine constitutes more than nutritional deprivation and thus it was not surprising to read that birth rates during the famine fell by 80%. However, the pregnancies that resulted also were certainly subjected to both extreme undernutrition and severe psychosocial stress, so it is almost surprising to find that the risk of schizophrenia and other neurodevelopmental disorders was not greater. Indeed, the results suggest a degree of adaptation and resilience that one might not have expected given the apparently high fetal susceptibility to subtle perturbations noted above.

Resilience in the face of adversity notwithstanding, we need to know if there are ways to ameliorate the fetal consequences of nutritional deprivation. At least one of the positive answers does not entail achieving an equitable distribution of the world’s financial and nutritional resources and ought to be feasible and relatively inexpensive. That answer is to provide our familiar micronutrient and friend of rapidly replicating cells, folate. Neugebauer points out that diet-triggered alterations in DNA methylation patterns may be buffered by having sufficient folate in the microenvironment of the replicating cells. Indeed, the genetic polymorphism methyltetrahydrofolate reductase, C677T, a key enzyme regulating intracellular folate availability, increases the risk of schizophrenia independent of undernutrition. Thus, famine seems to impact the DNA in a manner similar to that associated with embryo culture. Of course, other mechanisms also may be at play in famine and assisted reproduction as well.

The take-home message here is three-fold. First, think folate always, but especially when diet appears to be inadequate. Second, circumstances that combine both undernutrition and psychosocial stress, such as famine, typically result in anovulation and infertility rather than an at-risk pregnancy. Third, the chance that environmental or lifestyle insults will alter fetal imprinting or DNA methylation patterns, augment later cardiovascular risk, or subtly impact neurodevelopment is probably greater when conditions are not so severe as to cause anovulation. In the absence of medical intervention, stress-induced anovulation prevents the worse case scenario.


  1. Susser ES, et al. Schizophrenia after prenatal exposure to the Dutch Hunger Winter of 1944-1945. Arch Gen Psychiatry. 1992;49:983-988.
  2. Jones P. Schizophrenia after prenatal exposure to the Dutch Hunger Winter of 1944-1945. Arch Gen Psychiatry. 1994;51:333-334.
  3. Hansen D, et al. Serious life events and congenital malformations: a national study with complete follow-up. Lancet. 2000;356:875-880. Erratum in: Lancet. 2001;30:357:2142.
  4. Haddow J, et al. Maternal thyroid deficiency during pregnancy and subsequent neuropsychological development of the child. N Engl J Med. 1999;341:549-555.
  5. Neugebauer R. Accumulating evidence for prenatal nutritional origins of mental disorders. JAMA. 2005;294:621-623.