Prenatal exposure to the 1944–45 Dutch ‘hunger winter’ and addiction later in life

Prenatal exposure to the 1944–45 Dutch ‘hunger winter’ and addiction later in life Ernst J. Franzek1 , Niels Sprangers1 , A. Cecile J. W. Janssens2 , Cornelia M. Van Duijn3 & Ben J. M. Van De Wetering1 Bouman Mental Health Care Rotterdam, the Netherlands,1 Center for Medical Decision Sciences, Depar tment of Public Health, Erasmus Medical Center Rotterdam, the Netherlands2 and Genetic Epidemiology Unit, Depar tment of Epidemiology and Biostatistics, Erasmus Medical Center Rotterdam, the Netherlands3 ABSTRACT Aims Prenatal exposure to severe famine has been associated with an increased risk of schizophrenia and affective disorders. We studied the relationship between prenatal exposure to famine during the Dutch hunger winter of 1944–45 and addiction later in life. Design A case–control study. Setting The Rotterdam city area during the Dutch hunger winter lasting from mid-October 1944 to mid-May 1945. From February 1945 to mid-May 1945 the hunger winter was characterized by a famine peak. Participants Patients are native Dutch addicted patients from the Rotterdam Addiction Treatment Program and controls are native Dutch inhabitants of Rotterdam, born between 1944 and 1947. Measurement Exposure to the whole hunger winter (< 1400 kcal/day) and the peak of the hunger winter (< 1000 kcal/day) was determined for each trimester of gestation. For each trimester the exposed/unexposed ratios were compared between patients and controls and quantified as odds ratios (OR). Findings The odds of first-trimester gestational exposure to famine during the total hunger winter was significantly higher among patients receiving treatment for an addictive disorder [OR = 1.34, 95% confidence interval (CI) 1.10–1.64]. Stratification by sex shows that the odds of exposure during the first trimester was significantly higher only among men (OR = 1.34, 95% CI 1.05–1.72), but not among women (OR = 1.26, 95% CI 0.88–1.81). The odds of exposure to the peak of the hunger winter during the first trimester of gestation were also significantly higher among addiction treatment patients (OR = 1.61, 95% CI 1.22–2.12). We did not find any significant differences for the second and third trimesters of gestation. Conclusion First-trimester prenatal exposure to famine appears to be associated with addiction later in life. The study confirms the adverse influence of severe malnutrition on brain development and maturation, confirms the influence of perinatal insults on mental health in later life and gives rise to great concern about the possible future consequences for the hunger regions in our world. Keywords Addiction, malnutrition, neurodevelopment, pregnancy, prenatal exposure to famine, schizophrenia. Correspondence to: Ernst J. Franzek, Bouman Mental Health Care, PO Box 8549, 3009 AM Rotterdam, the Netherlands. E-mail: [email protected] Submitted 5 May 2007; initial review completed 8 August 2007; final version accepted 25 October 2007 INTRODUCTION In September 1944 the final and most dramatic episode of World War II began to hit the Netherlands. Hoping to capture strategic bridges over the Rhine to open a pathway for a rapid invasion into Germany, the Allied forces launched a parachute attack behind the Nazi forces near the city of Arnhem. However, the operation failed with major losses. Immediately afterwards the German authority imposed a total embargo on occupied Netherlands in retaliation for the Dutch support to the Allied forces. At that time the food situation in the western part of the Netherlands was already bad but subsequently worsened dramatically. Although the embargo was lifted from transports over water in early November 1944 the famine intensified progressively because the winter was unusually early and severe. All the waterways were frozen and barges could not sail out. Food rations declined to extremely low levels between February and May 1945, resulting in an individual average daily official ration below 1000 calories. During the weeks before liberation in early May 1945 the famine reached its peak. The normal average daily food intake is 2300 kcal for women and 2900 kcal for men with an active life-style. RESEARCH REPORT doi:10.1111/j.1360-0443.2007.02084.x © 2008 The Authors. Journal compilation © 2008 Society for the Study of Addiction Addiction, 103, 433–438 At least 22 000 people of the 3.5 million inhabitants of the western part of the Netherlands died because of the famine [1]. Initially, pregnant and lactating women were entitled to supplementary rations; however, at the peak of the famine these extra rations could no longer be provided. After the liberation the food situation improved very rapidly and by June 1945 the rations had already risen to over 2000 calories per person per day [2]. Despite the disastrous famine, babies were still conceived and born. Thus the circumscribed place and period of the Dutch famine created a natural ‘experiment’ in which the individuals being conceived or having been in various states of gestation during this period can be studied with respect to the effects of maternal malnutrition on health in adult life. A number of chronic somatic diseases in the offspring in later life, such as glucose intolerance [3], high blood pressure [4], obesity [5], heightened atherogenic lipid profiles [6], higher risk of coronary disease [7], increased prevalence of obstructive airways disease [8], reduced plasma fibrinogen and factor VII concentrations [9], microalbuminuria and higher adult mortality [10,11], have been found to be associated with maternal famine exposure during gestation. The same has been reported for major psychiatric disorders such as schizophrenia [12–14], schizophrenia spectrum disorders [15], major affective disorders [16,17] and antisocial personality disorder [18]. Thus, prenatal famine exposure may be a major contributing factor for the development of a variety of somatic, neurodevelopmental and neuropsychiatric disorders. Modern brain research indicates that susceptibility to addiction can involve neurodevelopmental abormalities [19,20]. We studied the association between prenatal exposure to the Dutch hunger winter of 1944–45 in the Rotterdam area and addiction in later life. We compared the odds of exposure/non-exposure during each trimester of gestation between addicted patients and inhabitants of the Rotterdam area, born between 1944 and 1947. METHODS Study population Patients included in this study are native Dutch patients born in the period 1944–47 registered with an addiction problem (alcohol and drug abuse) in the database of the Dutch Mental Health Care Organization, Bouman GGZ, based in the Rotterdam area. Controls are native Dutch inhabitants born in the period of 1944–1947 living in Rotterdam on 1 January 2004. Table 1 gives an overview of the number of patients and controls included in the study. Definition of exposure The Dutch hunger winter lasted from mid-October 1944 until 12 May 1945 [2]. During this period the average daily food intake was below 1400 kcal. We defined the period from 15 October 1944 to 12 May 1945 as the ‘whole hunger winter’. The Dutch hunger winter was characterized by a famine peak in the months from February 1945 until the end of the hunger winter. During these months the average daily food intake was below 1000 kcal and the effects were most drastic at the end of the hunger winter, because of the continuing starvation process. We defined the period from 1 February 1945 to 12 May 1945 as the ‘peak of the hunger winter’. In research studies on the effects of famine on gestation it is common to divide gestation into three trimesters. A pregnancy lasts approximately 280 days; each trimester lasts about 94 days. We assumed that all pregnancies lasted a full 282 days. Based on this assumption, we obtained the dates of conception. Figure 1 shows for each trimester of gestation the periods that define exposure and non-exposure to the whole hunger winter and the peak of the hunger winter: individuals who were conceived within those periods were considered as exposed for the trimester. For each individual we determined the exposure status for the first, second and third trimesters of gestation. Individuals are considered as exposed during a specific trimester of gestation if they were exposed to the effects of the hunger winter for a minimal period of 2 weeks during that trimester. Individuals are considered unexposed when they were conceived during the equivalent period in the subsequent year. We narrowed the definition of the controls explicitly to the equivalent date of birth period in the subsequent year because of possible confounding factors such as a possible relationship between season of birth and addiction disorders, and environmental and social effects such as the Dutch heroin epidemic of the early 1970s [21–26]. Figure 1 shows that there is an overlap between the periods that define exposure/non-exposure during first-, Table 1 Number of addicted patients registered in the database of Bouman GGZ and number of living native inhabitants of Rotterdam (controls) born between 1944 and 1947. Patients Controls Men Women Men Women 1944 168 101 1987 2138 1945 174 67 1654 1643 1946 266 138 3197 3341 1947 270 104 2852 2877 Total 878 410 9690 9999 434 Ernst J. Franzek et al. © 2008 The Authors. Journal compilation © 2008 Society for the Study of Addiction Addiction, 103, 433–438 second- and third-trimester exposure, i.e. some of the subjects are exposed during two trimesters. For exposure to the peak of the hunger winter, we therefore stratified the population additionally into five groups: exposure to the first, second or third trimesters, to the first and second trimesters and to the second and third trimesters (see Fig. 2). Data analysis For each trimester we calculated odds ratios (OR) and 95% confidence intervals (CI) as the exposed/unexposed ratio in patients divided by the exposed/unexposed ratio in controls. Using Pearson’s c2 test we tested whether the exposed/unexposed ratios were different between patients and controls (P < 0.05, two-tailed). RESULTS Table 2 shows the number of exposed and unexposed patients and controls for each trimester of gestation. The odds of first-trimester gestational exposure to the famine effects of the total hunger winter (< 1400 kcal/day) among patients was significantly higher than the odds of exposure among controls (OR = 1.34, 95% CI 1.10– 1.64; P < 0.01). Stratification by sex shows that the odds of exposure during the first trimester was significantly higher only among men (OR = 1.34, 95% CI 1.05–1.72; P < 0.05), but not in women (OR = 1.26, 95% CI 0.88– 1.81). There were no significant differences between the odds of exposure to second- and third-trimester gestational famine (<1400 kcal/day) of patients and controls. Table 3 shows the results of gestational exposure to the peak of the hunger winter with an average food intake of less than 1000 kcal/day. The odds of exposure during the first trimester of gestation was significantly higher among patients (OR = 1.61, 95% CI 1.22–2.12; P < 0.001). Stratification by sex again shows that the odds of exposure differ significantly only in men: the OR in men was 1.63 (95% CI 1.16–2.27; P < 0.01) compared to an OR of 1.40 (95% CI 0.85–2.33; P > 0.05) in women. There were no significant differences between the odds of exposure in patients and controls during the second and third trimesters of gestation. Finally, Table 4 presents the ORs for the stratification during five periods of exposure. The table shows that the odds of exposure to famine during the first trimester remains significantly higher only among male patients than among male controls (OR = 1.82, CI 1.18–2.80; P < 0.01). None of the other comparisons yielded a statistically significant difference. =Exposed during whole hunger winter Exposed Exposed Exposed Exposed Exposed Exposed UnExposed UnExposed UnExposed UnExposed UnExposed UnExposed Month of conception First Trimester Second Trimester Third Trimester Jan-44 Feb-44 Mar-44 Apr-44 May-44 Jun-44 Jul-44 Aug-44 Sep-44 Oct-44 Nov-44 Dec-44 Jan-45 Feb-45 Mar-45 Apr-45 May-45 Jun-45 Jul-45 Aug-45 Sep-45 Oct-45 Nov-45 Dec-45 Jan-46 Feb-46 Mar-46 Apr-46 May-46 Jun-46 =Exposed during peak hunger winter = Total hunger winter (< 1400 kcal/day) and peak of the hunger winter (1000 < kcal/day) Figure 1 Definition of exposure and non-exposure to the whole hunger winter and peak of the hunger winter for the first, second and third trimesters of gestation Exposed 3rd only 1st only 2nd & 3rd 1st & 2nd 2nd 3rd only 1st only 2nd & 3rd 1st & 2nd 2nd Exposed Exposed UnExposed UnExposed UnExposed Month of conception First trimester Second trimester Minimal overlap Third trimester May-44 Jun-44 Jul-44 Aug-44 Sep-44 Oct-44 Nov-44 Dec-44 Jan-45 Feb-45 Mar-45 Apr-45 May-45 Jun-45 Jul-45 Aug-45 Sep-45 Oct-45 Nov-45 Dec-45 Jan-46 Feb-46 Mar-46 Apr-46 May-46 Jun-46 =Peak of the hunger winter (< 1000 kcal/day) Figure 2 Definition of exposure and non-exposure to the peak of the hunger winter for the first, second and third trimesters of gestation and for exposure during two trimesters (first and second, or second and third) Exposure to hunger winter and addiction 435 © 2008 The Authors. Journal compilation © 2008 Society for the Study of Addiction Addiction, 103, 433–438 DISCUSSION Our study is the first to show an association between prenatal exposure to famine and addiction in later life. Before discussing the implications of our study, some methodological issues need to be addressed. First, we cannot exclude potential misclassification of addiction. Control data of the inhabitants of the Rotterdam city area were Table 2 Exposure to the whole hunger winter (< 1400 kcal/day) in addicted patients and controls. Controls Patients OR (95% CI) First trimester All Exposed 2048 154 1.34 (1.10–1.64)* Unexposed 5152 289 Men Exposed 1038 109 1.34 (1.05–1.72)** Unexposed 2493 195 Women Exposed 1010 45 1.26 (0.88–1.81) Unexposed 2659 94 Second trimester All Exposed 2679 184 1.14 (0.94–1.37) Unexposed 5319 321 Men Exposed 1330 133 1.21 (0.97–1.52) Unexposed 2599 214 Women Exposed 1349 51 0.96 (0.68–1.35) Unexposed 2720 107 Third trimester All Exposed 3148 224 1.06 (0.89–1.28) Unexposed 4123 276 Men Exposed 1572 158 1.09 (0.87–1.36) Unexposed 2048 189 Women Exposed 1576 66 1.00 (0.72–1.38) Unexposed 2075 87 CI = confidence interval; OR = odds ratio. *P < 0.01, **P < 0.05. Table 3 Exposure to the peak of the hunger winter (<1000 kcal/day) in addicted patients and controls. Controls Patients OR (95% CI) First trimester All Exposed 917 81 1.61 (1.22–2.12)*** Unexposed 2879 158 Men Exposed 484 59 1.63 (1.16–2.27)** Unexposed 1387 104 Women Exposed 433 22 1.40 (0.85–2.33) Unexposed 1492 54 Second trimester All Exposed 1343 87 1.16 (0.89–1.51) Unexposed 3229 180 Men Exposed 663 61 1.15 (0.83–1.58) Unexposed 1545 124 Women Exposed 680 26 1.15 (0.72–1.85) Unexposed 1684 56 Third trimester All Exposed 1915 134 1.15 (0.92–1.43) Unexposed 3453 211 Men Exposed 954 98 1.24 (0.95–1.63) Unexposed 1692 140 Women Exposed 961 36 0.93 (0.62–1.40) Unexposed 1761 71 CI = confidence interval; OR = odds ratio. **P < 0.01, ***P < 0.001. 436 Ernst J. Franzek et al. © 2008 The Authors. Journal compilation © 2008 Society for the Study of Addiction Addiction, 103, 433–438 obtained from the municipal statistics department and these data do not contain information about the health status of the inhabitants. It is therefore possible that the control population also included inhabitants who are addicted and perhaps even inhabitants who were registered in our database. If, however, misclassification of diagnosis had affected the results, it would probably lead to an underestimation of the effect. Secondly, we were unable to ascertain if all patients and controls were born in the Rotterdam area, and hence were exposed to the hunger winter. Because this problem holds for both patients and controls, we do not expect that the results were biased by this. Thirdly, although our study included 1288 patients and 19 689 controls, this sample may not have been large enough to demonstrate subgroup effects. The effect of malnutrition is probably underestimated, because patients were already in their 60s and there may have been excess mortality among this group during the preceding 20–30 years. Finally, because we had no data on comorbidity, we were unable to assess whether famine during gestation led to addiction, or whether this was an indirect effect explained by other psychiatric disorders such as schizophrenia [12–15]. One could argue that the effect found in this study is confounded by the presence of comorbidity. However, the prevalence of these disorders in the population of Bouman GGZ is low and it is very unlikely that this can explain the reported associations. Although we found a statistically significant association only with exposure during the first trimester of gestation, we also cannot rule out an effect of famine during all trimesters. All ORs in men and most ORs in women were higher than 1. Moreover, among men the OR for exposure during the third trimester was 1.39 and borderline statistically significant (Table 4). This could mean that exposure to prenatal food rationing leads to a higher risk of addiction in all trimesters, and that the risk is strongest for exposure to famine during the first trimester. A related issue is the presence of a dose–effect relationship: is greater exposure to famine during gestation associated with a higher risk of addiction later in life? On one hand, we found that the association was stronger for first-trimester exposure during the peak of the hunger winter compared to exposure to the total hunger winter. On the other hand, the effect of gestational exposure to the first trimester only was larger than exposure to the first and second trimesters. Thus, longer exposure to famine did not result in a higher risk of addiction. This supports the view that the first trimester is crucial in the Table 4 Prenatal exposure to the peak of the hunger winter (< 1000 kcal/day) in addicted patients and controls, with minimal overlap between birth cohorts Controls Patients OR (95% CI) Only 1st trimester Men Exposed 273 38 1.82 (1.18–2.80)** Unexposed 772 59 Women Exposed 245 13 1.32 (0.69–2.55) Unexposed 823 33 1st and 2nd trimesters Men Exposed 211 21 1.36 (0.80–2.34) Unexposed 615 45 Women Exposed 188 9 1.53 (0.69–3.40) Unexposed 669 21 Only 2nd trimester Men Exposed 69 7 1.14 (0.46–2.89) Unexposed 203 18 Women Exposed 101 4 0.77 (0.24–2.49) Unexposed 215 11 2nd and 3rd trimesters Men Exposed 383 33 1.03 (0.66–1.60) Unexposed 727 61 Women Exposed 391 13 1.11 (0.56–2.20) Unexposed 800 24 Only 3rd trimester Men Exposed 571 65 1.39 (0.99–1.96) Unexposed 965 79 Women Exposed 570 23 0.83 (0.50–1.37) Unexposed 961 47 CI = confidence interval; OR = odds ratio. **P < 0.01. Exposure to hunger winter and addiction 437 © 2008 The Authors. Journal compilation © 2008 Society for the Study of Addiction Addiction, 103, 433–438 development of the reward system in the human brain that is involved mainly in addictive behaviour [20]. In conclusion, we have found for the first time an association between famine during gestation and participating in treatment for addiction in adulthood. Further studies are needed to confirm these findings and to investigate other possible relationships between addiction and prenatal development, such as the influence of an unbalanced diet, influenza and other stressors. Acknowledgements We would like to express our gratitude to the ‘Centrum voor Onderzoek en Statistiek van de gemeente Rotterdam’ (Centre of Research and Statistics of the Municipality of Rotterdam) for their excellent service and for providing the birth dates of the native Dutch inhabitants. References 1. de Jong L. Nederland Tijdens de Tweede Wereldoorlog [The Netherlands During the Second World War]. Den Haag: Sdu Publishers; 1981. 2. Burger C. G. E., Sandstead H. R., Drummond J. C. Malnutrition and Starvation inWestern Netherlands, September 1944 to July 1945. Part I and II. The Hague: General State Printing Office; 1948. 3. Ravelli A. C., van der Meulen J. H., Michels R. P., Osmond C., Barker D. J., Hales C. N. et al. Glucose tolerance in adults after prenatal exposure to famine. Lancet 1998; 351: 173–7. 4. Roseboom T. J., van der Meulen J. H., Ravelli A. C., van Montfrans G. A., Osmond C., Barker D. J. et al. Blood pressure in adults after prenatal exposure to famine. J Hypertens 1999; 17: 325–30. 5. Ravelli A. C., van der Meulen J. H., Osmond C., Barker D. J., Bleker O. P. Obesity at the age of 50 years in men and women exposed to famine prenatally. Am J Clin Nutr 1999; 70: 811–16. 6. Roseboom T. J., van der Meulen J. H., Osmond C., Barker D. J., Ravelli A. C., Bleker O. P. Plasma lipid profiles in adults after prenatal exposure to the Dutch famine. Am J Clin Nutr 2000; 72: 1101–6. 7. Roseboom T. J., van der Meulen J. H., Osmond C., Barker D. J., Ravelli A. C., Schroeder-Tanka J. M. et al. Coronary heart disease after prenatal exposure to the Dutch famine, 1944–45. Heart 2000; 84: 595–8. 8. Lopuhaa C. E., Roseboom T. J., Osmond C., Barker D. J., Ravelli A. C., Bleker O. P. et al. Atopy, lung function, and obstructive airways disease after prenatal exposure to famine. Thorax 2000; 55: 555–61. 9. Roseboom T. J., van der Meulen J. H., Ravelli A. C., Osmond C., Barker D. J., Bleker O. P. et al. Plasma fibrinogen and factor VII concentrations in adults after prenatal exposure to famine. Br J Haematol 2000; 111: 112–17. 10. Painter R. C., Roseboom T. J., Bleker O. P. Prenatal exposure to the Dutch famine and disease in later life: an overview. Reprod Toxicol 2005; 20: 345–52. 11. Painter R. C., Roseboom T. J., van Montfrans G. A., Bossuyt P. M., Krediet R. T., Osmond C. et al. Microalbuminuria in adults after prenatal exposure to the Dutch famine. J Am Soc Nephrol 2005; 16: 189–94. 12. Susser E. S., Lin S. P. Schizophrenia after prenatal exposure to the Dutch Hunger Winter of 1944–45. Arch Gen Psychiatry 1992; 49: 983–8. 13. Susser E., Neugebauer R., Hoek H. W., Brown A. S., Lin S., Labovitz D. et al. Schizophrenia after prenatal famine. Further evidence. Arch Gen Psychiatry 1996; 53: 25–31. 14. 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 1959–1961. JAMA 2005; 294: 557– 62. 15. Hoek H. W., Brown A. S., Susser E. The Dutch famine and schizophrenia spectrum disorders. Soc Psychiatry Psychiatr Epidemiol 1998; 33: 373–9. 16. Brown A. S., Susser E. S., Lin S. P., Neugebauer R., Gorman J. M. Increased risk of affective disorders in males after second trimester prenatal exposure to the Dutch hunger winter of 1944–1945. Br J Psychiatry 1945; 166: 601–6. 17. Brown A. S., van Os J., Driessens C., Hoek H. W., Susser E. S. Further evidence of relation between prenatal famine and major affective disorder. Am J Psychiatry 2000; 157: 190–5. 18. Neugebauer R., Hoek H. W., Susser E. Prenatal exposure to wartime famine and development of antisocial personality disorder in early adulthood. JAMA 1999; 282: 455–62. 19. Leshner A. I. Addiction is a brain disease, and it matters. Science 1997; 278: 45–7. 20. Volkow N. D., Fowler J. S., Wang G. J. The addicted human brain: insights from imaging studies. J Clin Invest 2003; 111: 1444–51. 21. Goldberg A. E., Newlin D. B. Season of birth and substance abuse: findings from a large national sample. Alcohol Clin Exp Res 2000; 24: 774–80. 22. Kunugi H., Nanko S., Watanabe H., Sekiba K., Kazamatsuri H. Season of birth of chronic alcoholics. J Psychiatr Res 1998; 32: 321–3. 23. Levine M. E., Wojcik B. E. Alcoholic typology and season of birth. J Addict Dis 1999; 18: 41–52. 24. London W. P. Alcoholism: theoretical consideration of season of birth and geographical latitude. Alcohol 1987; 4: 127–9. 25. Modestin J., Ammann R., Wurmle O. Season of birth: comparison of patients with schizophrenia, affective disorders and alcoholism. Acta Psychiatr Scand 1995; 91: 140–3. 26. Watson C. G., Tilleskjor C., Kucala T., Jacobs L. The birth seasonality effects in nonschizophrenic psychiatric patients. J Clin Psychol 1984; 40: 884–8. 438 Ernst J. Franzek et al. © 2008 The Authors. Journal compilation © 2008 Society for the Study of Addiction Addiction, 103, 433–438

DMU Timestamp: March 29, 2019 18:11