Psychological distress during early gestation and

0 downloads 0 Views 64KB Size Report
Sep 25, 2007 - Carsten Obel1,6, Tine Brink Henriksen2, Niels Jørgen Secher3, Brenda ... larger datasets (Mills et al., 2003; Obel et al., 2003) and seems.

Reprod. Human Reproduction pp.Hum. 1–4, 2007

Advance Access published September 25, 2007


Psychological distress during early gestation and offspring sex ratio Carsten Obel1,6, Tine Brink Henriksen2, Niels Jørgen Secher3, Brenda Eskenazi4 and Morten Hedegaard5 1

Department of General Practice, Institute of Public Health, University of Aarhus, 8000 Aarhus C, Denmark; 2Department of Paediatrics, Aarhus University Hospital, Skejby, 8200 Aarhus N, Denmark; 3Department of Obstetrics and Gynecology, Hvidovre University Hospital, 2650 Hvidovre, Denmark; 4School of Public Health University of California, Berkeley, CA 94720, USA; 5Department of Obstetrics and Gynecology, Rigshospitalet, 2100 Copenhagen, Denmark 6

Correspondence address. Tel: þ45 89426059; Fax: þ45 86124788; E-mail: [email protected]

Keywords: psychological distress; stress; pregnancy; sex ratio; GHQ

Introduction In developed countries 5% more boys than girls are delivered (Biggar et al., 1999). However, several recent studies indicate that the male to female ratio (the sex ratio) is declining (Moller, 1996; van der Pal-de Bruin et al., 1997; Marcus et al., 1998). This decline is of public health interest because sex ratio can be regarded as a sentinel health indicator (Davis et al., 1998). The reason for this development is unknown. Smoking around the time of conception has been proposed as a possible cause (Fukuda et al., 2002), but this has not been confirmed in larger datasets (Mills et al., 2003; Obel et al., 2003) and seems an unlikely explanation since the prevalence of smoking during pregnancy is declining (Wisborg et al., 1998). Ecologic studies have shown that environmental exposures to flooding, earthquake or war were followed by a reduced sex ratio (Lyster, 1974; Fukuda et al., 1998; Graffelman and Hoekstra, 2000). In line with this, recent reports found a tendency to a lower sex ratio in the offspring of women, who were pregnant around the terrorist attack of September 11, 2001 (Catalano et al., 2005a, 2006). A common link between these studies

could be exposure to psychological stress. The strongest support for this hypothesis is a Danish registry-based study, which found fewer boys in the offspring of women exposed to severe stress around or just prior to conception (Hansen et al., 1999). However, only very rare and extreme stress exposures were considered. To explain the secular trend of decreasing male to female ratio, more moderate stress exposure needs to be considered. The effect of such levels of psychological distress has so far only been studied indirectly in an ecological design, e.g. following the use of antidepressants and anxiolytic drugs (Catalano et al., 2005b). In the present report, we examine the association between more moderate and common levels of maternal psychological distress and the sex ratio in the offspring in a follow-up design.

Materials and Methods Between August 1, 1989 and September 30th, 1991, a populationbased cohort of pregnant women was established at the Department

# The Author 2007. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved. For Permissions, please email: [email protected]

Page 1 of 4

Downloaded from by guest on June 7, 2013

BACKGROUND: Exposure to severe stress in early pregnancy is associated with a lower male to female ratio (sex ratio), but whether more moderate levels of psychological discomfort have the same kind of effect is unknown. In a population based follow-up study, we aimed to test whether psychological distress was associated with the sex ratio in the offspring. METHODS: From 1989 to 1992, a cohort of 8719 Danish-speaking pregnant women were followed until delivery. Questionnaires were administered to the women in early pregnancy and 6629 (76%) completed the 30item version of the General Health Questionnaire (GHQ). RESULTS: We found an overall male to female ratio (sex ratio) of 1.03. There was an inverse dose response association (test for trend P < 0.01) between GHQ score and sex ratio. Each 5-point increase in the GHQ score was associated with a decreasing odds of having a boy [Odds ratio (OR) 5 0.93, 95% CI 0.89–0.98]. Mothers scoring in the upper quartile of the GHQ had 47% boys as compared with 52% in the undistressed groups (Risk difference 54.8%, 95% CI 1.9–7.7%) resulting in a significantly lower sex ratio of 0.85 compared with 1.07 (OR 0.82, 95% CI 0.72–0.94). CONCLUSIONS: Our results suggest that not only severe stress, but also more moderate and common levels of psychological distress, may decrease the sex ratio in the offspring. Stress during pregnancy is a likely candidate involved in the decreasing sex ratio observed in many countries.

Obel et al.

of Obstetrics, Aarhus University Hospital, Denmark. Details about the data collection have previously been published (Hedegaard et al., 1993). Briefly, the study was based on information from women who during pregnancy attended the Hospital for routine antenatal care. The women were sent by mail self-administered questionnaires twice during pregnancy, prior to their routine antenatal visits at approximately the 16th and the 30th week of gestation. The present paper is based only on exposure-information related to the first trimester collected at the visit around 16 gestational weeks of pregnancy by two questionnaires. One was a brief questionnaire designed to provide details for the women’s medical record including information on medical history, demographic characteristics and smoking and alcohol habits, and the other was a more extensive research questionnaire including the 30-item version of the General Health Questionnaire (GHQ) (Goldberg, 1972). We used the GHQ (Goldberg, 1972), which was developed as a screening instrument for general populations. The score can be interpreted to indicate the severity of psychological disturbance. The content and construct validity have previously been tested in a Danish population (Bech, 1987), and in the present study, the reliability of the GHQ was found to be high (Crohnbach’s alfa .0.90). The 30 items asked about different aspects of psychological distress experienced within the previous month, i.e. approximately during the third to fourth gestational month. The possible answers were (i) ‘not at all’, (ii) ‘not more than usual’, (iii) ‘somewhat more than usual’ or (iv) ‘much more than usual’. On this basis, the women were asked to indicate the level of distress compared with a personal (internal) reference. Hence, women who were more chronically distressed were unexpected to obtain the highest scores. Each item in the GHQ was dichotomized with zero points for the two lower categories of answers on a given item, and 1 point otherwise, as recommended by Goldberg et al. (1997). The sum across the 30 items was calculated, resulting in a score with a potential value between 0 and 30. The scores of the women in our study ranged from 0 to 29. The median value was 2, the upper quartile was a score above 5 and the upper centile was a score above 9. Because of the skewed distribution, GHQ was analysed as a categorical variable. To investigate a possible dose response association, we categorized the GHQ scores a priori in equal steps of the score (0, 1–5, 6–10, 11–15, 16 –20, 21– 25, 26 –30). Women, who were unable to read Danish were excluded from the study (n ¼ 470). Furthermore, women with twin pregnancies were excluded (n ¼ 123), leaving 8719 women eligible for the study. Of these, 6643 women (76.2%) completed the questionnaire. The sex of the newborn in 14 deliveries was not reported, leaving 6629 cases for the analyses. Women were asked to record the date on which they completed the questionnaires. After delivery, the attending midwife recorded the information about the delivery and the newborn on a structured coding sheet. This information included weight, length, head circumference, APGAR score and sex of the child. A research midwife reviewed the medical records, validated and completed the coding sheet by adding information on the post-partum period.

out to compare the sex ratios according to the level of distress and adjust for potential confounding factors. Potential confounders were selected based on previous reports. Factors changing the estimate by 10% or more were included in the final model (Greenland, 1989). Risk differences (RD) and odds ratios (OR) were calculated for unadjusted analyses and OR’s for adjusted analyses. Both were provided with 95% confidence intervals (CI). Women who reported no distress (GHQ score of zero) were used as the reference group.

Statistical analysis


Data analyses were performed in SPSS , version 11.0. Chi-squared test for independence within contingency tables and Chi-squared test for trend were used for categorical data. Statistical significance was defined as a two-sided P-value of ,0.05. Multivariate logistic regression analyses were carried Page 2 of 4

Among participants, 3364 (50.7%) women delivered a boy, whereas 3265 (49.3%) women delivered a girl resulting in a sex ratio of 1.03. The sex ratio in the entire cohort was 1.05. The sex ratio among non-participants was higher, but not statistically different from that of the participants (OR ¼ 1.07, 95% CI 0.96– 1.18). Table 1 describes the sex ratio of the infants in relation to obstetric, behavioural and demographic factors. Only maternal height was significantly associated with the sex of the newborn. We found a slightly lower sex ratio among the youngest and the oldest mothers and fathers. Women who had received infertility treatment also had a reduced sex ratio. The association between psychological distress and the sex ratio in the offspring is shown in Table 2. We found a decreasing proportion of boys with higher levels of psychological distress. The sex ratio in the reference group was 1.07 compared with a ratio of 0.67 in women who reported the highest level of psychological distress. The test for trend suggested a dose-response relationship (P , 0.001), with each 5-point increase in the GHQ score being associated with a decreasing chance of having a boy (OR ¼ 0.93, 95% CI 0.89 –0.98). Women in the upperquartile of GHQ scores, i.e., those who scored greater than 5, had 47% boys (resulting in a sex ratio of 0.85) and were thus 5% less likely to have a boy (RD ¼ 4.8%, 95% CI: 1.9– 7.7%) as compared with those who reported no symptoms of distress at all. They had significant lower odds of delivering a boy (OR 0.82, 95% CI 0.72 – 0.94). The association between the GHQ score and the sex of the baby could potentially be confounded. In order to investigate this possibility, we conducted a series of multivariate logistic regression analyses with the sex of the baby as the outcome. In Table 2, we present data adjusted for all the covariates in Table 1. However, none of the potential confounders were able to change the crude OR estimates by 10% or more. The unadjusted data (only containing the GHQ score) was therefore regarded as the most valid.

We found fewer boys among the offspring of women who reported moderate to high levels of psychological distress. Among the children of women in the upper quartile of GHQ scores, we found 47% boys compared with 52% in the unstressed women and a significant dose response association.

Downloaded from by guest on June 7, 2013



Maternal stress and offspring sex ratio

Table 1: Offspring sex ratio according to potential confounders (The Aarhus Birth Cohort, n ¼ 6629) Boys, n (%)

Girls, n (%)

P-value (Chi-squared test)

51 (51.5) 513 (49.8) 1355 (48.7) 904 (48.8) 442 (51.2)

0.94 1.01 1.05 1.05 0.95


286 (50.4) 1024 (48.3) 1023 (47.4) 823 (51.5)

0.98 1.07 1.11 0.94


259 (49.5) 645 (45.5) 1027 (50.9) 850 (49.5) 412 (51.7)

1.02 1.20 0.96 1.02 0.93


176 (47.2) 1341 (49.7) 1136 (48.9) 371 (51.0) 161 (47.5)

1.12 1.01 1.05 0.96 1.11


2184 (49.2) 175 (51.3) 300 (46.9) 339 (50.7) 190 (50.7)

1.03 0.95 1.13 0.97 0.97


1709 (50.6) 1421 (48.0) 113 (45.4)

0.98 1.08 1.20


327 (50.0) 1002 (49.2) 1829 (49.2)

1.00 1.03 1.03


1752 (49.8) 1056 (48.6) 366 (49.6) 87 (46.8)

1.01 1.06 1.02 1.14


3058 (49.0) 138 (53.3)

1.04 0.88


Our findings are in accordance with the findings of Hansen et al. (1999), who showed a reduced sex ratio among women who experienced trauma (i.e. death or admission to hospital of an older child due to cancer, or death or admission to hospital of the spouse due to cancer or myocardial infarction) during

Table 2: Offspring sex ratio according to GHQ scores (Adjusted analyses for the potential confounders of Table 1) GHQ score 0 1 –5 6 –10 11– 15 16– 20 21– 25 26– 30




Sex ratio

Unadjusted OR

Adjusted OR

2319 2828 860 379 154 74 15

1201 1468 409 175 72 33 6

52 52 48 46 47 45 40

1.07 1.08 0.91 0.86 0.88 0.80 0.67

1 1.00 (0.90; 1.12) 0.84 (0.72; 0.99) 0.80 (0.64; 0.99) 0.82 (0.59; 1.13) 0.75 (0.47; 1.19) 0.62 (0.22; 1.75)

1 1.00 (0.89; 1.13) 0.89 (0.75; 1.05) 0.82 (0.65; 1.04) 0.87 (0.61; 1.23) 0.56 (0.33; 0.95) 0.57 (0.16; 1.96)

Page 3 of 4

Downloaded from by guest on June 7, 2013

Maternal age (years) ,20 48 (48.5) 20– 24 517 (50.2) 25– 29 1428(51.3) 30– 34 949 (51.2) .34 422 (48.8) Paternal age (years) ,24 281 (49.6) 25– 29 1096 (51.7) 30– 34 1134 (52.6) .34 774 (48.5) Maternal height (cm) ,160 264 (50.5) 160 –164 774 (54.5) 165 –169 991 (49.1) 170 –174 868 (50.5) .174 385 (48.3) Maternal pre-pregnant weight (kg) ,50 197 (52.8) 50– 59 1358 (50.3) 60– 69 1189 (51.1) 70– 79 357 (49.0) .79 178 (52.5) Maternal smoking (cigarettes/day) 0 2254 (50.8) 1 –4 166 (48.7) 5 –9 340 (53.1) 10– 14 329 (49.3) .14 185 (49.3) Marital status Married 1669 (49.4) Cohabiting 1539 (52.0) Living alone 136 (54.6) Maternal education (years) ,10 327 (50.0) 10– 12 1033 (50.8) .12 1890 (50.8) Parity 0 1763 (50.2) 1 1118 (51.4) 2 372 (50.4) .2 99 (53.2) Infertility treatment No 3180 (51.0) Yes 121 (46.7)

Sex ratio, males/ females

the 12 months prior to pregnancy or during the first trimester. They found the lowest sex ratio in the newborns of women exposed to stress during the first trimester. Women, who experienced the death of a spouse or an older child, gave birth to babies with a sex ratio of 0.65. This is comparable to the ratio found among the few women in the present study (1%), who reported very high levels of distress. Our findings also agree with a number of ecological studies that have found a lower sex ratio in populations exposed to severe stress (Lyster, 1974; Fukuda et al., 1998; Graffelman and Hoekstra, 2000; Catalano et al., 2006). The participation rate in the present study was 76%. The sex ratio in newborns of women who declined participation was slightly higher than among the responders. Less than 10% of all women had prenatal chromosome testing of their fetus. Results of these tests were usually unavailable until the questionnaire had been completed. The women were therefore unaware of the sex of the fetus at the time of accepting or declining participation and we found no reason to suspect that selection bias can explain our findings. We tested potential confounding effects from a number of other factors suspected to influence the sex ratio. However, none of these factors affected the association between the GHQ score in early pregnancy and the sex ratio in the offspring. We are unaware of other potential confounders that may have explained our findings. Trivers and Willard’s hypothesis that maternal conditions such as stress could influence the sex distribution of the offspring (Trivers and Willard, 1973) has found support in a number of animal experiments (Pratt and Lisk, 1989; Bacon and McClintock, 1999). They suggest that the effect may be mediated by corticosteroids (Pratt and Lisk, 1990). Psychological distress may activate the maternal hypothalamic – pituitary – adrenal axis and expose the fetus to high levels of cortisol. The GHQ was used as a measure of psychological distress in the present study and the instrument has previously been validated in pregnant women (Sharp, 1988). GHQ is a reliable measure of psychological discomfort and screens for minor psychiatric disorders (Goldberg et al., 1997). However, documentation of the association of GHQ markers with biological measures of human stress such as serum corticosteroid levels is quite limited (Aardal-Eriksson et al., 1999). As our measures of psychological distress were not defined by a specific event, we were unable to define an exact time window of exposure. We collected a measure of psychological distress related to the third to fourth month of pregnancy. This measure may be a weak indicator of peaks of stress during the

Obel et al.

Funding This project was funded by the Danish Medical Research Council.

References Aardal-Eriksson E, Eriksson TE, Holm AC, Lundin T. Salivary cortisol and serum prolactin in relation to stress rating scales in a group of rescue workers. Biol Psychiatry 1999;46:850–855. Bacon SJ, McClintock MK. Sex ratio bias in postpartum-conceived Norway rat litters is produced by embryonic loss in midpregnancy. J Reprod Fertil 1999;117:403–411. Bech P. Quality of life in psychosomatic research. A psychometric model. Psychopathology 1987;20:169– 179. Biggar RJ, Wohlfahrt J, Westergaard T, Melbye M. Sex ratios, family size, and birth order. Am J Epidemiol 1999;150:957–962. Byrne J, Warburton D. Male excess among anatomically normal fetuses in spontaneous abortions. Am J Med Genet 1987;26:605– 611. Catalano R, Bruckner T, Gould J, Eskenazi B, Anderson E. Sex ratios in California following the terrorist attacks of September 11, 2001. Hum Reprod 2005a;20:1221–1227. Catalano R, Bruckner T, Hartig T, Ong M. Population stress and the Swedish sex ratio. Paediatr Perinat Epidemiol 2005b;19:413–420. Catalano R, Bruckner T, Marks AR, Eskenazi B. Exogenous shocks to the human sex ratio: the case of September 11, 2001 in New York City. Hum Reprod 2006.

Page 4 of 4

Davis DL, Gottlieb MB, Stampnitzky JR. Reduced ratio of male to female births in several industrial countries: a sentinel health indicator?. JAMA 1998;279:1018–1023. Fukuda M, Fukuda K, Shimizu T, Andersen CY, Byskov AG. Parental periconceptional smoking and male: female ratio of newborn infants. Lancet 2002;359:1407–1408. Fukuda M, Fukuda K, Shimizu T, Moller H. Decline in sex ratio at birth after Kobe earthquake. Hum Reprod 1998;13:2321–2322. Goldberg DP. The Detection of Psychiatric Illness by Questionnaire. London: Oxford University Press, 1972. Goldberg DP, Gater R, Sartorius N, Ustun TB, Piccinelli M, Gureje O, Rutter C. The validity of two versions of the GHQ in the WHO study of mental illness in general health care. Psychol Med 1997;27:191– 197. Graffelman J, Hoekstra RF. A statistical analysis of the effect of warfare on the human secondary sex ratio. Hum Biol 2000;72:433–445. Greenland S. Modeling and variable selection in epidemiologic analysis. Am J Public Health 1989;79:340–349. Hansen D, Moller H, Olsen J. Severe periconceptional life events and the sex ratio in offspring: follow up study based on five national registers. BMJ 1999;319:548– 549. Hedegaard M, Henriksen TB, Sabroe S, Secher NJ. Psychological distress in pregnancy and preterm delivery (see comments). BMJ 1993;307: 234–239. Henriksen TB, Wilcox AJ, Hedegaard M, Secher NJ. Bias in studies of preterm and postterm delivery due to ultrasound assessment of gestational age. Epidemiology 1995;6:533–537. Jongbloet PH. The ovopathy concept for explaining the secondary sex ratio. Hum Reprod 2004;19:1036– 1037. Lyster WR. Altered sex ratio after the London smog of 1952 and the Brisbane flood of 1965. J Obstet Gynaecol Br Commonw 1974;81: 626–631. Marcus M, Kiely J, Xu F, McGeehin M, Jackson R, Sinks T. Changing sex ratio in the United States, 1969– 1995. Fertil Steril 1998;70:270– 273. Mills JL, England L, Granath F, Cnattingius S. Cigarette smoking and the male-female sex ratio. Fertil Steril 2003;79:1243– 1245. Moller H. Change in male:female ratio among newborn infants in Denmark. Lancet 1996;348:828–829. Obel C, Henriksen TB, Hedegaard M, Bech BH, Wisborg K, Olsen J. Periconceptional smoking and the male to female ratio in the offspring– re-assessment of a recently proposed hypothesis. Int J Epidemiol 2003;32:470–471. Pratt NC, Lisk RD. Effects of social stress during early pregnancy on litter size and sex ratio in the golden hamster (Mesocricetus auratus). J Reprod Fertil 1989;87:763–769. Pratt NC, Lisk RD. Dexamethasone can prevent stress-related litter deficits in the golden hamster. Behav Neural Biol 1990;54:1– 12. Sharp DJ. Validation of the 30-item General Health Questionnaire in early pregnancy. Psychol Med 1988;18:503– 507. Trivers RL, Willard DE. Natural selection of parental ability to vary the sex ratio of offspring. Science 1973;179:90– 92. van der Pal-de Bruin KM, Verloove-Vanhorick SP, Roeleveld N. Change in male:female ratio among newborn babies in Netherlands. Lancet 1997;349:62. Wisborg K, Henriksen TB, Hedegaard M, Secher NJ. Smoking habits among Danish pregnant women from 1989 to 1996 in relation to sociodemografic and lifestyle factors. Acta Obstet Gynecol Scand 1998;77:836–840. Submitted on April 2, 2007; resubmitted on June 11, 2007; accepted on July 17, 2007

Downloaded from by guest on June 7, 2013

entire first trimester of pregnancy and we cannot determine if there is a critical period within the first trimester. The fewer boys born to women in the groups experiencing higher stress levels may be due to a differential susceptibility to miscarriage. Unfortunately, we had no data on the stress levels of women whose pregnancies ended in miscarriage or on the chromosomes of the spontaneously aborted fetuses, but it has been reported that spontaneously aborted fetuses have a higher male to female ratio than newborns (Byrne and Warburton, 1987). Even in early pregnancy, a male fetus has a faster growth rate (Henriksen et al., 1995) which may explain the differential vulnerability. We measured the distress level in first trimester but this may also reflect stress levels around the time of conception. It has recently been proposed that nonoptimal conditions at this time for male zygotes could cause non-optimal maturation and suboptimal implantation resulting in a higher risk of early loss of male fetuses (Jongbloet, 2004). In conclusion, our findings support animal evidence by showing that the male to female ratio, also among humans may be decreased by stress during the first trimester, not only at extreme levels, but at levels that affect a substantial part of the pregnant population. Our findings suggest that stress may contribute to the shift in human sex ratio observed in developed countries over the last decades (Moller, 1996; van der Pal-de Bruin et al., 1997; Marcus et al., 1998).

Suggest Documents