Available online at www.sciencedirect.

com

Psychiatry Research 160 (2008) 380 386 www.elsevier.com/locate/psychres

Emotional state and dreams in pregnant women

Alfredo Mancuso a , Antonio De Vivo a,, Giusi Fanara b , Salvatore Settineri b , Annamaria Giacobbe a , Alfonsa Pizzo a
a

Department of Gynecological, Obstetrical Sciences and Reproductive Medicine, University of Messina, Messina, Italy b Department of Neurosciences, Psychiatric and Anaesthesiological Sciences, University of Messina, Messina, Italy Received 29 December 2006; received in revised form 23 May 2007; accepted 7 June 2007

Abstract The aim of this study was to investigate the frequency of recall and the content of dreams during pregnancy, as well as their correlation with socio-demographic, obstetric and physicianpatients relationship variables, emotional state and duration of labour. A questionnaire, designed to analyse background characteristics, was given to 290 women in the third trimester of gestation. The psychiatric analysis of anxiety and depression was performed using the Hamilton Rating Scale for Anxiety and the Montgomery sberg Depression Rating Scale, while dreams were divided into masochistic and pleasant according to Beck's criteria. Oneiric activity was found to be associated with age 35 years, higher family income, higher educational level, and a satisfactory physicianpatient relationship. Masochistic content was associated with age b 35 years, quality of information and frequent thoughts of delivery. Concerning the emotional state, depression levels were higher in women reporting masochistic dreams, while no difference in anxiety levels was found. Labour duration was shorter in the dreamer group and in patients with masochistic dream content. These findings may indicate that, also in pregnancy, the number and the content of dreams are influenced by women's mood and that the evaluation of the oneiric activity might represent a useful tool for clinicians either to investigate the women's emotional state or to predict its repercussions on the course of labour. 2007 Elsevier Ireland Ltd. All rights reserved.
Keywords: Anxiety; Depression; Dreaming; Labor; Pregnancy

1. Introduction Dreams are symbolic manifestations with an emotional meaning linked to lived experiences, wishes and personal needs. Pregnancy, with its numerous anatomicfunctional changes that also involve the mental

Corresponding author. Viale Principe Umberto 38, 98122 Messina, Italy. Tel.: +39 090675807; fax: +39 090695201. E-mail addresses: antonio.devivo@gmail.com, giodoacri@libero.it (A. De Vivo).

sphere, represents a period in a woman's life in which oneiric activity is more frequent (Blake and Rietmann, 1993). These changes, added to the internal unconscious conflicts and basic anxieties linked to pregnancy, make dreams more accessible and more easily recalled. Dreams could therefore be considered a mirror of a pregnant woman's experiences and may reflect her emotional state, one what often contains anxiety and apprehension (Kron and Brosh, 2003). For these reasons dreams in pregnancy have been reported to be characterised frequently by unpleasant content (Ablon, 1994). In pregnancy, dreams with

0165-1781/$ - see front matter 2007 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.psychres.2007.06.005

A. Mancuso et al. / Psychiatry Research 160 (2008) 380386

381

unpleasant or masochistic content (Beck, 1967) become a psychological means for coping with the pregnancy and the approaching birth (Cartwright, 1986), just as reported for people with past or present mood disorders (Bears et al., 2000). This study was designed to test three hypotheses. One is that the occurrence and the content of dreams in pregnancy are influenced by demographic and obstetric variables, and by the physicianpatient relationship. The second hypothesis is that dreams and their contents reflect emotional state in pregnancy. Lastly, since it has been reported that sleep disorders in the general population modify the number and the content of dreams and that sleep disturbance can influence the duration of labour (Lee and Gay, 2004), then the third hypothesis of this study was to verify any possible repercussions of dreams and their content on the duration of labour. 2. Methods The study was carried out at the Department of Gynecological, Obstetrical Sciences and Reproductive Medicine in Messina University Hospital between January and July 2005. A questionnaire, written in collaboration with the Department of Neurosciences regarding the psychological aspects, was given consecutively to 290 pregnant women in the third trimester of gestation who were attending our Department for a prenatal consultation. In order to obtain a uniform sample, subjects were enrolled between 35 and 36 weeks of gestation; this is when pregnant women contact us for streptococcus group B screening. The women who agreed to participate gave their written consent. The questionnaire contained three sections. The first part included questions about socio-demographic background: age (b 35 years, 35 years), marital status (married, unmarried), level of education (high: high school, university; low: compulsory schooling), family income (middlehigh: $18,000, low: b $18,000) and occupation (employed, unemployed). The second part contained questions regarding the obstetric history: parity (nulliparous, multiparous), previous miscarriages (yes/no) and presence of obstetric complications in the current pregnancy (yes/no). The question used for the latter item was: Have you had any obstetric complications? (The following complications were mentioned: threatened abortion, gestational diabetes mellitus, gestational hypertension, intrauterine growth restriction, placenta praevia, and preterm labour). The third part focussed on the physicianpatient relationship: physician gender, a personal evaluation

(sufficient/insufficient) of the quality of information provided by the physician about the course and possible complications of pregnancy and delivery, regularity of prenatal controls (regular/occasional) and a subjective assessment of relations with the physician (satisfactory/ unsatisfactory). Women were also asked about the kind of delivery they preferred (vaginal birth/Cesarean section) and how frequently they thought about the actual delivery (rarely/frequently). A dichotomous reply was included for all the above questions. Two self-rating instruments were attached to the questionnaire in order to assess anxiety and depression. A specialist in psychiatry (G.F.) explained to patients how to respond to these scales. The Hamilton Rating Scale for Anxiety (HAM-A) (Hamilton, 1959) (Cronbach alpha reliability 0.92) was used to test the anxiety level of all patients. This includes 14 items, designed to evaluate mental and somatic symptoms with a score of 04 for each item and a total score ranging from 0 to 56. Women's mood was analysed using the Montgomerysberg Depression Rating Scale (MADRS) (Montgomery and sberg, 1979) (Cronbach alpha reliability 0.88) containing 10 items (score 06 per item) with a total score from 0 to 60. Since their introduction, these scales have become widely used and accepted outcome measures for the evaluation of anxiety and depression in clinical trials (Smith et al., 1990; Asher et al., 1995; Kersting et al., 2004). Data on dreams were collected by instructing all participants to write down in detail their dreams on awakening on a sheet of paper attached to the questionnaire. All women were asked to return the questionnaire, the two scales (HAM-A and MADRS) and the description of their dream within 15 days. Women who did not report a dream were called non-dreamers, while those who did were classified as dreamers. In order to highlight depressive symptoms, to analyse the content of a dream rather than to rely solely on the dreamer's reported emotion (as happens with Hall and Van de Castle's coding system), and to facilitate understanding by obstetricians, a simple dream content analysis was performed dividing women's oneiric activity into pleasant and masochistic. We borrowed this term from Beck (Beck, 1967) and used his scale to identify masochistic dreams. According to Beck, the term masochistic dreams refers to a set of unpleasant dreams with specific content. A dream can be considered masochistic if it contains one of the following elements: negative representation of the self, physical discomfort and harm/injury, thwarting of the dreamer's objective, deprivation (disappointment, loss, a lack), physical attack, non-physical attack, the dreamer is

382

A. Mancuso et al. / Psychiatry Research 160 (2008) 380386 Table 1 (continued) Non-dreamers Dreamers n (%) Thoughts about delivery Rare 22 (39.3) Frequent 66 (34.4) n (%) 34 (60.7) 1 126 (65.6) 1.2 0.62.3 0.5 OR 95%CI P

Table 1 Background characteristics: non-dreamers vs. dreamers (n = 248) Non-dreamers Dreamers n (%) Age b 35 years 35 years 80 (41.7) 8 (14.3) n (%) 112 (58.3) 1 48 (85.7) 4.28 1.99.5 140 (64.2) 1 20 (66.7) 1.11 0.52.5 108 (60) 1 52 (76.5) 2.16 1.14.1 24 (46.2) 1 136 (69.4) 2.65 1.44.9 0.72 0.0004 OR 95%CI P

Marital status Married 78 (35.8) Unmarried 10 (33.3) Financial income Middlehigh 72 (40) Low 16 (23.5) Educational level Low 28 (53.8) Middlehigh 60 (30.6) Occupation Unemployed Employed Parity Primiparas Multioparas

0.79

0.017

0.023

38 (38) 50 (33.8)

62 (62) 1 98 (66.2) 1.2

0.5

46 (35.9) 42 (35)

82 (64.1) 1 78 (65) 1.04 0.61.7 0.62.3 0.64 0.51.9

0.87

Previous miscarriage No 72 (36.4) Yes 16 (32) Current pregnancy Planned 82 (36.3) Unplanned 6 (27.3) Obstetric complications Yes 20 (35.7) No 68 (35.4) Physician gender Male 48 (36.9) Female 40 (33.9) Prenatal controls Regular 8 (66.7) Occasional 80 (33.9) Quality of information Sufficient 78 (36.1) Insufficient 10 (31.3) Relations with physician Unsatisfactory 10 (71.4) Satisfactory 78 (33.3) Preferred kind of delivery Vaginal 70 (35.7) Cesarean 18 (34.6) section

126 (63.6) 1 34 (68) 1.2

0.56

144 (63.7) 1 16 (72.7) 1.5

0.4

36 (64.3) 1 124 (64.6) 1

0.96

82 (63.1) 1 78 (66.1) 1.14 0.61.9

0.62

4 (33.3) 1 156 (66.1) 3.9

0.03 1.113.3 0.62.8

excluded, superseded or abandoned, dreamer gets lost, dreamer is punished, or dreamer fails. To establish the reliability of ratings of dream content as masochistic, the dream reports were presented in a randomised order to two raters (G.F. and S.S.), blind to the other information, who scored each report for the presence or absence of masochism. The rate of agreement was 97.5% with a Kappa value of 0.95. The statistical analyses were based on the first author's rating (G.F.). The duration of labour for women who delivered vaginally, from the onset of active phase to the fetal expulsion (Cunningham et al., 2001), was recorded and analysed in relation to the presence and type of oneiric activity. Prior to the final analysis, the data relating to pregnant women meeting the exclusion criteria were eliminated. These criteria were: previous cesarean section, psychiatric or neurological disorders, induction of labour in the current pregnancy and preterm birth (b 37 weeks of gestation). Statistical analysis was performed using the SPSS 13 software package. Association between categorical variables was tested by means of logistic regression. The odds ratios and their corresponding confidence intervals (95%CI) were estimated, and all the statistically significant variables in the univariate analysis were incorporated in a multivariate logistic regression model. The total scores of the two psychiatric tools (HAM-A, MADRS) and the duration of labour were reported as mean standard deviation (S.D.). The subgroups for these variables were compared using the Student t test, for parametric continuous data, and the MannWhitney

138 (63.9) 1 22 (68.8) 1.2

0.6 Table 2 Multivariate analysis: non-dreamers vs. dreamers OR Age Financial income Educational level Regularity of prenatal controls Physicianpatient relationship 5.3 4.4 3.3 2.4 6.6 95%CI 2.212.5 2.19.5 1.57.1 0.414.3 1.432.3 P value 0.0002 0.0001 0.002 0.34 0.018

4 (28.6) 1 156 (66.7) 5

0.008 1.516.4 0.51.9

126 (64.3) 1 34 (65.4) 1.1

0.88

A. Mancuso et al. / Psychiatry Research 160 (2008) 380386 Table 3 Background characteristics: pleasant dreams vs. masochistic dreams (n = 160) Pleasant n (%) Age b 35 years 35 years Marital status Married Unmarried Masochistic OR n (%) 2.3 1 1.24.6 0.016 95%CI P Table 3 (continued) Pleasant n (%) Masochistic OR n (%) 2.9 1.36.6 95%CI P

383

Thoughts about delivery Frequent 48 (38.1) 78 (61.9) Rare 22 (64.7) 12 (35.3)

0.007

42 (37.5) 70 (60.5) 28 (58.3) 20 (41.7)

58 (41.4) 82 (58.6) 12 (60) 8 (40)

2.12 0.815.5 0.12 0.93.5

U-test, for skewed data. The chi-square test was used for categorical variables. A P-value b 0.05 was considered to be significant. 3. Results Forty-two of the 290 women enrolled in the study were excluded because of incompleteness of replies or because they fell within the exclusion criteria. The remaining 248 women were thus considered in the analysis. Out of 248 pregnant women, 160 (64.5%) reported oneiric activity and were included in the dreamer-group, while the remainder (35.5%) who did not report dreams were put into the non-dreamer group. Concerning demographic characteristics, the univariate analysis showed a statistically significant association between the presence of dreams and the following: age 35 years (OR 4.3; 95%CI 1.99.5, P = 0.0004), middlehigh family income (OR 2.2; 95% CI 1.14.1, P = 0.017) and high educational level (OR 2.7; 95%CI 1.44.9, P = 0.0023). No significant associations were found for the other demographic variables (marital status and occupation) (Table 1). As far as obstetric characteristics are concerned, no significant associations were found between the variables considered and presence/absence of dreams (Table 1). On analysing the physicianpatient relationship, oneiric activity was found to be correlated with the regularity of prenatal controls (OR 3.9; 95%CI 1.113.3, P = 0.03) and with satisfactory personal relations with the physician (OR 5; 95%CI 1.516.4, P = 0.008) (Table 1). As regards the preferred mode of delivery, our data showed a higher percentage of dreams in patients who indicated Cesarean section (65.4% vs. 64.3%) and in those who declared thinking frequently of childbirth (65.6% vs. 60.7%), but without a significant association (OR 1.1, 95%CI 0.51.9, P = 0.88; OR 1.2, 95%CI 0.6 2.3, P = 0.5). In the multivariate logistic regression model, which was applied to all the statistically significant variables in the univariate analysis, the variable regularity of prenatal controls was no longer found to be correlated with the presence of dreams, while the correlations for the other four variables (age, family income, educational level and the judgement of own

Financial income Low 18 (34.6) 34 (65.4) Middlehigh 52 (48.1) 56 (51.9) Educational level Middlehigh 58 (42.6) 78 (57.4) Low 12 (50) 12 (50) Occupation Employed Unemployed Parity Primiparas Multiparas

1.7

0.11

1.34 0.63.2 1.35 0.72.6 1.48 0.82.8 2.17 1.14.7 1.32 0.53.7 0.52.2 0.93.4 0.21.4 1.312.8

0.5

40 (40.8) 58 (59.2) 30 (48.4) 32 (51.6)

0.34

32 (39) 50 (61) 38 (48.7) 40 (51.3)

0.22

Previous miscarriage No 50 (39.7) 76 (60.3) Yes 20 (58.8) 14 (41.2) Current pregnancy Planned 62 (43.1) 82 (56.9) Unplanned 8 (50) 8 (50) Obstetric complications No 54 (43.5) 70 (56.5) Yes 16 (44.4) 20 (55.6) Physician gender Male 30 (36.6) 52 (63.4) Female 40 (51.3) 38 (48.7) Prenatal controls Regular 68 (43.6) 88 (56.4) Occasional 2 (50) 2 (50) Quality of information Insufficient 4 (18.2) 18 (81.8) Sufficient 66 (47.8) 72 (52.2) Relations with physician Unsatisfactory 0 (0) 4 (100) Satisfactory 70 (43.7) 86 (56.3) Preferred kind of delivery Vaginal 50 (39.7) 76 (60.3) Cesarean section 20 (58.8) 14 (41.2)

0.048

0.6

1 1

0.9

1.8

0.06

1.3

0.8

4.1

0.014

NC

NC

NC

2.17 1.14.7

0.048

384

A. Mancuso et al. / Psychiatry Research 160 (2008) 380386

physicianpatient relationship) were still statistically significant (Table 2). Considering only the dreamer group, dream content (in line with Beck's criteria) was defined to be masochistic in 90 cases (56.2%) and pleasant in the remaining 70 (43.8%). When the kind of dream, masochistic or pleasant, was matched against demographic data, obstetric history and physicianpatient relations, the only significant associations found were for: age b 35 years (OR 2.3; 95%CI 1.24.6, P = 0.016), previous miscarriage (OR 2.2; 95%CI 1.14.7, P = 0.048) and sufficient quality of information (OR 4.1; 95%CI 1.312.8, P = 0.014) (Table 3). Moreover, dream type was found to be associated with preference of vaginal childbirth (OR 2.2; 95%CI 1.14.7, P = 0.048) and more frequent thoughts about own delivery (OR 2.9; 95%CI 1.36.5, P = 0.007). The multivariate logistic regression highlighted that the statistically significant variables from univariate analysis, with the exception of previous miscarriage and preference of vaginal delivery, were independently correlated with the masochistic content of dreams (Table 4). With regard to psychiatric evaluation, the HAM-A scale showed no significant difference in anxiety levels between dreamers and non-dreamers (14.5 7 vs. 12.9 6.6; t = 1.79, df = 246, P = 0.08) or between patients with masochistic and pleasant dream content (15.35 7.3 vs. 13.46 6.6; t = 1.68, df = 158, P = 0.2). The MADRS scale for mood analysis revealed a higher mean score of depression among women reporting masochistic dreams (13.7 8.1 vs. 6.6 5.1; U = 1306, P = 0.001), while no significant difference was found between dreamers and non-dreamers (10.7 5.1 vs. 9.6 4.4; t = 1.77, df = 246, P = 0.09). The total number of women who delivered vaginally was 144 (58.1%): 94 patients (65.3%) belonged to the dreamer group, while the remaining 50 women (34.7%) were in the non-dreamer group. The two groups were found to be similar as regards the frequency of pluriparas (60.6% vs. 54%, 2 = 0.35, P = 0.55). The duration of labour was longer in the second group (nondreamer) with a significant difference between the two mean total values (272.35 87.7 min vs. 228.88
Table 4 Multivariate analysis: pleasant dream vs. masochistic dreams OR Age Previous miscarriage Quality of information Preferred kind of delivery Thoughts about delivery 2.25 2.16 6.6 2.02 5.4 95%CI 1.024.7 0.94.9 1.824.7 0.84.7 2.113.6 P value 0.04 0.07 0.005 0.1 0.0003

97.5 min; t = 2.64, df = 142, P = 0.009). In the dreamer group, the frequency of pluriparas did not differ significantly between the masochistic and the pleasant dream groups (61% vs. 58.8%, 2 = 0, P = 0.99), but the duration of labour was found to be shorter in the masochistic dream group than in the pleasant dream group (214.48 98.9 min vs. 294.12 57.1 min; U = 345, P = 0.0018). 4. Discussion Oneiric activity could play an important role in the process of pregnancy (Sered and Abramovitch, 1992), but the literature on dreams in pregnancy is limited, based on small groups and not statistically supported. Pregnancy, with its anatomicfunctional and psychological changes, might determine either an increase in the frequency of dreams or a modification of their content. Some authors have reported that the majority of pregnant women declare that they dream (Condon, 1987; Sered and Abramovitch, 1992; Blake and Rietmann, 1993), or rather recall their dreams on awakening, and our results (64.5%) are in line with this. The ability to recall one's own dreams more easily is probably due to sleep disorders linked to various factors, including psychological ones (Schredl et al., 2003). Pregnancy, by modifying the physiology and the intensity of sleep (Hedman et al., 2002), might therefore lead to a high rate of dream recall. However, since the psychological evaluation of the two groups, dreamers non-dreamers, did not reveal any significant differences in anxiety and depression, one could deduce that the high frequency of dreams during pregnancy is a paraphysiological event linked to changes during pregnancy, without considering it as an expression of a real psychoemotive disorder (Dagan et al., 2001). Moreover, our study shows that dreaming correlates with age over 35 years, a higher educational and financial level, and a satisfactory relationship with own physician. For the first three variables, one might think that these could modify the women's psychological approach to pregnancy, while the datum concerning the physicianpatient relationship is harder to interpret. A different consideration should be made for the analysis of dream content. Our results confirm not only that pregnant women have a high incidence of unpleasant dreams (56.2%), as reported by Ablon (Ablon, 1994), but also that these dreams are associated with higher scores on the MADRS. This suggests that, in a similar way to the general population (Beauchemin and Hays, 1996), dream content in pregnancy is related to mood state, and that it becomes an important sign of the psycho-emotional state

A. Mancuso et al. / Psychiatry Research 160 (2008) 380386

385

of patients. However, this association was not found with anxiety. This is probably due to the particular behaviour of anxiety in pregnancy: it uniformly affects women with no distinction regarding background, it is intrinsic to the pregnant state (linked to the hormonal state and to the unpredictability of pregnancy), and it increases in the third trimester of gestation. This kind of anxiety, called gestational anxiety in the literature, is probably common both in dreamers and in non-dreamers, so no difference between the two groups was found. Furthermore, the instrument used to analyse dream content, by dividing the content into masochistic and pleasant, is more suitable as an indicator of a depressive profile than of an anxious profile. Unpleasant content was found to be associated with age b 35 years, more frequent thoughts about delivery and a sufficient quality of information about pregnancy. A possible explanation for the different level of depression in younger women could be linked either to the higher incidence of depressive disorders in this age group (Nguyen et al., 2005) or inadequate cognitive maturity provoking a less adaptive response to a new stressor event such as pregnancy. As regards the other two statistically significant variables in multivariate analysis, i.e. frequent thoughts of childbirth and more information about pregnancy, these are probably the way for these women to cope with pregnancy and delivery, and are the expression of an internal emotional process that the physician should take into consideration. Frequently, however, the specialist fails to understand this condition, and in providing technical information about pregnancy and delivery, does not help the woman work through this process. Moreover, contrary to previous reports (Sered and Abramovitch, 1992), masochistic dreams were not independently correlated with variables linked to obstetric factors such as parity, previous miscarriage, unplanned pregnancy and presence of obstetric complications. It has been reported that the rate of psychological disorders following previous adverse events, such as miscarriage, tends to fall to community levels within 1 year (Broquet, 1999). It is therefore likely that pregnant women close to childbirth have already resolved their internal conflicts and are focussing their attention on the current experience. Finally, as regards the correlation between dreaming and duration of labour, our findings show that patients with oneiric activity in the third trimester of gestation have shorter labour times than patients who do not report dreams. This difference is more evident in patients with unpleasant dreams. Kron and Brosh (Kron and Brosh, 2003) reported that pregnant women

who experienced masochistic dreams were less likely to develop post-partum depression, suggesting that these women do emotional work that readies them for birth and motherhood. It is probable that women who report unpleasant dreams have resolved their internal conflicts at an earlier stage and are well predisposed to face labour and delivery. Given the results of this study, the evaluation of oneiric activity and especially of dream content might represent a useful tool for clinicians either to investigate mood better or to predict its repercussions in the course of labour. In this respect, it is desirable that the obstetrician be aided in his task by other healthcare workers such as psychologists, midwives and nurses, in order to facilitate pregnant women in their internal emotional working process. However, further studies are needed to corroborate our results. References
Ablon, S.L., 1994. The usefulness of dreams during pregnancy. International Journal of Psychoanalysis 75, 291299. Asher, I., Kaplan, B., Modai, I., Neri, A., Valevski, A., Weizman, A., 1995. Mood and hormonal changes during late pregnancy and puerperium. Clinical and Experimental Obstetrics & Gynecology 22, 321325. Bears, M., Cartwright, R., Mercer, P., 2000. Masochistic dreams: a gender-related diathesis for depression revisited. Dreaming 10, 211219. Beauchemin, K.M., Hays, P., 1996. Dreaming away depression: the role of REM sleep and dreaming in affective disorders. Journal of Affective Disorders 41, 125133. Beck, A.T., 1967. Depression: Clinical Experimental and Theoretical Aspect. Harper & Row, New York. Blake, R.L., Rietmann, J., 1993. The pregnancy-related dreams of pregnant women. Journal of the American Board of Family Practice 6, 117122. Broquet, K., 1999. Psychological reactions to pregnancy loss application of the 30-item General Health Questionnaire and the Edinburgh Postnatal Depression Scale. Primary Care Update for Ob/Gyns 6, 1216. Cartwright, R., 1986. Affect and dream work from an information processing point of view. The Journal of Mind and Behavior 7, 281297. Condon, J.T., 1987. Altered cognitive functioning in pregnant women: a shift towards primary process thinking. British Journal of Psychology 60, 329334. Cunningham, G., Gant, N.F., Leveno, K.J., Gilstrap, L.C., Hauth, J.C., Wenstrom, K.D., 2001. Williams Obstetrics, 21th edition. McGraw-Hill, New York, p. 430. Dagan, Y., Lapidot, A., Eisenstein, M., 2001. Women's dreams reported during first pregnancy. Psychiatry and Clinical Neurosciences 55, 1320. Hamilton, M., 1959. The assessment of anxiety states by rating. The British Journal of Medical Psychology 32, 5055. Hedman, C., Pohjasvaara, T., Tolonen, U., Suhonen-Malm, A.S., Myllyla, V.V., 2002. Effects of pregnancy on mothers' sleep. Sleep Medicine 3, 3742.

386

A. Mancuso et al. / Psychiatry Research 160 (2008) 380386 Schredl, M., Wittmann, L., Ciric, P., Gotz, S., 2003. Factors of home dream recall: a structural equation model. Journal of Sleep Research 12, 133141. Sered, S., Abramovitch, H., 1992. Pregnant dreaming: search for a typology of a proposed dream genre. Social Science & Medicine 34, 14051411. Smith, R., Cubis, J., Brinsmead, M., Lewin, T., Singh, B., Owens, P., Chan, E.C., Hall, C., Adler, R., Lovelock, M., 1990. Mood changes, obstetric experience and alterations in plasma cortisol, beta-endorphin and corticotrophin releasing hormone during pregnancy and the puerperium. Journal of Psychosomatic Research 34, 5369.

Kersting, A., Dorsch, M., Wesselmann, U., Ludorff, K., Witthaut, J., Ohrmann, P., Hornig-Franz, I., Klockenbush, W., Harms, E., Arolt, V., 2004. Maternal posttraumatic stress response after the birth of a very low-birth-weight infant. Journal of Psychosomatic Research 57, 473476. Kron, T., Brosh, A., 2003. Can dreams during pregnancy predict postpartum depression? Dreaming 13, 6781. Lee, K.A., Gay, C.L., 2004. Sleep in late pregnancy predicts length of labor and type of delivery. American Journal of Obstetrics and Gynecology 191, 20412046. Montgomery, S.A., sberg, M., 1979. A new depression scale designed to be sensitive to change. The British Journal of Psychiatry 134, 382389. Nguyen, C.T., Fournier, L., Bergeron, L., Roberg, P., Barrette, G., 2005. Correlates of depressive and anxiety disorders among young Canadians. Canadian Journal of Psychiatry 50, 620628.