The effects of miscarriage and other ‘unsuccessful’ pregnancies on feelings early in a subsequent pregnancy

Full Terms & Conditions of access and use can be found athttp://www.tandfonline.com/action/journalInformation?journalCode=cjri20

Download by: [ Josephine Green] Date: 02 November 2015, At: 01:40

Journal of Reproductive and Infant Psychology

ISSN: 0264-6838 (Print) 1469-672X (Online) Journal homepage: http://www.tandfonline.com/loi/cjri20

The effects of miscarriage and other ‘unsuccessful’pregnancies on feelings early in a subsequentpregnancy

Helen Statham & Josephine M. Green

To cite this article: Helen Statham & Josephine M. Green (1994) The effects of miscarriageand other ‘unsuccessful’ pregnancies on feelings early in a subsequent pregnancy, Journal ofReproductive and Infant Psychology, 12:1, 45-54, DOI: 10.1080/02646839408408867

To link to this article: http://dx.doi.org/10.1080/02646839408408867

Published online: 11 Dec 2007.

Submit your article to this journal

Article views: 49

View related articles

Citing articles: 25 View citing articles

/oumal of RrpmLm'vc and Infmr Plychology Vol. 12, pp. 45-54 (1994) Q, 1994 by the Society for Reproducriveand I&r Psychology

02G4-6838194/0 10045-1 0%05.00

The effects of miscarriage and other 'unsuccessfbl' pregnancies on feelings early in a subsequent pregnancy

HELEN STATHAM and JOSEPHINE M. GREEN Crntrr for Family Research, Uniucrsi~ of Cmbridge, Free ScbmI Lam. Cambridge CB2 3RF. UK

Abstract Data from pregnant women recruited to a prospective study of the psychologid and social d f e c t s of screening for fctal abnormality were andyscd according to previous reproductive experiences. In chis paper we report on the effects of thcsc previous experiences on women's feeling early in this pregnancy. Women who had experienced a previous unsuccessful pregnancy were more anxious, both generally and specifically, about the possibility of something being wrong with the baby and about the possibility of miscarriage. Women with unsuccessful pregnancies and no living children had the highest trait anxiety, when measured a d y in pregnancy. We show a number of ways in which women who have experienced a previous pregnancy failure differ from women who have not; thcsc difirenca vary with a number of fictors, notably the type of unsu-ful pregnancy, whether or not the woman has living children, the timing of the unsucccssful events, and whether a woman has also had a non-medical termination of pregnancy. We also t ~ p o ~ on a number of differences between women who have and have not been pregnant before.

Introduction

There is a growing recognition of the distress felt by women, and their partners, if a pregnancy fails. Late miscarriage (Lovell, 19831, termination for abnormality (Statham, 1992) and increasingly, early miscarriages (Iles, 1989) are more likely than previously to be viewed as a 'perinatal bereavement' with women experienang similar reactions to those women whose babies have been stillborn o r died in the neonatal period. It is also increasingly recognized that the impact of any unsuccessful pregnancy (including terminations for nonmedical reasons) may last well beyond the period of acute grief around the time of the event.

Women's reactions to the difFerent pregnancy losses have been reported according to different perspectives. The impact of miscarriage has been described in two ways: a large 'lay' literature has drawn on women describing their own experiences and feelings about miscarriage, and, indeed, this is the literature which has resulted in the wider recognition of the practical and emotional needs of miscarrying women (Oakley Hal, 1930; Moulder, 1990); a second approach has examined the short- term psychological and psychiatric responses to miscarriage (Friedman and Gath, 1989; Neugebauer et aL, 1992); research on recurrent miscarriage has focused on investigating aetiology and methods of treatment (reviewed by Gannon, 1994). For other types of reproductive &lure, individuals have written about their experiences (e.g. Neustater and Newson, 1986 for non-medical terminations); there have been psychiatricoutcome studies (e.g. Iles, 1989 for terminations for fetal abnormality); and there is a large literature of 'retrospective studies' (see Statham, 1992).

It is rhrough retrospective studies that parents have been abk to declare the effect of their previous experience on a subsequent pregnancy: "The next pregnancy was often described as a period of intense anxiety not relieved until the safe delivery of a normal child" (Elder and Laurence, 1991). However, we know of no study in which the nature and extent of this anxiety in a subsequent pregnancy has been quantified. Furthermore, little is known about this anxiety relative to other pregnant women, since most of the studies in this area la& comparison groups. The 1824 pregnant women who were recruited to the Cambridge Prenatal Screening Study (Green H a L , 1993~) included women who had experienced a range of reproductive experiences as well as women who had never been pregnant before. In this paper, we present data from this unselected group ofwomen which examines the ways in which different reproductive experiences, and the time at which they occurred, influence feelings early in a subsequent pregnancy.

Dow

nloa

ded

by [

Jose

phin

e G

reen

] at

01:

40 0

2 N

ovem

ber

2015

46 H e h Smtham andjosepbine M. Gmm

Methodology The Gmbridge Prenatal Screening Study

The purpose of the study was to chart the knowledge, attitudes, anxieties and experiences of pregnant women from before their first hospital appointment through to the postnatal period, with a particular focus on screening for fetal abnormality during routine antenatal care. A series of four postal questionnaires, designed for the study, allowed us to look at women’s experien’ces in the context of other things happening in their lives. Other Fdctors that have been taken into account have included attitudes to abortion, relationships, life events and coping strategies. For further details of the study, see Green etaL (1992, 1993a-c), Statham etaL (1992). This paper presents data from the first questionnaire only, which was completed before the first hospital visit.

Rrcruitmrnt of sample Pregnant women were recruited to the study between January and March 1990 in nine District hospitals, all within 60 miles of Cambridge (UK). Women were recruited via a letter and questionnaire forwarded to them with notification of their first hospital appointment. Many, but not all. of the 1824 women who joined the study were recruited by about the 1Oth-12th week of pregnancy; data in this paper are restricted to women who were less than 16 weeks pregnant at the time of recruitment and who answered questions relating to reproductive history (maximum n = 1496, but not all women answered all of the questions).

Measures The ‘trait’ scale of the Spielberger State-Trait Anxiety Inventory (Spielberger et al., 1970) was used to assess anxiety as a personality characteristic. Specific worries were measured using the Cambridge Worry Scale (Statham etd, 1992; Green ct aL, 199%); this list of items of possible concern to pregnant women (16 at the first questionnaire) is scored from 0 (not a worry) to 5 (extremely worrying) and covers pregnancy, health and socio-economic matters. Three items will be considered in detail in this paper: ‘the possibility of miscarriage’; ‘giving birth’; and ‘the possibility ofsomething being wrong with the baby’. An adjective checklist was used (Green ctal., 1991) in which women were asked to circle all of the words in a list of 21 that described their feelings at a given time. In this paper we will report on the choice of just two words, ‘confident’ and ‘anxious’. Data will also be presented showing women’s responses to questions’ about how they felt on finding they were pregnant and whether they thought of their baby as being a person.

Am& Data were analysed using the Statistical Package for the Social Sciences (SPSSX on the University of Cambridge’s main frame computer. Categorical data were analysed using the chi- squared test and continuous data by analysis of variance. Analysis of covariance was used to control for the effects of age and trait anxiety where necessary.

Samplr c b a r a c t d c s Prryiow reproductive events

Table 1 shows the sample dassified according to experience of the following reproductive events: (i) ‘successful’ pregnancies, i.e. those that resulted in the birth of a heathy liveborn child; (ii) miscarriages; (iii) other ‘unsuccessful’ pregnancies. These were primarily stillbirth, birth of an ill or handicapped child, and terminations for medical reasons (including maternal health) which occurred in roughly equal numbers.

One hundred and forty women reported that they had undergone a termination of pregnancy for which there were no medical indications; these have been distinguished from ‘unsuccessful’ pregnancies in that the premature cessation of the pregnancy is presumed to be the desired

Dow

nloa

ded

by [

Jose

phin

e G

reen

] at

01:

40 0

2 N

ovem

ber

2015

Mkarriagr andjklings in nrbseqrrmtpnpmy 47

Previous rtproductivc history (acluding women with non-medial terminations of pregnancy). Table 1.

n %

Women with no living children GpN No previous pregnancies 505 37 GpU Unsuccessful’ only 25 2

GpS Successful. pregnancies only 498 37

GpSlM Succcssful + miscarrhge 172 12

GpM Miscarriage only 77 6 Women with living children

GpSlU Succcssful + unsuccessful 79 6

Total 1356 100 ~

‘See ten for definitions and explanations.

outcome, in contrast to miscarriages and those events that we have called ‘unsuccesshl’. In general, they appear to have virtually no impact on psychological outcome measures (see Green et aL, 1993c) but there are a small number of situations where they do prove to be relevant. The data presented in this paper exclude women who have undergone non medical terminations except where specifically mentioned. Table 1 defines the abbreviations by which we will refer to the groups of women with different reproductive cxperiences. In the remainder of this paper the phrase ‘unsuccessful pregnancy’,. unless qualified, will be used to refer to unsuccessful pregnancies other than miscuriages.

Thirty six women had experienced both miscarriages and other unsuccesshl pregnancies. They have been classified according to the other unsuccessful pregnancy (n = 27 in Group S/U, = 9 in Group U). Thirty-four per cent (284/847) of the women who had been pregnant before had experienced a miscarriage and 12% (1041847) an unsuccessful pregnancy. Overall, 42% of this sample who had been pregnant before had undergone a previous pregnancy loss.

Demographic charactm.sta*cs Fifty-six per cent of the sample had ceased full-time education at the present minimum age or younger, 31 % had some additional education up to the age of 18, and the remaining 13% had had higher education (vs 9% for this age group nationally, Central Statistical Office, 1991). Given the higher levels of education in the region where the study took place, the sample is broadly representative of the population from which it was drawn, exccpt for an under-representauon of non-Caucasian women. Women with unsuccessful pregnancies (Groups U and S/U) were more likely to have had less education (72% and 65%, x2 = 18.5, df = 10, p < 0.05). Overall, 59% of the sample were in paid employment; women with living children were less likely to be so than those without children. The sample ranged in age from 16 to 43 years, mean 26.9 years, SD 5.0. As one would expect, women with children were older than those without, and women who had not had any form of unsuccessful pregnancy were younger than those who had. Women who had not been pregnant had the youngest mean age (24.8, SD 4 3 , with 19% of this group being aged 20 years or less, compared with 11% of the entire sample.

Results Trait anxiety

Trait anxiety scores differed significantly berween the groups, as shown in Table 2. Not all of the women listed in Table 1 completed the trait scale.

There was an overall difkrencc in trait anxiety scores between women with and without living children (mean scores 38.0, SD 8.0 and 39.0, SD 8.4 respectively, F = 5.2, df = 1 , p < 0.05). However, there was no difference between women who had not been pregnant before and those who had only had successful pregnancies (Groups N and S). Women with previous pregnancies

.

Dow

nloa

ded

by [

Jose

phin

e G

reen

] at

01:

40 0

2 N

ovem

ber

2015

48 Hekn Smtlxrm and Josephine M. Green

Table 2. Tnit Anxiety according to reproductive history.

Tnit anxiety n Man SD

Women with no living children GpN No previous pregnancies 487 38.7 8.3 GpU Unsuccessful* only 24 41.4 9.3

GpS Successfui* prcgnamics only 474 38.3 8.2 GpSlU Succashl+ unsuccasful 76 39.2 8.2 GpSlM Successful + miscarriage 165 36.5 7.4

GpM Miscarriage only 74 40.4 9.1 Women with living children

Whole Sample 1300 38.4 8.2

*See ten for definitions and explanations.

F = 3.6, &= 5, p < 0.01

but no living children (Groups U and M) had higher trait anxiety scores (mean score 40.6, SD 9.1) than those who had never been pregnant before (F= 4.3, df = 1, p < 0.05).

80

75

Feelings about bringpregnant Seventy-three per cent of the sample gave positive responses in answer to the question ‘how did you feel about finding that you were pregnant?’; 24% had ‘mixed feelings’ and 3% were negative. Fewer women with previous unsuccessful pregnancies were positive (Group U: 56% and S/U: 53%) and they were more likely to state that they had mixed feelings ( U 44% and S/U: 42%; x2 = 35.0, df = 10, p < 0.0001). Women who had had previous miscarriages gave h o s t identical responses to women with no problem pregnancies.

-

-

‘Do you think ofyour baby as bring a person yet?’ The groups differed in their responses to this question, and Figure 1 shows that there are two effects operating.

Firstly, women with children were consistently more likely to answer ‘yes’ than those without children (75% vs 66Y0, x2 = 12.5, df = 1, p < 0.001). Secondly, within these groups, women who had had miscarriages or other unsucctsshl pregnancies were more likely to see their baby as a

&I- 70

60

55

85 I

GP N G p U GpM G p S G p S N G p S M Figure 1. Percentage of each group answering ‘ya’ to the question ‘Do you think of your baby as being a

penon yet?’ See Table 1 for description of groups.

Dow

nloa

ded

by [

Jose

phin

e G

reen

] at

01:

40 0

2 N

ovem

ber

2015

person than those who had not: 75% of women with previous misariages answered ‘yes’ as did 8 1 % of those with other unsuccessful pregnancies, compared with 68% of other women, x2 = 9.2, df = 2, p < 0.01).

Adjrctivc Chcck-lrit Choosing the words ‘anxious’ and ‘confident’ from the adjective check-list also varied both with whether or not women had living children, and whether or not they had had unsuccessful pregnancies or miscarriages as shown in Table 3.

Table 3. Percentage of women describing thcmsclvcr as ‘confident’ and ‘anxious’ according to reproductive history.

Reproductive Percent of womcn circling:

GpN’ 13 52 8 56

14 57 GPU

23 34 GPM

10 56 GPS

GpSlM 17 57 Total sample 17 47 x’ 24.1 68.6 df 5 5 P < 0.001 < 0.0001

history ‘confident’ ‘anxious’

GpSlU

‘See rexc for description of groups.

Twenty per Cent of those with children were confident, compared with 13% of those without (x‘ = 13.3, df = 1 , p < 0.001); women with children were less likely to be anxious than women without(41%against54%,x2=21.9,df= l,p<O.OOOl), butthisis accountedforentirelybythe smaller number ofwomen with only successful pregnancies (Group S) reporting they were anxious as shown in Table 3.

Womk in carlyprepnty Figure 2 and Table 4 show the way in which women’s worries about miscarriage, giving birth and the possibility of something being wrong with the baby reflected their previous experiences. Differences shown in Table 4 remained when the analyses of variance for specific worries were repeated controlling for trait anxiety and age.

Further analyses showed that worries were lower for women with a successful pregnancy. Two-way analyses of variance yielded Fvalues of 71.1, 169.0 and 54.3 (df = 1 , all p c 0.0001) for

Worries about miscarriage ((mist'), giving birch (‘birch’) and the possibility of something being Table 4. wrong with the baby (‘baby’) according to previous reproductive history.

Rcproductive Worry about history ‘Misc’ SD ‘Birch’ SD ’ ‘Baby’ SD

GpN+ 2.7 1.7 3.1 1.5 4.0 1.4 2.3 1.8 3.9 1.4 3.0 1.6 1.9 1.7 1.9 1.7 3.0 1.8 1.7 1.7 GpSlU

GpSIM 3.2 1.8 1.7 1.7 Sample M a n 2.6 1.8 2.4 1.8 F 32.6 39.1 df 5 5 P < 0.0000 < 0.ooog ‘See rexr for description of groups.

GPU GPM GPS

2.8 1.5 3.8 1.5 3.0 1.5 2.2 1.5 2.9 1.6 2.4 1.5 2.6 1.6

13.6 5

< 0.0000

Dow

nloa

ded

by [

Jose

phin

e G

reen

] at

01:

40 0

2 N

ovem

ber

2015

50 Helm Stathvrm and Josq&u M. Gmm

s/

Miscamage Wow about birth Worry about baby

U G p N m G p U U G p M GpS U G p S N H G p S / M

something being wrong with the baby’. Women were grouped according co heir previous reproducrive history as dcscribai in Table 1.

the effect of having, or not having living children on each of the above worries with the type of pregnancy Failure significant for worries about miscarriage (F= 60.5, df = 2 , p < 0.0001) and about the possibility of something being wrong with the baby (F= 11.5, df = 2, p < 0.0001).

Among the women without living children, mean worry scores about giving birth were higher for Groups N and M, (3.1, SD 1.6) than for those in Group U (2.3, SD 1.8). i.e. those without the experience were more worried than those with ( F = 5.5, df = 1, p c 0.05) Worry about the possibility of something being wrong with the baby was specifically a greater worry for those women with unsuccessfil pregnancies other than miscarriage (F= 66.3, df = 2, p < 0.01). Worry about miscarriage, however, was equally high for women with previous miscarriages (Groups M and S/M, mean 3.4, SD 1.7) and other unsuccessful pregnancies (Groups U and SIU, mean 3.3, SD 1.8) with both greater than for women without a failed pregnancy (mean 2.44, SD 1.7; F = 43.5, df = 2, p < 0.0001). Within Groups U and S/U, women who had also experienced a miscarriage had higher mean scores for worry about miscarriage than women who had not undergone a miscarriage but this was only significant for Group S/U (mean scores 3.6, SD 1.7 and

Figure 2. Mean scores for worry about ‘rhe possibility of miscarriage’. ‘giving birrh‘ and ‘the possibiliry of

2.7, SD 1.8, F= 4.3, df= l,p< 0.05).

Timing of miscamizgcs and other unsucccss+lprpancus The data presented so far suggest that a successful pregnancy mitigates some of the effects of a miscarriage/unsuccessful pregnancy. The data were therefore further subdivided to investigate the importance of the order of these events, i.e. is the effect of a successful pregnancy only observed if the baby has been born since the miscarriage or unsuccessful pregnancy? This analysis is limited to those women who had had at least one miscarriage or other unsuccessful pregnancy, i.e. Groups U, M, S/U and S/M and who gave dates of those events. These groups have been subdivided as follows: groups M and U are as defined previously, i.e. women whose only experiences of pregnancy have ended in miscarriage or otherwise unsuccessfully; in groups S pre-M and S pre-U, the failed pregnancy was the most recent event but had been preceded by a successful pregnancy; in groups S post-M and S post-U, a successful pregnancy had occurred since the failed event. No further differences in trait anxiety emerged from this analysis; women with either unsuccessful event were most anxious if they had no living children. The pattern of worries shown in Table 5 , however, shows clearly that with regard to worry about both miscarriage and the possibility of something being wrong with the baby. a successful pregnancy is only protective if it has occurred since the unsuccessful event.

Dow

nloa

ded

by [

Jose

phin

e G

reen

] at

01:

40 0

2 N

ovem

ber

2015

Miscankgc and&htgs in mbseqwntpregna~ 5 1 The effect of the timing of a succasful pregnancy on tnit anxicty and worries for women with Table 5.

previous miscarriages or unsuccessful pregnancies. Worry about

Tnit ‘Misc’ ‘Birch’ ‘Baby’ Gmup n M a n SD M a n SD M a n SD Mean SD

Ma 77 40.4 9.1 3.9 1.4 3.0 1.6 3.0 1.5 S pre-M 92 36.3 7.4 3.6 1.6 1.6 1.6 2.7 1.5 S post-M 59 35.8 7.0 2.4 1.8 1.6 1.7 1.9 1.3 P 7.3 15.7 17.8 8.5 df 2 2 2 2

U’ 24 41.4 9.3 4.0 1.4 2.3 1 .8 3.8 1.5 P s pre-u 26 36.3 7.4 3.7 1.5 1.5 1.6 3.4 1.4 s post-U 17 39.8 6.6 2.3 1.9 1.6 1 .a 2.3 1.6 P 2.2 6.6 1.6 5.2 df 2 2 2 2 P ‘See text for dercripciin of groups

< 0.001 < 0.0001 < 0.0001 < 0.001

ns < 0.01 ns c 0.01

A successfL1 pregnancy that preceded a miscarriage or other unsuccessful pregnancy does little to reduce both worry about miscarriage and about something being wrong with the baby relative to the scores ofwomen with no successfd pregnancy. For all of the measures, there were no overall differences between women whose unsuccessful event was within the last year, compared with those for whom it was longer ago.

Prrviour non-medical tmnimtions o f p r p n r y A small number ofwomen in each of the reproductive history groups reported to us that they had undergone a non-medical termination, as shown in Table 6. Trait anxiety and worry scores are also shown for these women in Table 6.

While there was no significant main effect of non-medical terminations on trait anxiety, (means are shown in Tables 2 and 6, F= 2.3, df = 1, p = ns) there was a significant interaction with reproductive history group (F= 2.4, df = 5 , p c 0.05): women with previous miscarriages (Groups M and SIM) had relatively higher trait anxiety scores if they had also had a previous non-medical termination. These analyses take no account of the timing of the non-medical termination relative to other events, although for many of the women in the sample the non-medical termination was an early reproductive event.

Analyses ofvariance for the worries described above showed no overall effectof a non-medical termination but did reveal that, for worry about miscarriage, there was a significant interaction

Table 6. Trait Anxiety and worries according to reproductive history for women who had undergone a termination of pregnancy for non-medial rasons.

Worry about Tmit ‘ M i d ‘Birrh’ ‘Baby’

Gmup n M a n SD M a n SD Mar. SD M a n SD 54 39.8 8.3 2.2 1.7 3.0 1.6 2.8 1.7

3 41.3 6.1 4.0 1.7 2.3 1.5 4.3 1.2 GPN

9 44.8 11.7 4.7 0.5 2.3 1.9 3.2 1.5 GPU

60 38.3 8.3 1.9 1.7 1 .a 1 .a 2.6 1.6 GPM GpN* GpSIU* 4 35.0 7.9 2.3 1.5 4.3 1.1 3.0 1 .o GpSfM 10 46.1 10.5 4.7 0.5 2.3 1.9 3.2 1.5 Sample mean 39.8 8.9 2.4 1.8 2.4 1 .a 3.6 1.3 F 2.0 7.0 3.5 1.3 df 5 5 5 5 b 0.09 < 0.0000 < 0.01 ns

‘See text for dcrriptiin of groups.

Dow

nloa

ded

by [

Jose

phin

e G

reen

] at

01:

40 0

2 N

ovem

ber

2015

52

between a woman’s reproductive history and a previous non-medical termination. Again, i t was women with previous miscarriages who had higher scores for worry about miscarriage if they had undergone a non-medical termination (F= 2.54, df = 5, p < 0.05 forinteraaion) while the other groups showed no such difference; rather, women who had undergone a non-medical termination appeared less worried about miscarriage than women who had never been pregnant before. However, when this analysis was repeated controlling for trait anxiety, this interaction was no longer significant (F= 2.0, df = 5 , p = 0.08).

Hekn Smthm and Josephine M. Gmn

Discussion

The data provided in this paper are unusual in that the women were recruited to the study as ‘ordinary pregnant women’ rather than because they were women with a failed pregnancy. This has allowed comparison of the feelings of women who have had previous pregnancies which have not resulted in the birth of a healthy baby both with women who have never been pregnant before and with women who have only experienced a successful pregnancy. Other studies in which the impact of previous pregnancy experiences have been examined differ in two ways: firstly they have tended to recruit because of the previous pregnancy experience and therefore, secondly, do not include women who have not been pregnant before (Rajan and Oakley, 1993; Niven rtaL, 1993).

Women who had had previous miscarriages, those who had had other unsuccessful events and those without failed pregnancies differed from each other in a number of ways. Overall trait anxiety scores are similar to those reported by Spielberger for young undergraduate females (Spielberger ct at!, 1970). However, women who had only had unsuccessful experiences (Groups U and M) had significantly higher trait anxiety scores. For the sample excluding women with non- medical terminations, these did not reach the levels reported by Spielberger for patients with neuropsychiatric disorders (means of 44.6-53.4 for different types of disorder); women with previous miscarriages and non-medical terminations of pregnancy did have mean scores at the lower end of this range (44.8 and 46.1). If we consider the data acluding women with non- medical terminations, they suggest that eirher (i) women who are of a particularly anxious disposition are more likely to have pregnancies that go wrong, or (ii) that basic personality characteristics can be changed by experiences, or (iii) that basic personality characteristics cannot be changed but that the Spielberger measure of trait anxiety is not a pure measure. The first explanation has been advanced (see the critical review by Gannon, 1994) but does not seem knvincing when we see that, in this study, those who have also had successful pregnancies show no such effect. Similarly, Grimm (1962) found that women who had suffered recurrent miscarriages showed personality characteristics closer to those of a control group once they had reached 35 weeks gestation. The additional effect of a non-medical termination would suggest that one of the other two options is a more likely answer; choosing between these is part of a wider debate that is beyond the scope of this paper. A subsequent paper (in preparation) which will report data from these same women throughout the rest of their pregnancies. will contribute further to this debate.

Women who had had previous miscarriages and those who had had other unsuccessful events differed fiom each other in a number of ways. The latter, hr example, were much more likely to have mixed felings about the current pregnancy while the responses of women who had previously miscarried were the same as those who had not had any problems. These findings are somewhat unexpected as are those concerning attitudes to the baby. Niven (1 992) has suggested that women with a previous reproductive loss “may cope with the subsequent pregnancy by trying to avoid attachment to the fetus” (p. 12). Our data show that women with previous losses were significantly more likely than other women to think of their baby as a person in early pregnancy. Our question is, of course, only measuring one element of attachment and is therefore not strictly comparable with other measures, e.g. Niven ct af (1993). Nevertheless, that our findings should

Dow

nloa

ded

by [

Jose

phin

e G

reen

] at

01:

40 0

2 N

ovem

ber

2015

Miscnrriuge undfiehngs in subwquortprepnry 53

show asignificant difference in theopposite direction is ofsome interest particularly since our data are consistent with Niven etd.’s (1993) for women of different parity.

The data concerning women’s specific worries are more predictable: a successful pregnancy reduces worries (Erickson, 1976); women with no previous pregnancies resemble women with mixed pregnancy experiences; women with failed pregnancies are more worried about both miscarriage and something being wrong with the baby than women without those experiences. For women with mixed pregnancy experiences, however, the one which determines their worries is the most recent: thus women whose last pregnancy resulted in miscvriage or was otherwise unsuccessful are only marginally less worried about the possibility of miscarriage and of something being wrong with the baby than those with no successful pregnancies. Worry about giving birth was high, even at this early stage of pregnancy, for women who had not given birth before.

Women’s responses to non-medical terminations have been widely studied, with most reports concurring that while guilt, sadness, and regret are commonly experienced prior to the abortion, these feelings are neither extensive nor long-lived (Lazarus, 1975; Urquhart and Templeton, 1991). Iles (1989) has indicated factors associated with an adverse outcome. For the women in this study, a non-medid termination of pregnancy had no overall effect on worries; there was however a small subgroup ofwomen in whom it was possible to observe a significant effect of such an event. These were women who had also experienced miscarriages. Whether or not the women had living children, those with a miscarriage and a non-medical termination had higher mean trait anxiety scores and were significantly more worried about miscarriage than those who had not had a non-medical termination. I t may be that these were women who were anxious before the termination or reacted adversely to it. Alternatively, some may have become particularly anxious since the miscarriage, believing it to be a sign that physical damage had occurred as a result of the operation or even a form of punishment that might be meted out again in this pregnancy.

Pregnancy failure is a frequent event; 42% of the women in this study, and 43% of the women in the Social Support and Pregnancy Outcome study (Rajan and Oadey, 1993) who had been pregnant before had experienced a previous pregnancy loss; our data argues that women’s anxieties and concerns in pregnancy should be recognized in the Context of their previous pregnancy experiences or lack of pregnancy experience. Rajan and Oakley (1993) have shown how the emotional well-being of women with a previous pregnancy loss cul be improved by socia S U ~ ~ O K

during a subsequent pregnancy. Appropriate care in those subsequent pregnancies (Bourne and Lewis, 1984; SANDS, 1991; SATFA, 1991; Gannon, 1994) needs to take account of the type of reproductive failure and when it occurred in order to meet women’s specific needs.

Acknowkdgmnts: We would like to acknowicdg the help of Claire Snowdon, our co-worker on the study from which this paper dnws its data, and the Health Promotion R a ~ d Trust, who funded the study.

References

Bourne, S. and Lewis. E. (1984) Pregnancy afier stillbirch or neonatal death. Lancer, ii: 31-33. Central Statisrid O&cc (1991) S&I T d 21. Government Statistical Service. HMSO, London. Elder, S.H. and Laurence. K.M. (1991) The impact of supportive intervention after 2nd trimester ter-

mination of pregnancy for fetal abnormality. P m t a l D~Z~XOJ~S, 11: 47-54. Eridcson, M.T. (1976) The influence of halth Factors on psychologid nriables prcdiaingcomplications of

pregnancy, labour and ddivuy. journu1ofPycbosomutic Rewarcb, 20: 21. Friedman, T. and Gath, D. (1 989) The psychiatric consequences of spontaneous abortion. Britirk/mtmal of

PyhLzhy, 155: 810-813. Gannon, K. (1994) Psychological fiaors in thc aaiology and treatment of recurrent miscarriage. A review

and critique. /mtrnulofReprodrrm’vc udZnfant P~ybokyy, 12: 55-64. Green, J.M., Richards, M.P.M., Kitzinger, J.H. and Coupland, V. (1991) Mother’s percepcions of their 6-

wcdt-old babies. Rdationship with antcnaml, intnpartum and postnatal factors. Irish journal of PzrchologK 12: 133-144.

Dow

nloa

ded

by [

Jose

phin

e G

reen

] at

01:

40 0

2 N

ovem

ber

2015

54 H e h Stathm and Josqhine M. Green

G m n . J.M.. Statham, H. and Snowdon, C. (1992) Screening for fetal abnormalities: attitudes and expcrienccs. In: Chard, T. and Richards, M.P.M. (Edc) Obstrtrics in tbe 1990s: Current COniTOVffJjCS.

McKcith Press, London. Green, J.M., Statham. H.M. and Snowdon, C. (1993a) Women’s knowledge of prenatal screening tats. 1:

Relationships with hospital scraning policy and demographic factors. JournalofRrpmductive andlnfint

Green. J.M.. Snowdon, C. and Statham. H. (1993b) Pregnant women’s attitudes to abortion and prenatal

Green, J.M., Statham. H. and Snowdon, C. (199%) P n g ~ n r y : A Testing Em. Report of the Cambridge

Grimm, E.R (1962) Psychologid investigation of habitual abortion. PIychormMric Medicine, 24: 369-378. Ilcs. S. (1 989) The loss of a d y prrgnancy. Baifimi Cliniurl Obstctria and G ~ c o i o g y , 3: 769-790. Lazarus, A. (1 985) Psychiatric sequdac to therapeutic termination of pregnancy. Jounurl of Psychmomatic

Lovcll. A. (1983) Some questions of identity: late miscarriage, stillbirth and pcrinad loss. Social Science and

Moulder, C. (1990) Miscarriage: Women’s l3prienres and Need. Pandora Press. London. Neugcbauer, R, Kline, J., O’Connor, P., Shrout, P., Johnson, J., Skodol, A.. Wicks, J. and Susscr, M. (1992)

Depressive symptoms in women in the six monrhs after miscarriage. American Journal ofOLtcm’cs and Gynecobg)c 166: 104-109.

Pzycholog)r 1 I: 11-20.

scraning. JoumlofRrpductive andZnfmtPr)rkobg)r 1 I: 31-39.

P m r d Screening Sm+. Gntre b r Family Research, Cambridge.

Obstchirs and Gpcoology, 4: 141-150.

Medkine, 1 7 755-761.

Ncustatcr, A. and Newson. G. (1986) Mired Feelings: Tbe Ey~~etjmre ofibortion Pluto Press, London Nivcn, CA. (1992) Ps~hobgical Gzre fw Families: Bcfow, DuringandAfict Birth. Butrenvorth-Hcinemann,

Nivcn, C.. Wiszniewski. C. and AIRoomi, L. (1993) Amchmcnt (bonding) in mothers of prcterm babies.

Oaklcy, A.. McPhcrson, A. and Robem, H. (1990) Mucarriage. Penguin, Harmondsworth. R a j a , L. and Oaklcy. A. (1993) No pills b r heartache the impomnce of social support for women who

SANDS (1 991) Guihline~fi Pmjkswd: Miscmiage, StiUbirth andNeomdDeath. Stillbirth and Neonatal

SATFA ( 1991) Support a& Tmnination f m Abnwmality: A Parma Handboo& SATFA. London. Spielbcrgcr, C.D., Goauch. RL. and Lushcne, RE. (1970) Tbe Statc-Trait Anrirry ZnvmtoT. Consulting

Psychologists Press. Palo Alto. CA. Statham. H.. Green. J.M. and Snowdon, C. (1992) Psychological and social aspecrs of screening for fetal

abnormality during routine antenatal care. In: Robinson, S., Thomson, A. and Tickncr, U. Rescarcband Midw& Proceedings of the 1992 Conference, London and Glasgow, pp. 44-62.

Statham, H. (1992) Professional undemanding and parents’ apcriencc of termination. In: Brock, D.J., Rodcck. C.H. and Ferguson Smith, MA. (Eb) PrrnataL h n u ‘ n g and Diagnoziz, pp. 697-702. Churchill Livingstone, London.

Urquharc, D.R and Templeton. A.A. (1991) Psychiatric morbidity and acceptability following medial and surgical methods of induced abortion. Britichjmtnral of Obsktrics and Gynarrolgy, 98: 396-399

Accrptcd November i993.

Oxford.

JoumalofRrproddvrrurdZnfant P+obgy, 11: 175-185.

s&cr pregnancy loss. Journal of Rrproduciive and Znfmt psych oh^ 1 1: 75-87.

D a t h Society, London.

Received April 1993.

Dow

nloa

ded

by [

Jose

phin

e G

reen

] at

01:

40 0

2 N

ovem

ber

2015