Rapid Responses to:
|
|
Rapid Responses: Rapid Responses published:
-
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif)
Depression after stillbirth?
- Nosib Priyanjali Hema Devi (22 July 1999)
-
Authors response to Devi et al
- PM Hughes, C D H Evans (23 July 1999)
-
Observations may simply reflect normal process of greiving
- Malcolm Griffiths (10 August 1999)
|
|
|||
|
Nosib Priyanjali Hema Devi, 3rd Year Medical Student Newcastle University Medical School
Send response to journal:
|
We read with interest the study by Hughes et al. . They reported an increase in depression and anxiety in the third trimester, a trend towards increased depression in one measure at one year post partum and increased anxiety/depression in women conceiving within 12 months of stillbirth. We have some concerns, however, over the validity and interpretation of their findings. Firstly, the finding of increased anxiety and depression following stillbirth is not really surprising. The increased depression at one year was only a non significant trend in one dichotomous variable and there was no difference in the mean anxiety/depression scores. Also many possible confounders were not measured and compared. .For example, social support, the closeness and support given by partners, and the proportion of planned versus unplanned pregnancies may have differed between cases and controls,. These could have created differences in vulnerability to depression and anxiety. The authors also subdivided the cohort into 2 groups: those who conceived within 12 months of having a stillbirth and those who waited longer. They then concluded that women conceiving within 12 months were at greater risk of depression.. However, the authors made no comparison in the distribution of confounders between cases and controls in the two groups. For example, those conceiving within 12 months after still birth may have had poorer material circumstances, less social support and lower educational levels. The authors hypothesised that mothers need a time to mourn following stillbirth and that without this they are more prone to depression. If this were the case we would expect women conceiving within nine months to be more depressed than those conceiving later. No such dose response was observed. Given these limitations we feel that the findings from this study have few implications for clinical practice. More rigorous studies (including in-depth qualitative studies) are needed to clarify the issues raised by the authors. Matthew Cheesman Suzanne Gill Priyanjali Hema Devi Nosib Sam Molyneux Mark Tones 3rd Year Medical Students University of Newcastle upon Tyne |
|||
|
|
|||
|
PM Hughes, senior lecturer in psychiatry st George's Hospital medical School, C D H Evans
Send response to journal:
|
We appreciate the interest that Devi and colleagues have taken in our results and agree with their conclusion that more studies are desirable but disagree with their dismissal of the paper as having "few implications for clinical practice". They note that, taking the group as a whole, the difference in depression between stillbirth and control groups was indeed only a trend. We make the point that one important clinical conclusion from this paper, the first real "evidence base" in this area, is that many, in fact most, mothers cope and recover well in terms of self-reported anxiety and depression, following the trauma of a stillbirth. That brings out the thrust of the evidence of the data concerning the possible relevance of time since loss. Devi and colleagues suggest that confounding variables may explain the significant findings. We did in fact measure the quality of marital relationship and found no between group differences. We did not measure whether pregnancy was planned or unplanned, which we agree would have been interesting. The evidence is that this is a weak contributor to perinatal depression (Martin CT, Brown GW, Goldberg DP, Brockington IF. Psychosocial stress and puerperal depression. J of Affective Disorders 1989; 16:283-293. Zuckerman B, Amaro A, Bauchner H, Cabral H.). Depressive symptoms during pregnancy: Relationship to poor health behaviours. Am J of Obstetrics and Gynaecology 1989;160:1107-1111) In the analysis reported in the table we measured the difference between pairs of mothers, each subject against her own control. As subjects and controls were carefully matched for age, material circumstances and educational levels this should give us a true picture of the between group differences. As can be seen from the table, the slow to conceive group was very similar to their controls, while the more rapid conceivers showed differences from their controls. These differences did reach significance in the third trimester and at one year post partum. As we mentioned in the paper at the end of the results section we did look at a 9 month, 15 month and 18 month divide. The fact that there was no signficant difference if the cohort was dichotomised at 9 months does not invalidate the finding of a considerable difference splitting at 12 months. Apart from the simple issue of not interpreting a non- significant finding as evidence of no difference (i.e. correct understanding of the meaning of a non-significant difference), there are two other substantive reasons that may contribute to this finding. The first is the simple substantive issue, in line with previous clinical thinking, that a year may be a significant time to mourn and assimilate loss and plan for a new child. The other is more prosaic and methodological: it happened that the 12 month split almost perfectly bisected the sample. Splits either side of this have less balanced sample sizes and hence their power to detect a difference as significant falls making misinterpretation of a non-significant finding even more of a problem. We tried to be cautious in our conclusions: there is a higher level of depression and anxiety in the more rapid conceivers, but we have carefully pointed out that those more vulnerable to depression may be those who chose to conceive more rapidly. However, we would expect to see more of a history of depression and there was no evidence of this in the rapid conceivers. Overall, only 2 of all 60 women had had treatment for depression previously, which does not seem to indicate a particular tendency to depression. We accept that this paper is not perfect, and that a RCT would be the theoretical ideal. However, in the real world it is unlikely that people would agree to such an intrusion into their private lives, so we must do what we can. As regards the need to a qualitative study we entirely agree. The reported study is only a small part of the data we collected; we also looked at infant development in terms of security of attachment and cognitive development, at maternal representation of attachment and resolution of mourning, at the quality of the marital relationship before the birth and a year later, at maternal PTSD, and the relationship between traditional 'good practice' after stillbirth and outcome in terms of the mothers' mental health and the infants' development. We are currently writing up the findings. It is also the case that following up 120 people, many from low socio -economic groups with no phone and a relaxed attitude to keeping appointments, was extremely labour intensive and cost a surprising amount of money. A bigger study would indeed be attractive but we doubt if we could get the funding for it. There has been a lot of opinion expressed about the 'right' time to have the next baby after stillbirth, but this study is the first one to systematically evaluate both the kinds of problems which may arise around the birth of the next child and the possibility that delaying pregnancy for a year may reduce some problems for the mothers. We think that our findings deserve to be taken seriously, and hope that there will be further research in the area. |
|||
|
|
|||
|
Malcolm Griffiths, Consultant Obstetrician & Gynaecologist Luton & Dunstable Hospital
Send response to journal:
|
I was pleased that the paper by Hughes and colleagues (26 June 1999)1 was given such prominence, as parental outcomes following stillbirth are somewhat under-researched. Much of the advice given to parents after such losses is either empirical (and probably incorrect) or based on out-dated or poorly conducted studies. The authors claimed to have found an association between short interval (less than 12 months) between stillbirth and next conception and increased vulnerability to depression and anxiety. They cautiously accept that personalities of parents who chose to conceive earlier might have been such that they were anyway more likely to be depressed and anxious. I wish though to point out a potentially fundamental flaw which may invalidate either conclusion and accordingly to allow other researchers in this important area to avoid a similar pitfall. The subjects of the study were recruited into the study during pregnancy, stillbirth- conception interval was calculated, psychological assessments were performed in the third trimester and postnatally. Parents are likely to have been more depressed and more anxious sooner after stillbirth than after a lengthy period of bereavement and adjustment. The methodology of the study fails to take account of the natural course of depression and anxiety following stillbirth. It might well be that the researchers have merely recorded the natural course of bereavement (unaffected by a subsequent pregnancy). Couples (say) a year after a stillbirth are more likely to be depressed and anxious than after (say) two years - whether a pregnancy has intervened or not. Further the researchers justify their concerns about an increased rate of depression/anxiety as depression may be associated with poor pregnancy outcome - various studies are cited in support of this. In particular the authors draw attention to research linking depression with poor compliance with antenatal care - my experience is that women who have experienced stillbirth are highly compliant with antenatal care. I am unaware of any studies that demonstrate poor obstetric outcomes associated with depression following stillbirth. Accordingly I do not believe that the paper justifies obstetricians and others now advising bereaved couples to particularly delay the next pregnancy. I will continue to advise couples that they should "wait until they feel ready and have adjusted to their loss". References 1 Hughes, PM, Turton, P, Evans, CDH Stillbirth as risk factor for depression and anxiety in subsequent pregnancy: cohort study. BMJ 318: 1721-1724 |
|||