Editorials
Mental health
Volume 172 - Issue 11

Evaluating policy and practice: what are the effects of early hospital discharge after childbirth?

Author:  Judith M Lumley

Med J Aust 2000; 172 (11): 524-525.
Published online: 5 June 2000
Share: Share on Blue Sky Share on Facebook Share on LinkedIn Share by email
Editorial

Evaluating policy and practice: what are the effects of early hospital discharge after childbirth?

Large, specific, randomised trials are the only way to answer this question

MJA 2000; 172: 524-525

Two years ago the Journal published an article which described early hospital discharge after birth as a major risk factor for postnatal depression.1 There was also a strongly worded editorial supporting that conclusion.2 In this issue the message is reversed, with the finding by Thompson and colleagues that there is no evidence of increased risk.3 This is no trivial disagreement. A quarter of a million women give birth in Australia each year, virtually all of them spending less time in hospital than their mothers did, with close to 40% having an "early discharge" by the definition used in both articles (within 72 hours of birth).4

Depression is a distressing and disabling condition for those directly affected, and particularly so in women with a new baby, who have just taken on a 24-hour-a-day, seven-day-a-week job. The effects of maternal depression may flow on to other vulnerable family members. Doctors and midwives, hospitals, policymakers and the general public need to know whether early discharge is safe or not.

One problem in answering this question is the paucity of evidence from randomised trials on length of hospital stay after childbirth. Not only are there relatively few trials, but those that have been published have rarely measured maternal health outcomes such as depression, breastfeeding duration, confidence, or breastfeeding problems. The limited evidence from trials shows either no difference in depression between women discharged early and late, or a lower proportion of women becoming depressed after early discharge.5,6

The article by Thompson and colleagues3 is the third Australian population-based study which has shown no relationship between early discharge and depression after birth.7-10 All three of these studies were large enough to detect a twofold increase in the odds of becoming depressed after early discharge, as was found in the Nepean hospital-based study published in the Journal two years ago.1 All three, and the Nepean study, used the Edinburgh Postnatal Depression Scale (EPDS) with the same cut point of a score of more than 12 for probable depression, though the Victorian studies measured the point prevalence with a single score of more than 12 at six or eight months after birth, the ACT group measured the period prevalence from eight through 16 and 24 weeks, and the Nepean group required two or more scores over 12, from six, through 12, 18 and 24 weeks, confirmed by a structured clinical interview to measure the period prevalence of major depression.

The difference in findings between the Nepean study and the other three studies is surprising. Thompson and colleagues suggest that an important contributing factor might be the routine provision of a postnatal visit from a midwife and more practical help at home in the ACT than at Nepean.3 However, this was not the case in the 1993/94 Victorian Survey of Recent Mothers, which found that only 66% of women who went home within 48 hours of birth, and only 27% of those going home on the third or fourth day, had a home visit from a midwife8 -- findings much closer to those of the Nepean study. It is also unlikely that temporal factors explain the discrepancy in findings, as the Victorian survey took place at a very similar time to the Nepean study. Public knowledge and expectations about postnatal stay would have been similar in both study populations.

In interpreting non-experimental descriptive and observational studies, the key problem is selection bias -- in what ways do women who leave hospital early differ from those who stay longer, and are these differences in themselves factors which have a bearing on women's chances of becoming depressed in the following months? Predictable factors associated with shorter postnatal stays include maternal age less than 25 years, multiparity, unassisted birth, birth at term, low medical risk, birth centre care, and not having private health insurance. Psychological predictors of depression were measured in both the ACT and Nepean studies; they were not significantly associated with length of stay.

The six vignettes in the Box describe some groups of women who will be over-represented (A, C, D, E) and under-represented (B, F) among new mothers going home early, despite having uncomplicated vaginal births at term. Young women are often over-represented in group D8 and women from diverse overseas backgrounds in group E.11 As all these women "chose" their length of stay, these vignettes draw attention to problems with the notion of "choice" as a key determinant. Some choices are constrained by factors outside the woman's control, especially social isolation and absence of a partner or other social support, which are common associations of depression at this life stage. We also know that women are not necessarily free to choose their length of stay, as almost a quarter of those discharged early in both the ACT study,3 and in Victoria,8,9 thought their stay had been too short.

The vignettes also remind us that the relative proportions of women from groups A, C and E in study populations could easily contribute to, or even explain, the differing associations between length of stay and subsequent depression reported from different studies. Given the inevitability of such selection biases, which are impossible to adjust for, if we really want to know whether shortening postnatal stay is safe and cost-effective, or whether domiciliary midwifery or other postnatal support improves outcomes for mothers and babies, there is no alternative but to test these policies in randomised trials.

Judith M Lumley
Director
Centre for the Study of Mothers' and Children's Health
La Trobe University, Melbourne, VIC

  1. Hickey AR, Boyce PM, Ellwood D, Morris-Yates AD. Early discharge and risk of postnatal depression. Med J Aust 1997; 167: 244-247.
  2. Buist A. Counting the costs of early discharge after childbirth. Med J Aust 1997; 167: 236-237.
  3. Thompson JF, Roberts CL, Currie MJ, Ellwood DA. Early discharge and postnatal depression: a prospective cohort study. Med J Aust 2000; 172: 532-536.
  4. Day P, Sullivan EA, Ford J, Lancaster P. Australia's Mothers and Babies 1997. (Perinatal Statistical Series No. 9). Sydney: AIHW National Perinatal Statistics Unit, 1997. (AIHW Cat. No. PER 12.)
  5. Waldenström U. Early and late discharge after hospital birth: fatigue and emotional reactions in the postpartum period. J Psychosom Obstet Gynaecol 1988; 8: 127-135.
  6. Carty EM, Bradley CF. A randomized, controlled evaluation of early postpartum hospital discharge. Birth 1990; 17: 199-204.
  7. Small R, Lumley J, Brown S. To stay or not to stay: are fears about shorter lengths of stay justified? Midwifery 1992; 8: 170-177.
  8. Brown S, Lumley J, Small R. Reasons to stay, reasons to go. Victorian women talk about early discharge. Melbourne: Centre for the Study of Mothers' and Children's Health, 1995: 27-54.
  9. Brown S, Lumley J. Reasons to stay, reasons to go: results of an Australian population-based survey. Birth 1997; 24: 148-158.
  10. Brown S, Lumley J, Small R. Early obstetric discharge: does it make a difference to health outcomes? Paediatr Perinat Epidemiol 1998; 12: 49-71.
  11. Yelland J, Small R, Lumley J, et al. Support, sensitivity, satisfaction: Filipino, Turkish and Vietnamese women's experiences of postnatal hospital stay. Midwifery 1998; 14: 144-154.


Make a comment




Examples of scenarios in which women select different lengths of stay
Group Description Length of stay Maternal health outcomes
A Healthy mother, healthy baby, good or excellent family support Chooses to go home at 24-48 hours to be with her family in a restful atmosphere and establish breastfeeding Likely to be good
B Healthy mother, healthy baby, good or excellent family support Chooses to stay in hospital for 5 or more days to establish breastfeeding, recover from the birth and have a good rest Likely to be good
C Mother with major responsibilities at home and little or no family support Goes home as soon as she can ?
D Mother who dislikes life in hospital Leaves as soon as she can, even despite strong advice to stay ?
E Mother who does not find the hospital able to provide her with enough support and is unable to get enough rest there Leaves much earlier than she had planned to Possibly poor
F Mother who has had a long, exhausting labour; baby who is difficult to feed; slow establishment of breastfeeding; mother loses self-confidence Stays much longer than she had planned to Possibly poor
Back to text

Author

  • Judith M Lumley

References

  1. Hickey AR, Boyce PM, Ellwood D, Morris-Yates AD. Early discharge and risk of postnatal depression. Med J Aust 1997; 167: 244-247.
  2. Buist A. Counting the costs of early discharge after childbirth. Med J Aust 1997; 167: 236-237.
  3. Thompson JF, Roberts CL, Currie MJ, Ellwood DA. Early discharge and postnatal depression: a prospective cohort study. Med J Aust 2000; 172: 532-536.
  4. Day P, Sullivan EA, Ford J, Lancaster P. Australia's Mothers and Babies 1997. (Perinatal Statistical Series No. 9). Sydney: AIHW National Perinatal Statistics Unit, 1997. (AIHW Cat. No. PER 12.)
  5. Waldenström U. Early and late discharge after hospital birth: fatigue and emotional reactions in the postpartum period. J Psychosom Obstet Gynaecol 1988; 8: 127-135.
  6. Carty EM, Bradley CF. A randomized, controlled evaluation of early postpartum hospital discharge. Birth 1990; 17: 199-204.
  7. Small R, Lumley J, Brown S. To stay or not to stay: are fears about shorter lengths of stay justified? Midwifery 1992; 8: 170-177.
  8. Brown S, Lumley J, Small R. Reasons to stay, reasons to go. Victorian women talk about early discharge. Melbourne: Centre for the Study of Mothers' and Children's Health, 1995: 27-54.
  9. Brown S, Lumley J. Reasons to stay, reasons to go: results of an Australian population-based survey. Birth 1997; 24: 148-158.
  10. Brown S, Lumley J, Small R. Early obstetric discharge: does it make a difference to health outcomes? Paediatr Perinat Epidemiol 1998; 12: 49-71.
  11. Yelland J, Small R, Lumley J, et al. Support, sensitivity, satisfaction: Filipino, Turkish and Vietnamese women's experiences of postnatal hospital stay. Midwifery 1998; 14: 144-154.