General obstetrics: Does size matter? A population-based study of birth in lower volume maternity hospitals for low risk women

Authors

  • Sally K Tracy,

    Corresponding author
    1. Australian Institute of Health and Welfare (AIHW), National Perinatal Statistics Unit, University of New South Wales, Sydney, Australia
      Dr S K Tracy, Australian Institute of Health and Welfare, National Perinatal Statistics Unit, University of New South Wales, Level 2, McNevin Dickson Building, Randwick, NSW 2031, Australia.
    Search for more papers by this author
  • Elizabeth Sullivan,

    1. Australian Institute of Health and Welfare (AIHW), National Perinatal Statistics Unit, University of New South Wales, Sydney, Australia
    Search for more papers by this author
  • Hannah Dahlen,

    1. Royal Prince Alfred and Liverpool Hospitals, Sydney, NSW, Australia
    Search for more papers by this author
  • Deborah Black,

    1. School of Public Health and Community Medicine, University of New South Wales, Sydney, Australia
    Search for more papers by this author
  • Yueping Alex Wang,

    1. Australian Institute of Health and Welfare (AIHW), National Perinatal Statistics Unit, University of New South Wales, Sydney, Australia
    Search for more papers by this author
  • Mark B Tracy

    1. Neonatal Intensive Care Unit, Nepean Hospital, Sydney, Australia
    Search for more papers by this author

Dr S K Tracy, Australian Institute of Health and Welfare, National Perinatal Statistics Unit, University of New South Wales, Level 2, McNevin Dickson Building, Randwick, NSW 2031, Australia.

Abstract

Objective  To study the association between volume of hospital births per annum and birth outcome for low risk women.

Design  Population-based study using the National Perinatal Data Collection (NPDC).

Setting  Australia.

Participants  Of 750,491 women who gave birth during 1999–2001, there were 331,147 (47.14%) medically ‘low risk’ including 132,696 (40.07%) primiparae and 198,451 (59.93%) multiparae.

Methods  The frequency of each birth and infant outcome was described according to the size of the hospital where birth took place. We investigated whether unit size (defined by volume) was an independent risk factor for each outcome factor using public hospitals with greater than 2000 births per annum as a reference point.

Main outcome measures  Rates of intervention at birth and neonatal mortality for low risk women in relation to hospitals with <100, 100–500, 501–1000, 1001–2000 and >2001 births per annum.

Results  Neonatal death was less likely in hospitals with less than 2000 births per annum regardless of parity. For multiparous low risk women in hospitals of 100 and 500 births per annum compared with hospitals of >2000 births per annum the adjusted odds of neonatal mortality [adjusted odds ratio (AOR) 0.36; 99% confidence interval (CI) 0.14–0.93]. For low risk primiparous women in hospitals with less than 100 births per annum, there were lower rates of induction of labour (AOR 0.62; 99% CI 0.54–0.73); intrathecal analgesia/anaesthesia (AOR 0.34; 99% CI 0.28–0.42); instrumental birth (AOR 0.80; 99% CI 0.69–0.93); caesarean section after labour (AOR 0.59; 99% CI 0.49–0.72) and admission to a neonatal unit (AOR 0.15; 99% CI 0.10–0.22) and for low risk multiparous women in hospitals with less than 100 births per annum: induction (AOR 0.69; 99% CI 0.62–0.76); intrathecal analgesia/anaesthesia (AOR 0.32; 99% CI 0.29–0.36); instrumental birth (AOR 0.52; 99% CI 0.41–0.67); caesarean section after labour (AOR 0.41; 99% CI 0.33–0.52); and admission to a neonatal unit (AOR 0.09; 99% CI 0.07–0.12).

Conclusions  In Australia, lower hospital volume is not associated with adverse outcomes for low risk women.

Ancillary