Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

Continuous Quality Improvement

Doing it properly
This is a thorny problem in healthcare. There's not really an awful lot laid down in terms of what to do and how to do it. There's not even much in the literature. A report of continuous quality improvement to reduce planned induction of labour from Vancouver [1] has interesting lessons, not only about how to do it, but also about how to think about it.


The study was carried out at Canada's busiest maternity hospital, with 7000 births a year and 125 nurses, 36 obstetricians and 124 family physicians. In the mid 1990s the induction rate was running at 23-25% of all births, and while there was little evidence about what was appropriate, this rate was thought to be too high. Initial attempts to draft guidelines and change a booking form foundered mainly due to a lack of clarity in guidelines and ownership of the systems: nurses were set to monitor doctors, and good working relations suffered.

Doing it properly

Doing it properly involved:


The number of inductions per 1000 deliveries fell after the new programme was introduced, for all indications (Table). The rate fell from 239 per 1000 deliveries (24%) to 183 per 1000 deliveries (18%). This is equivalent to the intervention preventing one induction for every 18 deliveries.

Table: Planned induction of labour per 1000 deliveries before and after instituting a continuous quality improvement programme

  Inductions per 1000 deliveries  
  15 months before implementation 9 months after implementation Inductions saved per 1000 women
Post term pregnancy 51 48 3
Premature rupture of membranes 53 51 2
Suspected foetal jeapordy 36 29 7
Pre-eclampsia 30 24 6
Maternal disease 20 14 6
Family circumstance/place of residence 16 1 15
Other 29 16 13
Not recorded 4 0 4
Total per 1000 deliveries 239 183 56


The Table deserves a lingering examination to see just where the fall in inductions occurred. The single largest category was one of logistics, essentially one of family convenience, and the second largest was of 'other', which included drug abuse with risk of elopement (sic). Over half the fall in inductions was those planned for non-medical reasons.

This provides a useful indication of how other units wanting to look at their induction rate might start. Go for the large numbers with simple answers. This group may be one such.

Overall, though, this sensitively written article demonstrates again the quality available in 'how to do it' work. The importance here is that it compares a successful and unsuccessful approach.

  1. S Harris et al. Induction of labour: a continuous quality improvement and peer review program to improve the quality of care. Canadian Medical Association Journal 2000 163: 1163-1166.
previous or next story in this issue