“Except When Medically Necessary” : Making informed choices about induction of labor

January 27th, 2011 by avatar

It’s not hard for women to find advice and recommendations to avoid induction of labor “except when medically necessary.”  But what do those words mean and who decides when an induction is medically necessary?

Lamaze’s Healthy Birth Practice Paper cites ACOG Guidelines that define medical induction of labor as necessary in the following circumstances:

  • your water has broken and labor has not begun.
  • your pregnancy is postterm (more than 42 weeks).
  • you have high blood pressure caused by your pregnancy.
  • you have health problems, such as diabetes, that could affect your baby.
  • you have an infection in the uterus.
  • your baby is growing too slowly.

Yet a systematic review of the highest quality research found evidence only to support the first three. Even in these three cases, differences in important health outcomes were small, study methodologies have been criticized, and some important questions remain unanswered.

For the rest of this list and other so-called “medical” reasons for induction, we simply lack scientific evidence that induction offers any clear health benefits, and for other conditions the available evidence suggests induction is more likely to harm than help.

When trade-offs are uncertain for a proposed course of treatment, that treatment is known as “preference-sensitive care.”  According to the Dartmouth Atlas of Health Care, a leader in studying practice variation, “Decisions about these interventions — whether to have them or not, and which ones to have — should thus reflect patients’ personal values and preferences, and should be made only after patients have enough information to make an informed choice, in partnership with the physician.”

As the Dartmouth Atlas has demonstrated for many types of medical and surgical interventions, however, decisions are more likely to reflect local practice patterns and the preferences of individual providers than the preferences of patients themselves. While the Dartmouth Atlas does not track induction rates, a 2004 study in New York State found that risk factors (at least those documented in birth certificate records) explained just 12.6% of the four-fold variation in induction rates across hospitals.

But what of the variation in the use of interventions when clear evidence suggests is harmful? Shouldn’t rates of those interventions be stable at or near 0%?  Take, for example, the rate of elective (non-medically indicated) deliveries before 39 weeks. In a landscape where clinical consensus is hard to come by, all of the major players from ACOG and ACNM to the March of Dimes, the National Quality Forum, and the Joint Commission, have gotten on the no elective deliveries before 39 weeks bandwagon. Yet this week a major hospital watchdog group, The Leapfrog Group, partnering with Childbirth Connection and the March of Dimes, released for the first time hospital rates of elective deliveries before 39 weeks and the results are all over the map (pun intended). Some hospitals are in the low single digits, rates we know are possible when quality improvement efforts are taken seriously, while others report up to half or more of all births between 37 and 39 weeks are electively delivered.

Women need individualized, evidence-based information about the likely benefits and harms when considering induction of labor in the face of complications or significant risk factors. Childbirth Connection has launched a new web resource dedicated to Induction of Labor to help fill this need. But evidence is just one piece of the puzzle. Women also need information about maternity care practice patterns in their communities, since this factor seems to affect their likelihood of induction more than any other. Leapfrog’s voluntary database of elective early delivery rates stands to drive significant quality improvement. Let’s hope it’s just the first step toward full transparency of maternity care quality.

Uncategorized , , , ,

  1. January 27th, 2011 at 12:41 | #1

    Amy- I wonder how Leapfrog chooses the hospitals included… In PA, there are few large maternity hospitals listed. In Pittsburgh there are none.

  2. January 27th, 2011 at 13:58 | #2

    @Vanessa – Leapfrog answers these questions on their site at http://www.leapfroggroup.org/tooearlydeliveries.

    “My hospital isn’t listed. Why is that?
    The Leapfrog Group works strategically with regionally-based employers and business health coalitions to ask hospitals in their respective regions to participate in the Leapfrog Hospital Survey (www.leapfroghospitalsurvey.org). Over 2,500 hospitals around the US are targeted annually, and many hospitals outside of those areas choose to respond regardless of not being asked. There are, however, some hospitals in the country that are not asked to respond to the Leapfrog Hospital Survey. Those hospitals which are not targeted are not listed on Leapfrog’s website.

    My hospital is listed as “Declined to respond.” Is there anything I can do about this?
    Leapfrog works with large employers and regionally-based business health coalitions across the country to ask hospitals to participate in the Leapfrog Hospital Survey. Those coalitions are often looking for support from individuals like you. You can find a list of those organizations and their contact information here: (http://www.leapfroggroup.org/for_members/members_resources/regional_roll_outs/1277465)

    Hospitals also respond to patient letters, and we encourage you to write a personal note. We recommend writing to the Hospital CEO or Director of Quality. If you don’t know who this is, or how to reach them, ask a hospital representative.”

  3. January 27th, 2011 at 15:10 | #3

    Thank you for another thoughtful piece!

    I would love to see Lamaze elaborate on “your water has broken but labor has not begun,” in terms of the # of hours that pass before induction is necessary/advised. I know that various studies show good outcomes with expectant management for 12, 24 or even up to 72 hours, but it would be great to have a clear recommendation from a group like Lamaze. In our community, care providers vary wildly on when they tell women to come in after SROM, and I’d like to have something more definitive to share with my students. How about it, wonderful Lamaze Intl. friends?

  4. January 27th, 2011 at 15:38 | #4

    I will be honest – there were numbers that surprised me here in MA & CT as well – both places I work. Spots I thought would be good were not that great (or even poor), spots I would have guessed needed improvement were already below the goal. And the most local place to me rocked it with a 1.1% rate.
    When I saw the numbers that surprised me I wondered how hospitals answered or viewed the phrase “not medically necessary.”

  5. January 27th, 2011 at 18:59 | #6

    Great post! “Except when medically necessary” should apply to many interventions, not just induction, of course.

  6. January 28th, 2011 at 14:09 | #7

    @Jessica English
    You might be interested in ACNM’s PROM at Term Position Statement (http://www.midwife.org/siteFiles/position/PROM_10_08.pdf) This was released after ACOG reversed their position on management of PROM. This blog covered that move by ACOG here: http://www.scienceandsensibility.org/?p=507

  7. January 28th, 2011 at 14:11 | #8

    @Karen Bayne
    Some of the rates may be surprising but these measures are rigorously tested and hospitals are given clear guidelines as to how to calculate and report them. The denominator is not all births but rather all births between 37-39 weeks excluding those with certain complications. If you’re interested, the guidelines for calculating the measure are here: http://manual.jointcommission.org/releases/TJC2010A/MIF0166.html

  8. February 1st, 2011 at 12:46 | #9

    This is a frustrating problem for obstetricians as well. We lack hard data to clearly indicate when we should deliver certain types of fetuses, diabetes, growth restriction, hypertension.

    A recent study showed that delivery at 39 weeks for non pre-eclamptic hypertension had similar neonatal outcomes as expectant management, and this practice did not seem to increase cesarean rates. Delivery earlier than this isn’t as well supported, yet is practiced by many docs.

    We really don’t know what to do with growth restricted fetuses with reassuring fetal heart rate tracings. Cord doppler studies appear to be important, but there isn’t a clear direction in the literature on them either. A study this month compared delivery vs expectant management for abnormal cord dopplers and growth restriction (together) and showed similar outcomes in either group, though one group had more babies die from prematurity and the other more from stillbirths.

    One of of the reasons that early delivery is favored is that a postnatal death from prematurity is somehow more palatable to all than an intrapartum stillbirth, and likely to plaintiff’s attorneys as well.

  9. avatar
    Michele Peterson
    February 2nd, 2011 at 11:57 | #10

    Amy, I have had a lot of clients induced for “Low Fluid” between 37 and 41 weeks. Can you provide resources that address this issue? I was surprised to not see it on the list when so many of my clients have been scared into an induction for that reason. Are others seeing this as well?

    ~Birth Doula

  10. avatar
    Christie B
    February 2nd, 2011 at 21:32 | #11

    Hi Michele – Birth Sense ran a post about inductions for oligo a bit ago that might be interesting to you: http://www.themidwifenextdoor.com/?p=1141

  11. February 3rd, 2011 at 12:28 | #12

    Michele, if you follow the link to the Childbirth Connection page, you’ll find some more info about low fluid (oligohydramnios):

    A single, small randomized controlled trial compared induction of labor with expectant management (watching fetal wellbeing closely until 42 weeks) for women with oligohydramnios at 41 weeks. The women were healthy and did not have other risk factors or complications. The study found no difference in maternal or newborn outcomes, but was too small to detect some important differences that may exist. No trials of induction for isolated oligohydramnios at other gestational ages were found.

    There are other studies (not randomized controlled trials, which are the gold standard) that suggest that inducing for low fluid may increase cesarean rates.

  1. January 27th, 2011 at 14:53 | #1
  2. January 31st, 2011 at 18:41 | #2