Whether Women Have Cesareans Is Mostly Arbitrary

 Regular contributor Henci Goer, author of several books including Optimal Care in Childbirth as well as the expert on Lamaze International’s “Ask Henci” site, takes a look at a recent study that examines the wide divergence in cesarean rates amongst U.S. hospitals.  Read Henci’s take and see what she concludes might be behind this rate variability. – Sharon Muza, Community Manager, Science & Sensibility



© Patti Ramos Photography

If any doubt remained that the likelihood of cesarean depends mostly on care provider philosophy and practices, a study of variation in cesarean rates in U.S. hospitals has laid it to rest. Investigators plotted cesarean surgery rates during 2009 by their percentile at 593 U.S. hospitals with at least 100 deliveries, comprising 817,318 women in all (Kozhimanni 2013). Rates ranged from 7% to 70%, a 10-fold variation.

Thinking that hospital factors might explain some of the variation, the investigators compared rates according to hospital size, whether the hospital was a teaching hospital, and whether it was rural. None had any effect. Average cesarean rates were similar to the overall average rate regardless of hospital characteristics.

Variation in population characteristics likewise could explain variation in cesarean rates. Accordingly, investigators looked at a more homogeneous low-risk subset of women who were at term (37 weeks or more), carrying one head-down baby, and who had no prior cesareans. This, they reasoned, should reduce the variation in rates. Wrong again. The range widened. Rates among low-risk women ranged from a little over 2% to nearly 37%, a 15-fold variation instead of a 10-fold one.

The investigators stopped with expressing concern over the large variation in cesarean rates, writing: “There is an urgent need to address maternity care quality in general and rising cesarean rates and variation in practice patterns in particular” (p. 531), but their data tell us something more: few hospitals had anything close to reasonable rates.

The mean cesarean rate among women overall was 33%. The World Health Organization holds that cesarean rates should not exceed 15% because research shows that as cesarean rates rise above this threshold, they necessarily are performed in less clear cut situations, and the risks of the surgery begin to outweigh its benefits. Beyond 15%, maternal and neonatal morbidity and mortality rise in parallel with further increase. Only 2 of the 593 hospitals had cesarean rates of 15% or less. Indeed, only 21 hospitals had rates of 20% or less.

In the low-risk population, the mean cesarean rate was 12%. The recent analysis of 18,084 women planning birth center births gives us a fix on whether this is a reasonable rate for low-risk women (Stapleton 2013). Of the 14,881 women admitted in labor to the 79 participating birth centers, 6% delivered by cesarean, and perinatal outcomes were equivalent to those in similar women planning hospital birth. Only 23 of the 593 hospitals had a cesarean rate of 6% or less in their low-risk cohort.

To be fair, the low-risk hospital dataset wasn’t able to identify women with problems that would increase their likelihood of cesarean but who would have been excluded from birth center care. The birth center data, however, provides a handle on the possible effect on cesarean rate. Six percent of women planning birth at the birth center were risked out because of pre-eclampsia, non-reassuring fetal testing, postdates, or prelabor rupture of membranes and no labor. Let us assume that these problems occurred at the same rate in the low-risk hospital population. Let us further assume that all women with these problems ended up with a cesarean, which is highly unlikely. Those assumptions would boost the birth center baseline cesarean rate of 6% by another 6% or to 12% for the low-risk hospital population. Even making this extreme assumption, 271 hospitals, nearly half, had rates greater than 12%.

What’s the take-home? Practitioners with appropriate cesarean rates are thin on the ground. Women need to seek out care providers whose judgment on when a cesarean is indicated can be trusted. (I should add that they are likely to have better luck with a midwife, but it isn’t a sure thing.) Women free of medical or obstetrical risk factors may wish to plan to birth in a free-standing birth center or at home because while individual practitioners’ rates may vary within institutions, a high hospital rate—true of nearly all of them—creates a cesarean–friendly culture.

How would you use this research study when teaching classes or working with clients or patients?  Do you think that women do enough research and investigation when selecting a provider and a birth facility? Please share your thoughts. – SM


Kozhimannil, K. B., Law, M. R., & Virnig, B. A. (2013). Cesarean Delivery Rates Vary Tenfold Among US Hospitals; Reducing Variation May Address Quality And Cost Issues. Health Aff (Millwood), 32(3), 527-535. doi: 10.1377/hlthaff.2012.1030 http://www.ncbi.nlm.nih.gov/pubmed/23459732

Stapleton, S. R., Osborne, C., & Illuzzi, J. (2013). Outcomes of care in birth centers: demonstration of a durable model. J Midwifery Womens Health, 58(1), 3-14. doi: 10.1111/jmwh.12003 http://www.ncbi.nlm.nih.gov/pubmed/23363029







Cesarean Birth, Guest Posts, informed Consent, Maternal Quality Improvement, Maternity Care, Medical Interventions, New Research, Push for Your Baby, Research , , , , , ,

  1. | #1

    Hi Henci – This study is fascinating. thanks, Kathy

  2. avatar
    Christine Morton
    | #2

    Thanks for sharing. It’s good to see what we found in California (www.cmqcc.org/white_paper) is also reflected in US data. Another study just out finds similar variation among hospitals in Massachusetts. (http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0057817)
    In some ways, the state level data is more powerful because the use of linked data sets (birth records and patient discharge data) allows for risk adjustment and to look at cesareans among a low risk population – first time mothers with a single, head down baby (NTSV Cesarean). These studies are laying the groundwork by showing how much room there is for improvement. As of Jan 1, 2014, The Joint Commission will require all hospitals with >1100 births/year to report rates of NTSV Cesarean, which may help hospitals start to work with their physicians and nurses on identifying ways to reduce these medically unnecessary surgeries.

  3. avatar
    Chad Thomas
    | #3

    Kozhimannil’s study highlights some really important variation. Speaking as a practicing OB, this variation does exist in the real world, being driven by personality, risk-tolerance, and patient population. We may see a decrease in these numbers and variation with the increasing transparency of data reporting cesarean rate on a per-provider basis. Another factor likely driving variation is variation in obesity and induction of labor, both factors that increase cesarean rate.

    I try hard to avoid cesareans at all cost, but even so, my rate of cesareans in “low-risk” patients approaches 14%. This may be driven by the fact that nulliparous, term, singleton, vertex patients are not necessarily low-risk. In fact, many of them are first children in patients with other medical co-morbidities.

    The other issue not raised by the article is that of cesarean on demand. Depending on the community, this can significantly change the rate of cesarean sections.

  4. avatar
    Theresa Morris
    | #4

    This is such an important topic, Henci! Thanks for drawing attention to this epidemic! In my book, Cut It Out: The C-Section Epidemic in America (being published in the fall by NYU Press), I make a similar argument, drawing upon over 100 interviews I conducted with physicians, midwives, nurses, and post-partum women. There are by far too many c-sections being done in the U.S., and this trend negatively affects the health of women and babies and also women’s future reproductive health and choices.

  5. | #5

    Chad, in response to cesarean by maternal request, along with ACOG’s statement released today about encouraging a vaginal birth over a cesarean in the case of maternal request for cesarean delivery, they stated that the rate of maternal request was 2.5% (request without medical indication.) The Listening to Mothers survey of several years ago interviewed 1600 women I believe, and one woman fell into that category. So it appears that the percentage of women wanting a cesarean without indication is rather low. Can you share your experience? Thanks.

  6. | #6

    I prefered a C /S with my second baby it was much less stress. this is the 21st Century not the 16th…..

  7. | #7

    While cesarean surgery is relatively safe as surgeries go, as with any surgery, there are potential harms as well as possible benefits, including severe or life-threatening harms. (http://childbirthconnection.org/article.asp?ck=10166) Undergoing cesarean when there is no indication or when the problem could have been solved by lesser means necessarily exposes a woman to those potential harms with no counterbalancing benefit.

  8. avatar
    Ahmie (@DragonMama)
    | #8

    Dr. Thomas, I beg to differ on how much risk is necessarily inherent in obese women. Anthropologically, it makes little sense. A well-fed woman in a setting our species has found ourselves in until very recently, would have less risk. If obesity is a result of overeating (instead of other environmental factors, as many now suspect), then obesity itself does not make much logical sense as an increase in perinatal risk – a well nourished woman and her baby should, theoretically, be better able to withstand the rigors of birth than an undernourished one. This bothers me greatly, whenever I see authority figures blame the c-section rate on “obesity”. The need for the c-section may, indeed, be higher among women who are obese, but this is a correlation that is too frequently marketed as causation in my observations. My hypothesis is that maternal diet has a much greater effect on the likely need for c-section than straight pre-pregnancy BMI. A skinny pregnant woman who subsists on fast food but happens to have a high metabolism would, in this model, be at more risk for a c-section than a woman who eats a balanced, nutrient-dense diet but has trouble maintaining a lower BMI for other reasons. But instead, providers treat the thin woman as lower risk than the obese one, without exploring the actual causes of the risk increase. This is a failure on the part of practitioners.

    For the record, I am obese by BMI standards and have been at least borderline obese throughout my entire childbearing years thus far. It’s not due to over-eating, it’s due to physical disability. I was not overweight until I was in an accident that damaged my knees when I was 19, and several other injuries that have restricted my physical mobility. I have had four entirely uncomplicated vaginal births, without medical interventions (I refused even an IV, listening to fetal heart tones and cervical checks were the only interventions performed). My eldest is the only one not born at home, and all of them have been above the median newborn weight but not macrosomatic (range: 7lbs 14oz to 9lbs even). While I feel very safe birthing at home with the attendants that I have (and being a few blocks from a hospital), this is not necessarily the preferred option for all women.

    Women should not be classified as “higher risk” by default because of their pre-pregnancy BMI, without accounting for other factors, and thin women should not be automatically classified as “low risk”. There needs to be MUCH more individualized and evidence-based care going on, particularly for the women who do not feel that they would like to birth at home and who do not have access to a freestanding birth center. There are other ways to manage birth, and risks, without interfering to the level that is the current default setting.

  9. | #9

    Pam Vireday did an excellent 3-part series for Science & Sensibility on the facts vs. beliefs surrounding high BMI women and pregnancy and childbirth: http://www.scienceandsensibility.org/?p=3030, http://www.scienceandsensibility.org/?p=3064, and http://www.scienceandsensibility.org/?p=3094.

  10. | #10

    The 2.5% number quoted in the ACOG release yesterday stood out to me, too. I thought that the Listening to Mothers Survey was the only study to specifically address “maternal request” surgery (and there was only one of those, for a 0.4% rate), as Sharon said; any other numbers out there on this are derived from questionable birth certificate data that does not identify maternal request, but rather assumes it as a default category–giving us a possibly inflated number. Can anyone shed any light on this? It may seem like a small thing (0.4% vs. 2.5%), but it seems like we’re hearing more and more from doctors that women are partly to blame for the CS rate. I’m interested to know what the evidence is for that claim. Chad, if your statement about the variation among communities is based on personal experience, I’d like to hear a little more, too! Thank you!

  11. | #11

    I know that Childbirth Connection has carried out Listening to Mothers III, and the report should be out soon. Let us hope that the new survey addressed elective cesarean. I should add, too, that studies have found that the women themselves often perceive medical or psychological indications for their so-called elective cesareans. Furthermore, a survey at one hospital found that 13% of cesareans during labor were performed according to their own report at “obstetrician request,” that is, without medical indication. Another 3% were reported as a joint decision with the woman, which given the power inequity in the relationship, I would add to the “obstetrician request” column, making 16% of intrapartum cesareans at this hospital elective at the doctor’s request. If that is going on in labor, you can only imagine what must be going on during prenatal consultations.

    McCourt, C., Weaver, J., Statham, H., Beake, S., Gamble, J., & Creedy, D. K. (2007). Elective cesarean section and decision making: a critical review of the literature. Birth, 34(1), 65-79.

    Gamble, J. A., & Creedy, D. K. (2000). Women’s request for a cesarean section: a critique of the literature. Birth, 27(4), 256-263.

    Kalish, R. B., McCullough, L., Gupta, M., Thaler, H. T., & Chervenak, F. A. (2004). Intrapartum elective cesarean delivery: a previously unrecognized clinical entity. Obstet Gynecol, 103(6), 1137-1141.

  12. avatar
    Chad Thomas
    | #12

    @Sharon and Cristen: ACOG’s reported rate of 2.5% for cesarean on maternal request is a national average. In the area of the Pacific Northwest where I practice, it’s lower. I’d say I’m seeing something around 0.5-1% in my practice. Many of the women requesting cesarean have non-medical reasons: logistical and psychological are the ones I see most frequently. Usually, with a little time spent clarifying their concerns, and reviewing the risks/recovery times, they end up opting for a vaginal delivery. I don’t have experience outside the Northwest (where we have a high-value on things being “natural”), but reported rates for other areas are higher, up to 70% in Brazil.

    @Ahmie – I’m sorry you interpreted my statement that obesity was a risk factor for higher rates of C-section as causative. I agree with you, it’s a correlation, probably related to overall higher average blood sugars and possibly to less space in the pelvis.

  13. | #13

    I think the real message is that women need to be their own advocates when it comes to the birth of their babies. Only through self-education will you have the knowledge to challenge decisions doctors are making for you. This is especially true where c-sections are concerned. C-sections can be a very necessary surgery, but many times they are not necessary. Unless you educate yourself and know what options there are and if an option can be exercised, you could end up having a Cesarean that is unnecessary.

  14. | #14

    A woman at the end of pregnancy or in labor isn’t in a good position to oppose a care provider trying to persuade her into a cesarean especially when the care provider, as the Aussie’s put it, starts “waving the shroud” if she resists. That is why, as I wrote in the take-home for my blog post, that I think the main reason to be well-informed about the issues is so the woman can choose a care provider whose judgment she can trust. Still, we don’t really disagree because in many cases, women don’t control who attends them in labor.

  1. | #1
  2. | #2

cheap oakleys fake oakleys cheap jerseys cheap nfl jerseys wholesale jerseys wholesale nfl jerseys