I think so. Correlations tend to show worse outcomes for c-sections, but the guidelines take this evidence too literally. Careful causal evidence finds that it’s more of a mixed bag. Moderate negative effects on subsequent births seem more robust. An elective c-section might be the defensible choice for someone’s final pregnancy.
Disclaimer: Not a doctor.
The official guidelines
C-sections cost about twice as much as vaginal delivery and are associated with worse health outcomes for infants and moms. These facts are concerning enough to the US Medicaid system that one of their “Improvement Initiatives” is reducing low-risk cesarean delivery. They write: “Cesarean delivery poses a greater risk of maternal morbidity and mortality for low-risk pregnancies when compared to vaginal births, a risk that ideally should be avoided.”
They lead seminars with slides like this, highlighting that everything bad is correlated with c-sections:
Hospitals have been falling in line. UCSF boasts: “Our view of labor and childbirth as a natural process has helped keep our overall Cesarean rate at 20 percent, among the lowest rates in California…Our threshold for making the decision to recommend a C-section is a lot higher than in other places.”
The American College of Obstetricians and Gynecologists (ACOG) has similar advice, and to their credit they spell out the evidence and their reasoning. The main paper they cite is this Canadian study of 2.3m vaginal and 46k c-section deliveries. “A large population-based study from Canada found that the risk of severe maternal morbidities…was increased threefold for cesarean delivery as compared with vaginal delivery.”
But what really should we make of the association between c-sections and bad outcomes? The observed correlation is ripe for reverse causality, for the same reason that going to the hospital is correlated with dying: c-sections are often performed because of some dangerous condition (e.g., preeclampsia). In the Canadian study, the treatment group was mothers who had scheduled a c-section due to breech position, and breech is correlated with other negative outcomes. In other studies, the c-sections could happen for reasons that are never recorded in hospital data, so even lots of control variables should make you worried about selection effects.
The causal effect of a c-section
A new paper by Card, Fenizia, and Silver, using data from California births, takes the question of causality seriously. It appears to be the only careful attempt at separating correlation and causation.
C-sections are not randomly assigned. How can we use observational data to arrive at a causal estimate? Their approach combines the facts that hospitals exhibit stable differences in their (risk-adjusted) c-section rates and that moms often give birth at the hospital closest to them.
Together, these mean that some moms will have c-sections by virtue of living close to a hospital that performs more of them. In essence, a causal estimate can be derived by comparing the outcomes of mothers who live near vs. far from high c-section hospitals.
You might be concerned that certain kinds of moms live next to certain kinds of hospitals. But in detailed tests, the authors find no signs of problematic selection along this distance-to-hospital dimension (although with these kinds of designs it’s reasonable to worry that something not measured could still bias the estimates).
And they show that the distance measure does have a large effect on c-sections for the group they study: low-risk first births. (Why restrict to this sample? High-risk births are often defaulted to c-sections, and a c-section in a first birth means you should probably only get c-sections in subsequent births—so this is the group that is most relevant to the policy discussion.)
With this distance-based variation in hand, the authors estimate causal effects on a range of infant and maternal outcomes. Contrary to the correlational evidence, the estimates are mixed.
C-section babies have:
Reduced probability of a low (<7) 5-minute Apgar score (it’s a massive reduction, about 100%)
Lower probability of readmission to the hospital in the first 28 days, but higher probability of an emergency department visit in the first year after birth. (These emergency room visits are mostly due to respiratory issues.)
Lower mortality, although this result is insignificant at conventional levels (p=0.10)
Mothers who get c-sections have:
Lower risk of perineal laceration and other traumas
No significant difference in the post-birth stay (95% confidence interval: -.03 days to +.18 days)
An estimated ~15% higher risk of any emergency room visit (off a base of 12%), but this result is not significant at conventional levels
Importantly, these results only apply to marginal cases: mothers whose delivery method was determined by the practices at the hospital. But this is arguably the perfect subgroup to study for the current debate about c-section rates. Remember the “threshold” that the UCSF doctor mentioned. Should it be higher or lower?
Also, these outcomes are limited; there’s lots more that we’d want to know. The study is under-powered to look at other consequences for the mothers (including mortality, although see notes at bottom) and does not have the data to look at long-run consequences for either mothers or children.
But this should all be very surprising and possibly concerning to the people putting those presentations together for Medicaid. Instead of c-sections being clearly bad for low-risk births, the conclusion here is that it is far from obvious—you trade some problems for other problems.
The favorable result in the Apgar score, which measures how lively the baby is just after birth, seems consistent with the mechanics of the two delivery methods. With vaginal birth, the baby can get stuck. This kind of problem is less likely with c-sections. This is also an imminently critical outcome: some babies with low Apgar scores will have permanent brain damage from hypoxic ischemic encephalopathy.
The respiratory issues for c-section babies also line up with a recent assessment which singles out “delayed clearance of lung liquid” as a culprit, although it seems more mysterious.
The outcomes for mothers are a little hard to piece together. The outcome I put most weight in is length of stay, as this should capture most of the very bad things depicted in the Medicaid slide above. The insignificant but positive estimates on emergency room visits, even taken at face value, are much smaller than the estimates in the Medicaid slide, which all suggest a doubling of risk or more.
Finally, for what it’s worth, the flat results for mothers and infants line up with a randomized trial of twin pregnancies: “planned cesarean delivery did not significantly decrease or increase the risk of fetal or neonatal death or serious neonatal morbidity, as compared with planned vaginal delivery,” and “We did not find that planned cesarean delivery was associated with a higher or lower risk of maternal death or serious maternal morbidity than planned vaginal delivery.”
Subsequent pregnancies
The ACOG guidelines are actually more worried about effects on subsequent pregnancies: “the downstream effects are even greater because of the risks from repeat cesareans in future pregnancies.” The main theory, from what I can tell, is that the c-section scar interferes with the formation of the placenta, resulting in placental issues and preeclampsia. These are dangerous conditions that greatly increase the chance of harm for both mother and baby. There could also be subtler issues getting nutrients to the baby that don’t register in checks of maternal health.
Their reference for this is Silver et al. I’m a little stunned at how simplistic the analysis is. They only study c-section births and look at the correlation between negative outcomes and the number of previous births. The problem is that all of this could just be a birth order effect, not a number-of-previous-c-sections effect.Second, while they find increased risk of placental malformation in subsequent pregnancies, the initial c-section might have happened because of this very issue. Surely moms who have a given issue are more likely to have it again, regardless of method of delivery.
More convincing evidence comes from Daltveit et al 2008, using Norwegian register data. They compare outcomes for the second pregnancy for women whose first birth was vaginal vs. Cesarean:
The All women comparisons don’t impose any restrictions. In the Selected women section, they leave out women who had the same condition in their first pregnancy. This addresses the important source of bias that Silver et al ignore. Still, most outcomes show a clear increase.
There are increases in preeclampsia and SGA (small for gestational age). These could be connected with malformation of the placenta. There are effects on placenta outcomes and uterine rupture, although these represent small absolute risks. The risk of placenta previa increases from 0.22% to 0.33%. The risk of uterine rupture increases from 0.006% to 0.2%.
The uterine rupture result is scary but this appears to be driven by women who attempt vaginal delivery after a c-section—something the authors don’t account for. Another source that compares people based on their planned method finds strong evidence of this: “The estimated incidence of uterine rupture was 2 per 10,000 maternities overall; 21 and 3 per 10,000 maternities in women with a previous caesarean delivery planning vaginal or elective caesarean delivery respectively.” So risks were practically the same for the caesarean-to-caesarean mothers.
One lingering confound is that, even among the Selected women, “confounding by indication” may still cause the previous c-section mothers to have worse prospects, biasing the comparison. A way around this is to look at outcomes for elective c-sections—mothers who are not getting the c-section because of a health need. This seems to show the same result. For example, in this case control study, mothers with placenta accreta were more likely to have had a previous elective c-section.
Conclusion
Where does that leave us? The Card et al paper should make us more agnostic about the costs of a c-section for a one-off, low-risk birth. But, while I could not find careful causal studies on subsequent births, the observational studies do suggest that they may suffer negative effects of a previous c-section. Based on this investigation, I now believe that choosing an elective c-section for a one-off kid, or your last pregnancy, could be the more defensible choice.
In discussing these ideas with doctors, I get a common response that “A c-section is major abdominal surgery!” From my perspective, given the tragic injuries that can befall mother and infant in the process of childbirth, Mother Nature hasn’t done herself any favors. Maybe we can, in fact, do better.
Notes
I am still looking for better cites on this, but it appears that maternal mortality for elective c-sections, is lower if anything in observational data. (Although the selection bias here could favor scheduled c-sections.)
As always, Emily Oster is the best place to start with questions like this. She believes the effects on placental malformation in subsequent pregnancies. See her thoughts here. In particular: “The good news is, if we look at one or two years after birth, there is no evidence of differences in recovery. If anything, the data might slightly favor C-sections.” In a big meta-analysis she cites, “The only two complications that differed across groups were urinary incontinence and pelvic organ prolapse. Both are actually more common for women who have had a vaginal delivery.”
Are c-sections underrated?
Summary
I think so. Correlations tend to show worse outcomes for c-sections, but the guidelines take this evidence too literally. Careful causal evidence finds that it’s more of a mixed bag. Moderate negative effects on subsequent births seem more robust. An elective c-section might be the defensible choice for someone’s final pregnancy.
Disclaimer: Not a doctor.
The official guidelines
C-sections cost about twice as much as vaginal delivery and are associated with worse health outcomes for infants and moms. These facts are concerning enough to the US Medicaid system that one of their “Improvement Initiatives” is reducing low-risk cesarean delivery. They write: “Cesarean delivery poses a greater risk of maternal morbidity and mortality for low-risk pregnancies when compared to vaginal births, a risk that ideally should be avoided.”
They lead seminars with slides like this, highlighting that everything bad is correlated with c-sections:
Hospitals have been falling in line. UCSF boasts: “Our view of labor and childbirth as a natural process has helped keep our overall Cesarean rate at 20 percent, among the lowest rates in California…Our threshold for making the decision to recommend a C-section is a lot higher than in other places.”
The American College of Obstetricians and Gynecologists (ACOG) has similar advice, and to their credit they spell out the evidence and their reasoning. The main paper they cite is this Canadian study of 2.3m vaginal and 46k c-section deliveries. “A large population-based study from Canada found that the risk of severe maternal morbidities…was increased threefold for cesarean delivery as compared with vaginal delivery.”
But what really should we make of the association between c-sections and bad outcomes? The observed correlation is ripe for reverse causality, for the same reason that going to the hospital is correlated with dying: c-sections are often performed because of some dangerous condition (e.g., preeclampsia). In the Canadian study, the treatment group was mothers who had scheduled a c-section due to breech position, and breech is correlated with other negative outcomes. In other studies, the c-sections could happen for reasons that are never recorded in hospital data, so even lots of control variables should make you worried about selection effects.
The causal effect of a c-section
A new paper by Card, Fenizia, and Silver, using data from California births, takes the question of causality seriously. It appears to be the only careful attempt at separating correlation and causation.
C-sections are not randomly assigned. How can we use observational data to arrive at a causal estimate? Their approach combines the facts that hospitals exhibit stable differences in their (risk-adjusted) c-section rates and that moms often give birth at the hospital closest to them.
Together, these mean that some moms will have c-sections by virtue of living close to a hospital that performs more of them. In essence, a causal estimate can be derived by comparing the outcomes of mothers who live near vs. far from high c-section hospitals.
You might be concerned that certain kinds of moms live next to certain kinds of hospitals. But in detailed tests, the authors find no signs of problematic selection along this distance-to-hospital dimension (although with these kinds of designs it’s reasonable to worry that something not measured could still bias the estimates).
And they show that the distance measure does have a large effect on c-sections for the group they study: low-risk first births. (Why restrict to this sample? High-risk births are often defaulted to c-sections, and a c-section in a first birth means you should probably only get c-sections in subsequent births—so this is the group that is most relevant to the policy discussion.)
With this distance-based variation in hand, the authors estimate causal effects on a range of infant and maternal outcomes. Contrary to the correlational evidence, the estimates are mixed.
C-section babies have:
Reduced probability of a low (<7) 5-minute Apgar score (it’s a massive reduction, about 100%)
Lower probability of readmission to the hospital in the first 28 days, but higher probability of an emergency department visit in the first year after birth. (These emergency room visits are mostly due to respiratory issues.)
Lower mortality, although this result is insignificant at conventional levels (p=0.10)
Mothers who get c-sections have:
Lower risk of perineal laceration and other traumas
No significant difference in the post-birth stay (95% confidence interval: -.03 days to +.18 days)
An estimated ~15% higher risk of any emergency room visit (off a base of 12%), but this result is not significant at conventional levels
Importantly, these results only apply to marginal cases: mothers whose delivery method was determined by the practices at the hospital. But this is arguably the perfect subgroup to study for the current debate about c-section rates. Remember the “threshold” that the UCSF doctor mentioned. Should it be higher or lower?
Also, these outcomes are limited; there’s lots more that we’d want to know. The study is under-powered to look at other consequences for the mothers (including mortality, although see notes at bottom) and does not have the data to look at long-run consequences for either mothers or children.
But this should all be very surprising and possibly concerning to the people putting those presentations together for Medicaid. Instead of c-sections being clearly bad for low-risk births, the conclusion here is that it is far from obvious—you trade some problems for other problems.
The favorable result in the Apgar score, which measures how lively the baby is just after birth, seems consistent with the mechanics of the two delivery methods. With vaginal birth, the baby can get stuck. This kind of problem is less likely with c-sections. This is also an imminently critical outcome: some babies with low Apgar scores will have permanent brain damage from hypoxic ischemic encephalopathy.
The respiratory issues for c-section babies also line up with a recent assessment which singles out “delayed clearance of lung liquid” as a culprit, although it seems more mysterious.
The outcomes for mothers are a little hard to piece together. The outcome I put most weight in is length of stay, as this should capture most of the very bad things depicted in the Medicaid slide above. The insignificant but positive estimates on emergency room visits, even taken at face value, are much smaller than the estimates in the Medicaid slide, which all suggest a doubling of risk or more.
Finally, for what it’s worth, the flat results for mothers and infants line up with a randomized trial of twin pregnancies: “planned cesarean delivery did not significantly decrease or increase the risk of fetal or neonatal death or serious neonatal morbidity, as compared with planned vaginal delivery,” and “We did not find that planned cesarean delivery was associated with a higher or lower risk of maternal death or serious maternal morbidity than planned vaginal delivery.”
Subsequent pregnancies
The ACOG guidelines are actually more worried about effects on subsequent pregnancies: “the downstream effects are even greater because of the risks from repeat cesareans in future pregnancies.” The main theory, from what I can tell, is that the c-section scar interferes with the formation of the placenta, resulting in placental issues and preeclampsia. These are dangerous conditions that greatly increase the chance of harm for both mother and baby. There could also be subtler issues getting nutrients to the baby that don’t register in checks of maternal health.
Their reference for this is Silver et al. I’m a little stunned at how simplistic the analysis is. They only study c-section births and look at the correlation between negative outcomes and the number of previous births. The problem is that all of this could just be a birth order effect, not a number-of-previous-c-sections effect. Second, while they find increased risk of placental malformation in subsequent pregnancies, the initial c-section might have happened because of this very issue. Surely moms who have a given issue are more likely to have it again, regardless of method of delivery.
More convincing evidence comes from Daltveit et al 2008, using Norwegian register data. They compare outcomes for the second pregnancy for women whose first birth was vaginal vs. Cesarean:
The All women comparisons don’t impose any restrictions. In the Selected women section, they leave out women who had the same condition in their first pregnancy. This addresses the important source of bias that Silver et al ignore. Still, most outcomes show a clear increase.
There are increases in preeclampsia and SGA (small for gestational age). These could be connected with malformation of the placenta. There are effects on placenta outcomes and uterine rupture, although these represent small absolute risks. The risk of placenta previa increases from 0.22% to 0.33%. The risk of uterine rupture increases from 0.006% to 0.2%.
The uterine rupture result is scary but this appears to be driven by women who attempt vaginal delivery after a c-section—something the authors don’t account for. Another source that compares people based on their planned method finds strong evidence of this: “The estimated incidence of uterine rupture was 2 per 10,000 maternities overall; 21 and 3 per 10,000 maternities in women with a previous caesarean delivery planning vaginal or elective caesarean delivery respectively.” So risks were practically the same for the caesarean-to-caesarean mothers.
One lingering confound is that, even among the Selected women, “confounding by indication” may still cause the previous c-section mothers to have worse prospects, biasing the comparison. A way around this is to look at outcomes for elective c-sections—mothers who are not getting the c-section because of a health need. This seems to show the same result. For example, in this case control study, mothers with placenta accreta were more likely to have had a previous elective c-section.
Conclusion
Where does that leave us? The Card et al paper should make us more agnostic about the costs of a c-section for a one-off, low-risk birth. But, while I could not find careful causal studies on subsequent births, the observational studies do suggest that they may suffer negative effects of a previous c-section. Based on this investigation, I now believe that choosing an elective c-section for a one-off kid, or your last pregnancy, could be the more defensible choice.
In discussing these ideas with doctors, I get a common response that “A c-section is major abdominal surgery!” From my perspective, given the tragic injuries that can befall mother and infant in the process of childbirth, Mother Nature hasn’t done herself any favors. Maybe we can, in fact, do better.
Notes
I am still looking for better cites on this, but it appears that maternal mortality for elective c-sections, is lower if anything in observational data. (Although the selection bias here could favor scheduled c-sections.)
As always, Emily Oster is the best place to start with questions like this. She believes the effects on placental malformation in subsequent pregnancies. See her thoughts here. In particular: “The good news is, if we look at one or two years after birth, there is no evidence of differences in recovery. If anything, the data might slightly favor C-sections.” In a big meta-analysis she cites, “The only two complications that differed across groups were urinary incontinence and pelvic organ prolapse. Both are actually more common for women who have had a vaginal delivery.”