Life-Saving Data That Is Nowhere To Be Found: Hospitals’ C-section Rates


The United States is the only developed nation in the world with a steadily increasing maternal mortality rate — and C-sections are to blame. Nearly 32% of babies are born via C-section in the United States, a rate of double or almost triple what the World Health Organization recommends. While C-sections are an incredibly important life-saving intervention when vaginal delivery is too dangerous, they are not devoid of risks for mom or for baby. Hospitals and doctors alike are aware, as it’s been widely reported that unnecessary C-sections are dangerous — and hospitals and doctors agree that the number one way to reduce this risk is to choose a delivery hospital with low a C-section rate. However, information on hospitals’ C-section rates is incredibly hard to find, which leaves women in the dark as they try to make this important choice.

In an effort to help women make informed decisions about where to deliver their babies, we set out to collect a comprehensive, nationwide database of hospitals’ C-section rates. Knowing that the federal government mandates surveillance and reporting of vital statistics through the National Vital Statistics System, we contacted all 50 states’ (+Washington D.C.) Departments of Public Health (DPH) asking for access to de-identified birth data from all of their hospitals. What we learned might not surprise you — the lack of transparency in the United States healthcare system extends to quality information, and specifically C-section data.

After contacting 51 DPHs, 44 departments provided some level of birth data upon request — but the majority of those shared C-section rates for their state’s counties or districts, which doesn’t help when a patient needs hospital-specific data in order to select where she’ll deliver. Some states, such as Alabama, California, Massachusetts, Pennsylvania, Vermont, and West Virginia, have very transparent data-sharing practices, posting the vital data on their websites for the public to access and use. Other states, such as Wisconsin, Missouri, and Nebraska, put obstacles to accessing their data in place, including charging fees, requiring a signed data use agreement, and sometimes demanding institutional review board exemption. Still, six states — Illinois, Kentucky, Georgia, Wyoming, and Connecticut — outright refused to share their data, citing verbiage in their state’s statutes as the rationale.

Type of data shared from each state’s Department of Public Health

While it’s clearly difficult for patients to access these quality data on their own, thankfully there are organizations working to address the issue of unnecessary C-sections head-on. On the West Coast, the California Health Care Foundation developed a comprehensive initiative to reduce unnecessary C-sections, and on the East Coast, Ariadne Labs, a research group out of Harvard Medical School, has an entire department devoted to “Addressing the world’s most common and consequential surgical error: the decision to perform a C-section.” While these programs are making great strides, the U.S. could be doing a lot more to give the patient a voice and promote informed healthcare decision-making.

For organizations who want improve health outcomes, reduce healthcare spending, or enhance patients’ interactions with the healthcare system, allowing access to a comprehensive, accurate dataset of hospital C-section rates should be a top priority. The publication of C-section data at the hospital level would allow women to make informed decisions about their healthcare, mitigate unnecessary adverse outcomes, and reduce healthcare spending. Public data might also help influence change among healthcare providers and hospitals. Healthcare advocates, payers, patients, researchers, and the public alike need to band together to change this opacity. Mothers’ lives depend on it.

Dani Bradley leads the research team at Ovia Health, a health technology company providing innovative solutions to support women and families.

10 replies »

  1. You have to offer some conjectures, pj. You can’t just fling out this aweful data without telling us what you think is going on….besides a decay in our “social capital”.

    Here are some confounding ideas.
    Because our fertility rates are higher, we are seeing a slightly less-well population that becomes pregnant.
    Our definition of post partum may be different. Eg we might be saying that pp includes folks who have had abortions at home at 20 weeks.
    Maybe we have so much more diversity in our cultures that some pp mothers are not doing something important or are doing something bad that is causing deaths ( like sex too soon after delivery)
    Maybe our excess deaths are from our excess C-sections?
    Maybe our attorneys are defining non-gestational deaths as pp deaths so as to litigate more of these cases. Like DVTs in hemoglibin SC disease.
    Maybe pp deaths here are counted even if mothers used totally unorthodox deliveries thst may be off the books on other OECD countries. Eg home deliveries by religious groups’ midwives.
    These kinds of reasons may explain some of your assertions, dont you think?

  2. The higher tier OECD countries nearly all have better public health reporting. They often catch problems we do not because they collect data and monitor more widely and effectively. What is not being said here is that in some cases the increase in maternal mortality increased suddenly in states that had at least average rates before they made changes that resulted in maternal mortality figures seen more often in second and third tier economies.


  3. Reportedly attributed to Saint Augustine “In the absence of justice, what is sovereignty but organized robbery?” Feudalism still plagues the expression of justice.

  4. “Something about chex and balances.”

    The Constitution says “We the people”, not, “We the majority party”. This system of winner take all is a faux democracy and nothing less than corporate feudalism run by large money.

    Yes, you had the checks (bank account) right.

  5. A high level moral hazard exists within the institutional co-dependency between the medical institutions heavy committed to providing Complex Healthcare and the financial institutions that pay for this healthcare. There is an entrenched paradigm paralysis that surrounds the rules they use to resolve the social dilemmas they encounter when trying to stabilize their economic status. The Federal budget proposed by President Trump included severe cuts to the NIH, especially cancer research. Fortunately, Congress renewed their existing funding with a mild increase. Something about chex and balances.

  6. “Like our nation’s HEALTH generally, maternal mortality has worsened as a result of the decline in our nation’s Social Capital, community by community. Without a nationally sanctioned and locally originated (and funded) strategy, our nation’s healthcare for women during a pregnancy, families with young children, the homeless or the disabled will continue to experience worsening levels of entrenched poverty, mass shootings, obesity, young adult suicide/homicide, substance addiction/mortality, mid-life depression and DECREASING longevity.”

    Don’t worry, Trump and Republicans with their higher moral standards will fix all this.

  7. The nation’s with lower levels of maternal mortality are the other 34 members of the Organization for Economic Cooperation and Development (OECD). They all have fairly robust Public Health traditions. The only long-term outcome data supporting the basis for PAP smears came from Denmark that represented @10 year outcomes and began before WWII started and finished a few years after WWII. There is only historically one data issue and that has to do with the WHO definition for a maternal death to a period of 6 weeks after the end of the pregnancy. Some of the States have variably used the one year after the end of the pregnancy. That issue has largely disappeared in the last ten years as the CDC has slowly worked out the reporting processes. Since about 2007, the national and state by state maternity mortality ratio data has not been publicly reported until 2016. The management of all that is a bit murky, at best, but seems to be slowly improving. At issue, some of the States with a really bad level of maternal mortality apparently threatened the CDC that they would stop reporting any public health data to the CDC, if they didn’t stop reporting maternal mortality data. This might have also involved a Congressional resolution occurring about the time that the CDC was prohibited from doing any epidemiologic analysis of Mass Shootings in 2007 (rescinded within the last 12 months).

    My impression is that Congress had a chance @2011 to fix all the maternal mortality issues with a good set of legislative actions. This was all triggered by the AMNESTY INTERNATIONAL USA study entitled “Deadly Delivery” in September 2009. Its a bit graphic, but so is the reality of this lingering issue. To fully understand the issues, the following might help. There were 86 mass shooting from 1985 through 1999. From 2000 through 2015, there were 206. The increase represents “””234%”””. For 1987 through 1996, the national maternity mortality ratio averaged 6.84 deaths per 100,000 live births. For 2005 through 2014, the average annual maternal mortality ratio was 16.36. The increase represented “””239%”””. Coincidence, maybe…probably not. The cost and gaps in quality within our nation’s healthcare are multi-factorial. Obviously, there are substantial issues to solve within our nation’s healthcare industry. But, its also possible that these are aggravated by very large problems within the social fabric of our nation’s communities. These are likely the most important factors driving our nation’s maternal mortality. The complexity is shown by the mass shooting data. I’ll stop on that note.

  8. pj, Of course high rates of maternal mortality can also be owing to “we report accurately and other countries don’t”. How can one prove that another nation is missing gathering the data to produce an accurate statistic?

  9. As you probably already know, there has been a CDC moratorium on releasing State by State maternal mortality data since about 2007. This thick layer of ICE was finally unleashed in October of 2016 when a ten year, State by Sate, data-set was finally published in the OBESTETRICS and GYNECOLOGY journal. The data-set has been largely ignored within the medical community because the CDC folks continue to passively down-grade its significance. I understand their reasons for it because Congress has largely failed in its responsibility to manage the underlying issues. An effort around 2012 failed in Congress, to be repeated since then at a lower level of transparency. The failed Congressional efforts began after AMNESTY INTERNATIONAL USA published it scathing study of maternal mortality, DEADLY DELIVERY, in September of 2009.

    The scientific community continues to believe that a protocol-based, command and control strategy will fix the problem. Truly, there are aspects of maternal mortality that will require this level of discipline. But, the underlying problem for maternal mortality has not been acknowledged. Its odd that we have made great progress for infant mortality but virtually none for maternal mortality. Given even the broadest use of the data, there are at least 500 women who die in conjunction with a pregnancy every year that would still be alive if they had been living in another nation before the start of their pregnancy. To further define the issues, African emigrant women who encounter a pregnancy soon after immigration do not experience the higher-risk of maternal mortality occurring among native African American women (more than 3 times worse than for white women).

    Remember that a pregnancy represents a biological state of immune tolerance during gestation. In effect, labor and delivery represents a process of immune rejection. Recently, JAMA ( https://doi:10.1001/jama.2018.7028 ) reported a population study regarding a significant association of stress-related disorders with the subsequent occurrence of an Autoimmune Disease. A reverse causality attribute of the study demonstrated that the use of an antidepressant medication for a new Post-Traumatic Stress Disorder episode prevented the occurrence of the increased “autoimmune disease” prevalence.

    Like our nation’s HEALTH generally, maternal mortality has worsened as a result of the decline in our nation’s Social Capital, community by community. Without a nationally sanctioned and locally originated (and funded) strategy, our nation’s healthcare for women during a pregnancy, families with young children, the homeless or the disabled will continue to experience worsening levels of entrenched poverty, mass shootings, obesity, young adult suicide/homicide, substance addiction/mortality, mid-life depression and DECREASING longevity. This could be implemented for $1.00 per citizen per year with a Congressional Charter for a semi-autonomous institution that follows the design principles of economist Ellinor Ostrom (see below citation).

    Here is my own rendering for a SOCIAL CAPITAL definition:
    …a community’s norms of Trust, Cooperation and Reciprocity that
    …its citizens spontaneously express for resolving the social dilemmas
    …they encounter within their community’s municipal life
    …WHEN Caring Relationships are persistently nurtured
    …within the social networks of the community’s citizens,
    …especially the enduring Caring Relationships occurring
    …within the Micro-Neighborhood Network of each citizen’s Family.

    Remember that a reverse causality has been identified for a large community between HEALTH and Trust. Read the Social Capital definition again for a couple of times. Try substituting ‘generational’ for ‘enduring.’ Yep, its a really big, hairy, audacious GOAL.

    Core design principles for the efficacy of groups http://dx.doi.org/10.1016/j.jebo.2012.12.010

    Trust and Health reverse causality http://doi:10.1136/jech-2015-205822