Masking Maternal Mortality
By Ina May Gaskin
Mothering, March/April 2008
Twenty-two-year-old Army Specialist Tameka McFarquhar had no family members nearby to help her when she was released from Samaritan Medical Center in Watertown, New York, a day after giving birth to her first child on December 14, 2004. A single mother, the Jamaican-born office clerk had been transferred Stateside from Army duty in South Korea after becoming pregnant with her daughter, Danasia Elizabeth. She never revealed her child’s father’s name.
On the night of December 19, McFarquhar spoke with her mother in Jamaica and told her that she had a headache. Worried about her, her mother advised her to drink some milk and keep herself warm.
That phone call was the last time that any family member or friend heard McFarquhar’s voice. No one could get her
to answer her phone or her apartment door. A concerned friend notified the Watertown police, who found no probable cause to break into the apartment. Finally, on Christmas morning, McFarquhar’s friend again contacted the police, who this time went to McFarquhar’s apartment, only to find a horrifying scene. McFarquhar had bled to death several days earlier, and baby Danasia had died of dehydration and starvation.
According to the Watertown Daily Times, the Jefferson County medical examiner said that the cause of McFarquhar’s death was placenta increta, a rare complication in which the placenta cannot be released in the normal way because it had burrowed into the uterine muscle instead of attaching only to the uterine lining, which is shed just after birth.1 But could it really have been an increta? Placentae increta must be removed surgically with the patient under deep anesthesia; if McFarquhar had indeed had this complication, she must have been discharged from the hospital with the placenta still inside her uterus. Who could believe that that had happened?
One possibility is that a very ?small bit of placenta or membrane was left inside McFarquhar’s uterus, a much more common occurrence that can indeed cause a late postpartum hemorrhage such as the one she suffered. Because it sometimes takes a few days for soreness and infection to develop, this complication could more easily have been missed than a placenta increta in any examination that took place during the 24 hours McFarquhar was still in the hospital following birth.
It’s not just single, first-time mothers who can die from lack of post-birth follow-up care. The same complication is likely to be what happened to Galit Schiller, a San Anselmo, California, mother of three, who, three days after giving birth in a hospital in June 2007, died in her husband’s arms of a massive post-birth hemorrhage.2
Maternity-care systems in countries with low maternal death rates (the US is not among these) plan for the certainty that some percentage of previously healthy women will be in danger of a late postpartum hemorrhage, uterine or perineal infection, breastfeeding problem, postpartum depression, or some other post-birth complication requiring special attention. These countries—Australia, England, the Netherlands, New Zealand, Norway, Northern Ireland, Scotland, Sweden, and Wales, to name just a few3—send specially trained nurses to make home visits to new mothers during the first ten days or so following birth. Had McFarquhar had such a visit, her death and that of her baby could almost certainly have been prevented, as incomplete expulsion of the placenta and membranes can rather easily be diagnosed by a trained professional before a life-threatening hemorrhage occurs.
Inexcusably, such home visits during the first week or ten days postpartum are rare in the US, even though most women here are discharged from the hospital too early for some problems to be detected. The exceptions to this rule are those mothers who had planned homebirths, since post-birth visits are considered necessary by the attending midwives. But for women giving birth in hospitals, it seems fair to ask why most US maternity services fail to recognize that postpartum home visits by midwives, nurses, or physicians are not luxuries, but necessities for every new mother. Making post-birth home visits part of the standard maternity-care package in the US is only one of the steps that our nation should take to reduce the maternal death rate.
Every three years, the British Royal College of Obstetricians and Gynaecologists publishes a book titled Why Mothers Die. Anyone in Wales, Scotland, England, and Northern Ireland can walk into a bookstore and buy the 400-page book, which is a sort of report card on the results of the combined maternity services of the four countries.4 As the public-?outreach component of the UK’s respected Confidential Enquiry into Maternal Deaths, which is now part of the Confidential Enquiry into Maternal and Child Health (CEMACH), each edition of Why Mothers Die is based on data drawn from every maternal death in the UK from causes stemming from pregnancy or birth during the preceding three years of available data. Each of the main causes of maternal deaths—hypertension, thromboembolism, hemorrhage, amniotic fluid embolism, infection, anesthesia problems, and injuries to the cervix, perineum, or vagina—gets its own chapter and includes at least one narrative of a case of such a death.
The UK claims a high degree of accuracy in determining how many maternal deaths occur each year; the annual Why Mothers Die report is actually considered the “‘gold standard’ in professional self-audit.”5 Sometimes cases involving substandard care are described in Why Mothers Die, but the names of hospitals or cities are never mentioned. Because the purpose of the CEMACH program (a rough equivalent to our own Centers for Disease Control) is to get at the truth, names and places are kept confidential so that results of the inquiries can’t be used in malpractice lawsuits. Why Mothers Die not only provides detailed, accurate numbers for each category of death, but also makes recommendations about what steps should be taken to ensure that the number of deaths will be reduced in the next three-year period. As of 1999, building on the excellent feedback provided by CEMACH, the UK maternity system has been able to reduce the number of maternal deaths each triennium.6 A slight, statistically insignificant rise in the death rate was reported in the 2002 edition of Why Mothers Die.7
What a different situation we have in the US. Here, while we take it for granted that everything possible is done to prevent maternal deaths, most of us haven’t a clue about what this effort requires. We don’t read much about maternal deaths in the news media or on the Internet, so we assume that women rarely die from pregnancy or birth. What we don’t realize is how infrequently the deaths of mothers that do occur are mentioned in the news. Tameka McFarquhar’s death, for instance, was mentioned only in the Watertown paper and a newspaper in Kingston, Jamaica—never nationally. More recently, Caroline Still Wiren’s death, from hemorrhage, after her first birth in a Florida hospital in spring 2007, made the news most likely because of her husband, Nyle Wiren’s, fame as a longtime football player for Tampa Bay.8 In early spring 2007, when two teachers from the same small-town New Jersey elementary school died within 15 days of each other after the cesarean births of their first babies, local news coverage of the story prompted national coverage on network television.9 But when a third New Jersey mother died after the cesarean birth of her twins two months later, we were back to the norm of no news coverage at all.10
Whereas the British get their 400 pages of detailed reporting and analysis every three years, with special attention given to making these public, our single statistic per year of the number of women who die within 42 days of giving birth, issued by the CDC and the National Center for Health Statistics and derived from the same sources, comes with no details, no identification of important trends, no analysis, and no recommendations. With so little media interest in this subject and nothing but that single number once a year from the CDC, it’s no wonder that we in the US have a false sense of security about maternal deaths.
Even the published maternal death rate is far from accurate. According to the CDC in 1998, there is so much misclassification in the US system of maternal death reporting that the actual number could be as much as three times greater than the number officially published each year.11 What an admission for the CDC officials to have to make. Think how it would be received if the Federal Aviation Administration had to admit that they count and investigate only half to a third of the plane-crash deaths that occur each year. For instance, by making a more careful survey, Michigan recently found that its supposed maternal mortality rate of 7.6 deaths per 100,000 births should be revised to 18 per 100,000 births. According to the physician who wrote that report, it is likely that other states significantly underestimate their maternal mortality as well.12 I have found no news coverage yet of these problems, except for a few articles that appeared in 1999 and 2000 in obstetrical trade publications.13–17
One striking reason for this great inaccuracy is that the 50 states aren’t required to use the same death-certificate form. Although there have been efforts since 1979 to get every state to ask the obvious question about whether the deceased woman had been pregnant in the weeks or months preceding her death, only 21 states include such a question.18 This is important, because the ability to gather accurate statistics depends on how each woman’s death certificate is filled out. Women of childbearing age may die from a host of causes, and all of these deaths need to be distinguished from those that are directly or indirectly related to a woman’s pregnancy or birth. An almost 30-year effort to get states to cooperate by including the pregnancy question on the death certificate still leaves more than half of states without it.19 Sadly for women, this results in an unknown number of deaths being misclassified each year.
What this means, in practical terms, is that if a woman with Tameka McFarquhar’s problem goes back to a hospital and then dies, she is likely to do so in an emergency room, operating room, or intensive-care unit. If this happens in a state in which the death certificate lacks the essential question, and the woman’s family does not insist on an autopsy, it is very possible that the certificate could list her death as due to hemorrhage without ever referring to her recent pregnancy. In the world of statistics, this would translate to one fewer maternal death—which would be false. A recent article in a major obstetrical journal revealed a rate of underreporting of maternal death in Massachusetts of ?93 percent.20
Another problem noted by the CDC is the extent to which “the completeness and quality of maternal death reporting could be improved if physicians completed the cause-of-death section of the death certificate more accurately.”21 Health-care systems that prioritize the prevention of maternal death make sure that anyone authorized to fill out a death certificate has had the training necessary to do it correctly. There is an old expression about large databases: “GIGO,” for “garbage in, garbage out.” Clearly, this is the situation in the US about this important subject. Why do we continue to allow this kind of sloppiness?
Incidentally, the US autopsy rate, which was about 50 percent in 1960, has dropped to less than 5 percent.22 In large part, this development has taken place for economic reasons. Autopsies, although they are necessary for medical research, medical education, and quality control, do not generate profit for hospitals.23 Numerous studies have shown that, in 40 percent of cases, autopsy reveals a different cause of death from what had previously been diagnosed.24 For this reason, in Austria, it is required by law that every hospital death be followed by an autopsy.25
Whenever a maternal death occurs in the UK, CEMACH is automatically notified, and a multidisciplinary team of individuals who do not work at the hospital where the death occurred is dispatched to review all of the woman’s records. In sharp contrast, when a maternal death takes place in the US, there is usually no review of the case external to the hospital in question, and all employees with knowledge of the death are warned to keep silent about it.
My own informal survey and research reveal that fewer than half of the 50 states still have mortality and morbidity review committees, but the findings of the committees of those states that do (e.g., Alaska, Colorado, Florida, Georgia, Maryland, New York26–31) are only beginning to be made available to the public. Some states, including my own, Tennessee, have never bothered to conduct maternal mortality and morbidity reviews to clarify the causes of preventable deaths. However, almost every state has statutes that protect a review committee’s reports, proceedings, and findings from legal discovery.32
With only an honor system to encourage accurate reporting within a profit-based health industry, should we be surprised that so many deaths are missed in the count? There are no penalties for misclassification, and national audits are not possible. The states report to the CDC only crude numbers, with no names or records attached, so there is no way to find out whether a given woman’s death has been included. Can we imagine banks and other financial institutions being trusted to function according to such an honor system?
The CDC did make another very important statement in its 1998 report, one that it has since had no reason to change: There has been no improvement in the maternal death rate since 1982, when it was reported to be 7.5 deaths per 100,000 live births.33 Our current maternal death rate is four times as high as it should be, and this statement, remember, is based upon our underreported figure. According to the Department of Health and Human Services, the rate should not exceed 3.3 deaths per 100,000 live births, whereas the rate in 2004 was more than 13 deaths per 100,000 births.34
In February 2007, the CDC issued a report predicting that maternal death rates in the future will likely rise—not because of actual increases in maternal mortality, but because more states may add to their death certificates the important question about whether or not a deceased woman had been pregnant within the year before her death.35 What evidence, I wonder, does the CDC have that would allow it to say that it is unlikely there will be an actual increase in maternal mortality rates? Yes, it’s possible that some of the rise in maternal mortality can be attributed to better reporting, but comparatively little has changed when we consider the many flaws in our chaotic methods of maternal death ascertainment. We still have the honor system, no penalties for misclassification or false reporting, little or no training for those who fill out death certificates, and very few autopsies. In addition, most states lack maternal mortality and morbidity review committees, and almost none has the power to look at medical records, or a way to audit the data.
Another vital point: We have a US Standard Certificate of Death. In 1979 and 1989, it was proposed that this certificate should include a question asking if the deceased had been pregnant in the year previous to death. Inclusion of this question has been shown to significantly increase the count of maternal deaths. Amazingly, this question was not adopted in the US Standard Certificate until 2003. However—and this is a big however —the federal government does not require that the states use the US Standard Certificate, and most still don’t! From 1996 to 1998, 16 states included a question related to pregnancy status, and by 2003, 21 states had such a question. But, according to the CDC, ?“in 2003, only four states could capture information consistent with the standard.”36 The UK has no problem with this kind of craziness. All four countries of the UK use the same forms while, when lives are literally at stake, we can’t get more than a handful of states to cooperate in this gathering of important public health information. What is wrong with us?
Here’s another shock: The rate of maternal death for black women in the US for 2004 was 34.7 deaths per 100,000.37 Most countries with rates that high are seeking help from international agencies. In fact, according to the World Health Organization, at least 29 other countries have maternal death rates lower than we do in the US.38 But even that ranking is based on our officially published rate, which is, by the CDC’s own admission, very much underreported.
The Safe Motherhood ?Quilt Project
About seven years ago, I felt powerfully impelled to follow the example of the AIDS Quilt in drawing attention to the issue of underreporting maternal deaths and lack of media interest in this problem. I began stitching on a quilt piece that grew into the Safe Motherhood Quilt Project (www.rememberthemothers.net). Whenever I receive documentation of a US woman’s death from pregnancy-related causes between 1982 and the present, I arrange for a quilt block to be made in her honor. Sometimes a family member or friend creates the block, and sometimes it is made by one of the many who have contributed their efforts to the project.
The Quilt was first exhibited at the Summit for Safe Motherhood, sponsored by the Centers for Disease Control, the American College of Obstetricians and Gynecologists, and the American College of Nurse-Midwives, and held in Atlanta, Georgia, on September 4–5, 2001. Since then it has been shown at the Oakland Museum, at Dartmouth-Hitchcock Medical Center, and many other sites in the US, as well as in Austria, Brazil, Canada, Costa Rica, England, France, Germany, Hungary, Iceland, Ireland, Italy, Mexico, Northern Ireland, Norway, and Scotland.
I’m sure that when enough US women are informed about the maternal death problem in our country, we can exert enough political pressure to fix it. Only when we are able to equal the UK’s CEMACH system of ascertaining and analyzing maternal deaths will we be able to find out the causes of preventable maternal deaths and then set about preventing them.
Ina May Gaskin, certified professional midwife, is director of The Farm Midwifery Center, in Summertown, Tennessee. She is the author of Spiritual Midwifery, now in its fourth edition, and Ina May’s Guide to Childbirth, now in its fifth printing since its release in early 2003.
1. John Golden, “Postnatal Problem Ruled to Be Cause of Soldier’s Death,” Watertown Daily Times (8 March 2005).
2. Jim Staats, “San Anselmo Rallies around Family that Lost Young Mother: Woman Died Just Days after Childbirth,” Marin Independent Journal (30 June 2007).
3. Personal communications (all 4 December 2007) with: Tine Greve, midwife, Oslo, Norway; Bodil Frey, childbirth educator, Gothenburg, Sweden; Thea van Tuyl, childbirth educator, Apeldoorn, Netherlands; Robyn Thompson, RM, RN, M&CHN, BachAppSc, BF Cons, Victoria, Australia; Lorna Davies, RM, Senior Lecturer in Midwifery at Christchurch Polytechnic Institute of Technology, formerly Senior Lecturer at Anglia Ruskin University, Essex, England.
4. Why Mothers Die 2000–2002: The Confidential Enquiry into Maternal Deaths in the United Kingdom (London: RCOG Press, 2004): www.cemach.org.uk/Publications/Saving-Mothers-Lives-Report-2000-2002.aspx
6. Why Mothers Die 1997–1999: The Confidential Enquiry into Maternal Deaths in the United Kingdom (London: RCOG Press, 2001).
7. See Note 4.
8. Kevin Graham and David Murphy, “Death Shocks Fans, Family,” St. Petersburg Times (18 May 2007).
9. Marie McCullough, “Joined in Birth, Death,” Philadelphia Inquirer (10 May 2007).
10. Personal communication with Maria Korfiatis-Barroso, New York City Chapter Leader, International Cesarean Awareness Network (24 September 2007).
11. David Johnson and Teresa F. Rutledge, “Maternal Mortality: United States, 1982–1996,” The Morbidity and Mortality Weekly Report 47, no. 34 (4 September 1998): 705–707.
12. Timothy F. Kirn, “Maternal Mortality Rates Grossly Underestimated,” Ob.Gyn. News (15 January 2000).
14. “Pregnancy-Related Deaths: Moving in the Wrong Direction,” OBG Management (January 1998).
15. Cindy Obenstine, “Maternal Mortality: No Improvement Since 1982,” ACOG Today 43, no. 7 (August 1999): 6–7.
16. Jeffrey C. King and Cynthia J. Berg, “Maternal Mortality: An Unsolved Problem,” Contemporary OB/GYN (September 1999): 144–146.
17. Michael McCarthy, “US Maternal Death Rates Are on the Rise,” The Lancet 348, no. 9024 (August 1996): 394.
18. Donna L. Hoyert, “Maternal Mortality and Related Concepts,” National Center for Health Statistics, Vital and Health Statistics Series 3, no. 33 (February 2007).
19. C. Berg et al., eds., Strategies to Reduce Pregnancy-Related Deaths: From Identification and Review to Action (Atlanta, GA: Centers for Disease Control and Prevention, 2001).
20. C. Deneux-Tharaux et al., “Underreporting of Pregnancy-Related Mortality in the United States and Europe,” Obstetrics and Gynecology 106, no. 4 (October 2005): 684–692.
21. See Note 19.
22. E. C. Burton, “The Autopsy: A Professional Responsibility in Assuring Quality of Care,” American Journal of Medical Quality 17, no. 2 (Mar–Apr 2002): 56-60.
23. Jack Hasson, et al. “Autopsy training Programs: To Right a Wrong,” Archives of Pathology and Laboratory Medicine 119, no.3 (March 1995): 289–291.
24. G. Lundberg, “Low-tech Autopsies in the Era of High-tech Medicine: Continued Value for Quality Assurance and Patient Safety,” Journal of the American Medical Association 280 (1998):1273–1274.
25. Hani K. Atrash et al., “Maternal Mortality in Developed Countries: Not Just a Concern of the Past,” Obstetrics and Gynecology 86 (1995): 700–705.
26. Sandra Mullin and Andrew Tucker, “Citywide Infant Mortality Rate was 6.1 in 2004, a decline of 6% from 2003,” New York City Department of Health and Mental Hygiene press release 095-05 (8 September 2005): www.nyc.gov/html/doh/html/pr/pr095-05.shtml.
27. “Findings of the Alaska Maternal-Infant Mortality Review, 2000,” Family Health Dataline 8, no. 1 (Anchorage, AK: May 2002): www.epi.hss.state.ak.us/mchepi/pubs/dataline/2002_01.pdf.
28. Colorado Department of Public Health and Environment, Women’s Health Section, “Maternal Mortality in Colorado 1990-1997” Colorado Maternal Mortality Review Committee Brief, no. 1(Denver, CO: August 2000).
29. Angel Watson, MPH, RHIA, Florida Department of Health, Division of Family Health Services, “Pregnancy-related Deaths during the Postpartum Period, 1999–2004: Pregnancy-Associated Mortality Review” (18 April 2007): www.doh.state.fl.us/family/mch/docs/pdf/PAMRDC041807.pdf.
30. Georgia Department of Human Resources (DHR), Division of Public Health, “Georgia Maternal Mortality Surveillance Fact Sheet” (undated): http://health.state.ga.us/pdfs/epi/mch/maternalmortality.fs.04.pdf.
31. Governor Robert L. Ehrlich, Jr., “Maryland Department of Health and Mental Hygiene, Family Health Administration, Center for Maternal and Child Health, Maternal Mortality Review Program 2006 Annual Report.”
32. See Note 19.
33. See Note 11.
34. Arialdi M. Miniño et al., “Deaths: Final Data for 2004,” National Vital Statistics Reports 55, no. 19 Centers for Disease Control (21 August 2007).
35. See Note 18.
36. Ibid, page 4.
37. See Note 34.
38. C. Abou Zahr et al., Maternal Mortality in 2000: Estimates Developed by WHO, UNICEF, and UNFPA (Geneva, Switzerland: Department of Reproductive Health and Research, World Health Organization: 2004), 16–17.