On Wednesday, March 13, 2024 the Washington Post posted on line at 7:28 PM EDT an article written by Sabrina Malhi and Dan Keating with the headline “Study says U.S. maternal death rate crisis is really a case of bad data.” (1) The article was citing a paper published on line on March 12, 2024 by faculty members from Obstetrics and Gynecology Departments, Pediatrics Departments and Biostatistics in The University of British Columbia, Robert Wood Johnson School of Medicine, Rutgers and Laval University. (2) The objective of the research described by the authors was to increase our understanding of the reason behind the high maternal mortality rates in the U.S to determine if the causes can be attributed to changes in obstetrical factors, maternal medical conditions or to the methods used in maternal death surveillance. They focused on the addition to death certificates of a checkbox that determines if the deceased was pregnant or not. The authors evaluated the methods used by the National Vital Statistics System against the method they designated which lists only defined pregnancy related deaths and reported that by using their system of calculation, the US mortality rate was 10.2/100,000 births in the period of 1990-2002 and 10.4/100,000 live births in the timeframe of 2018-2021 and concluded that the US rates are stable and not rising and suggested that the concerns for the very high maternal mortality rates reported by the US government agencies was due to our current system of data collection or as the Washington Post headline stated due to “bad data”. In carefully evaluating the article in the American Journal of Obstetrics and Gynecology, I was not convinced that the US maternal death rate is all about bad data. They defined the definition of maternal death as a death “caused, related to or aggravated by pregnancy or its management”. I found it difficult to accept their approach that accepted preeclampsia as caused by pregnancy but that a preexisting condition of hypertension was not. The physiology of pregnancy which increased blood volume and changes in cardiac dynamics does and can aggravate hypertension and if not managed appropriately can lead to death. They did not appear to understand that cancer during pregnancy often alters the treatment of the disease in ways that can alter the ultimate management and thus the outcome of the disease. To me many of their premises regarding pregnancy and pregnancy outcomes are flawed. As a clinician and a former Chair of Obstetrics and Gynecology Departments I do know that birth and death certificates often do have inaccurate data and it is important to work to minimize these errors. My experience is that the medical profession takes the data on birth and death certificates seriously and health care facilities have established ways to achieve the collection of accurate reporting. In all the places I worked trained clerks collected data from the patient’s chart and entered this onto the birth or death certificate and the physician reviewed this carefully prior to signing it. If the purpose of the article is to call attention to the need to improve vital statistics reporting that is fine. In reading the paper I am willing to give the authors a bit of credit in this regard. They also point out that although their reported findings stated that by their analysis the maternal mortality rates in the U.S. are stable, they are still high relative to the rates of other countries. They also report that they found that the rates for women of color are disparate compared to that of white. They call for further research. Interestingly, the authors have previously published an article on the same topic in another journal so this is not the first time that they have discussed surveillance system issues (3). I am not happy with the Washington Post article and the suggestion that the maternal mortality “crisis is really a case of bad data”. This is a great disservice to our country and does not bode well for our current national debates on health care. I worry that many persons will say to themselves that maternal mortality is not a problem in the U.S. and we thus do not have to do anything to reduce it. The fact remains that our rates are higher than other developed countries, and that health disparities exist. Others in obstetrics and gynecology have written comments regarding the Washington Post article. The American College of Obstetricians and Gynecologists has put out a press release. (4) Finally, I read many of the comments listed online by readers of the Washington Post news article. It is always hard to know when reading these types of comments if they are being made by trolls or by serious people. However, in many cases I was saddened by the lack of understanding of the importance of the collection of vital statistics and how this information is used by medical and scientific professionals. This important data leads to research studies and in the case of maternal deaths have led to the initiation of statewide maternal death review boards. These boards are also able to evaluate material on the death certificate will ignore “bad data” such 70- year -old who is incorrectly listed as pregnant. This data is important in the understanding of health and disease trends and leads to further studies. Individuals who conduct research using vital statistics are trained to eliminate incorrect data and to do analysis is data that has been “cleaned” using standard and accepted techniques. I was also disturbed by comments that were biased toward certain populations such as low income or low educated groups and racial minorities. One person commented that death certificates are a waste of time. The fact that this was an individual who said he was a physician was really distressing. Lastly, it does need to be pointed out that whether or not a particular death is classified as a death that is either directly or indirectly related to pregnancy or occurs during pregnancy, a child has lost a mother and the father has lost a partner, no matter who they are or where they live.
References:
(1) Malhi, Sabrina, Keating, Dan. Study says U.S. maternal death rate crisis is really a case of bad data. Washington Post, online March 13, 2024 at 7:28 p.m. EDT
(2) Joseph KS, Lisonkova S, Boutin A, Brandt JS, Schisterman EF, Anath CV, et al. Maternal mortality in the United States: are the high and rising rates due to changes in obstetrical factors, maternal medical conditions, or maternal mortality surveillance? AJOG Published on line March 12, 2024DOI:https://doi.org/10.1016/j.ajog.2023.12.038
(3) Joseph KS, Boutin A, Lisonkova S, Muraca GM, Razaz N, John S, Mehrabadi A, Sabr Y, Ananth CV, Schisterman E. Maternal Mortality in the United States: Recent Trends, Current Status, and Future Considerations. Obstet Gynecol. 2021 May 1;137(5):763-771. doi: 10.1097/AOG.0000000000004361. PMID: 33831914; PMCID: PMC8055191.
(4) Zahn, C. https://www.acog.org/news/news-releases/2024/03/despite-new-manuscript-incontrovertible-evidence-proves-unacceptably-high-us-maternal-mortality-rate