Studies - General Research

Deaths Associated with Abortion Compared to Childbirth — a Review of New and Old Data and the Medical and Legal Implications

J Contemp Health Law Policy. 2004 Spring;20(2):279-327.

Reardon DC1, Strahan TW, Thorp JM Jr, Shuping MW. 2004
PubMed PMID: 15239361 [ ]

Comment re: “The comparative safety of legal induced abortion and childbirth in the United States” by Elizabeth G Raymond, David A Grimes, Obstet Gynecol, 119 (2012)

David Reardon | Mar 05 2014 16:17 EST
The authors describe the objective of this review1 “is to provide an updated assessment of the safety of abortion relative to delivery.” But those familiar with the literature should be immediately struck by the fact that they fail to cite, much less discuss, the large record linkage studies which contradict their conclusions.

For example, our study of 173,279 low income women in California, linking medicaid records for pregnancy treatments with death certificates, revealed significantly higher rates of death associated with abortion compared to childbirth.2 Specifically, we found an elevated relative risk of death associated with abortion for all causes (RR=1.62), suicide (2.54), natural causes (1.44), circulatory diseases (2.87), and cardiovascular disease (5.46). Moreover, these effects persisted over several years.

Even more striking was the authors’ failure to note or discuss six or more even better known studies by Gissler et al examining the entire population of women in Finland3-8. They, too, used centralized records to link death certificates with treatments for pregnancy, and they, too, found significantly higher rates of mortality associated with abortion compared to childbirth.

The Finland studies are especially important in that the authors have conclusively demonstrated that the methodology used by Raymond and Grimes, a simple comparison of reported mortality rates, is unreliable.5 Without record linkage, 94% percent of deaths associated with abortion (in the first year alone) could not be identified. 4

Moreover, the methods used to collect data on deaths associated with abortion and childbirth in the United States are inconsistent and incomparable in their own right.

Indeed, in response to an inquiry about the appropriateness of comparisons of the kind used by Raymond and Grimes in this review, Dr. Julie Louise Gerberding, director of the CDC, wrote in July of 2004 that maternal mortality rates and abortion mortality rates ”are conceptually different and are used by the CDC for different public health purposes.”9

The problems with the Raymond/Grimes approach, and a more complete discussion of the findings of record linkage studies in regard to abortion mortality rates are found in 2004 review of the literature10. The bottom line is that Raymond and Grimes have chosen to ignore all the research which challenges their conclusion that abortion is safer than childbirth, and they are basing their conclusions on data sets which are not truly comparable.

It is also notable that while it is customary for review papers to consult previously published reviews to address issues raised by previous reviewers of the topic, Raymond and Grimes chose not only to ignored our review10 but also a second major review published in Obstetrical & Gynecological Survey11 which also had conclusions contrary to theirs.

This should lead readers to be skeptical of the authors claim that: “We systematically reviewed the past decade of PubMed publications for relevant data.” Indeed, in light of the body of literature which is out there, and easily found, and has often been raised in the court cases which Dr. Grimes serves as an expert witness, it is patently clear to those of us who have published in this field that the Raymond/Grimes “systematic review criteria” were carefully and precisely constructed to exclude consideration of studies published our California study (published in 2002, examining data through 1997) and those of the population of Finland.

Despite this artifice, it is an indisputable fact that every study which has employ record linkage has found that mortality rates associated with childbirth are significantly lower than those associated with abortion.2-8

Finally, while the authors can be excused for not being aware of additional record linkage studies in press at the time their “review” was published, it is noteworthy that two studies of the entire population of childbearing women in Denmark between 1998 and 2005 have also higher death rates associated with abortion compared to childbirth.

The first of these Denmark studies found that compared women who deliver a first pregnancy, women who abort a first pregnancy have a significantly elevated risk of death within the first 180 days and this elevated risk of death persists for at least ten years12. The second revealed that there is also a dose effect associated with abortion, with each exposure of abortion contributing an additional 50% (approximately) increased risk of death over the period examined.13

I sincerely hope these authors will use PubMed Commons to publish a thoughtful defense of why they excluded record linkage studies from their review. And secondly, in light of the studies mentioned here, to explain if and how they can persist in their conclusion that the best medical evidence indicates that abortion is 14 times safer than childbirth.


(1) Raymond, Elizabeth G.; Grimes, David A. The Comparative Safety of Legal Induced Abortion and Childbirth in the United States. Obstetrics & Gynecology. 119(2, Part 1):215-219, February 2012. PMID: 22270271

(2) Reardon DC, Ney PG , Scheuren FJ, Cougle JR, Coleman, PK, Strahan T. Deaths associated with pregnancy outcome: a record linkage study of low income women. Southern Medical Journal, August 2002, 95(8):834-841. PMID: 12190217

(3) Gissler M, Berg C, Bouvier-Colle MH, Buekens P. Pregnancy-associated mortality after birth, spontaneous abortion, or induced abortion in Finland, 1987-2000. Am J Obstet Gynecol. 2004 Feb;190(2):422-7. PMID: 14981384

(4) Gissler M, Berg C, Bouvier-Colle MH, Buekens P.Methods for identifying pregnancy-associated deaths: population-based data from Finland 1987-2000. Paediatr Perinat Epidemiol. 2004 Nov;18(6):448-55. PMID:

(5) Gissler M, Kauppila R, Meriläinen J, Toukomaa H, Hemminki E. Pregnancy-associated deaths in Finland 1987-1994–definition problems and benefits of record linkage. Acta Obstet Gynecol Scand. 1997 Aug;76(7):651-7. Review. PMID:

(6) Gissler M, Hemminki E. Pregnancy-related violent deaths. Scand J Public Health. 1999 Mar;27(1):54-5. PMID:

(7) Gissler M, Berg C, Bouvier-Colle MH, Buekens P. Injury deaths, suicides and homicides associated with pregnancy, Finland 1987-2000. Eur J Public Health. 2005 Oct;15(5):459-63. PMID:

(8) Gissler M, Hemminki E, Lönnqvist J. Suicides after pregnancy in Finland, 1987-94: register linkage study. BMJ. 1996 Dec 7;313(7070):1431-4.

(9) Letter from Julie Louise Gerberding to Walter Weber, July 20, 2004. Reply.pdf responding to Weber’s April 30, 2004 request for a reassessment of pertinent statistical measures of mortality rates associated with pregnancy outcome.

(10) Reardon DC, Strahan TW, Thorp JM, Shuping MW. Deaths associated with abortion compared to childbirth: a review of new and old data and the medical and legal implications. The Journal of Contemporary Health Law & Policy 2004; 20(2):279-327. PMID: 15239361

(11) Shadigian EM; Bauer ST. Pregnancy-Associated Death: A Qualitative Systematic Review of Homicide and Suicide Obstetrical & Gynecological Survey. 2005. 60:183-190.

(12) Reardon DC, Coleman PK. Short and long term mortality rates associated with first pregnancy outcome: population register based study for Denmark 1980-2004. Med Sci Monit. 2012 Sep;18(9):PH71-6. PMID: 22936199

(13) Coleman PK1, Reardon DC, Calhoun BC. Reproductive history patterns and long-term mortality rates: a Danish, population-based record linkage study. Eur J Public Health. 2013 Aug;23(4):569-74. doi: 10.1093/eurpub/cks107. Epub 2012 Sep 5. PMID: 22954474
[5 Mar 2014, ]