-
Views
-
Cite
Cite
R. Klemetti, M. Gissler, M. Niinimäki, E. Hemminki, Birth outcomes after induced abortion: a nationwide register-based study of first births in Finland, Human Reproduction, Volume 27, Issue 11, November 2012, Pages 3315–3320, https://doi.org/10.1093/humrep/des294
Close - Share Icon Share
Abstract
Is the perinatal health of first-born children affected by the mother's previous induced abortion(s) (IAs)?
Prior IAs, particularly repeat IAs, are correlated with an increased risk of some health problems at first birth; even in a country with good health care quality.
A positive association between IA and risk of preterm birth or a dose–response effect has been found in some previous studies. Limited information and conflicting results on other infant outcomes are available.
Nationwide register-based study including 300 858 first-time mothers during 1996–2008 in Finland.
All the first-time mothers with a singleton birth (obtained from the Medical Birth Register) in the period 1996–2008 (n = 300 858) were linked to the Abortion Register for the period 1983–2008.
Of the first-time mothers, 10.3% (n = 31 083) had one, 1.5% had two and 0.3% had three or more IAs. Most IAs were surgical (88%) performed before 12 weeks (91%) and carried out for social reasons (97%). After adjustment, perinatal deaths and very preterm birth (<28 gestational week) suggested worse outcomes after IA. Increased odds for very preterm birth were seen in all the subgroups and exhibited a dose–response relationship: 1.19 [95% confidence interval (CI) 0.98–1.44] after one IA, 1.69 (1.14–2.51) after two and 2.78 (1.48–5.24) after three IAs. Increased odds for preterm birth (<37 weeks) and low birthweight (<2500 g and <1500 g) were seen only among mothers with three or more IAs: 1.35 (1.07–1.71), 1.43 (1.12–1.84) and 2.25 (1.43–3.52), respectively.
Observational studies like ours, however large and well-controlled, will not prove causality.
In terms of public health and practical implications, health education should contain information of the potential health hazards of repeat IAs, including very preterm birth and low birthweight in subsequent pregnancies. Health care professionals should be informed about the potential risks of repeat IAs on infant outcomes in subsequent pregnancy.
National Institute for Health and Welfare and the Academy of Finland. No competing interests.
- pregnancy
- quality of care
- conflict of interest
- abortion, induced
- child
- finland
- health personnel
- infant
- low birth weight infant
- mothers
- personal satisfaction
- social role
- social welfare
- surgical procedures, operative
- health education
- public health medicine
- premature birth
- perinatal period
- birth
- funding
- first birth
- birth outcome
Comments
Sir,
We read with great interest the recently published nationwide register-based study of first births in Finland, which concluded that sexuality education should contain information on the potential hazards of repeat induced abortion (IA), including very preterm birth and low birthweight in subsequent pregnancies. The authors adjusted for 7 confounders, and found a significant association of several adverse pregnancy outcomes when comparing women with 0 vs 3 IA's. With the exception of very preterm birth, no dose- response relationship was demonstrated.
We found it interesting that a similar study in 2006 by Raatikanien et al using data from a representative region of Finland found no association between IA and birth outcomes. One explanation may be that both multiparas and primaparas were included in Raatikanien et al's analysis. A second is that the 2006 analysis adjusted for 14 confounders, notably maternal medical conditions such as diabetes and toxemia - two major factors related to both spontaneous and indicated preterm birth. A Cochrane review of risk factors for preterm birth identified many of the same factors as important in risk scoring systems. We are concerned that important potential confounders such as these were not adjusted for in the current study.
The authors rightly noted that observational studies can only report associations, not causality and we wonder if the association between multiple IA's and preterm births persists once medical comorbidities and other factors are controlled for. The population-based database with which the authors have the privilege of working may contain valuable, yet untapped information that may shed further light on this interesting and important question.
Thank you,
Colleen M. Krajewski, Roxanne Jamshidi, Anne E. Burke, Jennifer A. Robinson, Michelle C. Fox
Division of Family Planning, Johns Hopkins Bayview Medical Center, Johns Hopkins Medical Institutions, Baltimore MD
Conflict of Interest:
Drs. Burke, Fox, and Jamshidi provide training for Nexplanon insertion. Dr. Krajewski (corresponding author) has no competing interests.