Perinatal outcomes after induced termination of pregnancy by methods: A nationwide register-based study of first births in Finland 1996-2013

RESEARCH ARTICLE Pe ina al ou comes a e induced e mina ion o p egnancy by me hods: A na ionwide egis e -based s udy o i s bi hs in Finland 1996–2013 Si u KC 1 *, Elina Hemminki 2 , Mika Gissle 3,4 , Su i M. Vi anen 1,5 , Reija Kleme i 6 1School o Social Sciences, Uni e si y o Tampe e, Tampe e, Finland, 2Depa men o Heal h and Social Ca e Sys ems, Na ional Ins i u e o Heal h and Wel a e, Helsinki, Finland, 3Depa men o In o ma ion Se ices, Na ional Ins i u e o Heal h and Wel a e, Helsinki, Finland, 4Depa men o Neu obiology, Ca e Sciences and Socie y, Ka olinska Ins i u e, S ockholm, Sweden, 5Depa men o Public Heal h Solu ions, Na ional Ins i u e o Heal h and Wel a e, Helsinki, Finland, 6Depa men o Wel a e, Na ional Ins i u e o Heal h and Wel a e, Helsinki, Finland *[email p o ec ed] Abs ac Backg ound Women wi h p e ious e mina ions o p egnancy (TOPs) be o e hei i s bi h ha e been associa ed wi h poo e pe ina al ou comes. Howe e , p e ious s udies on he pe ina al ou - comes by he me hod in p e ious TOPs a e inconsis en . Objec i e To examine he pe ina al ou comes o he i s - ime mo he s wi h single on bi hs, by he me hod o p e ious TOP (medical and su gical s no TOP, and su gical s medical). Me hod This is a na ionwide egis e -based s udy including 419,879 i s - ime Finnish mo he s wi h single on bi h du ing he ime pe iod 1996–2013. Mo he s ha ing hei i s bi h we e iden i- ied om he Medical Bi h Regis e and linked o he Abo ion Regis e by hei iden i ica ion numbe s. Mul inomial logis ic eg ession analysis was pe o med o examine he isk o p e- e m bi h, low bi h weigh , small o ges a ional age and pe ina al dea h by he me hod in p e ious TOPs. Resul s Among he i s - ime mo he s, 87.0% had no his o y o TOPs, 3.2% had a his o y o medical TOP(s), 9.2% had a his o y o su gical TOP(s) and 0.6% had a his o y o bo h (medical and su gical) TOP(s). No signi ican di e ences in pe ina al ou comes we e ound among he women wi h su gical TOPs, compa ed o he women wi h no TOPs. In unadjus ed analysis, inc eased odds o p e e m bi h and low bi h weigh we e ound when compa ing women ha ing p e ious su gical TOPs wi h medical TOPs. E en a e he adjus men o po en ial PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0184078 Sep embe 1, 2017 1 / 10 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 OPEN ACCESS Ci a ion: KC S, Hemminki E, Gissle M, Vi anen SM, Kleme i R (2017) Pe ina al ou comes a e induced e mina ion o p egnancy by me hods: A na ionwide egis e -based s udy o i s bi hs in Finland 1996–2013. PLoS ONE 12(9): e0184078. h ps://doi.o g/10.1371/jou nal.pone.0184078 Edi o : Co nelis B Lambalk, VU medisch cen um, NETHERLANDS Recei ed: No embe 22, 2016 Accep ed: Augus 17, 2017 Published: Sep embe 1, 2017 Copy igh : ©2017 KC e al. This is an open access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License, which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal au ho and sou ce a e c edi ed. Da a A ailabili y S a emen : Da a a e a ailable om he Na ional Ins i u e o Heal h and Wel a e and pe mission o use hem was ecei ed om he THL e hical commi ee. Con iden iali y was main ained while doing analysis. Pe mission o ge simila da a can be applied om THL Na ional Ins i u e o Heal h and Wel a e, mo e in o ma ion: se ice elephone o esea ch au ho isa ion applica ions el. +358 29 524 6677. Pe mission o ge simila da a can be applied om THL Na ional Ins i u e o Heal h and Wel a e, mo e in o ma ion: URL o esea ch au ho isa ion applica ions, con ounde s, odds o p e e m bi h <37 weeks (OR = 1.19, 95% CI = 1.04–1.36) and low bi h weigh <2500 g (OR = 1.16, 95% CI = 1.00–1.35) emained signi ican . A e es ic ing da a o he single TOP, he esul s we e simila ; OR o bo h p e e m bi h and low bi h weigh was 1.18 (95% CIs = 1.02–1.36 and 1.01–1.38). Conclusion Pe ina al ou comes did no di e among he mo he s wi h su gical TOPs compa ed o he mo he s wi h no TOPs, while he ou comes we e poo e a e su gical TOP(s) han a e medical TOP(s). In oduc ion In Eu ope, e mina ion o p egnancies (subsequen ly TOPs) a e common, and in Wes e n Eu opean coun ies mos TOPs a e pe o med be o e he i s bi h [1]. Finland has low a e o TOP, and in 2014 he a e was 8.5 pe 1,000 women aged 15–49 yea s old. The highes a e was among women aged 20–24 yea s old (16.8 pe 1,000 women), which is well below he mean age (28.6 yea s old) o a woman’s i s childbi h [2]. A e mina ion o p egnancy can be pe o med by su gical (dila a ion and u e ine e acua- ion) o medical (an ip oges in mi ep is one and misop os ol) me hods. In Eu ope, medical TOPs began in F ance in 1998 [3]. Mi ep is one ecei ed au ho iza ion in Finland in 2000. Since hen, he e ha e been inc eased use o medical e mina ion o p egnancy and i was nea ly 90% in 2014 [2]. Whe he o no an induced e mina ion o p egnancy p io o he i s bi h ad e sely in lu- ences he ou come o ha bi h has been p e iously deba ed [4–8]. The e a e e idences o an inc eased isk o p e e m bi h wi h many TOPs p io o he i s bi h [5–8], bu hese esul s e e o he ime pe iod when mos TOPs we e su gical. Howe e , some s udies did no ound an associa ion be ween p e ious TOPs and p e e m bi h/ low bi h weigh [4,9–11]. Few s udies ha e conside ed he me hod o TOPs, wi h ega d o he ou comes in subse- quen bi h [9,12–18]. Some s udies ha e epo ed a highe isk o p e e m bi h and low bi h weigh a e su gical TOPs, when compa ed o medical TOPs [15,17], bu o he s ha e ound no inc eased isk in ou comes be ween hese me hods [14,16,18]. Al hough no o he s udies ha e aken in o accoun he numbe o TOPs when compa ing hese me hods, a s udy om China [15] has epo ed an inc eased isk o p e e m bi h among hose mo he s wi h epea ed su gical TOPs, compa ed o hose mo he s wi h epea ed medical TOPs. A p e ious s udy om Finland ound an inc eased isk o poo e pe ina al ou comes a e many TOPs, howe e da a was oo scan o s udy he ou comes by he me hod o abo ion [8]. Thus he pu pose o his s udy was o examine he pe ina al ou comes o i s - ime mo he s wi h single on bi h by he me hod o TOP: medical and su gical s no TOP, and su gical s. medical TOP(s), while adjus ing o con ounding ac o s. Addi ionally, compa isons we e made be ween hose mo he s wi h only one p e ious TOP in hei ep oduc i e his o ies. Me hods The s udy was app o ed by he e hical commi ee o Na ional Ins i u e o Heal h and Wel a e (THL). A posi i e s a emen om THL e hics commi ee (22.10.2009), a posi i e s a emen wi h ega d o he amendmen o he da a, and a pe mission o use he da a we e ecei ed P e ious p egnancy e mina ion and pe ina al ou comes PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0184078 Sep embe 1, 2017 2 / 10 h ps://www. hl. i/en/web/ hl i-en/s a is ics/ in o ma ion- o - esea che s and email add ess o sec e a y, [email p o ec ed]. Funding: This wo k was suppo ed by The Finnish Cul u al Founda ion, Cen al Fund, G an numbe : 00160409, ecei ed by SK. Na ional Ins i u e o Heal h and Wel a e is he unde ha suppo ed us o publica ion. Compe ing in e es s: The au ho s ha e no compe ing in e es s. om THL (25.10.2014). The in o ma ion used in his s udy we e anonymized p io he analysis. In his popula ion-based coho s udy, we used he na ionwide Medical Bi h Regis e (MBR) and Abo ion Regis e (AR), which we e main ained by he Na ional Ins i u e o Heal h and Wel a e (THL). All he mo he s ha ing had hei i s bi h du ing he ime pe iod anging om 1996 o 2013 we e iden i ied om he MBR, and hese mo he s we e linked o he AR o de e mine he TOPs (1983–2013) hey had p io o hei i s bi h. The MBR was s a ed in 1987, and i con ains in o ma ion abou each mo he ’s backg ound cha ac e is ics, ca e du ing p egnancy and deli e y, and newbo n ca e up o 7 days o age [2]. The AR has been unc ional since 1950, and compu e ized da a a e a ailable since 1983 [19]. The egis e con ains in o ma ion on a woman’s backg ound, ges a ional age, indica ion o TOPs, da es, p ocedu es and complica ions occu ing du ing he p ocess [19]. O e all, he in o ma ion in bo h egis e s is ela i ely comple e, and he da a quali y is high [19–20]. Fo his s udy, he mo he s we e di ided in o ou s udy g oups by hei TOP his o ies and me hods: no p io TOP, medical TOPs only, su gical TOPs only and bo h ypes o TOPs. The medical TOPs included TOPs pe o med using mi ep is one alone, o in combina ion wi h misop os ol. Su gical TOPs included TOPs pe o med using ei he dila a ion and cu e age o acuum aspi a ion. Mo he s who had unde gone mul iple TOPs using bo h medical and su gi- cal me hods we e included in bo h ypes o TOPs. Only he mo he s wi h success ul TOPs we e included. The p opo ion o ailed TOPs is e y low; 0.4% in 2009–2015 acco ding o he Abo ion Regis e . The ou come measu es, ges a ional age a bi h, bi h weigh , small o ges a ional age and pe ina al dea h we e e ie ed om he MBR. The ges a ional age a bi h in he MBR is he cli- nicians’ bes es ima e a bi h, based on ul asound examina ion(s) and he da e o las men- s ua ion. The bi h was de ined as p e e m i he ges a ional age a bi h was less han 37 weeks, e y p e e m i he ges a ional age a bi h was less han 32 weeks and ex emely p e- e m i he ges a ional age was less han 28 weeks. Bi h weigh s o less han 2,500 g ams and 1,500 g ams we e de ined as low bi h weigh and e y low bi h weigh , espec i ely. Small o ges a ional age (SGA) was de ined acco ding o sex-speci ic Finnish s anda ds o newbo n in an s be ween 24 and 43 ges a ion weeks [21]. Pe ina al dea hs e e ed o s illbi hs om 22 weeks o ges a ion and ea ly neona al dea hs un il he end o he i s week a e bi h. The backg ound cha ac e is ics o women we e ecei ed om he MBR and hey e e o he ime o he bi h o he baby. The u bani y o he ma e nal municipali y o esidence was ca ego ized acco ding o S a is ics Finland, and he ca ego ies we e u he g ouped in o u ban, semi-u ban, u al and ab oad. In he MBR, ma i al s a us o mo he s was ca ego ized in o se en ca ego ies; ma ied and li ing oge he wi h spouse, egis e ed pa ne ship, ma ied and li ing sepa a ed om spouse, ne e ma ied, di o ced, widowed and unknown. These we e u he ca ego ized in o h ee g oups; ma ied/cohabi ing, unma ied/single and unknown. In his s udy, we ea ed all a iables as ca ego ical a iables. In o ma ion conce n- ing socioeconomic s a us o he mo he s was incomple e. So, ma e nal smoking and u bani y o municipali y we e used o explain he socioeconomic s a us o mo he s. The s a is ical so wa e, SPSS 23, was used o he analysis. C oss abula ions based on he s udy g oups we e calcula ed and chi-squa e es we e used o s udy s a is ical signi icance. The le el o s a is ical signi icance was se a p<0.05. Those mo he s wi h p e ious su gical TOPs and hose wi h p e ious medical TOPs we e sepa a ely compa ed o women wi h no p e ious TOP, adjus ing o di e ences in hei backg ound cha ac e is ics using a mul i a i- a e logis ic eg ession (odds a io and 95% con idence in e al). The po en ial con ounde s we e selec ed on he basis o he p e ious li e a u e on he ma e nal isk ac o s o bi h ou - comes, and hei a ailabili y and quali y in he egis e s. Mo he s wi h p e ious TOPs (su gical P e ious p egnancy e mina ion and pe ina al ou comes PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0184078 Sep embe 1, 2017 3 / 10 o medical) we e compa ed o hose wi hou p e ious TOP, a e adjus ing o he ma e nal age, ma i al s a us, u bani y o he municipali y o esidence, his o y o smoking du ing p eg- nancy and he yea o childbi h. Mo he s wi h p e ious su gical TOPs we e compa ed o hose wi h p e ious medical TOPs, a e adjus ing o he ma e nal age, ma i al s a us, u bani y o he municipali y o esidence, his o y o smoking du ing p egnancy and he yea o childbi h. In he second model, addi ional adjus men s (numbe o p e ious TOPs, ges a ional age a TOP and yea o las TOP) we e added. Since he numbe o women ha ing had bo h medical and su gical TOPs was small when compa ed o he o he g oups, hese mo he s we e excluded om he eg ession analysis. Las ly, a sub-analysis o hose mo he s ha ing had only one p e ious su gical o medical TOP was conduc ed, a e adjus ing o he same con ounde s as abo e. Resul s A o al o 419,879 i s - ime mo he s ha ing had single on bi h om 1996–2013 we e iden i- ied om he MBR. Acco ding o he AR, 365,356 (87.0%) o he mo he s had no his o y o TOP, 13,450 (3.2%) had his o ies o medical TOP(s), 38,659 (9.2%) had his o ies o su gical TOP(s) and 2,414 (0.6%) had his o ies o bo h medical and su gical TOP(s). The backg ound cha ac e is ics di e ed wi h ega d o se e al aspec s be ween hese subg oups (Table 1). The mo he s wi h his o ies o p e ious TOPs we e mo e o en younge , single, u ban esiden s and smoke s han he mo he s wi hou p e ious TOP. When compa ed o he mo he s wi h p e i- ous medical TOPs, he mo he s wi h p e ious su gical TOPs had mo e epea ed TOPs, had hei las TOP in ea lie yea s, and he ime di e ence be ween hei i s bi h and las TOP was longe (Table 2). When compa ed o hose mo he s wi h no p e ious TOP, he pe ina al ou comes we e poo e among hose mo he s wi h p e ious su gical o bo h ypes o TOPs, bu no among he mo he s wi h p e ious medical TOPs only (Table 3). The incidence o p e e m bi h was lowe among he mo he s wi h p e ious medical TOPs, when compa ed o hose mo he s wi hou p e ious TOPs. The unadjus ed logis ic eg ession analysis showed inc eased isk o all ypes o p e e m bi h and low bi h weigh a e su gical TOPs and dec eased isk o all s udied pe ina al ou - comes a e medical TOPs, when compa ed o he mo he s wi hou p e ious TOP (Table 4). Compa ed o he mo he s wi h p e ious medical TOPs, he mo he s wi h p e ious su gical TOPs had inc eased isk o all s udied ou comes, wi h he excep ion o SGA and pe ina al dea h. A e adjus ing o he sociodemog aphic ac o s, he mo he s wi h p e ious medical TOPs had dec eased isk o p e e m bi h and low bi h weigh when compa ed o he mo he s wi h no p e ious TOPs (Table 4). The inc eased isk o ad e se ou comes when compa ing he mo he s wi h su gical TOPs o he mo he s wi h no TOPs, did no emain signi ican , a e con olling o backg ound cha ac e is ics. Howe e , mo he s wi h su gical TOPs had ma gin- ally inc eased isk o SGA compa ed o he mo he s wi h no TOPs. The mo he s wi h p e ious su gical TOPs had highe isk o p e e m bi h and newbo n wi h low bi h weigh han hose mo he s wi h p e ious medical TOPs (Table 4). A e an addi ional adjus men o he numbe o p e ious TOPs, he ges a ional age a he ime o TOP, yea o he las TOP, isk o p e e m bi h (<37 weeks) and isk o newbo n wi h low bi h weigh emained signi ican (Table 4). A e es ic ing he analysis o hose mo he s ha ing had only one p e ious TOP, and adjus ing o con ounde s, he esul s did no change: he mo he s ha ing had only one P e ious p egnancy e mina ion and pe ina al ou comes PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0184078 Sep embe 1, 2017 4 / 10 p e ious su gical TOP had inc eased isks o p e e m bi h and low bi h weigh when com- pa ed o he mo he s wi h only one p e ious medical TOP (Table 4). Discussion The i s - ime mo he s wi h p e ious TOP we e much younge , single and mo e o en smoke s han he mo he s wi hou p e ious TOP. In addi ion, he mo he s wi h p e ious medical TOPs had a educed isk o p e e m bi h and low bi h weigh when compa ed wi h he mo he s wi h no p e ious TOP. All he poo ou comes measu ed, wi h he excep ion o small o ges a- ional age and pe ina al dea hs, we e mo e common among he mo he s wi h p e ious su gical TOPs han among he mo he s wi h p e ious medical TOPs. This was also ue among hose mo he s who had gone h ough only one TOP be o e hei i s bi h. Ou na ionwide s udy co e ed all he i s - ime mo he s ha ing had single on bi hs du ing he ime pe iod anging om 1996 o 2013, and all o he TOPs pe o med in Finland du ing he ime pe iod anging om 1983 o 2013. Because ou da a did no include TOPs be o e 1983, some women migh ha e been classi ied w ongly in o he “no TOP g oup”. Howe e , we assume ha he e a e only ew such cases and his will hus no a ec ou esul s. The quali y Table 1. Backg ound cha ac e is ics o Finnish i s - ime mo he s in 1996–2013 by p e ious TOPs and me hods. Cha ac e is ics No p io TOP (n = 365356) Medical TOP (n = 13450) Su gical TOP (n = 38659) Bo h (n = 2414) To al (n = 419879) P- alue* % % % % % Ma e nal age Mean (SD) 27.4 (5.2) 26.1 (5.3) 28.2 (5.5) 27.4 (5.5) 27.4 (5.2) <0.0001 19 5.8 8.3 4.6 3.6 5.8 20–24 24.5 35.7 22.6 28.1 24.7 25–29 36.9 31.4 32.7 32.6 36.3 30–34 23.6 17.2 26.3 23.2 23.7 35–39 7.7 5.9 11.2 9.6 7.9 40 1.5 1.4 2.6 2.9 1.6 Ma i al s a us Ma ied/cohabi ing 86.0 78.6 77.5 74.4 84.9 <0.0001 Unma ied/Single 12.6 20.8 20.3 25.1 13.7 Unknown 1.4 0.6 2.2 0.6 1.4 Type o esidence U ban 71.1 73.3 73.3 76.0 71.4 <0.0001 Semi-u ban 14.2 14.1 14.1 12.2 14.1 Ru al 14.1 11.9 13.1 11.2 13.9 Ab oad 0.6 0.7 0.4 0.5 0.6 Smoking s a us No smoking 82.7 63.5 67.6 54.8 80.5 <0.0001 S opped smoking in hi d imes e 4.8 11.9 6.3 11.6 5.2 Smoked a e i s imes e 10.5 22.6 23.8 31.5 12.3 Bi h yea o child 1996–2000 26.7 3.9 36.1 7.9 26.7 <0.0001 2001–2005 27.2 12.4 33.2 18.5 27.2 2006–2010 29.1 43.9 22.9 45.3 29.1 2011–2013 17.0 39.8 7.9 28.3 16.9 *P- alue om chi squa e es and s a is ical signi icance a he le el o 0.05 h ps://doi.o g/10.1371/jou nal.pone.0184078. 001 P e ious p egnancy e mina ion and pe ina al ou comes PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0184078 Sep embe 1, 2017 5 / 10 o he da a om he THL and AR is conside ed o be e y high and eliable [19,20], and ea lie s udies ha e compa ed he in o ma ion in he medical eco ds wi h he AR and ound ha 95% o he in o ma ion ma ched and 99% da a co e age [19,21]. Ou la ge da a se enabled us o s udy he ou comes in di e en subg oups, based on he me hod o e mina ing p egnancy. E en hough we we e able o s udy he ou comes by he me hod we we e no able o sepa a e di e en medical (mi ep is one/misop os ol) o su gical me hods (dila a ion, cu e age o ac- uum aspi a ion) because he egis e do no con ain as de ailed in o ma ion. Fu he mo e, his is an obse a ional s udy and i canno p o ide e idences o causali y. O e all, ou indings ha he demog aphic and ep oduc i e p o iles o i s - ime mo he s wi h his o ies o p e ious TOPs di e ed om hose o i s - ime mo he s wi hou his o ies o p e ious TOPs a e in line wi h p e ious s udies [8,11,22,23]. In he mul i a ia e logis ic eg es- sion analysis, we we e able o adjus o se e al backg ound a iables; howe e , we could no adjus o he socioeconomic posi ion o he mo he s due o incomple e da a (da a no shown). Table 2. His o y o TOPs o Finnish i s - ime mo he s in 1996–2013 by hei me hods. His o y o TOPs Medical TOP (n = 13450) Su gical TOP (n = 38659) Bo h TOP (n = 2414) To al (n = 54523) P- alue* % % % % Numbe o TOPs 1 90.6 87.5 0.0 84.4 <0.0001 2 8.1 10.5 71.3 12.6 3 1.3 2.0 28.7 3.0 Ges a ional age a TOP <12 weeks 77.0 90.4 61.4 85.8 <0.0001 12 weeks 23.0 9.6 38.6 14.2 Yea o las TOP 1987–1994 3.3 33.1 4.9 23.9 <0.0001 1995–1998 3.3 29.4 7.2 21.4 1999–2003 25.0 28.2 28.9 27.4 2004–2013 68.3 9.2 59.1 27.3 Di e ence be ween i s bi h and las TOP 5–38 mon hs 50.2 24.6 53.4 32.8 <0.0001 39–81 mon hs 35.4 33.0 34.0 33.7 82–310 mon hs 14.4 42.4 12.6 33.5 *P- alue om chi squa e es and s a is ical signi icance a he le el o 0.05 h ps://doi.o g/10.1371/jou nal.pone.0184078. 002 Table 3. Incidence o pe ina al ou comes among he Finnish i s - ime mo he s in 1996–2013 by p e ious TOPs and me hods. Pe ina al ou comes No p io TOP (n = 365356) Medical TOP (n = 13450) Su gical TOP (n = 38659) Bo h (n = 2414) To al (n = 419879) n /1000 n /1000 n /1000 n /1000 n /1000 Ex emely p e e m bi h <28 weeks 1122 3 38 3 154 4 11 5 1325 3 Ve y p e e m bi h <32 weeks 1843 5 57 4 232 6 10 4 2142 5 P e e m bi h <37 weeks 17041 47 515 38 1882 49 105 44 19542 47 Ve y low bi h weigh <1500 g ams 2805 8 90 7 341 9 20 8 3256 8 Low bi h weigh <2500 g ams 12439 34 417 31 1447 37 88 36 14391 34 Small o ges a ional age 17897 49 687 51 2047 53 136 56 20767 49 Pe ina al dea h 1622 4 58 4 206 5 13 5 1899 5 h ps://doi.o g/10.1371/jou nal.pone.0184078. 003 P e ious p egnancy e mina ion and pe ina al ou comes PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0184078 Sep embe 1, 2017 6 / 10 In Finland, smoking has been ound o be a good p oxy o he socioeconomic posi ion [24]; he e o e, we used u bani y o municipali y and mo he ’s smoking ins ead. P e iously, i has been epo ed ha unde going se e al TOPs be o e a woman’s i s bi h co ela ed wi h poo e pe ina al ou comes in subsequen bi hs [5–8]. Howe e , we compa ed he mo he s wi h his o ies o su gical and medical TOPs, bu adjus ed o he numbe o p e i- ous TOPs. Mo eo e , we conduc ed a subg oup analysis o hose ha ing had only one su gical o medical TOP. We also adjus ed o he ges a ional age a he ime o TOP and he yea o he las TOP, which has no been done in p e ious s udies [12,15,17,25]. Table 4. C ude and adjus ed ORs and 95% con idence in e als o pe ina al ou comes o Finnish i s - ime mo he s acco ding o he me hod o TOP. Pe ina al ou comes Odds Ra ios and 95% Con idence In e al Medical s no TOP 1 Su gical s no TOP 1 Su gical s medical TOP 2 Su gical s medical TOP, only one TOP 3 OR 95% CI OR 95% CI OR 95% CI OR 95% CI Ex emely p e e m bi h <28 weeks C ude Model 0.91 0.66–1.26 1.30 1.10–1.54 1.43 0.99–2.04 1.32 0.91–1.92 Adjus ed Model I 0.94 0.68–1.30 1.16 0.98–1.38 1.20 0.79–1.81 1.12 0.73–1.73 Adjus ed Model II 1.43 0.90–2.30 1.39 0.84–2.30 Ve y p e e m bi h <32 weeks C ude Model 0.83 0.64–1.08 1.19 1.04–1.37 1.44 1.07–1.92 1.33 0.98–1.80 Adjus ed Model I 0.88 0.67–1.15 1.03 0.89–1.18 1.23 0.87–1.71 1.13 0.80–1.61 Adjus ed Model II 1.24 0.85–1.81 1.23 0.83–1.84 P e e m bi hs <37 weeks C ude Model 0.81 0.74–0.89 1.05 1.00–1.10 1.29 1.17–1.43 1.26 1.13–1.40 Adjus ed Model I 0.83 0.76–0.91 1.00 0.95–1.05 1.16 1.03–1.30 1.15 1.02–1.29 Adjus ed Model II 1.19 1.04–1.36 1.18 1.02–1.36 Ve y low bi h weigh <1500 g ams C ude Model 0.87 0.70–1.07 1.15 1.03–1.29 1.33 1.05–1.68 1.25 0.98–1.60 Adjus ed Model I 0.88 0.71–1.10 1.00 0.89–1.12 1.16 0.89–1.52 1.11 0.84–1.47 Adjus ed Model II 1.27 0.93–1.73 1.27 0.91–1.76 Low bi h weigh <2500 g ams C ude Model 0.91 0.82–1.00 1.10 1.04–1.17 1.22 1.09–1.36 1.21 1.08–1.36 Adjus ed Model I 0.86 0.78–0.95 0.98 0.93–1.04 1.16 1.02–1.32 1.16 1.01–1.32 Adjus ed Model II 1.16 1.00–1.35 1.18 1.01–1.38 Small o ges a ional age C ude Model 0.96 0.88–1.03 0.92 0.88–0.97 0.96 0.88–1.05 0.96 0.88–1.05 Adjus ed Model I 1.05 0.97–1.14 1.07 1.02–1.12 0.99 0.89–1.10 0.99 0.89–1.10 Adjus ed Model II 1.00 0.89–1.13 1.01 0.89–1.14 Pe ina al dea h C ude Model 1.03 0.79–1.34 0.83 0.72–0.96 0.81 0.60–1.08 0.81 0.60–1.08 Adjus ed Model I 0.97 0.74–1.27 0.98 0.85–1.14 1.13 0.80–1.60 1.13 0.80–1.60 Adjus ed Model II 1.00 0.67–1.48 1.00 0.67–1.48 Adjus ed Model I- adjus ed o socio demog aphic ac o s; ma e nal age, ma i al s a us o mo he s, a ea o esidence, smoking s a us and yea o child bi h Adjus ed Model II- adjus ed o numbe o p e ious TOPs, ges a ional age a TOP and he yea o las TOP 1 No TOP g oup is used as e e ence g oup 2 Medical g oup is used as e e ence g oup 3 Medical g oup is used as e e ence g oup and includes mo he s wi h only one medical and one su gical TOP h ps://doi.o g/10.1371/jou nal.pone.0184078. 004 P e ious p egnancy e mina ion and pe ina al ou comes PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0184078 Sep embe 1, 2017 7 / 10 The e a e ew s udies examining he long- e m consequences o medical TOPs [14,16]. Howe e , ou inding o a educed isk o p e e m bi hs among he mo he s wi h p e ious medical TOPs, when compa ed o he mo he s wi h no p e ious TOP, is consis en wi h a p e- ious s udy om China [13]. In addi ion, some o he s udies [14,15,17] indi ec ly suppo ou inding o no inc eased isk o p e e m bi hs among he mo he s wi h his o y o medical TOPs. Ha ing had a TOP e lec s e ili y, and his may explain he be e ou comes among hose mo he s ha ing had medical TOPs compa ed o he mo he s wi hou his o y o TOPs. Poo e ou comes a e su gical TOPs migh be due o he eason ha he medical TOPs cause less physical auma o he ce ix and he less endome ial damage han he su gical TOPs [6,13,26,27]. As in some p io s udies, ou unadjus ed esul s showed an inc eased isk o p e e m bi hs among hose mo he s wi h su gical TOPs, when compa ed wi h hose mo he s ha ing had no p io TOPs [15,17,18,25]. Howe e , in ou s udy, he signi icance was los a e adjus ing o he sociodemog aphic ac o s. Simila o some p e ious s udies [9,15,17,25], we ound a highe isk o p e e m bi hs among he mo he s wi h p e ious su gical TOPs, when compa ed o he mo he s wi h p e i- ous medical TOPs. Medical TOPs may cause less ha m o he u e us han su gical TOPs, which can esul in be e bi h ou comes la e [13,27]. A ecen e iew and me a-analysis om 21 s udies also suppo s ou indings wi h ega d o he associa ion be ween p e e m bi hs and su gical TOPs [28]. In con as , some p e ious s udies ha e no ound an inc eased isk o p e e m bi hs in subsequen bi hs among mo he s wi h p e ious su gical TOPs, when compa ed o mo he s wi h p e ious medical TOPs [11,14,29]. Howe e , some o hose s udies did no con ol o po en ial con ounde s, and some we e based on sel - epo ed TOPs, which may in oduce ecall bias [13,14]. Con a y o some o he p e ious esea ch, ou s udy epo ed an inc eased isk o low- bi h weigh among he mo he s wi h p e ious su gical TOPs, when compa ed o he mo he s wi h p e ious medical TOPs [12–14,16]. Howe e , ew s udies ound a posi i e associa ion be ween su gical TOPs and he isk o low bi h weigh [4,9], which migh suppo ou indings. Conclusion Pe ina al ou comes did no di e among he mo he s wi h su gical TOPs compa ed o he mo he s wi h no TOPs, while he ou comes we e poo e a e su gical TOP(s) han a e medi- cal TOP(s). I is impo an o s udy he e ec s o he di e en me hods used o e mina ing p egnancy o de e mine he sa es me hod. This could be o impo ance o heal hca e p o es- sionals in e ms o clinical decision making and counselling women seeking e mina ion o p egnancy, wi h espec o he me hod used o e mina ion. Au ho Con ibu ions Concep ualiza ion: Si u KC, Elina Hemminki, Mika Gissle , Reija Kleme i. Da a cu a ion: Mika Gissle . Fo mal analysis: Si u KC. Funding acquisi ion: Si u KC, Reija Kleme i. In es iga ion: Si u KC, Reija Kleme i. Me hodology: Si u KC, Reija Kleme i. P ojec adminis a ion: Si u KC, Mika Gissle , Reija Kleme i. P e ious p egnancy e mina ion and pe ina al ou comes PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0184078 Sep embe 1, 2017 8 / 10 Resou ces: Si u KC, Su i M. Vi anen, Reija Kleme i. So wa e: Si u KC. Supe ision: Su i M. Vi anen, Reija Kleme i. Valida ion: Si u KC. Visualiza ion: Si u KC, Elina Hemminki, Mika Gissle , Reija Kleme i. W i ing – o iginal d a : Si u KC. W i ing – e iew & edi ing: Si u KC, Elina Hemminki, Mika Gissle , Su i M. Vi anen, Reija Kleme i. Re e ences 1. Gissle M, F on ei a I, Jahn A, Ka o H, Mo eau C, Oli ei a da Sil a M, e al. Te mina ions o p egnancy in he Eu opean Union. BJOG. 2012; 119(3): 324–32. h ps://doi.o g/10.1111/j.1471-0528.2011.03189. xPMID: 22129480 2. Na ional Ins i u e o Heal h and Wel a e. Induced abo ion. Induced abo ions 2015. h ps://www. hl. i/ en/web/ hl i-en/s a is ics/s a is ics-by- opic/sexual-and- ep oduc i e-heal h/abo ions/induced- abo ions. Accessed on 2 Ap il, 2016. 3. C einin MD. Medical abo ion egimens: His o ical con ex and o e iew. Am J Obs e Gynaecol. 2000; 183: 3–9. 4. Zhou W, So ensen HT, Olsen J. Induced abo ion and low bi h weigh in he ollowing p egnancy. In J Epidemiol. 2000; 29: 100–106. PMID: 10750610 5. Hen ie L, Kaminski M. Impac o induced abo ion on subsequen p egnancy ou come: he 1995 F ench na ional pe ina al su ey. BJOG. 2001; 108(10):1036–42. PMID: 11702834 6. Ancel PY, Lelong N, Papie nik E, ApheSau el- Cubizolles MJ, Kaminski M. His o y o induced abo ion as a isk ac o o p e e m bi h in Eu opean coun ies: esul s o he EUROPOP su ey. Hum Rep od. 2004; 19(7):734–40. 7. Shah PS, Zao J. Induced e mina ion o p egnancy and low bi h weigh and p e e m bi h: a sys ema ic e iew and me a-analyses. BJOG. 2009; 116(11): 1425–42. h ps://doi.o g/10.1111/j.1471-0528.2009. 02278.x PMID: 19769749 8. Kleme i R, Gissle M, Niinima ¨ki M, Hemminki E. Bi h ou comes a e induced abo ion: a na ionwide egis e -based s udy o i s bi hs in Finland. Hum Rep od. 2012; 27(11): 3315–20. 9. A ash HK, Hogue CJ. 11 The e ec o p egnancy e mina ion on u u e ep oduc ion. Baillie es Clin Obs e Gynaecol. 1990; 4(2): 391–405. 10. F ank PI, McNamee R, Hanna o d PC, Kay CR, Hi sch S. The e ec o induced abo ion on subsequen p egnancy ou come. B J Obs e Gynaecol. 1991; 98(10): 1015–24. PMID: 1751433 11. Raa ikainen K, Heiskanen N, Heinonen S. Induced abo ion: no an independen ou come o p egnancy ou come, bu a challenge o heal h counseling. Ann Epidemiol. 2006; 16(8): 587–92. h ps://doi.o g/10. 1016/j.annepidem.2006.01.007 PMID: 16621599 12. Yimin C, Wei Y, Weidong C, Xianmi W, Junqing W, Lin L. Mi ep is one-induced abo ion and bi h weigh in he i s subsequen p egnancy. In J Gynaecol Obs e . 2004; 84(3): 229–35. h ps://doi.o g/ 10.1016/S0020-7292(03)00338-2 PMID: 15001370 13. Chen A, Yuan W, Mei ik O, Wang X, Wu SZ, Zhou L, Luo l, Gao E sheng, Cheng Y. Mi ep is one- induced ea ly abo ion and ou come o subsequen wan ed p egnancy. Am J Epidemiol. 2004; 160(2): 110–17. h ps://doi.o g/10.1093/aje/kwh182 PMID: 15234931 14. Vi k J, Zhang J, Olsen J. Medical abo ion and he isk o subsequen ad e se p egnancy ou comes. N Engl J Med. 2007; 357(7): 648–53. h ps://doi.o g/10.1056/NEJMoa070445 PMID: 17699814 15. Liao H, Wei Q, Duan L, Ge J, Zhou Y, Zeng W. Repea ed medical abo ions and isk o p e e m bi h in he subsequen p egnancy. A ch Gynaecol Obs e . 2011; 284(3): 579–86. 16. Mannis o J, Men ula M, Bloigu A, Hemminki E, Gissle M, Heikinheimo O, Niinimaki M. Medical e sus su gical e mina ion o p egnancy in p imig a id women- is he nex deli e y di e en ly a isk? A popula- ion-based egis e s udy. BJOG. 2013; 120: 331–37. h ps://doi.o g/10.1111/1471-0528.12034 PMID: 23126244 P e ious p egnancy e mina ion and pe ina al ou comes PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0184078 Sep embe 1, 2017 9 / 10