Ethnic differences in folic acid supplement use in a population-based cohort of pregnant women in Norway

RESEARCH ARTICLE Open Access E hnic di e ences in olic acid supplemen use in a popula ion-based coho o p egnan women in No way Ta ja I. Kinnunen 1 , Line Sle ne 2 , Ch is ine Somme 3 , Ma ine C. Pos 4 and Anne Ka en Jenum 5* Abs ac Backg ound: Pe i-concep ional use o olic acid supplemen s is ecommended o p e en neu al ube de ec s. Co ec supplemen use seems o be less common among e hnic mino i ies. We examined e hnic di e ences in olic acid supplemen use be o e and du ing p egnancy and possible e ec modi ica ion by educa ion o planning o p egnancy. Me hods: The pa icipan s we e 811 heal hy p egnan women om a popula ion-based coho s udy in Oslo, No way in 2008–2010. E hnici y was ca ego ized o i e g oups (Eu opean, Middle Eas e n, Sou h Asian, Eas Asian, A ican). Da a on olic acid supplemen use we e ob ained om hospi al eco ds and emaining da a by a ques ionnai e. Logis ic eg ession analyses we e adjus ed o age, pa i y, planning o p egnancy, educa ion and No wegian language skills. Resul s: Be o e p egnancy, 30.1% o Eu opean women and 7.1 o 13.6% o women in he o he e hnic g oups used olic acid supplemen s (p< 0.001). The adjus ed odds a io (OR) o supplemen use was 0.55 (95% con idence in e al 0.31; 0.96) o Sou h Asian and 0.42 (95% con idence in e al 0.19; 0.94) o Middle Eas e n women compa ed wi h Eu opean women. Du ing p egnancy, supplemen use was mos common in Eu opean women (65. 7%) and leas common in Middle Eas e n (29.4%) and A ican women (29.0%) (p< 0.001). Compa ed wi h Eu opean women, all o he e hnic g oups had lowe adjus ed odds (OR 0.30 o 0.50, p< 0.05 o all) o supplemen use among women wi h high school o less educa ion, bu no among mo e educa ed women. Planning o p egnancy did no modi y he associa ion be ween e hnici y and supplemen use. Conclusions: Few women used olic acid supplemen s be o e p egnancy. Educa ional le el modi ied he associa ion be ween e hnici y and supplemen use du ing p egnancy. Public heal h campaigns should ocus on inc easing awa eness especially in e hnic mino i y g oups wi h low educa ional le el. Keywo ds: E hnici y, P egnancy, Folic acid, Die a y supplemen s, Educa ion, Unplanned p egnancy Backg ound Neu al ube de ec s such as spina bi ida and anencephaly a e among he mos common ype o se ious bi h de ec s, occu ing when he neu al ube ails o close du ing he i s weeks o ges a ion [1, 2]. The o e all p e alence o neu al ube de ec s was 9.1 pe 10.000 bi hs in Eu ope and has no dec eased since 1991 [3]. In No way, 9.3 pe 10.000 bi hs we e a ec ed by a neu al ube de ec in 1991–2011 [3]. Folic acid supple- men a ion has been shown o educe he isk o neu al ube de ec s ema kably [4] and i may also ha e o he heal h bene i s [5–8]. Recommenda ions o pe i-concep ional olic acid sup- plemen a ion a y in Eu ope [9]. In No way, all women planning a p egnancy o who may become p egnan , a e ecommended o use olic acid supplemen s (0.4 mg/day) a leas om one mon h be o e concep ion un il h ee mon hs ges a ion [10]. I is impo an o s a supplemen- a ion weeks be o e concep ion o achie e adequa e ola e s a us by he i s ew weeks a e concep ion. Howe e , a * Co espondence: [email p o ec ed] 5 Ins i u e o Heal h and Socie y, Depa men o Gene al P ac ice, Facul y o Medicine, Uni e si y o Oslo, Oslo, No way Full lis o au ho in o ma ion is a ailable a he end o he a icle © The Au ho (s). 2017 Open Access This a icle is dis ibu ed unde he e ms o he C ea i e Commons A ibu ion 4.0 In e na ional License (h p://c ea i ecommons.o g/licenses/by/4.0/), which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o he C ea i e Commons license, and indica e i changes we e made. The C ea i e Commons Public Domain Dedica ion wai e (h p://c ea i ecommons.o g/publicdomain/ze o/1.0/) applies o he da a made a ailable in his a icle, unless o he wise s a ed. Kinnunen e al. BMC P egnancy and Childbi h (2017) 17:143 DOI 10.1186/s12884-017-1292-x coho including 22500 e hnic No wegian women om yea s 2000–2003 showed ha only 12% o he women used he supplemen s be o e p egnancy and 70% a some poin du ing p egnancy [11]. A popula ion-based c oss- sec ional s udy om Oslo, No way, epo ed ha 17% o women had s a ed using he supplemen s be o e p eg- nancy and 58% had used hem a some poin du ing p eg- nancy [12]. Ano he ala ming inding in his s udy was ha non-Wes e n immig an women we e clea ly less likely o use he supplemen s han Wes e n women (2 s. 22% be o e p egnancy, 19 s. 73% du ing p egnancy, espec i ely). Howe e , he s udy did no adjus o impo an con ounde s such as planning o p egnancy o educa ion, and ca ego iza ion o e hnici y was e y c ude. E hnic di e ences in olic acid supplemen use be o e and/o du ing p egnancy ha e been s udied in a ew o he Eu opean coun ies (Ne he lands, Belgium, I eland and he Uni ed Kingdom) [13–20]. While hese s udies included a ious e hnic g oups, he main indings we e simila showing ha supplemen use was less common among mos e hnic mino i y g oups han among he compa ison g oups. Howe e , which and how many a i- ables we e con olled o in he analyses a ied subs an- ially be ween he s udies. In hese s udies and he No wegian s udy [11], a ious o he ac o s such as low educa ion, low socio-economic s a us, poo skills in local language, young age, unplanned p egnancy, mul ipa i y and la e booking in an ena al ca e we e iden i ied as isk ac o s o no using olic acid supplemen s be o e o du ing p egnancy. None o hese s udies epo ed i hese isk ac o s modi ied he ela ionship be ween e h- nici y and supplemen use, which migh ha e signi ican implica ions o implemen a ion o he indings. Iden i y- ing possible e ec modi ie s helps o a ge public heal h messages and in e en ions o he mos ulne able subg oups no using he supplemen s. The aim o he p esen s udy was o desc ibe e hnic di - e ences in olic acid supplemen use be o e and du ing p egnancy in a popula ion-based coho o p egnan women in Oslo, No way, and whe he possible di e ences we e modi ied by educa ion o planning o p egnancy. Me hods S udy design and popula ion The s udy was o iginally se up o in es iga e p edic o s o ges a ional diabe es melli us and e al g ow h. The me hods o his popula ion-based p ospec i e coho s udy ha e been desc ibed in de ail p e iously [21]. The s udy popula ion included p egnan women who a ended p ima y ca e Child Heal h Clinics o an ena al ca e in G o uddalen, in Oslo, No way. A la ge p opo - ion o he popula ion in his a ea has e hnic mino i y backg ound and 75 o 85% o p egnan women u ilize he an ena al se ices o hese clinics. The popula ion ep esen s he main e hnic g oups li ing in Oslo. To be eligible o he s udy, women had o li e in one o h ee ci y dis ic s in G o uddalen, plan o gi e bi h a one o he wo s udy hospi als, be a ≤20 weeks’ges a- ion, be able o communica e in No wegian o any o he eigh languages o which all he in o ma ion ma e- ials and ques ionnai es we e ansla ed (A abic, English, So ani, Somali, Tamile, Tu kish, U du and Vie namese), and be able o gi e in o med w i en consen . Exclusion c i e ia we e p e-p egnancy diabe es o o he diseases equi ing in ensi e hospi al ollow-up du ing p egnancy, and pa icipa ion in he same s udy du ing a p e ious p egnancy las ing ≥22 weeks. A o al o 823 women we e ec ui ed o he s udy be- ween May 2008 and May 2010. Pa icipa ion a e was 74% among all eligible women and a ied by e hnici y (Eu opean 82%, Sou h Asian 73%, Middle Eas e n 65%, A ican 64% and Eas Asian 63%). The inal sample size was 811 women a e excluding 12 women om Cen al o Sou h Ame ica due o small g oup size. This s udy was conduc ed acco ding o he guidelines laid down in he Decla a ion o Helsinki and all p ocedu es in ol ing human subjec s we e app o ed by he Regional E hics Commi ee and he No wegian Da a inspec o a e. W i - en in o med consen was ob ained om all subjec s. Da a collec ion Ou come a iables The main ou come a iables we e olic acid supplemen use be o e and du ing p egnancy. These da a we e e ie ed om hospi al eco ds. Women we e asked ques ions on supplemen use in an in e iew a he ul asound sc eening in he hospi al a 17 o 19 week’s ges a ion, as pa o he ou ine epo o he Medical Bi h Regis y o No way. The wo ques ions elici ed whe he o no he women had used olic acid con ain- ing supplemen s be o e o du ing p egnancy by he ime o he isi . In one o he hospi als, only a “yes”op ion was included in he o m and “no”was impu ed o women wi h a missing alue (conce ns 16.6% o p e- p egnancy da a and 13.7% o p egnancy da a in he whole coho ). No da a on he exac iming, dose o du a ion o use we e ob ained o he p esen s udy. E hnici y and backg ound a iables A he inclusion isi a 15 weeks’ges a ion on a e age, he s udy midwi es in e iewed he pa icipan s and illed in a ques ionnai e elici ing ques ions on e hnic o igin and o he backg ound in o ma ion [21]. When needed, p o essional in e p e e s we e used. E hnici y was de ined based on he pa icipan ’s coun y o bi h o he mo he ’s coun y o bi h i he mo he was bo n ou side o Eu ope o No h Ame ica. Fo he p esen Kinnunen e al. BMC P egnancy and Childbi h (2017) 17:143 Page 2 o 9 s udy, e hnici y was ca ego ized o i e g oups: Eu ope (in- cluding No h Ame ica), Middle Eas (including No h A ica and Cen al Asia), A ica (excep o No h A ica), Sou h Asia and Eas Asia. Eu opean women we e used as he e e ence g oup. By e hnici y, 82.6% o hem we e No wegian, 3.4% Swedish o Danish, 2.6% o he Wes e n Eu opean (including h ee whi e No h Ame ican women) and 11.3% Eas e n Eu opean. Wes e n and Eas e n Eu opean women we e i s analyzed as sepa a e g oups, bu since hey had e y simila olic acid supplemen use and he numbe o Eas e n Eu opean women was qui e small (n= 43), he g oups we e me ged in he inal analyses. Age and weeks’ges a ion a he inclusion isi we e used as con inuous a iables. Du a ion o esidence in No way was ca ego ized o 0–1 yea s ( ecen immig an s) o ≥2 yea s o women no bo n in No way. No wegian lan- guage skills we e elici ed wi h i e esponse op ions (poo , qui e poo , a e age, qui e good o good) and we e u he ca ego ized o “poo o qui e poo ”,“a e age o qui e good”,o “good”( he alue “good”was impu ed o all e h- nic No wegians). Educa ional le el was ca ego ized as “high-school o less”(i.e. less educa ed) o “college o uni- e si y”(i.e. mo e educa ed) based on six o iginal ca ego - ies. Pa i y was ca ego ized o “nullipa ous”o “pa ous” women. The ques ion on planned p egnancy had h ee e- sponse op ions (yes, no, pa ly) which we e e-ca ego ized o “yes”o “no”(including “pa ly”). By “pa ly”we mean “no using con acep i es, wi h a wish o become p egnan , al houghno planning oge p egnan a aspeci ic ime”. S a is ical me hods Desc ip i e da a a e p esen ed as means and s anda d de i- a ions (sd) o numbe s and pe cen ages. O e all e hnic di - e ences in he c ude pe cen ages o olic acid supplemen usewe e es edusing wo-sidedχ 2 – es , sepa a ely among all women and s a i ied by educa ional le el and planning o p egnancy. E hnic di e ences in supplemen use be o e anddu ingp egnancy( hedependen a iables) we e ana- lyzed u he by logis ic eg ession models. Unadjus ed ana- lyses we e pe o med o each co a ia e. Adjus ed models we e ca ied ou using En e -me hod, adjus ing simul an- eously o age, pa i y, planning o p egnancy and educa- ionalle el(Model1).Model2wasaddi ionallyadjus ed o No wegian language skills o demons a e how inclu- sion o language skills changes he es ima es. These a i- ables we e chosen as co a ia es as hey we e associa ed wi h e hnici y in he p esen da a and wi h olic acid sup- plemen use in p e ious li e a u e. We pe o med p elimin- a y analyses in which socio-economic s a us sco e [22, 23] was used ins ead o educa ion and he esul s we e essen- ially he same. In he inal models, educa ion was used as an indica o o socio-economic s a us as hey we e s ongly associa ed and educa ional le el is easie o in e p e han he socio-economic s a us sco e [22]. To es o e ec modi ica ion be ween e hnici y and planning o p egnancy and be ween e hnici y and educa- ion, he e hnici y a iable was e-ca ego ized o ‘Eu opeans’ s. ‘non-Eu opeans’ o inc ease s a is ical powe in he ana- lyses. The p oduc e ms e hnici y*planning o p egnancy and e hnici y*educa ion we e added o each eg ession model (Model 2) one by one. Planning o p egnancy did no modi y he associa ion be ween e hnici y and supplemen use be o e o du ing p egnancy (p=0.31andp = 0.89, espec i ely, o he p oduc e ms). The p oduc e m o e hnici y*educa ion was also no s a is ically sig- ni ican o supplemen use be o e p egnancy (p=0.28). Howe e , he p oduc e m was s a is ically signi ican o e hnici y and educa ion when supplemen use du ing p eg- nancywas heou come(p= 0.013). The e o e, he adjus ed Model 2 was also s a i ied by educa ional le el. The esul s o he logis ic eg ession models a e p esen ed as odds a- ios (OR) wi h 95% con idence in e al (CI), p- alues and adjus ed R 2 ,andp- alues <0.05 we e conside ed as s a is i- cally signi ican . All analyses we e conduc ed using he SPSS s a is ical so wa e package e sion 23 (SPSS Inc., Chicago, IL, USA). Resul s Backg ound cha ac e is ics o he pa icipan s a e de- sc ibed by e hnici y in Table 1. The mean age a ied om 28 o 31 yea s be ween he e hnic g oups. Women om A ica and Eas Asia we e mos likely o be ecen immig an s and A ican women we e leas likely o ha e good No wegian language skills. Highe educa ional le el was leas common among A ican and Middle Eas e n women and mos common among Eu opean women. The pe cen age o pa ous women was lowe in Eu opean women han in he o he e hnic g oups. The pe cen age o women wi h planned p esen p egnancy was highes among Eu opean and lowes among Eas Asian women. A ican and Eas Asian women we e included in he s udy a 17.3 and 16.3 weeks’ges a ion, espec i ely, whe eas Eu opean women we e included al eady a 14.3 weeks’ges a ion on a e age. Folic acid supplemen use be o e p egnancy The use o olic acid supplemen s be o e p egnancy was uncommon in all e hnic g oups, bu Eu opean women (30.1%) used he supplemen s mo e o en han he o he e hnic g oups (7.1 o 13.6%) (Fig. 1a). When s a i ying by educa ional le el, simila e hnic di e ences we e ob- se ed a bo h le els. Supplemen use seemed o be mo e common among mo e educa ed han among less educa ed women in mos e hnic g oups. When s a i ied by planning o p egnancy, Eu opean women we e mos likely and Middle Eas e n and A ican women we e leas likely o ha e used he supplemen s be o e p egnancy among women wi h a planned p egnancy. Howe e , no Kinnunen e al. BMC P egnancy and Childbi h (2017) 17:143 Page 3 o 9 s a is ically signi ican di e ences we e obse ed in sup- plemen use be ween e hnic g oups among women wi h an unplanned p egnancy. Compa ed wi h Eu opean women, all o he e hnic g oups had dec eased odds o using he supplemen s in he unadjus ed logis ic eg essions (Table 2). Lowe age, being pa ous, no planning p egnancy, lowe educa ional le el and poo e No wegian language skills we e also as- socia ed wi h no using he supplemen s be o e p eg- nancy. Adjus men o hese a iables a enua ed he e hnic di e ences in supplemen use. In Model 1, lowe odds o using he supplemen s we e obse ed among Middle Eas e n (OR 0.26) and Sou h Asian women (OR 0.40). In Model 2, when u he adjus ed o No wegian language skills, he same e hnic g oups di e ed om he e e ence g oup, and being pa ous, no planning p eg- nancy and poo e No wegian language skills emained independen ly associa ed wi h no using he supple- men s. O all 159 women who s a ed using olic acid supplemen s be o e p egnancy, 156 (98%) con inued using hem du ing p egnancy. Folic acid supplemen use du ing p egnancy The p e alence o using olic acid supplemen s du ing p egnancy was highes among Eu opean women (65.7%) and lowes among A ican (29.0%) and Middle Eas e n women (29.4%) (Fig. 1b). S a i ica ion by educa ion showed ha hese e hnic di e ences we e s a is ically signi ican among less educa ed women only. Lowe educa ion was ela ed o no aking supplemen s among all o he e hnic g oups, bu no among Eu opeans. When s a i ying by planning o p egnancy, Eu opean women we e mo e likely o ha e used he supplemen s han all o he e hnic g oups ega dless o whe he o no he p egnancy had been planned. Compa ed o Eu opean women, all o he e hnic g oups had lowe odds o supplemen use du ing p eg- nancy in he unadjus ed model and Model 1 (Table 3). In Model 2, he ORs we e somewha close o one, bu s ill di e en om he e e ence g oup excep o he Eas Asian women. Lowe educa ion and poo e lan- guage skills we e also associa ed wi h no using he sup- plemen s. When Model 2 was s a i ied by educa ion due o e ec modi ica ion, each e hnic mino i y g oup and women wi h poo e No wegian language skills had lowe odds o supplemen use among he less educa ed women only. No e hnic di e ences we e obse ed in supplemen use among he mo e educa ed women, bu age was in e sely associa ed wi h use. Discussion This is he i s s udy o epo pe i-concep ional olic acid supplemen use in speci ied e hnic g oups in Scandina ian coun ies. Few women used olic acid sup- plemen s be o e p egnancy and he use was pa icula ly low among women o non-Eu opean o igin. A e adjus - ing o planning o he p egnancy, educa ion, No wegian language skills, and pa i y, Middle Eas e n and Sou h Asian women we e s ill less likely o use he supple- men s be o e p egnancy han Eu opean women. The Table 1 Backg ound cha ac e is ics o he pa icipan s by e hnici y, means (sd) o numbe s (%) a Eu ope n= 379 (46.7%) Sou h Asia n= 200 (24.7%) Middle Eas n= 126 (15.5%) A ica n= 62 (7.6%) Eas Asia n= 44 (5.4%) n Mean (sd) /% n Mean (sd) /% n Mean (sd) /% n Mean (sd) /% n Mean (sd) /% Age (yea s) 379 30.2 (4.5) 200 28.2 (4.6) 126 29.0 (5.5) 62 27.7 (5.2) 44 30.8 (4.9) Bo n in No way, n (%) 299 78.9 41 20.5 7 5.6 1 1.6 1 2.3 Recen immig an s b , n (%) 8 12.3 17 10.7 14 12.0 14 23.0 9 20.9 No wegian language skills, n (%) Poo o qui e poo 8 2.1 34 17.0 29 23.2 16 25.8 11 25.0 A e age o qui e good 24 6.3 80 40.0 58 46.4 31 50.0 19 43.2 Good 346 91.5 86 43.0 38 30.4 15 24.2 14 31.8 Educa ional le el, n (%) High school o less 133 35.4 136 68.3 101 81.5 54 87.1 27 61.4 College o uni e si y 243 64.6 63 31.7 23 18.5 8 12.9 17 38.6 Pa i y ≥1, n (%) 175 46.2 116 58.0 82 65.1 35 56.5 26 59.1 Planned p egnancy, n (%) 245 69.2 125 65.4 73 63.5 32 60.4 18 46.2 Weeks’ges a ion a he inclusion isi 379 14.3 (2.4) 200 15.6 (3.9) 126 15.0 (3.3) 62 17.3 (4.9) 44 16.2 (3.9) a A o al o 59 women had a missing alue o planned p egnancy and he numbe s a ied by e hnici y as ollows: Eu opeans 25 (6.6% o all Eu opeans), Sou h Asian 9 (4.5%), Middle Eas e n 11 (8.7%), Eas Asian 5 (11.4%) and A ican 9 (14.5%). The o al numbe s o missing alues o he o he a iables we e: ecen immig an s (n= 17 among women no bo n in No way), educa ional le el (n= 6), No wegian skills (n= 2) and o he a iables (n=0) b <2 yea s’ esidence in No way among women who we e no bo n in No way Kinnunen e al. BMC P egnancy and Childbi h (2017) 17:143 Page 4 o 9 main inding was ha associa ion be ween e hnici y and olic acid supplemen use du ing p egnancy was s ongly modi ied by educa ional le el. While all non-Eu opean e hnic mino i y g oups had lowe odds o supplemen use du ing p egnancy among he less educa ed women, no e hnic di e ences we e obse ed among he mo e ed- uca ed women a e adjus ing o con ounde s. Ou esul s a e mainly consis en wi h p e ious s udies on e hnic di e ences in olic acid supplemen use be o e and/o du ing p egnancy in Eu ope. Se e al s udies used a dicho omous ca ego iza ion o e hnici y and epo ed lowe use o supplemen s among non-Wes e n han among Wes e n women [12, 14, 17–19]. A compa ison o ou esul s wi h a p e ious s udy in Oslo [12] om yea 2001 shows ha e hnic di e ences in supplemen use ha e pe sis ed, bu he p e alence o use ha e sligh ly inc eased especially among e hnic mino i y women, bo h be o e and du ing p egnancy. Simila o ou esul s, la ge s udies om he Ne he lands [13] and I eland [15] compa ing se e al e hnic g oups showed ha Middle Eas e n and A ican women we e less likely o use he sup- plemen s han Wes e n o Wes e n Eu opean women. In I eland, women om Eas e n Eu ope and Sou h Ame ica also used less supplemen s han Wes e n Eu opean women [15]. A e y la ge s udy (n= 466860) om he Uni ed Kingdom ound ha Sou h Asian, O ien al (Eas Asian) and A o-Ca ibbean women we e less likely o use olic acid supplemen be o e p egnancy han Caucasian women [20]. A smalle B i ish s udy (n= 402) ound ha supplemen use was less common in Wes Indian, A ican and Asian women han in Caucasian women [16]. The p e alence o use was gene ally lowe be o e p egnancy bu highe du ing p egnancy han in ou s udy. The o he s udies did no epo compa able da a [13, 15]. Ou s udy was he second o analyze Sou h Asians and Eas Asians as sepa a e g oups. I is no ewo hy ha all he included a b Fig. 1 aand b. E hnic di e ences in olic acid supplemen use be o e and du ing p egnancy. Among all women and s a i ied by educa ional le el and planning o p egnancy, c ude pe cen ages Kinnunen e al. BMC P egnancy and Childbi h (2017) 17:143 Page 5 o 9 non-Wes e n e hnic subg oups had lowe use o he supplemen s han he Wes e n e e ence g oup in hese s udies [13, 15, 16, 20]. Howe e , he esul s o he o he s udies a e no di ec ly compa able o ou esul s due o pa ly di e en e hnic g oups and adjus men o di e en con ounde s. Ou s udy is he i s o desc ibe e ec modi ica ion be ween e hnici y and o he co a ia es in ela ion o olic acid supplemen use. We obse ed e ec modi ica- ion be ween e hnici y and educa ion and i emained s a is ically signi ican in he adjus ed models. Only he less educa ed e hnic mino i y women we e signi ican ly less likely o use supplemen s du ing p egnancy han Eu opean women. P o iciency in he No wegian lan- guage also played a ole in non-Eu opean women, sug- ges ing ha he less educa ed non-Eu opean women may no ha e known he impo ance o using he sup- plemen s in ea ly p egnancy due o language ba ie s. Mo e educa ed non-Eu opean women may ha e ound he in o ma ion om o he sou ces ega dless o hei language skills. One s udy sugges s ha he olic acid supplemen use could be mo e s ongly ela ed o p o iciency in he local language han e hnici y as such [13]. The s udy ound ha p o iciency in Du ch was he s onges de e minan o knowledge on he impo ance o olic acid supplemen use du ing he pe i- concep ional pe iod and knowledge was he s onges de e minan o ac ual use. Educa ional le el also p edic ed knowledge on olic acid supplemen s. Highe educa ion has been ela ed o he use o olic acid supplemen s also in o he s udies [14, 17, 19]. Al hough planning o p egnancy modi ied he associ- a ion be ween e hnici y and c ude p e alence o olic acid supplemen use be o e p egnancy (Fig. 1a), no s a is ically signi ican e ec modi ica ion was obse ed in he ad- jus ed models. Howe e , planning o p egnancy was s ongly associa ed wi h supplemen use be o e p egnancy ega dless o e hnici y, which has also been obse ed in some o he s udies [11, 14, 15, 24] bu no in all [13]. In e es ingly, pa ous women we e less likely o use olic acid supplemen s han nullipa ous women bo h in ou s udy and se e al p e ious s udies [11, 13–15, 18, 19, 24] al hough pa ous women should ha e hea d abou he ben- e i s o supplemen use du ing hei p e ious p egnancy. Women who ha e p e iously deli e ed heal hy child en migh i ialize he isks and be eluc an o ollow ecom- menda ions on heal hy beha io , as discussed by an Eijsden [13]. Table 2 Logis ic eg ession models o he use o olic acid supplemen s be o e p egnancy a Supplemen use s, n (%) Unadjus ed models Adjus ed Model 1 Adjus ed Model 2 n= 749, R 2 = 0.22 n= 749, R 2 = 0.22 OR (95% CI) pOR (95% CI) pOR (95% CI) p E hnici y Eu ope 114 (30.1) Re e ence Re e ence Re e ence Sou h Asia 23 (11.5) 0.30 (0.19; 0.49) <0.001 0.40 (0.23; 0.67) 0.001 0.55 (0.31; 0.96) 0.037 Middle Eas 9 (7.1) 0.18 (0.09; 0.37) <0.001 0.26 (0.12; 0.56) 0.001 0.42 (0.19; 0.94) 0.034 A ica 7 (11.3) 0.30 (0.13; 0.67) 0.003 0.42 (0.17; 1.08) 0.072 0.71 (0.26; 1.93) 0.50 Eas Asia 6 (13.6) 0.37 (0.15; 0.89) 0.027 0.60 (0.23; 1.53) 0.28 1.02 (0.38; 2.74) 0.98 Age, yea s - 1.04 (1.001; 1.08) 0.046 1.01 (0.97; 1.06) 0.62 1.02 (0.97; 1.07) 0.43 Pa i y 0 95 (25.2) Re e ence Re e ence Re e ence ≥1 64 (14.7) 0.51 (0.36; 0.73) <0.001 0.65 (0.43; 0.98) 0.04 0.62 (0.40; 0.94) 0.023 Planned p egnancy No 20 (7.7) 0.23 (0.14; 0.38) <0.001 0.26 (0.16; 0.44) <0.001 0.26 (0.15; 0.43) <0.001 Yes 132 (26.8) Re e ence Re e ence Re e ence Educa ional le el High school o less 54 (12.0) 0.32 (0.22; 0.46) <0.001 0.60 (0.38; 0.93) 0.021 0.65 (0.42; 1.02) 0.058 College o uni e si y 105 (29.7) Re e ence Re e ence Re e ence No wegian language skills Poo o qui e poo 5 (5.1) 0.15 (0.06; 0.36) <0.001 0.26 (0.10; 0.72) 0.009 A e age o qui e good 19 (9.0) 0.27 (0.16; 0.44) <0.001 0.46 (0.25; 0.83) 0.010 Good 135 (27.1) Re e ence Re e ence a Unadjus ed models we e pe o med sepa a ely o each a iable in he Table. Adjus ed Model 1 included e hnici y, age, pa i y, planning o p egnancy and educa ional le el. Adjus ed Model 2 addi ionally included No wegian language skills. S a is ically signi ican indings (p< 0.05) a e highligh ed wi h bold Kinnunen e al. BMC P egnancy and Childbi h (2017) 17:143 Page 6 o 9 Table 3 Logis ic eg ession models o he use o olic acid supplemen s du ing p egnancy a Supplemen use s, n (%) Unadjus ed models Adjus ed Model 1 Adjus ed Model 2 Adjus ed Model 2, women wi h high school o less Adjus ed Model 2, women wi h college o uni e si y n= 749, R 2 = 0.14 n= 749, R 2 = 0.17 n= 419, adj. R 2 =0.20 n= 330, adj. R 2 =0.09 OR (95% CI) pOR (95% CI) pOR (95% CI) pOR (95% CI) pOR (95% CI) p E hnici y Eu ope 249 (65.7) Re e ence Re e ence Re e ence Re e ence Re e ence Sou h Asia 84 (42.0) 0.38 (0.27; 0.54) <0.001 0.42 (0.28; 0.61) <0.001 0.59 (0.39; 0.90) 0.015 0.50 (0.28; 0.90) 0.020 0.60 (0.31; 1.19) 0.14 Middle Eas 37 (29.4) 0.22 (0.14; 0.34) <0.001 0.28 (0.22; 0.87) <0.001 0.43 (0.25; 0.73) 0.002 0.32 (0.17; 0.61) 0.001 0.71 (0.25; 2.00) 0.51 A ica 18 (29.0) 0.21 (0.12; 0.39) <0.001 0.26 (0.14; 0.50) <0.001 0.42 (0.21; 0.84) 0.014 0.31 (0.14; 0.69) 0.004 1.56 (0.25; 9.82) 0.63 Eas Asia 19 (43.2) 0.40 (0.21; 0.75) 0.004 0.44 (0.22; 0.87) 0.018 0.69 (0.33; 1.44) 0.32 0.30 (0.10; 0.86) 0.025 1.80 (0.53; 6.04) 0.35 Age, yea s - 0.99 (0.96; 1.02) 0.40 0.96 (0.92; 0.99) 0.019 0.96 (0.93; 0.998) 0.037 0.97 (0.93; 1.02) 0.28 0.94 (0.88; 0.996) 0.037 Pa i y 0 214 (56.8) Re e ence Re e ence Re e ence Re e ence Re e ence ≥1 193 (44.5) 0.61 (0.46; 0.81) <0.001 0.82 (0.59; 1.15) 0.25 0.80 (0.57; 1.12) 0.20 0.92 (0.57; 1.49) 0.73 0.65 (0.40; 1.07) 0.093 Planned p egnancy No 125 (48.3) 0.78 (0.58; 1.06) 0.11 0.88 (0.63; 1.22) 0.44 0.90 (0.64; 1.25) 0.51 0.93 (0.60; 1.44) 0.74 0.84 (0.50; 1.44) 0.53 Yes 268 (54.4) Re e ence Re e ence Re e ence Re e ence Re e ence Educa ional le el High school o less 188 (41.7) 0.45 (0.34; 0.60) <0.001 0.64 (0.45; 0.91) 0.012 0.70 (0.49; 1.000) 0.050 College o uni e si y 217 (61.3) Re e ence Re e ence Re e ence No wegian language skills Poo o qui e poo 23 (23.5) 0.18 (0.11; 0.30) <0.001 0.30 (0.17; 0.54) <0.001 0.31 (0.16; 0.59) <0.001 0.28 (0.07; 1.12) 0.072 A e age o qui e good 72 (34.0) 0.31 (0.22; 0.43) <0.001 0.51 (0.34; 0.76) 0.001 0.42 (0.25; 0.71) 0.001 0.62 (0.32; 1.22) 0.17 Good 312 (62.5) Re e ence Re e ence Re e ence Re e ence a By 17 o 19 weeks’ges a ion. Unadjus ed models we e pe o med sepa a ely o each a iable in he Table. Adjus ed Model 1 included e hnici y, age, pa i y, planning o p egnancy and educa ional le el. Adjus ed Model 2 addi ionally included No wegian language skills. S a is ically signi ican indings (p< 0.05) a e highligh ed wi h bold Kinnunen e al. BMC P egnancy and Childbi h (2017) 17:143 Page 7 o 9 The s eng hs o ou s udy include he popula ion- based coho design, he high p opo ion o e hnic mi- no i y g oups and he ela i ely high, al hough no equal, pa icipa ion a es in each e hnic g oup. The popula ion was ound o be ep esen a i e o he main e hnic g oups in Oslo, dec easing he likelihood o selec ion bias [21]. The da a on supplemen use we e collec ed by mid-p egnancy when women we e mo e likely o e- membe possible supplemen use han i he da a we e collec ed la e in p egnancy. Howe e , we do no know when he women s a ed using supplemen s and i hey s opped using hem by 12 weeks’ges a ion as ecom- mended. O he possible limi a ions we e ha we ha e no da a on subjec i e easons o nonuse, o example unawa eness on he impo ance o olic acid supple- men s. We canno exclude he possibili y ha women wi h poo No wegian language skills migh ha e misun- de s ood he ques ions on supplemen use and, al hough obliga o y, ansla o s may no always ha e been used a he hospi als, leading o a po en ial in o ma ion bias. Addi ionally, “no”was impu ed o women wi h missing da a on supplemen use om one o he pa icipa ing hospi als (14 o 17% o he o al popula ion). Howe e , his has p obably no in oduced a bias since he esul s we e essen ially simila when we used aw hospi al da a in which missing alues we e no coded as “no”. A o al o 62 women had missing da a on planned p egnancy o ano he a iable and we e he e o e no included in he adjus ed models. The pe cen ages o missing alues o planned p egnancy a ied sligh ly by e hnici y (Table 1) and women wi h missing alues migh be di e en om he o he women. We also ca ego ized educa ion in o wo b oad ca ego ies as ew Eu opeans had eally low educa ion. The e o e, wi hin he ca ego y o “less edu- ca ed women”, 89% o he Eu opean and 63% o he non-Eu opean women had inished high school and 14% o he non-Eu opean women had ≤7 yea s o educa ion only. Low numbe s in some e hnic g oups also educed s a is ical powe especially in he s a i ied analyses. Ou indings ha e se e al implica ions o clinical and public heal h p ac ice. Fi s ly, e hnic mino i y women o childbea ing-age need in o ma ion on he impo ance o using olic acid supplemen s be o e and du ing p egnancy. Secondly, while p o iciency in local language is e y im- po an and should be p omo ed among immig an s, he in o ma ion should also be a ailable o women who canno ead o unde s and he local language. The in o - ma ion could a leas be ansla ed in o se e al majo lan- guages. Thi dly, since a la ge p opo ion o p egnancies a e no planned and also a mino i y o Eu opean women use he supplemen s be o e p egnancy, public heal h mes- sages o use olic acid supplemen s should be aimed a all women o childbea ing age who may become p egnan . The in o ma ion could be dissemina ed o example on websi es p o iding o icial heal h in o ma ion and ecom- menda ions o p egnan women, a school heal h educa- ion classes, in s uden heal h ca e o ia mass media campaigns. Fou hly, pa ous women should be eminded du ing la e an ena al o pos pa um isi s abou he impo ance o olic acid supplemen use be o e possible subsequen p egnancies. Addi ionally, a po en ial, bu con- o e sial op ion would be o conside ood o i ica ion wi h ola e as some au ho s ha e sugges ed [9]. Conclusions P e-p egnancy use o olic acid supplemen s was uncom- mon in gene al and subs an ially lowe among Middle Eas e n and Sou h Asian women as compa ed o Eu o- pean women, when adjus ed o se e al con ounde s. Educa ional le el s ongly modi ied he associa ion be- ween e hnici y and olic acid supplemen use du ing p egnancy. Among less educa ed women, all non- Eu opean e hnic g oups we e less likely o use olic acid supplemen s du ing p egnancy han Eu opean women. No signi ican e hnic di e ences we e obse ed in women wi h highe educa ion. Al hough awa eness o he impo ance o olic acid supplemen use should be inc eased among all women o childbea ing age, e hnic mino i y g oups need special a en ion. Abb e ia ions CI: Con idence in e al; OR: Odds a io; Sd: S anda d de ia ion Acknowledgemen s We a e e y g a e ul o all he women pa icipa ing in he s udy and o s udy pe sonnel and o he s a a he child heal h clinics in S o ne , G o ud and Bje ke dis ic s in Oslo. We also hank he Da a Acquisi ion G oup a he Heal h Se ices Resea ch Uni a Ake shus Uni e si y Hospi al o help in e ie ing da a om hospi al eco ds. Funding The STORK G o uddalen S udy has been suppo ed by he Resea ch Council o No way, he Sou h-Eas e n No way Regional Heal h Au ho i y, No wegian Di ec o a e o Heal h and collabo a i e pa ne s in The Ci y o Oslo, S o ne , G o ud and Bje ke adminis a i e dis ic s. The unde s had no ole in he design, analysis o w i ing o his a icle. A ailabili y o da a and ma e ials The da ase s analysed du ing he cu en s udy may be a ailable om he p inciple in es iga o (Anne Ka en Jenum) on easonable eques . Au ho s’con ibu ions TIK analysed he da a and w o e i s e sion o he manusc ip , CS compiled he da ase , MCP ca ied ou p elimina y da a analyses, LS and AKJ we e esponsible o designing he s udy and da a collec ion, and AKJ is he p incipal in es iga o o he s udy. All au ho s pa icipa ed in o mula ing he esea ch ques ions, e ising he analyses and he manusc ip , and ead and app o ed he inal e sion o he manusc ip . Compe ing in e es The au ho s decla e ha hey ha e no compe ing in e es s. Consen o publica ion No applicable. Kinnunen e al. BMC P egnancy and Childbi h (2017) 17:143 Page 8 o 9 E hics app o al and consen o pa icipa e All p ocedu es in ol ing human subjec s we e app o ed by he Regional E hics Commi ee and he No wegian Da a inspec o a e. W i en in o med consen was ob ained om all subjec s. Publishe ’sNo e Sp inge Na u e emains neu al wi h ega d o ju isdic ional claims in published maps and ins i u ional a ilia ions. Au ho de ails 1 Facul y o Social Sciences /Heal h Sciences, Uni e si y o Tampe e, Tampe e, Finland. 2 Depa men o Child and Adolescen s Medicine, Ake shus Uni e si y Hospi al, Lø enskog, No way. 3 Depa men o Endoc inology, Mo bid Obesi y and P e en i e Medicine, Oslo Uni e si y Hospi al, Oslo, No way. 4 Heal h Agency, Ci y o Oslo, Oslo, No way. 5 Ins i u e o Heal h and Socie y, Depa men o Gene al P ac ice, Facul y o Medicine, Uni e si y o Oslo, Oslo, No way. Recei ed: 15 Sep embe 2016 Accep ed: 23 Ma ch 2017 Re e ences 1. Copp A, S anie P, G eene N. Neu al ube de ec s: ecen ad ances, unsol ed ques ions, and con o e sies. Lance Neu ol. 2013;12(8):799–810. 2. Yi Y, Lindemann M, Colligs A, Snowball C. Economic bu den o neu al ube de ec s and impac o p e en ion wi h olic acid: a li e a u e e iew. Eu J Pedia . 2011;170(11):1391–400. 3. Khoshnood B, Loane M, Walle H, A iola L, Addo M, Ba isic I, e al. Long e m ends in p e alence o neu al ube de ec s in Eu ope: popula ion based s udy. BMJ. 2015;351:h5949. 4. De-Regil L, Pena-Rosas J, Fe nandez-Gaxiola A, Rayco-Solon P. E ec s and sa e y o pe iconcep ional o al ola e supplemen a ion o p e en ing bi h de ec s. Coch ane Da abase Sys Re . 2015;12:CD007950. 5. Nilsen RM, Vollse SE, Rasmussen SA, Ueland PM, Dal ei AK. Folic acid and mul i i amin supplemen use and isk o placen al ab up ion: a popula ion- based egis y s udy. Am J Epidemiol. 2008;167(7):867–74. 6. Ro h C, Magnus P, Schjolbe g S, S ol enbe g C, Su en P, McKeague IW, e al. Folic acid supplemen s in p egnancy and se e e language delay in child en. JAMA. 2011;306(14):1566–73. 7. Su en P, Ro h C, B esnahan M, Haugen M, Ho nig M, Hi z D, e al. Associa ion be ween ma e nal use o olic acid supplemen s and isk o au ism spec um diso de s in child en. JAMA. 2013;309(6):570–7. 8. Me aye C, Milne E, Docke y JD, Cla el J, Pombo-de-Oli ei a MS, Wesseling C, e al. Ma e nal supplemen a ion wi h olic acid and o he i amins and isk o leukemia in o sp ing: a Childhood Leukemia In e na ional Conso ium s udy. Epidemiology. 2014;25(6):811–22. 9. Cawley S, Mullaney L, McKea ing A, Fa en M, McCa ney D, Tu ne M. A e iew o Eu opean guidelines on pe iconcep ional olic acid supplemen a ion. Eu J Clin Nu . 2016;70(2):143–54. 10. Helsedi ek o a e . Anbe alinge om kos hold, e næ ing og ysisk ak i i e . 2014. Repo No.: Rappo IS-2170. 11. Nilsen RM, Vollse SE, Gjessing HK, Magnus P, Mel ze HM, Haugen M, e al. Pa e ns and p edic o s o olic acid supplemen use among p egnan women: he No wegian Mo he and Child Coho S udy. Am J Clin Nu . 2006;84(5):1134–41. 12. B aekke K, S a A. Pe iconcep ional use o olic acid supplemen s in Oslo. Ac a Obs e Gynecol Scand. 2003;82(7):620–7. 13. an Eijsden M, an de Wal MF, Bonsel GJ. Folic acid knowledge and use in a mul i-e hnic p egnancy coho : he ole o language p o iciency. BJOG. 2006;113(12):1446–51. 14. Timme mans S, Jaddoe VW, Mackenbach JP, Ho man A, S eege s- Theunissen RP, S eege s EA. De e minan s o olic acid use in ea ly p egnancy in a mul i-e hnic u ban popula ion in The Ne he lands: he Gene a ion R s udy. P e Med. 2008;47(4):427–32. 15. McGui e M, Clea y B, Sahm L, Mu phy D. P e alence and p edic o s o pe iconcep ional olic acid up ake–p ospec i e coho s udy in an I ish u ban obs e ic popula ion. Hum Rep od. 2010;25(2):535–43. 16. B ough L, Rees G, C aw o d M, Do man E. Social and e hnic di e ences in olic acid use p econcep ion and du ing ea ly p egnancy in he UK: e ec on ma e nal ola e s a us. J Hum Nu Die . 2009;22(2):100–7. 17. Ba aka M, S eu bau S, Leemans L, Foulon W, Laubach M, Coomans D, e al. De e minan s o olic acid use in a mul i-e hnic popula ion o p egnan women: a c oss-sec ional s udy. J Pe ina Med. 2011;39(6):685–92. 18. Sikkens J, an Eijsden M, Bonsel G, Co nel M. Valida ion o sel - epo ed olic acid use in a mul ie hnic popula ion: esul s o he Ams e dam bo n child en and hei de elopmen s udy. Public Heal h Nu . 2011;14(11):2022–8. 19. Mannien J, de Jonge A, Co nel M, Spel en E, Hu on E. Fac o s associa ed wi h no using olic acid supplemen s p econcep ionally. Public Heal h Nu . 2014;17(10):2344–50. 20. Bes wick JP, Hu ly WJ, Mo is JK, Wald NJ. P e en ion o neu al ube de ec s: a c oss-sec ional s udy o he up ake o olic acid supplemen a ion in nea ly hal a million women. PLoS One. 2014;9(2):e89354. doi:10.1371/ jou nal.pone.0089354. eCollec ion 2014. 2014;9(2):e89354. 21. Jenum AK, Sle ne L, Voldne N, Vangen S, Mo k id K, Ande sen LF, e al. The STORK g o uddalen esea ch p og amme: a popula ion-based coho s udy o ges a ional diabe es, physical ac i i y, and obesi y in p egnancy in a mul ie hnic popula ion. Ra ionale, me hods, s udy popula ion, and pa icipa ion a es. Scand J Public Heal h. 2010;38(5 Suppl):60–70. 22. Somme C, Sle ne L, Jenum AK, Mo k id K, Ande sen LF, Bi keland KI, e al. E hnic di e ences in ma e nal die a y pa e ns a e la gely explained by socio-economic sco e and in eg a ion sco e: a popula ion-based s udy. Food Nu Res. 2013:57. doi:10.3402/ n . 57i0.21164. P in 2013. 23. Sle ne L, Jenum AK, Mo k id K, Vangen S, Holme IM, Bi keland KI, e al. Ma e nal li e cou se socio-economic posi ion and o sp ing body composi ion a bi h in a mul i-e hnic popula ion. Paedia Pe ina Epidemiol. 2014;28(5):445–54. 24. Nilsen RM, Leoncini E, Gas aldi P, Alleg i V, Agos ino R, Fa a elli F, e al. P e alence and de e minan s o p econcep ion olic acid use: an I alian mul icen e su ey. I al J Pedia . 2016;42(1):65. doi:10.1186/s13052-016-0278- z. 2016;42:65. • We accep p e-submission inqui ies • Ou selec o ool helps you o ind he mos ele an jou nal • We p o ide ound he clock cus ome suppo • Con enien online submission • Tho ough pee e iew • Inclusion in PubMed and all majo indexing se ices • Maximum isibili y o you esea ch Submi you manusc ip a www.biomedcen al.com/submi Submi you nex manusc ip o BioMed Cen al and we will help you a e e y s ep: Kinnunen e al. BMC P egnancy and Childbi h (2017) 17:143 Page 9 o 9