Durham Research Online Deposited in DRO: 05 January 2009 Version of attached file: Accepted Version Peer-review status of attached file: Peer-reviewed Citation for published item: Ball, H. L. (2006) ’Parent-infant bed-sharing behavior : effects of feeding type, and presence of father.’, Human nature : an interdisciplinary biosocial perspective., 17 (3). pp. 301-318. Further information on publisher’s website: http://dx.doi.org/10.1007/s12110-006-1011-1 Publisher’s copyright statement: The original publication is available at www.springerlink.com Use policy The full-text may be used and/or reproduced, and given to third parties in any format or medium, without prior permission or charge, for personal research or study, educational, or not-for-profit purposes provided that: • a full bibliographic reference is made to the original source • a link is made to the metadata record in DRO • the full-text is not changed in any way The full-text must not be sold in any format or medium without the formal permission of the copyright holders. Please consult the full DRO policy for further details. Durham University Library, Stockton Road, Durham DH1 3LY, United Kingdom Tel : +44 (0)191 334 3042 — Fax : +44 (0)191 334 2971 http://dro.dur.ac.uk Durham Research Online Deposited in DRO: 05 January 2009 Version of attached file: Accepted Peer-review status of attached file: Peer-reviewed Citation for published item: Ball, H. L. (2006) 'Parent-infant bed-sharing behavior : effects of feeding type, and presence of father.', Human nature : an interdisciplinary biosocial perspective., 17 (3), pp. 301-318. Further information on publishers website: http://dx.doi.org/10.1007/s12110-006-1011-1 Publishers copyright statement: The original publication is available at www.springerlink.com Use policy The full-text may be used and/or reproduced, and given to third parties in any format or medium, without prior permission or charge, for personal research or study, educational, or not-for-profit purposes provided that : ƒ ƒ ƒ a full bibliographic reference is made to the original source a link is made to the metadata record in DRO the full-text is not changed in any way The full-text must not be sold in any format or medium without the formal permission of the copyright holders. Please consult the full DRO policy for further details. Durham University Library, Stockton Road, Durham DH1 3LY, United Kingdom Tel : +44 (0)191 334 2975 | Fax : +44 (0)191 334 2971 http://dro.dur.ac.uk Ba ll: Be d -sha ring b e ha vio ur Fo rthc o m ing in Huma n Na ture 2006 Pa re nt- Infa nt Be d- sha ring Be ha vio ur: e ffe c ts o f fe e ding type , a nd pre se nc e o f fa the r. Dr He le n L. Ba ll, Pa re nt-Infa nt Sle e p La b ., De p a rtme nt o f Anthro p o lo g y, Unive rsity o f Durha m , DH1 3HN UK . +44 191 334 0260 h.l.b a ll@ d ur.a c .uk KEYWO RDS: BED-SHARING , BREASTFEEDING , FO RMULA-FEEDING , MO THER-INFANT BEHAVIO UR, INFANT SLEEP A BSTRA C T An e vo lutio na rily info rme d p e rsp e c tive o n p a re nt-infa nt sle e p c o nta c t c ha lle ng e s re c o m m e nd a tio ns re g a rd ing a p p ro p ria te p a re nt-infa nt sle e p p ra c tic e s b a se d o n la rg e e p id e mio lo g ic a l stud ie s. In this stud y re g ula rly b e d -sha ring p a re nts a nd infa nts p a rtic ip a te d in a n in-ho me vid e o stud y o f b e d -sha ring b e ha vio ur. Te n fo rmula -fe e d ing a nd te n b re a stfe e d ing fa milie s we re filme d fo r 3 nig hts (a d justme nt, d ya d ic a nd tria d ic nig hts) fo r 8 ho urs p e r nig ht. Fo r b re a stfe d infa nts, mo the r-infa nt o rie nta tio n, sle e p p o sitio n, fre q ue nc y o f fe e d ing , a ro usa l a nd sync hro no us a ro usa l we re a ll c o nsiste nt with p re vio us sle e p -la b stud ie s o f mo the r-infa nt b e d -sha ring b e ha vio ur, b ut sig nific a nt d iffe re nc e s we re fo und b e twe e n fo rmula a nd b re a stfe d infa nts. While b re a stfe e d ing mo the rs b e d -sha re d with the ir infa nts in a c ha ra c te ristic m a nne r tha t p ro vid e d se ve ra l sa fe ty b e ne fits, fo rmula -fe e d ing mo the rs b e d sha re d in a m o re va ria b le m a nne r with c o nse q ue nc e s fo r infa nt sa fe ty. Pa te rna l b e d -sha ring b e ha vio ur intro d uc e d furthe r va ria b ility. Ep id e mio lo g ic a l c a se -c o ntro l stud ie s e xa mining b e d -sha ring risks a nd b e ne fits d o no t no rm a lly c o ntro l fo r b e ha vio ura l va ria b le s tha t a n e vo lutio na ry vie wp o int wo uld d e e m c ruc ia l. This stud y d e mo nstra te s ho w p a re nta l b e ha vio ur a ffe c ts the b e d -sha ring e xp e rie nc e a nd ind ic a te s tha t c a se s a nd c o ntro ls in e p id e m io lo g ic a l stud ie s sho uld b e m a tc he d fo r b e ha vio ura l, a s we ll a s so c io -d e m o g ra p hic , va ria b le s. 2 List o f Fig ure C a ptio ns Fig ure 1: C ha ra c te ristic sle e p p o sitio n o f b re a stfe e d ing m o the rs a nd infa nts List o f Ta b le s Ta b le 1: So c io -d e m o g ra p hic c ha ra c te ristic s o f fa milie s p a rtic ip a ting in stud y Ta b le 2: C o m p a riso n o f b re a stfe e d ing m o the r-infa nt d ya d s b e d -sha ring in o wn ho m e s (c urre nt stud y), ho sp ita l sle e p la b in C a lifo rnia (Ric ha rd , Mo sko e t a l. 1996) a nd ho sp ita l sle e p la b in Bristo l UK (Yo ung 1999). Ta b le 3: C o m p a riso n o f b re a stfe e d ing m o the r-infa nt p a irs in d ya d ic a nd tria d ic (w ith fa the r) sle e p ing a rra ng e me nts Ta b le 4: Infa nt he a d -c o ve ring o n d ya d ic a nd tria d ic b e d -sha ring nig hts. Ta b le 5: C o m p a riso n o f b e d -sha ring c ha ra c te ristic s fo r b re a stfe d a nd fo rmula -fe d infa nts. BIO G RAPHIC AL SKETC H The a utho r is a Se nio r Le c ture r in Anthro p o lo g y a nd Dire c to r o f the Pa re nt-Infa nt Sle e p La b a t the Unive rsity o f Durha m. O rig ina lly tra ine d in p rim a te e tho lo g y, she ha s b e e n re se a rc hing p a re nt-infa nt b e ha vio ur, infa nt sle e p sa fe ty, b re a stfe e d ing , a nd so c ia l sle e p fo r 10 ye a rs. AC KNO WLEDG EMENTS Tha nks a re d ue to the p a re nts a nd infa nts who a llo we d the ir b e d ro o m s to b e inva d e d b y vid e o c a me ra s, Ela ine Ho o ke r who c o d e d the vid e o ta p e s, a nd Ka re n Ma c Do na ld who p ro d uc e d the illustra tio n. C o nve rsa tio ns w ith Jim Mc Ke nna , Je a nine Yo ung , Sa lly Ba d d o c k a nd Ma rtin Wa rd -Pla tt g re a tly imp ro ve d this p a p e r. The Nuffie ld Fo und a tio n a nd the Fo und a tio n fo r the Stud y o f Infa nt De a ths (p ro je c t 227) p ro vid e d fund ing . 3 INTRO DUC TIO N An a nthro p o lo g ic a l p e rsp e c tive o n infa nt sle e p b io lo g y, inc lud ing the e vo lutio na ry und e rp inning s a nd d e ve lo p me nta l b e ne fits o f mo the r-infa nt c o -sle e p ing , ha ve b e e n d e sc rib e d in num e ro us p ub lic a tio ns b y Mc Ke nna a nd c o lle a g ue s (Mc Ke nna , Mo sko e t a l. 1990; Mc Ke nna , Tho m a n e t a l. 1993); Mc Ke nna e t a l. 2001; Mo sko e t a l. 1997) a nd tho se o f se ve ra l o the r b io lo g ic a l a nthro p o lo g ists ( Ba ll 2002, 2003; Ba ll e t a l. 1999, 2000; Hrd y 2000; Sm a ll 1992, 1998; Tre va tha n a nd Mc Ke nna 1994). The b e ha vio ura l a nd p o lyso m no g ra p hic stud ie s re p o rte d b y Mc Ke nna , Mo sko , a nd c o lle a g ue s thro ug ho ut the nine tie s sup p o rte d the lo g ic a l c o nc lusio n d e rive d fro m p hysio lo g ic a l, e vo lutio na ry, histo ric a l, p syc ho lo g ic a l a nd c ro ssc ultura l d a ta , tha t sle e p ing in c lo se p ro ximity to its mo the r o ug ht to b e in a n infa nt’ s b e st inte re sts. A ra ng e o f stud ie s ha ve d e m o nstra te d the b e ne fits o f p ro lo ng e d intima te inte ra c tio n (skin-to skin c o nta c t, ka ng a ro o c a re ) b e twe e n m o the r a nd infa nt in the imme d ia te p o st-na ta l p e rio d , he lp ing infa nts re c o ve r ra p id ly fro m b irth-re la te d fa tig ue (Lud ing to n-Ho e , 1999), e nc o ura g ing sp o nta ne o us b re a stfe e d ing (G o m e z Pa p i, 1998), p ro m o ting c o ntinue d b re a stfe e d ing (De C ha te a u 1977), he lp ing ne wb o rns p re se rve e ne rg y a nd a c c e le ra ting me ta b o lic a d a p ta tio n, a nd thus inc re a sing the we ll b e ing o f the ne wb o rn infa nt (C hriste nsso n e t a l. 1992). Skin-to skin c o nta c t is a lso a sso c ia te d with a sig nific a nt inc re a se in ma te rna l o xyto c in le ve ls (Ma tthie se n, 2001; Nisse n, 1995) with sig nific a nc e fo r ute rine c o ntra c tio n, m ilk e je c tio n, a nd mo the r-infa nt inte ra c tio n; a nd with le ss ma te rna l a nxie ty a nd mo re e ffic ie nt p a rtic ip a tio n o f m o the rs in c a ring fo r the ir ne wb o rn infa nts (Via l-C o urm o nt 2000). The se stud ie s d e m o nstra te the p hysio lo g ic a l im p o rta nc e o f m o the r-infa nt p hysic a l c o nta c t o n infa nt d e ve lo p me nt a nd m o the r-infa nt we ll-b e ing a nd d ra w a tte ntio n to the e vo lve d inte ra c tive na ture o f the p o stp a rtum p e rio d . Mo the r-infa nt sle e p c o nta c t o ve r the first fe w mo nths o f life is a lo g ic a l e xte nsio n o f p o st-p a rtum skin-to -skin o r ka ng a ro o c a re with c o nse q ue nc e s fo r the d e ve lo p m e nt o f infa nt sle e p b io lo g y, a nd m a te rna l fe e d ing p hysio lo g y (Ba ll 2003; Mc Ke nna a nd Mc Da d e 2005). 4 In Euro -Ame ric a n so c ie ty, c o -sle e p ing is o fte n syno nymo us with b e d -sha ring -- i.e . p a re nt(s) a nd infa nt sha ring a n a d ult b e d fo r sle e p -- a ltho ug h c o -sle e p ing re fe rs to a b ro a d e r p he no me no n o f mo the r-infa nt sle e p c o nta c t invo lving p hysic a l p re se nc e a nd se nso ry inte rc ha ng e – b ut d o e s no t ne c e ssita te sle e p ing o n a ‘ we ste rn’ style b e d , o r e ve n o n the sa m e sle e p surfa c e (Mc Ke nna a nd Mo sko 2001). Be d -sha ring a s p ra c tic e d in the UK a nd US is the re fo re o ne fo rm o f mo the r-infa nt c o -sle e p ing – a nd o ne tha t ha s a ttra c te d muc h c o ntro ve rsy, b e ing imp lic a te d in b o th a c c id e nta l infa nt d e a ths a nd SIDS in va rio us c o ntro lle d a nd unc o ntro lle d stud ie s o f sud d e n a nd une xp e c te d infa nt d e a ths ( Bya rd 1994; C a rp e nte r, e t a l. 2004; Dra g o a nd Da nne nb e rg 1999; Mitc he ll a nd Sc ra g g 1993; Na ka mura , Wind e t a l. 1999). So while c lo se m o the r-infa nt c o nta c t a t nig ht a p p e a rs to b e e vo lutio na rily a nd p hysio lo g ic a lly a d a p tive – it ha s b e e n a rg ue d tha t we ste rn b e d d ing a nd sle e p ing a rra ng e m e nts m a y no t b e . Unc o ntro lle d stud ie s (suc h a s tho se b a se d o n the d a ta fro m the US C o nsume r Pro d uc t Sa fe ty C o mm issio n, e .g . Dra g o a nd Da nne nb e rg (1999); Na ka m ura , Wind e t a l. (1999)) te ll us little a b o ut the re la tive risks a nd b e ne fits o f va rio us sle e p ing a rra ng e me nts fo r infa nts. C a se -c o ntro l stud ie s (e .g . Bla ir e t a l 1999; C a rp e nte r e t a l 2004; Ta p p in e t a l 2005) a re e xp e c te d to b e m o re rig o ro us, ho we ve r to d a te c a se -c o ntro l stud ie s e xa mining SIDS a nd infa nt sle e p sa fe ty ha ve p ro d uc e d c o nflic ting re sults re g a rd ing b e d sha ring , a nd a lm o st a ll suc h stud ie s ha ve fa ile d to m a tc h c a se s a nd c o ntro ls o n the b a sis o f b e ha vio ura l va ria b le s suc h a s fe e d ing c ho ic e , p a re nta l a lc o ho l a nd d rug c o nsum p tio n , a nd fre q ue nc y o r typ e o f b e d -sha ring a nd e m p ha sise the ma tc hing o f infa nts o n so c io d e m o g ra p hic va ria b le s a nd the lo c a tio n o f the ir ‘ la st’ o r ‘ re fe re nc e ’ sle e p . We ha ve p re vio usly re p o rte d o n the e ffe c ts o f b re a st vs. fo rmula fe e d ing up o n infa nt sle e p c ha ra c te ristic s a nd the p ro p e nsity to b e d -sha re (Ba ll 2003). In this p a p e r we p re se nt the re sults o f a stud y inve stig a ting ho w b e ha vio ura l va ria b le s suc h a s fe e d ing c ho ic e a nd p a rtne r p re se nc e in the b e d m a y a ffe c t m o the r-infa nt sle e p b e ha vio ur a nd the re fo re infa nt sle e p sa fe ty. C o mp a riso n is ma d e with the d e ta ile d o b se rva tio ns o f mo the r-infa nt b e d -sha ring 5 b e ha vio ur fro m Mc Ke nna a nd Mo sko ’ s stud ie s o f b re a stfe e d ing La tina mo the rs a nd the ir b a b ie s (Mc Ke nna 1990; Mc Ke nna a nd Mo sko 1994; Mc Ke nna , e t a l. 1997; Mo sko e t a l. 1996; Mo sko , e t a l. 1997a ; Mo sko e t a l. 1997b ; Mo sko e t a l. 1997c ; Ric ha rd e t a l. 1997). The se stud ie s we re c o nd uc te d in a ho sp ita l sle e p la b in C a lifo rnia using p o lyso mno g ra p hic mo nito ring a nd vid e o o b se rva tio ns o f b o th mo the rs a nd infa nts sle e p ing to g e the r in a na rro w ho sp ita l b e d . Mo the r-infa nt d ya d s who no rm a lly b e d -sha re d a nd b re a stfe d we re fo und to sle e p in c lo se p ro ximity with a hig h d e g re e o f mutua l o rie nta tio n a nd a ro usa l o ve rla p . Whe n b e d -sha ring , mo the rs p la c e d infa nts in the sup ine p o sitio n fo r sle e p a nd infa nts b re a st-fe d mo re , a nd c rie d le ss, tha n o n nig hts whe n the y sle p t a p a rt fro m the ir m o the rs. Be c a use o f the c o nfine s o f the na rro w ho sp ita l b e d in whic h mo the rs a nd infa nts sle p t, ho we ve r, it wa s unc le a r whe the r the o rie nta tio n, p ro ximity, a ro usa l, a nd b re a stfe e d ing fre q ue nc ie s re p o rte d we re a rte fa c ts o f sle e p ing in a rtific ia lly c ra mp e d c o nd itio ns – o r a n inna te fe a ture o f no rma l b e d -sha ring fo r b re a stfe e d ing mo the rs a nd infa nts. A stud y c o nd uc te d in Bristo l UK, a d d re sse d so m e o f the se issue s. Yo ung (Yo ung 1999) e xa m ine d mo the r-infa nt b e d -sha ring in a ho sp ita l sle e p la b d e sig ne d to re se mb le a d o m e stic b e d ro o m . Mo the rs a nd infa nts b e d -sha re d in a full-size d o ub le b e d , a nd we re re c o rd e d via infra -re d vid e o , with infa nts b e ing mo nito re d p o lyso mno g ra p hic a lly. In te rms o f c lo se p ro ximity, mutua l o rie nta tio n, sup ine infa nt sle e p ing p o sitio n, sle e p sta te c o nc o rd a nc e , a nd b re a stfe e d ing fre q ue nc y, the se d a ta sup p o rte d the find ing s o f Mc Ke nna a nd Mo sko (Yo ung 1999). Bo th o f the a b o ve -me ntio ne d stud ie s o b se rve d mo the rs a nd b re a stfe d infa nts o nly, a nd d id so in a ho sp ita l e nviro nm e nt. The stud y re p o rte d he re wa s d e sig ne d to furthe r e xp lo re 3 a re a s: a ) to e xa m ine the b e d -sha ring b e ha vio ur o f b re a stfe e d ing m o the r-infa nt p a irs in the ir ho m e e nviro nm e nt (d ya d ic b e d -sha ring ); b ) to e xa mine the e ffe c ts o f inc lusio n o f fa the rs in the b e d -sha ring sc e na rio (tria d ic 6 b e d -sha ring ); a nd c ) to e xa m ine the b e d -sha ring b e ha vio ur o f no n-b re a stfe e d ing m o the r-infa nt p a irs. We re p o rt he re o n a ll 3 p a rts o f this stud y. We hyp o the sise d tha t b e c a use the ho me e nviro nme nt wa s m o re fa m ilia r to b o th m o the rs a nd infa nts the y mig ht sle e p mo re d e e p ly a nd e xhib it fe we r a ro usa ls a t ho m e tha n the y ha d in the p re vio us sle e p la b stud ie s. We a lso hyp o the sise d tha t the p re se nc e o f fa the rs b e d -sha ring with the m o the r-infa nt d ya d m ig ht a ffe c t p ro ximity a nd o rie nta tio n o f the infa nt. We furthe r p re d ic te d tha t fa the rs’ p re se nc e mig ht a ffe c t the mo the rs’ re sp o nse -tim e to infa nt ro o ting o r fussing b e ha vio ur, a nd p o ssib ly inc re a se a ro usa ls in the m o the r-infa nt d ya d d ue to the d isturb a nc e c a use d b y a n a d d itio na l b e d -p a rtne r. Fina lly, a ltho ug h the b e ha vio ur o f no n-b re a stfe e d ing b e d -sha re rs ha d no t b e e n d o c ume nte d p rio r to this stud y, a ne c d o ta l e vid e nc e ha d le d us to q ue stio n whe the r m o the rs who d o no t b re a stfe e d the ir infa nts b e d -sha re d iffe re ntly fro m tho se who b re a stfe e d . We hyp o the sise d tha t the la c k o f a b re a st-b a b y a xis o f inte ra c tio n 1 (i.e . a m o ng m o the rs who ha ve ne ve r b re a stfe d ) wo uld e xtra p o la te to d iffe re nc e s in b e d -sha ring c o mp a re d with b re a stfe e d ing m o the rs. METHO DS Ethic s a p p ro va l wa s o b ta ine d fro m b o th the Unive rsity Ethic s C o mm itte e , a nd the Lo c a l NHS Re g io na l Ethic s C o m m itte e . Re c ruitm e nt Fa m ilie s with infa nts, who re g ula rly b e d -sha re d fo r who le o r p a rt nig hts, we re re c ruite d via p o ste rs in lo c a l he a lth c linic s a nd c a lls fo r vo lunte e rs in the lo c a l p re ss, p ro d uc ing a se lf- 1 We ha ve o b se rve d tha t m o the rs who b re a stfe e d c a rry, c ud d le a nd inte ra c t with the ir infa nts he ld a t b re a st-he ig ht, whe re a s m o the rs who d o no t b re a stfe e d c a rry, c ud d le a nd inte ra c t with the ir infa nts he ld a t sho uld e r/ fa c e he ig ht. We na m e d the p rim a ry fo c us o f c o nta c t b e twe e n the m o the r a nd the b a b y the ‘ a xis o f inte ra c tio n’ . 7 g e ne ra te d p o o l o f vo lunte e rs fro m whic h we se le c te d c a nd id a te s who ma tc he d o ur inc lusio n c rite ria : no n-smo king mo the r a nd fa the r w ith infa nts und e r 6 m o nths o f a g e w ho we re b re a stfe d (fo r p a rts a a nd b o f stud y) a nd fo rm ula fe d (fo r p a rt c o f stud y). Fa m ilie s who m e t the se c rite ria we re visite d a t ho me b y o ne o f the re se a rc h te a m who furthe r e xp la ine d the na ture o f the stud y, a nswe re d p a re nts’ q ue stio ns, a nd sub se q ue ntly o b ta ine d p a re nts’ sig na ture s o n a n info rm e d c o nse nt fo rm . Pro c e dure Prio r to vid e o -ta p ing p a re nts we re a ske d to c o mp le te a se rie s o f sle e p lo g s in o rd e r to a sse ss the c irc um sta nc e s o f b e d -sha ring a nd the d e g re e to whic h it w a s p ra c tic e d in the c o urse o f a no rma l we e k. Po rta b le infra -re d vid e o e q uip m e nt w a s the n insta lle d in the fa m ilie s’ ho m e s a nd p a re nts we re p ro vid e d with ve rb a l a nd writte n o p e ra ting instruc tio ns. Pa re nts we re re q ue ste d to b e g in re c o rd ing whe n the y we nt to b e d , a nd to ta p e until the y a wo ke fo r the m o rning , o r fo r the 8-ho ur d ura tio n o f the ta p e , w hic he ve r c a m e so o ne r. Pa re nts we re p ro vid e d with a re m o te c o ntro l with whic h the y c o uld ha lt the re c o rd ing a t a ny time d uring the nig ht. Fa milie s we re vid e o e d fo r a n a d justme nt nig ht (to e limina te ‘ first nig ht e ffe c t’ a nd to te st the vid e o se t-up ) a nd a me mb e r o f the re se a rc h te a m c he c ke d the vid e o the fo llo wing mo rning , a d justing the c a me ra se t-up if ne c e ssa ry. Fa m ilie s we re the n vid e o e d fo r 2 te st nig hts in ra nd o mly a llo c a te d d ya d ic a nd tria d ic b e d -sha ring c o nfig ura tio ns. O n the d ya d ic nig ht we filme d mo the rs a nd infa nts sle e p ing to g e the r (fa the rs sle p t e lse whe re ), a nd o n tria d ic nig hts we filme d the infa nt sle e p ing with b o th p a re nts. Fo llo wing filming p a re nts we re o ffe re d the o p p o rtunity to re vie w the ir ta p e s a nd to e ra se a ny p o rtio ns the y d e sire d in a c c o rd a nc e with Ro ya l C o lle g e o f Physic ia ns g uid e line s o n use o f vid e o ta p e d m a te ria l (o ne m o the r e ra se d a 30 se c o nd p o rtio n). No p hysio lo g ic a l m o nito ring o c c urre d . As a g ra tuity p a re nts re c e ive d a £10 g ift vo uc he r fo r a we ll-kno wn b a b y sto re fo r e a c h nig ht o f filming c o m p le te d . A na lyse s 8 A re se a rc he r tra ine d in e tho lo g ic a l o b se rva tio n te c hniq ue s c o d e d the vid e o s in re a l-time . Fo r e a c h ta p e the fre q ue nc y a nd d ura tio n o f c e rta in ke y b e ha vio urs (e .g . b re a stfe e d ing ) we re c o d e d a nd e xp re sse d a s a p ro p o rtio n o f the to ta l numb e r o f ho urs o b se rve d . Fo r sta te a nd p o sitio na l b e ha vio urs sc a n sa m p le s we re c o d e d e ve ry 3-m inute s thro ug ho ut the e ntire nig ht, a nd sc a n fre q ue nc ie s we re c a lc ula te d a s p ro p o rtio ns o f the to ta l numb e r o f sc a ns ma d e , a llo w ing w ithin fa m ily a nd b e tw e e n fa m ily c o mp a riso ns to b e m a d e . RESULTS Twe nty fa milie s we re vid e o ta p e d b e d -sha ring a t ho me with the ir infa nts: te n invo lving c urre ntly b re a stfe e d ing mo the rs a nd infa nts a nd te n invo lving mo the rs a nd infa nts who ha d ne ve r b re a stfe d . No sig nific a nt d iffe re nc e s in nig ht-time c a re -g iving p a ra me te rs we re d e te c te d fro m the sle e p lo g s o f the b re a st a nd fo rmula fe e d ing fa milie s. All ho me s ha d c e ntra l he a ting , a nd re c o rd ing s o f the two g ro up s o f fa milie s we re e ve nly d istrib ute d a c ro ss the ye a r. The so c io -d e mo g ra p hic c ha ra c te ristic s o f the fa m ilie s in the two sub -sa m p le s a re sho wn in Ta b le 1. Pa re nts o f b re a stfe d infa nts we re o ld e r tha n tho se o f fo rmula -fe d infa nts, ha d e xp e rie nc e d slig htly m o re ye a rs o f e d uc a tio n, a nd ha d g re a te r ho use ho ld inc o me s (re fle c ting the c urre nt p a tte rn o f so c io d e m o g ra p hic d iffe re nc e s b e twe e n b re a stfe e d e rs a nd no n-b re a stfe e d e rs in the UK (Ha mlyn 2002)), ho we ve r the o nly d iffe re nc e tha t re a c he d sta tistic a l sig nific a nc e b e tw e e n the two g ro up s wa s ho use ho ld inc o me . 9 Fig ure L C ha ra c te ristic sle e p po sitio n o f b re a st- fe e ding m o the rs a nd infa nts. Mo the r- infa nt b e d- sha ring a t ho m e Da ta summa rise d fro m the vid e o -ta p e s o f the in-ho me b e d -sha ring nig hts fo r the 10 b re a stfe e d ing b e d -sha ring mo the r infa nt d ya d s a re p re se nte d in Ta b le 2 a lo ng sid e d a ta re p o rte d b y Ric ha rd e t a l (Ric ha rd , Mo sko e t a l. 1996) o n 6 ro utine ly b e d -sha ring b re a stfe d infa nts a nd the ir m o the rs who we re o b se rve d in the C a lifo rnia sle e p la b , a nd d a ta re p o rte d b y Yo ung (Yo ung 1999) o n 5 ro utine ly b e d -sha ring b re a stfe d infa nts a nd the ir mo the rs who w e re o b se rve d in the Bristo l sle e p la b . TABLE 2 HERE A re m a rka b ly c o nsiste nt p ic ture e me rg e s fro m the se thre e stud ie s. Bre a stfe e d ing b e d -sha ring mo the rs sle p t fa c ing the ir infa nts fo r 71-76% o f the nig ht, infa nts sle p t fa c ing the ir mo the rs fo r 87-97% o f the nig ht. Re g a rd le ss o f b e d -size mo the rs a nd infa nts sle p t in c lo se p ro ximity to o ne a no the r, o fte n to uc hing , b ut g e ne ra lly no m o re tha n 20c m (infa nt a rm -le ng th) a p a rt. The fre q ue nc ie s o f fe e d ing b o uts, a nd to ta l le ng th o f fe e d ing we re g re a te r in the C a lifo rnia sle e p la b stud y, tha n in b o th the Bristo l sle e p la b stud y, a nd the Durha m in-ho m e stud y, ho we ve r the infa nts in the C a lifo rnia stud y we re o f a yo ung e r (a nd m o re lim ite d ) a g e -ra ng e tha n the UK stud ie s, whic h ma y a c c o unt fo r the d iffe re nc e . Infa nts ra re ly sle p t in the p ro ne p o sitio n while b e d -sha ring in a ll 3 stud ie s. The g re a te st d iffe re nc e b e twe e n the Durha m in-ho me stud y 10 a nd the Bristo l a nd C a lifo rnia Sle e p La b stud ie s invo lve s the p ro p o rtio n o f the nig ht infa nts sp e nt in a la te ra l a nd sup ine sle e p p o sitio n. In the p re se nt stud y b e d -sha ring b re a stfe e d ing infa nts sle p t p re d o mina ntly in a la te ra l p o sitio n, whe re a s in the p re vio us stud ie s infa nts p re d o m ina ntly sle p t in a sup ine p o sitio n. O ur p re d ic tio n tha t the ho me e nviro nme nt wo uld b e a sso c ia te d with fe we r a ro usa ls a nd le ss m utua l a wa re ne ss tha n o b se rve d in sle e p la b stud ie s wa s no t sup p o rte d . Inc lusio n o f fa the rs The e ffe c t o f the fa the r’ s p re se nc e in b e d with the m o the r-infa nt d ya d wa s e xp lo re d b y c o mp a ring infa nt sle e p p o sitio n, mo the r-infa nt p ro ximity, o rie nta tio n a nd b re a stfe e d ing d a ta fo r the d ya d ic a nd tria d ic sle e p nig hts: TABLE 3 HERE As sho wn in Ta b le 3 a sig nific a nt d iffe re nc e wa s o b se rve d re g a rd ing the lo c a tio n o f the b a b y in the b e d . O n tria d ic nig hts b a b ie s we re p re d o m ina ntly p o sitio ne d in the c e ntre o f the b e d , b e twe e n b o th p a re nts (a ltho ug h o ne m o the r e xc lusive ly sle p t he r infa nt o n the sid e o f the b e d o n the tria d ic nig ht), while o n d ya d ic nig hts m o st m o the rs sle p t in the c e ntre o f the b e d the mse lve s, a nd mo ve d the infa nt fro m sid e to sid e a s ne e d e d fo r b re a stfe e d ing . Re g a rd ing sle e p b e ha vio ur, m o the rs fa c e d the ir infa nts (o n a ve ra g e ) fo r thre e -q ua rte rs o f sle e p tim e o n b o th the d ya d ic a nd tria d ic b e d -sha ring nig hts, a nd the a mo unt o f time fo r whic h infa nts fa c e d the ir mo the rs wa s no d iffe re nt o n the two te st nig hts. Furthe rmo re infa nts a nd mo the rs sp e nt o ve r 70% o f b o th nig hts in p hysic a l c o nta c t with o ne a no the r, a nd b re a stfe e d ing a c tivity wa s simila r o n b o th nig hts. C o ntra ry to o ur p re d ic tio n the fa the r's p re se nc e d id no t, the re fo re , a lte r the c ha ra c te ristic p ic ture o f b e d -sha ring a s p ra c tise d b y b re a stfe e d ing m o the rs a nd the ir infa nts. Altho ug h b re a stfe e d ing m o the rs sle p t fa c ing , a nd in c lo se p ro ximity to the ir infa nt, fa the rs 11 te nd e d to sle e p fa c ing a wa y fro m the ir infa nt, a nd we re g e ne ra lly b e yo nd the ir infa nt’ s re a c h (o ve r 20c m a wa y). The re wa s o ne e xc e p tio n, a fa the r who sp e nt 83% o f sle e p -tim e to uc hing his infa nt, who in turn sp e nt 90% o f sle e p time fa c ing his fa the r. This fa the r-infa nt d ya d sp e nt 53% o f sle e p time in a fa c e -to -fa c e o rie nta tio n (c o mp a re d to 5% b e twe e n infa nt a nd m o the r), ho we ve r the m o the r sub se q ue ntly re p o rte d tha t she ha d ma stitis o n the tria d ic sle e p nig ht tha t m a d e sle e p ing with p re ssure o n the a ffe c te d b re a st (fa c ing the c e ntre o f the b e d ) unc o mfo rta b le . The b e ha vio ur o b se rve d in this fa m ily the re fo re wa s a typ ic a l o f the ir no rma l nig ht-time c a re -g iving p ra c tic e s. Mo the rs e xhib ite d c o nsiste ntly sync hro no us a ro usa l p a tte rns w ith the ir infa nt: 9 o f 10 m o the rinfa nt p a irs a ro using simulta ne o usly o n b o th d ya d ic a nd tria d ic nig hts, so me p a irs d o ing so up to nine tim e s o n a sing le nig ht. Two fa the rs e xhib ite d 100% sync hro no us a ro usa ls with the ir b e d -p a rtne rs o n the tria d ic nig ht, a ltho ug h the m a jo rity o f fa the rs sle p t thro ug h ha lf o f the ir infa nts’ a ro usa ls. Two fa the rs e xhib ite d no sync hro ny with the m o the r-infa nt p a ir, ne ve r sho wing a ny sig ns o f wa king o n the tria d ic nig ht. Sync hro ny o f fa the rs’ sle e p sta te s d uring tria d ic b e d -sha ring the re fo re a p p e a rs to b e va ria b le a nd id io sync ra tic . O n b o th d ya d ic a nd tria d ic nig hts m o the rs sp e nt la rg e p o rtio ns o f sle e p tim e in a d istinc tive sle e p ing p o sture (Fig ure 1), sp o nta ne o usly e xhib ite d b y a ll b re a stfe e d ing m o the rs in this stud y. The se m o the rs a d o p te d a la te ra l sle e p ing p o sitio n fa c ing the ir infa nt, with the m o the r’ s kne e s d ra w n up und e r the infa nt’ s fe e t, a nd m o the r’ s up p e r a rm p o sitio ne d a b o ve the infa nt’ s he a d (e ithe r und e r o r o n to p o f the p illo w). In c o ntra st mo st fa the rs ne ve r sle p t in this p o sitio n, o r d id so fo r o nly sho rt p e rio d s o f time . This p o sitio n c le a rly fa c ilita te d b re a st-b a b y c o nta c t in the se b e d -sha ring b re a stfe e d ing p a irs. Tha t this p o sitio n a lso p ro vid e s se ve ra l b e ne fits re la ting to b e d -sha ring sa fe ty is d isc usse d b e lo w FIG URE 1 HERE 12 Using inte rva l sa mp ling we c o d e d the p o sitio n o f b e d d ing a nd p a re nts' limb s re la tive to the infa nt’ s b o d y, he a d a nd fa c e , a nd the d ire c tio n a nd he ig ht o f the infa nt in the b e d o n b o th d ya d ic a nd tria d ic b e d -sha ring nig hts. We o b se rve d no insta nc e s, o n e ithe r d ya d ic o r tria d ic sle e p nig hts, o f p a re nts a c tua lly o r p o te ntia lly o ve rla ying the ir infa nts. De sp ite the fa c t tha t the re we re p illo w s in the b e d o n a ll o c c a sio ns no b re a stfe d infa nts w e re o b se rve d w ith the ir fa c e in the p illo w s, a nd p illo w s ra re ly to uc he d the to p o f a n infa nt's he a d -- d ue ma inly to the c ha ra c te ristic p o sitio n in whic h the se b re a stfe e d ing m o the rs we re o b se rve d to sle e p . C o ve r p o sitio n o n b a b ie s wa s hig hly va ria b le b o th fro m fa mily to fa mily, a nd b e twe e n d ya d ic a nd tria d ic nig hts. Be d -c o ve rs we re g e ne ra lly lo we r o n a n infa nt's b o d y whe n sle e p ing with the m o the r o nly, o r whe n sle e p ing o n the o utsid e o f the m o the r o n the tria d ic nig ht, tha n whe n sle e p ing b e twe e n b o th p a re nts. Ho we ve r, the re wa s no sig nific a nt d iffe re nc e in the me d ia n p ro p o rtio n o f sc a ns in whic h b a b ie s ha d b e d -c o ve rs o ve r the ir he a d s o n the d ya d ic a nd tria d ic nig hts (Ta b le 4), o r the me d ia n p ro p o rtio n o f time b a b ie s sp e nt with the c o ve rs a b o ve ve rsus b e lo w the ir w a ists. TABLE 4 HERE Mo the rs p re d o m ina ntly ke p t the b e d -c o ve rs a t the ir o wn wa ist-he ig ht o n the d ya d ic sle e p nig ht, a nd a ltho ug h c o ve rs we re hig he r o n the m o the r whe n the fa the r wa s sha ring the b e d (d ue to fa the rs ke e p ing the c o ve rs a t c hin he ig ht), the re wa s no sig nific a nt d iffe re nc e in c o ve r he ig ht o n the tw o nig hts. O n the d ya d ic nig ht the re wa s a te nd e nc y fo r infa nts to ro ta te fro m a ve rtic a l to wa rd s a ho rizo nta l p o sitio n in the b e d , whic h wa s no t o b se rve d o n the tria d ic nig ht, p re suma b ly d ue to the g re a te r sp a c e a va ila b le in the b e d in the fa the r's a b se nc e . Tha t infa nts sp e nt p ro p o rtio na lly le ss o f the d ya d ic nig ht tha n the tria d ic nig ht sle e p ing p e rp e nd ic ula r to the he a d b o a rd wa s sig nific a nt. So m e fa m ilie s sle p t und e r d uve ts o r c o m fo rte rs, while o the rs use d b la nke ts, b ut a ll we re suffic ie ntly fle xib le tha t the y d id no t c o ve r the infa nt's he a d whe n the y we re p ulle d up und e r 13 the p a re nts' c hins. No ne o f the infa nts vid e o ta p e d wo re mo re tha n a stre tc hy sle e p suit w he n b e d -sha ring , a nd no ne a p p e a re d unusua lly swe a ty in the c o urse o f e ithe r d ya d ic o r tria d ic b e d -sha ring nig hts, a ltho ug h w itho ut a te m p e ra ture m o nito r a tta c he d to the infa nts o ur a sse ssm e nts he re a re b a se d o n lim ite d e vid e nc e . Be ha vio ur o f no n- b re a stfe e ding b e d- sha ring dya ds Vid e o ta p e s o f 10 fa milie s with infa nts who we re no t b re a stfe d we re a na lyse d a s a b o ve , a nd o ur hyp o the sis tha t the b e ha vio ur o f b e d -sha ring no n-b re a stfe e d ing mo the rs a nd infa nts wo uld d iffe r fro m tha t o f b re a stfe e d ing d ya d s wa s sup p o rte d . Ta b le 5 c o mp a re s the b e ha vio ura l re sults fo r b re a stfe e d ing a nd no n-b re a stfe e d ing b e d -sha re rs. Sig nific a nt d iffe re nc e s we re fo und b e twe e n the tw o g ro up s in re la tio n to the p ro p o rtio n o f the nig ht m o the rs fa c e d the ir infa nts a nd c o nse q ue ntly the p ro p o rtio n o f the nig ht m o the rs a nd infa nts sp e nt in fa c e -to -fa c e o rie nta tio n; infa nt sle e p ing p o sitio n (b re a stfe d infa nts sle p t p re d o mina ntly o n the ir sid e s, fo rmula -fe d infa nts sle p t o n the ir b a c ks); he ig ht o f infa nt in b e d re la tive to m o the r (b re a stfe d infa nts we re a t c he st he ig ht, fo rm ula -fe d infa nts we re a t fa c e he ig ht); a nd fe e d ing a nd wa king fre q ue nc y (b re a stfe d infa nts wo ke a nd fe d mo re fre q ue ntly tha n fo rmula -fe d infa nts). TABLE 5 HERE DISC USSIO N Bre a stfe e d ing m o the r-infa nt d ya d s ha ve no w b e e n o b se rve d d isp la ying the sa me b e d sha ring b e ha vio ur in 3 se p a ra te stud ie s und e r d iffe re nt c o nd itio ns sug g e sting tha t a ro b ust p ic ture o f b e d -sha ring b e ha vio ur is e me rg ing . The find ing tha t a c ha ra c te ristic sle e p p o sitio n is c o nsiste ntly a d o p te d b y b e d -sha ring b re a stfe e d ing mo the rs sug g e sts tha t a d o p tio n o f this sle e p p o sitio n is a n instinc tive e vo lve d b e ha vio ur. As fe ma le g re a t a p e s a re a lso d o c ume nte d to a d o p t this p o sitio n w he n sha ring sle e p ing ne sts with the ir sm a ll infa nts (G o o d a ll 1968) it is p o ssib le tha t this b e ha vio ur e vo lve d in the c o nte xt o f infa nt p ro te c tio n a nd 14 sa fe ty 2 . Huma n mo the rs sle e p ing c urle d up a ro und the ir infa nts in this ma nne r e ffe c tive ly p ro vid e a b a rrie r with the ir up p e r le g s tha t p re ve nts the b a b y m o ving d o wn the b e d und e r the c o ve rs. The y a lso fo rm a b a rrie r with the ir up p e r a rm a b o ve the b a b y's he a d , tha t p re ve nts the b a b y fro m m o ving up the b e d a nd into (o r und e r) the m o the r's p illo w (s). The b a b y sle e p s fla t o n the m a ttre ss (no t p ro p p e d b y p illo w s), e ithe r in a la te ra l o r sup ine p o sitio n 3 . The fa c t tha t the m o the r is c urle d a ro und he r infa nt m e a ns she c a nno t ro ll fo rwa rd s o nto it, a nd ne ithe r c a n a ny o the r b e d -sha re rs lie o n the b a b y w itho ut lying o n the mo the r a lso . This p o sitio n is id e ntic a l to tha t d e sc rib e d b y Mo sko e t a l (1997c ); Ric ha rd e t a l (1996), a nd d e p ic te d b y Yo ung (1999). We e xa m ine d whe the r sle e p ing with b o th p a re nts invo lve d mo re ha za rd o us situa tio ns fo r a n infa nt tha n sle e p ing w ith the m o the r a lo ne . De sp ite the fa c t tha t the re we re p illo w s in the b e d o n a ll o c c a sio ns no infa nts we re o b se rve d with the ir fa c e in the p illo ws, a nd p illo ws ra re ly to uc he d the to p o f a n infa nt's he a d -- d ue to the c ha ra c te ristic p o sitio n in whic h the se b re a stfe e d ing d ya d s we re to sle e p . Po te ntia l infa nt o ve rhe a ting is so me time s c ite d a s a risk fa c to r in tria d ic b e d -sha ring d ue to p o ssib le the rma l-c o up ling b e tw e e n two a d ult b o d ie s (Tuffne ll e t a l. 1996). The o b se rva tio n tha t m o st infa nts sp e nt the va st m a jo rity o f b o th nig hts in c lo se p ro ximity to the m o the r, w ith the fa the r a t so m e d ista nc e a w a y, sug g e sts tha t the p re se nc e o f the fa the r in the b e d w ill ha ve a ne g lig ib le imp a c t o n the te mp e ra ture o f the infa nt -- a ltho ug h size o f the b e d a nd p a re nta l b o d y size will b o th b e fa c to rs a ffe c ting this re la tio nship . 2 Se ve ra l p ie c e s o f a rtwo rk thro ug ho ut 19 th a nd 20 th c e nturie s ha ve a lso d e p ic te d m o the rs a nd infa nts sle e p ing to g e the r in this p o sitio n e .g . Elwe ll’ s ‘ The Firstb o rn’ (Fe re n’ s Art G a lle ry, Hull, UK), a nd Mo hd e rso n’ s ‘ Mutte r mit kind ’ (Be rlin, G e rm a ny) . 3 It is p o ssib le tha t b e c a use the b a b ie s in the sle e p la b stud ie s we re wire d -up fo r p o lyso mno g ra p hic m o nito ring the sup ine p o sitio n m a y ha ve b e e n m o re c o m fo rta b le , thus a ffe c ting sle e p p o sitio n. Alte rna tive ly we m a y ha ve use d d iffe re nt c rite ria to d isting uish b e twe e n la te ra l a nd sup ine p o sitio ning . 15 A p ic ture e me rg e s o f so me no ta b le d iffe re nc e s in sle e p ing a rra ng e me nts b e tw e e n b re a stfe e d ing b e d -sha ring p a re nts a nd infa nts a nd fo rmula -fe e d ing p a re nts a nd infa nts. The m o st striking o f the se invo lve the infa nt's b o d y p o sitio n, a nd his/ he r p o sitio n in the b e d . Fo rmula -fe d infa nts p re d o mina ntly sle p t sup ine , w hile a ll b ut o ne b re a stfe d infa nt sp e nt the ma jo rity o f the nig ht sle e p ing in a la te ra l p o sitio n. This sig nific a nt d iffe re nc e a p p e a rs to re fle c t the e a se o f b re a stfe e d ing in the la te ra l p o sitio n. The sup ine sle e p ing p o sitio n is re c o mme nd e d fo r infa nts in a ll we ste rn c o untrie s d ue to the inc re a se d risk o f c o t d e a th a m o ng infa nts sle e p ing p ro ne (e .g . AAP 2000; Be a l 2000; Dwye r 1996; Esm a il e t a l. 1995; Fle m ing e t a l. 1990; G ib so n e t a l. 2000; Ma rke sta d e t a l. 1995; Sc ra g g , e t a l. 1996; Willing e r e t a l. 2000). Altho ug h re c o m me nd e d in the US until re c e ntly, the la te ra l sle e p p o sitio n ha s b e e n d isc o ura g e d fo r UK infa nts fo r a ro und a d e c a d e , a s se ve ra l c a se -c o ntro l SIDS stud ie s ha ve ind ic a te d tha t sle e p ing o n the sid e is a sso c ia te d with a g re a te r risk o f SIDS tha n sup ine sle e p (Fle m ing e t a l. 2000; Sc ra g g a nd Mitc he ll 1998; Ska d b e rg e t a l. 1998; Wig fe ld e t a l. 1992). It wo uld se e m, the re fo re , tha t whe n b e d -sha ring , fo rmula -fe d infa nts a re mo re c o mp lia nt with b a c k-to -sle e p g uid e line s tha n b re a stfe d infa nts. The issue is c o mp lic a te d , ho we ve r, in tha t e p id e mio lo g ic a l stud ie s e xa mining sle e p ing p o sitio n ha ve no t d o ne so in the c o nte xt o f b e d sha ring . Se ve ra l o f the a b o ve -m e ntio ne d a utho rs (e .g . Fle m ing e t a l. 2000); Ska d b e rg e t a l. 1998; Wig fie ld , e t a l. 1992) ha ve no te d tha t infa nts sle e p ing a lo ne who a re p o sitio ne d la te ra lly m a y ro ll fo rw a rd s into the p ro ne p o sitio n (the re b y inc re a sing the ir risk o f SIDS), ho we ve r a n infa nt sle e p ing in a la te ra l p o sitio n ne xt to its m o the r’ s b o d y is una b le to ro ll fo rwa rd s. It is c urre ntly unkno wn whe the r la te ra l sle e p ing in this c o nte xt is a lso a sso c ia te d with a n inc re a se d SIDS risk. The re la tio nship b e twe e n the sle e p ing p o sitio n o f b a b ie s who p ra c tic e b re a stfe e d ing re la te d b e d -sha ring a nd SIDS-risk is the re fo re a to p ic tha t re q uire s furthe r e xp lo ra tio n in future e p id e mio lo g ic a l stud ie s. Infa nt 'he ig ht' in the b e d (i.e . c lo se ne ss to p a re nta l fa c e s) a lso d iffe re d sig nific a ntly b e twe e n b re a stfe d a nd fo rmula -fe d b a b ie s with p a re nts o f the la tte r e xhib iting a te nd e nc y to p la c e the ir infa nts hig h in the b e d , with infa nts' fa c e s le ve l with p a re nts' fa c e s, a nd infa nts e ithe r 16 p o sitio ne d b e twe e n, o r o n to p o f, the p a re nts' p illo ws. Bre a stfe d infa nts we re c o mmo nly p o sitio ne d fla t o n the ma ttre ss, b e lo w p illo w he ig ht, le ve l with the m o the r's c he st. The p re d o mina nt lo c a tio n in the b e d fo r a ll infa nts wa s the mid d le o f the b e d , b e tw e e n b o th p a re nts, ho we ve r b re a stfe d infa nts a lso sp e nt tim e sle e p ing o n the o utsid e o f the mo the r, while fo rmula -fe d infa nts te nd e d no t to b e mo ve d a ro und the b e d . The c ha ng ing lo c a tio n o f the b re a stfe d infa nts re fle c ts the fa c t tha t so m e m o the rs m o ve d the ir b a b y to the ir le ft o r rig ht to fa c ilita te fe e d ing fro m a p a rtic ula r b re a st. It wa s a p p a re nt, ho we ve r tha t so me m o the rs we re a b le to fe e d fro m e ithe r b re a st while re m a ining in the sa me p o sitio n. As wo uld b e e xp e c te d (se e Ba ll 2003), fe e d ing fre q ue nc y a nd fe e d ing d ura tio n d iffe re d sig nific a ntly b e twe e n b re a stfe e d ing a nd fo rmula -fe e d ing b e d -sha re rs, the la tte r fe e d ing o n a ve ra g e o nc e p e r nig ht, while the fo rm e r fe d a t le a st tw ic e , so me infa nts fe e d ing fo ur o r m o re tim e s d uring the nig ht. Fe e d ing fre q ue nc y is re la te d to a ro usa l fre q ue nc y a nd a ro usa l sync hro ny b e twe e n b e d -sha ring m o the rs a nd infa nts, with b re a stfe e d ing m o the rs a nd infa nts e xp e rie nc ing sig nific a ntly g re a te r a ro usa l fre q ue nc ie s d uring the nig ht tha n fo rmula -fe e d ing m o the rs a nd infa nts, to g e the r with m o re sync hro no us a ro usa ls. Fo rmula -fe e d ing mo the rs d id no t ha ve the sa me d e g re e o f o rie nta tio n to wa rd s the ir infa nts d uring sle e p a s b re a stfe e d ing mo the rs, a nd sp e nt sig nific a ntly le ss time b o th fa c ing the ir infa nt, a nd in mutua l fa c e -to -fa c e o rie nta tio n, tha n d id b re a stfe e d ing mo the r-infa nt p a irs. The re wa s no sig nific a nt d iffe re nc e fo und in the o rie nta tio n o f fa the rs to the ir b re a st o r fo rmula -fe d infa nts. Ma ny o f the b e d -sha ring sa fe ty issue s d isc usse d a b o ve re g a rd ing b re a stfe d infa nts a lso a p p ly to fo rmula -fe d infa nts. Ho we ve r, g ive n the d iffe re nc e s in b e d -sha ring p a ra me te rs suc h a s infa nt p o sitio n a nd p a re nta l o rie nta tio n b e twe e n b re a st a nd fo rmula -fe d infa nts, it is use ful fo r c e rta in a sp e c ts o f b e d -sha ring sa fe ty to b e c o nsid e re d se p a ra te ly fo r e a c h g ro up . 17 Se ve ra l o f the fo rmula -fe d infa nts we re p la c e d with the ir he a d s (a nd sho uld e rs) o n the ir p a re nts’ p illo ws fo r the ma jo rity o f the nig ht. Bre a stfe d infa nts, o n the o the r ha nd , we re ne ve r p la c e d o n o r ne a r the p illo ws. We sug g e st tha t mo the rs o f fo rmula -fe d infa nts (p a rtic ula rly m o the rs who ha ve ne ve r b re a stfe d ) sle e p with the ir infa nts in the m a nne r o f sle e p ing with a no the r a d ult (e .g . fa c e s a t sa m e he ig ht, no c o nsiste nt ‘ p ro te c tive ’ sle e p ing p o sitio n), while m o the rs o f b re a stfe d infa nts sle e p with the ir b a b ie s in a m o re ‘ p ro te c tive ’ m a nne r – b a b y p o sitio ne d a t b re a st-le ve l, with m o the r o rie nte d to wa rd s b a b y in a c ha ra c te ristic p o sitio n. The b re a stfe e d ing m o the rs in this stud y sle p t w ith the ir infa nts in a sa fe r ma nne r tha n the no nb re a stfe e d ing m o the rs. C o ve ring the fa c e with b e d -c lo the s wa s a n infre q ue nt o c c urre nc e fo r the ma jo rity o f infa nts in b o th g ro up s a nd ne ithe r b re a st no r fo rmula fe d b a b ie s we re m o re like ly to ha ve the ir he a d s c o ve re d d uring b e d -sha ring , d e sp ite the fa c t tha t the b re a stfe d b a b ie s in g e ne ra l sle p t lo we r in the b e d tha n d id the fo rmula -fe d b a b ie s. The se find ing s illustra te tha t b e d -sha ring is no t a ‘ g e ne ric ’ b e ha vio ur fo r a ll m o the rs a nd infa nts. The sle e p re la tio nship o b se rve d (in this a nd o the r stud ie s) b e twe e n b re a stfe e d ing mo the rs a nd infa nts is q ua lita tive ly a nd q ua ntita tive ly d iffe re nt fro m tha t o b se rve d b e twe e n fo rmula -fe e d ing mo the rs a nd infa nts. The p a tte rning o f the se d iffe re nc e s is c o nsiste nt with o ur und e rsta nd ing o f the p hysio lo g ic a l me c ha nism s m e d ia ting m a te rna l a nd infa nt b e ha vio ur, in tha t b re a stfe e d ing m o the rs e xp e rie nc e a ho rm o na l fe e d b a c k c yc le whic h p ro m o te s c lo se c o nta c t with, he ig hte ne d re sp o nsive ne ss to wa rd s, a nd b o nd ing with infa nts in a wa y tha t is missing fo r mo the rs who d o no t b re a stfe e d (Hrd y 1999, Sm a ll 1999). Furthe rm o re , infa nts who a re b re a stfe d instinc tive ly o rie nt to wa rd s m a te rna l b re a st o d o ur (Va re nd i a nd Po rte r 2001), a n instinc t whic h wo uld c o ntrib ute to the m a inte na nc e o f the c ha ra c te ristic sle e p ing p o sitio n o b se rve d fo r b re a stfe e d ing m o the r-infa nt d ya d s. The se c ha ra c te ristic d iffe re nc e s in sle e p b e ha vio ur whe n b e d -sha ring ma y we ll a ffe c t the 18 o utc o m e s o f c a se -c o ntro l stud ie s re g a rd ing infa nt sa fe ty o r SIDS-risk whe re c a se s a nd c o ntro ls a re no t ma tc he d fo r infa nt fe e d ing b e ha vio ur. The re sults p re se nte d he re ind ic a te tha t no t o nly sho uld c a se s a nd c o ntro ls b e m a tc he d fo r va ria b le s kno w n to a ffe c t b e d -sha ring b e ha vio ur, b ut tha t d e ta ile d d a ta sho uld a lso b e c o lle c te d o n the m a nne r o f b e d -sha ring b y e a c h fa mily invo lve d (e .g . infa nt p o sitio n in the b e d ) re g a rd ing b o th the ir no rm a l p ra c tic e , a nd o n the nig ht in q ue stio n. C O NC LUSIO N This stud y c o nfirm s in the ho me e nviro nme nt the p ic ture o f b e d -sha ring a m o ng b re a stfe d mo the r-infa nt d ya d s re p o rte d in sle e p la b s b y two o the r re se a rc h g ro up s. Fo r fo rm ula -fe d infa nts we fo und tha t the b e d -sha ring p ic ture is so me wha t d iffe re nt d ue to the la c k o f a b re a st-b a b y o rie nta tio n d uring sle e p . Princ ip a l sa fe ty issue s re g a rd ing b e d -sha ring re la te to the c ha ra c te ristic d iffe re nc e s o f the se two g ro up s – the d a ng e r o f a c c id e nta l suffo c a tio n b y a n infa nt sle e p ing o n o r ne a r the p a re nta l p illo w s is c a use fo r c o nc e rn re g a rd ing fo rm ula -fe d infa nts, while the te nd e nc y o f b re a stfe d infa nt to a ssum e a la te ra l ra the r tha n sup ine sle e p p o sitio n while b e d -sha ring m ig ht c a use c o nc e rns re g a rd ing the p o ssib ly e le va te d risk o f SIDS in no n-sup ine sle e p ing infa nts. This issue c o uld b e a d d re sse d in a c a se -c o ntro l stud y whe re the c a se s a re m a tc he d fo r fe e d ing b e ha vio ur. The fa the r’ s p re se nc e in the b e d d id no t a p p e a r to p re se nt a ny unive rsa l sa fe ty imp lic a tio ns fo r b e d -sha ring infa nts, ho we ve r the fa c t tha t the re wa s g re a t ind ivid ua l va ria tio n o b se rve d in p a te rna l a ro usa b ility in re sp o nse to infa nts d uring the nig ht wo uld b e wo rthy o f furthe r e xp lo ra tio n. An imp o rta nt im p lic a tio n o f this stud y is tha t e p id e m io lo g ic a l inve stig a tio ns o f the risks o f SIDS a nd a c c id e nta l d e a ths in a b e d -sha ring c o nte xt sho uld inc lud e the m a tc hing o f c a se s a nd c o ntro ls b y the fe e d ing b e ha vio ur o f the infa nts stud ie d . The re sults o f this stud y, a nd a n e vo lutio na ry-info rme d vie wp o int, p re d ic t tha t b e d -sha ring ma y b e sa fe r fo r b re a st-fe d tha n ne ve r-b re a stfe d infa nts. BIBLIO G RAPHY AAP (2000). "Ta sk Fo rc e o n Infa nt Sle e p Po sitio n a nd Sud d e n Infa nt De a th Synd ro m e : 19 C ha ng ing c o nc e p ts o n sud d e n infa nt d e a th synd ro me : imp lic a tio ns fo r infa nt sle e p ing e nviro nme nt a nd sle e p p o sitio n." Pe d ia tric s 105(3): 650-656. Ba ll, H. L. (2002). "Re a so ns to b e d -sha re : w hy p a re nts sle e p w ith the ir infa nts." Jo urna l o f Re p ro d uc tive a nd Infa nt Psyc ho lo g y 20(4): 207-222. Ba ll, H. L. (2003). "Bre a stfe e d ing , b e d -sha ring a nd infa nt sle e p ." Birth 30(3): 181-188. Ba ll, H., E. Ho o ke r, PJ Ke lly. (2000). "Pa re nt-Infa nt C o sle e p ing : fa the rs' ro le s a nd p e rsp e c tive s." Infa nt a nd C hild De ve lo p me nt 9: 67-74. Ba ll, H. L., E. Ho o ke r, PJ Ke llyl. (1999). "Whe re w ill the b a b y sle e p ? a ttitud e s a nd p ra c tic e s o f ne w a nd e xp e rie nc e d p a re nts re g a rd ing c o sle e p ing with the ir ne wb o rn infa nts." Am e ric a n Anthro p o lo g ist 10 (1): 143-151. Be a l, S. (2000). "Sud d e n Infa nt De a th Synd ro m e in So uth Austra lia 1968-97. Pa rt I: c ha ng e s o ve r tim e ." Jo urna l o f Pa e d ia tric s a nd C hild He a lth 36(6): 540-7. Bya rd , R. W. (1994). "Is c o -sle e p ing in infa nc y a d e sira b le o r d a ng e ro us p ra c tic e ? " Jo urna l o f Pa e d ia tric s a nd C hild He a lth 30: 198-199. C a rp e nte r, R. G ., L. M. Irg e ns, e t a l. (2004). "Sud d e n une xp la ine d infa nt d e a th in 20 re g io ns in Euro p e : c a se c o ntro l stud y." The La nc e t 363: 185-191. C hriste nsso n, K., Sile s, C , Mo re no , L., Be la uste q ui, A., De La Fue nte , P., La g e rc ra ntz, H., Puyo l, P., Winb e rg , J. (1992). "Te m p e ra ture , m e ta b o lic a d a p ta tio n a nd c rying in he a lthy fullte rm ne wb o rns c a re d fo r skin-to -skin o r in a c o t." Ac ta Pa e d ia tric a 81(6-7): 488-93. De C ha te a u, P. a nd B. Wib e rg (1977). "Lo ng -te rm e ffe c t o n mo the r-infa nt b e ha vio urs o f e xtra c o nta c t d uring the first ho ur p o st p a rtum . II. A fo llo w up a t thre e m o nths." Ac ta Pa e d ia tric a Sc a nd ina via 66(2): 145-151. Dra g o , D. A. a nd A. L. Da nne nb e rg (1999). "Infa nt me c ha nic a l suffo c a tio n d e a ths in the unite d sta te s, 1980-1997." Pe d ia tric s 103 (5): 1-8. Dwye r, T. (1996). "Sud d e n infa nt d e a th synd ro m e : a fte r the "b a c k to sle e p " c a mp a ig n." British Me d ic a l Jo urna l 313: 180-181. Esm a il, A., P. C . La m b e rt, e t a l. (1995). "Pre va le nc e o f risk fa c to rs fo r Sud d e n Infa nt De a th Synd ro me in So uth Ea st Eng la nd b e fo re the 1991 na tio na l 'b a c k to sle e p ' he a lth 20 e d uc a tio n c a m p a ig n." Jo urna l o f Pub lic He a lth Me d ic ine 17 (3): 282-289. Fle m ing , P. J., P. Bla ir, e t a l. (2000). Sud d e n Infa nt De a th Synd ro m e : Mo d ifia b le risk fa c to rs a nd the wind o w o f vulne ra b ility. Sle e p a nd Bre a thing in C hild re n: A De ve lo p me nta l Ap p ro a c h. G . M. Lo ug hlin, J. L. C a rro ll a nd C . L. Ma rc us, Ma rc e l De kke r: 443-464. Fle m ing , P. J., R. G ilb e rt, e t a l. (1990). "Inte ra c tio n b e twe e n Be d d ing a nd Sle e p ing Po sitio n in the Sud d e n Infa nt De a th Synd ro m e : A Po p ula tio n Ba se d C a se -C o ntro l Stud y." British Me d ic a l Jo urna l 301: 85-89. G ib so n, E., C . A. De m b o fsky, e t a l. (2000). "Infa nt sle e p p o sitio n p ra c tic e s 2 ye a rs into the "Ba c k to Sle e p " C a m p a ig n." C linic a l Pe d ia tric s 39(5): 285-289. G o me z Pa p i, A. Ba ig e s-No g ue s., M. T.; Ba tiste Fe rna nd e z, M. T.; Ma re a G utie rre s, M. M.; Nie to Jura d o , A.; C lo sa Mo na ste ro lo , R. (1998). "Ka ng a ro o m e tho d in d e live ry ro o m fo r fullte rm b a b ie s." Anna ls Esp a na Pe d ia tric a . 48(6): 631-3. Ha mlyn, B. (2002). Infa nt Fe e d ing 2000: a surve y c o nd uc te d o n b e ha lf o f the De p a rtm e nt o f He a lth, the Sc o ttish Exe c utive , the Na tio na l Asse m b ly fo r Wa le s a nd the De p a rtme nt o f He a lth, So c ia l Se rvic e s a nd Pub lic Sa fe ty in No rthe rn Ire la nd . Lo nd o n, TSO . Hrd y, S. B. (1999). Mo the r Na ture : A histo ry o f mo the rs, infa nts a nd na tura l se le c tio n. Pa nthe o n, Ne w Yo rk. Lud ing to n-Ho e , S. M. A. G . C ., Sim p so n, S., Ho lling she a d , A., Arg o te , L.A., Re y, H. (1999). "Birthre la te d fa tig ue in 34-36-w e e k p re te rm ne o na te s: ra p id re c o ve ry with ve ry e a rly ka ng e ro o (skin-to -skin) c a re ." Jo urna l o f O b ste tric , G yne c o lo g ic a nd Ne o na ta l Nursing 28(1): 94-103. Ma rke sta d , T., B. Ska d b e rg , e t a l. (1995). "Sle e p ing p o sitio n a nd sud d e n infa nt d e a th synd ro me (SIDS): e ffe c t o f a n inte rve ntio n p ro g ra m m e to a vo id p ro ne sle e p ing ." Ac ta Pa e d ia tric a 84: 375-378. Ma tthie se n, A., A. Ra nsjo -Arvid so n, Nisse n, E.; Uvna s-Mo b e rg , K. (2001). "Po stp a rtum m a te rna l o xyto c in re le a se b y ne wb o rns: e ffe c ts o f infa nt ha nd ma ssa g e a nd suc king ." Birth 28(1): 13-9. Mc Ke nna , J., M. And o lina , e t a l. (2001). "Be d sha ring / b re a stfe e d ing m o the rs a nd infa nts: 21 a d a p ta tio n o r p a tho lo g y? " Am e ric a n Jo urna l o f Physic a l Anthro p o lo g y 70th Annua l Me e ting Ab stra c ts. Mc Ke nna , J. a nd S. Mo sko (2001). Mo the r-infa nt c o sle e p ing : to wa rd a ne w sc ie ntific b e g inning . Sud d e n Infa nt De a th Synd ro m e : p ro b le m s, p ro g re ss & p o ssib ilitie s. R. Bya rd a nd H. Kro us. Ne w Yo rk, Arno ld : 258-272. Mc Ke nna , J. J. (1990). "Evo lutio n a nd Sud d e n Infa nt De a th Synd ro m e (SIDS): Pa rt 1: Infa nt re sp o nsivity to p a re nta l c o nta c t." Hum a n Na ture 1(2): 145-177. Mc Ke nna , J. J. (2000). C ultura l influe nc e s o n infa nt a nd c hild ho o d sle e p b io lo g y, a nd the sc ie nc e tha t stud ie s it: to wa rd a m o re inc lusive p a ra d ig m . Sle e p a nd Bre a thing in C hild re n: A De ve lo p me nta l Ap p ro a c h. G . M. Lo ug hlin, J. L. C a rro ll a nd C . L. Ma rc us, Ma rc e l De kke r. Mc Ke nna , J. J. a nd N. J. Be rnsha w (1995). Bre a stfe e d ing a nd infa nt-p a re nt c o sle e p ing a s a d a p tive stra te g ie s: a re the y p ro te c tive a g a inst SIDS. Bio c ultura l Pe rsp e c tive s. P. Stua rt-Ma c a d a m a nd K. De ttwyle r (e d s). Ne w Yo rk, Ald ine De G ruyte r: 265-305. Mc Ke nna , J.J. a nd T. Mc Da d e (2005) “ Why b a b ie s sho uld ne ve r sle e p a lo ne : A re vie wo f the c o -sle e p ing c o ntro ve rsy in re la tio n to SIDS, b e d -sha ring a nd b re a stfe e d ing .” Pa e d ia tric Re sp ira to ry Re vie w s 6: 134-152. Mc Ke nna , J. J., S. Mo sko , e t a l. (1990). "Sle e p a nd a ro usa l p a tte rns o f c o -sle e p ing huma n m o the r/ infa nt p a irs: a p re lim ina ry p hysio lo g ic a l stud y with imp lic a tio ns fo r the stud y o f infa nt d e a th synd ro me (SIDS)." Am e ric a n Jo urna l o f Physic a l Anthro p o lo g y 83: 331-347. Mc Ke nna , J. J. a nd S. S. Mo sko (1994). "Sle e p a nd a ro usa l, sync hro ny a nd ind e p e nd e nc e , a m o ng m o the rs a nd infa nts sle e p ing a p a rt a nd to g e the r (sa m e b e d ): a n e xp e rim e nt in e vo lutio na ry me d ic ine ." Ac ta Pa e d ia tric a Sup p le me nt 397: 94-102. Mc Ke nna , J. J., S. S. Mo sko , e t a l. (1997). "Be d sha ring p ro m o te s b re a stfe e d ing ." Pe d ia tric s 100: 214-219. Mc Ke nna , J. J., E. B. Tho m a n, e t a l. (1993). "Infa nt-p a re nt c o -sle e p ing in a n e vo lutio na ry p e rsp e c tive : imp lic a tio ns fo r und e rsta nd ing infa nt sle e p d e ve lo p me nt a nd the Sud d e n Infa nt De a th Synd ro m e ." Sle e p 16(3): 263-281. 22 Mitc he ll, E. A. a nd R. Sc ra g g (1993). "Are infa nts sha ring a b e d with a no the r p e rso n a t inc re a se d risk o f Sud d e n Infa nt De a th Synd ro m e ? " Sle e p 16(4): 387-389. Mo sko , S., C . Ric ha rd , e t a l. (1997a ). "Infa nt a ro usa ls d uring mo the r-infa nt b e d sha ring : imp lic a tio ns fo r infa nt sle e p a nd Sud d e n Infa nt De a th Synd ro m e re se a rc h." Pe d ia tric s 100 (5): 841-850. Mo sko , S., C . Ric ha rd , e t a l. (1997b ). "Ma te rna l sle e p a nd a ro usa ls d uring b e sha ring with infa nts." Sle e p 20(2): 142-150. Mo sko , S., C . Ric ha rd , e t a l. (1997c ). "Ma te rna l p ro xim ity a nd infa nt C O 2 e nviro nme nt d uring b e d sha ring a nd p o ssib le imp lic a tio ns fo r SIDS re se a rc h." Am e ric a n Jo urna l o f Physic a l Anthro p o lo g y 103: 315-328. Na ka mura , S., M. Wind , e t a l. (1999). "Re vie w o f ha za rd s a sso c ia te d with c hild re n p la c e d in a d ult b e d s." Arc hive s o f Pe d ia tric a nd Ad o le sc e nt Me d ic ine 153: 1019-1023. Nisse n, E. L., G ., Wid stro m , A.M., Uvna s-Mo b e rg , K (1995). "Ele va tio n o f o xyto c in le ve ls e a rly p o st p a rtum in wo m e n." Ac ta O b ste tric ia e t G yne c o lo g ic a Sc a nd ina vic a 74(7): 530-3. Ric ha rd , C ., S. Mo sko , e t a l. (1996). "Sle e p ing p o sitio n, o rie nta tio n, a nd p ro xim ity in b e d sha ring infa nts a nd mo the rs." Sle e p 19(9): 685-690. Sc ra g g , R. K. R. a nd E. A. Mitc he ll (1998). "Sid e sle e p ing p o sitio n a nd b e d sha ring in the sud d e n infa nt d e a th synd ro me ." Anna ls o f Me d ic ine 30: 345-349. Sc ra g g , R. K. R., E. A. Mitc he ll, e t a l. (1996). "Infa nt ro o m -sha ring a nd p ro ne sle e p p o sitio n in sud d e n infa nt d e a th synd ro me ." La nc e t 347: 7-12. Ska d b e rg , B. T., I. Mo rild , e t a l. (1998). "Ab a nd o ning p ro ne sle e p ing :e ffe c ts o n the risk o f Sud d e n Infa nt De a th Synd ro m e ." The Jo urna l o f Pe d ia tric s 132(2): 340-343. Sm a ll, M. (1992). "A Re a so na b le Sle e p ." Disc o ve r Ap ril: 83-88. Sm a ll, M. F. (1998). O ur b a b ie s o urse lve s - ho w b io lo g y a nd c ulture sha p e the wa y we p a re nt. Ne w Yo rk, Do ub le d a y De ll Pub lishing G ro up Inc . Ta p p in D, Ec o b R, Bro o ke H. (2005) Be d sha ring , ro o m sha ring , a nd sud d e n infa nt d e a th synd ro m e in Sc o tla nd : a c a se -c o ntro l stud y. Jo urna l o f Pe d ia tric s. 147(1):32-7 23 Tre va tha n, W. R. a nd J. J. Mc Ke nna (1994). "Evo lutio na ry e nviro nm e nts o f hum a n b irth a nd infa nc y: insig hts to a p p ly to c o nte mp o ra ry life ." C hild re n's Enviro nme nts 11(2): 88-104. Tuffne ll, C . S., S. A. Pe te rse n, e t a l. (1996). "Hig he r re c ta l te m p e ra ture s in c o -sle e p ing infa nts." Arc hive s o f Dise a se in C hild ho o d 75: 249-250. Va re nd i, H., R.H. Po rte r (2001) “ Bre a st o d o ur a s the o nly m a te rna l stimulus e lic its c ra wling to wa rd s the o d o ur so urc e .” Ac ta Pa e d ia tric a 90 (4): 372-375. Via l-C o urm o nt M. The "Ka ng a ro o " wa rd Me d Wie ku Ro zwo j. 2000;4(2 Sup p l 3):105-17 Wig fie ld , R. E., P. J. Fle m ing , e t a l. (1992). C a n the Fa ll in Avo n's Sud d e n Infa nt De a th Ra te b e Exp la ine d b y C ha ng e s in Sle e p ing Po sitio n? British Me d ic a l Jo urna l. 304: 282-283. Willing e r, M., C . W. Ko , e t a l. (2000). "Fa c to rs a sso c ia te d with c a re g ive rs' c ho ic e o f infa nt sle e p p o sitio n, 1994-1998: the Na tio na l Infa nt Sle e p Po sitio n Stud y." Jo urna l o f the Ame ric a n Me d ic a l Asso c ia tio n 283(16): 2135-42. Yo ung , J. (1999). PhD the sis: Nig ht-tim e b e ha vio ur a nd inte ra c tio ns b e twe e n m o the rs a nd the ir infa nts o f lo w risk fo r SIDS: a lo ng itud ina l stud y o f ro o m sha ring a nd b e d sha ring . Institute o f Infa nt a nd C hild He a lth, Unive rsity o f Bristo l. Ta b le 1: So c io -d e m o g ra p hic c ha ra c te ristic s o f fa milie s p a rtic ip a ting in stud y To ta l sa m ple Bre a stfe e de rs No n- b re a stfe e de rs Mo the r’ s a g e 28 ye a rs 30 ye a rs 25 ye a rs ns Fa the r’ s a g e 31 ye a rs 34 ye a rs 27 ye a rs ns Infa nt’ s a g e 17 wks 18 wks 15 wks ns Mo the r’ s e d uc a tio n Up to a g e 18 Up to a g e 20 To a g e 17 ns Fa the r’ s e d uc a tio n Up to a g e 18 Up to a g e 19 To a g e 18 ns Ho use ho ld inc o me £21,000 £29,000 £10,000 p =0.025 Pa rity 1.5 1.6 1.3 ns 24 Ta b le 2: C o m p a riso n o f b re a stfe e d ing m o the r-infa nt d ya d s b e d -sha ring in o wn ho m e s (c urre nt stud y), ho sp ita l sle e p la b in C a lifo rnia (Ric ha rd , Mo sko e t a l. 1996) a nd ho sp ita l sle e p la b in Bristo l UK (Yo ung 1999). C urre nt study C a lifo rnia study Bristo l study (in ho m e ) Ho sp ita l la b Ho sp ita l la b 10 p a irs 6 p a irs 5 p a irs Ma te rna l a g e 28.8 (19-36) ye a rs 27 (18-37) ye a rs 35 (25-39) ye a rs Infa nt a g e 2-6 m o nths 2-4 m o nths 1-5 m o nths 80 ho urs ~48 ho urs 30 ho urs Do ub le to King Ho sp ita l Do ub le b e d 4ft 6in to 6ft b e d , 3 ft 4ft 6in Mo the r fa c e d infa nt 76% (m e d ia n) 74% 71% Infa nt fa c e d mo the r 97% (m e d ia n) 87% 89% Fa c e -to -fa c e o rie nta tio n 65% (m e d ia n) 68% 56% Infa nt sup ine 25% (m e d ia n) 62% 72% (m e d ia n) Infa nt la te ra l 74% (m e d ia n) 38% 13% (m e d ia n) Infa nt p ro ne 0% (m e d ia n) 0% 0% (m e d ia n) Pro ximity (he a d ) No d a ta <20c m 20-60c m Pro ximity (b o d y) <20 c m (91%) No d a ta <20c m (99%) Bf fre q ue nc y (who le nig ht) 3.0 (m e d ia n) 4.7 (m e a n) 3 (m e d ia n) To ta l fe e d ing tim e d uring nig ht 36 mins (m e d ia n) 56 mins (m e a n) 34 mins (m e d ia n) Num b e r o f ro utine ly b e d -sha ring (RBS) m o the r-infa nt p a irs Ho urs o f RBS m o the r-infa nt b e d sha ring o b se rva tio na l d a ta Size o f b e d fo r b e d -sha ring Pro p o rtio n o f sle e p time fo r whic h: 25 Ta b le 3: C o m p a riso n o f b re a stfe e d ing m o the r-infa nt p a irs in d ya d ic a nd tria d ic (w ith fa the r) sle e p ing a rra ng e me nts Dya dic nig ht Tria dic nig ht 8:06 ho urs 8:00 ho urs Mo the r fa c ing infa nt 76% 73% NS Infa nt fa c ing m o the r 97% 65% NS Fa c e -to -fa c e o rie nta tio n 65% 47% NS Infa nt sup ine 25% 40% NS Infa nt la te ra l 74% 54% NS Infa nt p ro ne 0% 0% NS Infa nt in mid d le o f b e d (b e twe e n p a re nts o n TN) 0% 65% P=0.028 Infa nt o n o utsid e o f b e d ne xt to mo the r 98% 35% P=0.028 To uc hing 71% 75% NS > 10c m a p a rt 20% 9% Ns Fe e d fre q ue nc y / nig ht 3 b o uts 2.5 b o uts NS To ta l fe e d ing tim e 36 minute s 31 minute s NS Numb e r ma te rna l a ro usa ls p e r nig ht 6 (ra ng e =3-9) 4 (ra ng e =3-5) P=0.037 Numb e r o f infa nt a ro usa ls p e r nig ht 5 (ra ng e =2-9) 3 (ra ng e =2-5) P=0.043 Numb e r o f mutua l a ro usa ls p e r nig ht 5 (ra ng e =2-9) 3 (ra ng e =1-4) P=0.028 Ho urs o f b e d -sha ring d a ta (m e d ia n) O rie nta tio n Infa nt sle e p p o sitio n Infa nt b e d p o sitio n Mo the r-infa nt p ro ximity Bre a stfe e d ing Awa ke ning fre q ue nc y All a ve ra g e s a re e xp re sse d a s me d ia ns. All te sts o f sig nific a nc e fo r within-sub je c t c o mp a riso ns we re c o nd uc te d using Wilc o xo n sig ne d ra nks te st. 26 Ta b le 4: Infa nt he a d -c o ve ring o n d ya d ic a nd tria d ic b e d -sha ring nig hts. Dya dic Tria dic Sig C o ve rs o ve r b a b y's he a d (whe n p a re nts 0% me d ia n 0% me d ia n NS a sle e p ) (0-45%) (0-47%) C o ve rs a b o ve b a b y's wa ist 96.1% m e d ia n 72.1 (m e d ia n) (27-100%) (28-100%) 15.0% m e d ia n 31.6% me d ia n (0-91.5%) (0-81.7%) Ba b y ro ta te d a c ro ss b e d (whe n p a re nts 1.9% m e d ia n 0% me d ia n Z=-2.023 a sle e p ) (0-76.9%) (0-37.4%) p =0.043 Infa nt fa c e a t m o the r's c he st he ig ht (whe n a ll 83.3% me d ia n 100% me d ia n NS a sle e p ) (38.9-100%) (36.9-100%) C o ve rs a t m o the r's c hin le ve l o r a b o ve 27 NS NS Ta b le 5: C o m p a riso n o f b e d -sha ring c ha ra c te ristic s fo r b re a stfe d a nd fo rm ula -fe d infa nts. Fo rm ula - fe d Bre a stfe d 7.6 ho urs 8.0 ho urs Mo the r fa c ing infa nt 59% 73% z=-1.96, p =0.05 Infa nt fa c ing m o the r 46% 65% ns Fa c e -to -fa c e o rie nta tio n 32% 47% z=-2.277, p =0.02 Fa the r fa c ing infa nt 46% 26% ns Infa nt fa c ing fa the r 20% 9% ns Fa c e -to -fa c e o rie nta tio n 8% 0% ns Infa nt sup ine 83% 40% z=-2.328, p =0.02 Infa nt la te ra l 6% 54% z=-2.414, p =0.02 Infa nt p ro ne 0% 0% ns To uc hing m o the r 60% 75% ns To uc hing fa the r 20% 6% ns Mo the r c urle d up ro und infa nt 25% 49% ns Fa the r c urle d up ro und infa nt 0% 0% ns Infa nt fa c e le ve l with m o the r's fa c e o r c hin 71% 0% z=-2.472, p =0.01 Infa nt fa c e le ve l with m o the r's c he st 29% 100% z=-2.387, p =0.02 Infa nt fa c e b e lo w m o the r's c he st he ig ht 0% 0% ns Ve rtic a l p o sitio n (p e rp e nd ic ula r to he a d b o a rd ) 94% 86% ns Ro ta te d b e twe e n 30-60o to e ithe r sid e 5% 13% ns Fe e d fre q ue nc y / nig ht 1.0 b o uts 2.5 b o uts z=-2.599, p =0.009 To ta l fe e d ing tim e 9 m inute s 31 m inute s z=-2.699, p =0.007 Num b e r m a te rna l a ro usa ls p e r nig ht 2 (ra ng e =0-4) 4 (ra ng e =3-5) z=-3.183, p =0.001 Num b e r o f infa nt a ro usa ls p e r nig ht 2 (ra ng e =0-3) 3 (ra ng e =2-5) z=-2.728, p =0.006 Num b e r o f m utua l a ro usa ls p e r nig ht 1 (ra ng e =0-2) 3 (ra ng e =1-4) z=-2.988, p =0.003 Ho urs o f b e d -sha ring d a ta (m e d ia n) Sig nific a nc e O rie nta tio n to m o the r* O rie nta tio n to fa the r* Infa nt sle e p p o sitio n Pa re nt-infa nt p ro xim ity Pa re nt-Infa nt sle e p ing p o sitio n He ig ht o f infa nt in b e d re la tive to m o the r Dire c tio n o f infa nt in b e d Fe e d ing Awa ke ning fre q ue nc y 28 Fig ure 1: C ha ra c te ristic sle e p p o sitio n o f b re a stfe e d ing m o the rs a nd infa nts 29