Exploring the influence of medical staffing and birth volume on observed-to-expected cesarean deliveries: a panel data analysis of integrated obstetric and gynecological departments in Germany

S öcke , A no; P a , Holge ; Schol en, Nadine; Kun z, Ludwig
A icle — Published Ve sion
Explo ing he in luence o medical s a ing and bi h olume on
obse ed- o-expec ed cesa ean deli e ies: a panel da a analysis o
in eg a ed obs e ic and gynecological depa men s in Ge many
The Eu opean Jou nal o Heal h Economics
P o ided in Coope a ion wi h:
Sp inge Na u e
Sugges ed Ci a ion: S öcke , A no; P a , Holge ; Schol en, Nadine; Kun z, Ludwig (2025) : Explo ing
he in luence o medical s a ing and bi h olume on obse ed- o-expec ed cesa ean deli e ies:
a panel da a analysis o in eg a ed obs e ic and gynecological depa men s in Ge many, The
Eu opean Jou nal o Heal h Economics, ISSN 1618-7601, Sp inge , Be lin, Heidelbe g, Vol. 26, Iss. 6,
pp. 987-1022,
h ps://doi.o g/10.1007/s10198-024-01749-0
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/330433
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h ps://c ea i ecommons.o g/licenses/by/4.0/
Vol.:(0123456789)
The Eu opean Jou nal o Heal h Economics (2025) 26:987–1022
h ps://doi.o g/10.1007/s10198-024-01749-0
ORIGINAL PAPER
Explo ing hein luence o medical s a ing andbi h olume
onobse ed‑ o‑expec ed cesa ean deli e ies: apanel da a analysis
o in eg a ed obs e ic andgynecological depa men s inGe many
A noS öcke 1,4,5 · Holge P a 1,4· NadineSchol en2,4,5· LudwigKun z3,4
Recei ed: 18 Sep embe 2024 / Accep ed: 3 Decembe 2024 / Published online: 21 Janua y 2025
© The Au ho (s) 2025
Abs ac
In oduc ion Cesa ean deli e ies accoun o app oxima ely one- hi d o all bi hs in Ge many, p omp ing ongoing dis-
cussions on cesa ean sec ion a es and hei connec ion o medical s a ing and bi h olume. In Ge many, he majo i y o
depa men s in eg a e obs e ic and gynecological ca e wi hin a single depa men .
Me hods The analysis u ilized quali y epo s om Ge man hospi als spanning 2015 o 2019. The ou come a iable was
he annual isk-adjus ed cesa ean sec ion a io—a me ic compa ing expec ed o obse ed cesa ean sec ions. Explana o y
a iables included annual coun s o physicians, midwi es, and bi hs. To accoun o case numbe - ela ed s a ing a ia ions,
ull- ime equi alen midwi e and physician s a posi ions we e no malized by he numbe o deli e ies. Uni- and mul i a ia e
panel models we e applied, complemen ed by mul iple ins umen a iable analyses, including wo-s age leas squa e and
gene alized me hod o momen s models.
Resul s Inco po a ing da a om 509 in eg a ed obs e ic depa men s and 2089 obse a ions, ep esen ing 2,335,839 deli -
e ies wi h 720,795 cesa ean sec ions (o e 60% o all inpa ien bi hs in Ge many), mul i a ia e model wi h ixed e ec s
e ealed a s a is ically signi ican posi i e associa ion be ween he numbe o physicians pe bi h and he isk-adjus ed
cesa ean sec ion a io (0.004, p = 0.004). Two-s age leas squa e ins umen a iable analysis (0.020, p < 0.001) and a sys em
GMM es ima o models (0.004, p < 0.001) alida ed hese esul s, p o iding compelling e idence o a causal ela ionship.
Conclusion The s udy es ablished a obus connec ion be ween he numbe o physicians pe bi h and he isk-adjus ed
cesa ean sec ion a io in in eg a ed obs e ic and gynecological depa men s in Ge many. While he cause o he e ec emains
unclea , one possible explana ion is a lack o specializa ion wi hin hese depa men s due o he combined p o ision o bo h
obs e ic and gynecological ca e.
Keywo ds Longi udinal design· Obs e ic ca e· Inpa ien sec o · Hospi al· O ganiza ional managemen · Medical s a ing
JEL Classi ica ion C23· I10· N34· P46
* A no S öcke
a no.s oeck[email p o ec ed]
1 Facul y o Human Sciences andFacul y o Medicine
andUni e si y Hospi al Cologne, Ins i u e o Medical
Sociology, Heal h Se ices Resea ch andRehabili a ion
Science, Chai o Quali y De elopmen andE alua ion
inRehabili a ion, Uni e si y o Cologne, Cologne, Ge many
2 Facul y o Medicine andUni e si y Hospi al Cologne,
Ins i u e o Medical Sociology, Heal h Se ices Resea ch
andRehabili a ion Science, Chai o Heal h Se ices
Resea ch, Uni e si y o Cologne, Cologne, Ge many
3 Depa men o Business Adminis a ion andHeal h Ca e
Managemen , Facul y o Managemen , Economics andSocial
Sciences, Uni e si y o Cologne, Cologne, Ge many
4 Cen e o Heal h Se ices Resea ch Cologne, In e acul y
Ins i u ion o  heUni e si y o Cologne, Cologne, Ge many
5 Cen e o Heal h Communica ion andHeal h Se ices
Resea ch, Depa men o Psychosoma ic Medicine
andPsycho he apy, Facul y o Medicine, Uni e si y Hospi al
Bonn, Bonn, Ge many
988 A.S öcke e al.
In oduc ion
O e he pas ew decades, he global p e alence o cesa -
ean sec ion (C-sec ion) bi hs has s eadily inc eased [1].
Some expe s and esea che s go so a as o cha ac e ize
his end as an endemic o pandemic o C-sec ions [2, 3].
This is no ewo hy gi en ha aginal deli e y is gene ally
ega ded as he p e e ed me hod compa ed o cesa ean
bi h [4]. In Ge many, he C-sec ion a e has doubled since
1991 and has s abilized a app oxima ely 30% since he
mid-2000s [5]. The eby, indica ions o a cesa ean sec-
ion can be di ided in o absolu e and ela i e indica ions.
This inc ease is p edominan ly a ibu ed o ela i e indi-
ca ions, such as b eech p esen a ion, bi h a es , impend-
ing e al hypoxia, and pos -sec ion condi ions [6], wi h
90% o C-sec ions lacking absolu e medical indica ions
[7]. Thus, unde s anding non-medical ac o s, including
o ganiza ional in luences, is impe a i e o a comp ehen-
si e s udy o deli e y me hods.
Gi en he impe a i e o heal h ca e p o ide s o align
hei ac ions wi h medical necessi ies and pa ien p e -
e ences, comp ehending he ac o s in luencing medical
p ac ice decisions is c ucial. In he case o C-sec ions,
ob iously ma e nal desi e is a key ac o [6, 8–10]. F om
he p o ide 's pe spec i e, nume ous ac o s in luence he
choice o bi h me hod. These encompass indi idual ac-
o s o he physician [3, 8, 11–14] o midwi e [15, 16], as
well as o ganiza ional ac o s a he hospi al o obs e ic
depa men le el [13, 17–22]. Addi ionally, con ex - ela ed
ac o s, such as he complexi ies o caseload [23], con ib-
u e o he decision-making p ocess. S udies on C-sec ion
a es and he quali y o obs e ic ca e e eal o ganiza ion-
speci ic di e ences linked o ac o s such as hospi al
owne ship, numbe o beds, and eaching ac i i y, bo h
in Ge many [17–19, 24, 25] and in e na ionally [26–30].
While a ia ions be ween depa men s wi h di e -
en o ganiza ional ac o s ha e been ex ensi ely s ud-
ied, unde s anding ela ionships and in e ac ions wi hin
obs e ic depa men s is equally i al. These in e -o gani-
za ional in luences can po en ially impac all depa men s
uni o mly, independen o speci ic o ganiza ional ac o s
such as cul u e, egion, socio-demog aphics, and pa ien
popula ion e ec s. F om an o ganiza ional s andpoin ,
comp ehending hese in luences is c ucial o making
in o med decisions abou medical p ac ices. O ganiza-
ions can enhance hei unde s anding o he ac o s shap-
ing medical p ac ice wi hin hei pu iew and implemen
adjus men s acco dingly [31].
This analysis ocuses on wo o ganiza ional ac-
o s, namely, medical s a ing and caseload olume, and
explo es hei in e play. Bo h ac o s ha e been acknowl-
edged o hei in luence on he quali y and quan i y o
obs e ic ca e in gene al [27, 32] and speci ically on
C-sec ions (e.g., olume o bi hs [33, 34] o he numbe
o physicians/deli e ies pe physician pe yea [35–37]).
O e all, medical s a ing le els ha e been less equen ly
in es iga ed han bi h olume as an explana o y a i-
able. Ou s udy enhances cu en unde s anding h ough a
longi udinal in es iga ion ha encompasses a signi ican
po ion o Ge man obs e ic depa men s and deli e ies.
Impo an ly, by employing an ex e nally e alua ed isk-
adjus ed C-sec ion a io, ou analysis add esses me hodo-
logical limi a ions by accoun ing o medical indica ions
on he pa o bo h mo he and child. Consequen ly, he
a io and he analysis emain obus , mi iga ing biases
induced by unde lying medical easons ha may p omp
a C-sec ion [38].
Ou esea ch ques ion del ed in o how he numbe o
ull- ime equi alen physicians and midwi es pe bi h,
coupled wi h he olume o bi hs, impac he isk-adjus ed
C-sec ion a io in Ge man combined obs e ic and gyneco-
logical depa men s. In doing so, we add ess me hodologi-
cal sho comings by ocusing on di e ences wi hin hospi al
depa men s, con ibu ing o he unde s anding o ac o s
independen o a iances be ween hospi al depa men s.
U ilizing a panel model da a se wi h indi idual and ime
e ec s, we con ol o depa men - and pa ien popula ion-
speci ic cha ac e is ics, ensu ing a comp ehensi e analysis
[39].
Me hods
Da a sou ce
This analysis was based on he s uc u ed quali y epo s o
Ge man hospi als, which se e as a co ne s one o ans-
pa ency in he inpa ien sec o in Ge many [40–44]. These
epo s a e compiled and disclosed in acco dance wi h Sec s.
136 and 137o he Ge man Social Secu i y Code (Sozi-
algese zbuch) V. They o e a comp ehensi e o e iew o
he s uc u es, se ices, and quali y o he espec i e hos-
pi al and i s specialis depa men s. Annual epo s include
documen a ion o pe o mance, o ganiza ional me ics, and
a ious quali y indica o s and sco es. De ails a e delinea ed
o he depa men le el. Al hough his da a se has been u i-
lized o c oss-sec ional analyses o a ious indica o s [18,
40–42, 45, 46], i s applica ion in he con ex o longi udinal
analyses wi h a speci ic ocus on obs e ics is no el. Du ing
he s udy pe iod, app oxima ely 30 quali y indica o s ela ed
o obs e ic ca e we e published (Appendix Table7). Ou
analysis cen e s on he quali y indica o o isk-adjus ed
C-sec ion a es pe hospi al.
In p epa a ion o his analysis, a panel da a se was
compiled using annually published quali y epo s. In hese
989
Explo ing hein luence o medical s a ing andbi h olume onobse ed‑ o‑expec ed cesa ean…
epo s each hospi al is assigned a unique iden i ica ion code,
and each si e wi hin a hospi al is designa ed a si e numbe .
Asp e iously ou lined [47, 48], iden i ica ion codes and si e
numbe s may unde go changes o e he yea s. The e o e,
au oma ed linking wi hou con en e i ica ion could yield a
success a e anging om 80 o 90%. Recognizing po en ial
changes in iden i ica ion codes and si e numbe s o e he
yea s, a manual ma ching p ocess was employed o ensu e
accu acy, supplemen ing a machine linkage ia iden i ica ion
codes and si e numbe s. The linkage inco po a ed hospi al
and si e add esses, bed numbe s, and he names o espon-
sible depa men and gene al hospi al manage s. Two inde-
penden esea che s pe o med he manual linkage, esol ing
dispu ed assignmen s h ough consensus. Each hospi al and
si e ecei ed a mas e iden i ica ion numbe , o ming he
ounda ion o he panel s udy.
As his analysis employs seconda y da a, and he da a a e
publicly a ailable, no e hical commi ee o e was deemed
necessa y.
S udy popula ion
In Ge many, he as majo i y o deli e ies occu in hos-
pi als, wi h less han wo pe cen aking place a home o
in bi hing cen e s. Obs e ic ca e is p edominan ly admin-
is e ed in depa men s specializing in bo h obs e ics and
gynecology simul aneously. Consequen ly, wo closely
ela ed bu inc easingly dis inc medical se ices a e o e ed
wi hin he same depa men s. Gi en he inhe en o ganiza-
ional dispa i ies be ween hose in eg a ed obs e ics depa -
men s and depa men s p ima ily dedica ed on obs e ic ca e,
we excluded obs e ic depa men s wi h a p ima y ocus on
obs e ic ca e om ou analysis. Impo an ly, he e a e no
sys ema ic di e ences be ween hese wo ypes o depa -
men s in e ms o pa ien popula ion o medical s anda ds.
The cu -o alue was de e mined h ough a da a-d i en
app oach, as in o ma ion on o ganiza ional ocus was no
consis en ly o con inuously included in he quali y epo s.
Addi ionally, depa men s wi h a ending physicians we e
excluded due o hei ypically highe C-sec ion a es in Ge -
many [17, 18, 25].
Ensu ing da a in eg i y, we iden i ied po en ial ou lie s
h ough Cook's dis ance and subsequen ly e i ied hem
manually. Closed depa men s we e con i med h ough
c oss- e e encing wi h p ess epo s. The exclusion was
deemed necessa y o a oid inaccu acies esul ing om sce-
na ios whe e a depa men , ope a ional o only six mon hs,
epo ed s a ing numbe s analogously o he en i e yea .
Consequen ly, s a ing igu es we e no p opo iona ely
epo ed in he quali y epo , and he numbe o deli e ies
was only accoun ed o du ing he speci ied 6-mon h pe iod.
This me iculous app oach was adop ed o main ain he accu-
acy and eliabili y o he da ase .
S udy pe iod
The obse a ion pe iod spans om 2015 o 2019, as he
ele an quali y indica o on he cesa ean sec ion a io was
in oduced in 2015, and 2019 ma ks he las yea be o e he
onse o he COVID-19 pandemic. Recognizing he po en-
ial impac o measu es associa ed wi h he pandemic on
daily hospi al p ac ices [49, 50], da a om 2020 onwa ds
was excluded o minimize bias. Mo eo e , he in oduc ion
o he Robson indica o o he quali y indica o in 2020 [51]
u he complica es longi udinal compa isons.
Measu es
Ou come a iable
The dependen a iable in he s udy was he isk-adjus ed
cesa ean sec ion a io a he depa men le el (quali y indi-
ca o no. 52249). This a io is calcula ed as he obse ed
numbe o C-sec ions di ided by he expec ed numbe o
C-sec ions. The decla ed quali y objec i e is he minimi-
za ion o cesa ean bi hs [51]. The nume a o includes all
obse ed cesa ean deli e ies wi hin a depa men , while
he denomina o comp ises isk-adjus ed expec ed cesa ean
deli e ies. In 2015 and 2016, all mo he s wi h a leas one
child bo n a e 24weeks we e included, while om 2017
onwa ds, only mo he s deli e ing be ween weeks 24 and 42
we e conside ed. Risk adjus men is pe o med by he Ins i-
u e o Quali y Assu ance and T anspa ency in Heal hca e
(Ins i u ü Quali ä ssiche ung und T anspa enz im Gesund-
hei swesen [IQTIG]). While indi idual pa ien da a a e con-
side ed in his p ocess, only agg ega ed, annual depa men -
le el da a we e accessible o his analysis open access.
No ably, he esea che s in ol ed in his s udy we e no
di ec ly engaged in he isk adjus men p ocess. The selec-
ion o isk ac o s was guided by Becke and Eissle [52] in
collabo a ion wi h he Fede al Pe ina al Medicine G oup.
Risk ac o s conside ed encompassed a iables such as
ma e nal age, comp ehensi e da a on in an and ma e -
nal heal h s a us, and in o ma ion on p e ious deli e ies
(Appendix Table22). Each yea , hese isk ac o s we e
adjus ed based on hei espec i e eg ession coe icien s.
Consequen ly, he isk adjus men unde aken compensa es
o he di e gen pa ien s uc u es ac oss di e en acili ies,
o e ing a mo e equi able basis o acili y compa isons. This
adjus men p o es pi o al in ensu ing a ai assessmen , as
pa ien s b ing indi idual isk ac o s, including concomi an
diseases, ha could sys ema ically in luence he quali y ou -
come. Th ough isk adjus men , ins i u ions wi h a highe
990 A.S öcke e al.
p e alence o high- isk cases can be s a is ically jux aposed
mo e equi ably wi h hose handling a la ge p opo ion o
low- isk cases, he eby acili a ing an unbiased analysis
[53]. In he con ex o he isk-adjus ed C-sec ion a io, al-
ues below 1 indica e ha a depa men is pe o ming ewe
cesa ean sec ions han expec ed. Con e sely, a a io abo e 1
signi ies ha mo e isk-adjus ed cesa ean sec ions a e being
pe o med han an icipa ed. Fo ins ance, a C-sec ion a io
o 1.1 would imply ha 10% mo e cesa ean sec ions we e
pe o med han expec ed. Fu he mo e, a e e ence ca ego y
is de ined o each yea , designa ing depa men s abo e he
90 h pe cen ile as conspicuous. The co esponding alue o
he e iew pe iod anged om 1.23 o 1.27.
Explana o y a iables
Th ee independen a iables we e u ilized: (1) he numbe o
deli e ies pe depa men (mo he s gi ing bi h) pe 1000,
(2) he numbe o ull- ime equi alen physicians in he
depa men pe 1000 deli e ies, (3) and he numbe o ull-
ime equi alen midwi es in he depa men pe 1000 deli -
e ies. S a ing le els ha e been di ided by 1000 o accoun
o olume ela ed di e ences in he depa men s. The da a
on deli e ies we e ex ac ed om he denomina o o he
quali y indica o o each depa men . The da a on bo h s a -
ing le els we e p esen ed sepa a ely o each depa men
wi hin a hospi al. S a ing da a o physicians we e agg e-
ga ed o esiden s and specialis s, collec i ely e e ed o as
physicians in subsequen discussions. Physician assis an s
we e excluded. As hese depa men s in eg a e obs e ics and
gynecology, physicians ep esen bo h special ies. In Ge -
many, he ‘Facha z s anda d’ (specialis s anda d) ensu es
high-quali y medical ea men s and p ocedu es, ypically
ca ied ou o supe ised by a specialis . Howe e , su i-
cien ly ained esiden s o assis an physicians may also
pe o m ea men s and p ocedu es independen ly, includ-
ing cesa ean sec ions. Since he e is no s ic h eshold o
de e mining when a physician has gained su icien expe i-
ence and knowledge, we op ed o include he o al numbe
o physicians in ou a iable. In addi ion, a midwi e has o
be p esen du ing labo in Ge many (‘Hinzuziehungsp lich ’;
midwi e's obliga ion o consul ).
Con ol a iables
Th ee o ganiza ional a iables se ed as con ol a iables
in he analyses: (1) hospi al owne ship (non-p o i , p i a e,
public), (2) eaching s a us (no eaching assignmen , aca-
demic eaching hospi al, uni e si y hospi al), and (3) pe i-
na al ca e le el ( egula obs e ic depa men s [ca e le el
4], pe ina al ocus [ca e le el 3], pe ina al cen e le el II
[ca e le el 2], pe ina al cen e le el I [ca e le el 1]). Regula
obs e ic depa men s p o ide s anda d pe ina al ca e om
36 + 0weeks o ges a ion wi hou an icipa ed complica ions.
Depa men s wi h a pe ina al ocus ca e o p egnan women
expec ing p ema u e in an s wi h an es ima ed bi h weigh
o a leas 1500g o wi h a ges a ional age om 32 + 0 o
less han o equal o 35 + 6weeks. Le el II pe ina al cen e s
se e p egnan women wi h an icipa ed p ema u e in an s
weighing be ween 1250 and 1499g o wi h a ges a ional
age om 32 + 0 o less han o equal o 35 + 6weeks. Le el
I pe ina al cen e s o e he highes le el o obs e ic ca e
o p egnan women expec ing p ema u e in an s wi h an
es ima ed bi h weigh unde 1250g o wi h a ges a ional age
om 29 + 0 o less han weeks. All h ee o ganiza ional a i-
ables we e equen ly ci ed as in luen ial ac o s in C-sec ion
a es [54, 55].
Model desc ip ion
We de eloped a ious s a ic and dynamic panel models
ea u ing ime- and indi idual-speci ic e ec s ( wo-way
e ec s), employing clus e - obus es ima o s o each
depa men :
(a) Fixed e ec s es ima o models
(b) Co ela ed andom e ec s es ima o models
(c) Two s age leas squa e es ima o models
(d) Sys em gene alized me hod o momen s es ima o
(Blundell/Bond [56] and A ellano/Bo e [57]) models
whe e
yi
= ec o o dependen a iable,
xi
= ec o
o independen a iables,
xi
= ec o o clus e means
independen a iables,
xi
= ec o o es ima ed inde-
penden a iables om he i s s age leas squa e
eg ession,
zi
= ec o o ( ime cons an ) independen
yi
=
𝛽0
+
𝛽1xi
+
𝛾
+
𝛼i
+
ei
yi
=
𝛽0
+
𝛽1xi
+
𝛽2xi
+
𝛽3zi
+
𝛾
+
𝛼i
+
ei
yi
=
𝛽0
+
𝛽1
x
i
+
𝛽2
z
i
+
𝛾
+
𝛼i
+e
i
xi
=
𝛿0
+
𝛿1
i
i
+
𝛿2
z
i
+
𝜓
+
𝜙i
+
𝜐i
Model:yi =𝛽0+𝛽1xi −1+𝛽2yi −1+𝛽3zi +𝛾 +𝛼i+ei
Di e ence:y
i
−y
i −1
=
𝛽1
(x
i −1
−x
i −2
)
+𝛽2(yi −1−yi −2)
+
𝛽3(
z
i
−z
i −1)
+
(
e
i
−e
i −1)
Le el:y
i
=𝛽
0
+𝛽
1
x
i −1
+𝛽
2
y
i −1
+
𝛽3
z
i −1
+
𝛾
+
𝛼i
+e
i

991
Explo ing hein luence o medical s a ing andbi h olume onobse ed‑ o‑expec ed cesa ean…
con ol a iables,
𝛽
= ec o o model coe icien s,
𝛾
= unobse ed ime-speci ic e ec ,
𝛼i
= unobse ed
depa men -speci ic e ec ,
ei
= indi idual e o e m,
𝛿
= ec o o pa ame e s o be es ima ed in he i s
s age leas squa e eg ession,
ii
= ins umen o inde-
penden a iable,
𝜓
= unobse ed ime-speci ic e ec s
in he i s s age,
𝜙i
= unobse ed depa men -speci ic
e ec s in he i s s age,
𝜐i
= e o e m in he i s s age.
Whe e en i ies (depa men s) a e deno ed as
i
=
1,
…
,n
and obse a ion pe iods (yea s)
=
2015,
…
, 2019
. The
e m
𝛾
inco po a ed he ime e ec on he dependen a i-
able, independen o obse able o unobse able di e -
ences be ween indi idual obse a ion uni s. This empo-
al e ec cap u ed he in luence o changes o e ime on
he dependen a iable and se ed as a global empo al
componen .
𝛼i
was he ixed unobse ed he e ogenei y o
each hospi al depa men , and
ei
signi ied he e o e m
o each hospi al depa men o e ime. S anda d e o s
obus o g oup-wise he e oscedas ici y and se ial co -
ela ion we e used. To choose be ween ixed e ec s o
andom e ec s models, we u ilized a obus Hausman-
like es .
O ganiza ional cha ac e is ics in he Ge man hospi al
sec o exhibi minimal a ia ion and emain ela i ely
ime-pe sis en . Including hem in a ixed e ec model
would ende he esul s alid only in he a e e en o
a shi in one o he ca ego ies, making i unsui able o
a ue compa ison be ween di e en o ganiza ional ac-
o s. Consequen ly, alongside a model wi h ixed e ec
es ima o s, we adop ed a co ela ed andom e ec s (CRE)
modeling app oach [58] o inco po a e and con ol o
o he o ganiza ional a iables, enhancing model sensi i -
i y and speci ici y. Acknowledging po en ial unmeasu ed
con ounde s and endogenei y wi h he explana o y a i-
ables in ou panel eg ession models, wo models wi h
ins umen a iables (IV) we e cons uc ed and analyzed
o add ess possible endogenei y.
While he panel s uc u e o he da a al eady accoun ed
o some aspec s o endogenei y [59], models wi h ins u-
men al a iable es ima ion we e employed o es ablish
po en ial causal e ec s. Recommended bes p ac ices
ad oca e o he inclusion o addi ional da a o add ess
endogenei y be o e eso ing o IV es ima ion [58–61],
ou da a sou ce had limi a ions in p o iding meaning ul
po en ial addi ional a iables. The i s model employed a
s a ic IV app oach wi h a wo-s age leas squa es (2SLS)
es ima o , using he numbe o nu sing s a pe 1000
deli e ies as an ex e nal IV. The second model employed
a dynamic app oach wi h gene alized me hod o momen s
(GMM) es ima o s, inco po a ing he lagged a ios o
obse ed o expec ed a es o cesa ean bi hs as an in e -
nal IV [56, 57, 62]. Addi ionally, o minimize da a loss
due o he unbalanced da ase and use mo e ins umen s
o mo e e icien es ima o s, we e e ed o he sys em
GMM es ima o ins ead a di e ence GMM es ima o [56].
Two-s ep GMM es ima o s we e chosen o hei obus -
ness o au oco ela ion and he e oscedas ici y [63].
S a is ical analysis
Da a p epa a ion ( idy e se package [2.0.0]) and analysis
(plm package [2.6–3], lm es package [0.9–40], g sum-
ma y package [1.7.2], modelsumma y package [1.4.3])
we e pe o med in R ( e sion 4.2.2) and R S udio ( e sion
2023.06.1 + 524).
Resul s
Da a inclusion
Following he exclusion o duplica e en ies, we iden i ied
da a om 912 depa men s pe o ming obs e ic ca e, p o-
iding 3627 obse a ions om he quali y epo s (Fig.1).
Regula o y au ho i ies censo ed 48 obse a ions due o da a
p i acy conce ns (less han ou deli e ies in he epo ing
yea ), hese we e excluded om u he analysis. Addi ion-
ally, 166 obse a ions lacked da a on he C-sec ion a io
quali y indica o o he espec i e yea , leading o hei
exclusion. O he emaining da a, 360 epo ed ha ing
a ending physicians in he espec i e yea , and hese depa -
men s we e excluded. Simila ly, depa men s p ima ily
o e ing obs e ic se ices we e excluded, wi h a da a-d i en
cu -o alue se a 1.4 imes he numbe o ull inpa ien
cases compa ed o bi hs (Appendix Fig.2). 377 obse a-
ions ell below he h eshold and we e ca ego ized as solo
obs e ic depa men s, subsequen ly excluded. A manual
check using Cook's dis ance me hod e ealed 60 en ies
wi h conspicuous da a, con i med h ough c oss-checking
wi h media epo s indica ing closu es wi hin he epo ing
yea . These depa men s we e consequen ly excluded. Las ly,
527 obse a ions lacked da a on he a iable o he numbe
o midwi es, esul ing in hei exclusion. The inal analysis
included 2089 obse a ions om 519 obs e ic depa men s.
In o al, he s udy popula ion wi h 2089 obse a ions
ep esen ed 2,335,839 mo he s gi ing bi h and 720,795
C-sec ion deli e ies (Table1). The panel is unbalanced. As
he s udy popula ion no longe co esponded o a ull su ey
o Ge man obs e ic depa men s, we alida ed he s udy
popula ion agains da a om IQTIG o he o e all numbe s
on bi hs and cesa ean deli e ies in obs e ic depa men s in
Ge many. O e he pe iod 2015 o 2019, he o al numbe o
mo he s gi ing bi hs inc eased om 713,563 o 745,941,
wi h a sligh dec ease in he C-sec ion a e om 31.42 o
30.85%. These ends we e mi o ed in he smalle s udy
992 A.S öcke e al.
Fig. 1 Flow cha o s udy
popula ion
993
Explo ing hein luence o medical s a ing andbi h olume onobse ed‑ o‑expec ed cesa ean…
popula ion (444,555–479,176 bi hs). The s udy popula-
ion co e ed 62.8% o all deli e ing mo he s and 62.4% o
all cesa ean deli e ies in Ge many. While he yea -by-yea
C-sec ion a e in he s udy popula ion was sligh ly lowe
han he na ional a e o all yea s excep 2015, he di e -
ences we e no signi ican ( es , p = 0.408). No ably, he
de ia ion in 2016 was a ibu ed o missing da a in he qual-
i y epo s, discussed in mo e de ail in he limi a ions sec-
ion. Addi ionally, solo obs e ic depa men s epo ed on
a e age cesa ean a es (30.3%) simila o hose o in eg a ed
depa men s (Appendix Table8).
Cha ac e is ics o s udy popula ion
The obse ed isk-adjus ed C-sec ions, on a e age, we e
ma ginally below he expec ed numbe , indica ing a a io
o 0.98 (median 1.00) (Table2). The numbe o deli e -
ies pe depa men a e aged 1118.2 bi hs. No ably, om
2015 o 2019, he mean numbe o deli e ies exhibi ed an
inc ease o 123.1, ising om 1031.5 o 1154.6 bi hs, wi h
an in e media e spike obse ed om 2015 o 2016. Howe e ,
he median, s anding a 913, consis en ly ailed he mean.
Rega ding cesa ean bi hs, he annual a e age pe depa -
men was 345, showing a modes inc ease o 21.7 om 2015
o 2016. Ye again, he median, a 240.3 cesa ean bi hs,
lagged behind he mean.
The a e age numbe o physicians pe depa men o e
he obse a ion pe iod was 12.5, wi h a median o 10.9. This
ep esen s an inc ease o mo e han one ull- ime equi alen
om 2015 o 2019. As ou lined ea lie , he o al numbe o
ull- ime equi alen physicians was di ided by he depa -
men 's numbe o bi hs o ensu e compa abili y ac oss
depa men s. The a e age numbe o ull- ime equi alen
physicians pe 1000 deli e ies pe depa men emained con-
s an a ound 12.5, wi h a median o 11.8, showing no signi i-
can change om 2015 o 2019. In e ms o midwi es, he
a e age numbe pe depa men ho e ed a ound 11 du ing
he obse a ion pe iod, wi h a median o 9.2. The a e age
numbe o ull- ime equi alen midwi es pe 1000 deli e ies
pe depa men was 10.7 (median 10.3). Howe e , om 2015
o 2019, a e age numbe o ull- ime equi alen midwi es
pe 1000 deli e ies dec eased by nea ly one. The a e age
numbe o ull- ime equi alen nu sing s a (excluding mid-
wi es) pe depa men was 24, wi h a median o 19.6. Pe
1000 deli e ies, he e we e 23.3 ull- ime equi alen nu sing
s a , wi h a median o 21.1. No ably, he e was a dec ease in
he a e age numbe o ull- ime equi alen nu sing s a pe
1000 bi hs om 24.7 in 2015 o 22.3 in 2019.
P i a e hospi als cons i u ed he mino i y in he s udy
popula ion a 16.6%, while non-p o i hospi als (36.3%) and
public hospi als (47.1%) comp ised he majo i y. Rega ding
academic eaching, 70.5% o hospi als we e lis ed as aca-
demic eaching hospi als, 7.6% we e uni e si y hospi als,
and 22% we e no engaged in academic eaching. Beyond
egula obs e ic depa men s, he landscape encompasses
acili ies specializing in pe ina al ca e, including le el 1
and 2 pe ina al cen e s. Addi ionally, 45% ca ego ized as
depa men s wi h he egula le el o pe ina al ca e. 19.4%
epo ed a pe ina al ocus, 8.9% we e a le el II pe ina al
cen e , and 26.8% we e a le el I pe ina al cen e — he high-
es o m o obs e ic heal h se ices o high in ense ca e.
Compa ed o in eg a ed obs e ic depa men s, solo
depa men s, on a e age, ha e a signi ican ly lowe isk-
adjus ed cesa ean sec ion a io (0.95; es , p = 0.001), mo e
deli e ies (1563.5; es , p < 0.001), and mo e cesa ean sec-
ions (473.7; es , p < 0.001), bu ewe ull- ime equi alen
physicians (9.7; 6.8 pe 1000 deli e ies; es , p < 0.001)
and mo e midwi es (13.4; 9.1 pe 1000 deli e ies; es ,
p < 0.001). The c ude cesa ean sec ion a e did no di e sig-
ni ican ly om each o he (mean solo depa men : 30.04%,
mean in eg a ed depa men : 30.41%; es : p alue = 0.371).
Solo depa men s we e mo e likely o be in ol ed in aca-
demic eaching (80.1%) and o ha e a highe pe ina al ca e
le el (ca e le el 4 = 33.1%; ca e le el 1 = 41.2%) (Appendix
Table9 o mo e de ails).
Table 1 Compa ison o bi hs and cesa ean sec ions be ween o al Ge man hospi al popula ion and s udy popula ion
IQTIG S udy popula ion
Deli e ies (all mo he s who
ha e had a leas one bi h o
a child)
Cesa ean deli e ies C-sec ion a e Deli e ies (all mo he s who
ha e had a leas one bi h o
a child)
Cesa ean deli e ies C-sec ion a e
2015 713,563 224,197 31.42 444,555 139,940 31.48
2016 753,289 235,096 31.21 420,158 130,253 31.00
2017 756,146 235,765 31.18 492,692 151,074 30.66
2018 749,024 229,676 30.66 499,258 152,461 30.54
2019 745,941 230,105 30.85 479,176 147,067 30.69
Σ/Ø 3,717,963 1,154,839 31.06 2,335,839 720,795 30.87
994 A.S öcke e al.
Table 2 Desc ip i e desc ip ion s udy popula ion
a n (%); mean/median (SD)
Cha ac e is ic O e all, N = 208912015, N = 43112016, N = 37612017, N = 43012018, N = 43712019, N = 4151
Risk-adjus ed
cesa ean a io
0.98/1.00 (0.18) 0.97/0.97 (0.19) 0.98/0.99 (0.18) 0.98/1.00 (0.19) 0.99/1.00 (0.18) 0.99/1.01 (0.18)
Numbe o deli -
e ies
1118.16/913.00
(695.53)
1031.45/828.00
(641.20)
1117.44/905.50
(701.29)
1145.80/934.00
(708.85)
1142.47/936.00
(702.03)
1154.64/950.00
(719.08)
Numbe o
C-sec ions
345.04/263.00
(240.27)
324.69/250.00
(229.90)
346.42/267.50
(240.73)
351.33/266.00
(241.12)
348.88/265.00
(242.04)
354.38/265.00
(247.48)
Numbe o ull-
ime equi alen
physicians
12.48/10.86 (7.10) 11.75/10.33 (6.57) 12.35/10.73 (7.30) 12.56/11.09 (7.12) 12.74/11.00 (7.17) 13.01/11.14 (7.32)
Numbe o ull-
ime equi alen
physicians pe
1000 deli e ies
12.46/11.75 (4.45) 12.85/11.84 (4.90) 12.32/11.64 (4.54) 12.22/11.50 (4.36) 12.40/11.84 (4.21) 12.51/11.81 (4.17)
Numbe o ull-
ime equi alen
midwi es
11.03/9.17 (7.56) 10.80/9.00 (6.80) 11.11/9.40 (7.56) 11.15/9.16 (7.59) 10.95/9.04 (7.71) 11.16/9.48 (8.13)
Numbe o ull-
ime equi alen
midwi es pe
1000 deli e ies
10.64/10.30 (4.53) 11.38/10.67 (4.35) 10.60/10.14
(4.09)
10.53/10.22 (4.86) 10.25/10.16 (4.31) 10.45/10.37 (4.88)
Numbe o ull-
ime equi alen
nu sing s a
23.97/19.60
(17.03)
23.33/19.80
(15.86)
24.60/19.70
(18.54)
24.30/19.65
(16.97)
24.00/19.68
(17.12)
23.64/19.30 (16.68)
Missing 49 40 2 2 3 2
Numbe o ull-
ime equi alen
nu sing s a pe
1000 deli e ies
23.34/21.06
(11.04)
24.69/22.25
(12.18)
23.99/21.90
(11.63)
23.19/21.17
(10.75)
22.76/20.54
(10.20)
22.27/20.12 (10.38)
Missing 49 40 1 2 3 2
Owne ship
Non-p o i 759 (36.33%) 166 (38.52%) 123 (32.71%) 165 (38.37%) 159 (36.38%) 146 (35.18%)
P i a e 346 (16.56%) 72 (16.71%) 67 (17.82%) 67 (15.58%) 72 (16.48%) 68 (16.39%)
Public 984 (47.10%) 193 (44.78%) 186 (49.47%) 198 (46.05%) 206 (47.14%) 201 (48.43%)
Teaching s a us
No eaching
assignmen
459 (21.97%) 97 (22.51%) 90 (23.94%) 98 (22.79%) 89 (20.37%) 85 (20.48%)
Academic each-
ing hospi al
1472 (70.46%) 309 (71.69%) 256 (68.09%) 298 (69.30%) 313 (71.62%) 296 (71.33%)
Uni e si y hos-
pi al
158 (7.56%) 25 (5.80%) 30 (7.98%) 34 (7.91%) 35 (8.01%) 34 (8.19%)
Pe ina al ca e
le el
Regula obs e ic
depa men
(ca e le el 4)
939 (44.95%) 201 (46.64%) 164 (43.62%) 193 (44.88%) 195 (44.62%) 186 (44.82%)
Pe ina al ocus
(ca e le el 3)
405 (19.39%) 73 (16.94%) 73 (19.41%) 87 (20.23%) 90 (20.59%) 82 (19.76%)
Pe ina al cen e s
le el II (ca e
le el 2)
185 (8.86%) 44 (10.21%) 36 (9.57%) 36 (8.37%) 36 (8.24%) 33 (7.95%)
Pe ina al cen e s
le el I (ca e
le el 1)
560 (26.81%) 113 (26.22%) 103 (27.39%) 114 (26.51%) 116 (26.54%) 114 (27.47%)
1001
Explo ing hein luence o medical s a ing andbi h olume onobse ed‑ o‑expec ed cesa ean…
he quali y epo s p o ed un easible. Consequen ly, he da a
o he yea 2016 con ain a bias due o he absence o hese
depa men s.
Rema kably, a conside able numbe o missing alues
pe aining o midwi es we e no ed, despi e egula o y man-
da es equi ing hei p esence a all hospi al bi hs in Ge -
many. The impe a i e o a midwi e's a endance ex ends
o all hospi al bi hs, including cesa ean sec ions (§ 4
Abs 1.-Hebammengese z). Howe e , ce ain depa men s
epo ed he p esence o a ending midwi es, leading o he
exclusion om he s udy. A no able pe cen age o hospi als
indica ed ei he a comple e absence o midwi es o epo ed
hem ou side obs e ic depa men s. A ecu ing obse a ion
was ha depa men s sla ed o closu e in he subsequen
yea ailed o u nish any da a on medical s a o he p e-
ceding yea . This phenomenon can be a ibu ed o he ime
lag o o e a yea in he da a collec ion and compila ion
p ocess. The o e all da a quali y in his aspec is subop imal,
po en ially se ing as a sou ce o bias.
Finally, i is c ucial o ecognize ha heal hca e, pa icu-
la ly obs e ic ca e, is signi ican ly in luenced by cul u al,
policy, and local ac o s [8, 33, 89, 90]. The complex and
di e gen na u e o he ela ionship be ween hese ac o s
unde sco es he challenge o d awing gene al and c oss-
na ional conclusions om he s udy esul s [37].
Implica ions o p ac ice
The indings unde sco e ha wi hin in eg a ed gynecological
and obs e ic depa men s, a highe numbe o physicians pe
deli e y signi ican ly co ela ed wi h mo e obse ed isk-
adjus ed C-sec ions han expec ed. The dual p o ision o
gynecological and obs e ic ca e in one depa men poses a
challenge, po en ially con ibu ing o an ele a ed C-sec ion
a io wi h an inc eased numbe o physicians pe bi h as
each physician may be less acquain ed wi h obs e ic ca e.
As a p ac ical app oach, di ec ing a en ion owa ds ocus
and specializa ion wi hin he medical eam could yield a
educ ion in he C-sec ion a io, po en ially in ol ing in a-
depa men s a ealloca ion allowing o a mo e special-
ized wo k o ce be ween he wo a eas o esponsibili y.
This app oach does no necessa ily imply educing he o al
numbe o physicians bu a he enhancing specializa ion
be ween obs e ics and gynecology ca e.
An insigh ul analysis o depa men s su passing he
e e ence alue (90 h pe cen ile) e eals ha hese depa -
men s (n = 118) exhibi an a e age o 15.1 physicians pe
1000 deli e ies, accompanied by 872.5 bi hs and 358.9
C-sec ions (41.1%). In con as , inconspicuous depa men s
(n = 1971) exhibi an a e age o 12.3 physicians pe 1000
deli e ies, along wi h 1132.9 bi hs and 344.2 C-sec ions
(30.4%). A a ge ed educ ion o physicians pe 1000 bi hs
by 3, aligning wi h he inconspicuous depa men a e age,
could an icipa e a modes educ ion o 0.012 in he obse ed
expec ed C-sec ions, based on ou esul s. On he o he hand,
depa men s in he 10 h pe cen ile (n = 217) deployed on
a e age 10.9 physicians pe 1000 deli e ies, accompanied
by 1216.5 bi hs and 259.7 C-sec ions (21.3%). Impo an ly,
as we es ima ed wi hin-e ec s he e ec size is smalle as
a be ween compa ison would sugges as hey inco po a e
depa men indi idual e ec s. As o depa men s ha
al eady ha e low C-sec ion a es (and a po en ial unde -
pe o mance o C-sec ions) i is unclea whe he a u he
inc ease in he numbe o deli e ies pe physician would
lead o imp o ed quali y o ca e.
These implica ions sugges ha a nuanced app oach o
physician s a ing, coupled wi h a s a egic ocus on spe-
cializa ion and a po en ial ealignmen o esou ces, could
con ibu e o achie e he objec i e o he quali y indica o
o minimizing C-sec ions wi hin in eg a ed obs e ic and
gynecological depa men s. I he goal, as a icula ed in he
quali y indica o , is o minimize he occu ence o cesa ean
sec ion bi hs, i is essen ial o conside addi ional ac o s
beyond he numbe o physicians pe bi h p esen ed in his
analysis. The assessmen mus also conside he p e e ences
o he woman gi ing bi h and ele an medical conside a-
ions, ensu ing ha hese c ucial aspec s a e no dis ega ded.
I is i al ha a na ow ocus on educing he cesa ean sec-
ion a e does no comp omise he heal h o women and
in an s.
Conclusion
Ou examina ion o in eg a ed Ge man obs e ic and gyneco-
logical depa men s e ealed a no ewo hy and obus posi-
i e associa ion be ween he numbe o physicians pe bi h
and an ele a ed a io o obse ed- o-expec ed cesa ean
sec ions. Addi ionally, ins umen al a iable analysis indi-
ca ed a po en ial causal e ec . Howe e , gi en he eliance
on annual, agg ega ed a e ages and he inhe en unce ain-
ies despi e he ins umen al analysis employed, a cau ious
app oach is essen ial when in e p e ing causali y. P elimi-
na y in e p e a ions sugges ha specializa ion, as indica ed
by he numbe o deli e ies pe physician, may in luences
he cesa ean sec ion a io wi hin in eg a ed obs e ic and
gynecological depa men s.
Appendix
See Tables6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
20, 21, 22 and Fig.2.

1002 A.S öcke e al.
Table 6 Dynamic IV models wi h sys em gene alized me hod o momen es ima o s
Ra io o obse ed o expec ed (O/E) cesa ean bi hs (2015–
2019) ( wo-way ixed e ec s model)
Uni a ia e model Mul i a ia e model Mul i a ia e model
wi h con ol a i-
ables
Lagged a io o obse ed o expec ed a io (O/E) o cesa ean bi hs 0.485*** 0.503*** 0.489***
p alue (< 0.001) (< 0.001) (< 0.001)
95% CI [0.355, 0.616] [0.374, 0.633] [0.383, 0.596]
SE (0.068) (0.069) (0.054)
Numbe o physicians pe 1000 deli e ies 0.004*** 0.004*** 0.004**
p alue (< 0.001) (< 0.001) (0.002)
95% CI [0.002, 0.006] [0.002, 0.007] [0.001, 0.006]
SE (0.001) (0.001) (0.001)
Numbe o midwi es pe 1000 deli e ies − 0.002 − 0.002
p alue (0.164) (0.105)
95% CI [− 0.004, 0.001] [− 0.004, 0.000]
SE (0.001) (0.001)
Numbe o deli e ies pe 1000 − 0.010+− 0.037***
p alue (0.095) (< 0.001)
95% CI [− 0.022, 0.002] [− 0.054, − 0.020]
SE (0.006) (0.009)
Owne ship: public ( e . ca ego y: owne ship p i a e) − 0.015
p alue (0.225)
95% CI [− 0.039, 0.009]
SE (0.013)
Owne ship: non-p o i ( e . ca ego y: owne ship p i a e) − 0.002
p alue (0.906)
95% CI [− 0.027, 0.024]
SE (0.014)
Pe ina al cen e s le el I (ca e le el 1) ( e . ca ego y: egula obs e ic depa men
(ca e le el 4))
0.051***
p alue (< 0.001)
95% CI [0.024, 0.078]
SE (0.014)
Pe ina al cen e s le el II (ca e le el 2) ( e . ca ego y: egula obs e ic depa -
men (ca e le el 4) 0.016
p alue (0.304)
95% CI [− 0.014, 0.045]
SE (0.016)
Pe ina al ocus (ca e le el 3) ( e . ca ego y: egula obs e ic depa men (ca e
le el 4) − 0.010
p alue (0.355)
95% CI [− 0.033, 0.012]
SE (0.012)
Teaching s a us: academic eaching hospi al ( e . ca ego y: eaching s a us: no
eaching assignmen ) 0.013
p alue (0.224)
95% CI [− 0.008, 0.034]
SE (0.011)
Teaching s a us: Uni e si y Hospi al ( e . ca ego y: eaching s a us: no eaching
assignmen ) 0.014
p alue (0.263)
1003
Explo ing hein luence o medical s a ing andbi h olume onobse ed‑ o‑expec ed cesa ean…
Table 6 (con inued)
Ra io o obse ed o expec ed (O/E) cesa ean bi hs (2015–
2019) ( wo-way ixed e ec s model)
Uni a ia e model Mul i a ia e model Mul i a ia e model
wi h con ol a i-
ables
95% CI [− 0.010, 0.037]
SE (0.013)
Num. obs 2520 2520 2520
Hansen–Sa gan es /J es (p alue) 6.983 (0.639) 11.725 (0.385) 12.310 (0.831)
A ellano–bond es /au oco ela ion es (1) (p alue) − 6.094 (< 0.001) − 6.058 (< 0.001) − 6.321 (< 0.001)
A ellano–bond es /au oco ela ion es (2) (p alue) − 0.101 (0.919) − 0.094 (0.925) − 0.253 (0.800)
Wald es o coe icien s (p alue) 110.125 (< 0.001) 148.589 (< 0.001) 321.429 (< 0.001)
Wald es o ime dummies (p alue) 3.130 (0.372) 3.084 (0.379) 4.702 (0.195)
S d. e o s HC1 HC1 HC1
+ p < 0.1, *p < 0.05, **p < 0.01, ***p < 0.001
1004 A.S öcke e al.
Table 7 Obs e ic quali y indica o s o he pe iod 2015–2019
QI-ID Indica o desc ip ion Yea s Quali y goal
330 An ena al co icos e oid he apy o p e e m bi hs wi h a p epa um inpa ien
s ay o a leas wo calenda days
2015–2019 F equen ini ia ion o an ena al co icos e oid he apy (lung ma u a ion induc-
ion) in bi hs wi h a ges a ional age o 24 + 0 o unde 34 + 0weeks, excluding
s illbi hs, and wi h a p epa um inpa ien s ay o a leas wo calenda days
50,046 Adminis a ion o an ibio ics in cases o p ema u e up u e o memb anes 2015 No lis ed in 2015
50,045 Pe iope a i e an ibio ic p ophylaxis in cesa ean sec ion deli e ies 2015–2019 High a e o pe iope a i e an ibio ic p ophylaxis in cesa ean sec ion deli e ies
52,243 Cesa ean bi hs 2015 No lis ed in 2015
52,249 Ra io o obse ed o expec ed a e (O/E) o cesa ean bi hs 2015–2019 Low a e o cesa ean bi hs
1058 Decision- o-deli e y in e al o e 20min in eme gency cesa ean sec ions 2015–2019 Ra ely a decision- o-deli e y in e al o mo e han 20min in eme gency cesa -
ean sec ions
319 Measu emen o umbilical a e y pH in single on li e bi hs 2015 No lis ed in 2015
321 Acidosis in ma u e single ons wi h umbilical a e y pH measu emen 2015–2019 Low a e o acidosis in single on li e bi hs wi h umbilical a e y pH measu e-
men
51,397 Ra io o obse ed o expec ed a e (O/E) o acidosis in ma u e single ons wi h
umbilical a e y pH measu emen
2015–2019 Low a e o acidosis in single on li e bi hs wi h umbilical a e y pH measu e-
men
51,826 Acidosis in p e e m single ons wi h umbilical a e y pH measu emen 2015 No lis ed in 2015
51,831 Ra io o obse ed o expec ed a e (O/E) o acidosis in p e e m single ons wi h
umbilical a e y pH measu emen
2015–2019 Low a e o acidosis in single on li e bi hs wi h umbilical a e y pH measu e-
men
318 P esence o a pedia ician a p e e m bi hs 2015–2019 F equen p esence o a pedia ician a he bi h o p e e m li e bi hs wi h a
ges a ional age o 24 + 0 o unde 35 + 0weeks
1059 C i ical ou come in ma u e newbo ns 2015 No lis ed in 2015
51,803 Quali y index o c i ical ou come in ma u e newbo ns 2015–2019 Ra e occu ences o child dea hs, 5-min Apga sco e below 5, pH below 7, and
Base Excess < − 16 in ma u e newbo ns
51808_51803 Le el 1: Ra io o obse ed o expec ed a e (O/E) o child dea hs 2018, 2019 No speci ied
51813_51803 Le el 2: Ra io o obse ed o expec ed a e (O/E) o child en wi h a 5-min
Apga sco e below 5
2018, 2019 No speci ied
51818_51803 Le el 3: Ra io o obse ed o expec ed a e (O/E) o child en wi h Base Excess
below − 16
2018, 2019 No speci ied
51823_51803 Le el 4: Ra io o obse ed o expec ed a e (O/E) o child en wi h acidosis
(pH < 7.00)
2018, 2019 No speci ied
322 Thi d- o ou h-deg ee pe ineal ea in spon aneous single on deli e ies 2015 No speci ied
51,181 Ra io o obse ed o expec ed a e (O/E) o hi d- o ou h-deg ee pe ineal
ea s in spon aneous single on deli e ies
2015–2017 Low numbe o mo he s wi h hi d- o ou h-deg ee pe ineal ea s in spon ane-
ous single on deli e ies
323 Thi d- o ou h-deg ee pe ineal ea in spon aneous single on deli e ies wi h-
ou episio omy
2015 No lis ed in 2015
324 Thi d- o ou h-deg ee pe ineal ea in spon aneous single on deli e ies wi h
episio omy
2015 No lis ed in 2015
52,244 Mo he s and child en discha ged home oge he 2015 No lis ed in 2015
52,254 Ra io o obse ed o expec ed a e (O/E) o mo he s and child en discha ged
home oge he
2015 No lis ed in 2015
1005
Explo ing hein luence o medical s a ing andbi h olume onobse ed‑ o‑expec ed cesa ean…
Table 7 (con inued)
QI-ID Indica o desc ip ion Yea s Quali y goal
331 Ma e nal mo ali y in he con ex o pe ina al su eys 2015–2019 Ra e occu ences o ma e nal dea hs
181,800 Quali y index o ou h-deg ee pe ineal ea s in single on deli e ies 2018, 2019 Low numbe o mo he s wi h ou h-deg ee pe ineal ea s in spon aneous o
aginal-assis ed single on deli e ies
181801_181800 Le el 1: Ra io o obse ed o expec ed a e (O/E) o ou h-deg ee pe ineal
ea s in spon aneous single on deli e ies
2018, 2019 No speci ied
181802_181800 Le el 2: Ra io o obse ed o expec ed a e (O/E) o ou h-deg ee pe ineal
ea s in aginal-assis ed single on deli e ies
2018, 2019 No speci ied
1006 A.S öcke e al.
Table 8 Compa ison o bi hs and cesa ean sec ions be ween in eg a ed and solo obs e ic depa men s
S udy popula ion (in eg a ed depa men s) Solo depa men s
Deli e ies (all mo he s who ha e had
a leas one bi h o a child)
Cesa ean deli e ies C-sec ion a e Deli e ies (all mo he s who ha e had
a leas one bi h o a child)
Cesa ean deli e ies C-sec ion a e
2015 444,555 139,940 31.48 90,690 28,866 31.83
2016 420,158 130,253 31.00 93,824 27,489 29.30
2017 492,692 151,074 30.66 91,709 28,149 30.69
2018 499,258 152,461 30.54 98,282 28,996 29.50
2019 479,176 147,067 30.69 107,058 32,398 30.26
Σ/Ø 2,335,839 720,795 30.87 481,563 145,898 30.32

1007
Explo ing hein luence o medical s a ing andbi h olume onobse ed‑ o‑expec ed cesa ean…
Table 9 Desc ip i e desc ip ion solo obs e ic depa men s
a n (%); mean/median (SD)
Cha ac e is ic O e all
N = 308a2015
N = 61a2016
N = 57a2017
N = 61a2018
N = 62a2019
N = 67a
Risk-adjus ed
cesa ean a io
0.95/0.95 (0.18) 0.97/0.99 (0.17) 0.92/0.92 (0.19) 0.95/0.96 (0.18) 0.94/0.94 (0.18) 0.95/0.96 (0.19)
Numbe o deli e -
ies
1563.52/1474.50
(709.62)
1486.72/1293.00
(716.81)
1646.04/1603.00
(785.32)
1503.43/1464.00
(634.92)
1585.19/1523.00
(681.82)
1597.88/1520.00
(734.48)
Numbe o
C-sec ions
473.69/437.00
(249.22)
473.21/436.00
(253.57)
482.26/437.00
(282.01)
461.46/433.00
(225.98)
467.68/426.50
(229.55)
483.55/448.00
(259.85)
Numbe o ull-
ime equi alen
physicians
9.66/8.34 (4.96) 9.28/8.25 (5.03) 9.73/8.60 (5.59) 9.19/8.04 (4.22) 9.74/8.56 (4.56) h ps://doi.
o g/10.32/8.34
(5.36)
Numbe o ull-
ime equi alen
physicians pe
1000 deli e ies
6.82/5.82 (3.31) 7.03/6.08 (3.68) 6.72/5.79 (3.52) 6.79/5.48 (3.33) 6.68/5.70 (3.03) 6.89/6.16 (3.09)
Numbe o ull-
ime equi alen
midwi es
13.39/12.79 (7.38) 13.22/12.30 (8.20) 13.72/12.80 (8.02) 12.61/12.64 (6.77) 13.24/12.89 (6.85) 14.12/13.60 (7.17)
Numbe o ull-
ime equi alen
midwi es pe
1000 deli e ies
9.07/9.20 (3.80) 9.57/9.19 (4.84) 8.95/8.86 (4.14) 8.70/9.29 (3.24) 8.79/9.08 (3.23) 9.32/9.52 (3.41)
Numbe o ull-
ime equi alen
nu sing s a
17.65/15.88 (9.93) 18.05/16.00 (9.89) 18.81/15.80
(12.93)
16.39/15.14 (7.99) 17.97/16.66
(10.31)
17.13/16.19 (8.26)
missing 1 0 0 0 0 1
Numbe o ull-
ime equi alen
nu sing s a pe
1000 deli e ies
12.17/10.93 (5.92) 13.51/12.29 (7.67) 12.23/10.54 (6.33) 11.41/10.94 (4.42) 12.25/10.58 (6.17) 11.53/11.09 (4.45)
Missing 1 0 0 0 0 1
Owne ship
Non-p o i 176 (57.14%) 34 (55.74%) 29 (50.88%) 34 (55.74%) 38 (61.29%) 41 (61.19%)
P i a e 31 (10.06%) 6 (9.84%) 8 (14.04%) 5 (8.20%) 5 (8.06%) 7 (10.45%)
Public 101 (32.79%) 21 (34.43%) 20 (35.09%) 22 (36.07%) 19 (30.65%) 19 (28.36%)
Teaching s a us
No eaching
assignmen
39 (12.66%) 8 (13.11%) 6 (10.53%) 10 (16.39%) 9 (14.52%) 6 (8.96%)
Academic eaching
hospi al
247 (80.19%) 51 (83.61%) 47 (82.46%) 47 (77.05%) 47 (75.81%) 55 (82.09%)
Uni e si y hospi al 22 (7.14%) 2 (3.28%) 4 (7.02%) 4 (6.56%) 6 (9.68%) 6 (8.96%)
Pe ina al ca e
le el
Regula obs e ic
depa men (ca e
le el 4)
102 (33.12%) 18 (29.51%) 23 (40.35%) 19 (31.15%) 20 (32.26%) 22 (32.84%)
Pe ina al ocus
(ca e le el 3)
31 (10.06%) 6 (9.84%) 4 (7.02%) 7 (11.48%) 7 (11.29%) 7 (10.45%)
Pe ina al cen e s
le el II (ca e
le el 2)
48 (15.58%) 11 (18.03%) 8 (14.04%) 10 (16.39%) 9 (14.52%) 10 (14.93%)
Pe ina al cen e s
le el I (ca e le el
1)
127 (41.23%) 26 (42.62%) 22 (38.60%) 25 (40.98%) 26 (41.94%) 28 (41.79%)
1008 A.S öcke e al.
Table 10 Uni- and mul i a ia e panel models wi h wo-ways andom e ec s on isk-adjus ed C-sec ion a io
+ p < 0.1, *p < 0.05, **p < 0.01, ***p < 0.001
Ra io o obse ed o expec ed (O/E) cesa ean bi hs (2015–2019) ( wo-way andom e ec s
model)
Uni a ia e models Mul i a ia e model
In e cep 0.893*** 0.964*** 1.022*** 0.918***
p alue (< 0.001) (< 0.001) (< 0.001) (< 0.001)
95% CI [0.852, 0.934] [0.924, 1.004] [0.994, 1.051] [0.852, 0.984]
SE (0.021) (0.020) (0.014) (0.034)
Numbe o physicians pe 1000 deli e ies 0.007*** 0.007***
p alue (< 0.001) (< 0.001)
95% CI [0.004, 0.010] [0.003, 0.010]
SE (0.002) (0.002)
Numbe o midwi es pe 1000 deli e ies 0.002 0.000
p alue (0.327) (0.912)
95% CI [− 0.002, 0.005] [− 0.004, 0.004]
SE (0.002) (0.002)
Numbe o deli e ies pe 1000 − 0.036*** − 0.017
p alue (< 0.001) (0.135)
95% CI [− 0.056, − 0.015] [− 0.039, 0.005]
SE (0.010) (0.011)
Num. obs 2089 2089 2089 2089
R20.061 0.000 0.019 0.059
R2 adj 0.061 0.000 0.019 0.057
AIC − 1282.3 − 1158.3 − 1201.5 − 1279.1
BIC − 1265.3 − 1141.3 − 1184.5 − 1250.9
S d. e o s HC1 HC1 HC1 HC1
Table 11 Fi s s age leas
squa e eg ession o numbe o
nu sing s a pe 1000 deli e ies
as an ins umen a iable o
numbe o physicians pe 1000
deli e ies
+ p < 0.1, * p < 0.05, ** p < 0.01, *** p < 0.001
Fi s s age eg ession ( ixed e ec s)
Numbe o nu sing s a pe 1000. deli e ies 0.239***
p alue (< 0.001)
95% CI [0.208, 0.269]
SE (0.016)
Num. obs 2037
R20.329
R2 adj 0.106
AIC 9101.9
BIC 9113.1
F-s a is ic 237.206
S d. e o s HC1
1009
Explo ing hein luence o medical s a ing andbi h olume onobse ed‑ o‑expec ed cesa ean…
Table 12 Endogenei y check
o ins umen a iable on isk-
adjus ed C-sec ion a io
+ p < 0.1, *p < 0.05, **p < 0.01, ***p < 0.001
Ra io o obse ed o expec ed (O/E) cesa ean bi hs
(2015–2019) ( wo-way ixed e ec s model)
uni a ia e model mul i a ia e model
Numbe o physicians pe 1000 deli e ies 0.014*** 0.014***
p alue (< 0.001) (< 0.001)
95% CI [0.009, 0.019] [0.009, 0.019]
SE (0.003) (0.003)
Numbe o midwi es pe 1000 deli e ies − 0.001
p alue (0.627)
95% CI [− 0.004, 0.002]
SE (0.002)
Numbe o deli e ies pe 1000 0.001
p alue (0.933)
95% CI [− 0.024, 0.026]
SE (0.013)
Endogenous pa om he IV − 0.006* − 0.006*
p alue (0.023) (0.021)
95% CI [− 0.012, − 0.001] [− 0.012, − 0.001]
SE (0.003) (0.003)
Num. obs 2037 2037
R20.066 0.067
R2 adj − 0.246 − 0.247
AIC − 3436.6 − 3433.6
BIC − 3419.7 − 3405.5
S d. e o s HC1 HC1
1010 A.S öcke e al.
Table 13 Compa ison o e ec
size lagged a iable o OLS,
ixed e ec and di e ence
GMM es ima o on isk-
adjus ed C-sec ion a io
+ p < 0.1, *p < 0.05, **p < 0.01, ***p < 0.001
Ra io o obse ed o expec ed (O/E) cesa ean bi hs
(2015–2019) ( wo-way ixed e ec s model)
OLS Panel model wi h
ixed e ec s
di e ence GMM
In e cep 0.144***
p alue (< 0.001)
95% CI [0.111, 0.177]
SE (0.017)
Lagged a io o obse ed o expec ed a io
(O/E) o cesa ean bi hs
0.864*** 0.046 0.510***
p alue (< 0.001) (0.241) (< 0.001)
95% CI [0.837, 0.891] [− 0.031, 0.123] [0.265, 0.754]
SE (0.014) (0.039) (0.125)
Numbe o physicians pe 1000 deli e ies 0.000 0.004* 0.003
p alue (0.769) (0.045) (0.239)
95% CI [− 0.001, 0.001] [0.000, 0.007] [− 0.002, 0.007]
SE (0.001) (0.002) (0.002)
Numbe o midwi es pe 1000 deli e ies 0.000 0.002 0.003
p alue (0.884) (0.336) (0.150)
95% CI [− 0.001, 0.001] [− 0.002, 0.006] [− 0.001, 0.006]
SE (0.001) (0.002) (0.002)
Numbe o deli e ies pe 1000 − 0.006* − 0.013 − 0.008
p alue (0.037) (0.670) (0.844)
95% CI [− 0.012, 0.000] [− 0.074, 0.048] [− 0.089, 0.072]
SE (0.003) (0.031) (0.041)
Num. obs 1479 1479 1024
R20.753 0.012
R2 adj 0.752 − 0.427
AIC − 2916.6 − 4015.8
BIC − 2884.8 − 3989.3
S d. e o s HC1 HC1
1017
Explo ing hein luence o medical s a ing andbi h olume onobse ed‑ o‑expec ed cesa ean…
Table 21 Uni- and mul i a ia e panel models wi h wo-ways di e ence gene alized me hod o momen es ima o s on c ude C-sec ion/bi h a io
Ra io o C-sec ion o all bi hs (2015–2019) ( wo-way ixed
e ec s model)
Uni a ia e model Mul i a ia e model Mul i a ia e model
wi h con ol a i-
ables
Lagged C-sec ion o deli e y a io 0.522*** 0.503*** 0.511***
p alue (< 0.001) (< 0.001) (< 0.001)
95% CI [0.310, 0.734] [0.289, 0.716] [0.295, 0.727]
SE (0.108) (0.109) (0.110)
Numbe o physicians pe 1000 deli e ies 0.001 0.001 0.001
p alue (0.117) (0.413) (0.433)
95% CI [0.000, 0.003] [− 0.001, 0.002] [− 0.001, 0.002]
SE (0.001) (0.001) (0.001)
Numbe o midwi es pe 1000 deli e ies 0.001 0.001
p alue (0.336) (0.310)
95% CI [− 0.001, 0.002] [− 0.001, 0.002]
SE (0.001) (0.001)
Numbe o deli e ies pe 1000 − 0.014 − 0.013
p alue (0.447) (0.471)
95% CI [− 0.050, 0.022] [− 0.050, 0.023]
SE (0.018) (0.019)
Owne ship: public ( e . ca ego y: owne ship p i a e) 0.000
p alue (0.995)
95% CI [− 0.051, 0.051]
SE (0.026)
Owne ship: non-p o i ( e . ca ego y: owne ship p i a e) 0.003
p alue (0.899)
95% CI [− 0.038, 0.043]
SE (0.021)
Pe ina al cen e s le el I (ca e le el 1) ( e . ca ego y: egula obs e ic depa men
(ca e le el 4)) 0.009
p alue (0.367)
95% CI [− 0.011, 0.029]
SE (0.010)
Pe ina al cen e s le el II (ca e le el 2) ( e . ca ego y: egula obs e ic depa -
men (ca e le el 4) 0.008
p alue (0.594)
95% CI [− 0.020, 0.035]
SE (0.014)
Pe ina al ocus (ca e le el 3) ( e . ca ego y: egula obs e ic depa men (ca e
le el 4) − 0.011
p alue (0.334)
95% CI [− 0.032, 0.011]
SE (0.011)
Teaching s a us: academic eaching hospi al ( e . ca ego y: eaching s a us: no
eaching assignmen ) − 0.006
p alue (0.509)
95% CI [− 0.022, 0.011]
SE (0.008)
Teaching s a us: Uni e si y Hospi al ( e . ca ego y: eaching s a us: no eaching
assignmen )
− 0.004
p alue (0.675)

1018 A.S öcke e al.
+ p < 0.1, *p < 0.05, **p < 0.01, ***p < 0.001
Table 21 (con inued)
Ra io o C-sec ion o all bi hs (2015–2019) ( wo-way ixed
e ec s model)
Uni a ia e model Mul i a ia e model Mul i a ia e model
wi h con ol a i-
ables
95% CI [− 0.021, 0.013]
SE (0.009)
Num. obs 1031 1031 1031
Hansen–Sa gan es /J es (p alue) 1.937 (0.858) 1.704 (0.888) 1.685 (0.891)
A ellano–Bond es /au oco ela ion es (1) (p alue) − 6.098 (< 0.001) − 6.225 (< 0.001) − 6.240 (< 0.001)
A ellano–Bond es /au oco ela ion es (2) (p alue) − 0.365 (0.715) − 0.494 (0.622) − 0.405 (0.686)
Wald es o coe icien s (p alue) 28.950 (< 0.001) 30.083 (< 0.001) 56.313 (< 0.001)
Wald es o ime dummies (p alue) 9.975 (0.019) 9.845 (0.020) 9.890 (0.020)
S d. e o s HC1 HC1 HC1
Table 22 Example o a depic ion o calcula ing he expec ed isk-adjus ed cesa ean sec ion a e by IQTIG o 2018. Own ansla ion. (Sou ce:
h ps:// iq ig. o g/ downl oads/ auswe ung/ auswe ung/ 2018/ 16n1g ebh/ QSKH_ 16n1- GEBH_ 2018_ QIDB_ V02_ 2019- 04- 11. pd , p. 19)
Re e ence p obabili y: 14.205% (odds: 0.165)
Risk ac o Reg ession coe icien S anda d e o Z alue Odds a io 95% con idence in e al
Cons an − 1.798342673800550 0.004 − 408.407 – –
Age 35–38yea s 0.034231814566637 0.008 4.439 1.035 1.019–1.051
Age > 38 0.286484745408656 0.011 24.971 1.332 1.302–1.362
Bi h isk: amnion in ec ion synd ome (suspec ed) 2.647566531445380 0.040 66.591 14.120 13.061–15.264
Bi h isk: diabe es melli us 0.354402849335064 0.014 24.645 1.425 1.386–1.466
Bi h isk: p ema u e bi h 0.356693940042668 0.018 20.341 1.429 1.380–1.479
Bi h isk: hype ensi e P egnancy diso de o
HELLP Synd ome
1.471801217071380 0.018 81.161 4.357 4.205–4.515
Bi h isk: pa hological CTG. poo e al hea
sounds. o acidosis du ing bi h (de ec ed by FBS)
0.922892533667843 0.007 124.693 2.517 2.480–2.553
Bi h isk: placen a p ae ia 3.414601549901360 0.061 55.693 30.405 26.962–34.287
Bi h isk: b eech posi ion 3.585447507902870 0.018 199.288 36.069 34.820–37.364
Bi h isk: ace/ o ehead P esen a ion 1.942056761249630 0.063 30.713 6.973 6.160–7.893
Bi h isk: ans e se/oblique posi ion 6.515521815847840 0.268 24.293 675.546 399.356–1142.746
Bi h isk: p e ious cesa ean sec ion o o he u e us
ope a ions
1.994488346490800 0.016 122.135 7.348 7.117–7.587
Mul iple p egnancy 1.453148908289270 0.024 59.963 4.277 4.078–4.485
Mo he 's eco d: hype ension o p o einu ia 0.241446328929522 0.025 9.548 1.273 1.212–1.338
Mo he 's eco d: placen al insu iciency 0.735298957679869 0.031 23.853 2.086 1.964–2.216
Mo he 's eco d: p e ious cesa ean sec ion o u e us
ope a ions
0.294514404293640 0.016 17.888 1.342 1.300–1.387
1019
Explo ing hein luence o medical s a ing andbi h olume onobse ed‑ o‑expec ed cesa ean…
Au ho con ibu ions Concep ualiza ion: A no S öcke , Ludwig Kun z,
Nadine Schol en; Me hodology: A no S öcke ; Fo mal analysis, In es-
iga ion, Da a Cu a ion: A no S öcke ; W i ing—o iginal d a p epa-
a ion: A no S öcke ; W i ing— e iew and edi ing: Ludwig Kun z,
Nadine Schol en, Holge P a ; Resou ces: Holge P a , Nadine
Schol en; Supe ision: Ludwig Kun z.
Funding Open Access unding enabled and o ganized by P ojek
DEAL.
Da a a ailabili y The analysis p o ided elied upon he hospi al quali y
epo s om Ge man acu e ca e hospi als. All he da a u ilized in his
analysis is accessible o he public and can be ound a www.g- ba. de/
quali ae s be ic h e.
Decla a ions
Con lic o in e es The au ho s ha e no ele an inancial o non- i-
nancial in e es s o disclose.
E hics app o al No applicable.
Open Access This a icle is licensed unde a C ea i e Commons A i-
bu ion 4.0 In e na ional License, which pe mi s use, sha ing, adap a-
ion, dis ibu ion and ep oduc ion in any medium o o ma , as long
as 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 licence, and indica e i changes
we e made. The images o o he hi d pa y ma e ial in his a icle a e
included in he a icle’s C ea i e Commons licence, unless indica ed
o he wise in a c edi line o he ma e ial. I ma e ial is no included in
he a icle’s C ea i e Commons licence and you in ended use is no
pe mi ed by s a u o y egula ion o exceeds he pe mi ed use, you will
need o ob ain pe mission di ec ly om he copy igh holde . To iew a
copy o his licence, isi h p://c ea i ecommons.o g/licenses/by/4.0/.
Re e ences
1. Boe ma, T., Ronsmans, C., Melesse, D.Y., e al.: Global epide-
miology o use o and dispa i ies in caesa ean sec ions. Lance
392, 1341–1348 (2018). h ps:// doi. o g/ 10. 1016/ S0140- 6736(18)
31928-7
2. Fai , T., Šťas ná, A., Kocou ko á, J., e al.: Has he cesa ean epi-
demic in Czechia been e e sed despi e e ili y pos ponemen ?
BMC P egnancy Childbi h 22, 469 (2022). h ps:// doi. o g/ 10.
1186/ s12884- 022- 04781-1
3. Owens, K.: When less is mo e: shi ing isk managemen in Ame -
ican childbi h. Ad . Med. Sociol. 20, 45–62 (2019). h ps:// doi.
o g/ 10. 1108/ S1057- 62902 01900 00020 008
4. Sandall, J., T ibe, R.M., A e y, L., e al.: Sho - e m and long-
e m e ec s o caesa ean sec ion on he heal h o women and
child en. Lance 392, 1349–1357 (2018). h ps:// doi. o g/ 10. 1016/
S0140- 6736(18) 31930-5
5. S a is isches Bundesam (2022) K anken hausen bindungen in
Deu schland. h ps:// www. des a is. de/ DE/ Themen/ Gesel lscha -
Umwel / Gesun dhei / K ank enhae use / Tabel len/ k ank enhau sen b
indun gen- kaise schn i . h ml. Accessed 20 Oc 2022
6. Schneide , H.: Risiko-Nu zen-Ve häl nis bei na ü liche Gebu
und elek i e Sec io. Gynäkologe 46, 709–714 (2013). h ps:// doi.
o g/ 10. 1007/ s00129- 013- 3179-x
7. Deu sche Gesellscha ü Gynäkologie und Gebu shil e (DGGG)
(Ed.). (2020). S3-Lei linie. Die Sec io caesa ea [S3 Guideline.
The caesa ean sec ion]. h ps:// www. awm . o g/ uploa ds/ x_ szlei
lini en/ 015- 084l_ S3_ Sec io- caesa ea_ 2020- 06_1_ 02. pd .
8. Be án, A.P., Temme man, M., Kingdon, C., e al.: In e en ions
o educe unnecessa y caesa ean sec ions in heal hy women and
babies. Lance 392, 1358–1368 (2018). h ps:// doi. o g/ 10. 1016/
S0140- 6736(18) 31927-5
9. Ki chengas , S., Ha mann, B.: Recen li es yle pa ame e s a e
associa ed wi h inc easing caesa ean sec ion a es among single-
on e m bi hs in Aus ia. In . J. En i on. Res. Public Heal h 16,
14 (2018). h ps:// doi. o g/ 10. 3390/ ije p h1601 0014
10. Loke, A.Y., Da ies, L., Li, S.: Fac o s in luencing he decision ha
women make on hei mode o deli e y: he Heal h Belie Model.
Fig. 2 Ra io deli e ies o ull inpa ien cases
1020 A.S öcke e al.
BMC Heal h Se . Res. 15, 274 (2015). h ps:// doi. o g/ 10. 1186/
s12913- 015- 0931-z
11. Wol , J.H.: Risk and epu a ion: obs e icians, cesa eans, and con-
sen . J. His . Med. Allied Sci. 73, 7–28 (2018). h ps:// doi. o g/ 10.
1093/ jhmas/ j x053
12. Mulchandani, R., Powe , H.S., Ca alla o, F.L.: The in luence o
indi idual p o ide cha ac e is ics and a i udes on caesa ean sec-
ion decision-making: a global e iew. J. Obs e . Gynaecol. 40,
1–9 (2020). h ps:// doi. o g/ 10. 1080/ 01443 615. 2019. 15876 03
13. Panda, S., Begley, C., Daly, D.: Clinicians’ iews o ac o s in lu-
encing decision-making o caesa ean sec ion: a sys ema ic e iew
and me asyn hesis o quali a i e, quan i a i e and mixed me hods
s udies. PLoSOne 13, e0200941 (2018). h ps:// doi. o g/ 10. 1371/
jou n al. pone. 02009 41
14. Facchini, G.: Fo ge ing-by-no -doing: he case o su geons and
cesa ean sec ions. Heal h Econ. 31, 481–495 (2022). h ps:// doi.
o g/ 10. 1002/ hec. 4460
15. Mona i, F., Di Ma io, S., Facchine i, F., e al.: Obs e icians’ and
midwi es’ a i udes owa d cesa ean sec ion. Bi h 35, 129–135
(2008). h ps:// doi. o g/ 10. 1111/j. 1523- 536X. 2008. 00226.x
16. Jose sson, A., Gunne ik, C., Sydsjö, A., e al.: A compa ison
be ween Swedish midwi es’ and obs e icians’ & gynecologis s’
opinions on cesa ean sec ion. Ma e n. Child Heal h J. 15, 555–560
(2011). h ps:// doi. o g/ 10. 1007/ s10995- 010- 0630-7
17. Science Media Cen e Ge many (2020) E s e S3-Lei linie Kaise -
schni : Hin e g und und egionale Da enanalyse. h ps:// www.
scien cemed iacen e . de/ alle- angeb o e/ in es iga i e/ de ai ls/ news/
e s e- s3- lei l inie- kaise schn i - hin e g und- und- egio nale- da en
analy se/. Accessed 21 Oc 2022
18. Zip el, L., Weidmann, C.: Ein luss ak o en au die Kaise -
schni a en in deu schen K ankenhäuse n in den Jah en 2015–
2017. Eine ökologische S udie (Fac o s In luencing he Caesa ean
Sec ion Ra es in Ge man Hospi als in he Yea s 2015–2017: an
Ecological S udy). Gesundhei swesen 84, 944–951 (2022). h ps://
doi. o g/ 10. 1055/a- 1531- 4998
19. G oss, M.M., Ma e ne, A., Be lage, S., e al.: In e ins i u ional
a ia ions in mode o bi h a e a p e ious caesa ean sec ion: a
c oss-sec ional s udy in six Ge man hospi als. J. Pe ina . Med. 43,
177–184 (2015). h ps:// doi. o g/ 10. 1515/ jpm- 2014- 0108
20. Hemminki, E., Heino, A., Gissle , M.: Should bi hs be cen alised
in highe le el hospi als? Expe iences om egionalised heal h
ca e in Finland. BJOG In . J. Obs e . Gynaecol. 118, 1186–1195
(2011). h ps:// doi. o g/ 10. 1111/j. 1471- 0528. 2011. 02977.x
21. Bi o, M.A., Knigh , M., Wallace, E., e al.: Is place o bi h asso-
cia ed wi h mode o bi h? The e ec o hospi al on caesa ean
sec ion a es in a public me opoli an heal h se ice. Aus . N. Z.
J. Obs e . Gynaecol. 54, 64–70 (2014). h ps:// doi. o g/ 10. 1111/
ajo. 12147
22. Joaquim, I., Pe ei a, L.N., Nunes, C., e al.: C-sec ions and hos-
pi al cha ac e is ics: a long e m analysis on low- isk deli e ies.
Res. Heal h Se . Reg. 1, 1–9 (2022). h ps:// doi. o g/ 10. 1007/
s43999- 022- 00014-2
23. Pe o a, C., Rome o, M., Sguasse o, Y., e al.: Caesa ean bi h
in public ma e ni ies in A gen ina: a o ma i e esea ch s udy on
he iews o obs e icians, midwi es and ainees. BMJ Open 12,
e053419 (2022). h ps:// doi. o g/ 10. 1136/ bmjop en- 2021- 053419
24. Mikolajczyk, R.T., Schmed , N., Zhang, J., e al.: Regional a i-
a ion in caesa ean deli e ies in Ge many and i s causes. BMC
P egnancy Childbi h 13, 99 (2013). h ps:// doi. o g/ 10. 1186/
1471- 2393- 13- 99
25. Kolip, P., Nol ing, H.-D., Zich, K.: Kaise schni gebu en—
En wicklung und egionale Ve eilung. Be elsmann S i ung,
Gü e sloh (2012)
26. Hanley, G.E., Janssen, P.A., G eyson, D.: Regional a ia ion in
he cesa ean deli e y and assis ed aginal deli e y a es. Obs e .
Gynecol. 115, 1201–1208 (2010). h ps:// doi. o g/ 10. 1097/ AOG.
0b013 e3181 dd918c
27. Guglielmino i, J., Deneux-Tha aux, C., Wong, C.A., e al.: Hospi-
al-le el ac o s associa ed wi h anes hesia- ela ed ad e se e en s
in cesa ean deli e ies, New Yo k S a e, 2009–2011. Anes h.
Analg. 122, 1947–1956 (2016). h ps:// doi. o g/ 10. 1213/ ANE.
00000 00000 001341
28. Wal he , F., Kues e , D., Biebe , A., e al.: A e bi h ou comes in
low isk bi h coho s ela ed o hospi al bi h olumes? A sys em-
a ic e iew. BMC P egnancy Childbi h 21, 531 (2021). h ps://
doi. o g/ 10. 1186/ s12884- 021- 03988-y
29. Kyse , K.L., Lu, X., San illan, D.A., e al.: The associa ion
be ween hospi al obs e ical olume and ma e nal pos pa um
complica ions. Am. J. Obs e . Gynecol. 207, 42.e1–17 (2012).
h ps:// doi. o g/ 10. 1016/j. ajog. 2012. 05. 010
30. Eps ein, A.J., Nicholson, S.: The o ma ion and e olu ion o physi-
cian ea men s yles: an applica ion o cesa ean sec ions. J. Heal h
Econ. 28, 1126–1140 (2009). h ps:// doi. o g/ 10. 1016/j. jheal eco.
2009. 08. 003
31. Plough, A., Hen ich, N., Gal in, G., e al.: Common challenges
managing bed and s a a ailabili y on labo and deli e y uni s in
he Uni ed S a es: a quali a i e analysis. Bi h 45, 303–310 (2018).
h ps:// doi. o g/ 10. 1111/ bi . 12342
32. Leona d, P.S.J., C ouse, D.L., Boud eau, J.G., e al.: P o ide ol-
ume and ma e nal complica ions a e caesa ean sec ion: esul s
om a popula ion-based s udy. BMC P egnancy Childbi h 20,
37 (2020). h ps:// doi. o g/ 10. 1186/ s12884- 019- 2709-5
33. Lee, K.-S., Kwak, J.-M.: E ec o pa ien isk on he olume-
ou come ela ionship in obs e ic deli e y se ices. Heal h Policy
(Ams e dam, Ne he lands) 118, 407–412 (2014). h ps:// doi. o g/
10. 1016/j. heal hpol. 2014. 05. 007
34. Bozzu o, L., Passa ella, M., Lo ch, S., e al.: E ec s o deli e y
olume and high- isk condi ion olume on ma e nal mo bidi y
among high- isk obs e ic pa ien s. Obs e . Gynecol. 133, 261–268
(2019). h ps:// doi. o g/ 10. 1097/ AOG. 00000 00000 003080
35. Clapp, M.A., Melamed, A., Robinson, J.N., e al.: Obs e ician
olume as a po en ially modi iable isk ac o o cesa ean deli -
e y. Obs e . Gynecol. 124, 697–703 (2014). h ps:// doi. o g/ 10.
1097/ AOG. 00000 00000 000473
36. Zbi i, S., Rozenbe g, P., Go ine , F., e al.: Cesa ean deli e y a e
and s a ing le els o he ma e ni y uni . PLoSOne 13, e0207379
(2018). h ps:// doi. o g/ 10. 1371/ jou n al. pone. 02073 79
37. Gombolay, M., Golen, T., Shah, N., e al.: Queueing heo e ic
analysis o labo and deli e y: unde s anding managemen s yles
and C-sec ion a es. Heal h Ca e Manag. Sci. 22, 16–33 (2019).
h ps:// doi. o g/ 10. 1007/ s10729- 017- 9418-2
38. Ramani, S., Halpe n, T.A., Ake man, M., e al.: A new index o
obs e ics sa e y and quali y o ca e: in eg a ing cesa ean deli -
e y a es wi h ma e nal and neona al ou comes. Am. J. Obs e .
Gynecol. 226, 556.e1-556.e9 (2022). h ps:// doi. o g/ 10. 1016/j.
ajog. 2021. 10. 005
39. Cu y, L.A., B aul , M.A., Linnande , E.L., e al.: In luencing
o ganisa ional cul u e o imp o e hospi al pe o mance in ca e
o pa ien s wi h acu e myoca dial in a c ion: a mixed-me hods
in e en ion s udy. BMJ Qual. Sa . 27, 207–217 (2018). h ps://
doi. o g/ 10. 1136/ bmjqs- 2017- 006989
40. de C uppé, W., Ge aed s, M.: Wie kons an hal en K ankenhäuse
die Mindes mengen o gaben ein? Eine e ospek i e, längsschni -
liche Da enanalyse de Jah e 2006, 2008 und 2010 (How S eady
a e Hospi als in Complying wi h Minimum Volume S anda ds?
A Re ospec i e Longi udinal Da a Analysis o he Yea s 2006,
2008, and 2010). Zen albl. Chi . 141, 425–432 (2016). h ps://
doi. o g/ 10. 1055/s- 0034- 13833 71
41. de C uppé, W., Ge aed s, M.: Mindes mengen un e sch e-
i en, Ausnahme a bes ände und ih e Konsequenzen ab 2018.
Komplexe Eing i e am Ösophagus und Pank eas in deu schen
1021
Explo ing hein luence o medical s a ing andbi h olume onobse ed‑ o‑expec ed cesa ean…
K ankenhäuse n im Zei e lau on 2006 bis 2014 (Falling Sho
o Minimum Volume S anda ds, Exemp ions and Thei Conse-
quences om 2018 Onwa ds. Complex P ocedu es on Oesophagus
and Panc eas in Ge man Hospi als om 2006 o 2014). Zen albl.
Chi . 143, 250–258 (2018). h ps:// doi. o g/ 10. 1055/a- 0573- 2625
42. Schol en, N., P a , H., Lehmann, H.C., e al.: Who does i i s ?
The up ake o medical inno a ions in he pe o mance o h om-
bolysis on ischemic s oke pa ien s in Ge many: a s udy based on
hospi al quali y da a. Implemen . Sci. 10, 10 (2015). h ps:// doi.
o g/ 10. 1186/ s13012- 014- 0196-7
43. Vo beck, L., Naumoska, D., Ge aed s, M.: Assozia ion on S uk-
u a iablen mi de Ve so gungsquali ä de K ankenhäuse in
Deu schland (Associa ion o S uc u al Va iables wi h Quali y
o Ca e in Ge man Hospi als). Gesundhei swesen 84, 242–249
(2022). h ps:// doi. o g/ 10. 1055/a- 1341- 1246
44. Messe , M., Reilley, J.T.: Quali ä sbe ich e als Ve mi lungsin-
s anz im We bewe b zwischen K ankenhäuse n: Pa ien en als
a ionale Ak eu e. Be l. J. Soziol. 25, 61–81 (2015). h ps:// doi.
o g/ 10. 1007/ s11609- 015- 0279-6
45. Volke , A., P a , H., Schol en, N.: Wha eally ma e s? O gan-
iza ional e sus egional de e minan s o hospi als p o iding
medical se ice cen es. Heal h Policy (Ams e dam, Ne he -
lands) 124, 1354–1362 (2020). h ps:// doi. o g/ 10. 1016/j. heal
hpol. 2020. 07. 011
46. Volke , A., S öcke , A., P a , H., e al.: Wha o ganisa ional and
egional ac o s in luence he ou pa ien p o ision o cu e ages in
Ge many? A longi udinal seconda y da a analysis using hospi al
quali y epo s da a om 2013 o 2019. BMJ Open 13, e072887
(2023). h ps:// doi. o g/ 10. 1136/ bmjop en- 2023- 072887
47. K aska, R.A., de C uppe, W., Ge aed s, M.: P obleme bei de
Ve wendung on Quali ä sbe ich sda en ü die Ve so gungs-
o schung (P oblems wi h Using Hospi al Quali y Repo s as a
Seconda y Da a Sou ce o Heal h Se ices Resea ch in Ge -
many). Gesundhei swesen 79, 542–547 (2017). h ps:// doi. o g/
10. 1055/s- 0035- 15559 53
48. Ji, L., Ge aed s, M. & de C uppé, W.: A heo e ical amewo k
o linking hospi alslongi udinally: demons a ed using Ge man
Hospi al Quali y Repo s 2016–2020. BMC Med Res Me hodol
24, 212 (2024).h ps:// doi. o g/ 10. 1186/ s12874- 024- 02317-zh
49. Woebe , K., Vande laan, J., Long, M.H., e al.: Midwi e y au on-
omy and employmen changes du ing he ea ly COVID-19 pan-
demic. J. Midwi e y Womens Heal h 67, 608–617 (2022). h ps://
doi. o g/ 10. 1111/ jmwh. 13400
50. Schmiedho e , M., De ksen, C., Die l, J.E., e al.: Bi hing unde
he condi ion o he COVID-19 pandemic in Ge many: in e iews
wi h mo he s, pa ne s, and obs e ic heal h ca e wo ke s. In . J.
En i on. Res. Public Heal h 19, 1486 (2022). h ps:// doi. o g/ 10.
3390/ ije p h1903 1486
51. Besch eibung de Quali ä sindika o en und Kennzahlen nach
QSKH-RL. Gebu shil e. Endgül ige Rechen egeln ü das E as-
sungsjah 2020: Besch eibung de Quali ä sindika o en und Ken-
nzahlen nach QSKH-RL Endgül ige Rechen egeln, Ve ö en li-
chungs e sion (2021)
52. Becke , A., Eissle , U.: Die s anda disie e p imä e Sec io a e
(SPSR) und ih e Anwendung im Quali ä smanagemen und ü
K anken- haus e gleiche. P ädik o en de p imä en Sec io als
Bei ag zu Ve sachlichung eine komplexen Diskussion (S and-
a dised p ima y Caesa ian sec ion a e (SPSR) and how i can be
applied in Quali y Managemen and o Hospi al). In e discipli-
na y Con ibu ions o Hospi al Managemen : Medicine, Pa ien
Sa e y and Economics (2013)
53. Chen, N., Pan, J.: The causal e ec o deli e y olume on se e e
ma e nal mo bidi y: an ins umen al a iable analysis in Sichuan,
China. BMJ Glob. Heal h 7, e008428 (2022). h ps:// doi. o g/ 10.
1136/ bmjgh- 2022- 008428
54. B azie , E., Bo ell, L.N., Huynh, M., e al.: Va ia ion and acial/
e hnic dispa i ies in Cesa ean deli e y a New Yo k Ci y hospi als:
he con ibu ion o hospi al-le el ac o s. Ann. Epidemiol. 73, 1–8
(2022). h ps:// doi. o g/ 10. 1016/j. annep idem. 2022. 06. 003
55. Hoxha, I., Zhubi, E., G ezda, K., e al.: Caesa ean sec ions in
eaching hospi als: sys ema ic e iew and me a-analysis o hos-
pi als in 22 coun ies. BMJ Open 11, e042076 (2021). h ps:// doi.
o g/ 10. 1136/ bmjop en- 2020- 042076
56. Blundell, R., Bond, S.: Ini ial condi ions and momen es ic ions
in dynamic panel da a models. J. Econom. 87, 115–143 (1998).
h ps:// doi. o g/ 10. 1016/ S0304- 4076(98) 00009-8
57. A ellano, M., Bo e , O.: Ano he look a he ins umen al a iable
es ima ion o e o -componen s models. J. Econom. 68, 29–51
(1995). h ps:// doi. o g/ 10. 1016/ 0304- 4076(94) 01642-D
58. An onakis, J., Bas a doz, N., Rönkkö, M.: On igno ing he andom
e ec s assump ion in mul ile el models: e iew, c i ique, and ec-
ommenda ions. O gan. Res. Me hods 24, 443–483 (2021). h ps://
doi. o g/ 10. 1177/ 10944 28119 877457
59. Papies, D., Ebbes, P., Van Hee de, H.J. (2017). Add essing
Endogenei y in Ma ke ing Models. In: Lee lang, P., Wie inga, J.,
Bijmol , T., Pauwels, K. (eds) Ad anced Me hods o Modeling
Ma ke s. In e na ional Se ies in Quan i a i e Ma ke ing. Sp inge ,
Cham. h ps:// doi. o g/ 10. 1007/ 978-3- 319- 53469-5_ 18
60. Ge mann, F., Ebbes, P., G ewal, R.: The chie ma ke ing o ice
ma e s! J. Ma k. 79, 1–22 (2015). h ps:// doi. o g/ 10. 1509/ jm. 14.
0244
61. Rossi, P.E.: In i ed pape —e en he ich can make hemsel es
poo : a c i ical examina ion o IV me hods in ma ke ing applica-
ions. Ma k. Sci. 33, 655–672 (2014). h ps:// doi. o g/ 10. 1287/
mksc. 2014. 0860
62. A ellano, M., Bond, S.: Some es s o speci ica ion o panel da a:
Mon e Ca lo e idence and an applica ion o employmen equa-
ions. Re . Econ. S ud. 58, 277 (1991). h ps:// doi. o g/ 10. 2307/
22979 68
63. Roodman, D.: How o do X abond2: An in oduc ion o di e ence
and sys em GMM in S a a. S and. Genom. Sci. 9, 86–136 (2009).
h ps:// doi. o g/ 10. 1177/ 15368 67X09 00900 106
64. Lee, D.S., McC a y, J., Mo ei a, M.J., e al.: Valid - a io in e ence
o IV. Am. Econ. Re . 112, 3260–3290 (2022). h ps:// doi. o g/ 10.
1257/ ae . 20211 063
65. Tu ne , L., Culli o d, D., Ball, J., e al.: The associa ion be ween
midwi e y s a ing le els and he expe iences o mo he s on pos -
na al wa ds: c oss sec ional analysis o ou ine da a. Women Bi h
J. Aus . Coll. Midwi es 35, e583–e589 (2022). h ps:// doi. o g/ 10.
1016/j. wombi. 2022. 02. 005
66. Gallaghe , L., Smi h, V., Ca oll, M., e al.: Wha would educe
caesa ean sec ion a es? Views om p egnan women and clini-
cians in I eland. PLoSOne 17, e0267465 (2022). h ps:// doi. o g/
10. 1371/ jou n al. pone. 02674 65
67. Boh en, M.A., Ho mey , G.J., Sakala, C., e al.: Con inuous sup-
po o women du ing childbi h. Coch ane Da abase Sys . Re .
7, CD003766 (2017). h ps:// doi. o g/ 10. 1002/ 14651 858. CD003
766. pub6
68. Han, K.-T., Kim, S.J., Ju, Y.J., e al.: Do hospi al cha ac e is ics
in luence Cesa ean deli e y? Analysis o Na ional Heal h Insu -
ance claim da a. Eu . J. Public Heal h 27, 801–807 (2017). h ps://
doi. o g/ 10. 1093/ eu pub/ ckx061
69. Gha a i-Sa a i, A., Chaiken, S.R., Packe , C.H., e al.: Cesa ean
deli e y a es by hospi al ype among nullipa ous and mul ipa ous
pa ien s. J. Ma e n. Fe al. Neona al. Med. 35, 8631–8639 (2022).
h ps:// doi. o g/ 10. 1080/ 14767 058. 2021. 19908 84
70. Clapp, M.A., James, K.E., Melamed, A., e al.: Hospi al olume
and cesa ean deli e y among low- isk women in a na ionwide
sample. J. Pe ina ol. 38, 127–131 (2018). h ps:// doi. o g/ 10. 1038/
jp. 2017. 173
1022 A.S öcke e al.
71. Hen schke , C., Mennicken, R.: The olume-ou come ela ionship
e isi ed: p ac ice indeed makes pe ec . Heal h Se . Res. 53,
15–34 (2018). h ps:// doi. o g/ 10. 1111/ 1475- 6773. 12696
72. Hugue , M.: Cen aliza ion o ca e in high olume hospi als and
inequali ies in access o ca e. Soc Sci Med 260, 113177 (2020).
h ps:// doi. o g/ 10. 1016/j. socsc imed. 2020. 113177
73. Ma in-A ibas, A., Escu ie , R., Bo às-San os, A., e al.: A com-
pa ison be ween midwi e y and obs e ic ca e a bi h in Spain:
ac oss-sec ional s udy o pe ina al ou comes. In . J. Nu s. S ud.
126, 104129 (2022). h ps:// doi. o g/ 10. 1016/j. ijnu s u. 2021.
104129
74. Thiessen, K., Nickel, N., P io , H.J., e al.: Ma e ni y ou comes in
mani oba women: a compa ison be ween midwi e y-led ca e and
physician-led ca e a bi h. Bi h 43, 108–115 (2016). h ps:// doi.
o g/ 10. 1111/ bi . 12225
75. Ca lson, N.S., Co win, E.J., He nandez, T.L., e al.: Associa ion
be ween p o ide ype and cesa ean bi h in heal hy nullipa ous
labo ing women: a e ospec i e coho s udy. Bi h 45, 159–168
(2018). h ps:// doi. o g/ 10. 1111/ bi . 12334
76. Hamlin, L., G unwald, L., S u di an , R.X., e al.: Compa ison o
nu se-midwi e and physician bi h ou comes in he mili a y heal h
sys em. Policy Poli . Nu s. P ac . 22, 105–113 (2021). h ps:// doi.
o g/ 10. 1177/ 15271 54421 994071
77. Sou e , V., Ne he y, E., Kopas, M.L., e al.: Compa ison o Mid-
wi e y and obs e ic ca e in low- isk hospi al bi hs. Obs e .
Gynecol. 134, 1056–1065 (2019). h ps:// doi. o g/ 10. 1097/ aog.
00000 00000 003521
78. Hoxha, I., Fejza, A., Aliu, M., e al.: Heal h sys em ac o s and
caesa ean sec ions in Koso o: a c oss-sec ional s udy. BMJ Open
9, e026702 (2019). h ps:// doi. o g/ 10. 1136/ bmjop en- 2018- 026702
79. Sandall, J., Sol ani, H., Ga es, S., e al.: Midwi e-led con inu-
i y models e sus o he models o ca e o childbea ing women.
Coch ane Da abase Sys . Re . 4, CD004667 (2016). h ps:// doi.
o g/ 10. 1002/ 14651 858. CD004 667. pub5
80. Hanahoe, M.: Midwi e y-led ca e can lowe caesa ean sec ion
a es acco ding o he Robson en g oup classi ica ion sys em. Eu .
J. Midwi e y 4, 7 (2020). h ps:// doi. o g/ 10. 18332/ ejm/ 119164
81. Ri o, J.C., Amyx, M., Ping ay, V., e al.: Obs e ical p o ide s’
p e e ed mode o deli e y and a i ude owa ds non-medically
indica ed caesa ean sec ions: a c oss-sec ional s udy. BJOG In .
J. Obs e . Gynaecol. 125, 1294–1302 (2018). h ps:// doi. o g/ 10.
1111/ 1471- 0528. 15122
82. Kea ney, L., Kynn, M., C aswell, A., e al.: The ela ion-
ship be ween midwi e-led g oup-based e sus con en ional
an ena al ca e and mode o bi h: a ma ched coho s udy. BMC
P egnancy Childbi h 17, 39 (2017). h ps:// doi. o g/ 10. 1186/
s12884- 016- 1216-1
83. And aczek, T., Magis e , S., Bau zmann, S., e al.: Gebu im
hebammengelei e en K eißsaal eines Pe ina alzen ums—
Le nku e, E gebnisse und Benchma k (Bi h in he Midwi e-Led
Deli e y Room o a Pe ina al Cen e - Lea ning Cu e, Ou comes
and Benchma k). Z. Gebu shil e Neona ol. (2023). h ps:// doi.
o g/ 10. 1055/a- 2082- 2176
84. Be ge , C.: Mü e liches und neona ales gebu shil liches Ou -
come on geplan en En bindungen in einem Hebammenk eißsaal:
Koho ens udie in einem Zen um de Maximal e so gung
in Deu schland. Hebamme 33, 12 (2020). h ps:// doi. o g/ 10.
1055/a- 1292- 2055
85. Ullah, S., Zae a ian, G., Ullah, F.: How o use ins umen al a i-
ables in add essing endogenei y? A s ep-by-s ep p ocedu e o
non-specialis s. Ind. Ma k. Manage. 96, A1–A6 (2021). h ps://
doi. o g/ 10. 1016/j. indma man. 2020. 03. 006
86. Hoxha, I., Sadiku, F., Lama, A., e al.: Cesa ean deli e y and
gende o deli e ing physicians: a sys ema ic e iew and me a-
analysis. Obs e . Gynecol. 136, 1170–1178 (2020). h ps:// doi.
o g/ 10. 1097/ AOG. 00000 00000 004172
87. F eeman, M., Sa a, N., Schol es, S.: Ga ekeepe s a wo k: an
empi ical analysis o a ma e ni y uni . Manag. Sci. 63, 3147–3167
(2017). h ps:// doi. o g/ 10. 1287/ mnsc. 2016. 2512
88. F iedman, A.M., Anan h, C.V., Huang, Y., e al.: Hospi al deli e y
olume, se e e obs e ical mo bidi y, and ailu e o escue. Am. J.
Obs e . Gynecol. 215, 795.e1-795.e14 (2016). h ps:// doi. o g/ 10.
1016/j. ajog. 2016. 07. 039
89. Smi h, V., Hannon, K., Begley, C.: Clinician’s a i udes owa ds
caesa ean sec ion: a c oss-sec ional su ey in wo e ia y le el
ma e ni y uni s in I eland. Women Bi h J. Aus . Coll. Midwi es
35, 423–428 (2022). h ps:// doi. o g/ 10. 1016/j. wombi. 2021. 08. 004
90. Kingdon, C., Downe, S., Be an, A.P.: Non-clinical in e en ions
o educe unnecessa y caesa ean sec ion a ge ed a o ganisa ions,
acili ies and sys ems: sys ema ic e iew o quali a i e s udies.
PLoSOne 13, e0203274 (2018). h ps:// doi. o g/ 10. 1371/ jou n al.
pone. 02032 74
Publishe 's No 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.