OBSTETRICS

OBSTETRICS

Obstetrician or Family Physician:

Are Vaginal Deliveries Managed Differently?
Haim A. Abenhaim, MD, MPH,1,2 Michel Welt, MD,1 Robert Sabbah, MD,1 Franois Audibert, MD, MSc2
1

Department of Obstetrics and Gynecology, LHpital du Sacr-Coeur, University of Montreal, Montreal QC

Department of Obstetrics and Gynecology, LHpital Ste-Justine, University of Montreal, Montreal QC

Abstract
Background: In Canada, obstetricians and family physicians both
provide obstetrical care. However, the effect of specialty
training on obstetrical outcomes of low-risk pregnancies has not
recently been evaluated. In this study we examine the role of
specialty training on the management of vaginal deliveries.
Methods: We conducted a cohort study on all vaginal deliveries
that took place at Sacr-Coeur Hospital between July 2000 and
June 2006. We compared baseline characteristics of
obstetricians and family physicians and used an unconditional
logistic regression model to estimate the adjusted relative risk
of undergoing different obstetrical interventions.
Results: Of a total 8807 vaginal deliveries, 1915 were conducted
by eight obstetricians and 6892 were conducted by 21 family
physicians. Apart from a higher rate of induction of labour in
patients of obstetricians, baseline characteristics were
comparable between the two groups. Overall rates of use of
instruments were similar in the two groups; however, family
physicians were less likely than obstetricians to perform an
episiotomy (odds ratio [OR] 0.47; 95% confidence intervals [CI]
0.410.55) but more likely to have patients who sustained a
perineal injury (OR 1.51; 95% CI 1.361.68). There were no
differences in the incidence of third- and fourth-degree tears,
and 5-minute Apgar scores were similar in both groups.
Conclusion: Obstetricians and family physicians differ in the
performance of episiotomies, and their patients differ in the
resulting type of perineal injury. Instrument use and neonatal
outcomes were similar in both groups. Major maternal and
neonatal morbidity are unlikely to differ whether women with
low-risk pregnancies are delivered by an obstetrician or a family
physician.

Rsum
Contexte : Au Canada, les soins obsttricaux sont offerts tant par
des obsttriciens que par des mdecins de famille. Cependant,
leffet de la formation spcialise sur les issues obsttricales
des grossesses nentranant que de faibles risques na pas
rcemment fait lobjet dune valuation. Dans le cadre de cette
tude, nous examinons le rle de la formation spcialise sur la
prise en charge des accouchements par voie vaginale.

Key Words: Obstetrician, family physician, general practitioner,

delivery outcomes
Competing Interests: None declared.
Received on April 2, 2007
Accepted on June 20, 2007

Mthodes : Nous avons men une tude de cohorte portant sur tous
les accouchements vaginaux stant drouls lHpital Sacr-Cur
entre juillet 2000 et juin 2006. Nous avons compar les
caractristiques de base des obsttriciens et des mdecins de
famille, et nous avons fait appel un modle de rgression logistique
inconditionnelle afin destimer le risque relatif corrig de subir
diffrentes interventions obsttricales.
Rsultats : Sur un total de 8 807 accouchements vaginaux, 1 915 ont
t mens par huit obsttriciens et 6 892 ont t mens par
21 mdecins de famille. Exception faite dun taux accru de
dclenchement du travail chez les patientes des obsttriciens, les
caractristiques de base taient comparables entre les deux
groupes. Les taux globaux de recours des instruments taient
semblables dans les deux groupes; cependant, les mdecins de
famille taient moins susceptibles que les obsttriciens
deffectuer une pisiotomie (rapport de cotes [RC], 0,47; intervalles
de confiance [IC] 95 %, 0,410,55), mais plus susceptibles davoir
des patientes ayant subi une blessure prinale (RC, 1,51; IC
95 %, 1,361,68). Aucune diffrence na t constate en ce qui
concerne lincidence des dchirures du troisime et du quatrime
degr; de plus, les indices dApgar cinq minutes taient semblables
dans les deux groupes.
Conclusion : Les obsttriciens et les mdecins de famille se
distinguent en matire dexcution dpisiotomies, et leurs patientes
se distinguent en matire de type rsultant de lsion prinale. Le
recours des instruments et les issues nonatales taient
semblables dans les deux groupes. Il est peu probable que des
diffrences soient constates en matire de morbidit maternelle et
nonatale majeure entre les patientes qui prsentent une grossesse
nentranant que de faibles risques et qui ont recours aux services
dun obsttricien et celles qui ont recours aux services dun mdecin
de famille.
J Obstet Gynaecol Can 2007;29(10):801805

INTRODUCTION

t has been forecast that in the next five to 10 years there will
be a significant reduction in the number of obstetriciangynaecologists practising in Canada.1 Since the majority of
births in Canada are considered low risk, it is reasonable to
expect that the proportion of deliveries performed by family
physicians who provide obstetrical care will need to increase
to meet the demand.

Currently, obstetrician-gynaecologists provide obstetrical care

to 75% of women who deliver in the province of Quebec, and
the remaining proportion is largely provided by family
OCTOBER JOGC OCTOBRE 2007 l

801

OBSTETRICS

physicians who practise obstetrics. Caregivers who have

been trained in obstetrics and gynaecology residency programs have had substantially more exposure to both
low-risk and high-risk obstetrical scenarios during their
training than those trained in family medicine programs.
Whether or not this plays a role in the management of
low-risk pregnancies is unclear. The purpose of our study
was to compare the management practices and the outcomes of low-risk vaginal deliveries of obstetrical caregivers
trained as family physicians with those of obstetriciangynaecologists in light of the emerging need for low-risk
obstetrical caregivers.
METHODS

We conducted a retrospective cohort study of all vaginal

deliveries that took place at Sacr-Coeur during a six-year
period from July 1, 2000 to June 30, 2006. At Sacr-Coeur
Hospital, obstetrician-gynaecologists provide a 24-hour
rotating on-call service. All women who have been followed
by an obstetrician will be delivered by the obstetrician on
call during the time of delivery. The family physicians practise differently: some practitioners attend all deliveries of
the patients to whom they have provided antenatal care, and
others have a rotating on-call system to provide obstetrical
care to the women who have received antenatal care from
their group. The on-call obstetrician is available at any time
to provide obstetrical expertise (evaluation and management of dystocia, fetal distress, or other emergency) to
family physicians when consulted.
The data used for the creation of our study cohort were
obtained from the labour and delivery room delivery summary database. The nurse present at the time of delivery
enters the following information into the database: baseline
clinical characteristics, the identity of the attending physician, interventions used, and obstetrical and neonatal
outcomes.
The study population consisted of all women who had
undergone vaginal delivery of a singleton, and it excluded
women who had undergone Caesarean section or who had
a multifetal gestation. Baseline characteristics included
maternal age, parity, and gestational age at delivery. We also
included induction of labour, use of oxytocin for labour
augmentation, and epidural analgesia as baseline characteristics; these were potential confounders for the second stage

ABBREVIATIONS
CS

Caesarean section

NICU

neonatal intensive care unit

OR

odds ratio

802

l OCTOBER JOGC OCTOBRE 2007

management (defined as being conducted by a family physician or obstetrician-gynaecologist) and delivery outcomes.
In cases when an obstetrician acted as a consultant to the
family physician, the delivery remained assigned to the
family physician.
Study outcomes were categorized as either management
outcomes or obstetrical and neonatal outcomes. Management outcomes included use of an internal scalp monitor
for fetal heart rate monitoring, performance of an
episiotomy, and use of instrumentation. An instrumental
delivery included any delivery for which a forceps, a vacuum, or both instruments were applied during the delivery.
Obstetrical and neonatal outcomes included the occurrence
of a perineal tear, third- and fourth-degree tear, and
5-minute Apgar scores of 7 or 3. Perineal trauma
included episiotomy and vaginal or perineal tear.
A two-step analysis was performed. We first compared
baseline characteristics between subjects delivered by family physicians and those delivered by obstetriciangynaecologists. Second, we compared management
differences and obstetrical and neonatal outcomes between
family physicians and obstetrician-gynaecologists. We conducted this analysis using a logistic regression model to
estimate the relative risk of the outcomes and adjust for
potential sources of confounding. In our analysis, we
adjusted for maternal age, gestational age, parity, induction
and augmentation of labour, use of epidural analgesia, and
birth weight. Approval for the study was obtained from the
hospitals ethics committee.
RESULTS

There were 8807 vaginal deliveries conducted between July

1, 2000 and June 30, 2006. Baseline characteristics are
shown in Table 1. Of these deliveries, 21.7% were performed by eight obstetricians, and 78.3% were performed
by 21 family physicians. There was no significant difference
in maternal age, gestational age, parity, use of epidural analgesia, rates of augmentation of labour, or birth weight
between deliveries attended by obstetricians and those
attended by family physicians. Family physicians appeared
less likely than obstetricians to induce labour in their
patients (23.8% vs. 30.3%).
Management differences are presented in Table 2. Family
physicians were significantly less likely than obstetricians to
place an internal scalp monitor, to perform an episiotomy,
or to use forceps for delivery. When including the use of the
vacuum, overall use of instruments for delivery was not different. Obstetrical and neonatal outcomes are presented in
Table 3. Family physicians were 51% more likely than
obstetricians to have a patient sustain a perineal injury, and
after combining perineal injuries with episiotomies, family

Obstetrician or Family Physician: Are Vaginal Deliveries Managed Differently?

Table 1. Baseline characteristics of 8807 vaginal deliveries taking place

at Sacr-Coeur Hospital
Delivery attended by
obstetrician (n = 1915)

Delivery attended by family

physician (n = 6892)

21

< 25

19.7

21.5

2524.9

31.0

33.6

3034.9

31.5

30.7

35+

17.8

14.2

Baseline characteristic
No. of Physicians
Maternal age

Gestational age (weeks)

< 37

4.8

4.4

3738+6

23.7

22.4

3940+6

61.2

61.8

41+

10.3

11.4

Multiparous

49.6

51.0

Induction of labour

30.3

23.8

Augmentation of labour

31.1

31.3

Epidural analgesia

63.6

61.9

< 2500

3.6

2.3

25003499

55.7

54.0

35003999

33.4

32.6

Birth weight category (gm)

4000+

7.36

physicians had a 21% greater likelihood of having patients

with perineal trauma. There were no observed differences
in 5-minute Apgar scores of 7 or 3.
DISCUSSION

Obstetrical care in Canada is provided by both obstetricians

and family physicians. With the anticipated increase in need
for obstetrical care providers, the role of the family physician in delivering obstetrical care may play an important
role, particularly in low-risk pregnancies. Whether or not
there are differences in the management of vaginal deliveries between family physicians and obstetricians is unclear
and has not recently been evaluated. Our findings suggest
that obstetrician-gynaecologists and family physicians may
differ subtly in their management of women who deliver
vaginally. Nevertheless, maternal third- and fourth-degree
tears and adverse neonatal outcomes indicated by abnormal
Apgar scores do not appear to be affected by these differences. Major maternal or neonatal morbidities in low-risk
pregnancies are unlikely to be affected by the type of
physician attending the delivery.
The first methodological concern in our study was to define
our study population. The initial decision to make was

11.1

whether or not we would include all deliveries and then

examine differences in CS rate as an additional outcome. In
this study we elected to examine only vaginal deliveries for
two reasons. First, in our institution the decision to perform
a CS is ultimately made by the on-call obstetrician. A comparison of CS rates between obstetricians and family
physicians is less likely to reflect differences in management
practices between the groups and more likely to reflect the
individual obstetricians decision-making in the management of a higher-risk pregnancy. Second, to evaluate rates
of CS appropriately, it would have been imperative to consider both the indication for CS and the time between the
onset of labour and delivery, since these would reflect differences in the management of labour. Because these variables were not available, we chose to limit our study to
vaginal births.
A direct comparison of obstetrical management between
obstetricians and family practitioners has previously been
conducted.24 Investigators in a Canadian study,4 using a
retrospective study design, demonstrated that the rate of
instrumental delivery by family physicians was lower than
the rate of instrumental delivery by obstetricians, with no
differences in neonatal outcomes. Although the authors of
this study describe important findings, their study examined
OCTOBER JOGC OCTOBRE 2007 l

803

OBSTETRICS

Table 2. Management of 8807 deliveries

Obstetricians %

Family Physicians %

Crude OR

Adjusted OR*

Internal scalp monitor

15.1

11.9

0.76 (0.660.88)

0.82 (0.710.95)

Episiotomy

16.8

8.7

0.47 (0.410.55)

0.47 (0.410.55)

Vacuum

19.7

17.4

0.86 (0.760.98)

0.91 (0.801.04)

Forceps

1.8

0.7

0.39 (0.250.61)

0.40 (0.260.63)

20.1

17.7

0.86 (0.750.97)

0.91 (0.791.04)

Outcome

Any instrument

*Adjusting for age, gestational age, parity, birth weight, induction of labour, and augmentation of labour.
Some women may have had both a vacuum and a forceps applied.

Table 3. Obstetrical and neonatal outcomes

Outcome

Obstetricians %

Family Physicians %

Crude OR

Adjusted OR*

58.4

67.0

1.45 (1.301.61)

1.51 (1.361.68)

Perineal injury

70.9

73.6

1.14 (1.021.28)

1.21 (1.071.36)

Third- or fourth-degree tear

2.9

2.4

0.83 (0.611.13)

0.84 (0.611.15)

5-minute Apgar 3

1.6

1.9

1.19 (0.801.76)

1.34 (0.892.02)

5-minute Apgar 7

0.5

0.7

1.42 (0.722.80)

1.62 (0.783.34)

Perineal trauma

*Adjusting for age, gestational age, parity, birth weight, induction of labour, and augmentation of labour.
Including episiotomies.

obstetrical practices that are different from current standards. For example, the rate of continuous electronic fetal
monitoring in their study was 45%, compared with over
90% in hospital settings today. As well, rotation and
mid-forceps deliveries occurred in over 10% of their population; the incidence of this practice has greatly decreased in
the last decade.
The model of an obstetrical unit with care provided in parallel by family physicians and obstetricians is not uncommon.5,6 This enables family physicians to provide obstetrical
care with the reassurance of continuously available specialist back-up for operative management of dystocia, fetal distress, or medical and surgical problems in labour and delivery. The functioning of such a model has previously been
evaluated in a study by Berman and colleagues.5 In this
study, investigators demonstrated that a structured method
for defining obstetric privileges for family physicians
resulted in a high correlation between defined privileges and
the care delivered. This resulted in an increase in the units
capacity to deliver obstetrical care to women with low-risk
pregnancies.
There are several limitations to our study. First, it did not
compare rates of CS for obstetricians and family physicians.
Although this is an outcome of concern, the available data
prevented a between the groups that took into account the
804

l OCTOBER JOGC OCTOBRE 2007

indication for CS. Furthermore, since the decision to perform CS is ultimately made by the obstetrician, we questioned the value of comparing such an outcome. Second,
we did not compare rates of admission to a neonatal intensive care unit between the two groups. Although this may
be an important benchmark of quality of care in tertiary care
centres, the low annual rate of transfers of babies from our
low-risk centre to an outside NICU limits our ability to
detect differences in management of vaginal deliveries.
Finally, there remains the possibility that some patients
choose to be followed by one type of physician rather than
the other. Although this factor is difficult to measure and
account for, major variables likely associated with physician
preference that are more likely to be important confounders
(parity, epidural use, and management of labour) were
recorded and adjusted for in our analysis, thereby limiting
the potential effect of preference.
In Canada, the involvement of family physicians in vaginal
deliveries has been declining.7 This trend raises concern
about overall capacity to meet the increasing demands to
provide obstetrical care in light of anticipated shortages of
obstetrician-gynaecologists.1 Family physicians play an
important role in all aspects of low-risk perinatal care810;
this role should be emphasized, and initiatives should be
taken to reverse the current trends in their declining
involvement in obstetrical care.

Obstetrician or Family Physician: Are Vaginal Deliveries Managed Differently?

REFERENCES
1. SOGC News Release. Shortage of OB/Gyns, Aboriginal Health Top
SOGC Presidents Agenda. SOGC Annual Clinical Meeting, Vancouver,
June 2227, 2006.
2. Deutchman ME, Sills D, Connor PD. Perinatal outcomes: a comparison
between family physicians and obstetricians.[see comment]. J Am Board
Fam Pract 1995;8(6):4407.
3. Franks P, Eisinger S. Adverse perinatal outcomes: Is physician specialty a
risk factor? J Fam Pract 1987;24:1526.
4. Reid AJ, Carroll JC, Ruderman J, Murray MA. Differences in intrapartum
obstetric care provided to women at low risk by family physicians and
obstetricians. CMAJ 1989;140:62533.
5. Berman DR, Johnson TR, Apgar BS, Schwenk TL. Model of family
medicine and obstetrics-gynecology collaboration in obstetric care at the
University of Michigan. Obstet Gynecol 2000;96:30813.
6. Chany Y, Souteyrand P, Meyer JL, Collange C, Bruhat MA. Joint
management of pregnancy at risk by the general practitioner and
obstetrician [article in French]. Rev Prat 1987;37:4415.

7. Kaczorowski J, Levitt C. Intrapartum care by general practitioners and

family physicians. Provincial trends from 19841985 to 19941995. [see
comment]. Can Fam Physician 46: 58792.
8. Franks P, Fiscella K. Primary care physicians and specialists as personal
physicians. Health care expenditures and mortality experience. [see
comment]. J Fam Pract 1998; 47:1059.
9. Khan-Neelofur D, Gulmezoglu M, Villar J. Who should provide routine
antenatal care for low-risk women, and how often? A systematic review of
randomised controlled trials. WHO Antenatal Care Trial Research Group.
Paediatr Perinat Epidemiol 1998;12 Suppl 2:726.
10. Tucker JS, Hall MH, Howie PW, Reid ME, Barbour RS, Florey CD. Should
obstetricians see women with normal pregnancies? A multicentre
randomised controlled trial of routine antenatal care by general practitioners
and midwives compared with shared care led by obstetricians. [see
comment]. BMJ 1996; 312(7030):5549.

OCTOBER JOGC OCTOBRE 2007 l

805