SUPPLEMENT TO
American Journal of Obstetrics and Gynecology
Founded in 1920
volume 186 number 5 MAY 2002
The Nature and Management of Labor Pain: Executive Summary
The Nature and Management of Labor Pain Symposium Steering Committee:
Donald Caton, MD, Maureen P. Corry, MPH, Fredric D. Frigoletto, MD, David P. Hopkins, MD,
MPH, Ellice Lieberman, MD, DrPH, Linda Mayberry, PhD, RN, Judith P. Rooks, CNM, MPH,
Allan Rosenfield, MD, Carol Sakala, PhD, MSPH, Penny Simkin, PT, and Diony Young
New York, NY
This report describes the background and process for a rigorous project to improve understanding of labor
pain and its management, and summarizes the main results and their implications. Labor pain and methods
to relieve it are major concerns of childbearing women, with considerable implications for the course, quality,
outcome, and cost of intrapartum care. Although these issues affect many women and families and have
major consequences for health care systems, both professional and public discourse reveal considerable
uncertainty about many questions, including major areas of disagreement. An evidence-based framework,
including commissioned papers prepared according to carefully specified scopes and guidelines for
systematic review methods, was used to develop more definitive and authoritative answers to many
questions in this field. The papers were presented at an invitational symposium jointly sponsored by the
Maternity Center Association and the New York Academy of Medicine, were peer-reviewed, and are
published in full in this issue of the journal. The results have implications for policy, practice, research, and
the education of both health professionals and childbearing women. (Am J Obstet Gynecol 2002;186:S1-15.)
Key words: Labor pain, labor pain management, intrapartum care, informed consent, evidencebased maternity care
Background: Origins and rationale for The Nature
and Management of Labor Pain
Labor pain and methods to relieve it are major concerns of childbearing women and their families, with considerable implications for the course, quality, outcome,
and cost of intrapartum care. Although these issues affect
approximately 4 million women and families annually in
the United States alone and have major consequences for
health care systems, both professional and public discourse revealed considerable uncertainty, with major
areas of disagreement and controversy about the safety
and effectiveness of some methods to relieve labor pain,
and inattention to others. In addition, little attention ap-
Convened by the Maternity Center Association, New York City.
The views expressed by the Steering Committee do not necessarily reflect
those of the agencies, institutions, and organizations with which its members are affiliated.
Reprint requests: Maureen P. Corry, MPH, Executive Director, Maternity
Center Association, 281 Park Ave South, New York, NY 10010.
0002-9378/2002 $35.00 + 0 6/0/123102
doi:10.1067/mob.2002.123102
peared to have been given to women’s access to a choice
of methods in US hospitals, to organizational factors and
decision-making processes that determine methods used,
and to actual patterns of use of pain relief methods.
Since 1918, the Maternity Center Association (MCA) has
provided national leadership for woman- and family-centered maternity care. In 1999, MCA began a long-term national program to promote evidence-based maternity care.
The Nature and Management of Labor Pain is a project of
MCA’s Maternity Wise™ program, which helps childbearing
women, maternity caregivers, policymakers, and the media
understand the best scientific evidence about the safety
and effectiveness of specific elements of maternity care,
undertake effective strategies for making care more consistent with the best evidence, and obtain sound answers to
important questions about safe and effective care.
The evidence-based paradigm1-3 offers an approach for
moving from less definitive to more definitive conclusions
about the effects of specific forms of care. This paradigm
recognizes that conventional reviews of the literature have
lacked the methodologic rigor required to minimize bias
and develop firm conclusions. The “systematic review” has
S1
S2 Symposium Steering Committee
been developed to address these limitations. Before carrying out such a review, the reviewer specifies parameters and
procedures that will be followed. These include interventions, outcomes, populations, and study designs that will
and will not be included, along with strategies for a
thorough search for relevant reports, for evaluating the
methodologic adequacy of identified relevant studies, and
for summarizing results of studies that will be included in
the analysis. The reviewer then conducts the review by adhering to these plans as closely as possible.4
A large body of research supports and refines this approach,5, 6 and there is growing international consensus
that a well-conducted systematic review provides the best
possible answers to specific questions about the effects of
care, given the available research base of primary studies. A
systematic review enables comparison of alternatives and
determination, with respect to a specific outcome, of
whether or not they are different, or whether uncertainty
remains because of limited, flawed, or contradictory studies. In the context of these rigorous methods, a finding of
uncertainty is as important for health professionals and individuals facing personal health care decisions as is a more
definitive finding of either a difference or no difference.
The Maternity Center Association initiated a project to
apply this evidence-based approach to questions about
labor pain and its relief by commissioning a series of papers, holding a symposium where the papers would be
presented and discussed by invited leaders representing
all fields involved with labor pain issues, and making the
peer-reviewed papers available to a broader audience.
Process and methods
The Maternity Center Association asked Judith Rooks
to direct the project and entered into partnership with
the New York Academy of Medicine for the purpose of
planning and holding the symposium. The meeting was
jointly sponsored by and held at the New York Academy
of Medicine.
A multidisciplinary steering committee was formed comprising 12 members with expertise in anesthesiology,
bioethics, childbirth education, consumer advocacy, epidemiology, journalism, labor support (doula care), midwifery, neonatology, nursing, obstetrics, pediatrics, physical
therapy, and public health. Steering Committee members
participated in all major decisions about content, authors,
and quality of the papers. The group endeavored to ensure
that papers presented at the symposium and published in
this issue would offer a panoramic view of the field and
avoid duplication where possible, would summarize the
best research currently available and minimize bias, and
that the project would address needs and interests of childbearing women and their families.
To help authors, symposium participants, and others
understand the values and concerns underlying this project, the Committee developed a statement of values,
May 2002
Am J Obstet Gynecol
principles, and objectives.7 This statement gives priority
to meeting needs and interests of women and their families, obtaining evidence-based information about labor
pain and methods for relief of labor pain, ensuring that
women and their caregivers have access to such information, ensuring that women have access to a choice among
alternative pain relief methods, honoring and supporting
choices women make, and identifying and addressing important gaps in the literature.
The committee next developed a view of the field of
labor pain management as a whole and of individual papers that collectively might characterize this field. The
view included both papers about the safety and effectiveness of specific methods of labor pain relief and several
papers providing a context for the methods papers. Contextual issues include the nature of labor pain, the social
history of labor pain management, the contribution of
pain and pain relief to women’s satisfaction with childbirth, and issues of access and choice. The committee
considered methods that are widely used in the United
States and methods that might warrant greater availability
and use (eg, those now widely used in other western industrial nations), and decided to focus on epidural analgesia, opioids, nitrous oxide, paracervical block, and
nonpharmacologic methods, including continuous labor
support. To help avoid both duplication and gaps in important topics, the committee developed explicit and distinct scopes for these papers.
Some of the most contested information in the field involves possible undesired effects of epidural analgesia.
Some investigators have concluded that changes in technique over time have eliminated many previous concerns, and others have concluded that many concerns
persist. The range of stated conclusions among experienced investigators in this field posed a dilemma about
whom to invite to conduct a review to answer these questions; the committee did not want the choice of author to
either determine or appear to determine the results of a
review of this very important pain relief method. After
much deliberation, we decided to invite 2 capable and
widely respected individuals who had in the past come to
somewhat different conclusions about effects of this pain
relief method to prepare parallel papers using the same
specified scope and the same general and rigorous guidelines for systematic reviews.
Because epidural analgesia has rapidly become the most
commonly used method of labor pain relief in the United
States, a third paper on this topic was planned to address
side effects and effects on intrapartum care, including cointerventions that are routinely used or more likely to be
used to monitor, prevent, or address undesirable effects.
A subcommittee of the Steering Committee examined
leading guidelines for preparing systematic reviews, identified 6 that are based on epidemiologic principles and
incorporate findings from the science of research review-
Volume 186, Number 5
Am J Obstet Gynecol
ing,8-13 and adapted the most current versions of these
then available to develop guidelines appropriate for use
by symposium authors. The resulting guidelines14 reflect
international consensus about these methods and include the following key steps: (1) developing a written
plan describing intended scope and procedures, (2) carrying out the planned review (including literature search,
validity assessment, data abstraction, data synthesis, interpretation of results), and (3) developing a report that
presents the decision rules adopted, steps taken, and results. Although some guidelines limit included studies to
controlled trials, the Steering Committee’s guidelines anticipated that it would be important to enable authors to
include observational studies for many key questions
relating to labor pain relief.
The committee asked authors of all papers about specific
methods of pain relief and the author of the paper about
women’s satisfaction with childbirth to use the project’s systematic review guidelines to develop their papers. Although
this method was not fully applicable to the remaining papers, other authors were encouraged to apply elements of a
systematic review as fully as possible. Principles with broad
application include establishing clear criteria for inclusion
and exclusion and designing and carrying out a systematic
search for relevant material. David Hopkins, MD, a member of the Steering Committee and an epidemiologist with
experience in systematic review methods, critiqued drafts
of all papers from this perspective and was available to provide technical support to the authors.
The Steering Committee then identified strong candidates for preparing the commissioned reviews. They were
asked to adhere as closely as possible to the detailed written scope for their respective papers. Lead authors were
given the option of working alone or inviting co-authors
of their own choosing. As with the Steering Committee,
the authors brought perspectives of many fields to this
project, including anesthesiology, epidemiology, labor
support, midwifery, nursing, obstetrics, pediatrics, physical therapy, and political science.
Several months before the symposium, the Steering
Committee met to review drafts of the papers. With the intent of helping the authors make their papers as strong as
possible, the committee provided detailed feedback and
asked authors to revise their papers before the meeting.
To ensure a balance of participants across all professions
and organizations involved with labor pain, symposium attendance was by invitation only. Participants included
anesthesiologists, anthropologists, childbirth educators,
consumer advocates, doulas, epidemiologists, family physicians, hospital administrators, midwives, neonatologists,
nurses, nurse-anesthetists, obstetricians, pediatricians, political scientists, public health practitioners, social workers,
and persons from federal health agencies. The committee
wanted to involve people who were in a strong position to
make substantive contributions to the dialogue at the
Symposium Steering Committee S3
meeting and to use the information provided at the meeting to improve labor pain management within their professional activities and organizations.
The symposium was held at the New York Academy of
Medicine on May 4-5, 2001. The lead authors presented
10 commissioned papers. Substantial time was devoted to
questions and comments. The dialogue was rich, engaging, and frequently impassioned, representing a very
broad spectrum of perspectives. Many participants later
said that they had been exposed to new and important
perspectives that they rarely encountered in their usual
professional activities. Despite strong personal preferences, most notably in favor of epidural analgesia or of
unmedicated labors, there was general agreement that
women’s preferences and choices should be honored.
In addition to the papers and discussions, the symposium
included 2 multidisciplinary panels. Leaders of 4 relevant
disciplines were asked to compare, contrast, and critique
the 2 parallel papers on unintended effects of epidural
analgesia. Stephen B. Thacker, Director of the Epidemiology Program Office at the Centers for Disease Control and
Prevention, responded as an epidemiologist. Lawrence J.
Saidman, former editor of the journal, Anesthesiology, and
professor at Stanford University School of Medicine, provided an anesthesiology perspective. Michael F. Greene, director of maternal–fetal medicine at Massachusetts General
Hospital and member of the editorial board of the New England Journal of Medicine, offered an obstetric perspective.
Roberta A. Ballard, professor of pediatrics at the University
of Pennsylvania School of Medicine, responded from the
perspective of neonatology.
A multidisciplinary concluding panel included Valerie
A. Arkoosh (anesthesiology, MCP Hahnemann University), Eugene Declercq (childbirth education, political
science, public health, Boston University), Margaret
Comerford Freda (nursing, Albert Einstein College of
Medicine), Fredric D. Frigoletto (obstetrics, Harvard
University), Lisa L. Paine (midwifery and public health,
Boston University), and Stephen D. Ratcliffe (family medicine, University of Utah). Because of space constraints,
summaries of the 2 panels are available elsewhere,15
along with remarks of Charles S. Mahan (public health,
obstetrics, University of South Florida), who was scheduled to serve as a discussant to the final symposium paper
but was unable to attend.
The Maternity Center Association’s Carola Warburg
Rothschild Award that recognizes outstanding contributions to the health and well-being of women and their
families was presented during the meeting. The award
was given to Iain Chalmers, Murray Enkin, and Marc J. N.
C. Keirse, whose pioneering work in evidence-based maternity care16-18 helped to establish the evidence-based
paradigm for health and medicine in general.19 On behalf of the group, Dr Enkin accepted the award, which
was given “for refining the scientific synthesis of research
S4 Symposium Steering Committee
evidence and leading a dedicated team in the application
of these ‘systematic review’ methods to the entire field of
maternal and newborn care, for disseminating their findings in a series of landmark publications, for helping to
ensure that women, clinicians, and policy makers can
make decisions about maternity care that are informed by
the best available research, and for their continuing efforts to determine the safety and effectiveness of maternity care for mothers and babies.”
After the symposium, authors had an opportunity to revise their papers in consideration of the fruitful discussions that had taken place at the meeting, before sending
manuscripts to the journal. With the exception of this executive summary and a commentary on the two parallel
papers about epidural analgesia, all papers published in
this issue were peer-reviewed.
As a consequence of the process just described, the papers are rigorous, comprehensive, and complementary.
Some are also long because of their ambitious scopes and
standards for reporting systematic reviews. Some authors
are making less central, but important, information, such
as details about excluded studies, available elsewhere.15
Results
The following summaries of The Nature and Management of Labor Pain project papers will provide an introduction and orientation to the papers themselves. These
overviews cannot substitute for the individual papers,
which are exceptionally comprehensive. Both the papers
developed with the systematic review methods and their
summaries below are generally longer than papers that,
out of necessity, used a more conventional approach to
review of literature. The summaries appear in the order
of presentation at the symposium and in this issue.
“The Nature of Labor Pain.” In the opening paper,
Nancy K. Lowe synthesizes the best work from an extensive literature on the essence, characteristics, and factors
that contribute to the phenomenon of labor pain. Dr
Lowe’s research in this area has focused on measurement
of labor pain and factors contributing to the experience
of labor pain. Her far-reaching and up-to-date overview of
both theory and empiric research is of interest to a wide
audience.20 Systematic review methods were not appropriate for the paper’s diverse questions and data sources.
Dr Lowe applies Chapman’s conceptual model of pain
to the phenomenon of labor pain. The experience of
labor pain is a highly individual reflection of variable
stimuli uniquely received and interpreted through an individual woman’s emotional, motivational, cognitive, social, and cultural circumstances. The complexity and
individuality of this experience suggest that a woman and
her caregivers may have a limited ability to anticipate her
labor pain experience before labor, and that standardized and limited approaches to labor pain management
may not meet the needs of many women. Choice among
May 2002
Am J Obstet Gynecol
a variety of methods and individualization of pain-related
care is desirable.
In the context of a project that seeks to improve
women’s access to safe and effective labor pain relief, Dr
Lowe’s synthesis makes an important contribution by
identifying modifiable factors that may alleviate labor
pain independent of administering or undertaking any
specific method of labor pain relief. Modifiable factors
that have empirically been shown to influence labor pain
include the following: environmental conditions, coping
strategies, fear, anxiety, expectations about the experience, and, above all, a woman’s sense of self-efficacy or
confidence in her ability to cope.
In a related discussion, Dr Lowe underscores the striking qualitative difference between pain in the context of
helplessness, suffering, and loss, and pain in the context
of coping resources, comfort, and a sense of accomplishment. Care and resources available to women as they look
toward their birthing experiences and during the time of
labor and birth may influence whether the sensory intensity of labor pain is experienced in a fundamentally negative or positive manner.
The choice of a tool for measuring labor pain is an important consideration in research to compare effects of
different methods for pain relief or assess the effect of
other factors that may affect how it is experienced. Dr
Lowe provides a thoughtful discussion of the leading
tools, expressing greatest appreciation for the McGill
Pain Questionnaire and the Short-form McGill Pain
Questionnaire. Although visual analogue scales (VASs)
are widely used, they have many shortcomings for measuring labor pain.
The paper also provides an overview of the causes and
transmission of labor pain stimuli, identifies the need for
better research to understand the purported harmfulness
of labor pain, and reviews the evidence on physiologic
and psychosocial factors that affect how it is experienced.
“Anesthesia for Childbirth: Controversy and Change.”
Donald Caton, the author of What a Blessing She Had Chloroform,21 is widely recognized as a chronicler of the social
history of labor pain relief in the United States and Europe during the last 200 years. With Michael A. Froelich
and Tammy Y. Euliano, he reviews changes in technology,
knowledge, and the values and interests of both professionals and women since the mid-19th century.22 This
paper was not amenable to systematic review methods.
The methods that have been available and used for
labor pain relief reflect a complex interplay of technology,
knowledge, and values and interests of both professionals
and childbearing women. At different times, widely different views have prevailed. Professional views have ranged
from understanding childbirth as “a physiologic process,
best managed with the least possible interference” to the
importance of using “aggressive practices,” requiring intensive anesthetic intervention. Women’s views have simi-
Volume 186, Number 5
Am J Obstet Gynecol
larly ranged from the idea that pain and the avoidance of
medication “might have physiologic, psychologic, or social
value” to freedom from labor pain as “a necessity, if not a
‘right’.” The authors argue that the present complex environment includes the full spectrum of professional and
consumer viewpoints.
Dr Caton and colleagues then address the topic of
epidural analgesia, which has rapidly become the most
common method of labor pain relief in the United States.
They argue that this method offers flexibility in balancing
the trade-offs of the degree of pain relief against the degree of interference with other functions and review
physicians’ attempts to find an optimal balance by modifying choice of drug, dosage (volume and concentration), and method of administration (eg, use of plastic
catheters and infusion pumps).
They conclude by reviewing many intriguing observations that have been made about the effect of epidural
analgesia on maternal and fetal/newborn physiology. Despite these observations, current knowledge in this area is
inadequate. Research about epidural analgesia has
largely bypassed these important matters in favor of questions about technique and outcomes.
“Unintended Effects of Epidural Analgesia During
Labor: A Systematic Review” and “The Effects of
Epidural Analgesia on Labor, Maternal, and Neonatal
Outcomes: A Systematic Review.” Ellice Lieberman and
Barbara L. Leighton were both invited to prepare independent systematic reviews of unintended effects of
epidural analgesia on specified labor, maternal, and
fetal/newborn outcomes. Dr Lieberman is a physician
and perinatal epidemiologist with a special research interest in effects of epidural analgesia and other labor
management practices on maternal and neonatal outcomes. Dr Leighton, an anesthesiologist and former
president of the Society for Obstetric Anesthesia and
Perinatology, has conducted studies investigating
epidural analgesia outcomes and technique. Dr Lieberman collaborated with Carol O’Donoghue,23 and Dr
Leighton and Stephen H. Halpern updated and expanded a previously published meta-analysis.24, 25
Both teams made and carried out a priori decisions consistent with conventions for systematic review methods, yet
their methodologic decisions varied in fundamental ways.
Because of high rates of protocol noncompliance and study
group crossover in the randomized controlled trials (RCTs)
of epidural analgesia as well as difficulty examining less
common outcomes with this study design, Lieberman and
O’Donoghue included both RCTs and better quality observational studies. Leighton and Halpern limited their review
to RCTs, except when examining 2 outcomes for which
only observational studies were available: breast-feeding
success and urinary incontinence. The search strategy of
Lieberman and O’Donoghue was somewhat broader and
outcomes of interest were considerably broader, but in ex-
Symposium Steering Committee S5
cluding abstracts, their criteria for inclusion eliminated
some studies that the other team included. The Lieberman
team provides detailed narrative assessments of individual
studies, and the Leighton team used the Jadad scale26 to
rate individual RCTs. In developing conclusions about specific outcomes, Lieberman and O’Donoghue also summarize included studies narratively, and Leighton and
Halpern use formal meta-analytic techniques. Leighton
and Halpern limited comparison groups to women receiving opioids, whereas Lieberman and O’Donoghue included other available comparisons that met other criteria
for inclusion. The commentary by Stephen B. Thacker and
Donna F. Stroup that follows the parallel papers provides a
fuller discussion of these methodologies.27
The Lieberman and O’Donoghue23 paper is longer
than the Leighton and Halpern25 paper for several reasons. First, it specifically addresses an array of methodologic challenges, with the aim of careful interpretation
and a transparent process. The authors provide explicit
rationale for the rules used to carry out the review and, by
examining strengths and weaknesses of individual studies, explicit rationale for their interpretation of the data.
This discussion is an important resource for investigators
working in this field and practitioners attempting to
make sense of the evidence. It clarifies, for example, that
“practice-based” outcomes that vary greatly across
providers, facilities, and communities help to explain
some variation across studies and some confusion in the
field about unintended effects of epidural analgesia.
Second, it examines a broader range of outcomes of
interest with respect to unintended effects of epidural
analgesia (see Table I).
Third, it includes a series of analyses to examine whether
specific alterations in epidural analgesia technique affect
the incidence of unintended outcomes (see Table II).
Lieberman and O’Donoghue23 ask what difference it
makes if (1) epidural analgesia is discontinued late in labor
versus continued during the time of pushing and birth, (2)
epidural analgesia is delayed until greater cervical dilation
is achieved versus initiated earlier in labor, (3) “light” or
“walking” epidural analgesia versus “standard” epidural is
used, (4) anesthetic agents are administered continuously
versus intermittently, and (5) combined spinal–epidural
technique versus epidural analgesia alone is used?
Drs Leighton and Halpern25 include 2 important comparisons in their review that are not addressed in the
other review. Their meta-analysis finds that mothers who
receive epidural analgesia had lower pain scores and were
more satisfied with their analgesia than mothers in the
opioid groups. (Ellen D. Hodnett’s paper in this issue,28
summarized later, examines a different outcome, overall
satisfaction with the childbirth experience.)
Table I summarizes conclusions of the 2 teams about
unintended effects of epidural analgesia when compared
with other methods of labor pain relief. As shown in this
S6 Symposium Steering Committee
May 2002
Am J Obstet Gynecol
Table I. Best available evidence about unintended effects of epidural analgesia: Summary of conclusions from 2 parallel
systematic reviews
Possible unintended effects
that have been studied
Length of 1st stage labor
Conclusions of Lieberman and O’Donoghue23:
Effects of epidural analgesia vs control group*
Conclusions of Leighton and Halpern25:
Effects of epidural analgesia vs control group†
Possible association. Existing data insufficient for conclusion. No difference in length (↑ use of oxytocin
augmentation after analgesia)
↑ Length
↑ Length
↑ Instrumental delivery
↑ Instrumental delivery
No difference in instrumental delivery for
Insufficient data to evaluate association with specific
dystocia
indications such as dystocia.
Possible association. Existing data insufficient for conclusion. Outcome not reported
Possible association. Existing data insufficient for conclusion. No difference in cesarean delivery rate
Outcome not reported
↓ Spontaneous vaginal birth
Length of 2nd stage labor
Instrumental vaginal delivery rate: forceps or
vacuum extraction
Fetal malpresentation
Cesarean delivery rate
Spontaneous vaginal birth
rate: neither cesarean nor
instrumental delivery
Maternal fever ≥38°C
↑ Maternal fever
(and increased
temperature of fetus)
Effects on fetus and newNo evidence for difference in meconium-stained fluid, low
born (except for behavumbilical cord pH, low Apgar scores
ioral and neurologic
↑ Fetal tachycardia. Existing data insufficient for conclusion
effects)
about other fetal heart rate abnormalities
No data available to evaluate differences associated with use
of usual doses of opioid. ↓ Need for naloxone found in
randomized trials was associated with atypical use of high
opioid doses close to birth in some studies
Consistent association with hyperbilirubinemia
Existing data suggest no difference in newborn retinal hemorrhage
Effects of epidural-related ↑ Neonatal sepsis evaluation and antibiotic treatment
fever
(2 studies)
In one study of low-risk women, 98% of febrile mothers used
epidural analgesia compared with 55% of afebrile mothers, and maternal fever was associated with: low 1-minute
Apgar scores, hypotonia after birth, bag and mask resuscitation, oxygen therapy in nursery, and possibly neonatal
seizure
One study found that unexplained neonatal seizure is associated with exposure to intrapartum fever
Neonatal behavioral and
Epidural-exposed versus unmedicated infants*
neurologic outcomes
Studies using more comprehensive Brazelton Neonatal
Behavioral Assessment Scale (NBAS) found differences
favoring nonmedicated infants; no difference when less
comprehensive tools used
Epidural- vs opioid-exposed infants†
Existing data find small differences that do not overall favor
one or the other exposure. Epidural-exposed infants
tended to perform better on auditory orientation and
habituation, and opioid-exposed infants had better
muscle tone
Perineal laceration
↑ 3rd- and 4th-degree perineal laceration
Postpartum hemorrhage
Existing data insufficient for conclusion
and retained placenta
Breast-feeding success
Existing data insufficient for conclusion
Longer-term problems in
No evidence to support difference in new-onset long-term
the mother: new-onset
back pain
long-term back pain,
Existing data insufficient for conclusion about urinary
urinary problems
retention or stress incontinence
↑ Maternal fever
No difference in fetal heart rate abnormalities or intrapartum meconium, low
5-minute Apgar scores, low umbilical
cord pH, severe asphyxia
(↑ maternal hypotension)
↓ Low 1-minute Apgar scores
↓ Need for naloxone
Outcomes not reported
Outcome not reported
Outcomes not reported
Outcomes not reported
No difference in breast-feeding success at 6 wk
No difference in new-onset mid- or lowback pain at 3 mo and 12 mo
↑ Urinary incontinence in immediate
postpartum period
No difference in urinary incontinence at 3
or 12 mo
↑, authors conclude evidence exists for increased association with epidural analgesia; ↓, authors conclude evidence exists for decreased
association with epidural analgesia.
*This review includes best available data from both randomized controlled trials (RCTs) and observational studies. Most women enrolled in comparison groups of RCTs received opioids. Control groups in observational studies often combined all other types of pain
relief or did not provide detailed information about the types of pain relief used by women not receiving epidural.
†All women enrolled in comparison groups in this review received opioids. In included RCTs, all epidural protocols called for the use
of bupivacaine (plus other drugs); nearly all opioid protocols were for intramuscular or intravenous meperidine (plus other drugs). Observational studies were used only to examine breast-feeding success and postpartum urinary incontinence because RCTs were not
found.
Symposium Steering Committee S7
Volume 186, Number 5
Am J Obstet Gynecol
Table II. Best available evidence about the impact of modification of technique on unintended effects of epidural analgesia: Conclusions of systematic review of Lieberman and O’Donoghue23
Modification of technique used
to minimize unintended
effects of epidural analgesia
Objectives of Modification
of Technique
Effects of Modified Technique
Discontinuing administration late in 1st stage vs
continuing during 2nd
stage
Delaying start of analgesia
until greater cervical dilation vs initiating with less
dilation
Using “light” (↓ concentration of local anesthetic
and addition of opioid)
vs “standard” doses
Administering continuous
infusion vs intermittent
infusion
Increase ability to push
Increase rate of spontaneous
vaginal birth
Existing data insufficient for conclusion about mode of delivery
and length of second stage labor
No evidence for difference in fetal outcomes
Improve labor progress
Increase rate of spontaneous
vaginal birth
Existing data insufficient for conclusion about fetal malposition
and instrumental vaginal delivery
Delay may reduce cesarean delivery but existing data insufficient
for definitive conclusion
No evidence for difference in: length of labor, instrumental and
cesarean delivery rates, or fetal outcomes
Using combined spinalepidural (CSE) analgesia
vs epidural alone
Rapid onset of pain relief
Decrease motor block
Reduce length of 2nd stage
Reduce instrumental delivery
Reduce pain experienced before
new bolus
Reduce hypotension with bolus
administration
table, the 2 teams agreed that epidural analgesia increases the likelihood of (1) longer second stage labor,
(2) instrumental delivery, and (3) maternal fever.
Two unintended outcomes were only addressed by
Leighton and Halpern,25 and thus also do not involve disagreement. They conclude that epidural analgesia
increases the likelihood of the use of oxytocin augmentation after analgesia and maternal hypotension.
Many outcomes were only addressed by Lieberman and
O’Donoghue,23 and thus also do not involve disagreement. They conclude that epidural analgesia (1) decreased spontaneous birth, (2) increased neonatal sepsis
evaluation and antibiotic treatment, (3) increased 3rdand 4th-degree perineal laceration, and (4) had consistent association with hyperbilirubinemia.
They also find that epidural analgesia may be associated
with the following outcomes (about which further research
is needed): (1) fetal malpresentation, (2) infant hypotonia
after birth, bag and mask resuscitation, oxygen therapy in
nursery, and unexplained neonatal seizure, in infants of
women with epidural-related fever, (3) decreased performance of infant on comprehensive Neonatal Behavioral Assessment Scale (NBAS) compared with nonmedicated
infants (but no difference when testing was done with less
comprehensive assessment tools), and (4) better infant auditory orientation and habituation but poorer muscle tone
on NBAS compared with infants exposed to opioids.
These researchers found no evidence for difference in
newborn retinal hemorrhage or new maternal backache.
No evidence for difference in cesarean or instrumental vaginal
deliveries
Existing data insufficient for conclusion about fetal heart rate abnormalities and length of labor
No evidence for difference in fetal outcomes
Existing evidence suggests no difference for cesarean or instrumental delivery rates, fetal malposition, length of labor
Possible association of CSE with fetal heart rate abnormalities but
data not conclusive
No evidence for difference in meconium-stained fluid, umbilical
cord pH
Available data were insufficient to reach conclusions
about association of epidural analgesia with postpartum
hemorrhage and retained placenta.
The 2 teams came to different conclusions about effects of epidural analgesia on the many outcomes that remain in Table I. These include the following:
1. Length of first stage labor: possible association, data
insufficient for conclusion (Lieberman team); no
difference (Leighton team)
2. Cesarean delivery rate: possible association, data insufficient for conclusion (Lieberman team); no difference (Leighton team)
3. Breast-feeding success: data insufficient for conclusion (Lieberman team); no difference at 6 weeks
(Leighton team)
4. Urinary problems: data insufficient for conclusion
about urinary retention or stress incontinence
(Lieberman team); increased urinary incontinence
after birth, no difference at 3 or 12 months
(Leighton team).
Attention to the authors’ respective methodologic decisions and study quality assessments helps to explain
these discrepancies. Readers are encouraged to consult
the 2 papers.
Table II summarizes results of the analysis of Lieberman
and O’Donoghue about the effect of modifications of
epidural technique on unintended effects of this method of
labor pain relief. Depending on the outcome and technique, this analysis finds that no evidence for a difference
S8 Symposium Steering Committee
exists, existing evidence suggests no difference, or existing
data are insufficient for a conclusion. These results call into
question the widespread assumption that these changes in
technique make a difference in outcome.
Given the rigorous process that has been used to develop the conclusions in these papers, the Steering Committee believes that clinicians, women, and policymakers
should have access to clear and full information about
both outcomes and conclusions about which there is no
disagreement, and outcomes and conclusions about
which disagreement and uncertainty persist. Women
need this information well in advance of labor and again
during labor.
“Epidural Analgesia Side Effects, Co-Interventions and
Care of Women During Childbirth: A Systematic Review.”
Linda J. Mayberry was invited to conduct reviews to identify
and describe the incidence of the most common intrapartum side effects of epidural analgesia, to describe the
effects of epidural analgesia on the use of other medical interventions during labor, and to examine associated modifications in nursing care. Dr Mayberry is a nurse-researcher
whose interests include appropriate intrapartum care for
women with epidural analgesia and promotion of evidence-based practice among intrapartum nurses. With
co-authors Donna Clemmens and Anindya De, she conducted a systematic review of RCTs that had a stated goal of
seeking methods for minimizing the incidence of side effects.29 They limited their review of the incidence of side
effects to studies published since 1990 to focus on agents
and techniques in current practice.
This review of side effects provides information about
the effect of modified epidural analgesia techniques. Notably, 18 of the 19 included studies examined “light” or
“walking” epidural analgesia, which includes opioids to
enable lower concentrations of local anesthetic with the
objective of limiting hypotension and lower extremity
motor blockade.
Several side effects were common and varied widely
among study groups, suggesting that these effects might
be associated with particular agents, dosages, drug combinations, or other factors. These side effects include:
1. Itching (17 studies): very common in groups receiving agents including opioids (had on average by
62%), whereas only 0% to 4% in groups receiving
analgesia without opioids had itching.
2. Voiding difficulties (4 studies): inability to void ranged
from 0% to 35% in 2 studies and 62% to 68% in another; 28% to 61% of women were catheterized in the
only study that reported this outcome.
3. Sedation (6 studies): 21% of the women on average
had sedation, but the range varied considerably
across groups (1%-56%).
4. Hypotension (16 studies): incidence in all study
groups ranged from 0% to 50%; groups in 6 studies
had hypotension rates of 24% or higher.
May 2002
Am J Obstet Gynecol
5. No ambulation (8 studies): 0% to 25% of women in
all groups did not walk at any time in labor in 2 studies, 30% to 45% did not walk in 2 studies, and 52%
to 85% did not walk in 4 studies.
Incidence of nausea and vomiting was relatively low
and may be unrelated to this method of pain relief. Although incidence of shivering was also low, this outcome
was examined in just 2 small studies.
One of the most important findings in the review by
Mayberry et al29 is that a large proportion of women who
receive what has been called a “light” or “walking”
epidural do not ambulate at all during labor. Even within
the 3 groups that were expressly encouraged to walk,
rates of no-ambulation were high (34%-85%). Possible
reasons for this general finding include: opioids contribute to drowsiness and fatigue, motor block interferes
with ability or stability, women are confined to bed by
tubes and cords connecting them to various devices,
nurses have other responsibilities and are not available to
assist with ambulation, and policies or advice of caregivers discourage ambulation.
The authors also examined the effect of epidural analgesia on the use of other interventions during labor. Electronic fetal monitoring, intravenous infusions, and
frequent blood pressure monitoring are used routinely
with epidural analgesia. The best available research finds
that routine use of these interventions could be avoided
with low-risk women without a specific indication such as
epidural analgesia.30 The authors discuss co-interventions
that women may have more frequently with epidural
analgesia, including oxytocin augmentation, bladder
catheterization, and drugs for hypotension. The questionable practice30 of directed, sustained pushing with
breath-holding may also be more frequent with epidural
analgesia.
The review concludes by discussing implications of
both the side effects and care used to monitor, prevent,
or address them for informed consent, the nursing care
required by women who have epidurals, the availability of
supportive care for women, hospital costs, and ambulation during labor.
Use of epidural analgesia may involve a complex cascade of intervention. For example, the authors note that
this method of pain relief may impair a sense of bladder
distention, a full bladder may impede uterine contractility, women may be at increased risk for bladder trauma or
lack of postpartum bladder tone, and catheterization may
introduce infection. Research is needed to clarify the interplay of these and other effects and interventions.
The authors stress the importance of giving women
who are considering the use of epidural analgesia clear
information during pregnancy about co-interventions
that would or may be required and side effects that may
occur. Those considering this pain relief method should
understand that for most laboring women, having an
Volume 186, Number 5
Am J Obstet Gynecol
epidural would make labor more technology-intensive
than it would otherwise need be. It would be inappropriate to imply that a woman who chooses a “light” or “walking” epidural will definitely be interested in and able to
walk. The term “walking” epidural is misleading and
should be discontinued. The appropriateness of the term
“light” is also called into question, given possibilities for
motor block and/or sedation.
“Parenteral Opioids for Labor Pain Relief: A Systematic
Review.” Although epidural analgesia is apparently becoming the most frequently used method of labor pain
relief in the United States, opioids are widely used as firstline labor pain medication that may preclude or precede
epidural and in settings where epidural analgesia is not
available. A recent survey found that parenteral opioids
are given to 39% of women in US hospitals with >1500
births per year, 56% in hospitals with 500 to 1500 births
per year, and 50% in hospitals with <500 births per year.31
Leanne Bricker was invited to prepare a systematic review
examining the safety and effectiveness of opioids for
labor pain relief. Dr Bricker is a clinical lecturer in obstetrics whose research interests include intrapartum
management and clinical trials; she is working in the
United Kingdom. As a member of the Cochrane Collaboration’s Pregnancy and Childbirth Review Group, she is
well versed in systematic review methods. In this issue, Dr
Bricker and co-author Tina Lavender review 48 RCTs that
met a priori criteria for relevance and quality.32
Most trials reviewed by Bricker and Lavender32 compared opioid with opioid, including different opioid
agents, dosages, routes or techniques of administration,
and opioid with codrugs added. Eleven trials compared
intravenous opioid with epidural analgesia, one compared intravenous opioid with paracervical block, and
only one was placebo-controlled.
The authors32 calculate that some included trials
were powered to address the primary outcome of clinical effectiveness, defined as maternal satisfaction with
pain relief in 1 to 2 hours; none were powered to address the primary outcome of safety, defined as need for
infant resuscitation after normal birth. Only 39% of the
trials reported data for the primary clinical effectiveness outcome, and 15% reported data for the safety outcome.
How effective are opioids for labor pain relief? Bricker
and Lavender32 conclude that there are “considerable
doubts about its effectiveness for maternal pain relief.”
Epidural analgesia is consistently more effective. In the
single (double-blind) trial comparing opioid with another agent, paracervical block was more effective for 1
hour after administration, after which there was no difference. In the single (double-blind) trial comparing opioid with placebo, more women were dissatisfied with pain
relief in the placebo group (71% vs 83%), but the opioid
results are not impressive.
Symposium Steering Committee S9
Although the included trials were underpowered to address the primary safety outcome, the authors32 note that
opioids readily cross the placenta and identify observational studies consistently showing adverse effects of opioids on newborns. They identify concerns about the use of
the opioid antagonist naloxone. They also point to a series
of studies that have associated self-destructive and addictive
behaviors later in life with fetal exposure to opiates during
labor. Opioids are also associated with troubling maternal
side effects, such as nausea, vomiting, and sedation, and
may also pose more serious risks to mothers. Little or no
data were available about many secondary, and arguably important, outcomes of interest, most notably breast-feeding
and mother-infant bonding.
Bricker and Lavender32 provide a series of analyses
comparing effects of opioids and epidural analgesia during labor. They conclude that opioids are associated with
greater pain and less satisfaction with pain relief method,
but shorter first- and second-stage labor, less oxytocin
augmentation, fewer fetal malpositions, and fewer instrumental vaginal deliveries. They found no difference in cesarean delivery rates, low 5-minute Apgar scores, and use
of naloxone in trials comparing these 2 methods.
Existing evidence provides very limited support for
widespread use of opiates for labor pain relief. Bricker
and Lavender32 conclude that if women and their caregivers do choose opioids, pethidine is the agent of choice
because it is familiar to caregivers throughout the world
and relatively inexpensive, whereas no convincing research evidence supports preference for a different opioid. They end their contribution by identifying a large
research agenda relating to the use of opioids in labor.
“Nitrous Oxide for Relief of Labor Pain: A Systematic
Review.” Recent surveys in Canada and the United Kingdom indicate that many women in those countries use nitrous oxide for labor pain relief. Sixty-eight percent of
Canadian hospitals responding to a national survey of
routine maternity care in 1993 reported that this method
was available for laboring women at their facility; pooled
facility-level estimates suggest that about 37% of Canadian women used this method during labor at the time of
the survey.33 Fully 99% of UK maternity units responding
to a national survey about labor pain and its relief in 1990
reported that nitrous oxide was available for laboring
women. In a simultaneous national survey of women just
after birth, 60% reported using nitrous oxide for labor
pain relief, and 85% of these rated pain relief with this
method as “very good” or “good.”34
A recent national survey of the use of pharmacologic
methods of labor pain relief in the United States did not
even mention nitrous oxide.31 Although formerly more
widely available in the United States, relatively few laboring women appear to have access to this method at this
time. Mark A. Rosen prepared a systematic review35 of the
safety and effectiveness of nitrous oxide to help consider
S10 Symposium Steering Committee
whether women in the United States might welcome and
benefit from increased access to this method of pain relief. Dr Rosen’s experience as director of a residency program in a hospital that offers nitrous oxide to laboring
women helped inform his review in this issue. This experience was appreciated at the symposium, where many
participants had little previous knowledge of this method
and expressed considerable interest.
In the introduction to his paper,35 Dr Rosen discusses
the possible mode of action of nitrous oxide, as well as the
agents and relatively simple and inexpensive equipment
used to administer it. Women self-administer this inhalation anesthetic through a face mask or mouthpiece, and a
demand valve closes when they are not exerting negative
pressure. It is a flexible method that may be used at any
time and for any duration during labor, if and as needed. It
can be used alone, as a complement to other pharmacologic methods (eg, before administration of epidural), or
in combination with nonpharmacologic measures. It takes
effect in about 50 seconds and can be used intermittently
with contractions or continuously. When used intermittently, women who anticipate their contraction and begin
to breathe into the mask or mouthpiece slightly ahead of it
obtain better relief than women who wait for the contraction to begin. Continuous use may cause a high level of maternal sedation, dizziness, or lightheadedness. Physicians
of various specialties, midwives, and nurses can all supervise the use of nitrous oxide.
To examine the effectiveness of pain relief with nitrous
oxide, Dr Rosen identified 11 RCTs meeting his criteria
for relevance and quality. Nine of these studies involved
either study groups or crossover designs with other inhalation agents. Because of the predominant comparison
with other inhalation agents and other limitations, the
best available studies are difficult to compare and summarize. In most studies, the great majority of women who
used nitrous oxide gave high ratings to their pain relief.
Many chose to continue to use nitrous oxide after the
study ended and indicated that they would use that
method again. A study comparing different concentrations found a dose-response effect. Although a placebocontrolled study found no difference in ratings of pain, it
was conducted in early labor and used an intermittent
technique beginning with the onset of a contraction that
limited the effectiveness of the drug at a time when pain
levels are generally low. Dr Rosen estimates that “although the efficacy of nitrous oxide seems limited compared with epidural analgesia ..., it appears to provide
analgesia at a level comparable with paracervical block,
and probably better than that provided by opioids.”
Seven trials used to assess effectiveness also included
data on adverse outcomes and were included in an analysis of safety. An additional 11 observational studies offered larger sample sizes and were included to examine
safety. Dr Rosen concludes, “After maternal use of nitrous
May 2002
Am J Obstet Gynecol
oxide for labor analgesia, infants are clinically unaffected.” Maternal dreams or drowsiness have been reported (0-24%), and women may also experience hazy
memory during labor. The greatest maternal risk is loss of
consciousness, a rare occurrence that can be prevented
by self-administration (hand falls away with drowsiness),
avoiding high drug concentrations, and limiting co-use
with opioids. The available literature is not amenable to
concise summary and definitive conclusions about other
outcomes, and we refer readers to Dr Rosen’s paper.35
This review also addresses concern about occupational
exposure of health care workers to nitrous oxide and other
anesthetic gases, as raised in a number of earlier studies. Dr
Rosen cites a meta-analysis concluding that the earlier studies do not establish an association between these agents and
outcomes of concern. He notes that hospital facilities in the
United States are well-ventilated, and that Nitronox machines that deliver a blend of nitrous oxide and oxygen
have an active scavenging device.
On the basis of his review and personal experience, Dr
Rosen offers recommended guidelines for the use of nitrous oxide during labor and birth. He concludes by
identifying priorities for research, including more rigorous studies and studies examining questions of timing of
administration, equipment modification, co-administration with other medications, supervision by different
types of health care providers, safety, effects on breastfeeding, and effects during different stages of labor.
Serious consideration of ensuring greater access to nitrous oxide in the United States is warranted. Women and
hospital administrators and staff may appreciate the availability of nitrous oxide for many reasons:
1. A large proportion of women appear to obtain adequate pain relief and to be satisfied with this
method.
2. As women self-administer it, they can maintain a
sense of control and reduce burdens on staff.
3. Effects appear to reverse rapidly when women stop
inhaling the drug.
4. It does not appear to interfere with labor physiology.
5. Use of co-interventions that require monitoring and
management and increase risks to the mother and
fetus/infant appears to be limited.
6. Once incorporated into practice, the technique is
simple and inexpensive.
“Paracervical Block for Labor Analgesia: A Brief Historical Review.” Paracervical block, a local bilateral injection near the cervix, is given to block pain during the first
stage of labor. The analgesic effect lasts for about 2 hours
or longer if codrugs are included to prolong the effect.
This method of labor pain relief is rarely available and
used in the United States at this time and appears to be
primarily used in rural areas and small towns. The Steering Committee wished to determine whether it had been
Volume 186, Number 5
Am J Obstet Gynecol
discontinued for good reason, or whether it might warrant greater availability. Mark A. Rosen prepared a review36 that examined the safety and effectiveness of
paracervical block. His review, which appears in this issue,
relies primarily and necessarily on earlier studies and uses
conventional review methods.
Concerns about the risks of fetal bradycardia have led
to attempts to better understand the effect of paracervical
block on maternal–fetal physiology and to decreased use
of this method. Currently, investigators lack consensus
about the physiologic basis for postparacervical block
fetal bradycardia. Tragic technique errors involving injection of paracervical block into the fetal head have been
another leading concern.
To examine the effectiveness of paracervical block, Dr
Rosen identified 4 RCTs meeting a priori criteria for relevance and quality. In these trials, about 75% of women
rate their pain relief as “good” or “excellent.” There is
considerable heterogeneity of technique in both of these
studies and actual practice.
Dr Rosen sought RCTs to determine the incidence of fetal
bradycardia. The evidence suggests that bradycardia occurs
in about 15% of fetuses exposed to paracervical block. In
most instances, the decrease in fetal heart rate is mild and
transient. Trial size does not provide a basis for identifying
the incidence of severe and harmful bradycardia.
Dr Rosen examined many types of studies to identify
other unintended outcomes of paracervical block. Possible maternal complications include abcess, hematoma,
and neuropathy. He found no evidence of adverse effects on labor progress, spontaneous vaginal birth, and
breast-feeding.
He concludes that “PCB is an effective and relatively simple technique, although the skill and experience of the operator are among the most important variables related to
both efficacy and safety.” The data reviewed do not clarify
whether specific training and technique can limit or eliminate the occurrence of severe bradycardia and other safety
concerns with paracervical block. Further research is
needed to address important safety questions.
“Nonpharmacologic Relief of Pain During Labor: Systematic Reviews of Five Methods.” Although the subject of nonpharmacologic measures for relief of labor pain is relatively
neglected in the health and medical literature, it is relevant
to virtually all childbearing women: those who choose the
most effective medications possible but need help before
these medications can be administered and take effect or
after they are discontinued, those who may welcome drugfree measures for pain relief as a complement to less effective medications, and those who prefer to labor entirely
without pain medications. National data are unavailable to
describe both childbearing women’s access to and use of
drug-free pain relief measures in the United States, which
appear to be quite limited and far from commensurate with
this universal relevance.
Symposium Steering Committee S11
Penny Simkin, who has done extensive work in the
field of nonpharmacologic measures for labor pain relief,37-41 was invited to prepare systematic reviews of selected measures. To narrow this vast topic, she limited her
scope to methods that require institutional support (ie,
specific skills, policies, and/or equipment) and have
been evaluated by methodologically adequate, controlled
prospective studies. By using these criteria, Penny Simkin
and co-author MaryAnn O’Hara present in this issue systematic reviews of the effects of the following nonpharmacologic measures on pain and related outcomes:
continuous labor support, baths, touch and massage, maternal movement and positioning, and intradermal water
blocks for relief of back pain.42
Simkin and O’Hara42 identify the following components of labor support: physical comforting, emotional
support, guidance and support for the woman’s partner, information, and facilitation of communication between the woman and the hospital staff. Although a
systematic review of 14 RCTs conducted throughout the
world finds that this type of care is associated with a
broad range of favorable outcomes and no identified
drawbacks,43 relatively few women in the United States
have access to this kind of care during labor. In this context, Simkin and O’Hara make an important contribution by conducting a new systematic review of the
effects of labor support on pain indicators and related
outcomes under the distinctive conditions existing in
North America. Their review of 9 North American RCTs
finds that “continuous labor support by a trained
layperson provides relief of pain and improves other
outcomes, to a greater degree in low-income women
who are not accompanied by a loved one than among
middle-class accompanied women.”
A growing number of hospitals in the United States appear to be making immersion in water available to laboring women. Simkin and O’Hara42 review 2 prospective
cohort studies and 7 RCTs assessing the effect of bathing
on labor pain and related outcomes. They conclude that
it is a “safe, popular, and promising method of temporary
pain relief in labor.” Their analysis provides important insights about such matters as water temperature, timing of
entry, water depth, duration of immersion, and physiologic effects. This discussion and the authors’ proposed
guidelines for baths during labor will be of particular interest to staff in facilities offering such baths, childbirth
educators, doulas, women themselves, and investigators
planning research in this area.
Because movement and positioning may be safe and
simple ways to enhance comfort during labor, Simkin
and O’Hara42 review 14 trials that investigate this question. Unfortunately, problems with the studies make interpretation difficult. For example, study protocols
directing women to move in specific ways beginning at
specific points for a specific length of time had high
S12 Symposium Steering Committee
rates of noncompliance. The authors conclude that
predominant use of upright positions during the first
stage of labor and squatting during the second stage
may speed labor and increase mothers’ comfort levels.
They underscore the need for trials that compare a policy of women’s freedom of movement with one of restriction to a labor bed.
Despite an extensive search, Simkin and O’Hara42
identified just one controlled study of the effects of touch
and one of massage on outcomes related to labor pain.
Although these studies do not enable a clear conclusion,
they suggest that reassuring touch and massage during
labor “may relieve pain, reduce anxiety, and enhance
labor progress, with no identified risks.”
A large proportion of women have low-back pain during labor, both with and without fetal occiput posterior
position.20 Simkin and O’Hara42 review 4 RCTs that assess
the effects of intradermal water blocks for relief of back
pain in labor. These well-designed studies are consistent
in demonstrating “a significant reduction in back pain
during labor from [four sterile water] injections into the
skin overlying the sacrum, but maternal satisfaction with
this method varies.”
The authors42 end this wide-ranging review with a summary of priorities for research and recommendations to
help hospitals and caregivers make these methods more
widely available. On the basis of their review, they propose a simple but far-reaching pain management protocol that combines nonpharmacologic methods with
pharmacologic relief, as desired and appropriate, and offers a woman “safe and effective choices that allow for
flexibility, individual attention, adequate pain relief, and
consideration of her psychosocial needs.”
“Pain and Women’s Satisfaction with the Experience of
Childbirth: A Systematic Review.” To develop a better understanding of the perspectives of women themselves,
Ellen D. Hodnett prepared a systematic review that examined factors contributing to women’s overall satisfaction with the childbirth experience, and the role of pain
and pain relief in this satisfaction. Dr Hodnett, an editor
of the Cochrane Collaboration’s Pregnancy and Childbirth Group, is the author of a series of systematic reviews
and primary studies examining effects of different maternity care arrangements on women and infants. Dr Hodnett’s resulting review in this issue is an important
contribution in both the present pain-related context
and with respect to the more general question of satisfaction in childbirth.28
Dr Hodnett reviews an extensive body of research. By
using established criteria for inclusion, her review is
based on 6 systematic reviews, 27 RCTs, and 29 observational studies of childbirth. Collectively, these studies describe the experience of more than 45,000 women.
Dr Hodnett concludes that 4 factors are remarkably
consistent in their association with childbirth satisfaction:
May 2002
Am J Obstet Gynecol
1. The amount of support a woman receives from caregivers
2. The quality of her relationship with her caregivers
(eg, good communication, rapport, and information; comfort expressing feelings)
3. Her involvement with decision-making
4. Her personal expectations (higher satisfaction is associated with both a childbirth experience that exceeds expectations and having high expectations;
lower satisfaction is associated with having and realizing low expectations).
The following factors appear to be considerably less important: age, socioeconomic status, ethnicity, childbirth
preparation, the physical birth environment, mobility,
medical interventions, continuity of care, and pain.
With respect to the role of pain and pain relief, Dr
Hodnett writes:
Caregivers frequently assume that optimum
pain relief during labor and birth is very important to most laboring women, and that those who
say they wish to avoid pharmacologic pain relief
measures are either martyrs or misinformed. However, the results concerning the impact of pain
and pain relief on childbirth satisfaction were consistent across a wide variety of circumstances —
when epidural analgesia was common or rare,
across a wide variety of study designs and methods,
in a variety of countries, over almost 30 years. Pain
and pain relief do not generally play major roles
in satisfaction with the childbirth experience, unless expectations regarding either are unmet.
Although pain and pain relief appear to have a rather
small effect on women’s overall satisfaction with childbirth, much care during labor and birth is related to pain.
Offering pain-related care consistent with the 4 preeminent factors noted previously (eg, providing good information and involving women in decision-making about
pain relief) may be highly related to satisfaction.
Dr Hodnett’s review has major implications for health
care managers and policymakers. “Females with delivery” is
the most common discharge on the US National Hospital
Discharge Survey, involving nearly 4 million hospitalizations in the United States every year.44 To attract women,
some hospitals have offered incentives (eg, champagne dinners) and have focused on cosmetic physical modifications
(eg, wallpaper). Facilities and caregivers can use Dr Hodnett’s research to ensure and publicize the availability of
services that are most appreciated by women.
This review also offers important information to
women who are selecting maternity caregivers and birth
settings. It is important that women understand that the
best available evidence suggests that their satisfaction
with this important life event will be enhanced by arranging for care that involves the 4 preeminent factors identified in the review. This consideration is especially
Volume 186, Number 5
Am J Obstet Gynecol
important in light of evidence suggesting that vivid and
largely accurate memories of the childbirth experience
endure through women’s lives.45, 46
Dr Hodnett’s review also addresses key methodologic
issues relating to childbirth satisfaction, including tools
for measuring it and methods used to develop and carry
out this research. It also provides guidance to help investigators evaluating methods of pain relief include a measure of satisfaction with the overall childbirth experience
into their research.
“Labor Pain Management in the United States: Understanding Patterns and the Issue of Choice.” Theodore R.
Marmor, a senior scholar of health policy, management,
and politics, was invited to examine questions concerning
related health system and policy issues. To what extent
are specific methods of labor pain relief available and
used in the United States? Why are some methods widely
available and used, whereas others are much less available
and used? How do caregivers, women, health care delivery systems, and costs affect these patterns? Dr Marmor
collaborated with co-author David M. Krol to prepare the
final paper in this issue.47 Systematic review methods
were not appropriate for these questions and the data
available to answer them.
The authors did not find any national data about the degree to which specific methods of labor pain relief are available in the United States that are comparable with detailed
surveys conducted in Canada and the United Kingdom.33, 34 A national survey to ascertain rates of the use of
selected pharmacologic methods in US hospitals was first
conducted in 1981 and repeated in 1992 and 1997. The survey examined parenteral, paracervical, spinal, epidural,
and combined spinal/epidural methods, but did not address nonpharmacologic measures and other pharmacologic measures. Marmor and Krol47 combine results from
the 3 surveys and comment on trends. During this period,
the use of epidural analgesia sharply increased. Hospitals
with more births per year have higher epidural rates than
hospitals with fewer births per year. Opioid use was also
high, with trends differing according to size of the service.
Rates of women who did not use pain medications declined
during this period.
The authors47 conclude that the range of choice available to women in the United States is restricted relative to
many other western industrial nations. They found no
data sources describing the preferences of women in the
United States for one or another method of labor pain relief and factors contributing to their preferences.
Marmor and Krol47 argue that the following factors
shape the availability and use of methods of labor pain relief in the United States: philosophy and professional
training of caregivers, practice settings, staffing constraints, economic rewards, and inclinations to avoid
pain. This discussion is largely based upon general knowledge of the US health care system and of factors that have
Symposium Steering Committee S13
repeatedly been shown to influence the provision of care
more generally and in other clinical areas, because few
empiric studies of care relating to labor pain are available
to address these matters.
A health policy and politics perspective suggests that
these inferences are highly likely to apply to the field at
hand. Nonetheless, rigorous, specific studies are needed
to answer these important questions with greater clarity
and confidence, and to address specific questions of access and choice relating to labor pain, childbearing
women, and maternity care in the United States.
Implications for practice
The Nature and Management of Labor Pain project has
used an open and systematic process to identify and assess
the best available research to answer leading questions
facing professionals and childbearing women. Until they
are superceded by new and updated reviews of equivalent
or greater rigor, these are arguably the best available answers to the specific questions addressed.
The Steering Committee believes that the major conclusions from these papers should inform pain-related decision making by clinicians, educators, administrators, and
policymakers in the fields with responsibility for childbearing women and their newborns. Colleagues in various fields
are actively involved in presenting results and examining
implications for their specialty or discipline. We encourage
action/implementation projects on the basis of this material. The Maternity Center Association has an ongoing program commitment to this area and welcomes the
opportunity to support or collaborate with such efforts.
These papers should also be used to develop clear and
full information for women about labor pain and about
the advantages and disadvantages of a variety of methods
for labor pain relief. This material should be available
during pregnancy to enable women to become familiar
with issues and options, to have their questions answered,
and to seek care arrangements according to their needs
and preferences. They should have access to this information again during labor as a part of an open and respectful informed consent process oriented toward
women rather than toward professional liability concerns. The Maternity Center Association has a section on
its Maternity Wise website48 that relies upon results of The
Nature and Management of Labor Pain project to help
women make informed decisions about labor pain relief.
Implications for research
As is clear from the accompanying papers, this identification and summary of the best available research clarifies that many critical gaps in knowledge remain. Space
limitations preclude an in-depth discussion here, and
readers are referred to the papers themselves. The committee would, however, like to identify one important research gap that is evident across all of the methods
S14 Symposium Steering Committee
papers. By and large, the comparison groups in available
studies of the effects of pain relief methods reflect needs
and interests of physicians and other caregivers rather
than those of laboring women. Many studies compare
one or another dosage, mode of administration, or agent
within the same class or drugs. With the exception of a series of trials that compare epidural analgesia with opioids,
few comparisons are available to help inform women
about the choices and options that they face. It would be
very helpful for women to understand the advantages and
disadvantages of epidural or opioid analgesia relative to
such options as nitrous oxide, labor support, and other
nonpharmacologic methods.
We also note that some national survey data relating to
labor pain and its relief that have not previously been available will become available later in 2002. In collaboration
with the Johnson & Johnson Pediatric Institute, the Maternity Center Association is conducting Listening to Mothers,
the first national US survey of the experiences of childbearing women and their assessments of these experiences. The
survey will include a number of pain-related issues, such as
population-based information about the use of various
pharmacologic and nonpharmacologic methods, women’s
knowledge and attitudes about various methods, and care
and outcomes associated with various methods. Information about survey reports will be available online.49
Finally, the Steering Commitee recommends the process
we have undertaken as a model for rigorously examining
other topics within maternity care and other areas of health
and medicine. We believe that the field and recipients of
care benefit from multidisciplinary perspectives, the evidence-based approach, a concerted and comprehensive examination of an entire field, and a process that engages all
stakeholders over time and includes repeated opportunities for feedback, dialogue, and refinement.
REFERENCES
1. Comparing the old world (before evidence based practice) and
the new world (with evidence based practice). BMJ [serial
online] 2000 Jan 15 320(7228):[1 screen]. Available at:
http://bmj.com/cgi/content/full/320/7228/0/DC1. Accessed
on: December 5, 2001.
2. Muir Gray JA. Evidence-based healthcare: how to make health
policy and management decisions, 2nd ed. Edinburgh (Scotland): Churchill Livingstone; 2001.
3. Sackett DL, Straus SE, Richardson WS, Rosenberg W, Haynes
RB. Evidence-based medicine: how to practice and teach EBM,
2nd ed. Edinburgh (Scotland): Churchill Livingstone; 2000.
4. Egger M, Davey Smith G, Altman DG, editors. Systematic reviews
in health care: meta-analysis in context, 2nd ed. London (UK):
BMJ Publishing Group; 200l.
5. Sutton AJ, Abrams KR, Jones, DR, Sheldon TA, Song F. Systematic reviews of trials and other studies. Health Technol Assess
1998;2:1-276. Available at: http://www.hta.nhsweb.nhs.uk/.
6. The Cochrane methodology register. In: The Cochrane Library,
issue 4. Oxford: Update Software; 2001.
7. Values, principles, and objectives of The Nature and Management of Labor Pain: an evidence-based symposium. Available at:
http://www.maternitywise.org/prof/pain.
8. Clarke M, Oxman AD, editors. Cochrane reviewers’ handbook
4.1.4 [updated October 2001]. In: The Cochrane Library,
May 2002
Am J Obstet Gynecol
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
issue 4. Oxford: Update Software; 2001. Available at:
http://www.cochrane.org/cochrane/hbook.htm.
NHS Centre for Reviews and Dissemination, University
of York. Undertaking systematic reviews of research on effectiveness: CRD’s guidance for those carrying out or
commissioning reviews, 2nd ed. York: CRD; 2001. Available at:
http://www.york.ac.uk/inst/crd/report4.htm.
Stroup DF, Berlin JA, Morton SC, Olkin I, Williamson GD,
Rennie D, et al, for the Meta-analysis of Observational Studies
in Epidemiology (MOOSE) Group. Meta-analysis of observational studies in epidemiology: a proposal for reporting. JAMA
2000;283:2008-12. Available at: http://jama.ama-assn.org/
issues/v283n15/toc.html.
Moher D, Cook DJ, Eastwood S, Olkin I, Rennie D, Stroup DF,
for the QUOROM Group. Improving the quality of reports
of meta-analyses of randomised controlled trials: the
QUOROM statement. Lancet 1999;354:1896-1900. Available at:
http://www.thelancet.com/journal/vol354/iss9193/contents.
Sutton AJ, Jones DR, Abrams KR, Sheldon TA, Song F. Systematic reviews and meta-analysis: a structured review of the
methodological literature. J Health Serv Res Policy 1999;4:49-55.
Cook DJ, Sackett DL, Spitzer WO. Methodologic guidelines for
systematic reviews of randomized control trials in health care
from the Potsdam Consultation on Meta-Analysis. J Clin Epidemiol 1995;48:167-71.
Guidance for preparing a systematic review paper for “The Nature
and Management of Labor Pain: An Evidence-based Symposium.”
Available at: http://www.maternitywise.org/prof/pain.
Available at: http://www.maternitywise.org/prof/pain.
Chalmers, I, editor. The Oxford database of perinatal trials. Oxford: Oxford University Press; 1988-92.
Chalmers, I, Enkin M, Keirse MJNC, editors. Effective care in
pregnancy and childbirth, 2 vols. Oxford: Oxford University
Press; 1989.
Enkin M, Keirse MJNC, Chalmers I. A guide to effective care in
pregnancy and childbirth, 1st ed. Oxford: Oxford University
Press; 1989.
Chalmers I, Enkin M, Keirse MJNC. Preparing and updating systematic reviews of randomized controlled trials of health care.
Milbank Quarterly 1993;71:411-37.
Lowe NK. The nature of labor pain. Am J Obstet Gynecol
2002;186:S16-24.
Caton D. What a blessing she had chloroform: the medical and
social response to the pain of childbirth from 1800 to the present. New Haven: Yale University Press; 1999.
Caton D, Frölich MA, Euliano TY. Anesthesia for childbirth: controversy and change. Am J Obstet Gynecol 2002;186:S25-30.
Lieberman E, O’Donoghue C. Unintended effects of epidural
analgesia during labor: a systematic review. Am J Obstet Gynecol
2002;186:S31-68.
Halpern SH, Leighton BL, Ohlsson A, Barrett JFR, Rice A. Effect
of epidural vs parenteral opioid analgesia on the progress of
labor: a meta-analysis. JAMA 1998;280:2105-10.
Leighton BL, Halpern SH. The effects of epidural analgesia on
labor, maternal, and neonatal outcomes: a systematic review. Am
J Obstet Gynecol 2002;186:S69-77.
Jadad AR, Moore RA, Carroll D, Jenkinson C, Reynolds DJ, Gavaghan DJ, et al. Assessing the quality of reports of randomized
clinical trials: is blinding necessary? Control Clin Trials
1996;17:1-12.
Thacker SB, Stroup DF. Methods and interpretation in systematic reviews: commentary on two parallel reviews of epidural
analgesia during labor. Am J Obstet Gynecol 2002;186:S79-80.
Hodnett ED. Pain and women’s satisfaction with the experience of childbirth: a systematic review. Am J Obstet Gynecol
2002;186:S160-72.
Mayberry LJ, Clemmens D, De A. Epidural analgesia side effects,
co-interventions and care of women during childbirth: a systematic review. Am J Obstet Gynecol 2002;186:S81-93.
Enkin M, Keirse MJNC, Neilson J, Crowther C, Duley L, Hodnett E,
et al. A guide to effective care in pregnancy and childbirth, 3rd
ed. Oxford: Oxford University Press; 2000.
Hawkins JL, Beaty BR, Gibbs CP. Update on obstetric anesthesia
practices in the U.S. [abstract]. Anesthesiology 1999;91:A1060.
Volume 186, Number 5
Am J Obstet Gynecol
32. Bricker L, Lavender T. Parenteral opioids for labor pain relief: a
systematic review. Am J Obstet Gynecol 2002;186:S94-109.
33. Levitt C, Harvey L, Avard D, Chance G, Kaczorowski J. Survey of
routine maternity care and practices in Canadian hospitals. Ottawa: Health Canada and Canadian Institute of ChildHealth; 1995.
34. Chamberlain G, Wraight A, Steer P, editors. Pain and its relief in
childbirth: the results of a national survey conducted by the National Birthday Trust. Edinburgh (Scotland): Churchill Livingstone; 1993.
35. Rosen MA. Nitrous oxide for relief of labor pain: a systematic review. Am J Obstet Gynecol 2002;186:S110-26.
36. Rosen MA. Paracervical block for labor analgesia: a brief historical review. Am J Obstet Gynecol 2002;186:S127-30.
37. Simkin P. The birth partner: everything you need to know to
help a woman through childbirth, 2nd ed. Boston (MA): Harvard Common Press; 2001.
38. Simkin P, Ancheta R. The labor progress handbook: early interventions to prevent and treat dystocia. Oxford: Blackwell Science; 1999.
39. Simkin P. Simkin’s ratings of comfort measures for childbirth.
Waco (TX): Childbirth Graphics; 1997.
40. Simkin P. Psychologic and non-pharmacologic techniques for
labor pain. In: Bonica JJ, McDonald JS, editors. Principles and
practice of obstetric analgesia and anesthesia, 2nd ed. Baltimore
(MD): Williams and Wilkins; 1995. p. 715-50.
Symposium Steering Committee S15
41. Simkin P. Non-pharmacological methods of pain relief during
labour. In: Chalmers, I, Enkin M, Keirse MJNC, editors. Effective
care in pregnancy and childbirth, vol 2. Oxford: Oxford University Press; 1989. p. 893-912.
42. Simkin P, O’Hara MA. Nonpharmacologic relief of pain during
labor: systematic reviews of five methods. Am J Obstet Gynecol
2002;186:S131-59.
43. Hodnett ED. Caregiver support for women during childbirth
(Cochrane review). In: The Cochrane Library, issue 4. Oxford: Update Software; 2001.
44. Popovic JR, Hall MJ. 1999 National Hospital Discharge Survey.
Advance data from vital and health statistics, No. 319. Hyattsville
(MD): National Center for Health Statistics; 2001.
45. Simkin P. Just another day in a woman’s life? Women’s long-term
perceptions of their first birth experience, part I. Birth
1991;18:203-10.
46. Simkin P. Just another day in a woman’s life? Women’s long-term
perceptions of their first birth experience, part II. Birth
1992;19:64-81.
47. Marmor TR, Krol DM. Labor pain management in the United
States: understanding patterns and the issue of choice. Am J Obstet Gynecol 2002;186:S173-80.
48. Maternity Center Association. How will I cope with labor pain?
Available at: http://www.maternitywise.org/mw/topics/pain/.
49. Maternity Center Association. Listening to Mothers. Available at:
http://www.maternitywise.org/.