safe motherhood - cmqcc

Safe Motherhood

Triennial Report | 2003-2005

[ 1 ]

Contents

Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

Abbreviations Used in this Report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

Composite Case Scenarios . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

CHAPTER 1: INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

Background of the Safe Motherhood Initiative . . . . . . . . . . . . . . . . . . . . . . . . . . 10

Th e Goals and Objectives of the Safe Motherhood Initiative . . . . . . . . . . . . . . . . 12

Achievements of the Safe Motherhood Initiative . . . . . . . . . . . . . . . . . . . . . . . . . 13

CHAPTER 2: THE SAFE MOTHERHOOD INITIATIVE: A MATERNAL

MORTALITY SURVEILLANCE SYSTEM . . . . . . . . . . . . . . 15

Identifi cation of Pregnancy-Related Deaths . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

Maternal Death Abstraction Form and Data Sources . . . . . . . . . . . . . . . . . . . . . 16

Multidisciplinary Review Team . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

On-site Review Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

CHAPTER 3: OUTREACH AND AWARENESS . . . . . . . . . . . . . . . . . . . . . 19

Th e Grand Rounds Curriculum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

Newsletter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

Website . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

CHAPTER 4: A LOOK AT THE FINDINGS . . . . . . . . . . . . . . . . . . . . . . . . 21

Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

Obstetric Care Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

System Recomendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

CHAPTER 5: DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

CHAPTER 6: LIMITATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

CHAPTER 7: FUTURE DIRECTIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

[ 3 ]

Safe Motherhood: Triennial Report

Foreword

Th e Safe Motherhood Initiative is a joint project of the American College of Obstetricians

and Gynecologists District II/New York in collaboration with the Bureau of Women’s

Health of the New York State Department of Health. Established in 2001, the mission of

the Initiative is to help prevent pregnancy-related deaths through improved understanding

of the causes and risk factors for maternal mortality. Utilizing the maternal death protocol

and accompanying abstraction form developed in year one, the Safe Motherhood Initiative

and the state’s Regional Perinatal Centers conduct quality assurance and quality improve-

ment activities related to maternal mortality. Th e Initiative was inspired by the signifi cant

racial disparities associated with maternal mortality. Th e key strategies for prevention of

maternal deaths in New York State include the development of a standardized system to

report and review pregnancy-related deaths along with the provision of recommendations

and training that have the direct goal of improving maternity care. Th e triennial report was

written and compiled by Neisha M. Torres, RN, MS, Donna Montalto Williams, MPP and

Jeff rey C. King, MD, FACOG.

[ 5 ]


AcknowledgmentsACOG would like to express deep appreciation to New York State Health Commissioner,

Antonia C. Novello, MD, MPH, DrPH for her recognition of maternal mortality as a

major public health concern in New York State. Without her vision, continued support

and commitment, this endeavor would not have been possible. In addition, the guidance

and generous support received from the Bureau of Women’s Health, New York State De-

partment of Health, specifi cally Barbara Brustman, EdD, Director; Mary Applegate, MD,

MPH, Medical Director; Wendy Shaw, RN, MS, Assistant Director and Linda Th ornton,

RN, MS, Director of the Perinatal Health Unit, were critical to the continued success of

the Initiative.

Special thanks go to the Healthcare Association of New York State and the Health and

Hospitals Corporation of New York City for their support and eff orts to promote aware-

ness and encourage project participation. Special thanks to the planning committee who

shared their vision and knowledge with ACOG. Th e planning committee’s wisdom and

expertise were vital to the success of the Safe Motherhood Initiative (SMI) and provided

the guidance and support needed to implement the project. Furthermore, ACOG is grate-

ful for the opportunity to work with Jeff rey C. King, MD, FACOG, who chaired the Safe

Motherhood Initiative. His vision, time and dedication have been invaluable to the devel-

opment and implementation of this Initiative.

Heartfelt gratitude is extended to those individuals and faculty who volunteered their time

to develop and implement the SMI and conduct maternal mortality reviews. Th rough grand

rounds presentations to maternal health care professionals, the faculty carried the SMI

message throughout New York State. Without their diligent work, expertise, and ongoing

commitment to keep mothers safe, the SMI and this report would not have been possible:

Finally, we convey particular recognition to John W. Choate, MD, FACOG, who helped

develop and launch this Initiative. As an advocate for safe motherhood, Dr. Choate dedi-

cated his life to improving perinatal care in the state. Dr. Choate passed away December

22, 2003. ACOG and the SMI will continue to remember Dr. Choate’s life through the

promotion of safe motherhood as a social investment that advances New York State’s public

health.

[ 6 ]

Abbreviations Used in this Report

ACOG American College of Obstetricians and Gynecologists, District II/NY

APCU Adequacy of Prenatal Care Utilization

BMI Body Mass Index

CDC Centers for Disease Control and Prevention

CEMD Confi dential Enquiries into Maternal Deaths

CME Continuing Medical Education

CT Computed Axial Tomography Scan

CVA Cerebrovascular Accident

DOH Department of Health

EMS/ER Emergency Medical Services/Emergency Room

HHC Health and Hospitals Corporation of New York City

ICU Intensive Care Unit

MDNF Maternal Death Notifi cation Form

MFM Maternal-Fetal Medicine

NCHS National Center for Health Statistics

NYC New York City

NYPORTS New York Patient Occurrence Reporting and Tracking System

NYS New York State

PAMR Pregnancy Associated Mortality Review

PHL Public Health Law

PIH Pregnancy Induced Hypertension

PPH Postpartum Hemorrhage

RCOG Royal College of Obstetricians and Gynaecologists

RPC Regional Perinatal Center

SMI Safe Motherhood Initiative

SPARCS Statewide Planning and Research Cooperative System

TOP Termination of Pregnancy

VRQC Voluntary Review of Quality of Care Program

VTE Venous Th romboembolism

p

[ 7 ]


Composite Case ScenariosTh ese vignettes are drawn from de-identifi ed cases to provide context for the reader.

An Italian-speaking primigravida presented at a hospital at 37 weeks gestation with a com-

plaint of shortness of breath. Her private obstetrician arrived and acted as a translator. Her

oxygen saturation on room air was 70% and she was noted to have slight edema and 3+

proteinuria. Cardiology was called and an echocardiogram was performed at her bedside,

revealing an enlarged right ventricle. She was transferred to the critical care unit where

she became fully dilated and delivered a live infant via vacuum extraction. Following the

delivery an angiogram was performed and it was ascertained that the patient suff ered from

a congenital heart defect. Th e patient was told of the seriousness of her condition and

that she would likely need a heart-lung transplant. She never left the critical care unit and

ultimately expired eight days after admission. It was later discovered that the patient had

been counseled by her health care team in her country of origin not to become pregnant

because of her heart defect. It was the impression of the review team that the patient and

family actively hid this information from the obstetrician who had followed her for the

index pregnancy.

An obese patient with a body mass index of >40 presented at a hospital in labor. She was 38

weeks gestation with mild contractions every 3-5 minutes. A repeat cesarean was carried

out without obstetrical or anesthetic complications. Th e subcutaneous tissue was irrigated,

the skin closed and the patient began a 24 hour course of Clindamycin IV. On post-opera-

tive day 3, a physician note indicated a foul smell at the site of incision. Th e wound was

cleaned with peroxide and re-dressed. Th e patient was taught how to clean the wound, with

specifi c instruction on lifting of excess abdominal fl aps of skin. Th e next day a nursing note

documents a foul smell again coming from the inguinal area. Th e patient was counseled

again on cleaning the wound and was discharged. Four days later the patient reported to

the clinic to have her staples removed. Th e wound was cleaned and re-cleaned with per-

oxide. Th e patient had no complaints. Two days later the patient was found dead at her

residence. Th e fi nal autopsy reported “sepsis related to wound infection and an enlarged

heart.”

A patient in her mid twenties, at 34 weeks estimated gestational age, arrived via ambulance

at a hospital with no obstetric service complaining of edema and a headache lasting for two

days. Th e patient eventually signed herself out against medical advice and contacted her

private obstetrician who urged her to immediately return to an emergency room that had

an obstetric service. She arrived at a second hospital and her blood pressure was found to

be 204/108. She began seizure activity and was treated with Magnesium Sulfate. A central

intravenous line was placed by anesthesia and the patient was cleared to undergo a cesarean

section. After delivery of a live baby, the patient was moved to the ICU where she remained

intubated and was treated for eclampsia, HELLP and renal failure. Sudden mental changes

were noted and a stat head CT showed an intracranial hemorrhage. Brain death criteria

were met and documented.

[ 9 ]


Chapter 1: Introduction

Maternal mortality is a tragedy that has prolonged eff ects on the mother’s partner, her

children and family, and the health care team.1 Th e Centers for Disease Control and Pre-

vention (CDC) reports that, over the past 20 years, an average of two to three (2-3) women

died of pregnancy-related complications each day in the United States.2 From 1900 to

1982, the rate† of pregnancy-related deaths declined dramatically. Despite this major public

health achievement, in 1982 the maternal mortality ratio reached a plateau, with no sub-

stantial change since then.3 (Figure 1.1) Studies indicate that as many as half of all preg-

nancy-related deaths could be prevented, if women had better access to health care, received

better quality care, and made changes in their health and lifestyle habits.2

Data from the CDC found that African-American

women are three times more likely to die from preg-

nancy-related complications than White, non-Hispanic

women.4 CDC’s decade-long study, conducted from

1991-1999, found that the overall pregnancy-related

mortality ratio was 11.8 deaths per 100,000 live births.

For African-American women the rate was 30.0 deaths

per 100,000 live births compared with 8.1 deaths for

White women. Th is gap is the largest and most diffi cult

disparity to understand in the area of maternal-child

health. Moreover, studies have shown that physicians

completing death certifi cates after a maternal death fail

to report that the woman was pregnant or had a recent

pregnancy in fi fty percent or more of the cases.5, 6, 7

†Th e Centers for Disease Control and Prevention, National Center for Health Statistics uses the term

“rate” when reporting this indicator of maternal mortality. Th e term “ratio” is used instead of rate when the

numerator includes some maternal deaths that are not related to live-born infants and thus not included in the

denominator.

Figure 1.12

*The national Healthy People 2010 objective is to reduce material deaths to 3.3 per 100,000 live births.

Maternal Mortality 1900-2002

[ 10 ]

Maternal mortality is a rare sentinel event, reported as a rate per 100,000 live births. In

2002 the National Center for Health Statistics reported a maternal mortality rate of 8.9

deaths per 100,000 live births. 8 According to the New York State (NYS) Community

Health Data Set, the maternal mortality rate for 2002 was 12.8 per 100,000 live births

or 32 maternal deaths. 9 Moreover, in 2002, there was a wide divergence in the maternal

mortality rate from upstate New York, 3.8 per 100,000 live births, when compared to New

York City, 22.9 per 100,000 live births. (Figure 1.2)

Th e maternal mortality rate in NYS exceeds the national rate and far exceeds the Healthy

People 2010 goal of reducing the maternal mortality rate to no more than 3.3 maternal

deaths per 100,000 live births.10 Th e higher rate in NYS is due in part to better case ascer-

tainment, particularly in NYC. Th e magnitude of the diff erence attributable to better case

ascertainment is unknown. Prevention of mortality attributable to pregnancy is a public

health priority for New York State.

Background to the Safe Motherhood Initiative

New knowledge of the causes of maternal mortality together with a renewed focus on

technology and improvements in hospital care has led to the emergence of a regional peri-

natal care structure. To ensure that mothers and newborns receive the care they need in

a timely, safe and eff ective manner, the statewide organization of maternal and newborn

health services has come to be known as regionalized perinatal care. In New York State

all hospital-based perinatal care services are required to participate in the regional system.

Each perinatal service is designated by the State Department of Health (DOH) as provid-

0

5

10

15

20

25

Upstate New York New York City New York State United States

Mat

ern

al D

eath

s pe

r 10

0,00

0 li

ve

birt

hs

Healthy People2010 Goal

Introduction

Figure 1.2 New York and United States: Maternal Mortal i ty Rate,† 8 2002 §

†Per 100,000 Live Births

§Dotted line represents the Healthy People 2010 goal of reducing maternal deaths to 3.3 per 100,000 live births.

[ 11 ]


ing Level I, Level II, or Level III perinatal care or the hospital is designated as a Regional

Perinatal Center (RPC). Since the early 2000’s, the new regional perinatal structure was

intended to result in greater access to more appropriate levels of care for maternity patients

and newborn infants.11 (Figure 1.3)

According to the NYS DOH Bureau of

Women’s Health, in 2005 there were 150 hos-

pitals in New York State with perinatal des-

ignations as follows: 19 hospitals constitute

17 regional perinatal centers, 71 Level I, 26

Level II, and 34 Level III hospitals.

Parallel to the regionalization of perinatal care

in 2002 the NYS DOH in conjunction with

the CDC, the Medical Society of the State of

New York and the American College of Ob-

stetricians and Gynecologists District II/NY

(ACOG) concluded a three-year retrospec-

tive maternal mortality study. Recommen-

dations from that study included the need to

strengthen regionalization of maternity care; reduce the number of high-risk pregnancies;

maximize ascertainment of maternal deaths; continue in-depth case reviews of incidents

of maternal death; educate community hospital staff on the maternal death review process;

and educate emergency room staff on obstetrical emergencies. 12

In 2001 the Commissioner of Health and the obstetric community through ACOG recog-

nized the potential benefi t of a uniform surveillance system devoted specifi cally to review

of in-hospital pregnancy-related deaths. Deaths had previously been reviewed using a va-

riety of methods including the state’s hospital-based review of sentinel events; hospital peer

review; or root-cause analysis by both the hospital of occurrence and the hospital’s RPC.

A new initiative was needed to implement prior recommendations while responding to

publications of the underreporting of maternal deaths; maternal mortality disparity among

diff erent racial and ethnic groups; and a required eff ort to reduce and prevent pregnancy-

related deaths. In 2002, the NYS Department of Health entrusted ACOG with the devel-

opment of a maternal mortality review study. Utilizing an expert committee, representing

local and state maternity care professionals, ACOG developed a statewide maternal mor-

tality surveillance system, entitled the Safe Motherhood Initiative (SMI).

†Jacobi Medical Center/Bellevue Hospital Center (joint RPC), Maimonides Medical Center, Montefi ore Medical

Center, Mount Sinai Hospital/NYU Medical Center (joint RPC), NY Presbyterian-Columbia, NY Presbyterian-

Cornell, Saint Vincent’s Catholic Medical Center of Staten Island, University Hospital of Brooklyn

†F igure 1.3

REGIONAL PERINATAL CENTERS/NETWORKS(SPOKES DRAWN TO AFFILIATE HOSPITALS)

CHILDREN’S (BUFFALO)STRONG MEMORIAL (ROCHESTER)CROUSE (SYRACUSE)ALBANY MEDICAL CENTERWESTCHESTER MEDICAL CENTER (VALHALLA)UNIVERSITY (STONYBROOK)LONG ISLAND JEWISH, WINTHROP UNIVERSITY, NORTH SHORE UNIVERSITY (NASSAU COUNTY)

NEW YORK CITY NETWORKS NOT SHOWN

[ 12 ]

Th e initiative defi ned “pregnancy–related death” in accordance with the national offi ces of

ACOG and the CDC as:

death of a woman, while pregnant or within one year of termination of pregnancy, irrespec-

tive of the duration and site of the pregnancy, from any cause related to or aggravated by her

pregnancy or its management, but not from accidental or incidental causes.13

Th e Goals and Objectives of the Safe Motherhood Initiative

Th e goals of the SMI were to reduce and prevent pregnan-

cy-related deaths in New York State. Th e methods selected

for accomplishing these goals included utilizing the state’s

regional perinatal centers to conduct quality assurance and

quality improvement activities related to maternal mortal-

ity. Th e SMI surveillance system was developed, refi ned

and tested during year one of the project. Th e objectives

for implementation in year two and three of the SMI in-

cluded:

1. Increase awareness of maternal mortality in New York State and educate health care

professionals about the SMI review protocol utilizing periodic bulletins, grand rounds

and electronic and print communication formats;

2. Perform voluntary hospital-based maternal death reviews to identify the factors con-

tributing to maternal mortality, off er quality assurance recommendations and provide

insight into disparities among populations of maternity patients;

3. Design and provide education to medical professionals to help prevent and/or treat the

contributing factors of maternal death discovered during the SMI review process.

Introduction

Maternal Mortality Facts2

Approximately 6 million American women

become pregnant each year, and more than

10,000 give birth each day.

Each day in the United States, between 2 and

3 women die of pregnancy-related causes.

Th e risk of pregnancy-related death has not

decreased signifi cantly since 1982.

Th e risk of death due to pregnancy varies

greatly in diff erent racial and ethnic groups.

[ 13 ]


Achievements of the Safe Motherhood Initiative

Th e SMI provides direction and a framework for regional perinatal centers to review

pregnancy-related deaths and provides quality improvement activities using a standardized

review protocol. During the initial three years, the SMI completed the following

activities:14

Year One Achievements• Convened the statewide SMI committee of experts including physicians, specialists,

midwives, coroners, and public health offi cials, all dedicated to safeguarding the health

of mothers.

• Created a nationally-recognized surveillance system for systematic hospital-based re-

view of pregnancy-related deaths, including a full review protocol, notifi cation and

abstraction forms and an accompanying instruction manual.

• Conducted statewide outreach and education on the SMI, including three confer-

ences and a satellite broadcast with continuing medical education (CME) credit for

participants.

• Obtained confi dentiality protection under section 206(1)(j) of the Public Health Law

(PHL) for review team participants from the Commissioner of Health.

• Maintained consensus among regional perinatal centers (RPC) and affi liate hospitals

on the need for ongoing local and regional hospital cooperation and collaboration

regarding the SMI.

Year Two Achievements• Implemented the voluntary, on-site, objective review of maternal deaths at hospitals

statewide.

• Th e SMI reviews brought together multidisciplinary teams of experts to review ma-

ternal deaths, enabling a more thorough review.

• Hosted a day-long forum to discuss specifi c maternal complications related to mor-

tality, including presentations on hypertensive disorders, associated cardiomyopathies,

and postpartum emboli. Th e need for increased aware-

ness and outreach regarding maternal mortality report-

ing was highlighted.

• Implemented an outreach and awareness campaign that

promoted the SMI as a model suitable for replication

by other states. Outreach included development and

delivery of a grand rounds curriculum and distribution

of SMI information to RPCs, chiefs of obstetric ser-

vice, risk management and quality-assurance offi cers.

• Facilitated RPC-affi liate hospital working relation-

ships, particularly in networks where prior relationships

were minimally developed.

• Almost 600,000 women die each year.

• 1,600 women die each day.

• One woman dies every minute.

World Health Organization

World death toll during

pregnancy and childbirth

[ 14 ]

‡An Institutional Review Board (IRB) is a committee whose purpose is to ensure that the rights and welfare

of human subjects are protected in all medical, behavioral and social science research. An IRB must review and

approve research involving human subjects in accordance with federal and state regulations. Th e IRB is respon-

sible for determining whether the research exposes subjects to unreasonable or unnecessary risk, and reviews

consent forms and processes to monitor progress of research.

Year Th ree Achievements• Identifi ed postpartum hemorrhage (PPH), embolism, sepsis and chronic disease as

some of the contributing factors associated with maternal death for the cases reviewed

by the SMI.

• Collaborated with New York City Department of Health and Mental Hygiene

(NYCDOHMH) and the NYSDOH to issue a postpartum hemorrhage alert to all

hospitals.

• Convened a subcommittee to collect, review and distribute sample PPH protocols to

all obstetrician-gynecologist chairs and nurse managers at obstetric hospitals state-

wide.

• Collaborated on the delivery of educational presentations on PPH, sepsis, pulmonary

embolism and preconception counseling.

• Addressed prophylactic measures, clinical management techniques and implementa-

tion of obstetric protocols and codes as areas for improvement in the NYS obstetric

community.

• Met with Health and Hospitals Corporation of New York City (HHC) to review the

SMI protocol, confi dentiality protections and the need to collect comprehensive data

from all hospitals statewide.

• Developed a comprehensive, aggregate maternal mortality review database.

• Continued outreach to all hospitals to increase reporting of deaths to the SMI and

updated the on-site hospital review process.

• Received annual New York State Department of Health Institutional Review Board

(IRB)‡ approval for the Safe Motherhood Initiative’s protocol and consent form.

Introduction

[ 15 ]


Chapter 2: Th e Safe Motherhood Initiative: a maternal mortalitysurveillance system

During this era of enhanced privacy protection of health information and concerns for

liability exposure, an in-depth maternal mortality review process could raise legal questions

about the sharing of detailed medical

information. Th erefore, to allay such fears

and facilitate participation in maternal

mortality reviews, the Department of

Health employed a section of the NYS

Public Health Law that authorizes the

Commissioner to designate representatives

to conduct confi dential medical studies.

Th e fi ndings of this research and the

information received by the designees must be kept confi dential and cannot be used as

evidence in any legal action.

NYS Public Health Law §206(1)(j) - Th e commissioner shall cause to be made such scientifi c studies

and research which have for their purpose the reduction of morbidity and mortality and the im-

provement of the quality of medical care through the conduct of medical audits within the state.

In conducting such studies and research, the commissioner is authorized to receive reports on forms

prepared by him and the furnishing of such information to the commissioner, or his authorized

representatives, shall not subject any person, hospital, sanitarium, rest home, nursing home, or other per-

son or agency furnishing such information to any action for damages or other relief. Such information

when received by the commissioner, or his authorized representatives, shall be kept confi dential and shall

be used solely for the purposes of medical or scientifi c research or the improvement of the quality

of medical care through the conduction of medical audits. Such information shall not be admissible as

evidence in any action of any kind in any court or before any other tribunal, board, agency, or person.

“Any pregnancy-related death is one too many”Secretary of the Department of Health and Human Services,

Tommy G. Th ompson

February 20, 2003 press release

[ 16 ]

In order to qualify for the confi dentiality protection and ensure the non-discoverability of

the information received, each individual maternal mortality review team member must be

approved by the Department of Health and offi cially designated by the Commissioner to

conduct such research. Moreover, each review team member is required to sign a confi -

dentiality agreement prior to participation in any on-site maternal mortality review. Th e

designation process is complex and can take more than three months.

Th e SMI’s hospital-based review process was modeled after national ACOG’s Voluntary

Review of Quality of Care Program (VRQC) and the Illinois maternal mortality review

process. 15 Th e review is structured to provide the utmost level of reliability and objectivity

possible using principles of eff ective peer review. Information is analyzed using both quan-

titative and qualitative approaches. Participation in the SMI is voluntary and available to

all New York State hospitals.

Identifi cation of Pregnancy-Related DeathsAlthough deaths attributable to pregnancy are rare, with an estimated average of 40 per

year in NYS, each death needs to be identifi ed and carefully reviewed. It is through the

maternal death notifi cation form (MDNF) that hospitals with obstetric services notify

ACOG, within three days of the occurrence. Th e MDNF was developed utilizing the New

York Patient Occurrence Reporting and Tracking System (NYPORTS) short form as a

model for reporting sentinel events.

Maternal Death Abstraction Form and Data SourcesTh e SMI maternal death abstraction form was fashioned through baseline assessments and

in-depth analysis of national and international maternal mortality surveillance systems and

projects. Special attention was given to surveillance systems from the New York City De-

partment of Health, Florida State Department of Health, United Kingdom’s Confi dential

Enquiries into Maternal Deaths (CEMD, now known as the Confi dential Enquiries into

Maternal and Child Health) and the former NYS DOH study of maternal mortality.

Th e de-identifi ed abstraction form captures information such as medical and social history,

demographic data, past pregnancies, prenatal-intrapartum-postpartum history, psychoso-

cial assessment, social services, coordination of care, staffi ng, death certifi cate and autopsy

report, if available. Th e 90 question abstraction form underwent extensive revisions and

the content and length refl ects multiple views from state, private, and local organizations

involved in its development. Th e purpose of the abstraction form is to enable reviewers to

conduct a medical review of all pregnancy-related deaths using a standardized form.

An instruction manual was developed to: (1) accompany the maternal death abstraction

form; (2) aid in implementation of the maternal death review process; and (3) train RPC

staff who will potentially adopt the SMI protocol as a method of reviewing sentinel events.

Th e Safe Motherhood Initiative: a maternal mortality surveillance system

[ 17 ]


Coding manuals from the National Center for Health Statistics (NCHS), NYS DOH and

ACOG’s antenatal medical record were used to create the instruction manual.

Multidisciplinary Review TeamFollowing the receipt of the MDNF, ACOG assembles a multidisciplinary team. Th e

review team is composed of a medical transcriptionist; two or more RPC representatives,

including a Maternal-Fetal Medicine specialist, a labor and delivery nurse or RPC nurse

coordinator, and sub-specialists as necessary (e.g., anesthesiologist, cardiologist, social

worker, pathologist, medical examiner, critical care specialist, etc.). Reliability of the review

is further enhanced by having an experienced ACOG team leader, a Fellow with expertise

in maternal mortality surveillance and peer review. Th is multidisciplinary approach allows

all aspects of patient care to be adequately and comprehensively reviewed.

When the maternal death occurs at an RPC, ACOG assigns a non-contiguous RPC team

to review the death. Th e non-contiguous RPC review team is composed of the same ex-

perts listed above.

On-site Review Process Th e teams, using the maternal death abstraction form, gather and evaluate patient risk fac-

tors in addition to clinical and systems issues that might have contributed to the maternal

death. Th is is not a punitive review of cases. Rather, the goal of the onsite review is to im-

prove care by identifying critical interventions and protocols that, if modifi ed, might have

impacted the outcome. Conversely, in gaining insight into the actual occurrence, the care

that was given to the patient can be shown to be unrelated to the outcome.

Within six to eight weeks of notifi cation of a maternal death, an on-site review of the preg-

nancy-related death is conducted at the hospital where care was provided using the SMI

standardized protocol and de-identifi ed chart abstraction form. Th e review team gathers

the minimum information necessary to accomplish a thorough study of each maternal

death.

Elements of the site visit include:

1. Entrance conference with hospital administrative staff ;

2. Discuss systems and policy/procedure changes resulting from in-house root-cause

analysis already implemented;

3. Review of hospital obstetric demographic data and policies as needed;

4. Review of medical record(s);

5. Review of hospital staffi ng logs, as needed;

6. Interview the chairman or director and nurse manager of obstetric services along with

health care providers who had contact with the patient including: physicians, MFM

specialists, critical care and/or emergency room physicians, labor and delivery nurses,

[ 18 ]

anesthesiologists, medical examiners, social workers, risk managers and midwives,

among others. Th ese interviews help defi ne the timeline of events surrounding the

maternal death and provide an understanding of the environment in which care was

delivered.

7. Exit conference with hospital administrative staff . Preliminary fi ndings and brief

recommendations may be off ered by the ACOG team leader prior to departure.

Following the review, the review team and ACOG staff draft a fi nal report. Th e fi nal hos-

pital report includes a summary of the occurrence, fi ndings, conclusions and recommenda-

tions.

Th e RPC and affi liate hospital may use the report as a foundation for implementing

changes to enhance quality of care within their respective departments of obstetrics and

gynecology.

Th e Safe Motherhood Initiative: a maternal mortality surveillance system

[ 19 ]


Chapter 3: Outreach and AwarenessSince the commencement of the initiative, ACOG has offered an array of informational activities to promote: (1) awareness and outreach regarding the incidence of maternal mortality; (2) knowledge of the SMI and its utility; and (3) increased case ascertainment of pregnancy-related complications and deaths in New

York State as reported to ACOG.

The Grand Rounds CurriculumDeveloped and presented by members of the SMI Planning Committee, educational activities provided information to hospitals statewide. The Grand Rounds curriculum contained case studies plus international, national, state, and regional maternal mortality statistics along with information on the SMI review protocol and research on racial differences regarding the incidence of maternal mortality. Educational resources, including program overview and the Maternal Death Notification Form, were distributed to program attendees.

The SMI grand rounds program was presented at 42 hospitals in the state including twelve (12) Regional Perinatal Centers. Furthermore, ACOG presented the program at sixteen (16) state and national organizations and specialty society meetings.

Similar to Grand Rounds, ACOG delivered SMI education at meetings of the Health and Hospitals Corporation, Inc. of New York City to promote participation through shared data on pregnancy-related deaths. New York City’s Health and Hospitals Corporation (HHC) includes 10 obstetric hospitals and a regional perinatal center (RPC) that is com-prised of two hospitals. Following extensive conversations regarding voluntary participa-tion and confidentiality protections, HHC provided the SMI with post-review data on maternal deaths that occurred during the 2004 calendar year.

In evaluating each educational program, ACOG recognized a lack of awareness among most obstetric providers regarding the value of participating in the SMI program. ACOG also learned of the need to allay hospital legal and risk management departments concerns

[ 20 ]

regarding confi dentiality. Hospitals were concerned that SMI information could be re-

leased to health department or other governmental offi cials and attorneys. Constant and

repeated information exchange was needed to allay these concerns.

NewsletterIn order to raise awareness regarding maternal mortality surveillance, a newsletter was de-

veloped for obstetric health care professionals and hospital quality assurance-performance

improvement managers. Launched in December 2003, the SMI newsletter provided in-

formation on project updates, maternal mortality research, quality assurance suggestions,

statistics and fi ndings. Nine (9) editions of the SMI newsletter have been distributed, with

mailings to more than 500 professionals since January 2005.

WebsiteACOG responded to the demands of obstetricians, maternity care professionals and con-

sumer advocates for information on the SMI with the development of a website (www.

acogny.org). Th is technology has played a role in increasing the speed and eff ectiveness

of SMI information exchange. Th e website has provided a successful means to share SMI

knowledge. Th e following information is available: project description, maternal death pro-

tocol, slide show, health alerts, SMI newsletters, and upcoming conferences/information.

Outreach and Awareness

[ 21 ]


Chapter 4: A Look at the Findings

MethodsDe-identifi ed data collected by ACOG included available information from the maternal

death notifi cation forms (MDNF) and on-site review information, including medical charts,

interviews, death certifi cates, autopsies and peer review committees. Between August 2003

and June 2005, maternal deaths were voluntarily reported and reviewed by ACOG in New

York State. Data were collected and collated using a system that maintained confi dentiality

while allowing for a comprehensive analysis of the pregnancy-related deaths.

Each death reported to ACOG was reviewed and classifi ed by the multidisciplinary review

team composed of clinically experienced obstetricians and gynecologists, labor and delivery

nurses, critical care specialists, and other physician specialists. Th e review team determined

the immediate and underlying cause of death, associated obstetric conditions, and the out-

come of pregnancy. A woman’s death was classifi ed as pregnancy-related if it occurred dur-

ing pregnancy or within one year of pregnancy and resulted from: (1) complications of the

pregnancy; (2) a chain of events that was initiated by the pregnancy; or (3) the aggravation

of an unrelated condition by the physiologic eff ects of the pregnancy or its management.

Race was defi ned as the race of the mother and was classifi ed as White, Black, or other.

Other races/ethnicities included Hispanic, Haitian, and Asian. Th e women’s ages at the

time of death were grouped into the following categories: 19 and younger, 20 to 24, 25 to

29, 30 to 34, 35 to 39, and 40 and over. Method of payment was categorized as uninsured/

self pay, Medicaid, or managed care/private health insurance. Medicaid managed care was

classifi ed as Medicaid.

Education information was based on the total years of education completed at the time

of death, and was broken into the following categories: less than 12 years of education, 12

years of education, or more than 12 years of education. Th e analysis by level of education

was restricted to those women age 20 years or older, the age at which most women would

be assumed to have graduated from high school. Marital status was classifi ed as unmarried

or married.

[ 22 ]

Th e analysis of prenatal care was based on the CDC’s publication of pregnancy-related

mortality surveillance, 1987-1990. 16 Levels of prenatal care were classifi ed using the Ade-

quacy of Prenatal Care Utilization (APCU) index developed by Kotelchuck17 and modifi ed

by the CDC. Pregnant women with “adequate plus” had other risk factors, which increased

the number of visits. Th e index can serve as an indicator that some medical condition re-

quired additional prenatal care. Th e index is defi ned as follows:

Mother’s risk and causes

of death were deter-

mined via consensus of

the SMI maternal mor-

tality review team. Af-

ter analysis of the avail-

able information during

the on-site review, the

SMI review team was

asked to categorize the

mother’s risk as: at some risk, high-risk or very high-risk. Only cases reviewed on-site by

the SMI (N=21) were included in this risk classifi cation analysis. Immediate cause of death

was categorized using the nine CDC categories in Table 4.4.

Classifi cation for chronic disease was based on the associated causes and obstetric compli-

cations as determined by physician members of the review team. Body mass index (BMI)

was calculated utilizing CDC’s BMI calculator.19 Obesity was determined using the BMI,

which accounts for both height and weight. Furthermore, obesity was identifi ed by a BMI

above 30; these women were classifi ed as having a chronic disease. Th e term chronic disease

used here is defi ned by the National Center of Health Statistics as conditions that are not

cured once acquired (such as heart disease and diabetes) and present three (3) months or

longer.20

Due to the small number of maternal deaths (N=37) reported to the Safe Motherhood

Initiative, all analyses were performed utilizing Microsoft Excel and cross referenced by at

least two individuals for accuracy.

Information regarding pregnancy-related deaths in this report is modeled after the CDC’s

February 21, 2003, MMWR Surveillance Summary4 concerning maternal mortality in the

United States and Florida’s Pregnancy Associated Mortality Review 1999-2002 Report21.

A Look at the Findings

‡Based on the American College of Obstetricians and Gynecologists Guidelines for Perinatal Care, 5th Edi-

tion18. Generally, a woman with an uncomplicated pregnancy is examined every 4 weeks for the fi rst 28 weeks

gestation, every 2 weeks until 36 weeks of gestation, and weekly thereafter. Th e frequency of follow-up visits is

determined by the individual needs of the woman and the assessment of her risks.

Level of prenatal care Month prenatal care began Percentage of recommended

visits of prenatal care‡‡,18

Adequate Plus ≤4th month of pregnancy ≥110%

Adequate ≤4th month of pregnancy 80%-109%

Inadequate ≤4th month of pregnancy <50%

or

≥ 5th month of pregnancy Not applicable

No Care None None

[ 23 ]


ResultsTh irty-seven maternal deaths were reported to ACOG Dis-

trict II/NY’s SMI from August 2003 through June 2005.

Since that time, additional data on 2004 deaths has been in-

cluded for statistical purposes. Of the 37 deaths, ACOG

conducted on-site reviews for 21 pregnancy-related deaths.

One of the 37 deaths was classifi ed as not pregnancy-relat-

ed and three did not meet criteria§§ for review by the SMI

program. Twelve cases were reviewed by an external review

program and the fi nal report for each case was submitted to

ACOG for SMI data incorporation and analysis.

A total of 33 pregnancy-related deaths were reviewed and

included in this analysis. Due to the absence of complete

documentation in the prenatal and intrapartum medical re-

cords, autopsy reports and death certifi cates, the information

analyzed in this report includes many unknowns. Table 4.1

Among the pregnancy-related deaths that were reviewed,

race/ethnicity had an approximately equal distribution.

More minority women were studied; black women and other

races/ethnicities represented 57.6% of the sample.

Th e ages of the pregnancy-related deaths ranged from 17

to 42 years with a mean age of 30.5 years. It was diffi cult to

determine whether English was the primary language of the

women reviewed; however, this does appear to be the case.

Of the sample, Medicaid appeared to be the primary method

of payment, followed by private insurance.

Of the pregnancy-related deaths reviewed, more than 80%

(n=28) occurred in the downstate region and 63.6% (n=21)

died at either a Level III or IV/RPC hospital. Autopsy in-

formation was available for only 16 maternal deaths.

Limited information obtained from the reviewed pregnan-

cy-related deaths revealed that more women in the sample

had higher levels of education and age. Women aged 30-34

had over 12 years of education. In the SMI sample, there

Table 4.1 Demographic characteristics of the pregnancy-related deaths

Categories n=33 (%)

Race

Black 10 (30.3)

White 8 (24.2)

Other 9 (27.3)

Unknown 6 (18.2)

Age

19 and younger 4 (12.1)

20-24 6 (18.2)

25-29 6 (18.2)

30-34 7 (21.2)

35-39 4 (12.1)

40 and over 6 (18.2)

English as a primary

language

Yes 15 (45.5)

No 9 (27.2)

Unknown 9 (27.2)

Region

Upstate 5 (15.2)

Downstate 28 (84.8)

Level of hospital

I 5 (15.2)

II 7 (21.2)

III 14 (42.4)

IV/RPC 7 (21.2)

Method of Payment

Uninsured/Self Pay 3 (9.1)

Medicaid 13 (39.4)

Managed Care/Private 12 (36.4)

Unknown 5 (15.2)

§§Insuffi cient prenatal or provider information available.

[ 24 ]

were more unmarried black women than white women. Overall, within the sample, most

of the women who died were unmarried.

Within the sample, 30 women had a live birth. Only 50% can be documented to have

begun prenatal care within the fi rst trimester. More than 50% of the pregnant women who

died received at least some prenatal care. Based on the levels of prenatal care developed by

Kotelchuck and modifi ed by the CDC, 18.2% (n=6) of the sample received ‘adequate plus’

prenatal care; 27.3% (n=9) were determined to have received ‘inadequate prenatal’ care and

the majority were not documented or unknown*** (Table 4.2)

Parity varied within

the sample and more

women were preg-

nant for the fi rst time

(27.3%) than in any

other single level of

parity. Of the wom-

en studied who had

a live birth (n=30),

more pregnancy-re-

lated deaths occurred

following a cesarean

section (69.7%) than

following a vagi-

nal delivery (21.2%).

Of the 21 maternal

deaths that were re-

viewed on-site by

ACOG, more than

half (57.1%) were determined by the review team to be ‘at some risk’, followed by ‘at high-

risk’ (33.3%). (Table 4.3)

Of the 33 pregnancy-related deaths reviewed by the SMI, the leading causes of death were

embolism (24.2%), PIH (24.2%), hemorrhage (15.2%), and infection (15.2%). (Table 4.4)

Th ere were no reported maternal deaths associated with anesthesia or CVA. Within the

sample size, more hemorrhage deaths occurred among black women than white wom-

Table 4.2 Maternal deaths by race, prenatal care, and adequacy of prenatal care use

Category Race

Black White Other Unknown Total

(n=33)

Began Prenatal Care

First trimester 7 4 3 1 15

Second trimester 2 2 - 4

Th ird trimester 1 1 - - 2

No care - - - - 0

Unknown 2 1 4 5 12

Adequacy of Prenatal Care Use

Adequate plus 4 2 - - 6

Adequate 1 1 1 - 3

Inadequate 2 4 3 - 9

No Care - - - - 0

Unknown 3 1 5 6 15

***Th is fi nding should be interpreted with caution since only 33 pregnancy-related deaths had available infor-

mation and incomplete prenatal care information existed among many pregnancy-related deaths reviewed.


[ 25 ]


Table 4.3 Selected maternal and pregnancy characteristics ofpregnancy-related deaths

Characteristic n = 33 (%) Characteristic n=33 (%)

Parity Timing of death after delivery

unknown 1 (3.0)

0 9 (27.3) < 24 hrs 9 (27.3)

1 6 (18.2) 24 hrs – 1 week 12 (36.4)

2 5 (15.2) 1 week- 42 days 8 (24.2)

3 6 (18.2) > 42 days 1 (3.0)

≥4 6 (18.2) Undelivered/TOP 3 (9.1)

Type of delivery Mother’s Risk (n=21)

Vaginal 7 (21.2) At Some Risk 12 (57.1)

Cesarean section 23 (69.7) At High-Risk 7 (33.3)

Primary 11 (47.8) At Very High-Risk 2 (9.5)

Repeat 12 (52.2)

TOP 1 (3.0)

Undelivered 2 (6.0)

en and women of other

races/ethnicities com-

bined. Yet, white women

died more often of em-

bolism than did black

women. Th e “other” cat-

egory includes: undiag-

nosed cardiac disease and

hemorrhage status post

pneumonectomy. Th ree

maternal deaths were un-

able to be classifi ed due

to multiple underlying

causes; the lack of avail-

able information such as

an autopsy report; or the

lack of consensus among

the maternal mortality

review team to conclu-

sively determine a cause

of death.

Of the pregnancy-related deaths reviewed by

the SMI, a live birth (81.1%) was the most

frequent pregnancy outcome followed by

stillbirth/fetal demise (16.2%). Th ere was

one termination of pregnancy. Of the wom-

en reviewed that had a stillbirth/fetal demise

(n=6), the leading causes of death were PIH

and infection.

For the cohort who delivered a live birth

(n=30), seventy percent (70%) died within

one week following delivery and thirty per-

cent (30%) died within twenty-four hours of

delivery. Th e women who died of PIH did

so between 24 hours to 1 week following de-

livery. Th e leading two causes of death among the portion of the sample that delivered via

cesarean section were embolism and PIH (26.1%). (Figure 4.5)

Table 4.4 The leading causes of pregnancy-related deaths reviewed by the SMI.

Cause of death n=33 (%)

Embolism 8 (24.2)

PIH 8 (24.2)

Hemorrhage 5 (15.2)

Infection 5 (15.2)

Cardiomyopathy 2 (6.1)

CVA None

Anesthesia None

Other 2 (6.1)

Unknown 3 (9.1)

[ 26 ]

Based on the 33 pregnancy-related deaths reviewed, 54.5% (n=18) had a his-

tory of chronic disease. Th e following chronic diseases were identifi ed: hyper-

tension, deep vein thrombosis, cardiac disease, diabetes, scleroderma, polycystic

ovarian syndrome, and sickle cell anemia. Of the cases with chronic disease,

half were Black and the remaining

were White or other races/ethnici-

ties. Th e most commonly reported

chronic disease was obesity, occur-

ring in 66.6% (n=12). Similar to

the overall causes of death, analysis

indicated that the leading cause of

death among obese women with a

BMI ≥ 30, was embolism (33.3%)

followed by PIH (25%) and hemor-

rhage (16.7%). (Table 4.5)

Pregnancies associated with ma-

ternal death are not consistently

reported within New York State,

as is the case across the country.

A review of the 2004 pregnancy-

related deaths reported to ACOG

via the Safe Motherhood Initia-

tive (SMI, n=25) compared to the

data identifi ed by Statewide Plan-

ning and Research Cooperative

System’s (SPARCS, n=28) revealed

that ACOG was not notifi ed of

47% of pregnancy-related deaths

that occurred in that one calendar

year.††† Table 4.6 illustrates the

most comprehensive one year data

analysis currently available in New

York State. While the SMI’s ac-

tive surveillance system reviewed

25 deaths and additional 28 deaths

were reported to ACOG through the SPARCS’ passive surveillance system, it

is estimated that these fi ndings still represent an underreporting of one year

data on pregnancy-related deaths in New York State.

Figure 4.5 Pregnancy-related deaths reviewed by cause of death and type of del ivery

0

2

4

6

8

10

Cause of Death

# of

Pre

gnan

cy-r

elat

ed D

eath

s

Vaginal

Cesarean Section

Em

bolism

Infe

ction

Hem

orrh

age

PIHCar

diomyo

pathy

Oth

erU

nknown

Table 4.5 Number of pregnancy-related deaths by cause of death and BMI

Cause of death BMI

Normal Overweight Obese Unknown

18.5 -24.9 25.0 – 29.9 ≥30

Embolism 2 - 4 2

Infection 1 1 1 2

Hemorrhage - 2 2 1

PIH 2 - 3 3

Cardiomyopathy 2 - - -

Other - - 1 1

Unknown - 1 1 1

Total 7 4 12 10

†††Hospital Discharge data was provided by the University at Albany, School of Public Health from the “Ad-

ministratively Releasable” fi les of SPARCs data. 2004 data was received through 9/05.


[ 27 ]


Recommendations

Th e recommendations pro-

vided herein are based on

the conclusions reached by

members of the mortal-

ity review team during the

twenty-one, on-site SMI

reviews. After careful con-

sideration of all available

facts at the time of each

review, the recommenda-

tions refl ect consensus from

the members of the review

team.

Th e overarching recom-

mendations relate to im-

proving systems, support

services, communication

and to improving the quali-

ty of obstetric services. On-

site reviews identifi ed mul-

tiple areas for improvement

which substantiated the need to promote quality

across the continuum of maternity care. Specifi cal-

ly, maternal deaths did not result from one isolated

contributory factor, but are the result of the com-

bined failure of various systems and other issues.

Th e recommendations of each review team member

were divided into obstetric care or systems catego-

ries. Overall, system recommendations were sig-

nifi cantly more common than obstetric recommen-

dations. It is not surprising that the most common

system recommendations were related to docu-

mentation and collaborative/consultative practice.

Within the obstetric category, the majority of the

recommendations related to assessment of high-

risk patients and the use of pharmacologic/medical

therapies. (Table 4.7)

Table 4.6 Numbers of Maternal Deaths by Source of Identifi cation and Selected Maternal and Pregnancy Characteristics: New York, 2004

Maternal and 2004 Cases By Source of Identifi cation

Pregnancy SMI Hospital Discharge All Sources

Characteristic N = 25 N = 28 N = 53

Race

White 5 7 12

Black 8 6 14

Other 10 2 12

Unknown 2 13 15

Age

15-24 8 5 13

25-29 3 3 6

30-34 5 7 12

35-39 4 4

40 and over 5 1 6

Source of Admission

MD/Clinic Referral 12 5 17

ER 8 8 16

Transfer 2 2

Unknown 5 1 6

Table 4.7 Distribution of overarching recommendations

Type of recommendations n (%)

Obstetric

Assessment of high-risk patients 9 (42.9)

Pharmacologic/medical therapies 9 (42.9)

Obstetric complications/emergencies 7 (33.3)

•obstetric hemorrhage 3 (14.3)

•sepsis 3 (14.3

Assessment of obstetric patients 6 (28.6)

Systems

Documentation 21 (100)

Collaborative/consultation practice 11 (52.4)

History of prenatal care 8 (38.1)

Support services 7 (33.3)

Grief counseling 7 (33.3)

Transfer to a higher level of care 6 (28.6)

CME/credentialing/privileges 3 (14.3)

Translation services/Cultural Competency 2 (9.5)

[ 28 ]

Obstetric Care Recommendations

Assessment of high-risk patients• Develop hospital protocols to encourage MFM consults for all high-risk obstetric

patients in a timely and effi cient manner. A backup system with another MFM spe-

cialist is recommended in cases when the primary MFM has been under prolonged

duty or is unavailable.

• Policies should be developed regarding the management of psychiatric or disruptive

patients in the labor and delivery unit.

Pharmacologic/medical therapies• Provide guidelines and continued education to maternity care professionals regard-

ing the appropriate use of pharmacologic and medical therapies, including but not

limited to the following: Magnesium Sulfate, Dilantin, Ativan, Valium, anticoagula-

tion therapy, antibiotic prophylaxis, intravenous fl uids, and peripheral vascular dop-

pler studies. Such guidelines and educational curricula should include the medical

indication(s) as well as nationally accepted pharmacologic dosages.

Obstetric complications/emergencies• Hospitals should develop a multi-specialty task force to respond to and manage ob-

stetric emergencies.

• Provide education to physicians and allied personnel to improve the management of

hypertensive disorders, obstetric hemorrhage, disseminated intravascular coagulation,

sepsis, pulmonary embolism and obese obstetric patients. Such education should

encourage nursing personnel to consistently communicate the patient’s status to the

managing physician(s), either attendings or residents.

• Mock codes are encouraged, on all shifts, to increase comfort level with procedures

and ensure consistency with responses to obstetric and medical emergencies.

• Patients with chronic diseases and/or potentially life-threatening pregnancy or non-

pregnancy-related complications should have preconception counseling by relevant

medical providers and it should be repeated in early pregnancy. Th is should be docu-

mented in the medical record.

Obstetric hemorrhage• Devise mechanisms to ensure that maternal observations result in accurate, timely

and appropriate estimation of blood loss, as well as early identifi cation of hypotension,

hypothermia and tachycardia. Drills for the management of obstetric hemorrhage

should be conducted on a regular basis.

• An ‘Obstetric Hemorrhage’ protocol should be developed at all delivery sites.

• All members of the healthcare team should be empowered to activate an obstetric

hemorrhage protocol


[ 29 ]


Sepsis• Develop hospital protocols for the evaluation and management of febrile patients and

patients requiring focused wound care, especially the obese obstetric patient.

• Develop systems to ensure that patients treated empirically for suspected infections

have antibiotic sensitivities verifi ed.

Assessment of obstetric patients• Any patient who is suspected of being in labor should be evaluated promptly in an

obstetric service area particularly when the patient’s prenatal record and/or provider

are not available. A focused history and physical exam should be carried out and the

following factors should be assessed and recorded in the patient’s permanent medical

record:

- Obstetric and medical history;

- Maternal vital signs;

- Frequency and duration of uterine contractions;

- Urinary protein and glucose concentrations;

- Cervical dilatation and eff acement, unless contraindicated;

- Fetal presentation and station of the presenting part.

• Provide education to obstetrician-gynecologists, midwives and registered nurses on

the proper assessment of obstetric patients who present at the hospital.

Systems Recommendations

Documentation• Strengthen the continuity of obstetric care through maintenance of a complete prena-

tal and perinatal medical record. Th e medical record should contain proper and con-

temporaneous documentation of vital signs, laboratory values, routine and emergency

interventions, physical examinations, consultations, orders, and a complete case man-

agement summary. Proper and thorough documentation practices should provide a

timeline of events surrounding the patient’s progress.

• Document in the medical record communication between members of an interdisci-

plinary health care team including nursing staff .

• In the cases of multiple physicians co-managing one patient, documentation must

include the written account(s) of face-to-face, verbal discussions at each critical point

in time; such as when there is any change in the patients’ status, or when there is a

change in the patients’ management plan.

• Promote chart documentation policies and education to ensure that physicians’ and

nurses’ documentation in the medical record is signed, dated, timed, and legible.‡‡‡

• Develop a policy to ensure complete emergency department records are included in

the patient’s medical record.

‡‡‡ A Joint Commission on Accreditation of Healthcare Organizations ( JCAHO) requirement.

[ 30 ]

• Improve prenatal care record documentation regarding communication with the pa-

tient on methods and associated risks surrounding the plan of delivery.

• Document all recommendations for follow-up care, medications, or consultation in

the medical record.

Collaborative/consultative practice• Improve methods for open communication among all perinatal providers, i.e., at-

tending physicians, nurses, midwives, residents in training, house staff , and consulting

physicians.

• Promote RPC-affi liate hospital communication and active consultation with a col-

laborative and/or lead physician at established levels of obstetric and pediatric risk.

In addition, inter-departmental consultations within the hospital and referrals among

other specialties will better serve the patients and diminish liability.

• Improve methods for co-management between patient’s obstetrician, MFM specialist

and the critical care or consultation team.

• Residents and hospital staff members should routinely update attending physician(s)

with a detailed description of the patient’s progress, particularly of the sequence of

events surrounding the patient’s care.

• If a multidisciplinary team is assigned to the care of the patient, a primary physician

should be designated for the coordination of the patient’s care to ensure that continu-

ous quality care is provided.

• Promote the identifi cation of leadership responsibilities during code situations. Sys-

tems should be developed to designate the members of the code team. Advanced

Cardiac Life Support (ACLS) training for the obstetrician-gynecologist, residents,

attendings, and hospital staff is optimal.

History of prenatal care• Strengthen the continuity of perinatal care by having a complete prenatal and perina-

tal medical record from the prenatal clinic or private physician’s offi ce at the deliver-

ing hospital when the patient is admitted.

• Th e use of a standardized and/or centralized medical record form is recommended

to ensure complete information is available for the institution receiving the patient.

Establishing an electronic health record would be optimal.

Support services• Th e department of obstetrics and gynecology, quality improvement personnel and the

blood bank should develop policies ensuring rapid availability and release of blood

products for transfusion in the event of obstetrical hemorrhage. Availability of Type

O Rh negative blood in the labor and delivery room at all times may be warranted.

Th ese policies should be balanced with the need to conserve valuable blood bank re-

sources. Th e hospital blood bank supervisory personnel should be immediately noti-

fi ed of the potential need for massive transfusion.

• Until blood loss is controlled by medical or surgical means, it is advisable to ensure

the rapid availability of suffi cient packed red blood cell (PRBC) units and other blood


[ 31 ]


products for rapid transfusion. It is recommended that blood banks establish guide-

lines that will expedite the release of blood products, e.g., possibly waive the need

for a second/confi rmatory type and cross specimen in obstetric emergencies such as

postpartum hemorrhage. Th e blood bank should also assign a specifi c individual to

work on obstetric emergency cases; this will facilitate and accelerate the dispatch of

blood products.

• A quality control monitoring plan with labor and delivery personnel should be es-

tablished for the performance of cesarean sections. Personnel in the obstetric de-

partment should be appropriately cross-trained to scrub and circulate for emergency

cesarean sections, as the opportunity arises.

• Encourage early triage of all obstetric patients by the anesthesia department for risk

classifi cation.

Grief counseling• Promote availability of continued services that facilitate coping and healing, such as

counseling, for hospital staff and family members impacted by a maternal death.

• Staff members and physicians who are in contact with the medical professionals in-

volved in the decedent’s care should be aware of emotional changes, including stress

levels, and should report any signifi cant behavior changes as appropriate.

Transfer to a higher level of care• Hospital protocols regarding the management of critical patients that require one-to-

one patient care by labor and delivery staff should be encouraged.

• Guidelines should be developed that facilitate timely transfer of women with pre-ex-

isting chronic disease, high-risk obstetric histories, or complications developing dur-

ing pregnancy to a higher level of service and/or hospital.

• Protocols should be developed for the evaluation and transfer of critical postoperative

obstetric patients to a unit with on-site hemodynamic monitoring systems, including

central venous pressure, arterial lines, and oxygen saturation monitoring with record-

ing.

• Encourage RPCs to investigate their local emergency medical service’s quality im-

provement programs to ensure appropriate care during transport of women with

perinatal emergencies. RPCs should reach out to facilities where care for obstetrical

patients is not common or available in order to increase education and planning, up-

date staff on the basic obstetric care requirements, and to encourage referral in a more

timely fashion to facilities and/or providers with broader obstetric experience.

CME/credentialing/privileges• Develop mechanisms to ensure attending physicians properly supervise residents and

licensed/unlicensed staff managing the patient.

• Strengthen mechanisms that support continued medical education, experience and

expertise to maintain competence in new and existing obstetric procedures and thera-

pies for all medical staff .

[ 32 ]

Translation services/cultural competency• Improve the availability of high quality, culturally sensitive and linguistically appro-

priate care related to labor and delivery, stress and grief management. Obstetric pro-

viders must seek interpretation and/or culturally appropriate services for patients with

limited English profi ciency.

• Educational opportunities should be off ered to all staff regarding cultural behaviors

associated with labor and delivery, stress, and grief. Th is program should provide staff

with methods to obtain, negotiate, and manage cultural information in every clinical

encounter, thereby reducing racial and ethnic disparities in health care.


[ 33 ]


Chapter 5: DiscussionTh e Healthy People 2010 objective is to reduce maternal deaths to 3.3 per 100,000 live

births.10 Although this same objective was repeated from the Healthy People 2000 report,

data from NYS DOH vital statistics indicate that NYS is not close to achieving this objec-

tive.9 Furthermore, statewide fi ndings from the 33 pregnancy-related deaths reviewed by

the SMI revealed a need to improve the quality of care obstetric patients receive within the

state.

Similar to data from the CDC, the leading causes of pregnancy-related deaths reviewed by

the SMI were embolism, PIH, infection, and hemorrhage. Given that the majority of the

pregnancy-related deaths reviewed occurred within the fi rst week following delivery, at-

tention is needed during the immediate postpartum period. Professional education should

describe warning signs for early identifi cation of postpartum obstetric complications and

emergencies. Practice drills are highly recommended.

Preconception/Prenatal Care

Th e analysis of prenatal care records available at the time of review suggests that women are

entering prenatal care during the fi rst trimester; however, prenatal care was not continu-

ous. More analysis of the contributing factors related to fragmented prenatal care obtained

by the women studied is needed. Some of the contributing factors may include: delay in

enrollment in government subsidized health insurance programs, dissatisfaction with care

leading to non-continuation of care, citizenship and relocation. Moreover, statewide eff orts

should be instituted that encourage health care providers to establish a follow-up system

for women who fail to keep scheduled appointments. Such a system might include follow-

up calls, home visits, or the assignment of case managers.

Of the 21 pregnancy-related deaths reviewed on-site by the SMI, 43% of the pregnancies

were determined by the review team to be at ‘high’ or at ‘very high-risk’. Th e data suggest

that the identifi cation of patients at high-risk needs to occur as early as possible, prefer-

ably during preconception, but certainly prior to a woman’s hospitalization and/or delivery.

Hospitals and physicians should consider scheduling or participating in education focused

[ 34 ]

on identifying and addressing the needs of patients at high-risk as part of their quality

improvement activities.

Conducting the SMI on-site review

Maternal deaths are rare and unexpected events within the area of obstetrics. When such

an event occurs, health care providers and institutions extensively evaluate the care given

to the particular woman whose family and personal characteristics they still recall. SMI

reviews have empowered individuals and institutions to be agents and advocates of change

within the obstetric community.

Th e most important principle revealed by the SMI on-site reviews has been the signifi cance

placed on the confi dential, anonymous, timely and non-threatening environment in which

the reviews are conducted and fi ndings analyzed. Th is non-punitive approach has led to

openness regarding the care provided, and facilitated the attainment of a more complete

picture of the precise sequence of events that led to the maternal death.

Health care provider interviews have proven to be one of the most important components

of the SMI process. Th e interviews tell the unique story behind each maternal death,

not otherwise discerned from medical chart review alone. For example, it is during the

interviews that information is obtained regarding the underlying reasons for patient “self-

discharge against medical advice” or why obstetric patients are “lost” to another specialty

service within the hospital. During the interviews, usually following a process of self refl ec-

tion, health care providers describe their contributions to the care of the woman and are

motivated to implement, or have already implemented, change within both the department

of obstetrics and the hospital. Furthermore, interviews have uncovered the communication

disconnect among specialists and sub-specialists in the care of obstetric patients, improved

understanding of the problems identifi ed, and provided recommendations for quality im-

provement in maternity care. Th e individuals and hospitals who voluntarily devote their

time and eff ort repeatedly expressed the long lasting educational and emotional healing

eff ect the SMI review process provides.

SMI reviews identifi ed the need to implement mechanisms that facilitate coping and heal-

ing for all staff aff ected by the maternal death. It was recognized that health care profes-

sionals involved in the decedent’s care were under signifi cant emotional stress and were

personally impacted by such catastrophic events. Professional counseling was consistently

recommended to help staff to continue to practice and cope with such outcomes.

Autopsy Reports

Strategies need to be developed to educate maternal-child health professionals, medical ex-

aminers/coroners, and family members regarding the importance of conducting autopsies

on all pregnancy-related deaths. Based on twenty-one reviewed cases, three deaths were

classifi ed as an unknown cause of death and only 16 had autopsy reports (verbal or written)

Discussion

[ 35 ]


available at the time of review. Th e availability of an autopsy report, while not listed as a

limitation of the SMI, would have assisted SMI review teams to properly assess, review, and

categorize deaths.

Th e lack of autopsy reports (verbal or written) may be related to the following factors:

1. Th e maternal death certifi cate is not completed in a timely manner.

2. Insuffi cient information is provided to the family members regarding the importance

of conducting an autopsy for better understanding of the causes related to her death.

3. Th e medical examiner/coroner may refuse the case because it is not properly defi ned

as a maternal death.

4. Even when an autopsy is ordered, the draft autopsy report and fi ndings are not pro-

vided routinely to the hospital where the death occurred and are therefore not in-

cluded in the obstetric chart.

Educational pamphlets such as those provided to the family following a fetal death could

be one of the multiple strategies developed to assist and educate families.

Embolism

As expected, one of the leading causes of death following a cesarean section of the studied

cohort was embolism. Th e SMI data suggest that a risk assessment of women undergoing

elective or emergency cesarean section should be performed and prophylaxis instituted, as

appropriate.

In 1995 the CEMD instituted recommendations with continued documented success in

the reduction of pregnancy-related deaths due to thromboembolism, during preconception,

pregnancy, intrapartum (vaginal and cesarean deliveries) and in the postpartum period.1

Th e CEMD recommends that all women undergo an assessment of risk factors for venous

thromboembolism (VTE) in the preconception period, early pregnancy and during the

postpartum period. Th ey identify the following factors to increase the risk for thromboem-

bolism: obesity, age over 35, pre-eclampsia, immobility, and excessive blood loss. Moreover

the CEMD recommends that clinicians caring for pregnant women who are over 35 years

and have a BMI greater than 30 and any other risk factors for VTE (such as pre-eclampsia

or immobility) should consider the use of low molecular weight heparin for three to fi ve

(3-5) days postpartum.

Th e table below, taken from the CEMD 1995 Working Party Report on Prophylaxis

Against Th romboembolism, and updated by the 2004 Royal College of Obstetricians and

Gynaecologists provides widely used recommendations for risk assessment in caesarean sec-

tion.22, 23 Th e SMI should work with statewide experts in adverse pregnancy outcomes and

maternal mortality to develop and model guidelines, like those published by the CEMD,

for the prevention of New York State pregnancy-related deaths due to embolism.

[ 36 ]

CEMD: Risk assessment profi le for thromboembolism in cesarean section.22, 23

Risk Recommendations

Low risk:

Early

mobilization

and hydration

• Elective cesarean section: uncomplicated pregnancy and no

other risk factors.

• Women undergoing elective cesarean section with uncomplicated

pregnancy and no other risk factors require only early mobilization

and attention to hydration.

Moderate risk:

Consider one

of a variety of

prophylactic

measures

• Age > 35 years

• Obesity (> 80 kg)

• Parity 4 or more

• Labor 12 hours or more

• Gross varicose veins

• Current infection

• Pre-eclampsia

• Immobility prior to surgery (> 4 days)

• Major current illness (e.g., heart or lung disease, cancer,

infl ammatory bowel disease, nephrotic syndrome)

• Emergency cesarean section in labor.

Women assessed as of moderate risk should receive subcutaneous heparin (doses are

higher during pregnancy) or mechanical methods. Dextran 70 is not recommended

until after the delivery of the fetus and is probably best avoided in pregnant women.

High risk:

Heparin

prophylaxis

with or without

leg stockings

• A woman with three or more moderate risk factors from above.

• Extended major pelvic or abdominal surgery (e.g., cesarean

hysterectomy).

• Women with personal or family history of deep venous thrombosis,

pulmonary embolism or thrombophilia, paralysis of lower limbs

• Women with antiphospholipid antibody (cardiolipin antibody or

lupus anticoagulant).

• Women assessed as high risk should receive heparin prophylaxis and,

in addition, leg stockings would be benefi cial.

• Prophylaxis until the fi fth postoperative day is advised (or until fully

mobilized if longer).

• Th e use of subcutaneous heparin as prophylaxis in women with an

epidural or spinal block remains contentious. Evidence from general

and orthopedic surgery does not point to an increased risk of spinal

hematoma.

Discussion

[ 37 ]


Obesity and Pregnancy

Similar to the fi ndings from Florida’s PAMR, over half (n=23) of the pregnancy-related

deaths reviewed occurred in obese women. In addition, the leading cause of death among

this sample of obese women was embolism followed by PIH and hemorrhage. Recent

research confi rms that morbid obesity is an independent risk factor for perinatal and ges-

tational complications.24 Moreover, another publication examined the eff ect of maternal

weight and obesity on labor progress and found a much slower labor progression before six

centimeters of cervical dilation in the women studied. According to the authors, “given

that nearly one half of women of childbearing age are either overweight or obese, it is

critical to consider diff erences in labor progression by maternal pre-pregnancy BMI before

additional interventions are performed.”25 Such a management strategy may reduce the

performance of cesarean deliveries in this high-risk population and thereby the post-opera-

tive complications.

Th ese fi ndings are particularly signifi cant due to the growing number of Americans who

are overweight and obese, which may potentially impact the number of pregnancy com-

plications and pregnancy-related deaths. Obese women are at increased risk for pregnancy

induced hypertension, cesarean delivery, delivery of large-for-gestational-age infants, and

stillbirths. Obesity among women of child bearing age is a risk factor that should be ad-

dressed so appropriate antenatal and intrapartum care can be provided. In addition, insti-

tutional changes within the obstetric community are recommended to educate perinatal

providers to address the needs of this high-risk population. Systems that will educate

women on the importance of attaining an ideal body weight prior to pregnancy and address

the need of pre-pregnancy advice and counseling among these obese women are vital to

reducing maternal obesity and potential pregnancy-related deaths.

Obstetric Hemorrhage

Since obstetric hemorrhage is one of the most preventable causes of maternal death, ACOG

in conjunction with the NYC DOHMH and NYS DOH issued a health alert focused on

methods to prevent maternal deaths through improved management of hemorrhage. As

a result, ACOG received numerous requests for sample hospital postpartum hemorrhage

(PPH) protocols. A SMI hemorrhage subcommittee was formed to select model protocols

that hospitals could adapt to meet their specifi c needs and resources. Th e subcommittee

reviewed eight PPH protocols from various level hospitals in NYS.

Th e committee also identifi ed the following core components that all PPH protocols should

include and/or consider:

• Schedule and conduct regular obstetric hemorrhage drills;

• Identify who will be called to respond to PPH;

• Designate staff to run the PPH response;

[ 38 ]

• Incorporate and utilize an obstetric fl ow sheet to help recognize when a patient is

approaching PPH status;

• Decision for transfusion must be based on accurate estimation of blood loss and not

dependent on laboratory values. Th e overall status of the patient must be considered:

vital signs, blood pressure, pulse, urine output, estimated blood loss up to that point,

attainment of homeostasis, and associated medical problems, etc;

• Schedule annual or bi-annual review of all obstetric protocols, especially PPH proto-

cols.

In February 2005, the above PPH protocol recommendations and the sample hospital

protocols were mailed to hundreds of hospital obstetric chairs, obstetric nurse managers

and hospital quality assurance personnel in New York State. Additional requests from

statewide obstetric providers continue to this day. Albany Medical Center conducted a

follow-up survey with their affi liate hospitals to determine how the PPH protocols and

recommendations were received. Utilizing a simple three question email survey, Albany

Medical Center found that their affi liate hospitals were: (1) aware of the ACOG postpar-

tum hemorrhage resources, (2) reviewing their current postpartum hemorrhage protocol,

and (3) moving forward in making modifi cations to improve their existing protocols.

In conjunction with the release of the PPH protocols and recommendations, ACOG sub-

committee members developed a curriculum and presented continuing medical education

(CME) programs throughout the state on the management of PPH. Th e PPH protocol

recommendations and CME programs have received a tremendous amount of positive

feedback within the New York State obstetric community.

Preconception Counseling

Because of the high rate of unintended pregnancy in the United States, the national offi ce

of the American College of Obstetricians and Gynecologists and the American Academy

of Pediatrics recommends that all health encounters during a woman’s reproductive years,

particularly those that are a part of preconception care, should include counseling on ap-

propriate health behaviors to optimize pregnancy outcomes.18 Th e ACOG analysis of 21

pregnancy-related deaths, with on-site review, revealed that 43% were determined to be ‘at

high’ or ‘at very high-risk’ and 54.5% had a history of chronic disease. Th is supported the

rationale for the development of a subcommittee on preconception counseling. Th e focus

of the subcommittee was to address chronic and pre-existing diseases as they relate to the

identifi cation of maternal risk that could potentially contribute to adverse pregnancy out-

comes, including maternal death.

Maternal risks of pregnancy in women with chronic disease are best addressed in the pre-

conception period. Care prior to pregnancy allows women with medical disorders to at-

tempt a pregnancy in optimal condition or decide to defer or avoid pregnancy when risks

are high. Since preconception care provides an opportunity to prevent maternal morbid-

Discussion

[ 39 ]


ity and mortality, the SMI partnered with the March of Dimes preconception campaign

to educate physicians and maternity care professionals. 26s Utilizing the March of Dimes

core curriculum for preconception counseling, ACOG adapted an educational program

that included case discussions of maternal deaths. Two statewide preconception counseling

conferences were provided to community health care workers and hospital-based maternity

professionals.

Obstetric Triage

Data obtained from the SMI on-site reviews revealed a need to study emergency medi-

cal services and emergency room (EMS/ER) policies regarding the triage and transfer of

obstetric patients. Th e SMI convened a subcommittee to develop and make recommen-

dations for EMS transfer and obstetric triage. Subcommittee participants included the

New York State Department of Health Bureau of Emergency Medical Services, the New

York Chapter of the American College of Emergency Physicians, the Regional Emergency

Medical Service Council and the SMI planning committee. Th e subcommittee concluded

with the following recommendations regarding obstetric triage:

• RPC’s must be encouraged to review EMS quality improvement programs to assure

appropriate care in perinatal emergencies.

• Facility education is needed for non-obstetric facilities, particularly regarding the

timely triage and transfer of obstetric patients to appropriate levels of care.

Following extensive research and discussion, it was found that little progress could be made

regarding EMS transfers due to current EMS structure and inconsistencies between paid

and volunteer districts.

Legislative/Policy

Th e SMI has brought RPC and affi liate hospitals together to promote quality improve-

ment and working relationships. It has been apparent that the SMI program has alleviated

some hospital competitive forces, thus encouraging and enhancing ongoing RPC-affi liate

hospital communication and collaboration.

Case identifi cation is dependent on voluntary hospital reporting because the SMI remains

separate from the RPC quality review process. Th is is particularly diffi cult for the con-

tinuation of the project, since confi dentiality concerns still exist, especially among hospital

quality assurance departments. Outreach continues to be needed to hospital risk manage-

ment and quality assurance departments to encourage hospital participation.

In order to improve the quality of maternity care delivered by obstetric, pediatric, and

primary care providers, maternal deaths must become reportable to an entity that can con-

duct surveillance and make regular reports and frequent recommendations. It is only by

[ 40 ]

continuing to study all of these rare, yet tragic, events that suffi cient information can be

gained to impact obstetric quality. While RPC’s are required to conduct quality assurance

and improvement activities with their affi liate hospitals, maternal mortality is not routinely

reviewed or studied in any systematic fashion. Without a statewide system that prompts

RPC’s to use the SMI review protocol, review maternal deaths, and share de-identifi ed

information, surveillance will be halted. Th e United Kingdom’s Confi dential Enquires

requires all maternal deaths to be reviewed and all care-givers must participate. Hence,

meaningful recommendations on achieving safe motherhood become possible.

Th e NYS DOH Bureau of Vital Statistics records maternal deaths in NYS based on in-

formation obtained from death certifi cates. Th e death certifi cate asks specifi c questions

regarding the patient’s pregnancy status at the time of death and within the last year before

the death. In several cases reviewed by ACOG, this information was not completed. Pre-

vious reports have also identifi ed the inaccuracy of NYS death certifi cate and pregnancy

check-box completion.27 It is very important for pregnancy-related information to be re-

corded correctly to more accurately refl ect the maternal mortality rates in New York State.

Discussion

[ 41 ]


Chapter 6: LimitationsLimitations related to the development, implementation, and analysis of data from the Safe

Motherhood Initiative should be recognized. An evaluation of the fi rst year of the SMI

project indicated that the 30 member planning committee, while increasing ownership and

buy-in, prolonged the development of the maternal death protocol and abstraction form.

A smaller planning group, composed of fewer committee members, may have expedited the

developmental process and assured focused discussion of controversial material. However,

fi nal consensus on the protocol may not have been so readily achieved.

From the outset, the Safe Motherhood Initiative has been considered a possible prototype

for other quality assurance activities by RPCs such as cesarean delivery or induction of

labor reviews. Despite eff orts made by ACOG and NYS DOH, commencement of the

on-site SMI reviews was delayed due to lack of project awareness, protracted designation

of maternal mortality review team members, and concerns regarding the project’s voluntary

nature and confi dentiality protections.

For example, at the request of the NYS DOH Division of Legal Aff airs, all prospective ma-

ternal mortality review team members from every Regional Perinatal Center in New York

State were required to sign a confi dentiality agreement in order to review maternal deaths.

Th is needed, yet lengthy, process delayed physician and hospital participation in the SMI

and delayed the initial project commencement. Despite ACOG’s extensive work with each

RPC, confi dentiality agreements have not been signed by key individuals who are needed

to participate in maternal death reviews.

Due to the uniqueness of each pregnancy-related death, review team members must in-

clude providers from other specialties such as radiologists, nephrologists, hematologists,

critical care specialists, medical examiners, nurse managers, and quality improvement per-

sonnel. Each additional maternal mortality review member must be reviewed by the NYS

DOH for approval prior to participating in a review. Mechanisms to facilitate and con-

dense confi dentiality protections will enhance the approval process, and the fl uidity and

timeliness of reviews.

[ 42 ]

Although the Maternal Death Notifi cation Form (MDNF) was modeled on the NYPORTS

short form for reporting sentinel events, hospital risk managers and quality improvement

personnel revealed concerns with utilization of the MDNF to report pregnancy-related

deaths to ACOG. While the information ACOG obtains from the MDNF is very similar

to information obtained from the NYPORTS form, the information transmitted to

ACOG is strictly confi dential and does not trigger a separate or independent review by the

Department of Health. Unfortunately, some hospitals have refused to participate in the

SMI for unfounded fear of DOH reprisal. Mechanisms should be implemented to modify

the MDNF and address these concerns.

Results and trends identifi ed in this report must be interpreted with caution since the data

included missing or unavailable information. Although the fi ndings obtained from the 33

maternal deaths reviewed by the SMI revealed useful information related to methods of

delivering obstetric care across the state, the quality and quantity of available information

was limited. SMI reviews discovered multiple challenges associated with availability, ac-

cess, and completeness of records and reports including: (1) the prenatal medical record; (2)

the medical record from previous hospitalizations and intrapartum care; (3) physician and

allied health care professional documentation; (4) the autopsy report at the time of review;

and (5) the death certifi cate.

Reporting of maternal deaths is inconsistent in New York State. Findings and recom-

mendations from the SMI will not be meaningful until there is additional, comprehensive

and long-term data available. A statewide system to report de-identifi ed maternal death

information to a central repository that incorporates an active (non-punitive) surveillance

system should be required. With accurate, long-term information, meaningful obstetric

recommendations can be made. ACOG’s participation in these reviews and its ability to

collect and secure confi dential information ensured that this review program had the sup-

port of the medical and health care community. ACOG members play a critical role in

maternal mortality review team discussions by interpreting information, explaining medi-

cal issues and identifying needed improvements while making clinical and practice changes

that can improve maternity care.

Th e SMI does not duplicate hospital or other peer review proceedings. SMI reviews are

confi dential and de-identifi ed cases are reported in the aggregate. Additionally, family,

economic, and psychosocial issues impacting the death are an important part of the review.

Th e SMI purpose is to improve maternity care by reviewing medical and systems issues

that contribute to maternal death, address gaps in care, and understand external issues that

impact outcomes.

Since the SMI’s focus was to review in-hospital pregnancy-related deaths, pregnancy-as-

sociated factors were not reported or reviewed. Consequently other contributing causes

including domestic violence, socioeconomic issues, homicide, suicide, vehicle accidents,

among other associated factors, were not identifi ed in the sampled cohort.28 An expansion

of the SMI activities to include the review of pregnancy-associated deaths will enable a

more comprehensive analysis of all maternal deaths.

Limitations

[ 43 ]


Chapter 7: Future directionsTh e lessons learned during the fi rst three years of the SMI, reported herein, should be

shared with maternal child health professionals who are in daily contact with women of

childbearing age, and those who have the opportunity to infl uence public and private poli-

cies related to the delivery of care to this population.

Providing consistent preconception and prenatal care to underserved women is vital to

preventing maternal mortality. New York State Health Commissioner Antonia C. Novello,

MD, MPH, DrPH, theorized that many immigrant women in the state are “frightened

of deportation.”29 Even women who are legal immigrants fail to get prenatal care in their

pregnancies. Th e CDC has found that failure to obtain prenatal care is a risk associated

with pregnancy-related death.4 Even with the limited number of reviewed cases, the SMI

data illustrates that continuous prenatal care was not necessarily received by all women

who died during pregnancy or after. In New York State, a more long-term analysis of all

maternal deaths is needed to ascertain the reasons behind gaps in the continuity of prenatal

care following an initial prenatal care visit.

In addition, standardized and centralized prenatal medical records would allow concurrent

access to patient health information, thus signifi cantly improving communication among

providers and improving the continuity of care.

Public health measures and education are encouraged. For example, education is needed

for health care providers and women regarding the risks associated with pregnancy. Such

education should begin with the importance of preconception care and end with care re-

quired during the immediate postpartum period. Education is also needed to underscore

the importance of identifying and treating pre-existing chronic diseases, such as hyperten-

sion and obesity, during the preconception period or in the fi rst trimester of pregnancy.

An expansion of the SMI to include “near miss” cases will further enhance knowledge of

chronic diseases and adverse outcomes related to pregnancy.

[ 44 ]

ACOG will continue to provide education to the NYS obstetric community regarding

postpartum hemorrhage, preconception counseling, pregnancy induced hypertension, and

cultural diversity. Other areas for further study and education include obesity and pregnan-

cy, and the prevention and management of embolism. Th e development of subcommittees

to address these and other identifi ed needs should be ongoing as information is collected

and trends are uncovered.

While the pregnancy-related deaths reviewed by the SMI identifi ed a variety of etiologies,

pregnancy outcomes, and system issues, some causes of death were undetermined due to

the absence of autopsy reports. Education regarding adverse pregnancy outcomes related to

maternal death will assist medical examiners and coroners to gain increased knowledge and

perspective. Th is, in time, will allow for more accurate diagnoses and improve the quality

and quantity of pregnancy-related autopsies in the state. Th e inclusion of medical examin-

ers and coroners as members of the review team will also provide insight into the factors

that may contribute to maternal death.

Improved maternal mortality case ascertainment and quality improvement is needed.

ACOG and the state should strive to enhance case identifi cation of pregnancy related

deaths by including record linkages between the Department of Health’s NYPORTS,

SPARCS and Vital Statistics’ death certifi cate data sets.

As NYS DOH and ACOG press for better case ascertainment, standardized data collec-

tion methods and instruments are needed. Resources will be needed to promote the SMI

as a model for replication by all hospitals with obstetric services while allowing fl exibility

to modify the general protocol as needed. ACOG strives for the attainment of uniform

data and a more long-term analysis of the intrinsic and external factors contributory to

pregnancy-related deaths.

Although RPCs are charged by the state to oversee quality improvement activities, con-

cerns regarding the amount of hospital eff ort required for each review need to be addressed

by the state. ACOG’s participation, as an objective third party, and the confi dential-

ity protections aff orded by the SMI, have facilitated hospital participation and enhanced

RPC-affi liate relationships.

In conclusion, the Safe Motherhood Initiative has been a vital, low cost investment that

promotes public health. Ongoing eff orts are needed to ensure women receive high quality

care regardless of race, socioeconomic status and/or geographic area. “Even one woman dy-

ing is too many women dying,” said Hani Atrash, MD, MPH, Chief, Pregnancy and Infant

Health at the National Center for Chronic Disease Prevention and Health Promotion.

Future Directions

[ 45 ]


References1 Department of Health, Scottish Executive Health Department, and Department of

Health, Social Services and Public Safety, Northern Ireland. Why Mothers Die. Sixth

Report on Confi dential Enquiries into Maternal Deaths in the United Kingdom, 2000-2002.

London: RCOG Press; 2004.

2 Centers for Disease Control and Prevention. Safe Motherhood Promoting Health for

Women Before, During, and After Pregnancy 2005. At a glance. 2005.

3 Callaghan WM, Berg CJ. Maternal mortality surveillance in the United States: Moving

into the 21st Century. JAMWA, 2002; 57:1–5.

4 Centers for Disease Control and Prevention. Surveillance Summaries. MMWR, No. SS-2,

2003:52; February 21, 2003.

5 Dye TD, Gordon H, Held B, Tolliver NJ, Holmes AP. Retrospective maternal mortality

case ascertainment in West Virginia, 1985 to 1989. American Journal of Obstetrics and

Gynecology. 1992; l67:72-76.

6 Centers for Disease Control and Prevention, Pregnancy-related mortality; Georgia. 1990-

1992. MMWR Morb Mortal Wkly Rep. l995; 44:93-96.

7 Atrash HK. Alexander S, Berg CJ. Maternal mortality in developed countries: Not just a

concern of the past, Obstet Gynecol. 1995; 86:700-705.

8 Kochanek KD, Murphy SL, Anderson RN, Scott C. Deaths: Final data for 2002. National

vital statistics reports; vol 53 no 5. Hyattsville, Maryland: National Center for Health

Statistics. 2004.

9 New York State Department of Health. 2002 Community Health Data Set. http://www.

health.state.ny.us/nysdoh/chac/cha/plots/matinf/death/mm65.htm

[ 46 ]

10 U.S. Department of Health and Human Services, Healthy People 2010, 2 Vols., With

Understanding and Improving Health and Objectives for Improving Health, 2nd ed., U.S.

Government Printing Offi ce, Washington, DC, November 2000.

11 Section 721.10 of Title 10 of NYCRR of the State of New York.

12 Applegate, M. Maternal Mortality Review in New York State. New York State Department

of Health. January 2000.

13 Berg C, Danel I, Atrash H, Zane S, Bartlett L (Editors). Strategies to reduce pregnancy

related deaths: from identifi cation and review to action. Centers for Disease Control and

Prevention, Atlanta 2001.

14 ACOG District II/ NY. Safe Motherhood Initiative: Response to reduce pregnancy-related

deaths. Final Report. October 2004

15 Peer review in obstetrics and gynecology by a national medical specialty society. Joint

Commission Journal on Quality and Patient Safety. 2003; vol 29, no. 2: 77-84.

16 Centers for Disease Control and Prevention, Pregnancy-related mortality surveillance,

United States, 1987-1990; MMWR Morb Mortal Wkly Rep. l997; 44:17-35.

17 Kotelchuck M. An evaluation of the Kessner Adequacy of Prenatal Care Index and a

proposed adequacy of prenatal care utilization index. Am J Public Health. 1994;84:1414-

20.

18 Th e American Academy of Pediatrics and Th e American College of Obstetricians and

Gynecologists. Guidelines for Perinatal Care. 5th ed., 2002.

19 Centers for Disease Control and Prevention. BMI - Body Mass Index: BMI Calculator.

http://www.cdc.gov/nccdphp/dnpa/bmi/calc-bmi.htm

20 Centers for Disease Control and Prevention, National Center for Health Statistics. Health

Conditions. http://www.cdc.gov/nchs/datawh/nchsdefs/healthcondition.htm#chronic

21 Florida Department of Health, Pregnancy-Associated Mortality Review (PAMR),

“PAMR Data Report 1999-2002.

22 Royal College of Obstetricians and Gynaecologists. Report of the RCOG Working Party

on Prophylaxis Against Th romboembolism in Gynaecology and Obstetrics. London: RCOG;

1995.

23 Royal College of Obstetricians and Gynaecologists. Th romboprophylaxis During Pregnancy,

Labour, and After Normal Vaginal Delivery. Guideline No. 37. London: RCOG; 2004.

24 Kristensen, J. et al. Pre-pregnancy weight and the risk of stillbirth and neonatal death.

BJOG 2005;112: 403-408.

References

[ 47 ]


25 Zhang, V.A. et al. Maternal pre-pregnancy overweight and obesity and the pattern of

labor progression in term nulliparous women. Obstet Gynecol; 104: 943-951.

26 March of Dimes. Professional and Researchers Continuing Education. http://www.

marchofdimes.com/professionals/682_15721.asp

27 Dorfman SF. Maternal mortality in New York City, 1981-1983. Obstet Gynecol 1990;76:

317-23

28 Dannenberg AL, Carter DM, Lawson HW, Ashton DM, Dorfman SF, Graham EH.

Homicide and other injuries as causes of maternal death in New York City, 1987 through

1991. Am J Obstet Gynecol. 1995;172: 1557-64

29 E Dozien, New York Post. January 17, 2003

American College ofObstetricians and GynecologistsDistrict II / NY152 Washington AvenueAlbany, New York 12210Phone: 518-436-3461Fax: 518-426-4728www.acogny.orgemail: [email protected]

Individual copies of this publication may be obtained by calling the American College of Obstetricians and

Gynecologists, District II/NY at 518-436-3461. It is also available on the internet at www.acogny.org

New York StateDepartment of HealthBureau of Women’s HealthEmpire State PlazaCorning Tower, Room 1805Albany, New York 12237Phone: 518-474-1911Fax: 518-474-7054www.health.state.ny.us

safe motherhood - cmqcc

Documents