JIlilllllllllllllllllllllllllllllll'llll L

3 1293 01020 1691

This is to certify that the

thesis entitled

PRENATAL CARE:
A COMPARATIVE ANALYSIS OF INFANT OUTCOMES

presented by

Myrth Crystal Condon

has been accepted towards fulfillment
of the requirements for

Master of Science Nursing

degree in

 

 

him

Major professor

 

Date ——1—1-/—1_4..LQ.4—_

0-7 639 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

LIBRARY
Mlchigan State
University

 

 

 

Puceuaswnu BOXtomnovothbdnckMMywmd.
TOAVOIDFINESMunonorbdmmm

I DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU I. An mm mm Opportunity lmwon
mm:

 

PRENATAL CARE: A COMPARATIVE ANALYSIS OF INFANT OUTCOMES

BY
Myrth Crystal Condon

AN ABSTRACT OF A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE IN NURSING

College of Nursing
1994

Carla L. Barnes

ABSTRACT
PRENATAL CARE: A COMPARATIVE ANALYSIS OF INFANT OUTCOMES
BY
Myrth Crystal Condon

This ex post facto study compared infant outcomes
of prenatal care provided by three different types of
health care providers: certified nurse midwives,
obstetricians, and general practitioners, to women with
either private insurance or Medicaid. Infant outcomes
were birth weight, gestational age, and admission into
neonatal intensive care for more than 24 hours. The
data were from 132 infant charts from a southern Michigan
city. Data analysis included chi square test, analysis
of variance, and logistic regression. Results found no
statistically significant differences for infant outcomes
of birth weight and rate of admissions. Gestational age
was found to be significant for type of provider (p=.01).
Clinically speaking, the adjusted means for gestational
age were all term births and were not considered
substantively different. Therefore, the findings
indicated that infant outcomes were the same regardless

of the type of provider.

TABLE OF CONTENTS

List of Tables . . . . . . . . . .
List of Figures. . . . . . . . . .
Introduction. . . . . . . . . . .
Background of the problem . . . . .
Statement of the problem . . . . .
Research question . . . . . . . .
Literature Review . . . . . . . . .
Prenatal care providers . . . . . .
Type of insurance . . . . . . . .
Type of provider and type of insurance .
Summary of type of provider and insurance
Infant outcomes . . . . . . . .
Research findings . . . . . . . .
Summary of infant outcomes . . . . .
Theoretical framework . . . . . .
Theoretical definitions of variables .
Methods . . . . . . . . . . . .
Research design . . . . . . . .
Sample . . . . . . . . . . .
Data collection . . . . . . . .

Protection of human subjects . . . .

iii

10

13

16

16

17

18

19

TABLE OF CONTENTS CONTINUED

Operational definition of variables

Demographic and background data

Assumptions . . . . .
Results . . . . . . .

Demographic variables .

Categorical demographic variables

Outcome variables . . .
Research questions . .
Findings . . . . . .
Discussion . . . . . .
Limitations to the study
Implications . . . .

Implications for advanced

Recommendations for further research

Appendix
UCRIHS approval letter .
List of references

References . . . . .

iv

nursing practice

19

21

22

23

23

24

26

32

33

34

34

35

36

39

42

43

LIST OF TABLES

Table 1 Demographic Variables . . . . . . . . . .
Table 2 Categorical Demographic Variables . . . .
Table 3 Birth Weight. . . . . . . . . . . . . . .
Table 4 Gestational Age . . . . . . . . . . . . .

Table 5 Admission into Neonatal Intensive Care
for More Than 24 Hours. . . . . . . . . .

Table 6 Analysis of Covariance for Birth Weight .
Table 7 Analysis of Covariance for Gestational Age
Table 8 Chi Square Test for Admissions. . . . . .

Table 9 Logistic Regression for Admissions. . . .

Figure 1

Figure 2

LIST OF FIGURES

Theoretical Model for Certified Nurse Midwives
Based on King . . . . . . . . . . . . . . . 12

Theoretical Model for Physician Providers
Based on the Traditional Medical Model . . 12

vi

Introduction

As our country currently undergoes health care
reform, there is a shortage of primary health care
providers. Many feel that this shortage should be
solved through the use of advanced practice nurses as
primary care providers (Edmunds, 1994; Pearson, 1994).
In fact, it has been calculated that as many as 60 to
80 percent of the activities currently performed by
physicians could be performed by advanced practice
nurses (Capan, Beard, & Mashburn, 1993; Pearson, 1994).

The term, advanced practice nurses, includes nurse
practitioners, clinical nurse specialists, certified
registered nurse anesthetists, and certified nurse
midwives (Pearson, 1994). This study focused on one
group of advanced practice nurses, certified nurse
midwives. Certified nurse midwives provided many
primary care services for women, but the focus for this
study was on prenatal care. Comparisons were made
between certified nurse midwives and other groups of
health care providers with respect to some infant

outcomes of this prenatal care.

Background of the problem

In 1984, access to prenatal care in a southern
Michigan city was compromised due to the closing of
several prenatal offices brought about by an exodus of
prenatal care providers who had either retired or left
practices. Economic factors, particularly increased
liability insurance and inadequate third party
reimbursement were the major reasons for closing practices
(Calhoun County Board of Health, 1991). In 1986, a
center for primary care was established to provide
access to prenatal care. The center used a
multidisciplinary team approach. In 1992, the center
was staffed by several certified nurse midwives, one
obstetrician who provided coverage for high risk cases
such as.cesarean births, nurses, social workers, and a
nutritionist. This city also had other obstetricians
and general practitioners who provided prenatal care,
and for the past six years, the prenatal care needs
have been met by three different types of providers:
certified nurse midwives, obstetricians, and general
practitioners. There have been no documented
comparisons of the infant outcomes of prenatal care
provided by these three groups.

e r em
The purpose of this study was to explore and compare

the infant outcomes of prenatal care provided in this

3

southern Michigan city by the three different types of
providers: certified nurse midwives, obstetricians, and
general practitioners, to women with either private
insurance or Medicaid. The comparison focused on the
infant outcome variables of birth weight, gestational age,
and admission into neonatal intensive care (NICU) for
more than 24 hours.
Researeh question

The aim of this study was to explore the following
questions: (1) Are the infant outcomes similar or
different across the three different types of providers?
(2) Are the infant outcomes similar or different across
the two different types of insurances?

Literature Review

Pr nat c e r v'de s

The literature review was limited to studies
involving types of prenatal care providers. Over the
past two decades, retrospective and prospective studies
have found that nurse midwives provide prenatal care
comparable to that of physicians as measured by infant
outcomes, mainly birth weight (Buhler, Glick, & Sheps,
1988; Chambliss, Daly, Medearis, Ames, Kayne, & Paul, 1992;
Dillon, Brennan, Dwyer, Risk, Sear, Dawson, & Wiele, 1978:
Heins, Nance, McCarthy, & Efird, 1990: McLaughlin,
Altemeier, Christensen, Sherrod, Dietrich, & Stern,

1992; Slome, Wetherbee, Daly, Christensen, Meglen, &

4

Thiede, 1976). Many of the studies involved nurse midwives
who worked in settings that offered a more comprehensive
prenatal care program than the standard "medical model"
of prenatal care (Haire & Elsberry, 1991; Heins et al.,
1991; McLaughlin et al., 1992; Neeson, Patterson, Mercer,
& May, 1983; Piechnik & Corbett, 1985). Services offered
in the nurse midwife affiliated comprehensive prenatal
care programs included child birth education (Haire &
Elsberry, 1991), nutritional counseling, activity
counseling, substance abuse counseling, and stress
reduction (Heins et al., 1990). Many of these settings
employed a team approach in providing the prenatal care
(Neeson et al., 1983; Piechnik & Corbett, 1985).

Five of these studies compared nurse midwives to
obstetricians and found no significant differences in
the infant outcome of birth weight (Dillon et al., 1978;
Heins et al., 1990; McLaughlin et al., 1992: Neeson et
al., 1983). Only one study compared nurse midwives to
general practitioners and found that the certified nurse
midwives provided better prenatal care as determined by
maternal and infant outcomes (Buhler, Glick, & Sheps,
1988). However, the study has limitations because it
based its findings on how the certified nurse midwives
and general practitioners documented their care, possibly
revealing that the midwives documented care given better

than the general practitioners. Two studies failed to

5
specify what type of physician was involved in the
comparison and found no significant differences in infant
outcomes of birth weight (Chambliss et a1, 1992; Slome
et al., 1976). None of the studies reviewed compared
all three prenatal care providers.
Type ef insurance

Few of the studies reviewed actually compared type
of insurance with infant outcomes. One study compared
three groups of women: those with Medicaid in a managed
care health plan, those with Medicaid in a fee for
service group, and those with private insurance in a
managed care health plan (Krieger, Connell, & LoGerfo,
1992). This study found that within the managed care
health plan, the women funded by Medicaid showed poorer
infant outcomes (birth weight) than the women with
private insurance (Krieger, Connell, & LoGerfo, 1992).
However, both Medicaid funded groups were similar in
infant outcomes (birth weight) (Krieger, Connell, &
LoGerfo, 1992).

Another study compared pregnancy outcomes of women
who received prenatal care funded through a health
maintenance organization (HMO) to women of the general
population (Quick, Greenlick, & Roghmann, 1981). The
findings suggested that for the geographical area
studied, the infant outcomes (birth weight, prematurity,

and infant mortality) for the HMO group were similar to

6
the general population; however, there was no mention
of the type of insurance funding for the general
population (Quick, Greenlick, & Roghmann, 1981).

Two other studies mentioned type of insurance, but
only the fact that it was held constant in one study
(Hulsey, Patrick, Alexander, & Ebeling, 1991) and was
used in matching in the other study (Piechnik & Corbett,
1985). Neither study compared type of insurance, nor
considered type of insurance when examining infant
outcomes.

Type of insurance was an indicator of socioeconomic
status. Women who qualified for Medicaid funding usually
had a lower socioeconomic status than women who were
privately insured. The fact that a woman was funded by
Medicaid may have influenced the infant outcomes, because
there are variables associated with lower socioeconomic
status such as potential for poor nutrition, poor weight
gain, increased stress, and low educational level of
the woman (Piechnik & Corbett, 1985).

Since a true measure of socioeconomic status was
beyond the scope of this study, this study included
type of insurance as a proxy variable for socioeconomic
status. Infant outcomes of each of the three prenatal
care providers were compared within the separate
insurance groups: the Medicaid group and the privately

insured group.

7
Type of pgovider and type pf insurance

None of the studies reviewed made any comparisons
between the type of provider and type of insurance. In
one study comparing Medicaid to private insurance, the
only type of prenatal care providers were obstetricians,
thereby providing no comparison (Quick, Greenlick, &
Roghmann, 1981). One other study that compared type of
insurance failed to mention the type of prenatal care
provider (Krieger, Connell, & LoGerfo, 1992).

Two studies mentioned that the majority of the
women who sought prenatal care were funded by Medicaid.
One study involved both certified nurse midwives and
obstetricians as providers (Haire & Elsberry, 1991),
while the other study involved only certified nurse
midwives as providers (Cavero, Fullerton, & Bartlome,
1991). However, neither study made a comparison of
type of insurance nor related it to infant outcomes.
Summary of type ef ppovider and insppence

Of the studies reviewed, none compared all three
types of prenatal care providers: certified nurse
midwives, obstetricians, and general practitioners.

Few of the studies made an attempt to compare type

of insurance (Krieger, Connell, & LoGerfo, 1992; Quick,
Greenlick, & Roghmann, 1981). None of the studies
reviewed compared both type of provider and type of

insurance and examined the subsequent infant outcomes.

8
This study examined both type of provider and type of
insurance and examined and compared the subsequent
infant outcomes.

Most of the reviewed studies were conducted in
facilities that were located in either the eastern states
(Baruffi, Dellinger, Stobino, Rudolph, Timmons, & Ross,
1984; Dillon et a1, 1978; Gann, Nghiem, & Warner, 1989:
Gortmaker, 1979; Haire & Elsberry, 1991) or western
states (Cavero, Fullerton, & Bartlome, 1991; Krieger,
Connell, & LoGerfo, 1992; Neeson et a1, 1983; Quick,
Greenlick, & Roghmann, 1981) or coastal states or
southern states (Heins et a1, 1990; McLaughlin et al,
1992; Piechnik & Corbett, 1985; Slome et a1, 1976).
None of the studies were based in the midwestern states.
Since regional differences in health care exist, this
study, conducted in southern Michigan, provides some
information applicable to the northern midwestern part
of the United States.

ut es

There have been many different documented infant
outcome variables listed in the literature. Of the
studies reviewed, the following outcomes were mentioned:
(1) Apgar scoring (Baruffi et al., 1984; Dillon et al.,
1978; Haire & Elsberry, 1991; Neeson et al., 1983; Slome
et al., 1976), (2) infant mortality (Baruffi et al.,

1984; Cavero, Fullerton, & Bartlome, 1991; Gortmaker,

9

1979; Haire & Elsberry, 1991), (3) stillborns (Haire &
Elsberry, 1991), (4) presence of anomalies or birth
injuries (Slome et al., 1976), (5) fetal distress at
delivery (Baruffi et al., 1984; Slome et al., 1976), (6)
recumbent length and head circumferences (Slome et al.,
1976), (7) length of infant hospital stay (Baruffi et al,
1984; Neeson et al., 1983), (8) admission into neonatal
intensive care (Haire & Elsberry, 1991; Neeson et al.,
1983), (9) gestational age measurements (Hulsey et al.,
1991; Neeson et al., 1983; Slome et al., 1976), and (10)
birth weight (Baruffi et al., 1984; Cavero, Fullerton,
& Bartlome, 1991; Dillon et al., 1978; Gann, Nghiem, &
Warner, 1989; Gortmaker, 1979; Haire & Elsberry, 1991;
Hulsey et al., 1991; McLaughlin et al., 1992; Neeson et
al., 1983; Piechnik & Corbett, 1985; Slome et al., 1976).
Research findings

Of the studies reviewed, the major findings
indicated that adequate levels of prenatal care provided
by both certified nurse midwives and physicians resulted
in lower incidences of low birth weight and prematurity.
Prenatal care was not significantly related to length
of infant stay, admission into a NICU, or a diagnosis
other than healthy newborn.
Snnnary of infant outcomes

Based on this literature review, birth weight and

gestational age were used often and were reliable outcome

10
measures of prenatal care and were included in this
study. The measures of length of infant hospital stay
and admission into a NICU were often used as indicators
of infant wellbeing after delivery (Neeson et al, 1983).
However, for this study these two measures were combined
as admission into a NICU for more than 24 hours. The
assumption was made, that if an infant was admitted to
a NICU for more than 24 hours, the infant’s health was
compromised in some way.
Tneopetical framework

Different types of health care providers often
vary in their approach to providing care. For this
reason, King’s Theory of Goal Attainment (King, 1981)
was used as the theoretical framework for the certified
nurse midwife providers. The traditional Medical Model
of care served as the framework for the obstetrician and
general practitioner providers.

According to King (1981), the Theory of Goal
attainment provides a description of the interactions
between nurse and client to attain mutually developed
goals. One commonly held goal is health. For this
study, the goal to be attained was a healthy infant.
Measures to determine a healthy infant were birth weight,
gestational age, and admission into a NICU for more than
24 hours.

The Theory of Goal Attainment is based on open

11
systems framework. The major focus in this theory is
on the interpersonal system which encompasses the
interactions between nurse and client within the health
care system. As depicted in Figure 1, the concept of
nurse represented the certified nurse midwife providers.
The concept of client was represented by the pregnant
woman who received prenatal care. The concept of the
client's socioeconomic status was represented by the
women’s type of insurance.

The nurse midwife provider and the client interacted
through the actions and activities that constituted
prenatal care. This interaction was represented in the
model by arrows from both nurse and client to the
interaction. This interaction between the type of
provider and client then led to the attainment of the
goal, a healthy infant.

The focus of the traditional Medical Model is the
identification of medical problems and subsequent
treatment (diagnosis--treatment--cure). In using the
Medical Model for this study, as depicted in Figure 2,
the concept of physician was represented by the general
practitioner and obstetrician providers. The concept of
patient referred to the pregnant woman who was the
receiver of prenatal care. This concept of patient also
included the concept of the woman’s socioeconomic status

as represented by the woman’s type of insurance.

12

 

(nurse)

 

Type of Provider
—Certified Nurse Midwives

 

 

 

 

InteractionL
LfiF___r

 

 

 

(client)

 

Type of Client by
Insurance Status
-Private
-Medicaid

 

 

 

 

(goal) = Infant Outcomes
-Birth weight
-Gestational age
-Admission into neonatal
ICU fornorem

 

 

 

Figure 1

THEORETICAL MODEL
FOR CERTIFIED NURSE MIDWIVES
BASED ON KING

 

(physician) =

 

Type of Provider
-Obstetrician
-General Practitioner

 

 

 

 

 

 

(patient) = Type of Client by Medical Problem
Insurance Status Diagnosis

-Private
-Medicaid

 

 

Treatment

 

 

 

 

 

 

(cure of medical
problem)

 

= Infant Outcomes
-Birth Weight
-Gestational Age
-Admission into neonatal
ICU for more than 24 hours,

Figure 2

THEORETICAL MODEL
FOR PHYSICIAN PROVIDERS
BASED ON THE TRADITIONAL MEDICAL MODEL

 

13

The concept of medical problem was represented by
the woman’s pregnancy. The concepts of diagnosis and
treatment constituted prenatal care as provided by the
physician providers. The arrows in the model dictated
that physicians were the directors of prenatal care for
the patient. The one way arrow from physician to the
medical problem, diagnosis, and treatment indicated that
the physicians were in command of the prenatal care
process. The one way arrow from medical problem,
diagnosis and treatment to patient represented a lack
of mutual interaction between physician and patient.
The arrow from patient points toward the cure of the
medical problem. The concept of cure of the medical
problem was represented by the infant outcomes. Thus
both the models, one based on King's theory and one
based on the Medical Model, had the same objective of
healthy infant outcomes.
Theoretical definitien ef variables

The variable of type of provider was defined as
one who has met certain educational requirements, as
set by law, and is licensed and/or certified to practice
in Michigan as a health care provider. Further, health
care was defined as actions taken by the provider in
order to restore, maintain, or promote a state of
wellness within the client. The client, in turn seeks

health care from a provider. The types of providers

14
for this study were certified nurse midwives,
obstetricians, and general practitioners.

For this study there were some commonalities that
existed among the three types of providers: (1) all
had privileges at the only hospital in the community
that offered obstetrical care; (2) all three groups of
providers saw both privately insured and Medicaid funded
women, although the ratio of Medicaid to private insurance
may have differed between individual providers; (3) all
used the same Hollister prenatal charting forms for
gathering history data, physical assessments, and lab
tests; (4) all referred their women to the same laboratory
for any required testing; (5) all adhered to the same
initial prenatal lab screening, tested for the presence
of beta strep, and followed other well established prenatal
protocols; and (6) all three types of providers also
managed women from low to moderate risk pregnancies.

For this study there was a regional high risk
perinatal center located in a nearby city where referrals
may be made. An assumption was made that any high risk
client was referred to this center. There were some
limitations to the scope of practice during the delivery
process for the nurse midwives and general practitioners.
For example, if a cesarean section were needed, both nurse
midwives and general practitioners would refer the

client to an obstetrician for the delivery. Since the

15
focus of this study was on the infant outcomes of prenatal
care and not the delivery process, the limitation to
the scope of practice for two types of providers should
be negligible for the infant outcome variables of birth
weight and gestational age. However, the mode of
delivery may have had some effect on the infant outcome
of admission into a NICU for more than 24 hours, and
this may have been a limitation.

The major differences among the three types of
providers were their educational backgrounds and past
experiences which may have influenced the manner in
which they provided prenatal care. Also different
approaches to providing prenatal care may have occurred
for different types of providers. For example, the
nurse midwives at this city’s center provided prenatal
care using a multidisciplinary team approach to offer an
array of services. The physician providers may have
offered these same services but did not use a team
approach.

The variable of type of insurance was a proxy
variable for socioeconomic status and can be defined
in terms of a client’s financial resources to obtain
health care. A client who had private insurance was
thought to have a higher socioeconomic status than a
client who had Medicaid. In order to receive Medicaid

certain requirements must have been met that pertained to

16
the number of persons in the family and the income level.

In this southern Michigan city the certified nurse
midwives, obstetricians, and general practitioners
provided care to women who were insured either privately
or through Medicaid. However, the percentage of
privately insured to Medicaid insured varied greatly
among the three types of providers. The percentage of
women with Medicaid was 92% of the clientele for the
nurse midwives, 31% for the obstetricians, and 30% for
the general practitioners.

The variable of infant outcomes referred to the
wellbeing or health of an infant upon birth, as a result
of or the consequence of the gestation. The infant’s
wellbeing was assessed in terms of birth weight,
gestational age, and a need for admission into a NICU
for more than 24 hours.

Methods

Research design

The study’s design was ex post facto. The effect of
the two independent variables of type of provider (three)
and type of insurance (two) were compared. This resulted
in a three by two design with six cells. The subsequent
dependent variable of infant outcomes was examined and
compared across the type of providers and the type of
insurance. The data were analyzed using the computer

software package SPSS/pc. Descriptive statistics, such

17

as cross tabs, histograms, measures of central tendency,
and frequencies were used to describe the sample and
infant outcomes. Inferential statistics, such as chi
square tests, were used to evaluate the research
questions. After descriptive statistics were completed
for birth weight and gestational age, these data were
collapsed into categories for use with chi square
statistics.
Sample

The population for this study consisted of the
medical records of infants born within this southern
Michigan city during 1992 since this was the last calendar
year before a major change in staff occurred at the
Women’s Center. A birth record was kept in the hospital’s
nursery. This record listed every live birth along with
the following data: birth date, infant name, medical
record number, type of health care provider, and type of
insurance. The birth record was used to select the sample
of infant charts. Stratified random sample selection
was used. Selection criteria were: type of provider
(i.e. certified nurse midwife, obstetrician, or general
practitioner), type of insurance (i.e. private or
Medicaid), and live birth. The birth record did not
include data on stillborns, fetal deaths, or neonatal
deaths; thus, only live births were included in this

study. Also women who self pay, versus having private

18
insurance or Medicaid, were not included because of the
possible diverse reasons for self payment.

The birth record was used to select the sample of
infant charts according to the two independent variables
of type of provider and type of insurance. If the birth
record had missing data, such as medical record numbers
and type of insurance, these charts were excluded from
the study. These missing data decreased the number of
available charts for the sample selection. The cell for
privately insured women who received prenatal care from
nurse midwives had the fewest number of available charts.
This cell contained 22 infant charts. Consequently 22
charts were then chosen to fill the other five cases by
stratified random selection according to type of provider
and insurance. Each of the cells contained 22 charts for
a total of 132 charts in the sample.

Deta eelleetion

The sample of charts was collected from the
time period beginning January 1, 1992, and ending
December 31, 1992. Data collection involved chart
reviews of infant charts. The charts were obtained from
the medical records department at the only delivering
hospital in the community. All data collection was done
by the principal investigator, who was familiar with the
charting forms used at this facility and had approval

to access the charts granted before data collection

19
occurred. All data were entered onto data collection
forms.
Preteetion of human subjects

After the birth record was used for the selection
of subjects, study numbers were assigned to each subject
and the subject's name was deleted to maintain
confidentiality. The medical record numbers were used
to obtain the charts from medical records. After all
the data from each chart were collected, each medical
record number was also deleted to prevent data being
traced back to the mother or infant. The only person
with access to the data before the assignment of study
numbers was the principal investigator.

Before any data collection was begun, approval
was granted by the Michigan State University Committee
on Research Involving Human Subjects (see Appendix).
Also, approval was granted by the Vice President of
Nursing in the southern Michigan city’s hospital.
Qpepatienai definition of variabiee

The first independent variable was type of prenatal
care provider. This was defined as either certified
nurse midwife, obstetrician, or general practitioner.
This variable was identified using the birth record
during sample selection. The provider was defined as
the person who provided the prenatal care, and was not

necessarily the person who delivered the infant.

20

The second independent variable was the type of
insurance and was also identified using the birth record
during sample selection. Type of insurance was either
private insurance or Medicaid insurance. In the state
of Michigan, eligibility requirements to receive Medicaid
are based upon a woman’s income level, how many people
are in the family, and the family’s total income.

During 1994, eligibility for Medicaid coverage for
pregnant women was 185% of the Federal Poverty Guideline
(National Governors' Association, 1994).

The following were the primary outcome variables
or dependent variables: infant birth weight, gestational
age, and admission into a NICU for more than 24 hours.
Infant birth weight was measured in grams. Gestational
age was measured in weeks at birth as determined by
Dubowitz assessment (Reeder & Martin, 1987). Both birth
weight and gestational age were found on the Dubowitz
assessment form in the infant's chart.

Birth weight was collapsed into the categories of
very low birth weight (VLBW), low birth weight (LBW),
normal birth weight (NEW), and high birth weight (HBW).
VLBW was defined as an infant weighing less than or equal
to 1500 grams, LBW was defined as an infant weighing more
than 1500 grams and less than or equal to 2500 grams,

NBW was defined as an infant weighing more than 2500 grams

and less than or equal to 4000 grams, HBW was defined as

21
more than 4000 grams (Klaus & Fanaroff, 1986; Korones,
1986). Data on gestational age were collapsed into
categories of preterm, term, and postterm. Preterm
had been defined as gestation of less than or equal to
37 weeks, term was 38 to 41 weeks gestation, and
postterm was more than 41 weeks (Merenstein & Gardner,
1993).

Admission into a NICU for more than 24 hours was
determined by counting the number of hours up to 25.
This information was found in Special Care Nursery
nurses’ notes by counting the number of hours the infant
spent in the nursery. If there were no Special Care
Nursery nurses’ notes on the infant's chart, the infant
was never admitted into the NICU.

emo ra i n bac round data

The following demographics were included to help
describe the sample population: amount of prenatal care
received by the women, admitted smoking, alcohol, and
drug use by the women, woman’s age, race, marital status,
gravida, para, and mode of delivery.

The amount of prenatal care was determined by the
months gestation at the first prenatal care visit
according to dates. The number of visits were counted
for the first two trimesters. These data were found on

the woman’s prenatal forms located in the infant's chart.

22

Admitted smoking, alcohol, and drug use were coded
as yes, no, or missing data. The woman’s age was
measured in years. The woman's race was either
Caucasian, Black, Hispanic, Asian, Native American,
other, or missing data. Marital status was either
single, married, divorced, separated, widowed, or missing
data. Gravida was determined by counting the number of
pregnancies including the present one. Para was
determined by counting the number of past pregnancies
that have produced an infant of viable age, commonly
defined as 20 weeks gestation (Reeder & Martin, 1987).
All of these data were found in the woman’s prenatal
record in the infant's chart. The mode of delivery was
either vaginal or by caesarean section which was found
on the infant’s birth record.
Aeeumptions

Three assumptions were made for this study. First,
all high risk pregnancies were referred to the regional
high risk perinatal center and were not delivered in
this southern Michigan city. Second, an infant admitted
into a NICU for more than 24 hours was compromised in
some way with regards to health. And last, prenatal
care visitation patterns established in the first two

trimesters would be maintained during the last trimester.

23
Results

Demegrephic yarieples

The following variables are shown in Table 1. The
ages of women ranged from 15 to 37 years old with a mean
age of 23.8 years old with a fairly normal distribution.
The month of gestation in which women had their first
prenatal care visit ranged from one month to eight months
gestation with a mean of 2.8 months. Only one woman
received her first visit at eight months and one at seven
months gestation. The number of prenatal visits ranged
from as few as 5 to as many as 30. The mean number of
visits was 17 and the mode was 20. The mean gravida
was 2.2 and the mean para was 0.8. In this sample, 36.4%
of the women were primiparous, 29.5% were gravid II, 18.2%
were gravid III, 9.1% were gravid IV, 4.5% were gravid V,

1.5% were gravid VI, and 0.8% were gravid VII.

Table 1

Demographic Variables

(n = 132)
Veziable Mean SD Range Skewness Knrtoeis
Maternal Age 23.8 5.097 15-37 0.433 -0.535
Months
Gestation 2.8 1.275 1- 8 1.634 2.643
Number of
Visits 17.0 4.671 5-30 -0.818 0.072
Gravida 2.2 1.307 1- 7 1.138 1.053

Para 0.8 0.983 0- 5 1.336 2.062

24

Categorical demognaphic variables

The following variables are shown in Table 2. During
pregnancy 30.3% of these women admitted to smoking, 3.8%
admitted to alcohol use, and 3% to drug use. There was
an 84.8% rate of vaginal deliveries and a 15.2% caesarean
birth rate. Of these women, 78.8% were Caucasian, 18.2%
were Black, and 3% were Hispanic. At the time of delivery,
47.7% were married, 46.2% were single (not previously

married), 3.8% were divorced, and 2.3% were separated.

Table 2

Categorical Demographic Variables

 

(n = 132)
Variables Percentage Numper
Maternal Behaviors
Smoking 30.3 40
Alcohol 3.8 5
Drugs 3.0 4
Mode of Delivery
Vaginal 84.8 112
Cesarean 15.2 20
Race
Caucasian 78.8 104
Black . 18.2 24
Hispanic 3.0 4
Marital Status
Single 46.2 61
Married 47.7 63
Divorced 3.8 5
Separated 2.3 3

All of these demographic variables were potential

confounding variables affecting the three dependent

25
infant outcome variables. Thus, statistical tests were
performed to see if cases in the six cells defined by
type of provider and insurance systematically differed
from each other in demographics. A chi square test was
done for the following categorical demographic variables:
smoking, alcohol use, drug use, gravida, para, race,
marital status, and mode of delivery. An analysis of
variance test was done for the following demographic
variables: maternal age, months gestation at the first
prenatal care visit, and the number of prenatal care
visits. Both the chi square and analysis of variance
tests were performed separately for each independent
variable of type of provider and type of insurance.

The variable race showed significant differences
for both type of provider (p=.003) and insurance
(p=.001). The nurse midwife group was comprised of 64%
Caucasian, 32% Black, and 4% Hispanic women; the
obstetrician group was comprised of 75% Caucasian, 21%
Black, and 2% Hispanic women; and the general practitioner
group was comprised of 98% Caucasian, and 2% Black women.
0f the privately insured women, 86% were Caucasian, 8%
were Black and 6% were Hispanic; while the Medicaid
funded group of women consisted of 71% Caucasian and
29% Black. The variables of marital status (p=.000),
maternal age (p=.000), and months gestation at the first

prenatal care visit (p=.013) also produced significant

26
differences, but only for type of insurance. Of the
privately insured women, 76% were married, 20% were
single, 1% were divorced, and 3% were separated: while
for the Medicaid funded women, 20% were married, 73%
werbleingle, 6% were divorced, and 1% were separated.
Therefore, the four demographic variables of race,
marital status, maternal age, and months gestation at
the first prenatal care visit were used as control
variables in the evaluation of the effect of provider
and insurance status on the three dependent infant outcome
variables of birth weight, gestational age, and admission
into a NICU for more than 24 hours.
02W

The birth weight of infants ranged from 1588 gm to
5131 gm with a mean weight of 3413 gm with a normal
distribution (see Table 3). The gestational age ranged
from 31 weeks to 42 weeks with a mean age of 39.5 weeks
and a negatively skewed distribution (see Table 4). For
this sample, 10.6% of the infants were admitted to the
NICU for more than 24 hours (see Table 5).

It had been planned that the data for both birth
weight and gestational age would be collapsed into the
categories of very low birth weight, low birth weight,
normal birth weight, and high birth weight, and
premature, term, and postterm respectively. However,

when the data were collapsed, there were no cases of

27
Table 3

Birth Weight

 

 

 

 

Mean (grams) SD
Private Insurance 3518 585.935
CNM 3512 519.771
OB 3440 562.979
GP 3602 679.525
Medicaid Insurance 3308 502.955
CNM 3263 345.350
OB 3187 629.093
GP 3475 470.393
Sample Population 3413 554.051
Sample Population Range 1588 - 5131
Skewness 0.416
Kurtosis 1.377
Table 4
Gestational Age
Mean (weeks) SD
Private Insurance 39.5 1.491
CNM 39.5 1.143
OB 39.0 2.011
GP 40.0 1.046
Medicaid Insurance 39.5 1.218
CNM 39.5 1.336
OB 39.4 1.136
GP 40.0 1.203
Sample Population 39.5 1.356
Sample Population Range 31.0 - 42.0
Skewness -2.285
Kurtosis 11.830

28
Table 5

Admission into Neonatal Intensive Care
for More Than 24 Hours

 

Percentage Number
Private Insurance 7.6 5
CNM 13.6 3
OB 4.5 1
GP 4.5 1
Medicaid Insurance 13.6 9
CNM 27.3 6
OB 9.1 2
GP 4.5 1
Sample Population 10.6 14

very low birth weight, three cases of low birth weight,
five cases of prematurity and five cases of postterm.
With so few cases it would be impossible to test
statistically for the presence of significant differences.
Therefore, the data for birth weight and gestational age
were not collapsed but were left as originally measured
in grams and weeks, respectively.

An analysis of covariance was done to assess the
effects of the two independent variables (type of
provider and type of insurance) on both mean birth weight
and mean gestational age. This analysis of covariance
also included the four control demographic variables
(race, marital status, maternal age, and months gestation
at the first prenatal care visit). For these analyses
of covariance the factor or categorical variables were

race and marital status, and the co-variate variables

 

29
were maternal age and months gestation at the first
prenatal care visit.

As shown in Table 6, no statistically significant
differences were found for the infant outcome variable
of birth weight, for type of provider, type of insurance,
marital status, race, maternal age, or months gestation

at the first prenatal care visit.

Table 6

Analysis of Covariance for Birth Weight

Senpce ef yariation Sum of Sgnapes DF F-Spa; Sig

Main Effects 2675727.681 7 1.273 .269
Provider 878647.455 2 1.463 .236
Insurance 275947.104 1 .919 .340
Covariates 1572003.789 2 2.618 .077
Race 765748.312 2 1.275 .283
Marital Status 80095.084 2 .133 .875
Maternal Age 801224.218 1 2.669 .105
Months Gestation 523229.034 1 1.743 -.189

As shown in Table 7, there were statistically significant
differences for the infant outcome variable of gestational
age. The control variable of months gestation at the

first prenatal visit was significant as was the independent
variable of type of provider. Since the focus of this
study was the effects of the independent variables of

type of insurance and type of provider, this discussion
will follow through with the variable of type of provider.
After adjusting for the effects of the covariates,

gestational age for the certified nurse midwife group

30
was 39.6 weeks. The adjusted mean gestational age for
the obstetrician group was 38.76 weeks, and for the

general practitioner group it was 40.23 weeks.

Table 7

Analysis of Covariance for Gestational Age

Sonrce of Variation Sum of Sgnapes DF F-Stat Sig

Main Effects 16.053 7 1.307 .253
Provider 12.0936 2 3.687 .028
Insurance .007 1 .004 .950
Covariates 13.185 2 3.758 .026
Race 5.121 2 1.460 .236
Marital Status .049 2 .014 .986
Maternal Age 4.454 1 2.538 .114
Months Gestation 10.592 1 6.037 .015

Therefore, the independent variable of type of
provider had some statistical effect on gestational age,
with infants delivered by general practitioners having
the longest average gestational age. However, from a
clinical standpoint, all three adjusted means were
considered term gestation by definition. Thus these
statistical differences should not be considered to have
clinical significance.

As shown in Table 8, a chi-square test was performed
to assess the effect of the two independent variables of
type of provider and type of insurance on the infant
outcome variable of admission into a NICU for more than
24 hours. There were no statistically significant

differences found for the type of insurance. However,

31
there were statistically significant differences for
type of provider. The percentages of admissions were
20.5% for the midwife group, 6.8% for the obstetrician

group, and 4.5% for the general practitioner group.

Table 8

Chi Square Test for Admissions

 

Variable Pearson DF Sig
Provider 6.87167 2 .032
Insurance 1.27845 1 .258

As shown in Table 9, a logistic regression was
done for the infant outcome variable of admission into
a NICU for more than 24 hours. A logistic regression
was done in order to use a statistical model appropriate
for a dichotomous outcome. It was also performed to
control for the four demographic variables (race, marital
status, maternal age, and months gestation at the first
prenatal care visit) and the two independent variables
(type of provider and type of insurance). This analysis
found no significant differences. None of the four
control variables contributed significantly to the
variable of admission into the NICU for more than 24
hours. Also, neither the type of provider nor type of
insurance were significant. Therefore, the significant
differences of admission into a NICU for more than 24

hours derived from the chi square test for type of

32
provider can be explained in terms of different demographic

variables of the women.

Table 9

Logistic Regression for Admissions

 

Variable DF Sig
Provider 2 .137
Insurance 1 .881
Race 2 .656
Marital Status 3 .791
Maternal Age 1 .417
Months Gestation 1 .782

Research guestions

Neither the infant outcome variables of birth weight
nor admission into a NICU for more than 24 hours were
found to be related to type of provider or insurance.

The birth weights differed at best by only 225 grams or
7 1/2 ounces. The rates of admission into a NICU for
more than 24 hours were similar across the different
types of insurances. The differences found across the
different types of providers (with the certified nurse
midwife group having the largest percent of admissions)
can be explained in terms of the demographic variables
of the women.

The infant outcome variable of gestational age
was found statistically different by the type of provider
but not by the type of insurance. The adjusted mean

gestational ages were 38.76 weeks for the obstetrical

33
group, 39.6 weeks for the certified nurse midwife group,
and 40.23 weeks for the general practitioner group.
Although these ages may vary statistically, from a
clinical standpoint, all were considered term gestation
by definition and thus were normal or desired infant
outcomes.

There was also statistical significance for the
controlling variable of the months gestation at the first
prenatal care visit as a source of variance for gestational
age. This finding supported existing literature that
stated the amount of prenatal care had significant impact
on gestational age (Hulsey et al, 1991).

Bindings

The results of this study lead to the conclusion
that the mean gestational age was statistically, although
not clinically, different; and that the type of provider
was significant as a source of variance: nurse midwives
delivered infants with higher gestational age than the
obstetrician providers and lower gestational age than
the general practitioner providers. No statistical
differences were found for the infant outcomes of birth
weight and admission into a NICU for more than 24 hours.
Thus, these two outcomes were similar across type of
providers. This was consistent with many of the studies
in the literature review (Neeson et al, 1983; Slome et

al, 1976).

34

No statistical differences were found for the infant
outcomes of birth weight, gestational age, and admission
into a NICU for more than 24 hours; so all were similar
across the different types of insurances.

This study found that the infant outcomes were the
same regardless of the type of prenatal care provider
and their subsequent approach to care. The theoretical
framework based on King's Theory of Goal Attainment
(1981) represented the certified nurse midwife group
providing care in an interactive approach. While the
theoretical framework based on the traditional Medical
Model for the two physician groups (obstetricians and
the general practitioners) represented the lack
of this interactive approach. However, when judged on
the same infant outcomes of care there were no differences
across type of providers regardless of their approach
to care.

Discussion
Limitations to the study

One problem occurred during data collection. The
woman’s prenatal care form was missing from many of the
charts. This omission made it impossible to determine
the weeks gestation at the first prenatal care visit
and to count the number of prenatal care visits.
However, the data were available on a birth certificate

form found in every chart. This form listed the months

35
gestation at which prenatal care began and the total
number of prenatal care visits. Use of the Kessner
Index or the Adequacy of Prenatal Care Utilization Index
(Kotelchuck, 1994) could have been included as a more
reliable measure for determining adequacy of prenatal
care. Different statistical tests had to be used than
originally planned because there was an insufficient
number of cases in each category for birth weight and
gestational age.

Some limitations to this study included the unknown
influence of the other multidisciplinary team members
who provided care with the nurse midwives at the women’s
center. Also, it was unknown whether or not the
obstetricians and general practitioners had access to
and utilized other such multidisciplinary team members.
Finally, there was no measure of the influence for the
mode of delivery (vaginal or cesarean) which could have
affected the infant outcome of admission into a NICU for
more than 24 hours.

W

In this study, infant outcomes were the same
regardless of the type of prenatal care provider. This
finding supported other studies that found nurse midwives
provided prenatal care comparable to that of physicians
for the infant outcome variables of birth weight (Dillon

et al, 1974; McLaughlin et al, 1992; Neeson et a1, 1983:

36

Slome et al, 1976), gestational age and admission into
a NICU (Neeson et al, 1983; Slome et al, 1976). The
prenatal care provided by nurse midwives was neither
better nor worse than that provided by physicians as
measured by infant outcomes; thus, nurse midwives can be
utilized in place of other health care providers, such
as physicians, to provide prenatal care. Nurse midwives
can, therefore, help solve the shortage of primary care
providers especially in the critical areas where access
to health care is limited, such as in rural areas.
Implications to; advanced nursing pteetiee

The argument has been raised that advanced practice
nurses, such as certified nurse midwives, can provide
health care at a much reduced cost to the public than
physicians. Cost analysis studies have been done
comparing advanced practice nurses, such as nurse
midwives, to physicians. Many of these studies indicate
that advanced practice nurses are more cost effective
than physicians. Several reasons are cited. First, it
is found that educational costs for advanced practice
nurses are one fifth the costs for physicians (McGrath,
1990). Second, average salaries for advanced practice
nurses are a third less than physicians (McGrath, 1990).
Third, advanced practice nurses have a much lower rate
of law suits than physicians (Pearson, 1994). And fourth,

the cost for advanced practice nurses to provide health

37
care is 39% less than for physicians (Pearson, 1994).
The lower cost is because advanced practice nurses choose
less invasive and less expensive procedures, treatments,
lab tests, and medications. Therefore, the advanced
practice nurses can provide primary health care more
cost effectively, because there is less initial cost for
their preparation (education), and less cost for
maintenance (salary and legal expense), and less health
care expense.

Even though the provision of care by nurse midwives
is cost effective, there are three recognized barriers
to advanced nursing practice. One barrier, in the state
of Michigan is that advanced practice nurses do not have
prescriptive authority which prohibits independent
nursing practice (Pearson, 1994). With independent
prescriptive authority certified nurse midwives could
independently manage clients throughout the prenatal
period, the labor and delivery process, and the post
natal period. High risk clients could be referred to
obstetricians for co-management, much in the same manner
as done by general practitioners.

Another barrier to advanced nursing practice
involves third party reimbursement. Currently in
Michigan, certified nurse midwives are eligible for third
party reimbursement by Medicaid and CHAMPUS. Certified

nurse midwives are also eligible for reimbursement by

38
Blue Cross/Blue Shield of Michigan, a private insurance
company, but only at 85% of the reimbursement level for
physicians. Also, reimbursement is limited by geographical
location in the state (Pearson, 1994). Consequently,
certified nurse midwives are not getting equivalent
reimbursement for equivalent care.

One may ask why should advanced practice nurses be
reimbursed at the same rate as physicians when advanced
practice nurses have been found to provide care at 39%
lower rates than physicians. The obvious answer is that
equivalent care provided by different types of providers
should be reimbursed equally. However, it could be argued
that physicians simply over charge for their services, and
that the physicians should also be reimbursed at the same
rate, albeit lower, as the advanced practice nurses.

A third barrier to advanced nursing practice is
the lack of sufficient educational programs to prepare
advanced practice nurses for primary care practice.

With respect to certified nurse midwifery education,

in the state of Michigan, there currently exists one
educational program. This program admits 14-15 students
per academic year. Nursing programs such as these need
to be increased in number and designed to admit more
students per academic year. These nursing programs also

need to be made accessible to nurses across the state.

39

In order to eliminate these barriers to advanced
nursing practice, nursing organizations must lobby to
ensure that policy changes are made. A first step is
to change state and federal legislation to allow for
independent prescriptive authority for advanced practice
nurses. A second step is lobbying efforts by nursing
organizations for health insurance companies to provide
reimbursement for advanced practice nurses, such as
certified nurse midwives, and particularly to provide
reimbursement at an equal rate for equal services for
all providers. Again lobbying efforts by nursing
organizations with state legislatures and federal
representatives are needed to increase the amount of
dollars allocated for nursing education. Finally, state
colleges and universities must be persuaded by nursing
organizations to develop educational programs for
advanced practice nurses, such as certified nurse
midwifery programs.

da '0 r urther re earc

The findings of this study have expanded the
literature base by including a comparison of three
different types of prenatal care providers as well as
examining two different types of insurances within one
study. However, additional research would be useful.
Future studies would include different outcome variables

such as those related to follow-up care with mothers and

40

infants. One of these variables includes breastfeeding.
Current research has found many health benefits of
breastfeeding for both mothers and infants (Janke, 1993).
In fact, breastfeeding is recommended by the American
Academy of Pediatrics, the American Public Health
Association, and the American Dietetic Association for
the first four to six months of life (Janke, 1993).
Therefore, nurse midwives should take an active role in
promoting breastfeeding. Studies could be conducted to
compare nurse midwife providers to other physician
providers in outcomes such as (1) the percentage of
women who chose to breastfeed over bottle feed, (2) the
duration of breastfeeding, (3) the provider’s knowledge
of breastfeeding, (4) the provider's ability to counsel
women having difficulties with breastfeeding, (5) the
provider's attitudes about breastfeeding, and (6) the
amount and type of education given to women.

Another area of study is prenatal care in settings
that employ a team approach (Haire & Elsberry, 1991;
Heins et al, 1991; McLaughlin et al, 1992; Neeson et al,
1983; Piechnik & Corbett, 1985). Future research could
be conducted to examine: the effects of a team approach
in providing prenatal care compared to that of a single
provider, the role of team members and their contributions
to the care provided, and the resultant outcomes for

mother and infant. Nurse midwives could be compared to

41
each other--those who provide prenatal care in a team
approach to those who provide prenatal care in solo

practices.

APPENDI X

MICHIGAN STATE
U N l v E R SIT Y

April 25, 1994

TO: M rth c. Condon
S 850 Fulton Road
Leonidas, Mi 49066
RE: IRBI: 94-177
TITLE: PRENATAL CARE: A COMPARATIV! ANALYSIS or INSANT
OUTCOMSS
REVISION REQUESTED: N/A
CATEGORY: -8
APPROVAL OATS: 04/22/94

The University Committee on Research Involving Human Subjects'(OCRIBS)
review of this project is complete. I am pleased to advise that the
rights and welfare of the human subjects appear to be adequately
protected and methods to obtain informed-consent are appropriate.
her:§o:g, the UCRIHS approved this project including any revision
.t 0v. 0

RIHEHAL: UCRIHS approval is valid for one calendar year, beginning with
the approval date shown above. Investigators planning to
continue a project ond one year must use the green renewal
form (enclosed with t e original a roval letter or when a
preject is renewed) to seek u at certification. There is a
manimum of four such expedite renewals possible. Investigators
wishing to continue a project beyond that time need to submit it
again or complete rev1ew.

REVISIONS: UCRIHS must review any changes in procedures involving human
subjects, rior to initiation of t e change. If this is done at
the time o renewal, please use the reen renewal form. To
revise an approved protocol at any other time during the year,
send your written request to the UCRIHS Chair. requesting revised
approval and referencing the project's 2R3 i and title. Include
in your request a description of the change and any revised
instruments, consent forms or advertisements that are applicable.

PROBLEMS I
anlGnS: Should either of the followin arise during the course of the
work. investigators must noti UCRIHS promptly: (l) roblems
(unexpected side effects comp aints. etc.) involving uman
subjects or (2) changes in the research environment or new
information indicating greater risk to the human sub ects than
existed when the protocol was previously reviewed an approved.

If we can be of any future help. '-ease do not hesit te to contact us
at (517)355-2150 or tax (517:3 6-i171.

Cr

Sincerel

    
  

 
   

avid 3. Bright, P .D.
UCRIHS Chair

Dtfl:pjm
cc: Carla L. Barnes

 

 

42

Ions-wellness

LI ST OF REFERENCES

43

References

Baruffi, G., Dellinger, W., Stobino, D., Rudolph,
A., Timmons, R., & Ross, A. (1984). A study of
pregnancy outcomes in a maternity center and a

tertiary care hospital. Americen Jentnel of Public
Health, 74(9), 973-978.

Buhler, L., Glick, N., & Sheps, S. (1988). Prenatal
care: A comparative evaluation of nurse-midwives and

family physicians. Cenedien Medieel Association
J urna 1 9, 397-403.

Calhoun County Board of Health. (1991). Repett

of the Ad Hoc Committee on Infant Mottality of the

Calhoun Connty Board of Health. Michigan: The Calhoun
County Board of Health.

Capan, P., Beard, M., & Mashburn, M. (1993). Nurse
managed clinics provide access and improved health care.

The Nurse Practitioner, 18(5), 50, 53-55.

Cavero, C., Fullerton, J., a Bartlome, J. (1991).
Assessment of the process and outcomes of the first
1,000 births of a nurse-midwifery service. Jentnei

gf Nurse Midyifety, 36(2), 104—110.

Chambliss, L., Daly, C., Medearis, A., Ames, M.,
Kayne, M., a Paul, R. (1992). The role of selection
bias in comparing cesarean birth rates between physicians

and midwifery management. Ametieen Qeurnal of
Qpetettics and gyneceiggy, §0(2), 161-165.

Dillon, T., Brennan, 8., Dwyer, J., Risk, A., Sear,
A., Dawson, L., & Wiele, R. (1978). Midwifery, 1977.

American Joutnai 0t Obstetrics end gynecology, 103(8),

917-922.

Edmunds, M. (1994). Time of risk, time of
opportunity. The Nntse Etactitiener, i9(8), 7-8, 11.

Gann, P., Nghiem, L., & Warner, S. (1989).
Pregnancy characteristics and outcomes of Cambodian

refugees. American Jeurnel of Pnblic Health, 72(9),
1251-1257.

44

Gortmaker, S. (1979). The effects of prenatal
care upon the health of the newborn. American Journal

9f Pubiic Health, 69(7), 653-660.

Haire, D., & Elsberry, C. (1991). Maternity care
and outcomes in a high-risk service: The North Central
Bronx Hospital experience. Bittny_ig(l), 33-38.

Heins, H., Nance, N., McCarthy, 3., & Efird, C.
(1990). A randomized trial of nurse-midwifery prenatal
care to reduce low birth weight. American Joutnal 0:

Obstetrics and Gynecology, 75(3), 341-345.

Hulsey, T., Patrick, C., Alexander, 6., & Ebeling,
M. (1991). Prenatal care and prematurity: Is there
an association in uncomplicated pregnancies? Bittn,
i§(3), 146-150.

Janke, J. (1993). The incidence, benefits and
variables associated with breastfeeding: Implications

for Practice. The_Nurse_Eractitignerl_1§(6). 22-23-

King, I. (1981). A tnegty fie: nutsing. New York:
John Wiley & Sons.

Klaus, M., & Fanaroff, A. (1986). gete_et_tne

nigh-tisk neonate (3rd ed.). Philadelphia: W. B.
Saunders.

Korones, S. (1986). High-tiek newborn infants
(4th ed.). St. Louis: C. V. Mosby.

Kotelchuck, M. (1994). An evaluation of the
Kessner Adequacy of Prenatal Care Index and a proposed
adequacy of prenatal care utilization index. Ametieen

qurnei of Public Heeltn, g4 (9), 1414- 1420.

Krieger, J., Connell, F., & LoGerfo, J. (1992).
Medicaid prenatal care: A comparison of use and outcomes
in fee-for-service and managed care. American Journal

gt Pubiic Heeitn, 82(2), 185-190.

McGrath, S. (1990). The cost effectiveness of

nurse practitioners. The unpee Eteetitioner, i5(7),
40-42.

McLaughlin, F., Altemeier, W., Christensen, M.,
Sherrod, K., Dietrich, M., & Stern, D. (1992). Randomized
trial of comprehensive prenatal care for low-income women:
Effect on infant birth weight. Begiettics, 82(1),

128-132.

45

Merenstein, G., & Gardner, S. (1993). Handboen
of neonatal intensive cate (3rd ed.). Chicago: Mosby.

National Governors’ Association. (1994). Medicaid
coverage options for pregnant women and infants. Jentnel

ef Matetnal cniid Heaitn, Update.

Neeson, J., Patterson, K., Mercer, R., & May, K.
(1983). Pregnancy outcome for adolescents receiving
prenatal care by nurse practitioners in extended roles.

goutnai of Adolescent Heeitn Care, 4(2), 94-99.

Pearson, L. (1994). Annual update of how each
state stands on legislative issues affecting advanced

nursing practice. The NBLSQ Eteetitioner, 12(1),
11-13, 17-18, 21.

Piechnik, S., & Corbett, M. (1985). Reducing low
birth weight among socioeconomically high-risk

adolescent pregnancies. Jentnel gt Nutse-Midwifery,
22(2), 88-98.

Quick, J., Greenlick, M., & Roghmann, K. (1981).
Prenatal care and pregnancy outcome in an HMO and
general population: A multivariate cohort analysis.

American gentnai ct Euplic fleaitn, 7;(4), 381-390.

Reeder, S., a Martin, L. (1987). Metetnity

nntsing; family, newpgrn, and women’s heaith care
(16th ed.). Philadelphia: J. B. Lippincott.

Slome, C., Wetherbee, H., Daly, M., Christensen,
K., Meglen, M., & Thiede, H. (1976). Effectiveness of
certified nurse-midwives. 'c Jo r a of

Qpetettics and gynecoiogy, i24(2), 177- 182.

"illllllflllllllllllli