lllllllllilllllwill.

\ 01565 0660

/"; ”it I. ’i \:
l' “/

 

 

 

LIBRARY
filéfiigan State
Unlnralty

 

 

 

This is to certify that the

thesis entitled

FACTORS WHICH ARE INVOLVED
IN THE DECISION TO INITIATE
BREASTFEEDING IN CHILDBEARING WOMEN

presented by

Pamela S. Steele

has been accepted towards fulfillment
of the requirements for

Master‘s degreein Science

 

Major professor

Date ’2 /6-/7é_

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

PLACE Ii RETURN BOX to romovo this checkout from your rooord. -
TO AVOID FINES rotum on or botoro doto duo.

DME‘DUE DATE DUE DATE DUE

JANfllig 199W

 

 

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

L__J___|
T—ll—jT—l

MSU in An Affirmativo ActioNEquol Opportunity institution

 

 

FACTORS WHICH ARE INVOLVED IN THE DECISION
TO INITIATE BREASTFEEDING IN CHILDBEARING WOMEN

BY

Pamela S. Steele

A THESIS

Submitted to
Michigan State university
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

College of Nursing
1996

ABSTRACT

FACTORS WHICH ARE INVOLVED IN THE DECISION
TO INITIATE BREASTFEEDING IN CHILDBEARING WOMEN

By

Pamela S. Steele

Despite health initiatives aimed at increasing
the number of women who initiate breastfeeding, the
percentage of women who choose to breastfeed
continues to decline. The purpose of this study
was to examine the factors which childbearing women
consider in their decision to breastfeed. The
sample were 39 subjects in a West Michigan
community. The perceived benefits of an infant
feeding method were found to play a significant
role in the infant feeding decision; the effect of
interpersonal influence approached significance.
while perceived barriers were not found to be a
factor in the infant feeding decision. The
importance of the role of the advanced practice
nurse in influencing the modifiable variables of
perceived benefits and intepersonal influence were
discussed in relation to the infant feeding

decision.

ACKNOWLEDGEMENTS

I wish to gratefully acknowledge the
assistance and support of my chairperson,
Linda Beth Tiedje, for her kindness in
helping me walk this path. I would
also like to thank my husband, Eric,
and son, Samuel, for allowing me

to find fulfillment in so many roles.

iii

 

TABLE OF CONTENTS

LIST OF TABLES ....................................... v
LIST OF FIGURES ..................................... vi
INTRODUCTION ......................................... 1
THEORETICAL FRAMEWORK ................................ 5
CONCEPTUAL DEFINITIONS OF STUDY VARIABLES ........... 10
REVIEW OF THE LITERATURE ............................ 14
METHOD .............................................. 25
Design ......................................... 25
Operational Definitions ........................ 28
Sample ......................................... 30
Field Procedures ............................... 31
Statistical Analysis of Data ................... 32
RESULTS/FINDINGS .................................... 33
Findings in Relation to the Model .............. 4O
Findings in Relation to the Literature ......... 40
Assumptions/Limitations ........................ 41
DISCUSSION .......................................... 45
Implications for Future Research ............... 46
Implications for Practice ...................... 48
Summary ........................................ 51
LIST OF REFERENCES .................................. 52
APPENDICES .......................................... 59

Appendix A: The Adapted Infant Feeding
Questionnaire ............................. 60

Appendix B: Consents From Participating
Clinics ................................... 66
Appendix C: Introduction and Consent Letter...70
Appendix D: UCRIHS Approval Letter ............ 72

iv

Table

Table
Table
Table

Table
Table
Table

Table

Table

LIST OF TABLES

Variables Associated with Breastfeeding
Initiation and Success (Adapted

from Janke, 1993) .................... 15
Frequencies of income ..................... 34
Frequencies of race ....................... 35

Crosstab table of feeding method
and race ............................. 35

Summary statistics for descriptive data...36
Summary statistics by feeding type ........ 37

Crosstab table of combined income by
site and selection of breastfeeding..37

Crosstab table of feeding method and
income ............................... 38

Summary statistics of benefits, barriers,
and interpersonal influences by
feeding method ....................... 38

LIST OF FIGURES

Figure l. The Revised Health Promotion Model ....... 6

Figure 2. Adaptation of Pender's Model to the
Decision to Breastfeed ............... 8

vi

Introduction

The selection of an infant feeding method is a
common, yet controversial dilemma facing childbearing
women today. A personal decision, the choice of infant
feeding method is confounded by such global influences
as government programs, employment policies, and
societal attitudes. Women who choose to breastfeed
commonly attempt to breastfeed despite inflexible work
schedules and lack of public support. Women who choose
to bottle—feed often feel defensive about choosing a
feeding method which the public simultaneously may judge
as second-rate. For those who decide to breastfeed, the
dilemma may be examined daily with conflicting
expectations between an appropriate length of
breastfeeding and meeting one's own social, emotional,
and professional needs. The infant feeding decision also
poses a dilemma for health care professionals. Health
care providers are often expected to provide the facts
about feeding methods, and yet are expected to remain
non-judgmental to parents once a feeding decision has
been made.

Exclusive breastfeeding is endorsed by the American
Dietetic Association and American Academy of Pediatrics
as the optimal source of infant nutrition for the first
four to six months and as a benefit to the lactating

woman (ADA, 1993; American Academy of Pediatrics (AAP),

l

2
1992). The benefits of breastfeeding for the infant

have been clearly demonstrated to include the
immunologic and nutritional properties of breast milk
(Cunningham, Jelliffe, & Jelliffe, 1991), and enhanced
intellectual, oral, and visual development for the
infant (Lucas, Morley, Cole, Lister, Leesen—Payne, 1992;
Makrides, Simmer, Goggin, & Gibson, 1993; Davis & Bell,
1991). Breastfeeding is also associated with a
decreased incidence of such acute and chronic childhood
diseases as otitis media (Saarinen, 1982), pneumonia,
bacteremia, and meningitis (Cunningham, Jelliffe, &
Jelliffe, 1991), insulin—dependent diabetes (Metcalfe &
Baum, 1992), and lymphoma (Davis, Savitz, & Graubard,
1988), as well as a decreased risk of sudden infant
death syndrome (Mitchell, Scragg, & Stewart, 1992).
Mothers who breastfeed not only benefit from the
convenience of breastfeeding and decreased costs of
infant feeding, but also the decreased risk of such
diseases as ovarian and premenopausal breast cancer
(Jones & Matheny, 1993; Rosenblatt & Thomas, 1993;
Newcomb, Storer, Longnecker, Mittendorf, Greenberg,
Clapp, et a1, 1994).

Benefits of breastfeeding to society include
decreased health care costs and environmental strain,
including decreased industrial and household waste which
accrue when babies are formula-fed (Dermer, 1995).

Dewey, Heinig, and Nommsen-Rivers (1995) in a matched

 

3

pairs study of breastfed and formula-fed infants found
that "the reduction in morbidity associated with breast-
feeding is of sufficient magnitude to be of public
health significance” (p. 696). In view of such
evidence, the Healthy People 2000 initiative set as its
goal that 75 percent of infants would be breastfeeding
at hospital discharge, and 50 percent still
breastfeeding by the age of five to six months (United
States Department of Health, Education, and Welfare,
1979).

Ironically, however, breastfeeding initiation rates
in the United States continue to decline from their peak
of 61.9 percent in 1982. The percentage of women who
initiated breastfeeding between 1984 and 1989 decreased
from 59.7% to 52.2%, while the percentage of women still
breastfeeding their infants at six months dropped from
23.8% to 18.1% (Ryan, Rush, Krieger, & Lewandowski,
1991). Commonly cited reasons for bottle-feeding
include conflicts with work or school schedules, the
convenience of formula, maternal or infant illness, lack
of milk, infant preference, breast pain, and
embarassment of public nursing (American Academy of
Pediatrics, 1982; Janke, 1988; Morse, 1989).

The advanced practice nurse in primary care must
evaluate the factors which child-bearing women are
considering in the selection of an infant feeding

method. By carefully assessing the many factors which

 

4

enter into the infant feeding decision, the advanced
practice nurse can identify modifiable factors to
encourage breastfeeding, and can better plan the care of
the childbearing family and the community as a whole.
The purpose of this study was to examine factors
involved in the infant feeding decision between women
who decide to initiate breastfeeding and women who
decide to initiate bottle-feeding. Factors which were
addressed in this study include interpersonal
influences, perceived benefits, and perceived barriers,
as well as individual characteristics.
Specific research questions addressed in this study
were:
1.) What are the differences in interpersonal
influences between women who decide to initiate
breastfeeding or bottle—feeding?
2.) What are the differences in perceived benefits
of an infant feeding method between women who decide to
initiate breastfeeding or bottle-feeding?
3.) What are the differences in perceived barriers
to an infant feeding method between women who decide to
initiate breastfeeding or bottle-feeding?
4.) What are the differences in individual
characteristics, including age, education, income,
parity, and race between women who decide to initiate

breast-feeding or bottle—feeding?

 

Theoretical Framework

The factors which influence the selection of
breastfeeding as the infant feeding method in
primiparous women have been adapted to the Health
Promotion Model by Pender (1996) for purposes of this
study. Because Pender's model includes both immutable
and modifiable variables (including perceived benefits,
perceived barriers, and interpersonal influences) which
influence health promoting behaviors, the model is
especially appropriate as a framework to study the
decision to breastfeed. Pender's model (See Figure 1)
includes various factors which influence behavioral
health outcomes, including individual characteristics
and experiences, and behavior—specific cognitions and
affect. Individual characteristics and experiences
include prior related behaviors and personal (including
biological, psychosocial, and sociocultural) factors.
Behavior-specific cognitions and affect consist of the
perceived self-efficacy, activity-related affect,
perceived benefits and perceived barriers to actions, as
well as interpersonal and situational influences. The
competing demands which are immediately confronting the
individual as well as the individual's commitment to a
plan of action also influence health promoting

behaviors.

Individual Behavior-Specific

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Characteristics Cognitions BMW“!
and Experiences and Afi'ect : Home
. Perceived benefits ,
r of action
Prior Perceived barriers Immediate competing
related —> ‘0 “5°“ ' demands
behavior (“low control)
and preferences
Perceived (high control)
‘ self-eficaey
._. Activity-related ,
afi‘ect
Personal factors; . . _ v
biological Commmnent Health
psychological “’ t0 8 . promoting
sociocultural In" : mom! plan of action f Theme:A
influences
—’ (family. peers, ‘—'
providers); norms,
support, models
—-)I
Situational influences;
Options
"' demand characteristics '-
aesthetics

 

 

 

 

 

 

Figure 1. The Revised Health Promotion Model. (Fender, 1996. p.67)

7

Pender conceptualizes behavior-specific cognitions and
affect as those characteristics described in the
literature review as modifiable, or changeable
variables. Pender (1996) states that behavior—specific
cognitions and affect ”constitute a critical 'core' for
intervention, as they are subject to modification
through nursing actions” (p. 68). The behavior-specific
cognitions of interpersonal influences, perceived
benefits and perceived barriers will be related to
infant feeding decisions in this study.

Sweeney and Gulino (1987) have adapted the Health
Promotion Model to explain breastfeeding practices
within a Hispanic population (See Figure 2). While some
of the variables are specific to their patient
population (primarily Hispanic), many are appropriate
for application to a variety of child-bearing
populations. Figure 2 depicts the factors which
influence the decision to breastfeed as conceptualized
within the Health Promotion Model.

In adapting the Health Promotion Model to
breastfeeding (Sweeney and Gulino,1987), individual
characteristics and experiences include personal factors
and prior related behaviors. Personal factors include
the woman's age, educational level, marital status,
prenatal employment, health status, and personal likes
and dislikes. Prior related behavior includes previous

breastfeeding behavior in the multiparous woman.

Individual
Characteristics
and Experiences

 

 

Prior

Related —>

Behavior

 

 

 

 

Individual
Charateristics;
Age.
Marital status.
Education.

Behavior-Specific

 

 

 

 

 

 

 

 

 

 

 

 

 

Race, Income

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

‘ Interpersonal Influences .
Husband/Poorer F—
Mother, M'otlwr-in-law, 1
Friends. Co-workers.
Plysician. Nurses

C0815 tions Behavioral
and Afi‘ect me
Perceived Benefits of Action
. _ Convenience, C ost. '
Health benefits. —’
Psychological benefit:
__lPercelved Barriers to Action _, C edrate
C omnience, Embarrassment. I trng
. . Demands and
Fear of pain. Lea/ong, Pref
Time C onstraints ercnces
’ Perceived __
Self-Emcee}!
Activity-Related
. Afi'ect .__.
' _ v L
Commitment Infant
to a Feeding
Plan of Action Decision
A A

 

 

 

 

 

 

 

 

qitnational Influences: Options

 

 

 

 

 

 

 

Figure 2. Adaptation of Pender’s Model to the Decision to Breastfeed
(Adapted fiom Sweeney & Gulino, 1987)

9

Personal factors included in this study include the age
and education of both the woman and her partner, as well
as combined income, race, and parity. Three behavior-
specific cognitions will be examined in this study:
interpersonal influence, perceived benefits, and
perceived barriers. The behavior-specific

cognition of interpersonal influence in the decision to
breastfeed includes the influence of friends, close
female relatives (mothers, sisters, mother-in-law,

etc ), the woman's husband or the baby's father, the
physician, nurses both in and out of the hospital, and
co-workers. The cognition of perceived benefits
includes perceptions of immunologic benefits, feelings
of attachment, feelings of naturalness, perceptions of
self—fulfillment, convenience, nutritional benefits,
perceived benefits of weight loss and sensations of
enjoyment. Factors included as perceived barriers to
breastfeeding include time constraints, inability to
measure feeding volume, feelings of responsibility,
breast pain, discomfort with feeding in front of
strangers, altered body perception, and perceptions of
discouragement from family and friends.

The variable individual characteristics falls
within Pender's category of individual characteristics
and experiences, which incorporates immutable factors.
The variables of perceived benefits, perceived barriers,

and interpersonal influences fall within Pender's

lO

category of behavior—specific cognitions and affect.
When perceived benefits to the action of breastfeeding
are high and perceived barriers are low, an individual
is likely to have a stronger commitment to a plan of
action and to engage in a health promoting behavior like
breastfeeding. Likewise a stronger sense of
interpersonal influence will increase commitment and the
likelihood of making a health promoting feeding
decision. Conversely, a woman may have a strong degree
of interpersonal influence from friends and family to
breastfeed, may cognitively identify perceived benefits
to breastfeeding, and yet may have increased perceptions
of barriers to action, such as feelings of loss of
control and therefore will be less likely to commit to
breastfeeding.

In summary, the variables of individual
characteristics, as well as perceived barriers,
perceived benefits, and interpersonal influences, which
have been derived from Pender's Health Promotion Model
(1996) and Sweeney and Gulino's adaptation of the Health
Promotion Model to breastfeeding (1987) will be examined
in relation to the infant feeding decision in this
study.

Conceptual Definitions of Study Variables

The key variables addressed in this study include

individual characteristics, interpersonal influences,

perceived benefits, perceived barriers, and their

ll

influence upon the decision to initiate breast or
bottle-feeding.
Individual Characteristics

Individual characteristics, as defined by Webster,
are ”the qualities which distinguish an individual from
others, distinguishing traits, or features"(1984, p.
239). For the purposes of this study, individual
characteristics are the demographic data, which include
the subject's age. her husband or partner's age, the
educational level of both parents, their combined
income, and the ethnic background of the subject. The
variables of individual characteristics, although
analyzed separately in this study, when viewed together
provide a picture of differences in socioeconomic
opportunities, educational influence, and problem-
solving abilities.

Interpersonal Influences

Interpersonal influences are conceptually defined
by Pender (1996) as “cognitions concerning the
behaviors, beliefs, or attitudes of others” (p. 70).
Pender notes that such cognitions may or may not
correspond with reality. As such, interpersonal
influences are a woman's perceptions of the beliefs and
values of others. Primary sources of interpersonal
influence identified by Pender include families, peers,
and health care providers. Such interpersonal

influences include social norms, social supports, and

l2
modeling (Pender, 1996). Pender notes that the

importance of interpersonal influence may vary between
cultures. An example of interpersonal influence might
include the woman who chooses to breastfeed not because
she has internalized the benefits of breastfeeding, but
because she feels obligated to fulfill family and
friends' expectations of what a good mother should do.
Interpersonal influence varies from the concept of
social support, which may include tangible, emotional.
or informational support. For the purpose of this study,
Pender's conceptualization of interpersonal influence
will be used, which focuses on the perceptions or
cognitions of the beliefs or values of others.
Perceived Benefits

Perceived benefits are defined by Pender (1996) as
"mental representations of the positive or reinforcing
consequences of a behavior" (p. 68), and are
conceptualized as powerful extrinsic or intrinsic
determinants of health—promoting behaviors. The
subjective perceptions of possible benefits to
breastfeeding are nutritional, immunologic, health,
emotional, economic, or social in nature. In the Health
Promotion Model, perceived benefits are proposed to
directly and indirectly motivate behavior by
"determining the extent of commitment to a plan of
action to engage in the behaviors from which the

anticipated benefits will result" (Pender, 1996, p. 69).

13

For the purposes of this study, Pender's
conceptualization of perceived benefits will be used,
which focuses on the cognitions of the positive
consequences of the behavior of breastfeeding.
Perceived Barriers
Conversely, perceived barriers are conceptualized
as real or imagined "perceptions concerning the
unavailability, inconvenience, expense, difficulty, or
time-consuming nature of a particular action” (Pender.
1996, p. 69). Barriers may be viewed as the costs or
blocks to a particular behavior. Pender (1996) states
“when readiness to act is low and barriers are high,
action is unlikely to occur" (p. 69). Perceived
barriers are also conceptualized as affecting health-
promoting behavior directly and indirectly by acting as
a block to action and by decreasing commitment to a plan
of action. Examples of common barriers to
breastfeeding include concerns about breast pain and
leaking, embarassment of breastfeeding in public, and
inconvenience. For the purposes of this study. Pender's
conceptual definition of perceived barriers will be
used, which focuses on the cognitions regarding the
inconvenience, difficulty, or possible problems
associated with breastfeeding
Breastfeeding
Breastfeeding has been conceptually defined across

a broad spectrum, from including women who supplement

l4

breastfeeding with unlimited formula to women who do not
supplement with a bottle. In general, breastfeeding is
described as feeding a child through the process of
maternal lactation. Oxytocin and suckling are
considered the primary biological facilitators of milk
production in the post-partum woman (Ladewig, London, &
Olds, 1990). For the purposes of this study,
breastfeeding will be conceptualized as any woman who
has initiated breastfeeding for any amount of time
between the infant's birth and the completion of the
survey at six to eight weeks post—partum.

Review of the Literature

In review of recent literature on the
characteristics of women who decide to breastfeed, two
classifications of variables emerge: immutable
variables, and modifiable variables. (See these
variables summarized in Table 1). The variable
individual characteristics in this study falls within
the category of immutable variables. while the variables
interpersonal influences, perceived benefits, and
perceived barriers fall within the category of
modifiable variables.

Studies which have been multivariate in nature have
found several key factors as significant predictors for
the decision to breast or bottle-feed. Gabriel,
Gabriel, & Lawrence (1986) found demographic factors and

cultural ideas were closely associated with the choice

15
Table 1. Variables Associated with Breastfeeding

 

Ini i tion and S c 35 d t 93
[mu tab! 9 Pariah] es Hodjfjabje Varietal es
Caucasian Intent to breastfeed a
Middle to upper class long time
Well educated Early first feeding
Married Committed to breastfeed
Early 205 to 305 Good support
Breastfed as infant Positive attitude
Healthy infant and mother toward breastfeeding

Non-smoker
Avoiding supplemental
feedings

of breast or bottle—feeding, as were perceived barriers
such as dietary and health related practices. Perceived
barriers. including perceptions of difficulties in
scheduling breastfeeding and worries about the demands
of breastfeeding were found to be significant predictors
influencing the duration of breastfeeding by Ekwo,
Dusdieker, Booth, & Seals (1984). A second study by the
same authors found three significant predictors for the
initiation of breastfeeding: positive maternal beliefs
about breastfeeding (perceived benefits), the absence of
maternal worries about breastfeeding (perceived
barriers). and higher levels of education (individual
characteristics) (Dusdieker, Booth, Seals, & Ekwo,
1985). Sweeney & Gulino (1987), in applying the Health
Belief Model to the breastfeeding decision found that
three of the components of the model were most useful in

predicting health-related outcomes--individual

perceptions, modifying factors, and the likelihood of

16
action. Melnyk (1988), although not implicitly

describing breastfeeding behaviors, states that
literature on barriers has found it to be one of the
most significant predictors of decisions relating to
health promoting behaviors.

Immutable variables for women who are more likely
to breastfeed include: Caucasian race (Gabriel. Gabriel,
& Lawrence, 1986; Kurinij, Shiono, & Rhoads, 1988),
middle to upper class (Ekwo, 1984; Grossman, 1990), well
educated (Feinstein, 1986; Ouarles,Williams, Hoyle,
Brimeyer, & Williams, 1994), married (Gabriel, Gabriel &
Lawrence, 1986; Grossman, 1990), and older women in
their 20's and 30's (Coreil & Murphy, 1988; Feinstein,
1986). Other immutable variables which are associated
with breastfeeding include women who likely were
breastfed themselves as infants (Arafat, Allen, & Fox,
1981; Gabriel, Gabriel, & Lawrence, 1986), and are
healthy with healthy infants (Ferris, 1987).

The effect of parity on the decision to breastfeed
is inconclusive. Ferris, McCabe, Allen, and Pelto
(1987) found that parity and previous breastfeeding
experience had no statistical effect on the feeding
method used at ten weeks post—partum, but that duration
of of previous breastfeeding experience was significant;
if a mother succesfully nursed before for a long
duration, she was more likely to breastfeed again.

However, others have found no association between

l7

breastfeeding and parity (Grossman, Fitzsimmons, Larson—
Alexander, Sachs, and Harter,1990; Gabriel, Gabriel, and
Lawrence,1986); and at least two studies have found more
breastfeeding among primiparous women than among
multiparous women (Rassin, Richardson, Baranowski,
Nader, Guenther, Bee, and Brown, 1984; Martinez, 1984).

Three broad categories of modifiable factors emerge
from the literature: Interpersonal influences on the
decision to breastfeed; the woman's perceived benefits
to breastfeeding; and the perceived barriers to
breastfeeding. These three factors have all been
explored extensively in the literature. Each of these
modifiable factors will be examined in the following
sections, as will the concept of the infant feeding
decision itself.

Interpersonal Influences

Interpersonal influences on breastfeeding have been
examined extensively in the literature, including such
interpersonal relationships as family, friends, and
health care providers. The role of the father on
influencing the decision to breastfeed has been
described in many studies. Fraley, Freed, & Schanler
(1992) found that fathers who were planning to have
their infants breastfeed were significantly more likely
to believe that breastfeeding was healthier for the
infant. Jordan and wall (1990) examined fathers'

perceptions of breastfeeding before and after the birth

18

of their children, and found fathers who expressed
support for breastfeeding before the birth of the child
often described feelings of being threatened or excluded
by breastfeeding after the birth of the child.

Husbands or partners have been usually identified by
Caucasian women as their primary source of support for
breastfeeding (Baranowski, Bee, Rassin et al, 1983;
Sweeney & Gulino, 1987).

Ethnicity was examined as a determinant of sources
of support. African-American women more often
identified close friends as sources of support for
feeding decisions, while Hispanic women more often
identified their mothers as the most influential
support. The husband or partner was the most
significant influence for Caucasian women (Baranowski,
Bee, Rassin, et al, 1983). Dusdieker, Booth, Seals, and
Ekwo (1985) found that the strongest predictor of
maternal breastfeeding anxiety was maternal worry about
lack of psychosocial support.

The interpersonal influence of health professionals
has also been the topic of several studies with
inconsistent results. One survey of health care workers
showed that one third of physicians never discussed
feeding plans with their prenatal patients (Lawrence,
1982). Another study found that women rarely report
receiving advice about feeding method from their health

care providers (Reif & Essock—Vitale, 1985).

19

Conflicting information about infant feeding method has
been given by some health care providers who provide
information about breastfeeding, yet advertise for
formula manufacturers by sending new mothers home with
formula packs (Jones & West, 1986; Bruce, Khan, & Olsen,
1991). When health care providers do provide positive
information about breastfeeding, women are more likely
to breastfeed (Hally, Bond, Crawley, Gregson, Philips, &
Russell, 1984).
Perceived Benefits

Several studies have addressed the link between
maternal beliefs in the perceived benefits of
breastfeeding and the decision to breastfeed. Besides
the immunologic, nutritional, intellectual, and physical
benefits described earlier, several studies have
demonstrated that women who breastfeed believe that
breastfeeding is "more natural" and results in better
bonding with the infant (Dix, 1991; Gielen-Carlson,
Faden, O'Campo, & Paige, 1992). The perceived
convenience along with decreased cost of breastfeeding
has been linked as a factor in the decision to
breastfeed (Yeung, Pennell, Leung, & Hall, 1981), yet
convenience was also the only positive benefit cited by
women who choose to formula-feed (Sarett, Bain, &
O'Leary, 1983; Yoos,1985; Arafat, Allen, & Fox, 1981).
Studies have characterized the perceived benefits of

infant feeding methods as ”infant centered” or “mother

20
centered“ (Switzky, Vietze, & Switzky, 1979; Yoos,

1985). Losch, Dungy, Russell, and Dusdieker (1995) in
describing these classifications describe women who
select breastfeeding as ”motivated more by concerns
about the baby and his or her welfare (e.g., providing
the healthiest nutrients and promoting a strong bond
with the infant)" (p. 511) than by concerns about
themselves. Yet, Dusdieker, Booth, Seals, and Ekwo
(1985) found that a mother's expectation of the
benefits for herself was the strongest predictor of
breastfeeding motivating factors. Interestingly, the
majority of formula-feeders also acknowledge the
superior health benefits of breast milk (Hally, Bond,
Crawley, Gregson, Philips, & Russell, 1984; Dix, 1991).
Those who choose to breastfeed overwhelmingly believe
that breast milk is healthier for their infants (Sarett,
Bain, & O'Leary, 1983; Gunn, 1984). In summary,
perceived benefits are cognitions of greater well-being
for the infant and the woman herself associated with
breastfeeding.
Perceived Barriers

Perceived barriers to breastfeeding have been
cited as a factor in the choice of infant feeding
method. Specific barriers include embarassment about
public breastfeeding (Gielen—Carlson, Faden, O'Campo, &
Paige, 1992), concerns about discomfort (Dix, 1991), and

worries about the problem of leaking of milk

21

(Morse,1989). Several studies evaluated other barriers
to breastfeeding, such as worries that their
professional and social lives would be restricted by
breastfeeding (Kearney & Crononwett, 1991; Gielen—
Carlson, Faden, O'Campo, & Paige, 1992; Yoos, 1985).
Concerns about the father not being involved in infant
feeding have also been cited as barriers to
breastfeeding (Adair,1983). In addition, Morse (1989)
describes the sexual stigma attached to breasts as an
aversive stimuli to breastfeeding for many women. In
summary, the perceived barriers to breastfeeding are
primarily cultural barriers, with the social impact of
employment and family, as well as the perceptions women
have developed within their culture regarding their
bodies, and particularly their breasts.
The Decision

The infant feeding decision itself has also been
examined in the literature. Hollen and Hobbie (1993)
conceptually define quality decision—making as
“searching and weighing of alternatives, careful
deliberation, planning, and follow-up, within realistic
constraints" (p. 771). Sarett, Bain, and O'Leary (1983)
summarized three separate studies on the infant feeding
decision between 1976 and 1980, and found that 85—92% of
mothers selected an infant feeding method before the end
of pregnancy. Approximately 5-7% remained undecided in

the last trimester. Ninety—six to ninety-seven percent

22
of mothers surveyed fed with the method which they had

previously selected. Dix's (1991) study on the infant
feeding decision showed 46% of women made the feeding
decision during pregnancy. while 41% had already decided
prior to conception. Jones and West (1986) found that
the earlier the decision is made to breastfeed, the
longer the duration of breastfeeding. Marchand and
Morrow (1994), using qualitative methodology, found five
common themes in the infant feeding decision:
knowledge, support, quantification (ensuring optimal
nutrition for the infant), satisfaction, and privacy.
Literature Critique

The literature regarding infant feeding decisions
is lacking in several areas. Cunningham, Jelliffe, and
Jelliffe (1991) in discussing the global epidemiological
benefits of breastfeeding note several glaring
methodological flaws in the literature. The most
obvious is the inconsistent definition of breastfeeding.
Studies vary from defining breastfeeding as exclusive or
complete feeding by breast (Ekwo, Dusdieker, & Booth,
1983); to limited supplementation with formula, defined
rigidly as less than or equal to one bottle-feeding per
day (Dusdieker, Booth, Seals, & Ekwo, 1985); to simply
attempting to breastfeed at least once after delivery
(Jones & West, 1986). Morse (1989) notes that until
the 1980‘s research failed to consider mixed feedings

(or a combination of breast and bottle) as a legitimate

23

category along with exclusive breast or bottle-feeding.
Sweeney & Gulino (1987) extended their definition of
breastfeeding to include ”those who breast—fed their
infants exclusively or in combination with supplementary
feedings of water, tea, formula, or other liquids" (p.
40).

The literature is also unclear in measuring
breastfeeding initiation versus duration as indicators
of successful breastfeeding. Many studies describe
breastfeeding behaviors, but determined study groups by
using exclusive breastfeeding or bottle-feeding groups
at six to ten weeks post-partum (Dusdieker, Booth,
Seals, & Ekwo, 1985; Ferris, McCabe, Allen, & Pelto,
1987; Grossman, Fitzsimmons, Larsen—Anderson, Sachs, &
Harter, 1990), which negate the decision of many women
to breastfeed only in the early weeks post—partum.
Successful breastfeeding is often not defined, or is
defined solely as how long a mother breastfeeds (see
Ferris, McCabe, Allen, & Pelto, 1987).

Small sample size and homogeneity are also common
methodological flaws found in the literature, which
limit generalizability of the research. Dix (1991)
examdned perceived barriers and benefits to
breastfeeding, but had a relatively small sample size
(n=81), which was homogeneous (young, single, African—
American, and multiparous). Likewise Yoos' (1985) study

of adolescent feeding selection involved a small sample

24

size (n=50) with a majority of subjects who were
Airican-American. Matich & Sims (1992) studied social
support variables in breast and bottle-feeders.

Although their sample was fairly large (n=159). 96% were
Caucasian women, which was representative only of its
small community.

One study stands out well in terms of methodology
and definitions of variables. Kurinj, Shiono, & Rhoads
(1988) were clear in defining breastfeeding as including
women who supplemented or intended to breastfeed. Their
study was a longitudinal study of 1,179 women, which
measured both initiation and duration of breastfeeding.
This study looked, however, only at demographics and
supplementation rates (objective data), not at
perceptions and attitudes (subjective data). This
suggests that studies need to be multivariate in nature
and study both objective and subjective data.

Many studies are simplistic in nature in
viewing the effect of only one variable on the decision
to breastfeed (Jones & West, 1993; Jordan & Wall, 1990;
Kearney & Cronenwett, 1991). .Although a few tools exist
to systematically examine the many factors influencing
breastfeeding including immutable variables, as well as
modifiable variables and perceptions (Dusdieker, Booth,
Seals, & Ekwo, 1985; Janke, 1994), replications to

validate such tools are lacking.

25

The purpose of this study was to examine several
factors which influence the choice of infant feeding
method. This study addressed a few of the weaknesses
identified in the literature by further validating the
use of an existing tool which examined modifiable
factors, multifactorial influences, and perceptions in
the decision to breastfeed (Janke, 1994). This study
looked at both immutable variables, which included
objective data including subjects' ages, race, income,
education, and parity, as well as modifiable variables.
Specific modifiable variables which were addressed in
this study included interpersonal influences, perceived
benefits, and perceived barriers and the effects of
these variables upon the decision to breastfeed. The
definition of breastfeeding was inclusive of any woman
who attempted breastfeeding, in order to focus
exclusively on the decision to breastfeed, regardless of
duration. The implications of identifying modifiable
factors in child-bearing women by the advanced practice
nurse in primary care were addressed in response to the
results of this study.

Method
Design

This study was a non-experimental design in which a
self-selected sample of post—partum.women completed a
one-time survey at six to eight weeks post-partum. The

decision to administer the test at six to eight weeks

26

post-partum was based on the timing of the standard
post—partum visit. The survey required about ten minutes
to complete and was administered at three different
health care or nutrition provider sites, to optimize
obtaining a heterogeneous sample. Surveys were given to
subjects on arrival to sites for post-partum or two-
month well child checks, with the instruction to return
surveys during that particular office visit.

The survey involved an adaptation of the Infant
Feeding Questionnaire (IFO) by Dusdieker, Booth, Seals,
& Ekwo (1985). The original IFO is a 64 item
questionnaire which explores motivation factors in
infant feeding for primigravidas' decision to initiate
breastfeeding. The IFO includes information in the
following broad categories: maternal worries about
family and personal health; maternal perceptions of the
importance of preventive health behaviors in determining
personal health; maternal perceptions of the benefits of
breastfeeding; maternal perceptions of worries and
anxieties regarding breastfeeding; the influence of
family, friends, media, and education on the feeding
decision; and perceptions of available support from
friends, family, and health care providers.

The IFO requires respondents to quantitate the
influence of a particular item on the infant feeding
decision, using a five point Likert scale. For example,

women are asked to indicate how important the idea that

27

their feeding method is convenient was in their
selection of a feeding choice. Responses range from ”of
no importance" (1) to "very important" (5). Cronbach's
alpha for individual subscales of the IFO ranged from
.70 to .92. Using the IFO, Dusdieker, Booth, Seals, and
Ekwo utilized multivariate techniques to create a causal
model identifying predictors in the infant feeding
choice. R2,for the entire model was .74.

The IFO was adapted by this researcher to be a 45
item questionnaire which explores factors in the
selection of infant feeding method in childbearing
women, regardless of parity or type of feeding method
selected. Questions regarding maternal worries about
personal and family health, and perceptions regarding
preventive health were not examined in the adapted IFO,
(a deletion of items one through twenty), since the
areas of perceived benefits, perceived barriers, and
interpersonal influences more concisely provided a
general overview of health perceptions. Questions were
rephrased in order to be universal for both breast and
bottle-feeders. The variable of perceived benefits in
the adapted IFO was derived from the original variable,
maternal perceptions of the benefits of breastfeeding.
The variable of perceived barriers was derived from the
orginal variable, maternal perceptions about worries and
anxieties regarding breastfeeding. The variable of

interpersonal influence was derived from the originial

28

variable of the influence of family, friends, media, and
education on the feeding decision (by deleting questions
related to information only in order to focus on the
perceptions of interpersonal influence). Several
questions were added in adapting the questionnaire,
including one question regarding the subject's
perception of when the infant feeding decision was made,
questions related to parity and previous breastfeeding
experience, questions related to smoking behavior and
its influence on feeding method, and questions related
to type of delivery and complications. These questions
were added since parity and timing of the decision were
not topics of study within the original tool, and to
provide more descriptive information about the sample.
Appendix A includes the adapted IFO. Demographic data
about the respondent are also asked, including the age,
income, and education of both parents, and the infant's
sex. The adapted questionnaire takes about ten minutes
to complete, and can be done independently by the
respondent.
Operational Definitions

Determination of sample groups was based on
responses to question one, which asks whether or not the
respondent breastfed her infant in the hospital. This
question makes the assumption that the majority of women
who initiate breastfeeding will have done so immediately

after the infant's birth, or while still hospitalized.

29

Regardless of duration of breastfeeding, women who
answered yes to question one were operationally defined
as included in the breastfeeding sample, while those who
responded no to question one were operationally defined
within the bottle-feeding sample.

The variable interpersonal influences was
operationally defined as a summation of items 22 through
30 of the adapted IFO, with scores ranging from 8 (a low
degree of interpersonal influence) to 40 (a high degree
of interpersonal influence). The variable perceived
benefits was operationally defined as a summation of
items 14 through 21 of the adapted IFO, with scores
ranging from 7 (a low degree of perceived benefits) to
35 (a high degree of perceived benefits). The variable
perceived barriers was operationally defined as a
summation of items 5-14 of the adapted IFO, with scores
ranging from 9 (a low degree of perceived barriers) to
45 (a high degree of perceived barriers). There were no
reverse scored items; therefore, higher numbers were
always assumed to imply higher perceived benefits,
higher perceived barriers, and more perceptions of
interpersonal influence.

In addition, demographic data, including the
respondent and her significant other's age, education,
income, smoking status, type of delivery, infant sex,
race or ethnicity, and previous breastfeeding experience

was assessed (questions 33—45 of the adapted IFO),

30

because all of these variables have been associated with
breastfeeding in the literature. One question was asked
to determine at what time the respondent felt that she

had made the decision regarding an infant feeding method

(question 4 of the adapted IFO).

Smile

The target sample for this study was breast-feeding
and bottle-feeding women in the West Michigan area.
The sample size was 40 post-partum women, and was based
upon 10 respondents for each variable. Subjects were
recruited from one family practice, one
obstetricallgynecological site, and the area WIC office
in the greater Holland area. The sites involved in the
study are: Lakewood Family Medicine, P.C., Ob~Gyn,
Associates, P.C., and the area Women, Infants, and
Children (WIC) Program office. Sites were selected with
the intention of creating socio-economic diversity and
better representing the Holland community. Lakewood
Family Medicine, P.C., is an established, large family
practice with six family practice physicians, one
physician's assistant, and one nurse clinician.
Lakewood has an estimated patient load of 25,000
patients, who are primarily middle to upper income. 0b—
Gyn Associates, P.C., is a relatively new obstetrics and
gynecology office with four obstetricianlgynecologists

on staff. Ob~Gyn, Associates provides care for clients

31

from a variety of socio-economic levels, including women
on Medicaid, as well as upper-income women. The area
WIC office provides nutritional coupons and counseling
to low—income women and children, and has a case load of
approximately 2800 clients per month.

Exclusion criteria for the study included: (1)
Women with a hospitalized or ill infant (2) non—English
speaking women due to lack of consistency in
translation. The determination of infant feeding method
was based on self-report from the respondents.

Field Procedures

The following procedures were implemented in this
study:

1. Three clinical liasons were established, with
one liason at each of the practice sites. Training of
the liason was done by the principal investigator.

2. As women presented to each site at the six to
eight week postpartum/well-child check, staff from that
site determined eligibility for the study and provided
eligible subjects with an envelope containing an
introduction letter (see Appendix B) which contained
information regarding the study, its purpose, and its
level of commitment. Voluntary consent was determined
by the completion and return of the survey. All surveys
(whether completed or not) were returned to the box
provided in the waiting room at each site. Subjects

were identified only with identification number and the

32

number of each site. No other identifying information

was used.

Statistical Analysis of Data

Statistical analyses of the data were done with the
SPSSX/PC+ computer program. Frequencies, means, and
percentages were used to describe study groups, as well
as chi-square tests to compare groups. Groups were
determined on the basis of question #1 as those who have
decided at birth to breastfeed and those who have
decided at birth to bottle—feed.

1. Comparisons of interpersonal influences between
groups using a Wilcoxon signed-rank test were done.

2. Comparisons of perceived benefits between
groups using a Wilcoxon signed-rank test were done.

3. Comparisons of perceived barriers between
groups using a Wilcoxon signed—rank test were done.

4. Frequencies were run on individual
characteristics, including age, education, and income of
subjects. These data on individual characteristics
(using sites as groupings and feeding method as groups)
were compared using Chi—square analyses.

Protection of Human Rights

The rights of all subjects involved in this study
were protected in accordance with the university
Committee on Research Involving Human Subjects (UCRIHS)

at Michigan State University guidelines. Approval #96—

33

406 was obtained from the Michigan State University
Human Subjects Review Committee prior to initiating any
data collection. Permission was obtained from the
director of each of the participating clinics. (See
Appendix B). Voluntary consent to participate in the
study was indicated by completion and return of the
survey, as indicated in the introduction and consent
form.(See Appendix C). Parental consent was not obtained
for minors participating in the study, as subject matter
involved infant feeding decisions and pre—natal care,
and was exempt under State of Michigan Act No. 153 of
1984, which maintains that minors do not need parental
consent for any treatment related to pregnancy.
Results/Findings

Most of the sample was obtained from Ob—Gyn,
Associates (48.7%). Lakewood Family Medicine, although
it had a much larger patient base, provided only 12.8%
of the sample. WIC participants made up the remaining
38.5%. The determination of feeding method, based on
question one, which asked about the feeding method used
in the hospital, showed that 64.1% of the sample had at
one time breastfed, while 35.9% had elected to bottle—
feed. The majority of subjects (53 %) had selected a
feeding method during pregnancy, while 41% reported the
decision prior to pregnancy. Five percent of subjects
made the feeding decision after the birth of the infant.

Fifty—nine percent of the subjects were multiparous,

34

while 41% were multiparous. One third of the sample
reported previous breastfeeding experience. The
majority (79.5%) delivered vaginally, while 20.5%
delivered by caesarian section. Non—smokers made up
84.6% of the sample, while smokers made up 15.4%.
Summaries of income and race are found in Tables 2 and
3.

Because the tool did not have mutually
exclusive categories of incomes, income categories were
collapsed. Since 185% of the poverty level is
approximately $25,000, this level was used to separate
high and low income for statistical analysis. Chi-
square analysis of income showed a p-value of .5164
(with/Z = 4.211 and D.F. =1). indicating that there was
no significant difference in income between women who
chose to breastfeed and women who chose to bottle-feed.
Table 3 shows racial distribution, with 66.7% of
respondents being Caucasian and 17.9% of the respondents
being Latino; although this sample is certainly not
generalizable, it is fairly representative of the

community under study. Chi-square analysis of race with

2 e ue cies ' e
19.2.1 6 WM Bergen;
Missing Value 1 2.6
less than $25,000 17 43.6

more than $25,000 21 53.8

35

Table 3. Frequencies of rage
Value Exeguency Percent

Missing Value 1 2.6
African-American 1 2.6
Asian 1 2.6
Caucasian 26 66.7
Latino 7 17.9
Native American 1 2.6
Other 2 5.1

collapsed categories comparing Caucasians and all other
minorities combined was significant, with a p-value of
.0231 (Z'=5.1579, D.F. =1), indicating there was a
racial difference in breastfeeding and bottle—feeding
mothers, with Caucasians more apt to breastfeed than all
other minorities.

Tables 5,6,and 7 (pages 36 and 37) provide another
picture of the sample. The picture of the ”average
respondent” which is emerges is one of a woman in her
mid—twenties, with a significant partner who is slightly
older. The average respondent and significant other
have completed high school, and possibly some college
education. Surprisingly, the average bottle—feeder and
her significant other were slightly older than the
average breast—feeder and her partner. However, breast—

feeders and their significant partners had slightly

     

s

Feeding African— Asian C ucasian Latino Native Other

 

 

wrican American
breast 1 0 19 3 1 1
bottle 0 1 7 4 o 1

 

total 1 l 26 7 l 2

36

b e 5. s t' t' 3 sc 1 ‘ve a
lien. Mean mm
Age 25.6 5.95 19
Sig. other age 28.2 6.20 24
Years of education 12.6 2.349 11
.Sig. other education 12.52 2.35 8
Total perc. barriers 19.92 6.62 26
Total perc. benefits 25.90 6.50 29
Total Personal Influence 21.56 7.38 26

more education than bottle-feeders. Chi—square
analyses showed no significant differences between
groups in age (if =2.74, D.F.=3, p: 4329), or age of
the significant partner (it =8 05, D.F.=4, p= 0897).
However, the amount of education was found to be
different between groups(:2I=24.23, D.F =5, p=.0002), as
was the amount of education of the partner (7L =10.94,
D.F.=4, p: 0272), with breast feeders and their partners
both having more education.

As shown in Tables 7 and 8, incomes at site one
were higher than at the other sites (as expected). WIC
(Site #3) incomes were positively skewed, which is
expected since the program serves low income families.
The higher survey completion rate at site one explains
the negatively skewed income distribution in the overall

sample population. Breastfeeding incidence rose with

37

Table 6. Summagy atatistics by feeding type
Baeaagfaeders
Mean Wm.- . n e

Age 25.54 5.88 19
Sig. other age 27.96 6.10 23
Years of Educ. 12.79 2.284 10
Sig. other Educ. 13.38 2.018 6
Egggle—faegers

Age 26.14 6.29 17
Sig. other age 28.21 6.68 20
Years of Educ. 12.36 2.50 10
Sig. other educ. 11.07 2.20 8

income; however, bottle—feeding had a bimodal
distribution; the overall distribution of bottle—
feeding cross—tabulated with income appears normal with
a peak in middle-income women, and then a second peak
which occurs in high—income women.

Because of the small sample size, normality
could not be assumed; therefore, nonparametric tests
were used, rather than t-tests in answering the research
questions 1,2, and 3. The nonparametric test used for
this study was the Wilcoxon signed-rank test, which
ranks the data instead of using actual means, and is
appropriate with ordinal data. The Likert scales used

which assess perceived benefits, perceived

b e 7. rosstab tabl o ' e ' e 't
a d s ect on o s e
Site < $5,000- $10,000— 315,000-325,000- $35,000- >
35000 310.000 315.000 325.000 335.000 350.000 than

 

 

 

350.000
01. 1 3 4 6 5
ELL 1 +2 1 i
03 2 2 2“ .4 3. 1
ngalz 3 3 2a9 7 7 7

 

38

Table 8. Cgosstab tabla of feeding mathod and ibcome
Feeding < $5,000—810.000-515.000—$25,000-$35,000— >
meth. 85000 $10,000 $15,000 $25,000 $35,000 $50,000 than

 

 

 

$50,000
ELEASE l 2 1 6 6 5 3
bottle 1 l 2 3 1 2 4
gblal 2 3 3 9 7 7 7

 

barriers, and interpersonal influences can be
conceptualized as ordinal data. This test is also
useful when outliers are present. Missing values are
omitted from the ranking and not included in the tests.
In this study the influence of each specific variable on
the infant feeding decision was examined; no attempts
were made to explore the interactions of each

variable upon the other variables (requiring
multivariate regression analyses).

Research question #1 asked: What are the
differences in interpersonal influences between women
who decide to initiate breastfeeding or bottle—feeding?
Using an alpha of p =.05, there was not enough evidence

to reject the null hypothesis. The p-value for

Table 9. Sammagy atatistica bf baaefits, barriazs. and
'nter ersonal influe ces feed' et od
Bottle Braaat
Median (Range) Median
Benefits 22 7-35 30
Barriers 19 9-45 19

Interpersonal influences 19 8-40 24

39

interpersonal influences was .0997; therefore, the null
hypothesis could not be rejected, although this variable
did approach significance. Median score for breast—
feeders was 24 compared to bottle-feeders' median of 18,
indicating breastfeeders perceived more interpersonal
influences on them to breastfeed.

Research question #2 asked: What are the
differences in perceived benefits of an infant feeding
method between women who decide to initiate
breastfeeding or bottle-feeding? Perceived benefits was
the only variable which was statistically significant.
The p-value for this variable equaled .0004; therefore,
there is evidence that perceived benefits did play a
role in the feeding decision. Medians were
significantly higher for breastfeeders with a median of
30 compared to bottle-feeders with a median of 22.
Results within this variable imply that breastfeeders
identified more benefits to breastfeeding than bottle-
feeders.

Research question #3 asked: What are the
differences in perceived barriers to an infant feeding
method between women who decide to initiate .
breastfeeding or bottle—feeding? In testing the variable
of perceived barriers, the p value was .4283. Therefore
the null hypothesis could not be rejected. The medians

for both groups were 19, which suggests that both breast

40

and bottle—feeders identified the same amount of
barriers.
Findings in Relation to the Model
The findings of this study were partially
consistent with the conceptual model. Behavior-specific
cognitions as described by Pender include the concepts
of perceived benefits, perceived barriers and
interpersonal influences. Perceived benefits did play a
role in the breastfeeding decision, with breastfeeding
mothers perceiving more benefits to breastfeeding.
Interpersonal influences came close to being
significant, with mothers who breastfed perceiving more
interpersonal influence to breastfeed. Perceived
barriers were not shown to play a significant role in
the decision to breastfeed, which contradicts the
variable as a factor within the Health Promotion Model.
Increased education of the woman who breastfed and her
partner fit consistently with Pender's concept of
individual characteristics and are related to the
decision to breastfeed.
Findings in Relation to the Literature
The findings of this study were also fairly
consistent with the literature. The decision to
breastfeed was made by the majority of women before or
during pregnancy, which is consistent with the
literature. Initiation rates were consistent with

national averages. The bimodal distribution of bottle—

41

feeding and income, in which there is a peak of bottle—

feeding with middle-income women, and a second peak of

bottle-feeding in high-income women was a surprise
finding which conflicts with the literature, which
describes increased breastfeeding behaviors as income
rises. The finding of increased benefits to
breastfeeding reported by breastfeeding women, is
consistent with the literature, although contrary to the

literature, bottle-feeding women within this study did

not acknowledge the health benefits of breastfeeding.

The importance of interpersonal influences was

consistent with previous literature. The lack of

significance for perceived barriers was an unexpected
finding in this sample, since past literature has
described perceived barriers as a significant factor in

the decision to breast or bottle-feed (Melnyk, 1988).

Assumptions
The following were identified as assumptions for
the purpose of this study:

1. The American Academy of Pediatrics recommendation
that all infants be exclusively breastfed for
the first four to six months is the standard
for infant feeding.

2. Self-reported data given by subjects was accurate

to the best of their knowledge.

42

Limitations

The results of this study are limited by several
methodological flaws. The sample was made up of an
extremely small group from within one West Michigan
community. Although attempts were made to create a
diverse sample using sites which provided health care to
different socio-economic and cultural groups, mostly
upper income women responded to the survey, and two-
thirds of the sample were Caucasian. Results are
therefore not generalizable to a larger population.

Recruitment of subjects was a significant problem
at two of the sample sites. Lakewood Family Medicine
staff verbalized understanding of field procedures.
However, due to fluctuations in intake personnel and
inconsistent support from the site liaison, all possible
subjects were not identified consistently. The primary
investigator worked with the clinical liaison to
increase study partipants; however, all eligible women
who presented at this site were not identified by staff.
Subjects from WIC were also limited due to programmatic
problems, in which WIC participants were rarely
encountered in the six to eight week post—partum time
frame at that site (and typically are seen only at birth
and three months to obtain food coupons).

Other threats to external validity include
respondents who may have given socially acceptable

responses because of perceptions that the investigator

43

or health care provider supported breastfeeding.
Experimenter effects may have been present in
respondents who had received health care from the
primary investigator. The primary investigator is a
clinician and active supporter of breastfeeding.

Threats to internal validity include maturation,
which involves changes in respondents' perceptions of
factors which influenced a decision which may have been
made months before the survey was completed. For
example, a subject may have forgotten why she really
chose a feeding method, since in most cases the decision
had been made months before, often before or during
early pregnancy. Selection biases may have been
exacerbated by office personnel, who may not have
presented the survey to all women presenting for post—
partum or well-child care at six—eight weeks post-
partum. In addition, because participation was
voluntary, participants self-selected their
participation by simply returning either completed or
incomplete surveys to each site. Since participation
was self-selected, women who felt strongly regarding the
feeding decision may have been more likely to
participate, while women who did not feel as strongly or
who were systematically different from respondents may
not have participated. Interestingly, no totally
incomplete surveys were received at any of the three

sites, although four surveys were received which had

44

completed less than 20 questions. These were dropped
from the analysis. Response rates were unable to be
computed due to the lack of information on women who
were actually eligible and asked to complete a survey.
The low subject number from site two may have been due
to low staff recruitment or to patient refusal to take
the research packet and read the information and consent
letter.

Another limitation of this study was that cost, the
influence of smoking, and medication use were not
included as a factor in the decision to breast or
bottle-feed. The original IFO did not include questions
regarding cost as a possible perceived benefit to
breastfeeding. In adapting the IFO, cost was not added,
although it could be viewed as a perceived benefit to
breastfeeding, or as a perceived barrier to bottle-
feeding. As discussed in the literature review, cost
may be an important factor, with many women viewing
breastfeeding as an inexpensive alternative to rising
costs of formula.

The other problem encountered involved missing
values. Because subjects were offered the choice of not
answering all the questions, and discontinuing the
survey at any time, many missing values were
encountered. SPSS does take into account missing
values, and although use of the Wilcoxon—signed rank

test minimized the problem, total completion of the

45

survey would have provided more accurate results. This
was partially a design flaw in that the researcher
offered possible options of interpersonal influences

which may not have been applicable to all participants.

Discussion

This study did find the variable of perceived
benefits to be a significant factor in the infant
feeding decision, with breastfeeding women perceiving
more benefits to breastfeeding than bottle—feeding
women. The variable of interpersonal influence
approached significance, with breastfeeding women
perceiving more interpersonal influence to breastfeed
than bottle-feeding women. The variable of perceived
barriers was not found to play a role in the infant
feeding decision in this study.

Although individual characteristics were analyzed
separately, as a composite they provide information
about what type of woman chooses to breastfeed.
Caucasian women with higher levels of education were
more likely to breastfeed. Perceptions of interpersonal
influence were also significantly different in this
group. In general, this reflects the increased access
to education, information and support available to
middle and upper-class Caucasian American women, and the
cultural influences of their peer group. Since

employment status was not assessed, results may reflect

46

a cohort which has the financial security to allow the
woman to remain at home, which eliminates some of the
role and time conflicts associated with employment and
breastfeeding.

Implications for Future Research

Results of this study create new opportunities for
future research. Larger sampling of multi—ethnic groups

‘
is necessary for further exploration of individual and
cultural characteristics which may affect the infant
feeding decision. Further research can be done to
better evaluate the timing and the process of decisions
related to infant feeding method. Perceptions and
attitudes which may affect the feeding decision
(including further research on perceived benefits,
perceived barriers, and interpersonal influences)
require further exploration. Outcomes research which is
specifically geared to the particular effects of
advanced practice nursing interventions is also
necessary, with possible questions related to
educational interventions (including childbirth
preparation classes and teaching at prenatal visits) by
the advanced practice nurse.

Although the adapted IFO provided a concise
overview of several of the factors involved in the
decision to breastfeed, including both cognitive—
perceptual and demographic data, this researcher would

expand the use of this tool to provide more attitudinal

47

data, including subjects' attitudes (positive or
negative) regarding breastfeeding, as well as perceived
attitudes from sources of interpersonal influence. For
example, this tool was not created to demonstrate the
perceived attitudes of the husband or partner toward
breastfeeding, although it did measure the woman's
perception of support from her husband or partner.

This study looked at several factors, including
perceived benefits, perceived barriers, interpersonal
influences, and individual characteristics, and the
influence of these individual factors on the decision to
breastfeed. Further research would include multi—
variate regression to examine the relative effect of
each independent variable on the dependent variable (the
decision to breastfeed) in the presence of other
variables. A strong methodological design would include
the effects of an advanced practice intervention upon
the decision to breastfeed.

In replicating this study, this researcher would
make several suggestions to strengthen the research. A
larger sample is needed, with more sample sites within a
community to decrease the influence of individual health
care providers and to gain better insight into the
community as a whole. Better recruiting of subjects
must be performed, means of obtaining better recruitment
might include more training of all staff as well as the

site liaison or the provision of incentives.

48

Administering the questionnaire during the actual
hospitalization might provide more recent recollections
of perceptions (less threat of maturation) involved in
the decision, and may provide a more accessible captive
audience, although short post—partum hospital stays
would make this difficult to implement.
Implications for Practice

The implications for the advanced practice nurse
(APN) in primary care are many. Since this study found
perceived benefits to be a significant factor in the
decision to breastfeed and the factor of interpersonal
influence to approach significance, the APN must examine
how best to influence these modifiable variables. In
the role of assessor, the APN must assess all pregnant
clients early in the pre—natal time period to determine
whether or not a decision has been made regarding an
infant feeding method, and how strongly each woman feels
regarding that decision, in order to tailor discussions
regarding breastfeeding to each client. In light of
study findings, it is particularly important to assess
the perceived benefits of breastfeeding from the
client‘s point of view to determine the client's
knowledge base. The APN must also assess the culture of
the client to identify perceived interpersonal
influences, including not only the partner and extended
family, but also social and institutional supports

including co-workers and religious affiliations.

49

As educator, the APN uses the role to influence the
modifiable variable of perceived benefits. Failure to
provide education regarding a woman's options ethically
deprives her of the opportunity to make an informed
decision. One possible method of education would be to
create a one—page flyer or brochure to describe the
immunologic, nutritional, emotional, and intellectual
benefits of breastfeeding for the baby, as well as the
emotional, nutritional, and health—protective effects
for the woman. The APN may be involved on a community
level by providing education regarding the benefits of
breastfeeding during childbirth and infant care classes.
Media presentations which describe specific research
studies which confirm the benefits of breastfeeding
would educate a largely uninformed public on the
benefits of breastfeeding. Since less education was
associated with bottle-feeders, the APN may emphasize
educational programs to target women educated at the
high school level or lower. Including the topic of
breastfeeding and its benefits in high school sex
education and family classes, as well as basic biology
curriculae would reach more women. The APN must also
target boys as well as girls in curriculae, since the
influence of the husband or partner has been found to
play a significant role in the decision.

As educator, the APN serves not only as an

interpersonal influence, but may make referrals to other

50

sources of support which may modify perceived
interpersonal influence, such as La Leche organizations
or new mom support groups. The APN may function as a
consultant to credible health care providers within the
client's socio-economic and cultural milieau to increase
the amount of influence that is placed upon prenatal
information regarding the feeding decision, emphasizing
the benefits of breastfeeding. The APN may collaborate
with community health leaders to promote breastfeeding
through the influence of public health and other nurses
involved with at—risk infant programs, such as maternal
support services (M83) and infant support services
(ISS).

In presenting education and discussion on the
benefits of breastfeeding, the APN serves as advocate
for the optimal nutrition of the infant. However, in
serving as an advocate for the mother, the APN must take
into account perceived benefits for the woman, and must
support the woman's own perceived needs, beliefs, and
values. Discussions of the benefits of breastfeeding
for the mother include the emotional satisfaction and
feelings of fulfillment associated with breastfeeding,
weight loss in the post—partum period, the financial
benefits of not having to purchase formula, and
convenience. A thorough assessment of the woman's
culture, perceived benefits, and interpersonal

influences will assist the APN in helping the client

51

make a feeding decision which will be right for the
client as well as the infant. Danner (1991) wisely
summarizes, “Respect for the individual must never be
lost and must be balanced with respect for the power of
breastfeeding (p.228).“
Summary

This study of post-partum women from a small West
Michigan community examined the factors of perceived
benefits, perceived barriers, interpersonal influences,
and individual characteristics upon the infant feeding
decision. More benefits to breastfeeding were perceived
by breastfeeding women. Perceptions of more
interpersonal influence on the decision to breastfeed
were present in breastfeeders. These findings were
consistent with the Health Promotion Model by Pender
(1996), as well as the literature. The influence of the
APN on the modifiable variables of perceived benefits

and interpersonal influences was discussed.

LIST OF REFERENCES

LIST OF REFERENCES

American Academy of Pediatrics, Committee on
Nutrition (1982). The promotion of breastfeeding.

Eediatrigsr_6215) 654-661-

American Academy of Pediatrics, Committee on
Nutrition. (1992). The use of whole cow's milk in

infancy. Pediatriggi_§2, 1105— 1109.

American Dietetic Association. (1993). Position
of the American Dietetic Association: promotion and
support of breast—feeding. Journai_gi_;he_émgrigan
Dietetic Association. 93, 467—469.

Arafat, 1., Allen, D.E., & Fox, J.E. (1981).
Maternal practice and attitudes toward breastfeeding.
ournal f Obstetric G neco o ' d c at Nurs'
i3, 68—72.

Baranowski, T., Bee, D. E. & Rassin, D. K. (1983).
Social support, social influence, ethnicity, and the

breastfeeding decision _Sgcial_§21enee_nediginer_l
367— 372

Bruce, N.G., Khan, 2., & Olsen, N.D.L. (1991).
Hospital and other influences on the uptake and
maintenance of breast feeding: The development of
infant feeding policy in a district. Enblig_flgal;ni
195‘ 357- 368.

Coreil, J., & Murphy, J.E. (1988). Maternal
commitment, lactation practices, and breastfeeding
duration. ' ‘

Hegnafal_flur§ingi_ll. 273-278-

Cunningham, A. S. Jelliffe, D. B. & Jelliffe, E.F.
(1991). Breast— —feeding and health in the 19803: a

global epidemiologic review. lgnrnal_gf_figdiairig§i
iifii 659- 666.

52

53

Danner, S. (1991). How do we influence the
breastfeeding decision? Birth‘_l§(4), 227-228.

Davis, D W. & Bell, P. A. (1991). Infant feeding
practices and occlusal outcomes. a longitudinal study.

l2urnal_2f_1he_Canadian_Dental_As§2211112n1 211 593— 594

Davis, M.K., Savitz, D.A., & Graubard, B.I. (1988).
Infant feeding and childhood cancer. Langgti_2(8607),
365-368.

Dermer, A. (1995). Overcoming medical and social

barriers to breast feeding Ansr1gan_Eamilx_£hx§1£1an1
51(4), 755- 761.

Dewey, K.G., Heinig, J., & Nommsen-Rivers, L.A.
(1995). Differences in morbidity between breast-fed and

formula—fed infants. Journal of Pediatrics. 126 (5),
696-702.

Dix, D.N. (1991). Why women decide not to
breastfeed. Birth, 18. 222—225.

Dungy, C.I., Losch, M., Russell, D., Christensen—
Szalanski, J. (1992). Effect of discharge samples on
duration of breast—feeding. Pediatrics, 90, 233—237.

 

Dusdieker, L B., Booth, B M., Seals, B.F., & Ekwo,
E. E. (1985). Investigation of a model for the
initiation of breastfeeding in primigravida women.

52g111.121en2erned191n21_2_(7) 695- 703

Ekwo, E. E. Dusdieker, L. B. & Booth B M. (1983).
Factors influencing intiation of breast- -feeding.

Amer12an_l2urnal_21.21sabled_£h1ldr_nr_lél 375- 379

Ekwo, E. E. (1984). Psychosocial factors
influencing the duration of breastfeeding by

primigravidas Ae1a_Eaed1a1r123.521nd1naxigar_snnnli
299. 241—247.

Feinstein, J. M. (1986). Factors related to early
termination of breastfeeding in an urban population.

Eediatr12§1_lflfz) 210- 215

Ferris, A.M. (1987). Biological and sociocultural
determinants of successful lactation among women in
Eastern Connecticut. Journal of the American Dietetic
Association, 87(3), 316-321.

54

Fraley, K. Freed, G. L, & Schanler, R. J. (1992L
Attitudes of expectant fathers regarding breast- feeding.

Eed1atric§1_2_1 224- 227

Gabriel, A., Gabriel, K.R., & Lawrence, R.A.
(1986). Cultural values and biomedical knowledge:
Choices in infant feeding, analysis of a survey. Sggial

§c1ence_and_nedic1nei_22(5). 501-509.

Gielen—Carlson,.A.. Faden, R.R., O'Campo, P., &
Paige, D.M. (1992). Determinants of breastfeeding in a

rural WIC population. W
11—15.

Grossman, L.K. (1990). The infant feeding decision
in low and upper income women. Clinical Pediatrics.
g2(1), 30-37.

Gunn, T.R. (1984). The incidence of breast feeding
and reasons for weaning. New Zealand Medicina:;ournal,
21. 360—363.

Hally, M.R., Bond, J., Crawley, J., Gregson, B.,
Philips, P., & Russell, I. (1984L Factors influencing
the feeding of first— born infants. Acaa Pediatrica

Scandinax1car_llr 33- 39

Hollen, P. J. & Hobbie W. L (1993). Risk taking
and decision making of adolescent long-term survivors of

cancer Qnc_lcgy_Eursing_E2rum1_ZQl5L 769- 775

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

for practice Hur§e_Eract111gner1_l§(6) 22-32-

Janke, J. R. (1994). Development of the breast-

feeding attrition prediction tool. Huraing_Ba§aarghi
ga(2), 100- 104

Johnson, C.A., Garza, C., & Nichols, B. (1984). A
teaching intervention to improve breastfeeding success.
Journal of Nutrition Education. 16, 19-20.

Jones, D.A. & West, R.R. (1993). Effect of a
lactation nurse on the success of breast-feeding: A
randomized controlled trial. ur l o ide '

and Community Health. 40. 45—49.

5'5

Jones. E.G.. & Matheny, R.J. (1993). Relationship
between infant feeding and exclusion rate from child

care because of illness. n f e
Diatatig A§§ggiatioa. 2;. 809-811.

Jordan, P.L., & Wall, V.R. (1990). Breastfeeding
and fathers: Illuminating the darker side. Bi£1h1_ll.
210-213.

Kearney, M., & Cronenwett, L. (1991).
Breastfeeding and employment . W921...
Gynecologic. and Neonatal Nursing, 20. 471—480.

 

Kurinj, N., Shiono, P.H., & Rhoads, G.G. (1988).
Breast—feeding incidence and duration in black and white
women. Pediatrics, 8i(3), 365—371.

Ladewig, P.A., London, M L., & Olds, 5.8. (1990).
Essentials of Maternal—Newborn Nursiag. (2nd Ed.i.
Addison-Wesley: New York.

Lawrence, R.A. (1982). Practices and attitudes
toward breast—feeding among medical professionals.
Pediatrigs. 70, 912—920.

Lawrence, R. (1991). Breast—feeding trends: A
cause for action. Pediatricsi_88a 867-868.

Losch, M., Dungy, C.I , Russell, D., & Dusdieker,
L.B. (1995). Impact of attitudes on maternal decisions
regarding infant feeding. MW
12§(4). 507—514.

Lucas, A.. Morley, R., Cole. T.J., Lister. G.. &
Lesson—Payne, C. (1992). Breast milk and subsequent
intelligence quotient in children born preterm. Langagi
212. 261—264.

Marchand, L., & Morrow, M.H. (1994). Infant
feeding practices: Understanding the decision-making
process. Clinical Research and Methods. 26(5), 319—324.

Makrides, M., Simmer, K., Goggin, M., Gibson, R.A.
(1993). Erythrocyte docosahexaenoic acid correlates
with the visual response of healthy, term infants.

Begiatric Resident. 33, 425-427.

56

Martinez, G.A. (1984). Trends in breast feeding in
the United States. In Bep_rt_of_ths_§urgegn_§eneral_s
Wgrkahop an Braast Egeding and Humaa Lagta agiga U. S.

Department of Health and Human Services.

Matich, J. R. & Sims, L. S. (1992L A comparison
of social support variables between women who intend to

breast or bottle feed S2c1al_§cienge_ned121nei_141
919-927.

Melnyk, K.A. (1988). Barriers. A critical review
of recent literature. Na;aing_fia_aa;gha_§_(4), 196-201.

Metcalfe, M.A., & Baum, J.D. (1992). Family
characteristics and insulin-dependent diabetes.
Archiveg of Disabled Children. 67, 731—736.

Mitchell, E.A., Scragg, R., & Stewart, A.W. (1992).
Four modifiable and other major risk factors for cot
death: the New Zealand study. Journal of Padiatric and
Child Health. 28i1). 53-58.

Morse, J.M. (1989). ”Euch', those are for your
husband!” examination of cultural values and assumptions
associated with breast-feeding. Health Care for Women
International. 11, 223-232.

Newcomb, P.A., Storer. B.F., Longnecker, M.P.,
Mittendorf, R., Greenberg, E. R. & Clapp, R. W. (1994).
Lactation and a reduced risk of premenopausal breast

cancer N___Eng1and_lournal_of_nedic1ne1_12_1 81- 87

Pender, N J., & Pender, A.R. (1987). Haalgh
Ercmot1gn_1n_flur§1ng_2racf1ce_(2nd Ed )- Appleton &

Lange: Norwalk, Connecticut.

Ouarles, A., Williams, P.D., Hoyle, D A., Brimeyer,
M., & Williams, A. R. (1994) Mothers' intention, age,
education, and the duration and management of

breastfeeding Ma1ernal:Qhild_Nurs1ng_lournalr_22(3L
102- 108.

Rassin, D.K., Richardson, C.J., Baranowski, T.,
Nader, P.R., Guenther, N., Bee, D.E., & Nader, P.R.
(1984). Incidence of breast-feeding in a low
socioeconomic group of mothers in the United States:
Ethnic patterns. Pediatrics. 73. 132—137.

 

57

Reif, M I., & Essock—Vitale, S.M., (1985).
Hospital influences on early infant—feeding practices.

Eediagrics, 76, 872-879.

Rosenblatt, K. A. & Thomas, D. B. (1993). Lactation
and the risk of epithelial ovarian cancer.

In1erna11_nal_J__rnal_2f_Epidemiol_gxr_221 192- 197

Ryan, A. 5. Rush, D. Krieger, F. W. & Lewandoswki,
G.E. (1991). Recent declines in breast- feeding in the
United States, 1984 through 1989. E§Q135£i£§1_§§;719-
727.

Saarinen, U.M. (1982). Prolonged breast feeding as
prophylaxis for recurrent otitis media. _Ag;a
Paediatrica Scandinavica 71, 256—571.

 

Sarett, H P., Bain, K.R., & O'Leary, J.C. (1983).
Decisions on breast—feeding or formula feeding and
trends in infant—feeding practices. American Journal of
Disabled Children. 137. 719-725.

Sudman, S., & Bradburn, N. (1983). Asking
Questions: A Practigal Guide to Questionnaire Design.
Jossey-Bass. San Francisco.

Sweeney. M., & Gulino, C. (1987). The health
belief model as an explanation for breast-feeding
practices in a Hispanic culture. v e ' u s'n

Sgiaaga. 9, 35-50.

Switzky, L.T.. Vietze, P., & Switzky, H.N. (1979).
Attitudinal and demographic predictors of breast-feeding
and bottle-feeding behaviors by mothers of six—week-old

infants. Esxcholggx_Beportr_3§. 3—14.
U.S. Department of Health, Education, and Welfare.
(1979). o e: s e ' c

on health promotion.and disease prevention. (DHEW
publication PHS 79-55071). Washington DC: U.S.
Government Printing.

Williams, E.L. & Hammer, L.D. (1995).
Breastfeeding attitudes and knowledge of pediatricians—
in—training. American Journal of Praventive Medicine

iiil), 26—33.

 

58

Yoos, L. (1985). Developmental issues and the
choice of feeding method of adolescent mothers. angaal
0 .9 ‘ II. 1" 0 00., 0. e 9. ‘09: v. _,S_!°. ‘
68—72.

Yeung, D.L., Pennell, M.D., & Leung, M. (1981).
Breastfeeding: Prevalence and influencing factors.

MW. 323—330.

APPENDICES

APPENDIX A

The Adapted Infant Feeding Questionnaire

59

60

ID 3
Site 3

 

MOTIVATION FACTORS IN INFANT FEEDING DECISIONS
(CONFIDENTIAL)

DIRECTIONS: Place a circle around the number at the point on
the scale which best describes your feelings for each
question. In the example below, if like to watch television
alot, then you would circle the 1. If you chose to circle
the 3. that would mean that your liking of television fell
between ”like somewhat" and "dislike somewhat." In other
words. you do not particularly like or dislike watching
television, or you have neutral feelings about watching
television.

 

4v"- uv- n 1' ' y e
nnmb§;§_ For example, if your choice were to fall between 4
and S, w w c ire e e e w i
feelingeiit... either 4 or 5.
lease ead a l estions ca e u l efore a sweri
them.
EXAMPLE: How well do you like to watch television?
Like Like Neutral Dislike Dislike
very much somewhat somewhat very much
2 3 4 5
\\\\~

\_

 

THINK BACK\TO WHEN YOUR BABY was BORN...

 

1. Did you breastfeed your baby in the hospital? Yes
No
2. If yes, for how long did you breastfeed?
3. Are you still breastfeeding your baby? Yes__
No__
4. IF YOU CAN REMZMBER EH£E_YOU DECIDED HOW YOU WERE GOING

TO FEED YOUR BABY, WAS IT
Before I became pregnant
During my pregnancy
After the baby was born
I can't remember when

61

WHEN YOU DECIDED HOW YCU WERE GOING TO FEED YOUR BABY. HOW
CONCERNED OR BOTHERED WERE YOU ABOUT THE FOLLOWING?

 

 

 

 

 

 

 

 

 

 

 

5. Knowing exactly how much milk your baby would take.

Not at all Somewhat Neutral Moderately Very
concerned concerned concerned concerned
1 2 3 4 5
6. Having enough time.

Not at all Somewhat Neutral Moderately Very
concerned concerned concerned concerned
1 2 3 4 5
7. Feeling totally responsible.

Not at all Somewhat Neutral Moderately Very
concerned concerned concerned concerned
1 2 3 4 5
8. Feeling tied down.

Not at all Somewhat Neutral Moderately Very
concerned concerned concerned concerned
1 2 3 4 5
9. Concerns about breast pain, engorgement, or leaking.

Not at all Somewhat Neutral Moderately Very
concerned concern concerned concerned
1 2 3 4 5
10. Concerns about breastfeeding in front of strangers.

Not at all Somewhat Neutral Moderately Very
concerned concerned concerned concerned
1 2 3 4 5
11. Loss of your figure.

Not at all Somewhat Neutral Moderately Very
concerned concerned concerned concerned
1 2 3 . 4 S
12. Lack of support rem or discouragement by family members.
Not at all Somewhat Neutral Moderately 'Very
concerned concerned concerned concerned
1 2 3 4 5
13. Lack of support or discouragement by friends.

Net at all Somewhat Neutral Moderately Very
concerned concerned concerned concerned

 

1 2 3 4 5

62

WHEN YOU DECIDED HOW YOU WERE GOING TO FEED YOUR BABY. HOW
IMPORTANT WERE The FOLLOWING TO YOU:

14. The fact that the baby could acquire or obtain some
immunity against many infections from breast milk.

 

Of no Of little Neutral Moderately Very
importance importance important important
1 2 3 4 5

15. The feeling that you and the baby would develop close
feelings for eachother by breast-feeding or bottle-feeding.

 

Of no Of 11 tle Neutral Moderately Very
importance importance important important
1 2 3 4 S
16. The feeling that breastfeeding was a natural way to feed
your child.

Of no Of little Neutral Moderately Very
importance importance important important
1 2 3 4 S

 

17. The desire to be a "complete" woman or feel self-
fulfilled.

 

 

 

Of no Of little Neutral Moderately Very
importance importance important important
1 2 3 4 S
18. Convenience.

Of no Of little Neutral Moderately very
importance importance important important
1 2 3 4 5
19. Nutrition.

Of no Of little Neutral Moderately Very
importance importance important important
1 2 3 4 S

20. The feeling that breastfeeding might help you regain
your figure.

Of no Of little Neutral Moderately Very
importance importance important important
1 2 3 4 5

 

21. The feeling that you might enjoy the sensation of

breastfeeding.
Of no Of little Neutral Moderately Very
importance importance important important

1 2 3 4

 

HOW MUCH INFLUENCE DID

63

THE FOLLOWING HAVE ON YOUR FEEDING

 

 

 

 

 

 

 

 

DECISION?

22. riends who had nursed their babies.

No influence Little Neutral Moderate A lot of
at all influence influence influence
1 2 3 4 S
23. Friends who had bottle-fed their babies.

No influence Little Neutral Moderate .A lot of
at all influence influence influence

1 2 3 4 S
24. Close female relatives. (Mother, sisters, mother-in-law,
etc.)

No influence Little Neutral Moderate .A lot of
at all influence influence influence
1 2 3 4 5
25. Your husband or baby's father.

No influence Little Neutral Moderate .A lot of
at all influence influence influence
1 2 3 4 S
26. People who came in contact with you prenatally ( health
aides. nurses. nutritionists, etc.)

No influence Little Neutral Moderate .A lot of
at all influence influence influence
1 2 3 4 S
27. Your physician(s).

No influence Little Neutral Moderate .A lot of
at all influence influence influence
1 2 3 4 S
28. Your nurse practitioner or nurse-midwife (answer if
applicable)

No influence Little Neutral Moderate A lot of
at all influence influence influence
1 2 3 4 S
29. Your co-workers (answer if applicable)

No influence Little Neutral Moderate .A lot of
at all influence influence influence
1 2 3 4 5
30. Lectures or childbirth classes you attended.

No influence Little Neutral Moderate .A lot of
at all influence influence influence
1 2 3 4 5

 

64

31. Organizations or other groups. (Please

 

specify

No influence Little Neutral Moderate .A lot of
at all influence 3 influence influence
1 2 4 S

 

32. Other influences. (Please

 

specify )

No influence Little Neutral Moderate A lot of
at all influence influence influence
1 2 3 4 5

 

33. Your age

34. YOur husband's or child's father‘s age

35. Years of school you have completed (Circle one)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 or more
(high school) (college)

36. Years of school your husband or child's father has
completed (Circle)
1 2 3 4 5 6 7 8 9 10 11 12 l3 14 15 16 16 18 19 or more
(high school) (college)

 

37. Are you a smoker? Yes____ No____
Did that influence your feeding decision? Yes
No
38. Did you have any problems with your labor and delivery?
Yes__,
No____
39. Type of delivery: vaginal C-Section
40. Your baby's sex: Male Female
41. Have you had any liveborne children before? Yes
No
42. If you answered yes to 39. did you breastfeed before?
Yes
No

43. If yes, for how long?

 

65

44. combined yearly family income (Check one)
_____1ess than $5,000
$5,000-510,000
510,000-515,000
515,000-325,000
525,000-535,000
535,000-550,000
____;more than $50,000

 

45. Your race or ethnicity:
African-American

 

Native American

Bi-raciall Multi-racial

Other

Thank you very much for your time. Please place completed
surveys in the box in the waiting room.

Asian
Chucasianl European-American
Latino

 

 

APPENDIX B

Consents From Participating Clinics

66

67

To: University Committee on Research involving Human Subjects
Re: Infant Feeding Motivators master's Thesis
From: All sites involved in the study

To Whom it May Concern,

This letter is intended to verify our agreement to participate as a
research study site for the thesis on motivation factors in infant feeding
decisions. ll has been explained to our practice that our commitment will
involve handing out survegs to post-partum women who are 6-8 weeks post-
deliverg, and providing space for a box designed to collect completed
surveys.

A copy of the questionnaire has been provided, as well as the
introduction letter which will be given to all participants. We understand
that the duration of the study will be for two months.

Please provide a brief description of your site
£5 __@a Magda: P!

 

 

 

 

 

 

 

Signature of physician or

C] I [NC SUD?:SOV’
égail 3f/ (12me

Date jéw/yé

 

l'C

lnferCcre

Cbnnnhfibdmhhhofi

Administration:
[0. in 130
”Ciel-baffled
WJAHIBQGI 49013

Tub-e: (616) 427-7937
FAX (616) 427-5180

mantle-Sire.
PD. Ian I”
”.maooia
(616)427-7967
W Band. WIC

whims-wet

loudneuwu

(616! 39943200
Mal, WIC

WdeSee-t
lawn Hubs. M 49022
C616) 927-”
Medal. 0mm!

cvawnnasnu
no. be 397
In Gain. m 4911 l
(616)461-6927
Medial. Wit

”1 W50.
Mica Ml 49047
[616} 782-3374
MC

I705 lessen Blvd. Suite E
Grand Hun. Ml 49417
€616) flee-W7
Mt

68

Pan Steele
22 West 25th Street
Holland, MI 49423

May 28, 1996

Dear Pam,

I'm writing you this letter to follow up on our phone
conversation of today regarding your request to use our
WIC site for your survey. I think it would work out fine
to ask the WIC clients to complete the survey if they
want to. WIC sees about 2800 clients per month. I
enclosed the income guidelines which.ycu.requested. I'll
wait to hear from you as to the dates and details of the
survey.

Sincerely,

(freed

Bna Gunnink
Health Center Manager

69

To: University Committee on Research involving Human Subjects
Fe infant Feeoing Hotivators master‘s Thesis
From- All sites invalved in the stucg

To Whom it may Concern,

This letter is intended to verify our agreement to participate as a
research studg site for the thesis on motivation factors in infant feeding
decisions. It has been explained to our practice that our commitment will
involve handing out survegs to post-partum women who are 6-8 weeks post-
deliverg, and providing space for a box designed to collect completed
survegs.

A copy of the questionnaire has been provided, as well as the
introduction letter which will be given to all participants. We understand
that the duration of the study will be for two months.

Please provide a brief description of your site:
{MW 1 ' )1 ”C. )

 

 

 

 

 

 

 

 

 
 
   

of physician or
ni c supervisor

 

APPENDIX C

Introduction and Consent Letter

70

71

Dear Potential PartiCipant.

I am a graduate student in the College of NurSing at Michigan
State University. My studies include the completion of my
maSter's thesis

I am uSing a questionnaire to examine factors which influence
women in the decision between breast and bottle—feeding. If
you are 5-8 weeks post-partum. and have a well infant. you
are eligible to be involved in this study.

The commitment involved is one-time. The questionnaire will
take about 10 minutes to complete. and can be returned while
you are still in the office today. Participation in this
study is voluntary; you may choose to participate or may
return the unan wered questionnaire without any penalty or
change in the health care provided to you. If you choose to
parti ipate. no info tion which could identify you will be
asked and all responses will be kept confidential. You may
refuse to answer certain questions or discontinue the
questionnaire at any time. You indicate your voluntary
agreement to participate by completing and returning this
questionnaire.

Your help and assistance is greatly appreciated. As a nurse.
I am interested in providing quality care to people and the
Holland communitj. Research in this area will provide
information which can help the nursing profession better care
for women and chil “en in the Holland area. For questions
regarding the study or additional information. you may
contacc me at (616) 396-0252.

Once again. thank you for considering participation in this
studv.

Sincerely.

W M ed, 69:0
Pamela Steele. R N., B.S.N.

APPENDIX D

UCRIHS Approval Letter

72

73

MICHIGAN STATE

UNlVERSITY

July is. 1996

 

ro: Pamela 5. Steele
22 w. 25th
Holland. MI 49423

RE: IRBQ: 96-406

?ZTLE: FACTORS WHICH INYLUENCE THE DECISION T0
BREAST-FEED IN CHILD-BEARING WOMEN

REVISION REQUESTED: N/A

CXTZGORY: l-C

APPfiOVRL 3822: 07/15/96

The university Committee on Research :nvolving Human Subjects'fUCRIHS)
review of this project is complete, I am pleased to adVise that the
rights and welfare of the hum_. subjects appear to be adequately
rotecced and methods to obtain informed consent are appropriate.
erefore. the UCRIHS approved this project and any revisions listed
ve.

2138113: UCRIES approval is valid for one calendar year..beginning wi
tne approval date shown above. Investigators planning to
onti.ue a prozect be ond one year must use the green renews.
form (enclosed with the original approval letter or 'hen a.
project is renewed) to seek u dated certification. There is
maximum of four suon expedite renewals possible. nvestiga'
wishing to continue a project beyond that time need to submi‘
again for complete reView.

RSVISIOHS: UCRlHS must review any changes in procedures involving.human
subjects. “rior to initiation of t e change. I: this is don:
tne_time of renewal, please use the green renewal_tcrm. To
revise an approved protocol at any other time during the yea;
send yopr written request to the. CRIHS C? ir. requesting re°
approva- and rererencing the project's IRE 8 and 31:19.. Inc

your request a description of the change and any revised
instruments, consent forms or advertisements that are app“c‘

---'

   
   
   
   
    

rnosisssl , ,
cannons: Should either of the followin arise during the course of the
worn. investigators must noti-v UCRIHS *romptly: (l) grobler
(unexpected side effects, complaints. etc.) involving uman

subjects or i2) changes in the research environment or new

information indicating greater risk to the human subiects the
existed when the protocol was previously reviewed and approvc

If we can be of any future hel . lease do not hesitate to contact us
at (517)355-2180 or FAX (517l4 2- 121.

mess-21m
FAX smoz-«m cc: Linda Beth Tiedg‘e

HinuunSIMHnuq
Dash-uueaumz
ammunkan

“Weadhummnn
aanflWWmflflfl