• TABLE OF CONTENTS
HIDE
 Title Page
 Acknowledgement
 Table of Contents
 Abstract
 Main














Title: Maintenance of breast self-examination skill and home practice /
CITATION THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00102816/00001
 Material Information
Title: Maintenance of breast self-examination skill and home practice /
Physical Description: Book
Language: English
Creator: Criswell, Eleanor Lee, 1950-
Copyright Date: 1981
 Record Information
Bibliographic ID: UF00102816
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 07843220
ltuf - ABS1435

Table of Contents
    Title Page
        Page i
    Acknowledgement
        Page ii
        Page iii
    Table of Contents
        Page iv
        Page v
        Page vi
    Abstract
        Page vii
        Page viii
    Main
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
Full Text





















MAINTENANCE OF BREAST SELF-EXAMINATION SKILL
AND HOME PRACTICE






BY

ELEANOR LEE CRISWELL


A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL
OF THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS OF THE
DEGREE OF DOCTOR OF PHILOSOPHY


UNIVERSITY OF FLORIDA


1981
















ACKNOWLEDGEMENTS


My largest debt of gratitude goes to H. S. Pennypacker

who chairs my doctoral committee and who generously offered

personal and professional support throughout the course of

this research. Dr. Pennypacker is also the Principal

Investigator of the grant from the National Cancer Institute,

Division of Cancer Control (CA 20791) which funded the

research. I am deeply appreciative to all the other members

of my committee for their interest and guidance: James

Johnston, Cecil Mercer, Yvonne Brackbill, Gerald Stein, and

Edward Malagodi. Dr. Stein served faithfully and expertly

as the medical consultant to this study.

Four women at the Center for Ambulatory Studies-Breast

Section deserve singular mention for their hard work and

devotion to this study. Helen Booth assisted not only with

the mammoth job of phone answering during subject recruit-

ment, but served as an observer in the early weeks of the

study as well. Lilian Dearing was an observer in the

beginning of the study and constructed many of the breast

models used. Sharon Brauns also helped make models and was

an observer during the last half of the study. Gloria

Branscum served as an observer through most of the study and

was an invaluable assist in data checking. These women

worked nights and weekends and cooperated with each other in










order to get the job done. Their contribution to this

study is enormous.

Finally, I received unfailing personal support from

Howard Criswell, Nell Criswell, and William Longo. I am

deeply grateful to them.


iii














TABLE OF CONTENTS

CHAPTER PAGE

ACKNOWLEDGEMENTS..................................... ... ii

LIST OF TABLES.......................................... v

LIST OF FIGURES......................................... vi

ABSTRACT ................................................ vii

ONE INTRODUCTION.................................. 1

TWO METHOD........................................ 8

Participants. ................................ 8
Settings ..................................... 9
Apparatus................................... 12
General Procedures........................... 15
Specific Group Procedures .... ............... 27

THREE RESULTS....................................... 29

Dependent Measures........................... 29
Performance on Wife's Breast................ 30
Performance on Breast Models................ 42
Home Practice. .............................. 50

FOUR DISCUSSION.................................... 55

REFErFENCES NOTES ................ ........................ 62

R FE' Er' jCES... .............. .............................. 63

APPENDICES. ............................................

A RAW DATA FROM TESTS ON THE WIFE'S BREAST
FOR ALL SUBJECTS ............................... 67

B RAW DATA FROM TESTS ON THE BREAST MODELS
FOR ALL SUBJECTS ............................... 83

C AGE, SIZE, BREAST CHARACTERISTICS, AND PRACTICE
AND PROMPT REPORT DATA FOR ALL SUBJECTS........ 99

BIOGRAPHICAL SKETCH ..................................... 108














LIST OF TABLES


TABLE PAGE

1 Demographic data.............................. 10

2 125 ml Breast model mistures ................. 14

3 Mean percent of breast examined during
pretest, posttests 1, 2, and 3............... 32

4 Mean number palpations during pretest,
posttests 1, 2, and 3........................ 35

5 Mean duration (sec) per cm breast tissue
during pretest, posttests 1, 2, and 3....... 37

6 Percentage of subjects palpating wife's
breast to criterion during pretest,
posttests 1, 2, and 3........................ 38

7 Mean number simulated lumps detected during
pretest, posttests 1, 2, and 3.............. 44

8 Mean number false positive responses during
pretest, posttests 1, 2, and 3.............. 46

9 Mean model examination duration (sec) during
pretest, posttests 1, 2, and 3............... 48

10 Mean percent false positive detections
during pretest, posttests 1, 2, and 3....... 49

11 Subject reports of breast exams and spouse
prompts .. ................................. 51
















LIST OF FIGURES

FIGURE PAGE

1 Sample month from take-home calendar........ 16

2 Group median percent of breast examined
during pretest, posttests 1, 2, and 3.... 31

3 Group median number palpations during
pretest, posttests 1, 2, and 3........... 34

4 Group median duration (sec) per cm breast
tissue during pretest, posttests 1, 2,
and 3...................................... 36

5 Group median percent change from posttest 1
to posttest 3 on percent of breast
examined................................. 43
















Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment
of the Requirements for the Degree of
Doctor of Philosophy


MAINTENANCE OF BREAST SELF-EXAMINATION SKILL
AND HOME PRACTICE

By

Eleanor Lee Criswell

June 1981

Chairman: H. S. Pennypacker
Department: Psychology


Procedures for teaching breast self-examination (BSE),

based on training the discrimination of lumps from normal

tissue in a silicone breast model and shaping BSE topography

on the trainee's own breast, have been previously empiri-

cally developed. These real breast plus breast model

training procedures (RBM) have been shown to produce breast

model examination performance superior to that produced by

other training procedures. With 26 husband-wife teams and

26 wives only, the present study examined the effects of

variants of these procedures on the maintenance of actual

BSE skill, directly measured before, immediately following,

and 3- and 6-months after training. Subject reports of home

BSE practice were also examined. Tests on the breast model

were incorporated to extend findings of earlier work.

Results indicate that RBM training is effective in producing


vii










skillful BSE performance. Subjects receiving only machine-

delivered auditory training without actual practice were

significantly less proficient at performing BSE during

follow-up sessions than subjects trained with RBM proce-

dures. Training in subjects' homes with RBM procedures

was associated with better maintenance of BSE skill than was

training in the laboratory. Training husband-wife teams

with RBM procedures was associated with more frequent BSE

practice than was training wives alone. Breast model

testing was sensitive to immediate effects of training, but

performance on the model is adversely affected by interim

events over time. RBM training is effective in a variety of

training conditions.


viii
















CHAPTER I
INTRODUCTION


It has been shown that manual breast examination, either

by a health professional at a yearly check-up or by a woman

at regular intervals in her home is effective in detecting

breast lumps (Greenwald, Nasca, Lawrence, Horton, McGarrah,

Gabriele, & Carlton, 1978; Holleb, Venet, Day, S Hoyt,

1960). Because some lumps may be manually detected early

when still small, it is possible to lower the rates of

radical surgery and death from breast cancer by regular

manual examinations (Hutchison 6 Shapiro, 1968; Wolfe,

1974). Two retrospective studies of women with breast

cancer found that those who reported regular breast self-

examination (BSE) had lower mortality from breast cancer

(Greenwald et al., 1978) and found lumps at a smaller

clinical stage than women in these samples who reported

infrequent or no exams (Feldman, Carter, Nicastri, 6 Hosat,

in press; Foster, Lang, Costanza, Worden, Haines, & Yates,

1978). Because BSE presents no physical risks, it is the

preferred modality for detecting lumps in women under 50

years of age (Shapiro, 1977). BSE is the best way for women

to detect lumps in the interval between professional breast

check-ups (Lesnick, 1977; Strax, 1976).










BSE training procedures are typically informal and

unstructured, and data from one study suggest that even

high risk women perform poorly after training (Wyatt, 1979).

In that study, 138 women with breast disease were observed

performing BSE, and the number of steps performed out of 13

steps which comprise a comprehensive exam was counted. The

mean number of steps performed was only 3.9 despite the fact

that 97.2% of the sample had heard of BSE, and 58% of those

women had previously been trained. The author contends that

if BSE performance of breast disease patients is poor, one

might assume worse BSE performance on the part of other

women. Better BSE training is obviously needed.

Systematic research developing BSE training procedures

has recently been undertaken (Adams, Hall, Pennypacker,

Goldstein, Hench, Madden, Stein, S Catania, 1976; Hall,

Adams, Stein, Stephenson, Goldstein, S Pennypacker, 1980;

Hall, Goldstein, & Stein, 1977; Madden, Hench, Hall,

Pennypacker, Adams, Goldstein, S Stein, 1978; Stephenson,

Adams, Hall, S Pennypacker, 1979; Bloom, Criswell,

Pennypacker, Catania, S Adams, Note 1; Pennypacker,

Neelakantan, Bloom, Criswell, S Goldstein, Note 2). Be-

cause palpating and reporting lump-tissue distinctions are

requisite components in BSE performance, a prototype

silicone breast model was developed to provide a training

medium which a woman could palpate and so experience the

feel of a lump in "normal" tissue (Hall et al., 1977; Madden

et al., 1978). Using this model, it was found that subjects










trained in palpation improved in detection performance

(Adams et al., 1976; Stephenson et al., 1979). In fact,

after training, most subjects found significantly smaller

lumps in the breast model than lumps usually presented by

women to physicians, an important advance in breast-

examination training. Another study found that trainees

taught palpation on the breast model improved in their

performance of detecting known lumps in live stimulus

women (Hall et al., 1980). Subsequent psychophysical

analyses led to the idea that breast models might be even

more effective in training if they approximate the feel of

the breast of the trainee (Bloom et al., Note 1). With

training procedures improved and refined, a training

comparison study was undertaken which found that subjects

taught breast examination by procedures which lead them to

feel their own breast and compare their tissue to normal

tissue and lumps embedded in a silicone breast model produce

better performance measured on the breast model than state-

of-the-art BSE training offered by the American Cancer

Society or training procedures which focus only on the

breast model or on the breast tissue alone (Pennypacker et

al., Note 2).

In addition to data concerning the immediate effects of

training, some data were available concerning the long-term

effects of training on performance on the breast model.

Bloom et al. (Note 1) recalled two subjects for testing six

months after several months of extended training and practice.










An increase in false alarm responding was observed, but

threshold data revealed little decrement in detection scores

as a function of lump size and hardness. Adams et al.

(1976) recalled five subjects for follow-up testing three

weeks after three to six daily sessions of practice on the

model. Detection data showed that performance remained

stable for three subjects and decreased slightly for the

other two subjects.

Finally, 58% of subjects who received only one 60-minute

session of various types of training and practice returned

for testing six months after training (Pennypacker et al.,

Note 2). Following very little practice and a long recall

interval, performance of subjects from all training groups

decreased in quality over the six-month period, but remained

better than pretest levels. Subjects trained in breast

examination by either combined real breast-plus breast

model procedures or real breast alone procedures did not

show significant performance decrements, but subjects

trained on the breast model alone did, suggesting that real

breast training promotes long-term maintenance of examina-

tion skill. Further, subjects in the real breast plus

breast model group were significantly better examiners than

subjects who had been trained with the American Cancer

Society pamphlet or who had received no training.

Thus, the effects of training procedures on breast

model performance had been studied, but no work had been

done concerning the effects of training on actual BSE






5


performance. One goal of the present study was to measure

these effects directly. To that end, a direct system of

measuring BSE performance was developed. The present study

also sought to measure decrements in BSE performance over

time.

A recent survey indicated that 77% of 772 randomly

selected women over 20 years of age had performed BSE, but

of those, only 22% reported regular monthly practice

(Gallup Organization, 1980). An additional interest of the

present study was to monitor subject self-reports of home

practice and to determine if training conditions would be

associated with frequent home BSE practice.

Further, the present study incorporated breast model

testing to provide replications and extensions of earlier

works and to systematically investigate the effects of

training on model examination performance three and six

months after training with a larger sample. The present

study also employed real breast plus breast model training

conditions and a training condition comparable to the

pamphlet condition used by Pennypacker et al. (Note 2)

except that the information that these subjects received

described real breast plus breast model training procedures

without providing practice.

The maintenance of behavior change acquired in train-

ing programs has been shown to be determined by environmen-

tal variables and is amenable to experimental study (e.g.,

Koegel & Rincover, 1977). The problem of decrements in










subject performance is the "Achilles heel" of psychology

(Atthowe, 1973). Although most training and research

programs fail to collect data on the long-term effectiveness

of their interventions (Hayes, Rincover, & Solnick, 1980),

many researchers note recidivism, relapse, and absence of

the desired behavior at follow-up sessions (Kazdin, 1975).

Baer, Wolf, and Risley (1968), Kazdin (1975), and Stokes and

Baer (1977) state that it is naive to assume that behavior

change persists without special maintenance programming, and

there are variables which may be manipulated to affect

behavior change maintenance. The suggested strategy is to

teach responses which will be reinforced by the subject's

natural environment, this may involve training or manipula-

ting aspects of the natural environment, and then design

stimulus events into the training session which prepare the

subject to emit the desired response in the non-training

setting (Criswell, Note 3). Thus, experimental manipula-

tions of components of BSE training might produce

differential effects on BSE performance, breast model

examination performance, and home BSE practice measured over

a six-month period.

In the present study, subjects were trained to perform

BSE, and stimulus conditions were incorporated into the

training session to prepare the subject for at-home practice.

Mode of training was varied; subjects were trained either by

real breast plus breast model procedures or by listening to

a tape-recorded description of the procedures. This was










done to permit a comparison between skill maintenance and

frequency of practice following proven good training and

training which might not produce the same quality perform-

ance. Some subjects participated with their husbands to

ascertain if these men, trained in BSE, would exert

discriminative or reinforcing control over their wives' BSE

performance, and thus increase the number of exams their

wives performed relative to wives with untrained husbands.

Finally, setting of training was varied. Some subjects

were trained at home in their beds to determine if the use

of more natural BSE training settings would produce more or

better BSE performance than would be exhibited by subjects

trained in the lab.
















CHAPTER II
METHOD


Subjects were recruited through advertisements placed

in local newspapers which requested persons over 21 years of

age to call for information about participating in a

research project. Selection criteria included: (a) married,

living with spouse, (b) expected to live in Gainesville for

the next six months, (c) did not perform breast exams as a

profession, (d) not pregnant or breast-feeding, and (e) not

formerly a subject at this laboratory. Subjects were

assigned to groups in rotating fashion, and six groups were

filled. Group 1 consisted of husband-wife teams who

received real breast plus breast model (RBM) training, and

all sessions took place in the lab. Group 2 consisted also

of husband-wife teams who received RBM training, but all

sessions took place in the subjects' homes. Group 3

consisted of husband-wife teams who received tape recorded

training, and all sessions took place in the lab. Group 4

consisted of wives only-their husbands were not involved at

all-who were trained with RBM procedures in the lab. Group

5 consisted of wives only who received RBM training; all

sessions took place in the subjects' homes. Group 6

consisted of wives only who received the tape recorded

training, and all sessions took place in the lab.









Of 91 subjects participating in Session 1, 78, or 86%

remained eligible and participated in all three sessions.

Data analysis is based on these 78 subjects. All subjects

in Groups 1 (n=20) and 5 (n=ll) remained eligible and

participated in all three sessions. One couple in Group 2

could not be located after Session 1, leaving nine couples

in this group. One couple in Group 3 quit because of the

wife's extended illness, and another couple separated from

each other after Session 2 and were thus ineligible, leaving

seven couples in this group. In Group 4, one woman when

contacted to schedule Session 2 stated that she did not want

to participate further because she had been slightly

embarrassed during Session 1, and two more women separated

from their husbands during the study, so 7 of 11 are

included in data analysis.

Table 1 presents demographic data for the six groups

and shows that the groups were fairly well balanced

demographically. Women in Group 5 were slightly older and

slightly larger-breasted while wives in Group 2 had firmer

breasts possibly making BSE more difficult. Risk factors

(over 40 years of age or with history of a breast problem)

were fairly evenly distributed across groups.


Settings

Laboratory Setting

Interviews, tests on the breast model, tape recording

training, and debriefing were conducted in a small office in












Table 1. Demographic data



Group n Median Age Number
age range subjects
over
age
40



1. Teams-RBM-lab
Wives 10 31 22-62 2
Husbands 10 32 26-81 3


2. Teams-RBM-home
Wives 9 30 22-55 2
Husbands 9 33 26-59 3


3. Teams-tape-lab
Wives 7 29 23-62 2
Husbands 7 33 24-68 2


4. Wives alone-RBM-lab 8 37 23-62 4


5. Wives alone-RBM-home 11 48 25-64 8


6. Wives alone-tape-lab 7 32 21-41 1


al=Soft; small amount nodular tissue
2=Medium firmness; average amount nodular tissue
3=Firm; nodular tissue predominant.











Table 1-extended


Number of subjects Number of Median breast characteristics
who had sought subjects size (cm) firmnessa nodularitya
medical attention reporting
for a breast monthly
problem BSE




6 (1 biopsy) 1 57.5 1.5 2
1 0



3 0 54.0 3.0 2
1 0



1 2 52.0 2.0 2
1 (biopsy) 0


5 (4 biopsies) 1 55.5 2.0 2


6 (1 biopsy) 2 62.0 2.0 2


3 (2 biopsies) 1 53.0 2.0 2









the Psychology Building. Tests on the wife's breast, breast

measurement, and RBM training were conducted in a nearby

room equipped with a storage cabinet, a sink, and a standard

medical examining table.

Home Setting

Interviews, tests on the breast model, and debriefing

were usually conducted in living or dining rooms. Tests on

the wife's breast, breast measurement, and RBM training

occurred in the subject's bed. There was wide variability

in stimulus conditions across homes, but relatively little

variability within homes across sessions.


Apparatus

Breast Models for Testing

The test model consisted of two silicone sections each

of 140 ml volume horizontally positioned on top of each

other. The diameter of each circular section was 14.5 cm;

it was about 1 cm deep, and the top section had a small

silicone nipple in the center on its top side. Each section

was made by hand-painting a steel mold with a thin (.5 mm)

layer of well mixed and aerated liquid silicone (Dow Corning

MDX4-4210 with 10% curing agent), curing this 30 min in a

3500F oven, filling it with 93.5 g liquid silicone (87.2 g

GE RTV 6159 with 6.3 g or 7.2% curing agent) and 46.5 g

simulated nodular breast tissue. Nodularity was made by

chopping or "scrambling" a 150 g batch of cured silicone

(133 g GE RTV 6159 with 17 g or 11.33% curing agent). The

filled skin was then cured. Lumps hand-cut from foam and










sponge materials were then placed into the cured mix, and a

flat skin made the same way as the mold skin but thinly

painted (.5mm deep) and cured on an 18 x 18 x .5 cm metal

plate was then glued on to become the back of the

model. After a final curing, the model was ready for use.

Take-Home Breast Models

These models were single-section 125 ml hemispheres

constructed in the same general way as the test models

except the steel mold was hemisphere-shaped. Each model

contained four lumps: 10 and 5 mm soft lumps and 7 and 5

mm hard lumps. Multiple models of three firmnesses (soft,

medium, and firm) each with three amounts of nodularity

(small, medium, and large) were constructed so the breasts

of subjects could be matched to a model. Table 2 gives the

mixtures used to create each of the nine breast models.

These mixtures were determined empirically and selected

because they represent the range of breast characteristics

most often observed in previous studies.

Real Breast Testing Apparatus

A 10 x 10 cm grid, partitioned into 100 squares, was

printed onto a transparency. Squares were numbered from

left to right, row by row, from 1 to 100, and square 46 was

marked for projection onto the wife's nipple. Projection

was done with a 3M overhead projector model CCAG.


1See Bloom et al. (Note 1) for procedures used to determine
hardness.











Table 2. 125 ml Breast model mixtures


Grams Grams
Amount of nodu- Grams curing
Firmness nodularity larity RTV gel agent



Soft small 12.5 105.44 7.06
(10% of volume)

medium 25.0 93.72 6.28
(20% of volume)

large 37.5 82.01 5.49
(30% of volume)


Medium small 12.5 104.90 7.60

medium 25.0 93.30 6.70

large 37.5 81.70 5.90


Firm small 12.5 104.46 8.04

medium 25.0 92.85 7.15

large 37.5 81.24 6.26










Home Calendars

A Calendater style 1010-0 12-month calendar was

modified by printing choices with checkblocks for subject

self-reports on each day. Figure 1 presents a sample month

from the calendar.


General Procedures

Sessions

Each subject participated in three sessions; Sessions 2

and 3 were held at 3-month intervals after Session 1. During

Session 1, the subject first signed the approved consent

form and payment contract, then was interviewed to obtain

demographic and pertinent personal data. Baseline data on

reported frequency of BSE was obtained at this time. If the

subject's husband was involved, he also signed forms and was

interviewed with his wife. Following this, the subject's

skill at breast examination was pretested in two ways.

First, she examined five breast models, then was taken to

the examination room (or her bedroom at home) where she

performed a breast exam on herself. If the subject's

husband was involved, he was pretested on the breast models

in another room by another observer while his wife was being

administered the same test. While the wife examined her own

breast during the pretest, the husband waited outside the

room. When the wife's pretest was concluded, the observer

called the husband in, and he was asked to perform a breast

exam on his wife. After the protests, the subject was






























SUN MON


May 1980

TUE WED! THU FRI SAT


1
w SSE 0
H ex W "0
W Wk H '0
H ok W .0
aomr 0


4
W ISE IO
H J w0
HeW k H
amer 0
oM a


5;
W SE [0
H M w 0
W Sk N H C
H Wk W 0l
omer 0


6
W SE O3
H n W 0
W ak Hw 0
ho CW 0
omer 0


7i
W ISE I0
H n W tO
W ask HM.
H Uk W 0
Who 0
Swgo
MMWO


8i
W BSE 0

W w '0
amer 0


2
WBSE C0
H MW 0
W Or n
H ak W, 0
omer n


9
W ISE 0
H w 0
W Uk mH 0
H -k W Q
Omer 0


3
W SE
MeW
H Ir W
W Uk H
H Uk W
.mtr


10
W SSE
H x W
W Uk H
H Uk W
oWer


11 121 13 14 15 16 17
0 wWSE W SE 0 WBSE 0 0 WSE E WSE o w SE
0 MH W 'O MHaW HnexW HOxW r' Hu.W D M W On H aW
0 HUW ;O 0 WkHO W k a WH0 COW k H W!O k MHI W
a3 1H k W 0 H wkW 10 1akW o H 1U W 0 H Wk W 0 H uk WI H 4k W
0 aMr 0 amer 3 omer O mer D 0o0M 0 0omer 0 oer

18 19j 20 21 22 23 24
O WSE 'o WSSE 0 WISE WIBSE o0 W SED wSE D WISE
SH*tWo i Hww o H0 Maw m .w )O Haw T0 H.w 'o Hmw
0 WwkH ; O WMH 0 WH-* W WkH O WknHOl W k H 0 W H
0 Hwkw C3 Mwkw0 W 3 HIW oa MH.w 0 H1Okw 0 H-awkw O kw
0 omr 0 aoar 03 awr o omer 0 mer 0 o Mer omer


25
W BSE 0
HaW 0
14 k W W0
w 7


26
W BSE '
HaW '0
W Uk H 0
H Uk W 0
other 0


271
w ISE 0
H1n W | O
W ,k H tO
H *k O
Mer 0


28
W SE 0O
H xW I0
W Uk H '0
H U.k W 0
oter 0


291
W BSE 0
H aW -0
W -kH O
H ak W i0
aen r |0


30
W ISE .
MaHW '0
W OakH C


31
W BSE
H O W
W OWN H
H Uk W
mamr


Figure 1. Sample month from take-home calendar










trained, and if the husband was involved, he participated

in the same training session with his wife. Following

training, the subject received a brief break. After the

break, the subject's skill was posttested. First, she

performed a breast exam on herself. Then she dressed and

was lead to the office (or other room in the house) and

asked to examine five breast models. If her husband was

involved, he waited outside the room while the wife was

examining herself, then he was called into the room by the

observer, and he performed a breast exam on his wife. He

then examined the five breast models while the wife was

administered the same test, but he was tested in another

office (or room in the house) by another observer. Finally,

the subject was given a calendar and instructions for the

interim, was paid and then dismissed. If her husband was

involved, he was given a calendar and instructions at the

same time in the same room with his wife. Each subject and

each husband, if involved, received $5.00 for completing

Session 1.

At Session 2, the subject turned in her calendar and was

interviewed briefly to gather pertinent interim information.

Using Session 1 procedures, she was then posttested on her

own breast and on the breast models. Finally, the subject

was given back her calendar, received instructions for the

next three months, was paid and then dismissed. Each subject

and each husband, if involved, received $5.00 for completing

the session plus $5.00 for turning in the calendar. If the











husband was involved, he participated in the session as he

had participated during Session 1.

Session 3 was conducted like Session 2 except that the

subject was re-interviewed after testing and then debriefed.

If the husband was involved, he was tested as he had been

during previous sessions, and he was included in the final

interview and debriefing along with his wife. Each subject

received $5.00 for completing the session plus $5.00 for

turning in the calendar, and her husband was similarly paid

if he was involved.

Real Breast Testing Procedures

The bare-breasted woman was supinely positioned on the

examination table (or on her bed at home) and her non-

preferred arm raised over her head so that her upper arm was

perpendicular to her body and resting on the table. The

overhead projector stood on a table 100 cm from the edge of

the examination table and was the same height as the

examination table. With the projector light illuminated,

the grid was aligned with square 46 covering the nipple of

the breast on the nonpreferred side. The grid was aligned

with a horizontal grid line straight across the clavicle.

The projector was then focused and the grid taped into

position. The observer's gridded score sheet was marked to

indicate total breast area; the borders were straight lines

at the clavicle, midline, braline, and axillary line. Care

was taken at subsequent test occasions to keep the total










breast area marks constant across tests. In the home

settings, the projector was positioned as close to 100 cm

away as possible, and positioning was constant across test

occasions. The subject had no informed knowledge of the

observer's grid or the observer's scoring task, nor could

she see the grid projected on her breast. Most husbands

could not see the projected grid and husbands had no

knowledge of the boundaries marked on the observer's score

sheet. Husband performance did not appear to be cued by

the grid.

The observer read the instructions after positioning

the subject. They were: "When I say 'GO,' examine your

(your wife's) (left or right, contralateral to preferred

side) breast as best you can. Use your (preferred) hand.

Tell me when you have finished. Do you have any questions?"

Observers answered only those questions which could be

answered by repetition of the instructions. One breast exam

constituted the test. During the test, the observer placed

a red mark on the score sheet in the square of the grid

corresponding to the location of the top of the subject's

middle finger, one mark for each palpation. The observer

also recorded exam duration (read from a stopwatch) and made

notes about hand topography. When a subject's husband was

involved, he waited outside the testing room while the wife

was tested, then he performed his exam while standing by the

examination table (or sitting on the bed at home) on the

wife's nonpreferred side. The same breast was examined by










both spouses and both examined with their preferred hand.

The wife was asked not to speak to her husband during the

test except to indicate discomfort.

Breast Model Testing Procedure

The subject and observer sat opposite each other at a

small table. Secured on the table was a 1.0 m square

vertical black felt screen (with armslits) which prevented

the subject from seeing the lumps in the transparent models.

Portable cardboard units were constructed for home testing

and placed on a suitable table in the home. The detection

bell was located on the table in front of the subject's

nonpreferred hand. All husbands were tested by the

experimenter (EC), and wives tested by another trained

observer.

When subject and observer were comfortably seated, the

observer read the following instructions:

Are you right or left-handed? I will be
sitting behind this screen. Please put
your (preferred) hand through the curtain
and rest it here off to the side while I
read the instructions. There is a breast
model behind the screen. Notice that the
model is circular, and there is a nipple
in the center. Do you feel that? Now
rest your hand off to the side. The breast
model may or may not have lumps in it.
When I say 'GO,' examine the model with
your (preferred) hand as best you can.
Take your hand off the model when you have
finished examining it. Here is the bell
(observer rings it). Every time you find a
lump, ring the bell with your (other) hand
(subject rings bell for practice). Do you
have any questions?

The observer answered subject questions by repeating










pertinent directions but additional instruction was not

given. As the subject examined the model, the observer

recorded true detections, false positive detections, dura-

tion of search, and made notes on search topography. When

the subject finished examining and took her hand off the

model, the observer recorded exam duration, changed the

model, and prepared for the next trial. Five trials

constituted the test, Trials number 3 on the pretest and

number 4 on the posttests were catch (no-lump) trials. On

the remaining four trials, four lumps were presented on each

trial in combinations of two large and two small, two hard

and two soft, and two deep and two surface lumps such that

each different lump was presented twice during the test

(e.g., the large soft lump was presented twice on the

surface and twice deep).

Observer Training

Four female observers (three lab assistants and the

project secretary) were individually trained and tested.

The experimenter planned BSE performances by determining

values for the variables of interest. These values served

as criterion or reference scores. Each observer practiced

recording trial data while the experimenter examined her own

breast according to the plan. Each observer scored a number

of different trials, and each was considered trained when

her scores matched the criterion scores. In addition to

initial teaching and testing sessions, periodic testing or

calibration sessions were held, and remedial training was










given if the observer's test scores fell outside an

acceptable range. In all, 109 such calibration trials were

administered to the four observers. In order, observers A,

B, C, and D administered 20%, 6%, 11%, and 62% of the 363

real breast tests to subjects and scored 22%, 16%, 23%, and

39% of the calibration trials. Calibration trials were

analyzed to determine if the observer scored correctly the

performance components of interest: use of three fingers,

vertical strip search pattern, small circular finger motion,

and systematic application of light and deep pressure.

Agreement with criterion scores was obtained on 92%, 100%,

94%, and 94% of trials for these measures respectively.

Between 95% and 105% agreement on number of palpations (a

continuous stroke at any location) used during the calibra-

tion trial was obtained on 87% of the trials.

Additional calibration trials were run during which the

experimenter stopped palpating, held her hand in place, and

checked that the observer recorded on the gridded answer

sheet the correct location of the palpation. Observers were

consistently accurate in their recording. When errors

occurred, the location mark was placed in an adjacent square.

The same four observers plus the experimenter adminis-

tered breast model tests. Observers B and E were skilled at


iThis type of observer training permits the use of single
observers. Because the observer's behavior has been cali-
brated against ("true") criterion values, a second observer
and interobserver agreement calculations are unnecessary.
For a full discussion of the advantages of using calibrated
single observers instead of multiple observers of the same
response, see Johnston and Pennypacker (1980).










breast model testing from other experiments, and the other

observers were individually trained until proficient.

During calibration trials, the experimenter examined the

model in a predetermined way, and the observer scored the

trial. Observer scores were then compared to the experi-

menter's criterion scores. In order, observers A, B, C, D,

and E (EC) administered 12%, 6%, 6%, 41%, and 35% of the 373

breast model tests to subjects and scored 10%, 0%, 40%, 49%,

and 0% of the 99 calibration trials. Scores of +1 for true

detections and +1 for false positive responses were obtained

on 97% and 97% of the calibration trials.

Breast Measurement Procedures

At Session 1, the breast on the nonpreferred side of

each woman was assessed for firmness and nodularity. The

experimenter palpated the supine woman's bared breast and

matched it to one in a series of 125 ml breast models which

varied systematically in firmness and nodularity. At

Session 3, the same breast was measured for size with a

metric tape measure with the woman standing. A straight

vertical line was measured from the clavicle across the

nipple to the braline and another line measured from midline

to axillary line around the fullest part of the breast.

These two measures summed represent breast size.

Real Breast Plus Breast Model (RBM) Training Procedures

After the pretest on the woman's breast, the observer

left the room, and the experimenter (EC) entered to begin

training. First, she taught the subject the feel of normal










tissue in the subject's breast and in the breast model. A

125 ml breast model which matched the subject's breast in

firmness and nodularity was selected at this time and used

throughout in training; the subject kept this model for at-

home practice. Next, the subject was taught to hold three

fingers together and examine the tissue with the tips

(flats) of the fingers. A circular finger motion was taught

which included three small circles at each spot examined,

one circle of light pressure to find surface lumps, and the

next two circles with deeper pressure to find deep lumps.

Next, lump characteristics were demonstrated using the

simulated lumps embedded in the breast models. The subject

was guided to feel hard, soft, large, small, fixed, movable,

surface, and deep lumps. The subject was taught to distin-

guish these lumps from her own normal tissue by examining

small areas of her breast and comparing her tissue to lumps

and normal tissue in the breast model. Finally, the verti-

cal strip search pattern was taught. The subject was told

to begin the exam at the junction of the midline and

collarbone, to examine using the circular motion down to the

braline in a straight strip, to then move in about an inch

toward the nipple and examine in another straight strip back

up to the collarbone, to move in and examine in another

strip back down to the braline and so on until the last

strip, the axillary line, was completed. The importance of

examining every bit of tissue was stressed. The subject then

practiced examining her own breast as the experimenter










praised or corrected examination technique. If her husband

was involved, he was trained along with his wife at the same

time. Training lasted about 25 minutes.

In the event that a subject asked a question about her

own breast tissue that might have been a lump, the experi-

menter would palpate the tissue. In instances where the

subject's tissue was not at all questionable, the experi-

menter would say so and add that she was not a physician,

and the subject was advised to consult her physician if she

was still concerned. If there was any doubt at all about

the tissue, the experimenter would advise that the woman

consult her physician. The experimenter answered questions

calmly, did not overrepresent her skill at distinguishing

lumps from normal tissue, and did not imply that the woman

might have cancer. Such subject questions arose during

testing, training, or during interviews at Sessions 2 and 3.

All subject questions were later discussed with the project

medical consultant, and subject concerns were properly

documented and followed-up by phone. Had a subject refused

to seek proper medical attention, she would have been termi-

nated from the study; however, this did not happen.

Tape Recording Training Procedures

After the real breast pretest at Session 1, .the woman

dressed and was taken into a small office. The experimenter

said: "You will now hear a tape recording which teaches you

how to do a breast exam. I will leave you alone and come


Gerald H. Stein, MD










back when the tape is over. Please push this stop button

when the tape is over." The subject then heard an 8-minute

tape recording of the experimenter explaining BSE proce-

dures. The information on the tape was similar to that

given subjects in the RBM groups, but the subject did not

receive guided instruction or practice on a breast model or

on her own breast. About ten minutes later, the experi-

menter returned, gave the subject a short break, then began

the real breast posttest. Few subjects asked questions,

but if they did, pertinent portions of the tape recording

were repeated. If husbands were involved, they listened to

the tape recording along with their wives.

Calendar Procedures

Each subject received a calendar and instructions

concerning the calendar were repeated at each session. The

subject was told to keep the calendar and a pencil near the

bed to make recording convenient, to check "W BSE" whenever

she examined herself, "H EX W" whenever her husband

examined her, "W ASK H" whenever she asked her husband to

examine her breasts, "H ASK W" whenever her husband asked

her if he could do her breast exam or whenever he reminded

her to examine her own breasts, and "OTHER" whenever she saw

or heard something that reminded her of BSE. If her husband

was involved, he was given a calendar and instructions at

the same time. The subject was told that standard medical

advice suggests performing BSE once a month, but that she

should practice BSE as often as she wished. The subject was











reminded that the experimenter was not looking for some

special pattern or combination of calendar marks, that she

was only interested in what actually happened.

At the beginning of Session 2, the calendar was

collected and checkmarks transcribed to a raw data sheet

while testing was in progress. When the calendar was

returned at the end of the session, the experimenter did not

question or comment on calendar marks and said only: "Your

instructions are the same for the next three months as they

were for the first three months" and reviewed the instruc-

tions. At the beginning of Session 3, calendars were again

collected, and at the debriefing, discussed in greater

detail. The subject was given a new calendar before leaving

the study.


Specific Group Procedures

Group 1: Teams-RBM-Lab

Husband-wife teams were recruited, and all sessions took

place in the lab. Training at Session 1 was accomplished by

RBM procedures previously described, and Sessions 2 and 3

were conducted as described above. At the end of Session 1,

the team was given the small breast model which was matched

to the wife's breast and used during training. They were

told to practice with the breast model as often as they

wished. Each spouse kept a self-report calendar as described

above.










Group 2: Teams-RBM-Home

Procedures for this group were identical to Group 1

procedures except that all sessions were held in subjects'

homes.

Group 3: Teams-Tape Recording-Lab

All sessions for this group were run in the lab. The

team was trained at Session 1 by listening to the tape-

recorded training. These subjects never saw or felt a

breast model (except during model tests), and they did not

receive a model for use at home. Each spouse kept a self-

report calendar as described above.

Group 4: Wives Alone-RBM-Lab

The procedures for this group were similar to Group 1

except that the spouse was not included in any session nor

was the subject informed that the behavior of her spouse was

of experimental interest. All sessions were run in the lab.

Training was accomplished by RBM procedures. The subject

took home the training model matched to her breast and was

instructed to keep a self-report calendar.

Group 5: Wives Alone-RBM-Home

Procedures for this group were the same as for Group 4

except that all sessions were run in the home.

Group 6: Wives Alone-Tape Recording-Lab

Procedures for this group were the same as for Group 3

except that the spouse did not participate in the sessions,

and the subject was not informed that her husband's be-

havior was of experimental interest. Each subject kept the

self-report calendar.















CHAPTER III
RESULTS


Dependent Measures

The dependent variables of special interest in this

study were selected to reflect: (a) skill at performing

BSE, (b) skill at detecting breast lumps in silicone breast

models, and (c) frequency of breast exams performed at home

over the 6-month period following BSE training. Measures

from real breast testing include: (a) area of breast ex-

amined, or the ratio of the number of grid squares examined

to number of grid squares covering the total breast area,

(b) duration of examination divided by breast size, (c) num-

ber of palpations, and (d) verbal descriptions of search

topography. Measures from breast model testing include:

(a) lumps detected, (b) detection responses which were false

positives or inaccurate, and (c) duration of examination.

Calendar measures include: (a) number of self-examinations

performed by the wife, (b) number of husband-performed

breast exams on his wife, (c) number of husband-delivered

verbal prompts to his wife to do BSE, and (d) number of wife

prompts to husbands.

Baseline data were collected at Session 1 during pre-

tests on the breast model and on the wife's breast and the

oral interview. Posttests immediately after training and











3- and 6-months later provided follow-up data. Frequencies

of home breast exams and prompts were collected from the

self-report calendars kept by each subject. Because many

subjects stated that they did not reliably report "OTHER",

these data are not included in analysis.


Performance on Wife's Breast

Figure 2 presents the group median scores for percent

of breast examined during all tests. Immediately after

training, the median score for area examined for all sub-

jects was about 75% of the breast. Husbands trained with

the tape recording improved to only 51%, but all similarly

trained wives improved to 75%. Subjects trained with RBM

procedures improved to between 76% and 83%. Wives trained

at home with RBM procedures but without their husbands were

the least improved on this variable during the first post-

test. Mean scores for this variable are presented in Table

3. An analysis of variance revealed no statistically sig-

nificant difference among the groups on this variable on the

first posttest scores, but the overall mean score for all

RBM groups was higher than that for all tape groups (t=2.90,

df=76, p<.05).

The largest immediate differences in performance be-

tween RBM- and tape-trained subjects were that most RBM-

trained subjects examined more breast area, palpated more,

spent more time per cm breast tissue, and used the proper

search pattern during Posttest 1. The median differences





























100-
o
wO
z 80-

x 60-
LJ
40-
z
o 20-


u'_ _ _ _ _ _ _ _ _ _


All
Subjects


RBM Tope Teom Alone RBM R8M
Lab Home
GROUPS


Teoms-RBM-Lob Teoms-RBM-Home WivesAlone Wives Alone
Wives Husbonds Wives Husbands RBM-Lob RBM-Home


Wives Aone Teams-Tape-Lob
Tape -Lob Wives Husbands


GROUPS


Figure 2. Group median percent of breast examined during pre-
test, posttests 1, 2, and 3


100-
0
w
z 80o

x 60-

S40-

S 20-
2


v


/I--/r*






















(. (D zr co (D O ULn' Ln c-



Ln CN CD L CC.4 0 0C Dl uL (N
"c- LO C O ['-( CD C l) LC) [-





'-i (.D LLC c C (Y) C C CC L.n
H- CC CN -I ,-'I C q (N -- ,-


C Nc n cH--- Ln u C c) Ln -
- CN CC CC U-c) ( c( --






H C CO C) CN C)O H CO l
H H^ (N H (N H q (N H H


C (D Co CN Ln co co co 1- 4
CO C(- U CO C-- D 1"- D cO [I-






CO CO CL CO (N 0 (N (N C CC





-. -.2- -t mO LO r --- :-- C:f -.


i)
A E
0 0




I I



SH H


Ea : 3



Cd > >
ci) "- o


a C

cc

00
a0 W
-, a
0
c ho
O

(
m


r> H

* <


L)








cc) (N

S0
)0 0 1
to E




CO M m
)
nO A E



4- s s



H H H
H H H
'ci~ :


ra E
0 o
ri c
H s

M m
C Ci


E e

0) 0)


HI (N m C) Z-











on these variables may be seen in Figures 2, 3, and 4. Fi-

gure 3 presents group medians for number of palpations on

each of the four test occasions and shows that subjects

trained with RBM procedures consistently used more palpa-

tions. Mean scores for this variable are presented in

Table 4. Differences between all RBM- and tape-trained sub-

jects were statistically significant for this variable on

the first posttest (t=4.48, df=76, p<.05). Wives trained in

the laboratory with RBM procedures were especially improved

when measured immediately after training.

Figure 4 presents group medians for exam duration (sec)

per cm breast tissue for each of the four tests. As shown

in Figure 4, subjects trained with RBM procedures spent more

time examining each cm of tissue than subjects trained by

the tape recording during the first posttest. Mean scores

for this variable are presented in Table 5. Statistical

tests of significance revealed across group differences for

Posttest 1 scores (F=5.06, df=5,72, p<.05) and differences

between the RBM and tape groups scores (t=3.32, df=76, p<.05).

Any training, especially RBM, slows the hand's movement.

Table 6 presents the proportion of subjects in each

group observed palpating to criterion with respect to the

four topographical variables of interest: using three fin-

gers, searching in a vertical strip pattern, moving fingers

in a small circular motion, and systematically applying

light and deep pressure at each small area of breast

































(i


RBM Tope Team Alone RSM-Lob RSM-Home


GROUPS


Teoms-RBM-Lob
Wives Husbands


Teams-RBM-Home WivesAlone WivesAlone
Wives Husbands RBM-Lab RBM-Home


Wives Alone Teams-Tope-Lob
Tope- Lob Wives Husbands


GROUPS


Figure 3.


Group median number
posttests 1, 2, and


palpations during pretest,
3


V,
z
80-

cr
w
J40-
z
z 20-
u 0.


r


i


:2-
j


Z\I






























C 0












cn o
,4 i









CC Co






0- C)
T-1











co co




Ln n





co o
CM CO
T Cl14


(O






L)






CO CC
S CN








C C
c (N
























H-t CO






CN CO
cC UL)













CC C
(n


C E nC
w m H
H a, I
I i 0)
s m
E C

0 H I I I
H Ac I 0) 0) 0)
I I 0) C C C
Z. 0 0 0
pa m (C H ,- -


M M 0 0 0
E E E ,-
o m > > >
0) 0) 0 *- *-H *H
- H EH 3 S: S


Ln





CCM






C 3
0
0 4





H H I
H H H
x0 C-14 n
mr- i-
dC)P

s &. s


Ln uc


Cl)


C
0)


in Ln
o o
CO O

\ V







II 1I














O O
II II




I I
H H









o 0
IOI II







o &


0 0





























I


u I_ _ _ _ _ _ _ _


All RBM Tape Team Alone
GROUPS


RBM- Lob R8M-Home


Teams-RBM-Lob Teams-RBM-Home WivesAlone WivesAlone
Wives Husbonds Wives Husbands RBM-Lob RBM- Home


Wives Alone Teams-Tope- Lob
Tope-Lab Wives Husbands


GROUPS


Figure 4. Group median duration (sec) per cm breast tissue
during pretest, posttests 1, 2, and 3


E
E 40.
U
V) 30-
z
-i
20-
w
10-


6.0-
50-
E
S40-
z 3.0-
2 20-
1.0-


'


v


P


i'

























C CH 0 ) 0 H 0 0 C

S 0"0O C4 O C-I O O

4 r 4 0- N N N O C 4 CO




CN







m












0
O U O C H O O H O



SC C- CDI LC 0 o Co

co Dc 4







































SO ,-. I. I O
H H H H 0 H H






U) CO I o U. C U) N C n 3 co C




S)0 0 0
C O0 iO ( C) O Co 0 co cC C



r0 D ON N
u) c4 4Ii4




0








o




C0 E C) .o &
o U) 0 Hl C C- 0 03
*H H H L- 0 4 4
4- I I (I M 0 !. CD C
) U) 0 H- I I 0 1 U U



0 I I 0 0 0 ? O U) 0
C l l ) H H H

I +I S U S O -4



: CUl C 0 C 0 0
CU a Ea E-i 3 1 4 < <

1 01
F-' F H) N CO D 0 0
Ca E- E- :: :3 3
cl0







38





Table 6. Percentages of subjects palpating wife's breast to criterion
during pretest, posttests 1, 2, and 3



Groups 3 Fingers Vertical Strips
Pre P1 P2 P3 Pre P1 P2 P3


1. Teams-RBM-lab 65 95 75 85 0 95 80 90

2. Teams-RBM-home 56 100 78 72 0 100 94 83

3. Teams-tape-lab 50 86 71 71 0 50 43 36

4. Wives alone-RBM-lab 88 100 88 75 0 100 50 75

5. Wives alone-RBM-home 64 91 82 82 0 100 82 91

6. Wives alone-tape-lab 83 100 86 57 0 33 29 29

All RBM (groups 1, 2, 4, 5) 65 96 79 79 0 98 81 86

All tape (groups 3, 6) 60 90 76 67 0 45 38 33

All RBM lab (groups 1, 4) 71 96 79 82 0 96 71 86

All RBM home (groups 2, 5) 59 97 79 76 0 100 90 86












Table 6-continued.


Light-Deep
Groups Small Circles Pressure
Pre P1 P2 P3 Pre P1 P2 P3


1. Teams-RBM-lab 15 90 80 75 0 90 25 25

2. Teams-RBM-home 60 89 67 83 0 89 30 56

3. Teams-tape-lab 14 93 57 64 0 43 07 0

4. Wives alone-RBM-lab 0 100 63 100 0 100 13 13

5. Wives alone-RBM-home 0 91 64 82 0 82 36 64

6. Wives alone-tape-lab 17 67 43 57 0 67 14 29

All RBM (groups 1, 2, 4, 5) 07 91 70 82 0 89 28 40

All tape (groups 3, 6) 15 85 52 62 0 50 10 10

All RBM lab (groups 1, 4) 11 93 75 82 0 93 21 21

All RBM home (groups 2, 5) 03 90 66 83 0 86 34 59










examined. The most striking effect of training with the

tape recording appears in the first posttest with respect

to the search pattern (vertical strips) and pressure re-

quirements. Evidently, these elements of proper BSE tech-

nique can not be as effectively instructed without demon-

stration.

Three months after training, as shown in Figures 2, 3,

and 4, decrements in performance were observed for many sub-

jects on breast area covered, number of palpations, and BSE

duration. All RBM-trained subjects continued to perform

better than tape-trained subjects during Posttest 2 and

mean differences, from Tables 3, 4, and 5, were statistically

significant for breast area examined (t=3.23, df=76, p<.05),

number of palpations (t=2.15, df=76, p<.05), and duration

per cm tissue (t=2.04, df=76, p<.05). As shown in Table 6,

most subjects continued to use three fingers and palpate

with small circles. However, nearly all subjects trained

with RBM procedures and only half of the tape-trained sub-

jects systematically applied the correct pressure.

Also during Posttest 2, slight though not statisti-

cally significant differences in breast examination skill

were observed among the various groups of RBM-trained sub-

jects. As shown in Figure 3, wives trained alone palpated

more than their team-trained counterparts, and home-trained

subjects scored higher than did lab-trained subjects. Fur-

ther, lab-trained subjects also lost slightly more skill in

covering breast area than did home-trained subjects over










the 3-month interval from Posttest 1 to Posttest 2 as shown

in Figure 2.

More decrements in performance were observed in the

first three than the last three follow-up months, but per-

formance continued to decline slightly. All RBM-trained

subjects were better examiners than tape-trained subjects on

measures of area examined (t=2.39, df=76, p<.05), number of

palpations (t=2.40, df=76, p<.05), and duration per cm

tissue (t=2.53, df=76, p<.05) during the final test of BSE

skill. Differences in median values for area examined and

number of palpations are clearly seen in Figures 2 and 3.

There were major differences in topography (see Table 3) on

the 6-month test with more RBM-trained than tape-trained

subjects examining as taught, but the largest difference was

in search pattern used; 86% of the RBM-trained versus 33%

of the tape-trained subjects used the vertical strip search

pattern.

Scores for duration per cm tissue were especially de-

creased from Posttest 1 levels for all groups during the

second posttest, but measures for this variable were little

changed after Posttest 2. In fact, as shown in Figure 4,

there was little difference between groups median scores on

this variable during Posttest 3. A related measure, palpa-

tion rate or number of palpations divided by exam duration,

shows that tape-trained subjects palpated .46 times per

second, and RBM-trained subjects palpated .45 times per sec-

ond, both about once every two seconds. The slow speeds of











palpating and long durations observed during Posttest 1 are

apparently not functional in the long run, and over time,

the hand comes to palpate comfortably once every two seconds.

However, following RBM training, the hand produces a better

exam. This is not surprising because the hand was specific-

ally trained during RBM training.

For subjects trained with RBM procedures, differences

were observed between home-trained and lab-trained groups

during the 6-month test. Home-trained subjects examined

each cm of tissue longer (t=2.14, df=55, p<.05), covered more

breast area (t=1.75, df=55, p<.05; one tailed), and used more

palpations. Observations on topography did not separate

these groups except that more home-trained than lab-trained

subjects, 59% versus 21%, systematically used light and deep

pressure.

Figure 5 shows the amount of change in percent of breast

area examined by groups from Posttest 1 immediately after

training to Posttest 3 given six months after training.

There appears to be no loss of skill on this variable only

for the subjects trained with RBM procedures at home. Some

training gains were lost with respect to this variable by

subjects in all other groups.


Performance on Breast Models

Table 7 presents group means for number of lumps detec-

ted in the breast models on each of the four test occasions.

Table 7 shows that the effect of all forms of training was to











































All R8M Tope Teoms Alone RSM RBM
Lob Home
GROUPS


Teams-RBM-Lob Teoms-RBM Home Wives Alone W sAlone
Wives Huband Total Wives Husbands Ttol RBM-Lab RBM-Honm


Wives Alone Teoms-Tbpe-Lob
Tape- Lob Wive Husbands Total


GROUPS


Figure 5.


Group median percent change from posttest 1 to
posttest 3 on percent of breast examined


W

(
SI
(- -1(
-I,







I,








Z -2(
a -2!









44
















o- C -1 CO m H Co (C CN






L- co 0 co o -- co
E0







co
C 0"
*0 11C 0 i


cO Q I I I






) o L
o- I1) 10 10
(Na O CD (N H (N H



a)
4'0 10 10 N 00N 10 10 0


aO 0 N










4-0 00







+-'
SD I I Q)
a)



















n-r 0 I I !
C O C)









z- C D 0 0 0
a)
M



















C C) U o U) 1 1 H1










increase mean number of lumps detected in the model. All

subjects found a mean of 6.8 lumps during the pretest and

8.68 lumps during Posttest 1; these scores were statis-

tically different (t=5.56, df=154, p<.05). The mean group

scores rose to between 7.7 and 10.1 lumps detected during

Posttest 1 immediately after training. However, all wives

trained with RBM procedures found a mean of 9.2 lumps while

all those trained by the tape recording found a mean of

7.9 (t=1.86, df=50, p<.05; one-tailed). This result repli-

cates an earlier finding that real breast plus breast model

procedures are associated with larger increases in lumps

detected than procedures which do not train the fingers on

real or simulated breast tissue (Pennypacker et al., Note 2).

In that study, the post-training mean for subjects trained by

RBM procedures on lumps detected in the breast model was 8.87

while the mean number of lumps detected by subjects trained

by reading the American Cancer Society training pamphlet was

only 6.93. Those values are comparable to the RBM and tape

group mean scores on Posttest 1 in this study.

Responding on the part of subjects trained with RBM

procedures was also characterized by less false positive

responses and longer durations on all breast model posttests

than was the case with subjects trained by the tape recording.

Mean scores for number of false positive responses are pre-

sented in Table 8. The posttest means for all RBM-trained

subjects were 8.0, 6.2, and 5.6, and means for tape-trained

subjects were 13.8, 12.2, and 8.6. These means, however,

























CM aT (0 Co (0 CO Co N
ai c U i) (D Co CO a-

CM

Co o C-1 (Co m -1 co 3-

C4 a- 1 a- :_ co C n co co

O



O 0 in o -


0 ( a- H Co Co ^-1 Co ^H


CN C
() f3
P )-


4 -1 o
0
D H C o

I- -
U) Co L) -1 CN Co Co Co 0 Co

C)CN


b0 .
n C' Cm -1 0
3 Co N H Co ( C a Co


M (C -i co ca i n ( ( 0) co
CH Co *H * *
S0 C C C Co C

0



a




U) M )l U)









&I I CD C) C
a o a
I CI I I


EC af a m) f




U) C3 M C) W C) H ilr l*r









M 0 *H ( Co a- 3 (0*









were significantly different only for Posttest 2 scores

(t=2.14, df=76, p<.05). Table 9 presents mean scores for

model examination duration. Statistically significant dif-

ferences in duration data occurred among all groups (F=3.05,

df=5,72, p<.05) and between all RBM and all tape-trained

subjects (t=2.04, df=76, p<.05; means 101.7, 78.5) during

Posttest 3. It may be seen from Tables 7, 8, and 9 that

detection skill worsens somewhat over the six-month period

following training.

Another striking result of breast model testing was

that on three measures of response accuracy, the three groups

of wives trained with their husbands were the most inaccurate

responders on the final model test. Overall, these wives

scored more false positive responses than did wives trained

alone (t=2.17, df=50, p<.05, means 8.3, 4.2) and more false

positive responses on the no-lump trial (t=2.08, df=50,

p<.05, means 3.1, 1.9). Table 10 presents group mean values

for percent false positive detections. Wives trained in

teams scored a higher percent of false positive detections

than did wives trained alone (t=2.13, df=50, p<.05, means =

41, 28). The mean false positive response percentages for

the three groups of wives trained with their husbands ex-

ceeded those percentages for all other groups (although an

analysis of variance was not significant) and suggest a

breakdown in detection sensitivity. It may be that the

presence of a trained husband at home maintains the wife's

attention to normal but lumpy breast tissue, and those wives









48









L- I a) CO C d Lt) 0)

mc C co c o o LC


(D01 vr v v








O3


0') 0-) r-II
coo 04 Co o o
(N - C - -



O I v I r I
O) Nco c N O) O N
In C o co H N0
VCi)
SCL,





04 ci)
,. D 0 0)) C o o cC _,
Q) L' __ C% CN CN a;








4 cm n co 0









o co 1- 0N zr a) LN f C(
cN Co C 0 a n C(0 Ca)










C- Lo N- (0 c LO


I) 0I











c )

0 F= o o C
0 -O C, N C' C' C' C' CD

N' co (0 H (0 (0

















CI I

w EC :3 o C-o E- :
So0







42 C1 -
Sc cc
0) O 0T) H0 CI
FH C







t '0 42 Cd r o
I C I I l







CD1 0 N- cjCo Zd In> (0









49
















-I in n- ] m n C
C' C4 C01 N CO C1N -I -1


No 0o N (N Co N N
C CO 0 C'O C C- CO (





C( C" C, m- C4 C-1 Cr CO CS C.



U (I Co Co C Co (N (D C(l L C 0
CIO co co Ln or m (n
U)


0

OO CDo (0 co CO co







O
@ O n C- CN O 0 ~ t- 04 C.













P,
( (N C N (N C oN (N (N C
U)
cI) CO CO CO CO Co CN Co CO NO






r-* o CS C m CS C




0co Co r Co Co Co Co (N












ro 0
.(2,
C-4
w s



0



OO







C C0 ,M C- -M C C
n ) IU

































r~ 0 I I I I I
(a > o> > o> > > >
a) .CO HCO C d CN Cr CN C









































E 0
ci
0 rd 0 H
ci) co co o oo IT)

































CO (N Co (










might indicate detections on normal but lumpy tissue in the

model. Thus, the breast model may be an ideal device on

which to assess the immediate effects of training, but home

practice may interfere with model performance after a long

period of time. Further, six months after training, the mean

percentage of false positive responses was 33% for subjects

trained with RBM procedures as compared to 27% for the tape-

trained subjects. Although this difference was not statis-

tically significant, it may suggest that the systematic and

attentive BSE accomplished by RBM-trained subjects interfered

with their performance on the breast models. A similar re-

sult is suggested by data from an earlier study (Pennypacker

et al., Note 2) wherein six months after training, subjects

with the greatest increase in false positive responding were

subjects who reported the most home practice. Thus, break-

downs in accurate model examination performance over follow-

up intervals must be viewed as unfortunate, but these

breakdowns may not suggest serious decrements in BSE skills

on real tissue.


Home Practice

Does training the husband engender more interval breast

exams than training only the wife? The answer is yes over

the long-term for husbands trained with RBM procedures.

Table 11 presents the median reported practice frequency

and median number of spouse prompts to perform breast exams

in each of the two, 3-month follow-up intervals. Median
























o o
c00


S00
rH r


00
.i .


00
00


oo o0o o0oo mo
_ rl -4- co Li CN CN -


00
O oC
..


00
CN H


S00
0 ZT


S00
o 6


C)

0 *H

E *4

+-1



Q) 44




















0 )





4-J
0 cl




+-)I ?-
M (1)
m 4-
*H l4-
~1 -^


r)






1-,
E
0
I *H


0

(n) 4-
3 0
= E-i


0
03
0
a
> m


C )
r.d
0) .Q
0>
*S


S00
00


S00
o 6


00

00


00 000
So d. I 0ouI
t- f _: (:)L


00

00


000 000
Do U L 0o


E
0




ow
O





c
0




*>


CO
H c




, X



OmdH
4-J 7jl


0 W -4iH
c > 4- 4
*H 0C
Z33 E-


H



CO


F0)
:,j

E

0
*H 0)
r0
( U


0o 0I 00o0 o00 mL o

C) c-N 0 C -IiD ; O










scores for reported practice in the first 3-month interval

show group differences, but in the second interval, group

medians are very close to one another. However, some con-

clusions may be drawn. For subjects trained with RBM pro-

cedures, wives trained alone reported more exams than team

members added together for the first 3-month interval.

Two subjects trained alone in the lab reported that their

untrained husbands examined them; the wife of one husband,

reported to have performed four exams on her, stated that

all his exams occurred during sexual afterplay. These exams

increased the median scores of that group. Only one subject

trained alone at home reported occasional husband exams,

and he was a nursing student.

In the second 3-month interval, the two team groups

trained with RBM procedures reported more exams than their

singly-trained counterparts. In addition, practice reports

of both groups of team wives increased as reports by wives

trained alone decreased. Within these teams, wives trained

in the lab reported more exams as their husbands reported

less exams. The contribution of these husbands decreased

over the 6-month period. For the home-trained teams,

however, the practice reports of both spouses increases and

reports of prompts remained stable and evenly divided be-

tween the spouses. Only one husband in this group reported

no exams compared to three husbands in the lab-trained group

reporting no exams or prompts in the 6-month period. It










appears that home training improved the effects of training

on reports of husband practice.

For wives trained alone with RBM procedures, a similar

enhancing effect of home training is seen. Home-trained

wives reported more exams than lab-trained wives in both

follow-up periods.

For subjects trained with the tape recording, there is

a suggestion that husbands continued to examine and prompt.

The large number of exams reported by these teams, however,

dropped in the second follow-up interval and equalled the

practice of the singly-trained wives. (The total median

score of the singly-trained wives group is inflated by

practice reports of one untrained husband). In addition,

the practice reports of both spouses trained with the tape

recording decreased, an effect not seen with teams trained

with RBM procedures. Training husbands with the tape re-

cording was not as effective as training husbands with RBM

procedures.

Qualitative evidence is available concerning the effects

of training on reported practice. Five women sought medical

attention for suspected lumps during the study (none found

cancer). Four were trained with RBM procedures. The fifth

was trained with her husband with the tape recording, and he

found the suspicious tissue. The only woman trained with

her husband with the tape recording who reported no exams

throughout the study was the only woman in that group who

had been told that she had fibrocystic breasts (meaning










cystic tissue was predominant throughout her breasts).

Women with diagnosed fibrocystic breasts trained with RBM

procedures, however, reported exams. Further, singly-

trained women trained with RBM procedures at home were

generally firmer breasted, perhaps indicating that as groups

these women had more difficult breasts to examine than other

groups. Reported practice, however, was not less than with

other groups. These observations suggest that RBM training

may lead to home practice by women with a variety of breast

characteristics.















CHAPTER IV
DISCUSSION


Other work undertaken at this lab (Pennypacker et al.,

Note 2) presents compelling evidence that RBM training pro-

cedures produce better immediate performance on the breast

model than do various other training procedures such as

reading training pamphlets or using either the breast model

or the subject's breast alone. We may now add that RBM

procedures also produce actual BSE performances superior to

that produced by training procedures which provide similar

information but no practice. Research in progress, employ-

ing direct observation of BSE performance as developed in

the present work, is assessing the effects of RBM, model

alone, and American Cancer Society pamphlet training proced-

ures.

This study also provides the first information on the

long-term effects of RBM training on BSE performance. BSE

performance decreased in quality over the six-month period,

but in most cases, subjects performed better during the

final test than they did during the pretest. Specifically,

subjects continued to examine most of the breast, especially

those trained at home, although over time, subjects tended

to miss the outermost boundary areas. Number of palpations

used during the exam remained higher than pretest levels.










In general, subjects spent less time per cm breast tissue

six-months after training than they had during the first

posttest, and exam durations were shorter. Most of the major

decrements in skill occurred in the first three months after

training. However, six months following only 30 minutes of

RBM training, BSE performance remained improved over pretest

levels.

Periodic retraining may be needed in some cases to

bring BSE performance more tightly under instructional con-

trol. The type of retraining offered should be determined

empirically because merely repeating training may not be as

effective as a modified training session (Criswell, Note 3).

Data from Pennypacker et al. (Note 2) provide evidence that

subjects originally in control groups did not perform as well

as RBM-trained subjects when they were retrained six months

later with RBM procedures. This suggests that certain train-

ing session events may be functional stimuli only at the

first session. In the case of BSE retraining, perhaps the

order of training components should be changed, or perhaps

only guided practice is required.

The data on breast model performance in the present

study replicate and extend the findings of Pennypacker et al.

(Note 2). Brief practice on the breast model during RBM

training does not lead to accurate performance after long

follow-up intervals. In fact, the data from the present

study show decreasing trends such that one might expect

pretest level performance had there been one more posttest.










Because lump-tissue discrimination are important to BSE

skill, and because a breast model with lumps is necessary to

teach the discrimination, we have recently increased the

amount of breast model practice during RBM training. To

provide for continued practice on the model, we now suggest

that a woman examine her breast model as a prelude to BSE.

Home training with RBM procedures was associated with

better maintenance of BSE skill and home practice than was

training in the lab. Training in the same setting where the

desired behavior would occur offered subjects reinforced

practice in the presence of natural simulus conditions.

Whether BSE performance would generalize to a new albeit

related setting was never an issue with home-trained subjects.

During home training sessions, the subject was comfortably

positioned on her own bed, her head rested on her own pillow,

her eyes met with furnishings in her own room which might

have acquired discriminative properties during training,

and immediately after training, she resumed her normal ac-

tivities. Home performance of lab-trained subjects remains

unassessed, however, so perhaps they too performed well at

home. But testing these subjects at home, thus providing

practice in the presence of an experimenter in a natural

setting, would have greatly confounded the treatment con-

ditions. Teaching and testing occurred in the same settings

for all subjects. In addition, the lab-based conditions

were designed to mimic usual clinic training. Clinic per-

sonnel do not assess performance at home.










Home BSE practice is encouraged, at least temporarily,

by BSE training, but is not dependent on excellent immediate

performance. Indeed, even those women trained with the tape

recording reported examining their breasts more often than

they did before training. Quality and practice frequency of

BSE are separate issues, and the important question sug-

gested by data from this study, and others, is: Is it better

for women to perform poor quality exams than not to examine

at all with respect to finding lumps in their own breasts,

or is high quality performance the only performance which

will detect breast lumps? Qualitative evidence from the

present study suggests that women with difficult breasts to

examine will benefit most in this regard from RBM training.

A long-term clinical trial of BSE training procedures would

provide the critical data.

RBM training procedures were associated with good BSE

performance six months after training, but in addition to

training, home practice affects the long-term maintenance of

BSE skill. Because amount of home practice was not controlled

in this study, conclusions concerning the practice frequency

which supports skill maintenance may not be drawn. Research

in progress in which subjects trained with RBM procedures,

who are asked not to practice at home, are recalled at vari-

ous intervals (once a week, twice a month, and once a month)

and observed practicing BSE. With practice frequency con-

trolled in this way, pertinent data will be gathered. The










long-term effects of RBM training may be more favorable with

more frequent practice.

Slight differences in BSE performance were observed

between women trained alone and women trained with their

spouses, with women trained alone performing slightly better.

Percent of breast examined did not appear to be sensitive to

the presence or absence of the spouse, but both groups of

wives trained alone used more palpations during an exam and

spent more time per cm of tissue than their team-trained

counterparts. This effect was not seen with tape-trained

subjects. Slight skill decrements, however, may be offset

by the benefits with respect to frequent home practice en-

gendered by team training. Data from this study suggest that

interested husbands who are taught breast-examination with

RBM procedures may support the BSE practice of their wives.

Many husbands who continued to examine their wives reported

that exams fit easily into sexual fore- or afterplay. But

over long periods of time, unless husbands are trained at

home, it appears that the contribution of the husband wanes

as time passes. Each couple brings to the training session

a unique, perhaps long-standing history, and a 30-minute

training session may not provide sufficient reinforcement

for team breast-examination at home. The six subjects in

the present study who reported no exams in the entire six-

month period were all trained in teams. Some subjects

stated that their husbands used their new skill at examina-

tion as a trick to initiate sex, or that looking for cancer









spoils the usual delight caused when the husband fondles the

wife's breasts. So, for many women, training without the

spouse would likely be more effective than training with

him. Individual preferences must be considered.

Because of the superiority in BSE skill and home prac-

tice of subjects in the home-trained groups in this study,

mass home training may be the desired method of dissemi-

nating BSE training. If BSE training could be mediated by

a computer or delivered by a speaker telephone with video

capabilities, and if homes were equipped with terminals,

programs, and other needed machinery, the stage might be

set for effective mass home training. Computer hard- and

software, perhaps suited to this purpose, is being developed

at this lab for use in clinics (Thomas, Goldstein, Stein,

& Pennypacker, in press), but tests of the effectiveness

and feasibility of these procedures have not yet been at-

tempted. Although we have compelling evidence that actual

practice is critical to the acquisition and maintenance of

BSE skill, we do not now know to what extent a human trainer

is critical in administering RBM training. Although tele-

visions are already widely available in homes, conventional

televised presentations do not include the shaping of indi-

vidual performance necessary in BSE training, and thus,

would not be expected to produce high quality BSE performance.

Research testing the effects of televised training is planned.

Many years into the future as more and more women are well-

trained in BSE, the situation might occur that mothers teach










their daughters how to examine their breasts as mothers

teach daughters how to brush their teeth. The home train-

ing setting may become commonplace in that way. In the

meantime, development of home training systems seems a

promising endeavor.

Finally, this study suggests that RBM training pro-

cedures are effective in a variety of training situations,

and no one RBM training situation need be recommended for

all women. Designing training settings like bedrooms is

suggested if home training is unavailable. It is likely

that the best delivery system would take into account the

willingness or hesitance of a partner residing with the

trainee to participate in BSE practice. For high risk

women, training in the home, with or without a partner,

should be encouraged.















REFERENCE NOTES


1. Bloom, H. S., Criswell, E. L., Pennypacker, H. S.,
Catania, A. C., S Adams, C. K. Major stimulus
dimensions determining detection of simulated breast
lesions. Manuscript submitted for publication, 1981.

2. Pennypacker, H. S., Neelakantan, P., Bloom, H. S.,
Criswell, E. L., & Goldstein, M. K. The effects of
selected training procedures on acquisition and
maintenance of skill in detecting simulated breast
cancer. Manuscript submitted for publication, 1981.

3. Criswell, E. L. The long-term maintenance of behavior
change. Unpublished manuscript, University of Florida,
1980.















REFERENCES


Adams, C. K., Hall, D. C., Pennypacker, H. S., Goldstein,
M. K., Hench, L. L., Madden, M. C., Stein, G. H., S
Catania, A. C. Lump detection in simulated human
breasts. Perception 9 Psychophysics, 1976, 20, 163-167.

Atthowe, J. A. Behavior innovation and persistence.
American Psychologist, 1973, 27, 34-41.

Baer, D. M., Wolf, M. M., 9 Risley, T. R. Some current
dimensions of applied behavior analysis. Journal of
Applied Behavior Analysis, 1968, 1, 91-92.

Feldman, J. G., Carter, A. C., Nicastri, A. D., 6 Hosat, M. S.
Breast self-examination, relationship to stage of breast
cancer at diagnosis. Cancer, in press.

Foster, R. S., Lang, S. P., Costanza, M. C., Worden, J. K.,
Haines, C. R., 9 Yates, J. W. Breast self-examination
practices and breast-cancer stage. The New England
Journal of Medicine, 1978, 299, 265-270.

Gallup Organization. Public awareness and use of cancer
detection tests. New York: American Cancer Society,
1980.

Greenwald, P., Nasca, P. C., Lawrence, C. E., Horton, J.,
McGarrah, R. P., Gabriele, T., S Carlton, K. Estimated
effect of breast self-examination and routine physician
examinations on breast-cancer mortality. The New
England Journal of Medicine, 1978, 299, 271-273.

Hall, D. C., Adams, C. K., Stein, G. H., Stephenson, H. S.,
Goldstein, M. K., & Pennypacker, H. S. Improved
detections of breast lesions following experimental
training. Cancer, 1980, 46, 408-414.

Hall, D. C., Goldstein, M. K., S Stein, G. H. Progress in
manual breast examination. Cancer, 1977, 40, 364-370.

Hayes, S. C., Rincover, A., S Solnick, J. W. The technical
drift of applied behavior analysis. Journal of Applied
Behavior Analysis, 1980, 13, 275-286.










Holleb, A. I., Venet, L., Day, E., S Hoyt, S. Breast
cancer detected by routine physical examination.
New York State Journal of Medicine, 1960, 60, 823-
827.

Hutchinson, G. B., S Shapiro, S. Lead time gained by
diagnostic screening for breast cancer. Journal of
the National Cancer Institute, 1968, 41, 665-681.

Johnston, J. M., & Pennypacker, H. S. Strategies and
tactics of human behavioral research. Englewood
Cliffs, New Jersey: Lawrence E. Erlbaum 8 Associates,
1980.

Kazdin, A. E. Behavior modification in applied settings.
Homewood, Illinois: The Dorsey Press, 1975.

Koegel, R. L., & Rincover, A. Research on the difference
between generalization and maintenance in extra-therapy
responding. Journal of Applied Behavior Analysis,
1977, 10, 1-12.

Lesnick, G. J. Detection of breast cancer in young women.
Journal of the American Medical Association, 1977,
237, 967-969.

Madden, M. C., Hench, L. L., Hall, D. C., Pennypacker, H. S.,
Adams, C. K., Goldstein, M. K., S Stein, G. H. Model
breasts for use in teaching breast self-examination.
Journal of Bioengineering, 1978, 2, 427-435.

Shapiro, S. Evidence of screening for breast cancer from a
randomized trial. Cancer, 1977, 39, 2772-2782.

Stephenson, H. S., Adams, C. K., Hall, D. C., S Pennypacker,
H. S. The effects of certain training parameters on
detection of simulated breast cancer. Journal of
Behavioral Medicine, 1979, 2, 239-250.

Stokes, T. F., & Baer, D. M. An implicit technology of
generalization. Journal of Applied Behavior Analysis,
1977, 10, 349-367.

Strax, P. Results of mass screening for breast cancer in
50,000 examinations. Cancer, 1976, 37, 30-35.

Thomas, R. G., Goldstein, M. K., Stein, G. H., S Pennypacker,
H. S. Computer-aided instruction of breast self-
examination. Proceedings of world association for
medical informatics, in press.






65



Wolfe, J. N. Analysis of 462 breast carcinomas.
American Journal of Roentgenology, Radium Therapy,
and Nuclear Medicine, 1974, 121, 846-853.

Wyatt, D. L. Performance issues in breast self-examination.
Unpublished master's thesis, University of Texas,
School of Public Health, 1979.





































APPENDIX A
RAW DATA FROM TESTS ON THE WIFE'S BREAST FOR
ALL SUBJECTS













Teams-RBM-Lab
Wives Breast Data


Subject Squares Squares Percent Total Duration/
# Available Examined Examined Palps Duration Size


Pretest


1.78
0.75
0.82
0.64
0.84
1.58
2.05
0.54
0.69
0.87


Posttest 1


201
208
231
112
191
143
339
217
117
159


3.41
3.71
3.12
2.24
2.77
2.20
5.75
4.72
2.17
3.06


Posttest 2

1
2
3
4
5
6
7
8
9
10


71
114
109
109
117
41
175
63
100
99


1.20
2.04
1.5
2.18
1.70
0.63
2.97
1.37
1.85
1.90












Teams-RBM-Lab
Wives Breast Data
(Continued))


Subject Squares Squares Percent Total Duration/
# Available Examined Examined Palps Duration Size

Posttest 3

1 30 20 67 49 122 2.07
2 42 23 55 44 92 1.64
3 42 23 55 48 87 1.18
4 20 18 90 37 59 1.18
5 35 29 83 65 115 1.67
6 30 21 70 27 30 0.46
7 24 17 71 51 155 2.63
8 24 14 58 38 81 1.76
9 35 20 57 37 110 2.04
10 25 12 48 31 83 1.60













Teams-RBM-Lab
Husband Breast Data


Subject Squares Squares Percent Total Duration/
# Available Examined Examined Palps Duration Size


Pretest


0.47
0.36
0.20
1.34
0.30
1.09
0.51
0.37
0.44
1.85


Posttest 1


295
129
178
136
387
86
153
144
160
170


5.00
2.30
2.41
2.72
5.61
1.32
2.59
3.13
2.96
3.27


Posttest 2

1
2
3
4
5
6
7
8
910
10


1.73
0.63
0.89
1.54
2.68
0.68
1.58
0.89
1.91
1.19












Teams-RBM-Lab
Husband Breast Data
(Continued)


Subject Squares Squares Percent Total Duration/
# Available Examined Examined Palps Duration Size


Posttest 3

1 30 24 80 57 173 2.93
2 42 17 40 22 38 0.68
3 42 21 50 31 59 0.80
4 20 19 95 32 82 1.64
5 35 28 80 113 209 3.03
6 30 13 43 26 34 0.52
7 24 18 75 33 88 1.49
8 24 11 46 16 33 0.72
9 35 23 66 41 88 1.63
10 25 18 72 39 69 1.33













Teams-RBM-Home
Wives Breast Data


Subject Squares Squares Percent Total Duration/
# Available Examined Examined Palps Duration Size


Pretest


2.18
0.48
0.86
0.80
0.30
1.43
0.43
0.83
0.36


Posttest 1


4.89
2.63
5.22
8.28
3.21
3.64
2.71
2.98
2.51


Posttest 2

1
2
3
4
5
6
7
8
9


79
103
150
180
155
117
144
52
39


1.41
1.91
3.41
3.60
2.92
2.02
2.09
1.00
0.48












Teams-RBM-Home
Wives Breast Data
(Continued)


Subject Squares Squares Percent Total Duration/
# Available Examined Examined Palps Duration Size

Pretest

1 42 21 50 26 85 1.52
2 24 20 83 42 107 1.98
3 30 24 80 65 150 3.41
4 30 22 73 56 288 5.76
5 20 18 90 46 63 1.19
6 42 16 38 30 106 1.83
7 36 32 89 65 113 1.64
8 30 24 80 63 111 2.13
9 48 28 58 44 61 0.75












Teams-RBM-Home
Husbands Breast Data


Subject Squares Squares Percent Total Duration/
# Available Examined Examined Palps Duration Size


Pretest


0.93
1.04
1.61
1.04
0.38
0.48
0.72
1.12
0.49


Posttest 1


174
194
239
321
260
142
154
243
123


3.11
3.59
5.43
6.42
4.91
2.45
2.23
4.67
1.52


Posttest 2

1
2
3
4
5
6
7
8
9


4.21
2.52
2.84
3.70
1.89
2.53
1.49
2.25
1.02












Teams-RBM-Home
Husbands Breast Data
(Continued)


Subject Squares Squares Percent Total Duration/
# Available Examined Examined Palps Duration Size


Posttest 3

1 42 29 69 59 59 1.05
2 24 20 83 66 224 4.15
3 30 25 83 50 139 3.16
4 30 20 67 34 178 3.56
5 20 19 95 31 74 1.40
6 42 18 43 36 153 2.64
7 36 26 72 53 106 1.54
8 30 22 73 61 179 3.44
9 48 35 73 73 117 1.44













Teams-Tape-Lab
Wives Breast Data


Subject Squares Squares Percent Total Duration/
# Available Examined Examined Palps Duration Size


Pretest


2.55
0.47
1.17
0.62
0.53
0.27
0.49


Posttest 1


207
142
362
115
58
157
65


4.22
2.68
6.96
1.92
1.23
1.99
1.38


Posttest 2

1
2
3
4
5
6
7


Posttest 3

1
2
3
4
5
6
7


104
64
262
87
27
60
53


2.12
1.21
5.04
1.45
0.57
0.76
1.13


1.88
0.64
3.63
1.40
0.51
1.06
1.21













Teams-Tape-Lab
Husband Breast Data


Subject Squares Squares Percent Total Duration/
# Available Examined Examined Palps Duration Size


Pretest


0.76
0.70
0.87
0.48
0.21
0.22
1.47


Posttest 1


161
117
180
55
52
116
122


3.29
2.21
3.46
0.92
1.11
1.47
2.60


Posttest 2

1
2
3
4
5
6
7


Posttest 3

1
2
3
4
5
6
7


41
115
131
19
32
100
42


0.84
2.17
2.52
0.32
0.68
1.27
0.89


56
140
154
15
23
74
62


1.14
2.64
2.96
0.25
0.49
0.94
1.32













Wives Alone-RBM-Lab
Breast Data


Subject Squares Squares Percent Total Duration/
# Available Examined Examined Palps Duration Size


Pretest


144
27
29
36
25
38
28
21


2.57
0.49
0.51
0.67
0.43
0.76
0.53
0.30


Posttest 1


6.39
3.33
5.79
9.19
5.83
5.42
5.98
2.04


Posttest 2

1
2
3
4
5
6
7
8


3.30
1.20
2.09
3.09
3.76
2.24
0.91
2.45












Wives Alone-RBM-Lab
Breast Data
(Continued)


Subject Squares Squares Percent Total Duration/
# Available Examined Examined Palps Duration Size


Posttest 3

1 20 16 80 50 128 2.29
2 30 16 53 47 104 1.89
3 24 10 42 30 68 1.19
4 30 17 57 54 137 2.54
5 30 25 83 49 225 3.88
6 30 21 70 60 115 2.30
7 30 21 70 38 65 1.23
8 49 30 61 80 133 1.87













Wives Alone-RBM-Home
Breast Data


Subject Squares Squares Percent Total Duration/
# Available Examined Examined Palps Duration Size


Pretest


62
226
17
23
48
93
20
55
32
26
39


0.94
3.32
0.27
0.31
0.89
1.58
0.32
0.83
0.71
0.47
0.67


Posttest 1


271
470
60
132
184
247
131
171
215
213
228


4.11
6.91
0.97
1.78
3.41
4.19
2.11
2.59
4.78
3.87
3.93


Posttest 2

1
2
3
4
5
6
7
8
9
10
11


1.83
3.59
0.76
1.15
2.31
4.27
1.31
1.92
2.73
1.24
1.59












Wives Alone-RBM-Home
Breast Data
(Continued)


Subject Squares Squares Percent Total Duration/
# Available Examined Examined Palps Duration Size


Posttest 3

1 36 33 92 73 161 2.44
2 49 36 73 100 228 3.35
3 35 12 34 19 38 0.61
4 25 18 72 44 103 1.39
5 56 32 57 44 66 1.22
6 30 26 87 71 272 4.61
7 42 26 62 80 169 2.73
8 35 33 94 84 189 2.86
9 30 23 77 58 106 2.36
10 20 18 90 29 72 1.31
11 35 20 57 24 91 1.57












Wives Alone-Tape-Lab
Breast Data


Subject Squares Squares Percent Total Duration/
# Available Examined Examined Palps Duration Size


Pretest


0.60
0.70
1.95
0.54
0.91
1.12
0.93


Posttest 1


61
211
78
126
149
85
121


1.15
3.30
1.39
2.14
3.24
1.73
2.63


Posttest 2

1
2
3
4
5
6
7

Posttest 3

1
2
3
4
5
6
7


0.77
2.86
0.63
1.61
1.61
1.16
1.67


0.42
2.22
0.70
1.15
1.33
0.86
1.74

































APPENDIX B
RAW DATA FROM TESTS ON THE BREAST MODELS FOR ALL SUBJECTS













Teams-RBM-Lab
Wives Model Data


Percent
False
Subject Lumps False False Positive Total Positive
# Detected Positive Catch Trial Responses Responses Duration


Pretest


115.2
154.4
72.8
43.2
60.0
97.0
111.0
40.2
46.5
62.8


Posttest 1


193.4
146.6
127.6
82.6
106.3
102.8
177.8
205.6
112.2
104.2


Posttest 2
1
2
3
4
5
6
7
8
9
10


132.6
113.2
69.0
75.4
64.8
61.2
163.8
137.4
116.0
91.6












Teams-RBM-Lab
Wives Model Data
(Continued)


Percent
False
Subject Lumps False False Positive Total Positive
# Detected Positive Catch Trial Responses Responses Duration


Posttest 3

1 7 2 2 9 22 108.8
2 9 5 2 14 36 91.6
3 11 0 0 14 00 52.8
4 2 4 2 6 67 59.4
5 8 12 7 20 60 78.2
6 2 0 0 2 00 44.2
7 14 12 0 27 44 124.6
8 11 30 9 46 65 156.2
9 7 12 4 20 60 114.2
10 6 16 5 26 62 75.8













Teams-RBM-Lab
Husbands Model Data


Percent
False
Subject Lumps False False Positive Total Positive
# Detected Positive Catch Trial Responses Responses Duration


Pretest


105.0
25.4
17.6
102.2
78.2
59.2
28.0
35.8
30.0
250.4


Posttest 1


97.8
108.0
124.4
114.8
156.2
85.2
96.4
110.6
95.6
194.0


Posttest 2


81.6
61.4
67.6
94.8
108.4
90.8
88.2
65.6
55.2
130.0












Teams-RBM-Lab
Husbands Model Data
(Continued)


Percent
False
Subject Lumps False False Positive Total Positive
# Detected Positive Catch Trial Responses Responses Duration


Posttest 3

1 8 5 1 14 36 61.6
2 4 0 0 4 00 58.6
3 7 4 3 13 31 50.6
4 5 0 0 9 00 68.4
5 6 2 0 10 20 83.8
6 5 0 0 5 00 41.4
7 6 10 0 19 53 100.4
8 7 1 1 9 11 54.0
9 7 3 0 12 25 71.4
10 10 15 2 26 58 111.6













Teams-RBM-Home
Wives Model Data


Percent
False
Subject Lumps False False Positive Total Positive
# Detected Positive Catch Trial Responses Responses Duration


Pretest


286.2
73.8
93.4
50.6
56.8
151.6
53.4
12.0
73.6


Posttest 1


143.0
96.8
233.8
210.2
130.0
295.6
113.2
82.4
143.8


Posttest 2


92.6
81.2
125.0
130.8
110.0
144.4
119.2
74.6
81.4













Teams-RBM-Home
Wives Model Data
(Continued)


Percent
False
Subject Lumps False False Positive Total Positive
# Detected Positive Catch Trial Responses Responses Duration


Posttest 3

1 3 1 0 5 20 75.0
2 4 0 0 4 00 75.8
3 6 8 5 14 57 215.2
4 6 4 3 11 36 143.8
5 7 8 3 15 53 119.4
6 10 5 3 17 29 178.2
7 10 11 4 22 50 72.8
8 10 22 8 33 67 83.0
9 5 5 1 10 50 66.2













Teams-RBM-Home
Husbands Model Data


Percent
False
Subject Lumps False False Positive Total Positive
# Detected Positive Catch Trial Responses Responses Duration


Pretest


103.0
36.2
146.4
84.0
54.0
29.2
45.8
69.8
10.2


Posttest 1


128.0
119.0
210.6
154.0
238.2
100.4
85.8
116.8
21.8


Posttest 2

1
2
3
4
5
6
7
8
9


131.2
147.4
181.2
94.0
85.2
109.6
96.0
129.8
71.0













Teams-RBM-Home
Husbands Model Data
(Continued)


Percent
False
Subject Lumps False False Positive Total Positive
# Detected Positive Catch Trial Responses Responses Duration


Posttest 3

1 9 3 2 13 23 107.0
2 8 16 3 24 67 205.2
3 6 9 2 15 60 189.4
4 6 6 3 12 50 94.8
5 6 1 0 7 14 106.8
6 6 1 0 7 14 128.0
7 6 0 0 7 00 72.2
8 11 5 0 19 26 146.8
9 6 0 0 7 00 75.6













Teams-Tape-Lab
Wives Model Data


Percent
False
Subject Lumps False False Positive Total Positive
# Detected Positive Catch Trial Responses Responses Duration


Pretest


102.0
31.8
96.8
79.4
65.0
49.2
51.6


Posttest 1


151.4
81.6
207.8
175.2
77.4
96.4
120.6


Posttest 2

1
2
3
4
5
6
7


Posttest 3

1
2
3
4
5
6
7


161.2
61.6
147.2
105.4
60.2
74.0
122.2


130.4
39.2
106.4
125.4
58.0
71.4
122.0












Teams-Tape-Lab
Husbands Model Data


Percent
False
Subject Lumps False False Positive Total Positive
# Detected Positive Catch Trial Responses Responses Duration


Pretest


137.0
45.6
74.2
42.6
23.4
35.3
68.4


Posttest 1


220.4
84.8
204.0
65.8
91.6
91.0
84.2


Posttest 2
1
2
3
4
5
6
7


Posttest 3

1
2
3
4
5
6
7


101.8
59.0
100.2
24.8
109.4
59.8
51.2


89.0
83.0
114.4
18.0
56.0
41.4
61.0




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - Version 2.9.8 - mvs