Frequency and adequacy of breast cancer screening among elderly hispanic women

PREVENTIVE MEDICINE 16, 76 l-774 (1987)

Frequency and Adequacy of Breast Cancer Screening among Elderly Hispanic Women’

JEAN L. RICHARDSON,DR.PH. ,2,* GARYMARKS,PH.D.,* JULIA M. SOLIS, PH.D.,t LINDA M. COLLINS, PH.D.,*

LOURDES BIRBA, M.S.G. ,t ANDJOHN C. HISSERICH, DR.P.H.*

*Department of Preventive Medicine, School of Medicine, University of Southern California, 1420 San Pablo Street, A-301, and tUniversity of Southern California Comprehensive Cancer Center,

1721 Griffin Avenue, Los Angeles, California 90033

Studies have demonstrated that Hispanic (relative to Anglo) women are at greater risk for late-stage breast cancer diagnosis. Screening irregularity may be a factor contributing to late-stage diagnosis, yet virtually nothing is known about the breast cancer-screening behavior of Hispanic women. We interviewed 600 elderly Hispanic women residing in Los Angeles to collect information on frequency of physician breast examinations and mammography and on regularity and competence of breast self-examination. Predictors of screening were also examined. Fifty percent of our sample indicated that they had had a breast exam within the past year; 12.5% had had a mammogram within the past year (74% never had had a mammogram); and 47% reported that they had performed breast self-examination within the past month. Few of the women were able to demonstrate adequate breast self-examination technique on a foam breast model, and only 1% found all five lumps present. Thus, although the observed frequency of screening and self-examination is comparable to national norms, it is unlikely that our subjects’ attempts at self-examination would lead to early breast cancer detection. Age, educational level, emotional reactions to screening, and media cues predicted screening behavior. Physician instruction in breast self-examination increased the frequency and adequacy of self-examination. Perceived susceptiblity to cancer, perceived benefits of early detection, and level of acculturation were not strong predictors. The extent to which our results generalize to other subpopulations of Hispanic women is discussed. 0 1987 Academic Press. Inc.

INTRODUCTION

Although Hispanic women are not at greater risk for breast cancer (22, 27), research has shown that they are at greater risk for presenting with larger tumors and with regional or distant metastases (10, 34). For example, an analysis of age- specific incidence data between 1975 and 1979 from the Los Angeles County/Uni- versity of Southern California Cancer Surveillance Program showed that, overall, 48% of Spanish-surnamed women were diagnosed with breast cancer at a non- local stage, whereas 42% of whites (other than Spanish surname) were diagnosed with non-local breast cancer (8). This difference held at all socioeconomic levels.

Because stage of breast cancer diagnosis is perhaps the most crucial factor in determining survival time (5), it is imperative to gain an understanding of the

r Supported by Grant CA35666 from the National Cancer Institute. The survey reported in this article was conducted as part of a larger intervention research program to promote the use of cancer screening and detection methods by elderly Hispanic women.

2 To whom reprint requests should be addressed.

761

009 l-7435/87 $3 .oo Copyright 0 1987 by Academic Press, Inc. All rights of reproduction in any form reserved.

762 RICHARDSON ET AL.

factors that contribute to late-stage diagnosis of this disease among Hispanic women. One hypothesis implicates screening behavior irregularity as a contributing factor. Surprisingly, however, virtually nothing is known about the breast cancer screening behavior of Hispanic women. Therefore, one purpose of the present study was to collect descriptive information about the frequency of physician breast examinations and mammography and the regularity and competence of breast self-examination (BSE) among a large group of elderly Hispanic women.

Routine breast cancer screening recommendations for women over 50 years of age consist of annual clinical palpation by a physician, annual mammography, and monthly BSE (29). With mammography and physician examination we assume that women who present themselves for screening will be correctly examined. This assumption, however, cannot be made for BSE (17). Several studies have found that many women perform BSE incorrectly (13, 24, 32). Most of these studies have been conducted with college women, hospital employees, or breast cancer patients (2, 36); none has been conducted with Hispanics. Further, only one study has used observed proficiency, lump detection, and self-reported behavior to describe fully BSE competence among white women (of unspecified ethnicity) (2).

A second purpose of the present research was to assess the degree to which past findings about the predictors of screening behavior among non-Hispanics generalize to an Hispanic population. Beliefs in susceptibility to breast cancer (7, 19), beliefs in screening efficacy (15, 18, 25, 28), emotional response to screening (2, 13, 20, 33), age (14, 28), education (14, 28), and training by a physician (15, 18, 25, 28) have each been found to be a significant predictor of screening among non-Hispanics. Few studies, however, have assessed the relative importance of cognitive, emotional, sociodemographic, teaching source, and media variables as predictors of the different types of screening behaviors. In addition, prior investi- gations of Hispanic populations have implicated language and traditional health- care beliefs as barriers to acquiring general health care (4, 9, 11, 16, 26, 35). This suggests that level of acculturation may be a unique barrier to breast cancer screening among Hispanic women.

METHOD

Data Collection

This survey of elderly Hispanic women was conducted in 1984 and 1985. A total of 17 publicly subsidized housing projects in Los Angeles participated in the study. Residents of these housing projects must demonstrate their ability to care for themselves; thus, they are a population of physically well elderly suitable to query regarding their use of screening services. All Hispanic women 55 years of age or older residing in these projects were contacted for an interview. Of the potential target population of 890 women, 603 were interviewed, resulting in a 68% response rate. The data from 3 women were not included in the analyses because they were under 55 years of age.

Participants were interviewed individually in their apartments. Each interview took approximately lr/2 hr to complete and was conducted in the participant’s

BREAST SCREENING AMONG HISPANIC WOMEN 763

preferred language; for 79% of the cases, the interview was conducted completely in Spanish, and for another 8%, it was conducted primarily in Spanish. The na- ture of the questions required careful selection and training of interviewers to enhance the rapport with the respondent and to reduce bias or error. We recruited seven bilingual, predominantly middle-aged women as interviewers and trained them in data collection procedures.

Subjects

The median age of the subjects was 71 years (range 55 to 92). Fifty percent were widowed, 30% were single or divorced, and 18% were married. The overall socioeconomic status of our sample was low, with a mean annual income of $5,772 and a mean educational level of 6 years; 14% had had no formal education. Ninety percent were currently unemployed. In terms of current or most recent employment, 67% worked as laborers, service workers, or operatives and 23% were homemakers with no outside employment.

Most of the participants (72%) were born in Mexico or another Latin-American country; however, these women had been residents of the United States for an average of 25 years. The women personally identified themselves as either Mex- ican (48%), Mexican-American (12%), American (ll%), Hispanic (4%), or Latin- American (2%); the remaining subjects identified themselves with other specific nationalities (e.g., Cuban, Chilean). Twenty-seven percent preferred to use Spanish with family and friends, 27% reported that they did not understand En- glish, and 39% did not speak English at all. On the other hand, 98% understood spoken Spanish well, but only 44% read Spanish well. In terms of friendship patterns, 41% indicated that all of their friends were Latino, 29% stated that most of their friends were Latino, and 25% said that their friends were equally mixed between Latin0 and Anglo; only 3% said most or all of their friends were Anglo.

Our subjects have, to a large extent, learned to utilize the social and health-care services available to them. This is evidenced by the fact that they have secured housing in public facilities. In addition, only 9% did not have health insurance of any type. Ninety-seven percent had a routine place to go for medical care. In 68% of cases it was a private physician’s office, and in 17% of cases it was a hospital outpatient clinic; the remainder were seen at public health clinics and health maintenance organizations. Ninety-six percent stated that they had seen a physician in the past year, and 61% said that they had had a complete physical examination in the past year. In summary, the Hispanic women in our sample have, for the most part, retained their primary language and ethnic identity and appear to be sophisticated in the use of health-care and public facilities.

Measures

Awareness of breast cancer and screening. Participants were asked whether they had ever heard of beast cancer or BSE (yes/no) or had ever read or seen on television that they should (a) have a breast examination from a physician, (b) have a mammography, or (c) practice BSE (yes/no). They indicated whether anyone (e.g., friend, spouse, doctor) had, in the past few years, suggested that


they engage in these three behaviors, and whether a physician or nurse had ever taught them BSE (yes/no).

Knowledge of breast cancer symptoms. Subjects were asked, in an open-ended format, to indicate verbally symptoms of breast cancer. For the purpose of re- cording responses, we had six symptoms listed on the questionnaire which the respondent could not see: lump in breast or under arm, change in the size of the breast, discharge from breast, nipple change, skin changes, and persistent pain and sense of discomfort. The number of symptoms correctly identified comprised the overall knowledge score for each subject. They were also asked how they would respond if they found a lump in their breast.

Recency of last breast examination and mammography. Recency of breast examination was measured by asking “When was the last time you had your breasts examined by a physician?” Open-ended responses were coded on a scale ranging from 1 (never) to 6 (less than six months ago). Recency of last mammography was measured in the same fashion.

Knowledge of breast self-examination. The women were asked three open- ended questions: (a) “How often do you think a woman should examine her breasts?” A response of “every month” was coded as correct; all other responses were coded as incorrect. (b) “When doing breast self-examination what position or positions do you think are best?” A response indicating that one should stand up and also lie on one’s back was coded as correct. (c) “Do you think a woman should leave her clothes on when she does breast self-examination?” A response of “no” was recorded as correct. The overall knowledge score for each subject was the number of correct responses.

Frequency and recency of breast self-examination. Frequency was measured by asking “How often have you usually examined your breasts?” Open-ended responses were coded from “never” (coded as 1) to “more than once a month” (coded as 7). Recency was measured by asking “When was the last time you examined your breasts?” Open-ended responses were coded from “never” (coded as 1) to “less than one month ago” (coded 6).

Assessment of breast self-examination peformance. To assess directly the quality of technique the women used when attempting BSE and to measure directly their ability to detect breast lumps, subjects were presented with a foam breast model (Spenco Co.) and asked to examine it as they would examine themselves. The life-size foam model, which was covered with a thin cloth, contained five lumps distributed throughout the breast. The model was placed on the table in front of the subject and she was asked to display BSE technique. The interviewer (who was thoroughly trained in appropriate BSE procedures) scored the subject on four examination procedures: (a) moved fingers in a circular motion around the nipple area; (b) used the flat part of two to four fingers to feel the breast; (c) put pressure on the breast model; and (d) squeezed the nipple area. For these ratings the interviewer used a scale ranging from 1 (definitely did not perform procedure) to 4 (definitely performed procedure). The interviewer also rated how comfortable the subject appeared when attempting the examination. After the interviewer rated the subject’s performance of breast examination using the model, she asked the subject to report the number of lumps detected in the


model. The subjects were asked “How many lumps do you feel in this breast? You may need to put more pressure on the model to feel the lumps.” No further instructions were given by the interviewer during the time that the subject was attempting the exam. To create a single BSE performance score, the subjective judgments by the interviewer and the number of lumps found were standardized and then averaged to form a single index for each subject.

Nervousness and embarrassment. Subjects indicated how nervous/embarrassed they would feel (a) to have a doctor examine their breasts and (b) to examine their own breasts. They were read four response alternatives that ranged from “not at all nervous/embarrassed” (scored as 1) to “very nervous/embarrassed” (scored as 4).

Perceived susceptibility. To measure subjects’ perceptions of susceptibility to cancer, those who indicated that they had never had any form of cancer (91.5%) were asked “Although nobody likes to think about it, some people do get cancer. How probable do you think it is that you might get cancer sometime in your life?” The four response alternatives (read to the subject) ranged from “not at all probable” (coded as 1) to “very probable” (coded 4).

Benefits of early cancer detection. The women were asked “How much do you think going to the doctor as soon as you notice a symptom that might be cancer increases a person’s chances of living ?” They were read four response options that ranged from “not at all” (coded as 1) to “a great deal” (coded as 4). They were then asked “Do you think that undergoing treatment which may have un- pleasant side effects is worth the possibility of being cured?” (yes/no). Finally, they were asked “For most people who develop cancer, if proper medical care is obtained, how propable is it that they will be cured?” The response options read to them (scored 1-4) ranged from “not at all probable” to “very probable.” Because these three measures correlated positively (r = 0.16 to 0.44, P < O.OOl), the items were standardized and then averaged to form a single overall score for each subject.

Acculturation. Two indicators of acculturation were used: understanding of En- glish and friendship patterns.

RESULTS

Knowledge and Behavior Pertaining to Breast Cancer and Its Symptoms

Ninety-one percent of the women had heard of breast cancer, and 3% (18 women) had had breast cancer at some time. Although over 60% of the sample stated that they did not know what caused breast cancer, the most frequent mis- conceptions of cause were injury or blow to the breast (20%) or some problem with breast feeding (6%). The women who had heard of breast cancer were asked to state all of the breast cancer symptoms they knew. Seventy percent mentioned a lump in the breast or under the arm, 23% stated persistent pain or discomfort, 10% mentioned a change in the size of the breast (swelling or shrinking), 9% said a discharge from the nipple, 8% mentioned skin changes (thickening, dimpling), and 7% mentioned nipple changes (elevated, inverted). In terms of number of symptoms mentioned, 25% of the women could not state any symptoms, 43%


stated one symptom, 21% mentioned two symptoms, 7% stated three symptoms, and only 4% mentioned four or more symptoms. Finally, 92% of the subjects said that if they were to find any lumps in their breasts they would call a doctor imme- diately; the remainder said that they would call the doctor only if the lump per- sisted for a few weeks or if it was painful.

Knowledge and Behavior Pertaining to Breast Exam and Mammography

As seen in Table 1, approximately 50% of the participants had had a breast exam by a physician within the previous year, but almost 16% had never had a breast exam. Fifty-seven percent of the respondents stated that, within the past few years, nobody suggested that they should have a physician breast exam; 41% said that a health professional had suggested such an exam; and 2% said a friend or relative had made this suggestion. Only six women (1) indicated that their spouse (living or deceased) had ever objected to them having a breast exam, and two of them had had it anyway. Seventy-six percent of the women said they had read or seen on television that women should have their breasts examined. Fi- nally, 45% indicated that having a breast exam would make them somewhat or very nervous, and 58% said that they would be more likely to have a regular breast exam if the examiner were a woman.

Nearly 74% of the women had never had a mammogram (see Table 1). Eighty- two percent of all the women indicated that nobody had ever suggested that they have a mammogram; 17% stated that a health professional had suggested this exam; and only 1% said that a friend or relative had made the suggestion. Inter- estingly, 58% had read or seen on television that they should obtain a mammogram.

Knowledge and Behavior Pertaining to Breast Self-Examination

There was high variability of BSE knowledge among the women. Approxi- mately 80% of the subjects said they had heard of BSE, but only 22% knew they should do BSE once a month, and 27% indicated that they did not know how often they should perform BSE. Only 8% knew they should do BSE in two positions: standing up, then lying down. Seventy-eight percent knew they should not do BSE with their clothes on.

Fifty-nine percent of the women stated that they had done BSE within the last year, and 47% said they had done it within the last month; 33% said they had never done BSE. Forty-seven percent said that they were not at all nervous or

TABLE 1 RECENCY OF LAST BREAST SCREENING

Physician breast exam (%) Mammography (%)

Breast self-exam (%)

O-l years ago 50.44 12.46 59.31 l-3 years ago 21.12 7.68 5.38 Over 3 years ago 12.91 6.14 2.26 Never 15.53 73.72 32.99


embarrassed about examining their own breasts. Nearly two-thirds of the women indicated that nobody, in the past few years, had suggested that they should examine their breasts. Finally, 39% had been taught BSE by a physician or other medical person, and 70% said that they had read or seen on television that they should do BSE.

Assessment of Breast Self-Examination Performance

Ten percent of the women refused to touch the breast model, 5% touched the model but made no attempt to examine it, 6% attempted the exam with embarrassment, and 79% appeared comfortable examining the model. With t tests we examined whether willingness to touch the breast model was related to age and/or acculturation. The 10% who refused to touch the model (compared with the rest of the sample) were older (76 vs 71; P < O.OOl>, had fewer Anglo friends (P < 0.05), and tended to have less understanding of the English language (P < 0.07). Even stronger differences for these three variables were obtained when t tests compared the 79% who appeared to be comfortable while examining the breast model to the rest of the women. Thus, both age and acculturation were related to whether the women attempted the examination and whether they felt comfortable examining the model.

Excluding the 10% of subjects who refused to touch the model, nearly 50% did not find any lumps in the model, 24% found one lump, 13% found two lumps, 8% found three, 4% found four, and only 1% were able to find all five lumps present. The subjective assessments of the four BSE procedures for the subjects who did not refuse to touch the model are shown in Table 2. Between 10 and 20% of the women were definitely observed to perform each of the four criterion behaviors.

TABLE 2 PERCENTAGEOFSUBJECTSJUDGEDTOPERFORMBREASTSELF-EXAMINATIONADEQUATELY

Definitely did this

Correlation with number

Seemed Seemed Definitely to not to did not

of lumps found

do this do this do this Y P

Squeezed the nipple area

Moved fingers in a circular motion toward the nipple

Used the flat part of two to four fingers to feel the model

Put pressure on the model, not just super- ficial touch

11.95 17.28 25.18 45.59 0.30 <O.OOl

11.40 46.14 18.38 24.08 0.36 <O.OOl

16.73 44.49 15.07 23.71 0.36 <O.OOl

18.78 46.59 14.73 19.90 0.42 <O.OOl

Nore. Subjects who refused to touch the breast model (10%) were not included.


For the most part, women did not squeeze the nipple area, and few moved their fingers in a circular motion, used the flat part of several fingers, or put pressure on the model. As seen in Table 2, each of the four subjective assessments of BSE performance correlated positively with the number of lumps detected in the breast model. Putting pressure on the model appeared to be a key factor in determining BSE adequacy in that it had the highest correlation with number of lumps found.

Relation between Knowledge and Screening Behavior

The correlations between breast cancer knowledge and screening are presented in Table 3. The screening behaviors (except for mammography) show moderate intercorrelation, and for the most part, knowledge correlates positively with behavior. The low correlations for mammography are likely due to the fact that there was little variability associated with this behavior; most of the women had never had a mammogram.

Predictors of Breast Cancer Knowledge and Screening

Although knowledge about breast cancer and BSE was related to screening behaviors, it is difficult or impossible (with the present data) to determine the direction of the relationships. For example, it is possible that a person who knows about breast cancer will decide that an appropriate response to her knowledge is to obtain breast cancer screening (i.e., knowledge causing behavior). On the other hand, it is equally possible that she may know little about breast cancer, and may present herself to her physician for a routine physical exam during which time the physician may do a breast exam, either explaining about breast cancer in the process or causing her to seek more information (i.e., behavior causing knowledge).

Given the uncertainty surrounding the direction of causality between knowledge and behavior, the regression analyses reported below did not include knowledge as a predictor variable. Rather, knowledge and behavior were examined as separate, dependent measures. Seven regression equations were tested, one for each knowledge and behavior construct. The predictors (see Table 4) were en- tered into the equations simultaneously.

TABLE 3 INTERCORRELATIONSFORKNOWLEDGEANDBEHAVIORVARIABLES

Variables 1 2 3 4 5 6 I

1 Knowledge of breast cancer - 2 Recency of last breast exam 0.15* - 3 Recency of last mammography 0.07 0.32** - 4 Knowledge of breast self-exam 0.16* 0.22** 0.10 - 5 Performance of breast self-exam 0.23** 0.20** 0.15* 0.33** - 6 Recency of last breast self-exam 0.21** 0.24** 0.17* 0.45** 0.44** - 7 Frequency of breast self-exam 0.22** 0.26** 0.15* 0.46** 0.40** 0.7P -

* P < 0.05. ** P < 0.01.

TABL

E 4

REGR

ESSI

ON

ANAL

YSIS

FO

R BR

EAST

CA

NCER

KN

OWLE

DGE

AND

SCRE

ENIN

G BE

HAVI

OR

Depe

nden

t va

riable

s Ag

e

Pred

ictor

va

riable

s > 2

Lear

ned

Lear

ned

Perc

eived

BS

E Br

east

Lear

ned

abou

t ab

out

Lear

ned

8

Unde

r- su

scep

ti-

Bene

fits

mak

es

exam

s ab

out

mam

mog

- br

east

BSE

Equa

tion

Latin

st

and

bility

to

of

ear

ly m

e m

ake

me

BSE

mph

y ca

ncer

fro

m

an

E Ed

uc.

frien

ds

Engl

ish

canc

er

dete

ctio

n ne

rvou

s ne

rvou

s on

TV

on T

V on

TV

M

.D.

R2

F P

z

Know

ledge

of

br

east

8

canc

er

ns

0.18

***

ns

ns

“S

ns

ns

0.19

***

0.08

4.

52

<O.O

l Re

cenc

y of

las

t br

east

exam

inatio

n i

ns

0.13

**

“S

ns

ns

0.10

**

-0.1

3***

ns

0.08

4.

58

<O.O

l Re

cenc

y of

las

t m

am-

2 m

ogra

m

0.13

**

ns

-O.?

***

ns

IlS

ns

ns

0.27

***

0.10

5.

93

<O.O

l z

Know

ledge

of

BS

E ns

ns

ns

0.

F**

-0.1

3***

0.18

***

0.17

***

0.19

10

.74

<O.O

l t;

Perfo

rman

ce

of

BSE

-0.1

8***

ns

ns

“S

“S

ns

0.17

***

0.31

***

0.24

14

.61

<O.O

l Re

cenc

y of

las

t BS

E -0

.09*

* ns

ns

ns

“S

ns

-()

y4**

* 0.

19**

* 0.

29**

* 0.

30

20.1

9 co

.01

Freq

uenc

y of

BSE

-O

.lO**

ns

%

“S

ns

ns

ns

-0.2

2***

0.09

* 0.

23**

* 0.

22

13.6

1 <O

.Ol

ij

Note.

En

tries

ar

e sta

ndar

dized

re

gres

sion

weig

hts.

Va

riabl

es

that

we

re

not

inclu

ded

in a

par

ticul

ar

regr

essio

n an

alys

is ar

e le

ft bla

nk.

ns,

not

signi

fican

t (P

>

0.10)

. BS

E,

brea

st se

lf-ex

amin

atio

n.

* P

< 0.

10.

z

** P

< 0.0

5. 9

***

P <

0.01

.


Two demographic variables, age and education, were each found to have rather specific effects. Age was inversely related to each BSE measure (knowledge, performance, recency, and frequency), but not related to knowledge of breast cancer, recency of last breast exam, or recency of last mammogram. On the other hand, education was directly related to these latter three measures, but not related to any of the BSE measures. The two acculturation variables did not predict any dependent measure of knowledge or screening behavior.

The two health belief variables, perceived susceptibility to cancer and perceived benefits of early detection of cancer, were not found to be strong and consistent predictors. Perceived susceptibility did not produce any significant effects, and perceived benefits of early detection was predictive only of recency of last breast examination and knowledge of BSE.

In comparison to these two cognitive variables, the affective variables (nervousness and embarrassment) appear to be stronger and more consistent predictors. Self-report that physician breast examination makes one nervous, predicted recency of last breast exam. Further, self-report that BSE makes one nervous was a significant predictor of BSE knowledge, recency, and frequency. Nervousness was not related to observed competence of BSE performance among those who attempted the exam. This seems to indicate that nervousness about this procedure does not relate to BSE competence but rather to aversion to examining oneself regularly.

Strong media effects were also found. Having read or seen on television that one should do BSE was a significant predictor of BSE knowledge, performance, recency, and frequency. Having read or seen on television that one should have mammograms was the strongest predictor of recency of last mammogram; television or printed information about breast cancer was predictive of breast cancer knowledge.

Finally, learning BSE from a physician was a significant predictor of knowledge, recency, and frequency of BSE. Also, physician instruction in BSE was the strongest predictor of performance of or competence at BSE. Thus, learning BSE from a physician is predictive of both doing BSE and the quality of BSE.

DISCUSSION

Screening Behavior

Our data indicate that the level of breast cancer screening among our sample of elderly Hispanic women is not dramatically different from that of the general population. National surveys indicate that between 42 and 58% of women ages 45 to 64 have had a breast exam by a physician within the last year, although this is somewhat lower for older women (23, 30). Fifty percent of our sample indicated that they had had a breast exam within the past year. National surveys also indicate that 17 to 41% (the actual percentage is probably at the lower end of this range) of women over 50 years have ever had a mammogram, and only 4 to 15% have mammograms annually (13, 2 1, 30). Twenty-six percent of our participants had had a mammogram sometime in their life, and 12.5% had had one within the past year. Although care was taken to distinguish mammograms from other X-


rays of the torso, it is possible that some of our respondents may have confused mammograms with other X-rays and erroneously overreported this screening activity. Finally, according to national surveys, over 90% of all women are aware of BSE, over 75% report practicing it at least once in the past year, but only 24% of all women practice BSE monthly (28). Among our sample, 79% had heard of BSE, 59% had done BSE sometime in the past year, and 47% reported they had done BSE within the past month.

The national survey data reported above, as well as the present survey data, are based on individual self-report. Self-report methodology is particularly vul- nerable to self-presentational biases (e.g., overreporting positive behaviors). In this regard, it is important to note that Spanish heritage groups in the United States (compared with the general U.S. population) have been found to have a greater tendency to “yea-say” on health behavior questionnaires (i.e., agree with an item regardless of the content of the item) (1). This suggests that the actual frequency of screening among our sample of Hispanic women may be somewhat lower than the values obtained in our survey.

Although many of our subjects reported doing BSE, albeit irregularly, few were able to demonstrate adequate BSE technique on a foam breast model. Even when they were instructed to press hard on the model and try to find all lumps, only 1% found all five lumps present and almost 50% found no lumps. Thus, although the self-reported frequency of BSE appears to be consistent with national norms, our test of the adequacy of the behavior, as actually conducted, indicates that it is performed very poorly and would be useless in detecting breast lumps.

Because the present study is the first to examine breast cancer screening among elderly Hispanic women, our data can provide an important comparison base for future research in this area. In this regard, our findings on the frequency of breast cancer screening may not generalize to other groups of elderly Hispanic women. The women we sampled were living more or less independently in housing projects located in a large urban environment. They were covered for health-care expenses by Medicare and Medicaid and were sophisticated in their use of health and public services. The frequency of screening activity among elderly Hispanics residing in rural areas or those in urban areas who are less knowl- edgeable about the social and health services systems may be much lower. These women may be more prone to delay screening because they may have less access to health care or may rely more on the use of home remedies. Surveys of these Hispanic subgroups are needed to provide a more comprehensive picture of the frequency and adequacy of breast cancer screening among these groups.

Predictors of Knowledge and Screening

Our data confirm that age and education play a role in screening. We also confirm that among Hispanic women, as among other women, physician instruction in BSE is a key variable to its correct and regular performance. Although the acculturation variables we examined (understanding of English and proportion of Hispanic friends) correlated with whether our subjects were willing to attempt the breast model examination, these variables did not predict breast cancer knowledge or screening behavior.


Perceived susceptibility to cancer and perceived benefit of early detection, two variables associated with value-expectancy theories such as the health belief model (6, 31), were not, for the most part, useful predictors of screening. On the other hand, emotional reactions to screening (nervousness and embarrassment) were stronger predictors. Finally, the role of television or printed media in in- forming women about breast cancer and encouraging them to obtain screening was shown to be very important. In fact, more women learned about mammography from media than from their physicians. Further attention should be given to the role of television as a health-education tool for the elderly, because this group may be especially receptive to this medium as a source of information about health-care guidelines.

CONCLUSIONS

The purpose of our study was to examine the regularity of screening for breast cancer among elderly Hispanic women. We found that the screening regularity of our sample, which resided in a large urban environment and had good access to health care, was similar to that of national norms. Most of our participants, however, were unable to demonstrate effective BSE technique, and it is unlikely that their own attempts at self-examination would lead them to detect breast cancer early. This performance outcome, however, may be due more to age than to ethnicity. That is, any ethnic group of elderly women may have performed as ours did.

The fact that our study did not demonstrate lower screening activity among Hispanics compared with non-Hispanics should not be interpreted to mean that screening behavior is unimportant as a potential cause of later-stage diagnosis of breast cancer among Hispanics. Further studies are needed to examine the extent to which our findings generalize to other subgroups of elderly Hispanic women, for example, those in rural environments. Our data do suggest that given adequate access to health care (coverage by Medicare and Medicaid and residence near health-care facilities), Hispanic women will use screening services to the same extent as Anglo women. Furthermore, we did not find unique beliefs or barriers to screening among this population.

Finally, our study has provided an initial examination of only one potential cause of late-stage diagnosis of breast cancer. There are other reasons besides screening that should be examined, such as failure to detect, recognize, and ap- propriately respond to cancer symptoms. Each of these potential reasons should be given further examination.

REFERENCES

1. Aday, L. A., Chiu, G. Y., and Anderson, R. Methodological issues in health care surveys of the Spanish heritage population. Amer. .I. Public Health IO, 367-374 (1980).

2. Alagana, S. W., and Reddy, D. M. Predictors of proficient techniques and successful lesion detection in breast self-examination. Health Psychol. 3, I13- 127 (1984).

3. American Cancer Society. Public attitudes toward cancer and cancer tests. CA-Cancer J. C/in. 30, 92-98 (1980).


4. Anderson, R., Lewis, S. Z., Giachello, A. L., Aday, L. A., and Chiu, G. Access to medical care among the Hispanic population of the southwestern United States. He&h Sot. Behav. 22, 78-89 (1981).

5. Axtell, L. M., Asire, A. J., and Myers, M. H. (Eds.). “Cancer Patient Survival,” No. 5. National Institutes of Health, Bethesda, MD, 1976.

6. Becker, M. H., Haefner, P D., Kasl, S. V., Kirscht, J. P., Maiman, L. A., and Rosenstock, I. M. Selected psychosocial models and correlates of individual health related behaviors. Med. Care 15 (Suppl.), 27-46 (1977).

7. Calnan, M. W., and Moss, S. The health belief model and compliance with education given at a class in breast self examination. J. Health Sot. Behav. 25, 198-219 (1984).

8. Cancer Surveillance Program. Unpublished data. University of Southern California, Comprehen- sive Cancer Center, Los Angeles, 1982.

9. Chavez, L. R., Cornelius, W. A., and Jones, 0. W. Mexican immigrants and the utilization of U.S. health services: The case of San Diego. Sot. Sci. Med. 21, 93-102 (1985).

10. Daly, M., Osborne, C., Clark, G., and McGuire, W. Mexican-American breast cancer patients have a worse prognosis (Abstract). Proc. Amer. Sot. C/in. 0~01. 2, C-18 (1983).

11. Femandez-Caballero, C., Otterbein, S. A., and Romano, T. L. The Spanish speaking patient and the EMS system. Emerg. Med. Serv. 7, 57-59 (1978).

12. Gallup Organization Inc. “Women’s Attitudes regarding Breast Cancer.” Gallup Organization, Princeton, NJ, 1974.

13. Gallup Organization. “Public Awareness and Use of Cancer Detection Tests: 1983 Survey.” Gallup Organization, Princeton, NJ, 1984.

14. Greer, S. Psychological aspects: Delay in the treatment of breast cancer. Proc. Royal Sot. Med. 67, 470-473 (1974).

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

16. Hooper, E. M., Comstock, L. M., Goodwin, J. M., and Goodwin, J. S. Patient characteristics that influence physician behavior. Med. Care 20, 630-638 (1982).

17. Kegeles, S. S. Education for breast self-examination: Why, who, what, and how? Prev. Med. 14, 702-720 (1985).

18, Keller, K., George, E., and Podell, R. N. Clinical breast examination and breast self-examination experience in a family practice population. J. Fum. Prac. 11, 887-893 (1980).

19. Kelly, P. T. Breast self-examinations: Who does them and why. J. Behav. Med. 2, 31-38 (1979). 20. Lieberman, S. “A Study of the Effectiveness of Alternative Breast Cancer Public Education

Programs.” Lieberman Research, Inc., New York, 1977. 21. Lieberman, S. “A Basic Study of Public Attitudes toward Cancer and Cancer Tests.” Lieberman

Research, Inc., New York, 1979. 22. Mack, T. M., Walker, A., Mack, W., and Bernstein, L. Cancer in Hispanics in Los Angeles

County. Natl. Cancer Inst. Monog. 69, 99-105 (1985). 23. Makuc, D. M. Changes in use of preventive health services, in “Health in the United States,

1981.” DHHS Publ. No. (PHS) 82-1232. Offtce of Health Research, Statistics, and Technology, Hyattsville, MD, 1981.

24. Mammon, J., and Zapka, J. G. Determining the quality of breast self examination and its relation- ship to other BSE measures, in “Progress in Cancer Control IV: Reseach in the Cancer Center” (C. Mettlin and G. P. Murphy, Eds.). A. R. Liss, New York, 1983.

25. Manfridi, C., Warnecke, R. B., Graham, S., and Rosenthal, S. Social psychological correlates of health behavior: Knowledge of breast self-examination and routine physician examination on breast cancer mortality. Sot. Sci. Med. 44, 433-440 (1977).

26. Marcos, L. R., Urcuyo, L., Kesslernan, M., and Alpert, M. The language barrier in evaluating Spanish-American patients. Arch. Gen. Psychol. 29, 655-659 (1973).

27. Menck, H. R., Henderson, B. E., Pike, M. C., Mark, T., Martin, S. P., and Soo Hoo, J. Cancer incidence in the Mexican American. J. Nat/. Cancer Inst. 5, 531-536 (1975).

28. National Cancer Institute. “Breast Cancer: A measure of Progress in Public Understanding.” DHHS Technical Report No. 81-2306, Washington, DC, 1980.


29. National Cancer Institute. “The Breast Cancer Digest” (Carol Case, Ed.). National Cancer Insti- tute Publication No 84-1691, Bethesda, MD, 1984.

30. National Cancer Institute. “Increasing the Use of Mammography and Breast Palpation for Early Detection of Breast Cancer.” Request for Application, Bethesda, MD, 1986.

31. Rosenstock, I. M. The health belief model and preventive health behavior. Health Educ. Monog. 2, 354-386 (1974).

32. Trotta, P. Breast self examination: Factors influencing compliance. Onto/. Nurs. Forum 7, 13- 17 (1980).

33. Turnbull, E. M. Effect of basic preventive health practices and mass media on the practice of breast self-examination. Nurs. Res. 27, 98-102 (1978).

34. Westbrook, K. C., Brown, B. W., and McBride, C. M. Breast Cancer: A critical review of a patient sample with a ten year follow-up. South. Med. J. 58, 543-548 (1975).

35. Young, J. C., and Garro, L. Y. Variation in the choice of treatment in two Mexican communities. Sot. Sci. Med. 16, 1453-1465 (1982).

36. Zapka, J. G., and Mamon, J. A. Breast self-examination by young women. II. Characteristics associated with proficiency. Amer. J. Prev. Med. 2, 70-78 (1986).

Documents

Frequency and adequacy of breast cancer screening among elderly hispanic women