 |
 |
 |
 |
|
“If I Needed It, They Would Have Sent Me:”
Linda M. Hunt Judith C.D. Longworth and Katherine B. de Voogd
April 2002
About the Authors: Dr. Linda M. Hunt is Associate Professor of Anthropology and is
jointly appointed to the Department of Anthropology and the Julian Samora
Research Institute. She holds a Ph.D. in anthropology from Harvard University.
Her research has been concerned with health and health care among Latinos
in Mexico and the United States, particularly focusing on issues of ethnicity
and health, the management of chronic illness, and the culture of biomedicine. Dr. Judith C.D. Longworth is an Assistant Professor at The University
of Texas Health Science Center in San Antonio in the Department of Family
Nursing Care. She holds a Ph.D. in Nursing from Wayne State University.
Her research is concerned with women's health issues for minority women,
particularly in areas of contraception, management of menopause, and urinary
incontinence. Katherine B. de Voogd is Project Coordinator in the Division of Genetics and Metabolic Disorders, Department of Pediatrics at the University of Texas Health Science Center at San Antonio. She holds an MA in Sociology from the University of Guelph in Ontario, Canada. Her research has focused on women’s health issues in Bolivia and the United States, with a special focus on how women make decisions concerning their health care and behaviors.
Suggested Citation Hunt, Linda M. (Ph.D.), Judith C.D. Longworth, (Ph.D., RN., CS., FNP.), and Katherine B. de Voogd, (M.A.) ““If I Needed It, They Would Have Sent Me:” Cancer Screening, Knowledge and Adherence Among Older Hispanic Women,” JSRI Research Report #31, The Julian Samora Research Institute, Michigan State University, East Lansing, Michigan, 2002. Abstract The purpose of this study was to explore factors, other than patient knowledge, that might explain low use of cervical and breast cancer screening among Hispanic women. A questionnaire was used to assess knowledge of screening recommendations and self-reported adherence among 70 older Hispanic women in Texas. Most had high knowledge levels, but this did not predict adherence. Fourteen women, all with high knowledge levels, also answered a semi-structured qualitative interview. Barriers to screening discussed in qualitative interviews included transportation, time, cost, and believing screening to be unnecessary following previous negative screening, or when sexual activity is absent. Reminders and referrals from primary care providers were key to reported adherence. Establishing policies and procedures to assure consistent cancer screening reminders and referrals may improve rates of cancer screening among women similar to those in our study, especially in settings where there is little opportunity to develop long-term patient-provider relationships.
A good deal of recent research and intervention development has focused
on trying to understand and improve cancer screening among Hispanic women.
The primary thrust of these efforts has been on identifying practical
and conceptual barriers to screening (Harmon, Castro, & Coe, 1996;
Hubbell, Chavez, Mishra, & Valdez, 1996; Martinez, Chavez, & Hubbell,
1997; Morgan, Park, & Cortes, 1995; Ramirez, Suarez, Laufman, Barroso,
& Chalela, 2000; Secker-Walker, Vacek, Hooper, Plante, & Detsky,
1999; Suarez, 1994) and developing educational and motivational campaigns
to encourage women to use cancer screening (Ramirez, Suarez, Laufman,
Barroso, & Chalela, 2000; Vellozzi, Romans, & Rothenberg, 1996;
Yancey & Walden, 1994; Zavertnik, 1993). However, the relationship
between patient knowledge, attitudes and beliefs and their cancer screening
behavior is not straightforward. These factors have not been consistently
shown to be specifically associated with low adherence to cancer screening
(Harlan, Bernstein, & Kessler, 1991; Modiano, Villar-Werstler, Meister,
& Figueroa-Valles, 1995; Perez-Stable, Sabogal, Otero-Sabogal, Hiatt,
& McPhee, 1992), nor has patient knowledge and motivation been found
to reliably predict cancer screening behavior (Coughlin, 1998; Fox, Stein,
Gonzalez, Farrenkopf, & Dellinger, 1998; Ornstein, Garr, Jenkins,
Rust, & Arnon, 1991; Zimmerman, 1997). We are left with the question:
What factors, other than patient knowledge, may also contribute to lower
rates of cancer screening among Hispanic women? In an effort to identify such factors, we conducted an exploratory study of a group of older low-income Hispanic women in San Antonio, Texas. We selected only women who were already knowledgeable about cancer screening recommendations, and divided our sample between those who reported timely screening and those who did not. Using open-ended interview techniques, we explored the circumstances under which the women reported having actually used cervical and breast cancer screening services in the past, and reasons they gave for not having had regular screening. By limiting this study to women with substantial knowledge about cancer and cancer screening recommendations, we were able to identify several factors, other than knowledge, which may affect whether or not such women pursue cancer screening.
Methods The study consisted of two phases. In phase one, community recruitment was undertaken to identify women who had high levels of knowledge about cancer screening recommendations. In phase two, open-ended interviews were conducted with a sub-sample of women identified as having high knowledge levels, but divided between those who reported having adhered to those recommendations and those who had not.
Data Collection The National Hispanic Leadership Initiative on Cancer (NHLIC) conducted
a community education intervention among low income Hispanics in several
cities in the United States, including San Antonio. With the cooperation
of NHLIC staff, we recruited women for phase one of the study through
the same churches, libraries, and senior activity centers that had participated
in their project. We administered a brief questionnaire to a convenience sample of 70 women from six of these locations. They were all low-income, self-identified Mexican or Mexican-Americans, over 50 years of age, with no previous history of cancer. The questionnaire included socio-demographic and general health information questions, as well as cancer screening knowledge and adherence items. The questionnaires were administered at the recruitment sites, and were read to the individual in either English or Spanish, according to her preference. Permission was obtained at that time to contact the women later should they be chosen for the in-depth interview of phase two. To assess cancer screening knowledge, we asked how often women over 40 should have four types of cancer screening: mammography, clinical breast examinations, self-breast examinations, and pap/pelvic examinations. Women were classified as having “high” knowledge if their questionnaire responses were consistent with the current recommendations of the American Cancer Society (ACS) for women over 40 years old (Smith, Mettlin, Davis, & Eyre, 2000) for at least three of the four types of cancer screening. To assess level of adherence we asked when was the last time they had each of these examinations, and how many times they had had them within the past five years. We classified women as having “High” adherence if they reported completing at least three of these cancer screening behaviors at a frequency consistent with the ACS recommendations. Those who did not were classified as having “low” adherence. Based on their responses to the phase-one questionnaire, we selected 14 women for the in-depth interviews of the second phase of the project. All those interviewed in phase two had high knowledge of cancer screening recommendations; six had high adherence levels and eight had low adherence. In the open-ended interviews of phase two, the women were asked a set
of standardized questions and were encouraged to answer as expansively
as they cared to. These questions focused on their understandings, interpretations,
experiences and personal history regarding mammography, clinical breast
examinations, pap/pelvic examinations, and self-breast examinations. For
example, we asked what made it easier and harder for them to have the
screening tests done, and about their personal experiences surrounding
the first and last times they had pap tests and mammography. We also asked
general questions about their concepts, expectations and perceptions regarding
cancer and cancer screening; their sense of personal risk, and how they
applied this knowledge to themselves in their decision to pursue, or not
pursue, cancer screening. These interviews took place in the women’s homes or another quiet location of their choice. To ensure consistency, all phase two interviews were conducted by one interviewer (KBD), in English or Spanish, based on the interviewee’s preference. The tape-recorded interviews ranged from one to two-and-a-half hours in length. Analysis Data from both phases of the study were entered into an SPSS database
(Norusis, 1993) which was used to generate simple descriptive statistics
of the samples from each phase. For the phase one sample we also assessed
general associations between knowledge and adherence levels and various
dichotomized socioeconomic variables, using Fisher’s Exact Test (2-Tail).
This statistic was chosen because it is useful for testing associations
in small samples with small expected values (Fisher & van Belle, 1993).
In examining these correlations our goal is simply to describe the relationship
between these variables for this particular sample. We do not intend to
imply any predictive power. The core of our analysis focused on the content of data from the phase
two open-ended interviews. Analysis of this material took place in several
steps (Miles & Huberman, 1994). First, we wrote summaries of each
case, and then developed a method for displaying the interview data, building
initial matrixes of blocks of text (quotations and summations) for each
patient. These matrixes included general statements about facilitators
and barriers to screening behaviors, as well as more specific factors
associated with reported screening histories. Next, we identified trends
and patterns between cases, which we then summarized into higher-level
matrixes, contrasting informants grouped by level of adherence. All phases of data processing and analysis were cross-checked for consistency in coding and classification during conference sessions in which all project staff participated. Any anomalies or discrepancies were addressed and resolved during these sessions. Results The women interviewed in both phases of the study were similar in their
demographic characteristics. According to program staff, these characteristics
are typical of women attending the community programs from which they
were recruited. Participants ranged in age from mid-50’s to 90, with
a mean age of 70. Most of the women had low levels of education, with
less than eight years of schooling, and were about evenly divided between
Spanish and English speakers. The majority of women were widowed, divorced,
or separated. Nearly all had some form of health insurance, however, those
in the phase one group relied more on Medicare/Medicaid, while the phase
two group included a higher percentage of women with private insurance
(See Table 1).
Ninety-one percent (64/70) of the women answering community recruitment
questionnaire of phase one had high knowledge of cancer screening recommendations;
that is, they correctly named at least three of the ACS screening recommendations
for women over 40. Thus, while women included in phase two of the study
were pre-selected to be those with high knowledge, this level of knowledge
was quite common among all the women we interviewed in these community
programs. More than three-fourths (54/70, 77%) of the women surveyed in
the community programs reported what we have defined as “high”
adherence to breast and cervical cancer screening recommendations, having
followed at least three of the ACS recommendations. Still, fully 20% (13/64)
of those with high knowledge reported low levels of adherence.
It is noteworthy that we did not find a significant relationship between
levels of knowledge and levels of adherence. It would seem that factors
other than knowledge of screening recommendations affect the adherence
of these women. In the open-ended interviews of phase two, we asked the 14 women (who
were all knowledgeable about screening recommendations) what made it easier
or more difficult for them to have cancer screening tests. Table 3 summarizes
the facilitators and barriers they identified, contrasting the responses
of those reporting high versus low adherence. Not surprisingly, women reporting high adherence named fewer barriers
to cancer screening than did those with low adherence. The high adherence
group mentioned primarily practical barriers, such as lack of transportation
or time, or the cost of the tests. In contrast, those reporting low adherence
identified more conceptual barriers to screening, including not believing
in the benefit of screening, and feeling that they were not personally
at risk, either because their previous screenings had been negative, or
because they were no longer sexually active. In that most (80%) of the
women over 75 were unmarried, such concepts may help explain why older
and unmarried women were found, in our analysis of the phase one questionnaire
responses, to be significantly less likely to report being adherent. Both adherence groups identified similar factors as facilitating their
cancer screening behavior. These included: wanting reassurance about their
health or their families encouraging them to be screened, and having transportation,
insurance, the time to go for tests, and access to a female provider.
Women in the low adherence group additionally noted that they would be
more likely to go for screening if they experienced symptoms. In addition to these general questions about facilitators and barriers
to cancer screening, we also asked the circumstances under which the women
had actually come to have cancer screening in the past. We asked: “When
was the first Pap Screening Test you ever had? Tell me about that. Why
did you have it done at that time?” We repeated these questions for
their most recent Pap test, as well as for their first and last mammography.
The women’s answers to these four questions are especially revealing.
Table 3 presents a summary of their answers to these questions.
Two points seem critical. First, for women in both the low and high adherence
groups, by far the most commonly named factor influencing their actual
screening behavior was whether or not they had received a referral for
the test from their primary care provider. Women who well knew the cancer
screening recommendations for women in their age group, but had failed
to have timely cancer screening, often made comments such as: “Well,
I had an appointment last month for my blood pressure medication. He didn’t
tell me I needed to have a Pap. I figured that if I needed it, he would
have sent me.” Or conversely, those who were up-to-date in their
cancer screening made remarks like: “When I come in for my annual
checkup, my doctor just automatically does a Pap test. So that’s
how I always get it done.” Thus, whether or not these women received
consistent and timely recommendations and referrals for cancer screening
was a central factor determining whether or not they reported actually
following the recommendations with which they were already familiar. A second important point is that women in the high adherence group were more likely to report that they had established an annual routine that included cancer screening. While for many, like the woman quoted above, this was part of their having an annual general checkup, several others noted that they knew when it was time for their screening because they would receive reminder postcards or a phone call from the clinic they normally attended.
Discussion The purpose of this study was to develop an in-depth understanding of
the concepts and experiences of a small group of older Hispanic women
who were knowledgeable about cancer screening recommendations, exploring
the circumstances under which they reported following or not following
those recommendations. Due to the small sample size, limited sampling
frame, and spontaneous nature of the topics discussed, our findings cannot
be generalized to a larger population without further research. We also
should note that relying on self-reports of screening may overestimate
actual screening behavior. Still, we have gained some insight into how
knowledge of screening recommendations comes to be enacted, at least for
this small group of women. We found that knowledge of cancer screening recommendations was already
quite high among the women who answered our questionnaire in the community
programs. It is likely that these high knowledge levels are due, at least
in part, to the community’s recent participation in the NHLIC cancer
screening education intervention. Nevertheless, 20% of those with high
knowledge of the screening recommendations did not report adhering to
them, indicating that knowledge alone many not assure adherent screening
behavior. The open-ended interviews gave us some insight into this apparent anomaly.
We found that women in both the high and low adherence groups emphasized
practical considerations such as time, cost and transportation, in discussing
their use of cancer screening services. The low adherence group also mentioned
some conceptual barriers to screening, such as not believing in the tests
themselves, or thinking that the tests are not necessary for women who
are not sexually active, who have no symptoms, or who have had normal
tests results in the past. While the facilitators and barriers named by these women are of interest,
perhaps more telling are the circumstances under which they reported actually
having received screening tests in the past. Both the high and the low
adherence groups indicated that primary care providers’ recommendations
and referrals played a key role in their screening behavior. Many women
in the high adherence group had come to view annual testing as routine
because they received consistent annual referrals and reminders. At the
same time, several women in the low adherence group said they assumed
they didn’t need the tests because their primary care provider had
failed to recommend them. Indeed, receiving a recommendation from a primary
care provider was the principal reason given by the low adherence group
for having had any cancer screening tests at all. It seems that, while
knowledge of screening recommendations did not assure adherence among
these women, a primary care provider discussing cancer screening with
them, assisting with referrals, and/or establishing an annual screening
routine was crucial to the past screening behavior they reported. The fact that minority poor women are unlikely to have a long-term relationship
with a primary care provider (Berk & Bernstein, 1982; Cornelius, 1997;
McCourt & Pearce, 2000; O’Malley, Mandelblatt, Gold, Cagney,
& Kerner, 1997; Phillips, Mayer, & Aday, 2000), could help explain,
at least in part, the low rates of cancer screening reported for Hispanic
women. Many recent studies have found that having a regular source of
health care is a key predictor of use of cancer screening services, particularly
for older and minority women (Bindman, Grumbach, Osmond, Vranizan, &
Stewart, 1996; Caplan & Haynes, 1996; Carney, Dietrich, & Freeman,
Jr., 1992; Ettner, 1996; Kagawa-Singer & Pourat, 2000; Kelaher &
Stellman, 2000; Laws & Mayo, 1998; Mandelblatt et al., 1999; O’Malley,
Mandelblatt, Gold, Cagney, & Kerner, 1997; Zambrana R.E., N. Breen,
S.A. Fox, & M.L. Gutierrez-Mohamed, 1999). This may be because women
without a regular provider are less likely to consistently receive referrals,
reminders and recommendations. They also would not have the opportunity
to develop a provider-patient relationship adequate for identifying, discussing
and correcting concepts and beliefs that may present barriers to screening.
These propositions are supported by recent studies which report improvements
in cancer screening rates through programs that provide cancer screening
education to primary care providers, establish a system of reminder phone
calls or postcards to patients, and enhance communication between providers
and within the health care system (Burack et al., 1994; Burack et al.,
1998; Costanza et al., 1992; Taplin et al., 2000). Our findings draw attention away from the usual focus on patient knowledge and motivation as determinants of cancer screening among older Hispanic women, and instead point toward possible inadequacies in provider-patient relationships and clinic procedures which may be disproportionately affecting this population. Further research with a representative sample would be necessary before the generalizability of these findings can be checked. Still, our study indicates that cancer screening for older Hispanic women might be enhanced if consistent and reliable referral procedures are followed. Establishing explicit institutional policies and procedures for achieving this goal would seem especially important in clinic settings where the primary care provider has little opportunity to develop long-term relationships with patients.
Acknowledgements The authors express their gratitude to James Layman for his assistance with translation of the interview tools and data collection, and to Juliet Bickley for her assistance with recruiting and interviewing community questionnaire subjects. Special thanks to Dr. Roberto Villarreal for his help in establishing community contacts and sharing the NHLIC information. The study was funded by an Institutional Research Grant (IRG) from the University of Texas Health Science Center at San Antonio.
References Berk, M.L. & A.B. Bernstein. (1982). “Regular source of care
and the minority aged.” Journal of the American Geriatrics Society,
30, 251-254.
|
|
|
 |
For more information, contact: |
