Traditionality and Cancer Screening Practices Among American Indian Women in Vermont

This article was downloaded by: [Temple University Libraries]On: 19 November 2014, At: 10:09Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Health Care for Women InternationalPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/uhcw20

Traditionality and Cancer ScreeningPractices Among American Indian Womenin VermontMary K. Canales a , William Rakowski b & Alan Howard ca Department of Nursing , University of Vermont , Burlington,Vermont, USAb Department of Community Health , Brown University , Providence,Rhode Island, USAc Academic Computing Services , University of Vermont , Burlington,Vermont, USAPublished online: 12 Mar 2007.

To cite this article: Mary K. Canales , William Rakowski & Alan Howard (2007) Traditionality andCancer Screening Practices Among American Indian Women in Vermont, Health Care for WomenInternational, 28:2, 155-181

To link to this article: http://dx.doi.org/10.1080/07399330601128544

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the“Content”) contained in the publications on our platform. However, Taylor & Francis,our agents, and our licensors make no representations or warranties whatsoever as tothe accuracy, completeness, or suitability for any purpose of the Content. Any opinionsand views expressed in this publication are the opinions and views of the authors,and are not the views of or endorsed by Taylor & Francis. The accuracy of the Contentshould not be relied upon and should be independently verified with primary sourcesof information. Taylor and Francis shall not be liable for any losses, actions, claims,proceedings, demands, costs, expenses, damages, and other liabilities whatsoever orhowsoever caused arising directly or indirectly in connection with, in relation to or arisingout of the use of the Content.

This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &

http://www.tandfonline.com/loi/uhcw20

http://dx.doi.org/10.1080/07399330601128544

Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Dow

nloa

ded

by [

Tem

ple

Uni

vers

ity L

ibra

ries

] at

10:

09 1

9 N

ovem

ber

2014

http://www.tandfonline.com/page/terms-and-conditions

http://www.tandfonline.com/page/terms-and-conditions

Health Care for Women International, 28:155–181, 2007Copyright © Taylor & Francis Group, LLCISSN: 0739-9332 print / 1096-4665 onlineDOI: 10.1080/07399330601128544

Traditionality and Cancer Screening PracticesAmong American Indian Women

in Vermont

MARY K. CANALESDepartment of Nursing, University of Vermont, Burlington, Vermont, USA

WILLIAM RAKOWSKIDepartment of Community Health, Brown University, Providence, Rhode Island, USA

ALAN HOWARDAcademic Computing Services, University of Vermont, Burlington, Vermont, USA

The study purpose was to examine the relationship between cancerscreening and traditionality, using a culturally specific surveyinstrument. American Indian (AI) women were randomly selectedfrom the Vermont Breast Cancer Surveillance System (VBCSS),a statewide mammography database. The 13 items that assessedtraditionality examined identity, cultural beliefs, customs, andhealth practices. The sample of 115 was predominately married,educated, middle class, insured, and rural. Significant associationswere found between traditionality and past mammography andclinical breast examination (CBE) behavior, as well as futuremammography intention. Results suggest that traditionality scalescan be useful for identifying differences in screening behavior,with further testing of traditionality items with other AI womenrecommended. International implications of the study findings arealso addressed.

American Indian (AI) women are a diverse group, with significantly differentcancer experiences depending on geographic region and tribal affiliation(Hampton & Friedell, 2001). Breast cancer is the second leading cause of

Received 1 November 2005; accepted 30 January 2006.The authors recognize funding for the project from the National Cancer Institute U01-

CA70013–07S1 (Dr. Canales) and K05-CA90485 (Dr. Rakowski). We thank the women whoparticipated in the study and the staff of the Vermont Breast Cancer Surveillance System, whoassisted with data management and recruitment of study participants.

Address correspondence to Mary Canales, 215 Pine Grove St. SW, Bemidji, MN, 56601,USA. E-mail: [email protected]

155

Dow

nloa

ded

by [

Tem

ple

Uni

vers

ity L

ibra

ries

] at

10:

09 1

9 N

ovem

ber

2014

156 M. K. Canales et al.

cancer death for AI women (Swan & Edwards, 2003), with cervical cancerincidence and mortality rates for this population of women among thehighest in the United States (Hampton & Friedell, 2001). For example, inthe Aberdeen Indian Health Service (IHS) area, the cervical mortality rate isfive times that of all races in the United States (15.6 vs. 3.0; Giroux et al.,2000). Death from breast and cervical cancer can be reduced substantially,however, if the tumor is diagnosed at an early stage (Department of Health& Human Services [DHHS], 2001).

Although mammography screening is currently the most effectivemethod for detecting early breast malignancies, AI women have lowscreening rates. According to the American Cancer Society (ACS, 2004), only37% of AI women 40 and above had a mammogram within the past year,compared with 57% of U.S. non-Hispanic White women, while only 52% ofAI women had a mammogram within 2 years compared with 72% of U.S.non-Hispanic White women. Although regular Papanicolauou (Pap) smeartesting can significantly reduce the risk of invasive cervical cancer, AmericanIndian/American Native (AI/AN) screening participation rates are also belowthose of non-Hispanic, White women. According to the ACS (2000), 78%of AI/AN women 18 years and above received a Pap test every 3 years,compared with 84% of non-Hispanic White women 18 years and above.

Considering the current disparities in screening rates, much work liesahead to reach the Healthy People 2010 goals of 70% of AI/AN womenaged 40 and above having a mammogram within 2 years and 90% of AI/ANwomen aged 18 and above receiving a Pap smear every 3 years (DHHS,2000) . Although efforts to increase screening rates among this population ofwomen have had some success (Brant, Fallsdown, & Iverson, 1999; Kottke &Trapp, 1998; Lanier, Kelly, & Berner, 1999; Smith, Christopher, & McCormick,2004), there is minimal understanding of what influences AI women’s breastand cervical cancer screening decisions or how these influencing factors maydiffer from those of the majority White population. The study being reportedhere begins to address these questions.

Traditionality was identified as an influencing factor in mammographydecision-making in a qualitative study with AI women in Vermont (Canales &Geller, 2004). To further examine the relationship between cancer screeningdecision making and traditionality, a culturally specific survey instrumentwas created, informed by the qualitative data, and pilot tested with a sampleof AI women in Vermont. The traditionality items included in the surveywere adapted from ethnic identity research with AI adolescents (Moran,Fleming, Somervell, & Manson, 1999) and AI women in Oklahoma (Solomon& Gottlieb, 1999).

In this study, traditionality refers to the extent to which an individualculturally identifies with his or her racial/ethnic heritage and encompassesself-identification, group identification, and the social environment (context).Solomon and Gottlieb (1999) suggest that the physical and cultural genocide

Dow

nloa

ded

by [

Tem

ple

Uni

vers

ity L

ibra

ries

] at

10:

09 1

9 N

ovem

ber

2014

Traditionality and Cancer Screening Practices 157

experienced by AI populations has resulted in a diversity of changes inthe cultural beliefs, practices, and customs of many AI nations. Althoughtraditionality often exists along a continuum, from American Indians whomaintain or prefer their tribal ways and identify as more traditional tothose who have become more acculturated to White, mainstream cultureand identify as less traditional, fluidity exists between these culturalpositions and is influenced by cognitive, behavioral, spiritual/affective, andsocial/environmental experiences (Walters, 1999). The purpose of the studybeing reported here was to investigate the relationship between traditionalityand cancer screening behavior among a sample of AI women in Vermont.

With regards to terminology, we have chosen to use the phrase“American Indian” to describe the sample population. “American Indian” isthe term currently used by many tribal groups in the United States to describethemselves as well as the term used by the U.S. Census Bureau when, forexample, presenting health indicators and statistical data (Roubideaux, 2002).

Following an overview of existing research with AI women related tocancer screening and traditionality, the results of a pilot study with AI womenin Vermont will be presented. We conclude the article with a discussion ofimplications for research and practice, directions for future research efforts,and limitations of the study.

LITERATURE REVIEW

The influence of traditionality on cancer-screening participation has receivedsome attention from researchers working with AI women (Canales, 2004b;Coe et al., 2004; Solomon & Gottlieb, 1999; Weiner, 1999). Traditionality,however, has not been consistently defined across studies. This review,therefore, approaches traditionality from a broad perspective and includesresearch that has examined the influence of cultural beliefs, attitudes,customs, or practices on AI women’s cancer screening participation.

Several studies have identified a positive role for culture in cancerprevention and control programs (Coe et al., 2004; Giuliano, Papenfuss, deZapien, Tilousi, & Nuvayestewa, 1998; Risendal, Roe, DeZapien, Papenfuss,& Giuliano, 1999). Giuliano and colleagues reported that AI cultural identitymay facilitate participation in breast cancer screening services. A breastand cervical cancer knowledge, attitudes, beliefs, and behaviors (KABB)survey was given to 314 women living on the Arizona Hopi reservation. Theresearchers found that women who reported attending cultural ceremoniessome of the time were nearly two times as likely to have had a mammogramin the past 2 years (OR = 1.8) and to report a yearly CBE (CBE)(OR = 2.4),while use of the Hopi language some of the time was associated with anearly twofold increase in the likelihood of having a mammogram in thepast 2 years (OR = 1.8). Although they did find that women who soughtthe services of a traditional healer were 40% less likely to report having a

Dow

nloa

ded

by [

Tem

ple

Uni

vers

ity L

ibra

ries

] at

10:

09 1

9 N

ovem

ber

2014


yearly CBE (OR = 0.61), a similar impact on mammography utilization wasnot found. The researchers concluded that, overall, a proportion of Hopiwomen were successfully using the strengths of both their Native cultureand Western medical systems to promote health (Giuliano et al., 1998).

Risendal and colleagues (1999) also suggested that a positive rolefor culture in cancer prevention and control programs exists for urban AIwomen. They examined the relationship between culture and breast cancerscreening in a study with 519 adult AI women in Phoenix, Arizona. Theresearchers assessed “cultural values,” specifically attendance at ceremonies(sociodemographic variable), and a fatalistic view toward cancer and cancerscreening (attitudinal variable; p. 508). In terms of the attitudinal variable,the researchers reported that the majority of women surveyed did nothave a fatalistic view of cancer. They did not believe they would dieof cancer; they viewed early detection as important; and they believedthat CBE and mammography can detect cancer early. In terms of thesociodemographic variable, they reported that attendance at ceremonies,such as social gatherings and tribal-specific religious ceremonies, was “apositive predictor of breast cancer screening (CBE)” in the age-adjustedanalyses (OR = 2.21, p. 508).

Coe and colleagues (2004) examined the influence of traditionalism onhealth risks related to cancer, specifically smoking, alcohol consumption,and obesity. Three dimensions of traditionalism were assessed: languageusage, cultural participation, and percentage of life spent off-reservation.They tested an instrument designed to analyze the relationship betweentraditionalism and health risk behaviors with a random sample of 559 women,aged 18 years and above, living on the Hopi reservation. Findings indicated asignificant positive relationship between traditionalism and health behaviors.The most traditional women were more likely to participate in health-promoting behaviors, such as employing Hopi practices to keep healthyand seeking the services of a traditional healer, and least likely to engagein health-risk activities, specifically smoking and alcohol consumption.The authors recommended that traditionalism be incorporated into healtheducation interventions and public health programs aimed at decreasingchronic disease rates among AI populations.

In contrast to the above studies, other researchers have found evidencesuggesting that traditionality may hinder participation in cancer screening.For example, Solomon and Gottlieb (1999), in their study with 196 AIwomen in Oklahoma, reported an inverse relationship between traditionalbehavior and cervical cancer screening. Women who scored higher ona traditional behavior scale were less likely to follow recommended Papscreening guidelines than those with lower traditionality scores. The authorssuggested that AI women in the study may have relied on cultural traditionsto deal with the complexity of their day-to-day lives, traditions that maydiscourage seeking health care when one is not ill.

Dow

nloa

ded

by [

Tem

ple

Uni

vers

ity L

ibra

ries

] at

10:

09 1

9 N

ovem

ber

2014


The issue of when to seek health care was also raised in a qualitativestudy evaluating the effectiveness of an early detection cervical cancerintervention with Cherokee and Lumbee women in North Carolina (Messer,Steckler, & Dignan, 1999). Although the researchers reported that culturalfactors negatively influenced women’s participation in the interventionprogram, only one “cultural factor” was identified: the “cultural belief”that “one does not go to the doctor unless one is sick” (p. 558). Theresearchers reported that this belief was common to both groups of womenand limited their participation in the program. According to the researchers,this “cultural belief” conflicted directly with program instructions for a yearlycervical cancer screen, and consequently attenuated the effectiveness ofthe intervention. Although the authors recommended that an assessmentof “cultural factors” be conducted prior to and throughout early detectionintervention projects, they did not define these factors in general, or for theseparticular tribes.

Seeking health care services also may be impeded by AI women’sstrong sense of privacy. Modesty has been identified as a factor limitingNative women’s willingness to be screened for cancer (Clarke et al., 1998;DHHS, 2002; Michielutte, Dignan, Sharp, Blinson, & Wells, 1999; Stevenet al., 2004). In a study with First Nations women, Clarke and colleaguesreported that the study participants were “embarrassed and intimidated bythe procedure” (Pap test) and they “objected to getting fully undressedand being exposed on the examining table” (p. 38). Steven and colleaguesreported similar results from their study with Ojibwa and Oji-Cree womenfrom Northwestern Ontario. The researchers found that the Native womenparticipants had a more intense sense of privacy than the Italian, Ukranian,or Finnish women in the study. The Native women resented having strangerstouch them and found breast and cervical examinations to be aversive andintrusive. The Native women reported they had not been informed aboutbreast and cervical cancer screening and, despite their privacy concerns,requested more information.

A lack of knowledge about cancer causation also has been identifiedas a possible hindrance to participation in cancer screening among AIwomen. Strickland and colleagues (1996) conducted a grounded theorystudy with 15 women from the Yakima tribe in eastern Washington togain an understanding of the meaning of the Pap test. Yakima womenassociated the Pap test with childbearing and sexual activity, rather thanwith cancer and cancer prevention. The researchers reported, “Many Indianpeople believe that the basis of illness is a loss of one’s sense of selfand identity” (Strickland, Chrisman, Yallup, Powell, & Squeoch, p. 147).The researchers reported that there was an identity continuum within thetribe, from very traditional to more acculturated women. They did not,however, link the varying degrees of traditionality with cervical screeningbehavior.

Dow

nloa

ded

by [

Tem

ple

Uni

vers

ity L

ibra

ries

] at

10:

09 1

9 N

ovem

ber

2014


Other research has pointed out that among certain tribes and individuals,thinking or speaking about cancer may cause its onset (Locust, 1986; Olsen,1993). In her study of AI‘s beliefs about health and unwellness, Locustsaid, “Many tribes believe that speaking about an illness may give it thepower to manifest or express itself in human form” (as cited in Olsen,1993, p. 15). Olsen reported, “Some traditional patients believe use of firstperson language involves one’s spirit and may cause them to get cancer”(1993, p. 47). In contrast, ethnographic studies with AI in California suggestthat cancer discourse is very complicated (Weiner, 1999). Speaking aboutcancer is influenced by many factors, including the amount of contact withcancer survivors, accessible information about cancer, and culturally sensitivediscussions with respected health care providers (Weiner, 1999). These dataindicate that thinking or speaking about cancer can both hinder or facilitateparticipation in cancer screening, depending on the situation.

A dual role for traditionality also was evident in the grounded theorystudy with AI women in Vermont (Canales, 2004b). The influence of“connecting to nativeness” on mammography decision making was identifiedas a salient factor in this study. In this context, “connecting to nativeness”referred to the extent to which a woman culturally identified with herracial/ethnic heritage: her degree of traditionality. The generated theory,“moving in between mammography,” suggested that degree of traditionalityinfluenced women’s decision making, acting both as a facilitator for, andan inhibitor of, participation in mammography screening (Canales, 2004b;Canales & Geller, 2004). The findings from the grounded theory studysuggested that a focus on traditionality may provide an avenue for examininghow women transition toward or away from routine screening. Traditionalityalso may play a role in how AI women perceive the use of technology inrelation to their Native beliefs.

This review of relevant literature supports the need for furtherinvestigation of the relationship between cancer screening decisions andtraditionality. Currently, it seems that traditionality is as likely to facilitatebreast and cervical cancer screening behavior as it is to act as an inhibitor.If ways to promote the former while intervening with the latter can beidentified, providers may be able to improve cancer-screening rates amongdiverse groups of AI women.

METHODS

Instrument

The study was designed to pilot test a mammography-screening instrumentdeveloped by the authors for use with AI women. Qualitative data from agrounded theory study conducted with 20 AI women in Vermont (Canales,2004a, 2004b; Canales & Geller, 2004) suggested that degree of traditionality

Dow

nloa

ded

by [

Tem

ple

Uni

vers

ity L

ibra

ries

] at

10:

09 1

9 N

ovem

ber

2014


influenced AI women’s screening decisions. The instrument was thereforedesigned to examine the relationship between mammography decisionmaking and traditionality. Item development was informed by the qualitativedata, the transtheoretical model (TTM) of behavior change as applied tomammography screening (Rakowski, Goldstein, & Dube, 1996), and existingresearch conducted with AI women (e.g., Brant et al., 1999; Foxall, Barron,& Houfek, 2001; Strickland et al., 1996).

In addition to the TTM-specific mammography questions, the surveyincluded 13 questions that assessed traditionality, 10 questions regardingcancer-screening practices (mammography, CBE, breast self-examination[BSE], Pap test), 2 questions requested breast cancer history (personal andfamilial), 6 questions that focused on hormone therapy use (prescription andover-the-counter [OTC]), and 4 questions related to demographic information(age, education level, yearly household income, and current marital status).The focus of this report is limited to an analysis of the relationship betweentraditionality and cancer screening practices (mammography, CBE, BSE, andPap test).

The traditionality items were adapted from existing scales (Table1). Eight questions, based on the bicultural ethnic identity scale (Moranet al., 1999), dealt with ethnic affiliation, current cultural activities, languageuse in childhood home, and importance of religious/spiritual beliefs. Theseethnic identity items have been used in other health-related research with AI,including patient satisfaction (Garroutte, Kunovich, Jacobsen, & Goldberg,2004) and use of biomedical services, traditional healing practices, or both(Buchwald, Beals, & Manson, 2000; Novins et al., 2004). The ethnic identityquestions were answered on a 4-point Likert scale, from 1, not at all, to4, a lot. The other traditionality items were adapted from Solomon andGottlieb (1999). One question asked respondents to choose the statementthat best described their ethnic identity and 4 questions assessed culturalhealth practices, which were answered on a 5-point Likert scale, from 1, notat all, to 5, all the time.

Measurement

We hypothesized that a relationship existed between traditionality and breastand cervical cancer screening behavior. To test this hypothesis, the followingdependent variables for breast and cervical cancer screening were used:

Ĺ Mammogram within the past 2 yearsĹ Intention to have a mammogram within the next 12 monthsĹ CBE within the past yearĹ BSE within the past monthĹ Pap test within the past 3 years

Dow

nloa

ded

by [

Tem

ple

Uni

vers

ity L

ibra

ries

] at

10:

09 1

9 N

ovem

ber

2014


TABLE 1 Traditionality Items

Not at all Rarely Occasionally Often All the time

I attend American Indianspiritual ceremonies.

�1 �2 �3 �4 �5

I participate in traditional Indianhealth practices such as sweatlodge or burning sage.

�1 �2 �3 �4 �5

I practice traditional Indian craftssuch as basket weaving.

�1 �2 �3 �4 �5

I seek guidance from atraditional Indian healer orelder.

�1 �2 �3 �4 �5

Not atall

A little Some A lot

Do you live by or follow the White/Angloway of life?

�1 �2 �3 �4

Do you live by or follow the AmericanIndian way of life?

�1 �2 �3 �4

Some families have special activities ortraditions (such as special meals,religious activities). In your family, howmany activities or traditions are based onWhite/Anglo culture?

�1 �2 �3 �4

In your family, how many activities ortraditions are based on American Indianculture?

�1 �2 �3 �4

How often were English/French spoken inyour home when you were growing up?

�1 �2 �3 �4

How often was a tribal language spoken inyour home when you were growing up?

�1 �2 �3 �4

How important is it for you to followreligious and/or spiritual beliefs that arebased on Christian beliefs such asCatholic, Protestant, Baptist, etc.?

�1 �2 �3 �4

How important is it for you to followreligious and/or spiritual beliefs that arebased on traditional Indian beliefs?

�1 �2 �3 �4

Which of the following best describes you? Please choose only one answer.a. I keep and follow the traditional ways and customs of my tribe.b. I follow some of the traditional practices and beliefs of my tribe.c. I purposefully follow both mainstream and tribal cultures in order to fit in.d. I am currently returning to or learning for the first time the traditional ways.e. I am like or similar to non-Indians in most ways.

The independent variables included demographic characteristics forage (59 years and below or 60 years and above), education (collegegraduate or not), marital status (married/partner or other), income (<$36,000or >$36,000), and traditionality. Traditionality was assessed using the 13traditionality items in the survey. This approach allowed us to examinetraditionality more broadly, in terms of an identity continuum, as well asmore specifically, in terms of cultural beliefs, customs, and health practices.

Dow

nloa

ded

by [

Tem

ple

Uni

vers

ity L

ibra

ries

] at

10:

09 1

9 N

ovem

ber

2014


Human Subjects

Human subjects’ approval for the study was received through the sponsoringuniversity and the peer review board for the Vermont Breast CancerSurveillance System (VBCSS), a statewide database of all women who havehad mammograms since 1994 (Geller, Worden, Ashley, Oppenheimer, &Weaver, 1996). This second approval was necessary because the VBCSSwas the recruitment source for potential participants. The VBCSS has beencollecting data since 1994, using well-established processes for data entry,validation, and error retrieval (Geller et al., 1996). In 2001 the VBCSS began toscan all patient questionnaires, further enhancing the efficiency and accuracyof data management processes.

Recruitment

In Vermont, the AI population accounts for 0.4% of the total population,or 24,200 (United States Bureau of the Census, 2001). In this primarilyhomogeneous state, where 96.8% of the population is White, AI are thethird largest racial group (U.S. Census). Although this is a small percent ofthe total population, sufficient numbers of AI women were available in theVBCSS for recruitment into the study. In addition to mammography history,the database contains demographic information including race, age, andmammography facility. A questionnaire, which is completed at the time ofeach mammogram, includes a consent signature line. Women who sign thisform authorize researchers to contact them in the future. All information fromthe questionnaire is entered into the database and maintained by well-trainedstaff.

There were three advantages of using the VBCSS for recruitment.One, the database provides access to a geographically dispersed, highlydiverse sample of AI women. Because the government does not recognizedistinct tribal nations or designate tribal lands, it is very difficult to recruita random sample of AI women in Vermont without a formal samplingresource. The VBCSS provided that formal resource. Second, profiles ofresponders and nonresponders could be compared using VBCSS data.Finally, mammography history could be assessed based on clinic recordsfor use in statistical analyses. Use of actual mammography dates, rather thanreliance on self-report, increases the accuracy of the results. Limitations ofusing the VBCSS will be addressed later in the article.

Potential participants were randomly selected from the VBCSS. Inclusioncriteria for the study were self-identification as AI, which women indicated byselecting “American Indian” as their racial category on the VBCSS question-naire. Because the database manager noted that some women answeredthis question differently over time (L. Riddell, personal communication,April 2004), women were only included in the sampling frame if “American

Dow

nloa

ded

by [

Tem

ple

Uni

vers

ity L

ibra

ries

] at

10:

09 1

9 N

ovem

ber

2014


Indian” was selected on two consecutive questionnaires. In order to examinescreening patterns over an extended time, we included women whose mostrecent mammogram was within the past year as well as women whose mostrecent mammogram was 3 or more years ago.

Based on these criteria, 300 AI women were randomly selected fromthe VBCSS for the sample pool. Our goal was to enroll 150 women into thestudy (50% response rate). It was anticipated that this response rate wouldbe achieved, as previous research using the VBCSS for recruitment was oftenin the 80% range (B. Geller, personal communication, November 2003).

After the 300 women were selected, the appropriate mammographyfacilities were faxed their names and they, in turn, provided the VBCSSstaff with the women’s contact information. The VBCSS staff then providedthe names and contact information to the first author. As the names werereceived, the first author mailed a packet containing a cover letter describingthe recruitment process and the purpose of the study, the survey itself,two bookmarks, and an AI-specific pamphlet on breast cancer. The initialmailing was completed over a 4-week period, during August–September2004. The first author conducted all follow-up efforts, including mailed postcards and telephone calls, over a two-month period, November–December2004. Despite these efforts, the final response rate was 43%, below the targetand therefore a limitation of the study.

Statistical Analyses

SAMPLE DESCRIPTION

To investigate whether the sample of responders may be biased, comparisonof several demographic variables between those who responded and thosewho did not respond was done. Age was compared via a t test, whilemammography history and geographic location were compared with a chi-square test.

CANCER SCREENING PRACTICES

The following variables having an ordinal scale were analyzed with a Mann-Whitney test to compare women who were on schedule with their mostrecent mammogram in relation to those who were not: most recent CBE,Pap test, and BSE.

TRADITIONALITY

Spearman’s rank-order correlation was used to test for significant correlationbetween the 13 traditionality variables and the following sets of variables:

Dow

nloa

ded

by [

Tem

ple

Uni

vers

ity L

ibra

ries

] at

10:

09 1

9 N

ovem

ber

2014


demographic (age, income, education), cancer screening (BSE, Pap smear,CBE), and mammography screening behavior. Mann-Whitney tests were usedto compare traditionality between those women who were on schedule forthe most recent mammogram and those who were not. Mann-Whitney testsalso were used to compare traditionality between women who plannedto have a mammogram in the next 12 months and women whose futureintention was questionable. All statistical analyses were conducted usingSPSS 12 for Windows (2004).

RESULTS

Sample Description

Data for this report are based on a final sample of 115 women, 40–82 yearsold (M = 57.8, SD = 8.4). There were no statistical differences betweenresponders and nonresponders in terms of age, mammography history, orgeographic location (data not shown; all p values > 0.1). Sociodemographiccharacteristics of the women showed that the sample was predominatelymarried, educated, middle class, insured, and rural. In Table 2 we presentthe sociodemographic characteristics of study participants.

Demographic variables also were analyzed with regards to traditionality.Of the 13 items intended to measure traditionality, analyses indicated that 7of the items were significantly associated with the demographic variables ofage, income, and education (Spearman’s rho). Younger women were morelikely to follow Indian religious beliefs (rho = −0.22, p < .05), and youngerwomen (rho = −0.24) and poorer women (rho = −0.27) were significantlymore likely to seek guidance from a traditional healer (p < .01). In contrast,wealthier women were significantly more likely to follow White/Anglo waysof life and traditions based on White/Anglo culture (rho = 0.28, p < .01,and rho = 0.24, p < .05, respectively). Older women were significantly morelikely to follow Christian religious beliefs (rho = 0.19, p < .05). Youngerwomen (rho = −0.29, p < .01) and more educated women (rho = 0.19, p <

.05) were significantly more likely to participate in traditional health practices.

Cancer Screening Practices

MAMMOGRAPHY

As noted earlier, all of the women had at least two mammograms during thepast 5 years, a criterion for recruitment into the study. There was, however,a range in terms of participation in regular screening. According to VBCSSrecords, 50.4% (58/115) of women had a mammogram within the past year,32.2% (37/115) within 1–2 years, and 17.4% (20/115) more than 2 yearsago. Using VBCSS data we also were able to compare women who wereon schedule with their most recent mammogram (<2 years) and those who

Dow

nloa

ded

by [

Tem

ple

Uni

vers

ity L

ibra

ries

] at

10:

09 1

9 N

ovem

ber

2014


TABLE 2 Sociodemographic Characteristics of Study Par-ticipants (n=115)

Characteristics n (%)

Age40—49 33 (29.2)50—59 48 (42.5)60—69 24 (21.2)70—79 6 (5.3)80 and above 2 (1.8)Missing 2 (1.7)

EducationLess than high school 5 (4.3)High school graduate 27 (23.5)Education past HS 36 (31.3)College graduate 22 (19.1)Education past college 25 (21.7)

IncomeLess than $10,000 11 (10.4)$10,000-$14,999 10 (9.4)$15,000-$35,999 37 (34.9)$36,000-$50,999 23 (21.7)$51,000-$75,000 13 (12.3)More than $75,000 12 (11.3)Missing 9 (7.8)

Marital statusMarried/partnered 79 (68.7)Separated 2 (1.7)Divorced 24 (20.9)Never married 8 (7.0)Widowed 2 (1.7)

Insurance statusPrivate insurance 52 (56.5)Medicare 15 (16.3)Medicaid 12 (13.0)Ladies first∗ 8 (8.7)Self-pay 5 (5.4)Missing 23 (20.0)

Facility locationRural 95 (82.6)Urban 20 (17.4)

∗Ladies First is the CDC Breast and Cervical Cancer ScreeningProgram in Vermont for underserved, eligible women.

were not on schedule (>2 years) in relation to most recent CBE, Pap test,and BSE practice. Although not significant, having a recent CBE (p < .08)and Pap test (p < .06) were positively correlated with being on schedule forthe most recent mammogram.

To analyze the relationship between mammography screening andtraditionality, the actual date of the most recent mammogram from the VBCSSwas used to divide the sample into two comparison groups: women onschedule (mammogram < 2 years) and women overdue (mammogram >

2 years). These two groups were compared (Spearman’s rho) to examine

Dow

nloa

ded

by [

Tem

ple

Uni

vers

ity L

ibra

ries

] at

10:

09 1

9 N

ovem

ber

2014


the correlation between traditionality and being on schedule for themost recent mammogram. Women who were on schedule for their mostrecent mammogram (<2 years) were significantly more likely to describethemselves as similar to non-Indians in most ways (rho = 0.23, p < .05). Incontrast, women who were overdue for their most recent mammogram (>2years) were significantly more likely to report following the Indian way oflife (rho = −0.21), to seek guidance from a traditional healer (rho = −0.22,both p < .05), and to attend spiritual ceremonies (rho = -0.25, p < .01).

To further examine the relationship between being on schedule formammography and degree of traditionality, the 13 traditionality itemsalso were tested for association (Mann- Whitney). These results wereconsistent with the correlation analyses. Women on schedule for their mostrecent mammogram (<2 years) were significantly more likely to describethemselves as similar to non-Indians in most ways (p < .05). Althoughthe numbers are not significant, these women also tended to follow theWhite/Anglo way of life (p < .07). In contrast, women who were overduefor their most recent mammogram (>2 years) were significantly more likelyto follow the Indian way of life, seek guidance from a traditional healer(both p < .05), and attend spiritual ceremonies (p < .01). Although the dataare not significant, these women also tended to follow traditions based onIndian culture (p < .06).

We also examined future intention for a mammogram in relation totraditionality. Because all of the women had a mammography history, thesample was divided into two groups: women who planned to have amammogram within the next 12 months (very likely and definitely responses)and women who did not plan to have a future mammogram or whoseintention was questionable (definitely not, perhaps, and not sure responses).The results of this analysis were consistent with the findings reported above.Planning to have a mammogram was higher among women who tendedto describe themselves as similar to non-Indians in most ways (p < .001)and who tended to follow the White/Anglo way of life (p < .05). Womenwho were more likely to attend spiritual ceremonies (p < .05), to follow theIndian way of life, to seek guidance from a traditional healer (both p < .01),and to participate in traditional health practices (p < .001) were less likelyto have a mammogram in the future.

CBE

Assessment of CBE participation relied on self-report data, as the VBCSS doesnot collect this information. Although current guidelines recommend a CBEevery year for women 40 years old and above (ACS, 2004), only 73% (84/115)of women reported a CBE within the past year, while 18% (20/115) had itwithin 1–2 years, and 5% (6/115) more than 2 years ago. Only 3% (3/115) ofwomen did not remember the time frame for their most recent CBE.

Dow

nloa

ded

by [

Tem

ple

Uni

vers

ity L

ibra

ries

] at

10:

09 1

9 N

ovem

ber

2014


With regards to CBE, two traditionality items were significantly corre-lated with delayed CBE. Women who grew up in a home where a triballanguage was spoken more often (rho = −0.22) and described themselvesas more traditional, were less likely to be up to date with their most recentCBE (rho = −0.24, both p < .05).

BSE PRACTICE

Assessment of BSE participation also relied on self-report data, as the VBCSSdoes not collect this information. The majority of women did not routinelypractice BSE. Twenty-four percent (27/115) of women reported practicingBSE monthly, while 49% (56/115) practiced it along a time frame fromevery 2–3 months to once a year. Of note is that while 16% (18/115) ofwomen did not practice BSE at all, 11% (13/115) practiced it weekly. Wealso examined traditionality in relationship to BSE practice and found nostatistically significant correlations.

PAP TEST

Assessment of Pap test participation also relied on self-report data, asthe VBCSS does not collect this information. Regular gynecologic cancerscreening was high among the sample population, with 97% (112/115) ofwomen reporting they had ever had a Pap smear, with the remaining 3%(3/115) either unable to recall it or never having had it. Based on currentscreening guidelines recommending a Pap smear every 3 years (AmericanCollege of Obstetrics & Gynecology [ACOG], 2004), 89% (102/115) of womenfollowed this recommendation. In terms of traditionality, no statisticallysignificant correlations were found between the traditionality items andcervical screening (most recent Pap smear).

Traditionality and Identity

Descriptive statistics for the 13 traditionality items suggest that there wasa range of identity among the sample, from a small percent of womenidentifying as most traditional (mean = 14%), to slightly more than a thirdidentifying as moderately traditional (mean = 33%), to slightly more thanhalf (mean = 53%) identifying as least traditional or similar to non-Indiansin most ways. Internal consistency reliability of the items was very strong(alpha = 0.90).

Because the traditionality items were organized into three separatesets of questions, we also examined if this range existed when each setof questions was analyzed separately. The eight questions adapted fromthe bicultural ethnic identity scale (Moran et al., 1999) also suggested a

Dow

nloa

ded

by [

Tem

ple

Uni

vers

ity L

ibra

ries

] at

10:

09 1

9 N

ovem

ber

2014


range of identity within the sample, with a very similar number of womenidentifying as most traditional (mean n = 15; 13%), slightly more than a thirdas moderately traditional (mean n = 43; 38%), and half as least traditional(mean n = 56; 50%). Although internal consistency reliability of these itemswas adequate (alpha = 0.79), it was less than the reliability reported byMoran and colleagues (alpha = 0.91).

A similar range was evident when responses to the set of culturalhealth practice questions were examined (Solomon & Gottlieb, 1999). Asmall number of women identified as most traditional (mean n = 16; 14%),one third identified as moderately traditional (mean n = 39; 34%), andslightly more than half as least traditional (mean n = 60; 52%). Internalconsistency reliability of these traditionality items was strong (alpha = 0.86),and consistent with the reliability measure of 0.81 reported by Solomonand Gottlieb (1999), who developed these items. When the single “bestdescription” question (Solomon & Gottlieb) was analyzed, the range shiftedslightly, with more women, 56% (n = 61/109) identifying as least traditionalcompared with 28% (n = 31/109) as moderate, and 16% (n = 17/109) asmost traditional.

DISCUSSION

Cancer Screening Participation

Our findings indicate that cancer screening participation rates among thissample of AI women were high. The mammography participation rate of82.6% surpasses the Healthy People 2010 goal of 70% of AI/AN womenreceiving a mammogram within the past 2 years (DHHS, 2000), and issignificantly higher than screening rates cited by the ACS (2004), where only52.2% of AI/AN women 40 and above had a mammogram within the past 2years. Participation rates among the study sample also met Healthy People2010 goals for cervical cancer screening, with 97% of AI women reportingever having a Pap test (goal = 97%) and 91% reporting one within the past3 years (goal = 90%). These cervical cancer-screening rates are also abovethose currently reported for AI/AN women, where 78.4% of women 18 andabove received a Pap test within 3 years (ACS, 2004).

Although the sample was predominately rural (83%), which has beenidentified as a screening barrier in previous research with AI/AN women(Foxall et al., 2001; Lantz et al., 2003), the women’s high level of education,insurance coverage, and income provided access to screening and healthcare services often not available to poor and uninsured populations ofAI/AN women. These high screening rates also may be related to the factthat because AI tribes in Vermont are not federally recognized, women arenot limited to IHS sites for screening services. The lack of mammographyequipment and screening programs within the IHS has been identified as

Dow

nloa

ded

by [

Tem

ple

Uni

vers

ity L

ibra

ries

] at

10:

09 1

9 N

ovem

ber

2014


a barrier to screening for AI/AN women living on-reservation and in urbanareas served by IHS (DHHS, 2003; Giuliano et al., 1998; Lantz et al., 2003;Risendal et al., 1999).

Although 26% of the sample was not meeting ACS guidelines forannual CBE, 92% of the sample reported a CBE within 2 years. This findingsuggests that rather than annually, CBE screening coincides with other cancerscreening services (mammogram and Pap test), as most current guidelinesrecommend mammography every 1–2 years (NCI; United States PreventiveServices Task Force [USPSTF]) and Pap testing every 2–3 years (ACOG; ACS).The finding that women were significantly more likely to be on schedule fortheir most recent mammogram if they also had a recent Pap test and CBEis consistent with previously reported research with AI women living inthe southwestern United States (Giuliano et al., 1998; Risendal et al., 1999)and supports the value of a coordinated approach toward cancer screeningbehavior. A coordinated approach regarding cancer screening procedures isrelevant for all countries in which the technology exists. This approach maybe especially valuable in rural and isolated regions, where distance, time,and transportation often limit access to screening services.

In contrast to other cancer screening behaviors, monthly BSE practicewas low among the sample population (24%). Although this findingcontradicts previous research, where higher rates of BSE were reportedamong AI women in Nebraska (Foxall et al., 2001) and Alaska Native women(Lanier & Kelly, 1999), it is consistent with findings from research with Nativewomen in Ontario, Canada, where Steven and colleagues (2004) reportedthat the Ojibwa and Oji-Cree women in their study felt uncomfortableperforming BSE. These researchers described the Native women as having“a more intense sense of privacy than other women” in the study, whichcontributed to their discomfort with BSE as well as CBE and cervicalcancer screening procedures (Steven et al., 2004, p. 308). Assessment ofthe relationship between privacy perceptions and participation in regularscreening may provide insight for developing new strategies to reach AIwomen currently not screened at regular intervals, including the 16% ofwomen in this study who never practiced BSE and the 17.4% who wereoverdue for a mammogram (> 2 years).

Although this study did not examine privacy and modesty specifically,it is an area that has been examined among diverse groups of womenin relation to cancer screening participation. For example, beliefs aboutprivacy influenced breast screening behavior among Southeast Asian womenin Canada (Bottorff et al., 1998) and older Chinese American women(Tang, Solomon, & McCracken, 2000), while young immigrant Muslimwomen reported concerns with physician recommendations for pelvic andPap smear exams and, subsequently, often avoided routine gynecologiccare (Matin & LeBaron, 2004). Assessment of privacy perceptions prior toproviding screening services may increase cancer screening participation

Dow

nloa

ded

by [

Tem

ple

Uni

vers

ity L

ibra

ries

] at

10:

09 1

9 N

ovem

ber

2014


among immigrant and religious women. This assessment may alert providersto women’s privacy concerns and provide an opportunity to discuss howthese concerns influence screening decisions.

With regards to cancer screening services, traditionality was associatedwith participation in CBE and mammography. Women who tended to de-scribe themselves as least traditional, and identified more with White/Angloculture, were significantly more likely to be on schedule for CBE andmammography. Conversely, women who tended to describe themselves asmost traditional, and identified more with AI culture, were significantly morelikely to be off schedule or overdue for CBE and mammography. Degreeof traditionality may account for the high rates of mammography screeningamong the sample population compared with previously reported partici-pation rates among AI populations. Since the majority of women identifiedmore with White/Anglo culture, they also may identify more with media andeducational messages promoting mammography screening as well as withoutreach efforts that target predominately White, non-Hispanic women.

Of concern was the association between traditionality and cancerscreening participation among women who most strongly identified withtheir Native culture. The results of this study were the opposite of previouslyreported research, in which cultural factors were positively associated withcancer screening participation. For example, Giuliano and colleagues (1998)reported that, although not statistically significant, attendance at culturalceremonies and use of the Hopi language were associated with increasedparticipation in biennial mammography and annual CBE. Risendal andcolleagues (1999) reported that AI urban women who attended ceremonies(social as well as tribal specific) were significantly more likely to be onschedule for CBE. Coe and colleagues (2004) also reported a positiveassociation between traditionalism and participation in healthy behaviors,with higher levels of traditionality associated with decreased smoking rates,alcohol use, and obesity.

However, Solomon and Gottlieb (1999), however, reported a significantcorrelation between higher levels of traditionality and lower cervicalscreening participation. Considering that some of the traditionality itemsused in this study were adapted from those developed by Solomon andGottlieb, it is possible that how traditionality was defined in these two studiesdiffered from the studies reporting a positive association. For example,Coe and colleagues (2004) interpreted traditionality as “a state of being,not a process of becoming” (p. 407), which contradicts the experience ofAI women in Vermont (Canales, 2004a) and elsewhere (Barrios & Egan,2002; Napholz, 2000), where reclaiming one’s Native identity was clearlya process. Traditionality is a complex concept, with varying interpretationsthat are difficult to measure quantitatively. Additional validation studies aswell as qualitative analyses are necessary to further elucidate the concept oftraditionality and its relationship to cancer screening.

Dow

nloa

ded

by [

Tem

ple

Uni

vers

ity L

ibra

ries

] at

10:

09 1

9 N

ovem

ber

2014


In previous research, seeking guidance from a traditional healer hasbeen interpreted as both a positive and inhibiting factor in relation to healthybehaviors. Risendal and colleagues (1999) reported that seeking guidancefrom a traditional healer was negatively associated with CBE participation,while Coe and colleagues (2004) interpreted this practice as a “diseaseprotecting behavior” (p. 403). The results of this study suggested that seekingguidance from a traditional healer interfered with regular screening andinfluenced future mammography intention. These conflicting results suggestthat it is important for providers to initially assess if a woman is seekingguidance from a traditional healer and then determine the influence of thehealer on past cancer screening behavior as well as on future intention.

The differences in the relationship between traditionality and cancerscreening participation among these studies are most likely influenced by thediverse samples of AI women represented: women residing on-reservation(Coe et al., 2004; Giuliano et al., 1998) and off-reservation (Risendal et al.,1999; Solomon & Gottlieb, 1999; current study); women in rural (Coe etal., 2004; Giuliano et al., 1998; current study) and urban areas (Risendalet al., 1999); women from the Southwest (Coe et al., 2004; Giuliano et al.,1998; Risendal et al., 1999) and Northeast (current study); and women fromfederally recognized tribes (Coe et al., 2004; Giuliano et al., 1998; Solomon& Gottlieb, 1999) and those who are not (current study). This diversity onceagain suggests that in order to develop and implement culturally sensitive andappropriate cancer-screening services, locally specific cultural information isessential (Weaver, 1997). As Weaver and Brave Heart (1999) point out, “Whileculturally based interventions may be helpful for some clients, they may notbe relevant for others” (p. 29).

Inclusion of locally specific cultural information in the assessment ofcancer screening messages and approaches is also important internationally,where many countries are becoming increasingly diverse. Providers cannotmake assumptions about how, or if, culture influences screening participa-tion. Assessment is key for determining the cultural influences on screeningperceptions and participation.

Finally, the relationship between traditionality and future intentiondeserves mention. Solomon and Gottlieb (1999) examined the relationshipbetween traditional behavior and intention to get a Pap test. Although“traditional behavior” was significantly correlated to “Pap test compliance”in their sample of AI women from northeast Oklahoma, there was nostatistical relationship between “traditional behavior” and intention to geta Pap test (Solomon & Gottlieb, p. 502). Previous studies examining therelationship between AI traditionality and future mammography behaviorwere not located; however, Rakowski and colleagues have demonstratedthat intentionality has a significant influence on whether women willcontinue to participate in routine cancer screening (Spencer, Pagell, &Adams, 2005). Rakowski’s (1993, 1996) research indicates that improvement

Dow

nloa

ded

by [

Tem

ple

Uni

vers

ity L

ibra

ries

] at

10:

09 1

9 N

ovem

ber

2014


in regular screening behavior requires questioning women about theirfuture screening plans as well as assessing past behavior. The significanceof the relationship between traditionality and intention to participate inmammography screening among this sample of AI woman suggests thattraditionality, as a possible indicator of future intention, is also important toconsider in future research efforts as well as within the clinical setting.

Internationally, future intention and traditionality deserve more attentionfrom researchers. This is an area where potential barriers to screeningcould be identified and interventions developed to enhance screeningparticipation.

Traditionality

Similar to previously reported research (Buehler, 1992; Canales, 2004b;Coe et al., 2004; Moran et al., 1999; Solomon & Gottlieb, 1999; Stricklandet al., 1996), there was a range of traditionality among the sample of AIwomen in the study, from a small group identifying themselves as mosttraditional (16%; n = 17/109), to slightly less than a third identifying asmoderately traditional (28%; 31/109), with slightly more than half (56%;61/109) describing themselves as similar to non-Indians in most ways.This range persisted when responses to traditionality items were analyzedtogether and in distinct groupings.

Solomon and Gottlieb (1999) reported similar results in their studywith AI women in northeast Oklahoma. Although 60% of women in theirconvenience sample were categorized as nontraditional based on responsesto a traditionality self-identification scale, this number decreased to 28%when only blood quantum was used as the indicator for traditionality. Coeand colleagues (2004) also reported a range of traditionality among theirrandom sample of women living on the Hopi reservation, although thedegree of traditionality was reversed, with approximately 36.3% categorizedas high traditionalism, 23.7% as medium, and 40% as low traditionalism. Thisreversal in traditionality (Coe et al.) may reflect differences between womenliving on a reservation compared with those who are more dispersed.

The complexity inherent within racial and ethnic populations oftenis obscured in health behavior research. “Although certain cultural char-acteristics may cluster within a given racial or ethnic group, it is at leastequally likely that substantial differences exist between individuals andsubgroups within these populations” (Kreuter, Lukwago, Bucholtz, Clark,& Sanders-Thompson, 2003, p. 134). As Weaver (1997) noted, althoughthe hundreds of AI groups in North America share some common culturalfeatures, they are also quite distinct. The range of traditionality foundamong the AI sample reflects the complexity inherent within the broadcategory “American Indian,” and suggests that identity variations need to

Dow

nloa

ded

by [

Tem

ple

Uni

vers

ity L

ibra

ries

] at

10:

09 1

9 N

ovem

ber

2014


be considered when conducting research and providing health care servicesto AI populations. As Weaver and Brave Heart (1999) point out, “Culturalidentity does not necessarily correlate precisely with blood quantum orcitizenship in a Native nation” (p. 25).

Within the international arena, every country has its own themesof traditionality within the racial and ethnic groups that comprise thepopulation. Traditionality is, therefore, a concept with multiple meaningsand constructions. These meanings and constructions need to be examined,defined, and measured within each specific cultural context for usefulnessin behavioral research, in general, and cancer screening in particular.

It is important to note that women who described themselves as moretraditional in the study tended to be younger and more educated. Thisis in contrast to previous research, in which increasing traditionalism wasassociated with increasing age and decreasing education (Coe et al., 2004).This trend in age and education suggests that some AI women in the studyare participating in an AI “renaissance,” a return to Native customs andtraditions, which has been occurring particularly among AI who have littleconnection to tribal organizations (Olsen, 1993, p. 4), which is the contextfor many AIs in Vermont.

Income was also a significant variable in terms of traditionality, withpoorer women more likely to seek guidance from a traditional healer. Thisfinding raises the question, Are poorer women seeking guidance from atraditional healer for cultural reasons, financial reasons, or a combinationof these reasons? Results from the qualitative study with AI women inVermont suggested that the ability to finance health care was a keyfactor influencing women’s health care decisions. Some AI women reliedon traditional health practices when finances were limited or unavailable(Canales, 2004a). Buchwald and colleagues (2000) reported that among theirsample of 869 urban, predominately low-income AI/AN, 70% (579) usedtraditional practices for health-related reasons. Although the monthly meanincome for users of traditional health practices in their study was only $908,cultural affiliation—living the Native way of life—was significantly associatedwith use of traditional health practices, rather than income (Buchwald et al.).

It is also important to note that accessing services from traditional healersrequires financial resources; these services are not ordinarily free and can bequite costly, depending on the services required (Kim & Kwok, 1998). Forexample, cost was cited as the main barrier to using Native healers amonga sample of Navajo IHS patients, where the cost of visiting a Native healerwas reported to range from $1 to $3000, with an average cost per visit of$388. The researchers noted that these costs were a “conservative estimate,”as they did not include associated expenses such as transportation, foodfor participants, or required materials (buckskin or herbs; Kim & Kwok, p.2245). The findings from these studies suggest that the relationship betweenincome and traditionality deserves further investigation.

Dow

nloa

ded

by [

Tem

ple

Uni

vers

ity L

ibra

ries

] at

10:

09 1

9 N

ovem

ber

2014


Limitations

Although the high breast and cervical cancer screening participation ratesamong the study participants is very encouraging, the findings are limitedto an AI population that is educated, middle class, and insured. Because arandomization process was used to select the sample from the VBCSS, it ispossible that AI women residing in Vermont have better access to cancerscreening services compared with AI women in other parts of the country,where access to cancer screening has been hindered by cost and lack ofinsurance (DHHS, 2003; Risendal et al., 1999).

Another limitation of the study is the small sample size. Although therewere no statistical differences between responders and nonresponders interms of age, mammography history, or geographic location, the overallresponse rate of 43% (n = 115) is not optimal. This may reflect thereluctance of women from racially and ethnically diverse populations toparticipate in traditional survey research (Coughlin, 1998; Napholz, 1998;Weaver, 1997), where there is no personal connection with the researchernor a connection with or endorsement from a member of the woman’s tribalcommunity. Although Native communities in Vermont are dispersed andoften fractious, connection to one’s community can influence decisions ofwhether to participate in research (Canales, 2004b).

Along similar lines, the small sample size may be related to use of theVBCSS database for recruitment. Although women sign a consent allowingtheir names to be accessed for future research, this approach may be lessacceptable to AI populations, who historically are more suspicious of surveyresearch (Harala, Smith, Hassel, & Gailfus, 2005; Weaver, 1997). This isparticularly relevant in Vermont, where the impact of the 1930s eugenicsmovement is still felt among some AI populations in the state (Gallagher,1999; Wiseman, 2001). In addition, when a woman received the researchpacket in the mail, she may have had concerns with how the data would beused or how her name was accessed. Although included in the cover letterwere an explanation of the study and recruitment processes as well as theresearcher’s contact information, this explanation may have been insufficientto overcome an individual woman’s reluctance to participate in the study.

Precautions were taken to include only women who consistently self-identified as AI; however, this still proved to be an issue that subsequentlylimited the sample size. For example, there were women who returnedblank surveys with notes stating that they were not “Indian enough” tobe in a research study and subsequently declined to participate. The firstauthor encountered a similar response during numerous follow-up callsrequesting women to complete the survey. Women expressed concern thattheir responses would “skew” the data since they were not practicing as“Indians,” were not aware of their family heritage except that there was“Indian blood” among their relatives, or that although they were “proud” oftheir “little bit of Indian blood,” it was not sufficient for this type of research.

Dow

nloa

ded

by [

Tem

ple

Uni

vers

ity L

ibra

ries

] at

10:

09 1

9 N

ovem

ber

2014


Some women were upset that their names were included in the recruitmentlist and questioned the first author as to how their names were accessed.When they were reminded of the VBCSS and their self-identification as AI,women were often surprised they had checked the box or admitted they hadforgotten.

These responses from women who declined to participate in the studyreinforce the fluid nature of racial categorization and suggest that womenperceived a distinction between racial and cultural identity. For example,there were women who did not consistently categorize their race as AI onthe VBCSS questionnaire and, consequently, were excluded from the studyinitially. There were also women who were invited to participate, based ontheir consistent categorization of race as AI, yet did not identify themselvesas such culturally, and subsequently declined to participate in the study.Researchers who rely on racial classification for identifying study participantsneed to recognize that for some individuals there is a distinction betweenracial and cultural identification, and how one identifies may change overtime.

Decisions of whether to use existing databases as recruitment sourcesfor research with AI women need to weigh the advantages, such as access tothe sample population, actual mammography dates, and ability to randomize,with the disadvantages of survey methodology, particularly the lack ofpersonal contact with the researcher, and reliance on standardized racialcategories. A possible compromise, if resources are available, is to conductindividual interviews, either in person or by telephone, and financiallycompensate women who participate.

A final limitation of using the VBCSS is that women who never hada mammogram were not included in the sample pool. Considering thatthe purpose of the VBCSS is to track all mammograms performed inthe state, reliance on the database for recruitment limited the sample towomen with a mammography history. An important area for future researchis the examination of the relationship between traditionality and cancerscreening beliefs and perceptions among AI women who have never had amammogram. This is a group of women for whom conventional outreachefforts have thus far been unsuccessful and who may benefit from effortsthat consider traditionality in their approach.

CONCLUSION

The results of this study suggest that further examination of the relationshipbetween traditionality and cancer screening behavior among AI womenis warranted. The significant associations between traditionality and pastmammography and CBE behavior, as well as future mammography intention,suggest that among women who identify more with their Native culture,

Dow

nloa

ded

by [

Tem

ple

Uni

vers

ity L

ibra

ries

] at

10:

09 1

9 N

ovem

ber

2014


traditionality may inhibit regular screening and increase the risk of lapsingfrom screening. This may be especially relevant considering that traditionalitywas stronger among younger, more educated AI women. Younger women,as a group, are at increased risk for invasive cervical cancer, while their riskfor breast cancer increases with age.

Insofar as every country exists within a context different from theUnited States, and every country has its own constructions of traditionality,these results also have international implications. Since traditionality is acontext-specific concept with multiple meanings, cancer researchers acrossthe globe are encouraged to define, measure, and investigate traditionalitywithin their specific cultural context, as well as consider it when designingand testing interventions for improving screening participation. Cliniciansare also encouraged to asses women’s level of traditionality and its possibleinfluence on screening participation and future intention. This is especiallyrelevant for primary care providers who care for the majority of poor andunderserved women worldwide.

Although the small sample size limits generalizability of the findings, theresults suggest that traditionality scales can be used to identify differencesin screening behavior in this racial/ethnic group. Further testing of thetraditionality scales with other groups of AI women is important forconfirming these results and identifying interventions to reach AI womenwho may not be participating in regular screening or who are at increasedrisk for lapsing from screening. Reaching these women is essential if theprogress made toward meeting Healthy People 2010 goals is to continueand breast and cervical cancer mortality among AI women is to bereduced.

REFERENCES

American College of Obstetricians and Gynecologists. (ACOG) (2004). Revised cervi-cal cancer screening guidelines require reeducation of women and physicians.Retrieved 5 November 2004, from www.acog.org/from home/publications/press releases

Barrios, P. G., & Egan, M. (2002). Living in a bicultural world and finding the wayhome: Native women’s stories. Affilia, 17(2), 206–228.

Bottorff, J. L., Johnson, J. L., Bhagat, R., Grewal, S., Balneaves, L. G., Clarke, H., &Hilton, B. A. (1998). Beliefs related to breast health practices: The perceptionsof South Asian women living in Canada. Social Science & Medicine, 47(12),2075–2085.

Brant, J. M., Fallsdown, D., & Iverson, M. (1999). The evolution of a breast healthprogram for Plains Indian women. Oncology Nursing Forum, 26(4), 731–739.

Buchwald, D., Beals, J., & Manson, S. M. (2000). Use of traditional health practicesamong Native Americans in a primary care setting. Medical Care, 38(12), 1191–1199.

Dow

nloa

ded

by [

Tem

ple

Uni

vers

ity L

ibra

ries

] at

10:

09 1

9 N

ovem

ber

2014


Buehler, J. A. (1992). Traditional Crow Indian health beliefs and practices. Journalof Holistic Healing, 10(1), 18–33.

Canales, M. K. (2004a). Connecting to Nativeness: The influence of AmericanIndian identity on women’s health care decisions. Canadian Journal of NursingResearch, 36(4), 18–44.

Canales, M. K. (2004b). Taking care of self: Health care decisions of American Indianwomen. Health Care for Women International, 25(5), 411–435.

Canales, M. K., & Geller, B. M. (2004). Moving in between mammography: Screeningdecisions of American Indian women in Vermont. Qualitative Health Research,14(6), 836–857.

Clarke, H. F., Rhea, J., Deschamps, M., Hislop, G., Band, P. R., & Atleo, R. (1998).Reducing cervical cancer among First Nations women. The Canadian Nurse,March, 36–41.

Coe, K., Attakai, A., Giuliano, A., Papenfuss, M., Martin, L., & Nuvayestewa, L. (2004).Traditionalism and its relationship to disease risk and protective behaviors ofwomen living on the Hopi reservation. Health Care for Women International,25(5), 391–410.

Coughlin, S. S. (1998). Implementing breast and cervical cancer preventionprograms among the Houma Indians of southern Louisiana: Cultural and ethicalconsiderations. Journal of Health Care for the Poor and Underserved, 9(1), 30–41.

Department of Health and Human Services (DHHS). (2003). Facing cancer inIndian country: The Yakama Nation and Pacific Northwest tribes. President’sCancer Panel, 2002 Annual Report. Bethesda, MD: National Institutes ofHealth/National Cancer Institute.

Department of Health & Human Services (DHHS). (2005). Healthy People 2010.Retrieved 21 February 2005, from http://www.healthypeople.gov/

Foxall, M. J., Barron, C. R., & Houfek, J. F. (2001). Ethnic influences on bodyawareness, trait anxiety, perceived risk, and breast and gynecologic cancerscreening practices. Oncology Nursing Forum, 28(4), 727–738.

Gallagher, N. (1999). Breeding better Vermonters: The eugenics project in the greenmountain state. Hanover, NH: University Press of New England.

Garroutte, E. M., Kunovich, R. M., Jacobsen, C., & Goldberg, J. (2004). Patientsatisfaction and ethnic identity among American Indian older adults. SocialScience & Medicine, 59, 2233–2244.

Geller, B. M., Worden, J. K., Ashley, J. A., Oppenheimer, R. G., & Weaver, D. L.(1996). Multipurpose statewide breast cancer surveillance system: The Vermontexperience. Journal of Registry Management, 23(4), 168–174.

Giroux, J., Welty, T. K., Oliver, F. K., Kaur, J. S., Leonardson, G., & Cobb, N. (2000).Low national breast and cervical cancer-screening rates in American Indian andAlaska Native women with diabetes. Journal of the American Board of FamilyPractice, 13, 239–245.

Giuliano, A., Papenfuss, M., de Zapien, J. D., Tilousi, S., & Nuvayestewa, L. (1998).Breast cancer screening among Southwest American Indian women living on-reservation. Preventive Medicine, 27, 135–143.

Hampton, J. W., & Friedell, G. H. (2001). Cancer surveillance: A problem for NativeAmericans and Appalachian populations. Cancer Supplement, 91(1), 242–246.

Dow

nloa

ded

by [

Tem

ple

Uni

vers

ity L

ibra

ries

] at

10:

09 1

9 N

ovem

ber

2014


Harala, K., Smith, C., Hassel, C., & Gailfus, P. (2005). New moccasins: Articulatingresearch approaches through interviews with faculty and staff at Native andnon-Native academic institutions. Journal of Nutrition Education Behavior, 37,67–76.

Kim, C., & Kwok, Y. S. (1998). Navajo use of native healers. Archives of InternalMedicine, 158, 2245–2249.

Kottke, T. E., & Trapp, M. A. (1998). Implementing nurse-based systems to provideAmerican Indian women with breast and cervical cancer screening. MayoClinical Proceedings, 73(9), 815–823.

Kreuter, M. W., Lukwago, S. N., Bucholtz, D. C., Clark, E. M., & Sanders-Thompson,V. (2003). Achieving cultural appropriateness in health promotion programs:Targeted and tailored approaches. Health Education & Behavior, 30(2), 133–146.

Lanier, A. P., & Kelly, J. J. (1999). Knowledge, attitudes, and behavior of AlaskaNative women regarding cervical and breast cancer. In D. Weiner (Ed.),Preventing and controlling cancer in North America: A cross-cultural perspective(pp. 71–83). Westport, CT: Praeger.

Lanier, A. P., Kelly, J., & Berner, J. (1999). The Alaska Native women’s health projectto reduce cervical cancer. In C. Sanchez. Glover & F. S. Hodge (Eds.), Nativeoutreach: A report to American Indian, Alaska Native, and Native Hawaiiancommunities (pp. 23–32). Bethesda, MD: National Institutes of Health/NationalCancer Institute.

Lantz, P. M., Orians, C. E., Liebow, E., Joe, J. R., Burhansstipanov, L., Erb, J.,& Kenyon, K. (2003). Implementing women’s cancer screening programs inAmerican Indian and Alaska Native populations. Health Care for WomenInternational, 24, 674–696.

Locust, C. S. (1986). American Indian beliefs concerning health and unwellness(Native American Research and Training Center, Monograph Series). Tucson:University of Arizona.

Matin, M., & LeBaron, S. (2004). Attitudes toward cervical cancer screening amongMuslim women: A pilot study. Women & Health, 39(3), 63–77.

Messer, L., Steckler, A., & Dignan, M. (1999). Early detection of cervical canceramong Native American women: A qualitative supplement to a quantitativestudy. Health Education & Behavior, 26(4), 547–562.

Michielutte, R., Dignan, M. B., Sharp, P. C., Blinson, K., & Wells, B. (1999).Psychosocial factors related to cervical screening among Lumbee women.American Journal of Health Behavior, 23(2), 115–127.

Moran, J. R., Fleming, C. M., Somervell, P., & Manson, S. M. (1999). Measuringbicultural ethnic identity among American Indian adolescents: A factor analyticstudy. Journal of Adolescent Research, 14(4), 405–426.

Napholz, L. (1998). Enhancing research participation and retention for threeethnically diverse groups. Journal of Cultural Diversity, 5(4), 117–119.

Napholz, L. (2000). Balancing multiple roles among a group of urban midlifeAmerican Indian working women. Health Care for Women International, 21,255–266.

Novins, D. K., Beals, J., Moore, L. A., Spicer, P., Manson, S., & AI-SUPERPFPTeam. (2004). Use of biomedical services and traditional healing options among

Dow

nloa

ded

by [

Tem

ple

Uni

vers

ity L

ibra

ries

] at

10:

09 1

9 N

ovem

ber

2014


American Indians: Sociodemographic correlates, spirituality, and ethnic identity.Medical Care, 42(7), 670–679.

Olsen, S. J. (1993). Cancer prevention and early detection in Native American andAlaska Native populations. In M. Frank-Stromborg & S. J. Olsen (Eds.), Cancerprevention in minority populations (pp. 1–77). St. Louis, MO: Mosby.

Rakowski, W., Dube, C. A., & Goldstein, M. G. (1996). Considerations for extendingthe transtheoretical model of behavior change to screening mammography.Health Education Research, Theory, and Practice, 11, 77–96.

Rakowski, W., Ehrich, B., Dube, C. E., Pearlman, D. N., Goldstein, M. G., &Peterson, K. K. et al. (1996). Screening mammography and constructs fromthe transtheoretical model: Associations using two definitions of the stages-of-adoption. Ann Behavioral Med, 18(2), 91–100.

Rakowski, W., Fulton, J. P., & Feldman, J. P. (1993). Women’s decision making aboutmammography: A replication of the relationship between stages of adoption anddecisional balance. Health Psychology, 12(3), 209–214.

Risendal, B., Roe, D., DeZapien, B. A., Papenfuss, M. S., & Giuliano, A. (1999).Influence of health care, cost, and culture on breast cancer screening: Issuesfacing urban American Indian women. Preventive Medicine, 29, 501–509.

Roubideaux, Y. (2002). Perspectives on American Indian health. American Journalof Public Health, 92(9), 1401–1403.

Smith, A., Christopher, S., & McCormick, A. (2004). Development and implemen-tation of a culturally sensitive cervical health survey: A community-basedparticipatory approach. Women & Health, 40(2), 67–86.

Solomon, T. G. A., & Gottlieb, N. H. (1999). Measures of American Indiantraditionality and its relationship to cervical cancer screening. Health Care forWomen International, 20, 493–504.

Spencer, L., Pagell, F., & Adams, T. (2005). Applying the transtheoretical model tocancer screening behavior. American Journal of Health Behavior, 29(1), 36–56.

Steven, D., Fitch, M., Dhaliwal, H., Kirk-Gardner, R., Sevean, P., Jamieson, J., &Woodbeck, H. (2004). Knowledge, attitudes, beliefs, and practices regardingbreast and cervical cancer screening in selected ethnocultural groups innorthwestern Ontario. Oncology Nursing Forum, 31(2), 305–311.

Strickland, C. J., Chrisman, N. J., Yallup, M., Powell, K., & Squeoch, M. D. (1996).Walking the journey of womanhood: Yakima Indian women and Papanicolaou(Pap) test screening. Public Health Nursing, 13(2), 141–150.

Swan, J., & Edwards, B. K. (2003). Cancer rates among American Indians and Alaskanatives. Cancer, 98(6), 1262–1272.

Tang, T. S., Solomon, L. J., & McCracken, L. M. (2000). Cultural barriers tomammography, clinical breast exam, and breast self-exam among Chinese-American women 60 and older. Preventive Medicine, 31, 575–583.

United States Bureau of the Census. (2001). State and county quickfacts. Retrieved22 August 2004, from http://quickfacts.census.gov/qfd/states/5000htlm

United States Preventive Services Task Force (USPSTF). (2002). Screening forbreast cancer: Summary of recommendations. Retrieved 5 May 2005, fromhttp://www.ahrq.gov/clinic/uspstf/uspsbrca.htm

Walters, K. L. (1999). Urban American Indian identity attitudes and acculturationstyles. Journal of Human Behavior in the Social Environment, 2(1/2), 163–178.

Dow

nloa

ded

by [

Tem

ple

Uni

vers

ity L

ibra

ries

] at

10:

09 1

9 N

ovem

ber

2014


Weaver, H. N. (1997). The challenges of research in Native American communities:Incorporating principles of cultural competence. Journal of Social ServiceResearch, 23(2), 1–15.

Weaver, H. N., & Brave Heart, M. (1999). Examining two facets of American Indianidentity: Exposure to other cultures and the influence of historical trauma.Journal of Human Behavior in the Social Environment, 2(1/2), 19–33.

Weiner, D. (1999). American Indian cancer discourse and the prevention of illness.In D. Weiner (Ed.), Preventing and controlling cancer in North America: Cross-cultural perspectives (pp. 55–68). Westport, CT: Praeger.

Wiseman, F. M. (2001). The voice of the dawn: An autohistory of the Abenaki nation.Hanover, NH: University Press of New England.

Dow

nloa

ded

by [

Tem

ple

Uni

vers

ity L

ibra

ries

] at

10:

09 1

9 N

ovem

ber

2014

Documents

Traditionality and Cancer Screening Practices Among American Indian Women in Vermont