Abstract
Background:
U.S. women's awareness and knowledge of gynecologic cancer have not been well studied, with the exception of cervical cancer screening and risk factors.
Methods:
Fifteen focus groups were conducted with women aged 40–60 years in Miami, New York City, Chicago, and Los Angeles.
Results:
Most participants said they had heard of cervical, ovarian, and uterine cancers but were unfamiliar with vaginal and vulvar cancers. The misconception that the Pap test screens for multiple gynecologic cancers was prevalent and engendered a false sense of security in some women. An annual Pap screening interval was most familiar to participants; some mentioned a shorter screening interval for high-risk women; few mentioned an extended screening interval. A few participants thought the pelvic examination could detect a variety of conditions, including ovarian cancer. Some knew that the human papillomavirus (HPV) could cause cervical cancer, but no other risk factors for specific cancers were mentioned with any consistency. Although some recognized unexplained vaginal bleeding as a symptom of cervical cancer, participants generally were unfamiliar with gynecologic cancer symptoms. Participants reported learning about the discussion topics from a variety of sources, including the mass media.
Conclusions:
Participants lacked critical knowledge needed to understand their gynecologic cancer risk and seek appropriate care. Pap tests and routine examinations offer ideal opportunities to educate women about the purpose of the Pap test as well as risk factors and symptoms associated with various gynecologic cancers. The reported influence of the mass media also supports the viability of multimedia educational strategies.
Introduction
To date, studies exploring U.S. women's knowledge about gynecologic cancer have focused primarily on cervical cancer screening and risk factors. A key knowledge gap described in studies spanning two decades is confusion about the purpose of the Papanicolaou (Pap) test, 1 –8 which has been used to detect cervical cancer in the United States for more than 60 years 9 and remains the only screening test recommended for population-based detection of any form of gynecologic cancer. 10 Although first-hand experience with the Pap test is almost universal in the United States, with 98% of women having had one or more, 11 many women reported they did not know its purpose and believed that it screens for multiple gynecologic cancers and sexually transmitted diseases (STDs), as well as other conditions. 1 –8 Similarly, many women remain unaware of cervical cancer risk factors, 12 –19 but advertisements and news reports about the human papillomavirus (HPV) vaccine seem to have increased awareness that HPV and sexual promiscuity are linked to cervical cancer. 1,20,21 Studies exploring women's knowledge of other gynecologic cancer topics found low awareness of the symptoms of cervical cancer 22,23 and ovarian cancer 24 and the risk factors associated with ovarian cancer 24 and uterine cancer. 25
Increasing women's knowledge and awareness of gynecologic cancer became a Congressional mandate in 2007 with the passage of the Gynecologic Cancer Education and Awareness Act, 26 or Johanna's Law. This legislation authorized the Centers for Disease Control and Prevention (CDC), in collaboration with the Department of Health and Human Services' Office on Women's Health, to establish the Inside Knowledge: Get the Facts About Gynecologic Cancer campaign. Inside Knowledge is designed to raise awareness among women and healthcare providers about the signs, symptoms, risk factors, and prevention strategies associated with the five main types of gynecologic cancers—cervical, ovarian, uterine, vaginal, and vulvar cancers. The campaign encourages women to pay attention to their bodies and know what is normal for them so they may recognize any persistent changes and seek care in an appropriate and timely manner.
The study reported here was undertaken by the CDC to supplement the limited existing literature and inform the development of the Inside Knowledge campaign. This focus group study of women aged 40–60 years in four U.S. cities was intended to provide a current, in-depth assessment of general awareness and knowledge of screening, risk factors, and symptoms related to cervical, ovarian, uterine, vaginal, and vulvar cancers.
Materials and Methods
Fifteen focus groups were conducted in 2009 with 132 women in Miami (4 groups, 36 participants), New York City (4 groups, 35 participants), Chicago (4 groups, 34 participants), and Los Angeles (3 groups, 27 participants). The cities were selected based on the diversity of their geographic region and demographic composition.
Participants
Participants were limited to women between the ages of 40 and 60 years and were recruited from the participant databases of the professional focus group facilities where the groups were held. A minimum of 7 and a maximum of 9 participants were included in each group. Participants identified themselves as white (40.2%), African American (32.6%), Hispanic (18.9%), and Asian (8.3%) (Table 1). They included women with all levels of educational attainment, including 7.6% who had not completed high school. Similarly, participants represented a variety of annual household income levels from <$25,000 (15.2%) to ≥$100,000 (11.4%). Most were employed either full-time (56.1%) or part-time (22.0%), and 81.8% reported that they had health insurance for routine care. All participants said they had had at least one Pap test, and most (87.1%) reported that they had a Pap test within the last 3 years.
Protocol
Each focus group lasted 1.5–2 hours and was led by a professional moderator using a semistructured discussion guide. Table 2 presents the order in which the topics reported here were discussed, the information about each topic given to participants during the groups, and the suggested moderator script for each topic. The symptoms associated with gynecologic cancers discussed during the groups were those recognized by the Inside Knowledge campaign, based on expert panel recommendations and the existing scientific evidence. 27 Three or more investigators monitored each focus group from behind a one-way mirror. All focus groups were audiorecorded, and verbatim transcripts were prepared. Participants who reported as scheduled received a small honorarium consistent with the standard for federally sponsored studies. At the end of the groups, each participant was given patient education fact sheets about cervical, ovarian, uterine, vaginal, and vulvar cancers to take home.
The symptoms discussed were those recognized by the Centers for Disease Control and Prevention's Inside Knowledge campaign at the time of the study. During this exercise the moderator was instructed not to use the word “symptoms,” given its possible negative connotation. To mitigate any order effects, the order of the symptoms on the list was changed midway through the study after the last focus group in the second city (group 8 of the 15 groups in the study).
Before participating in the study, each participant signed a consent form informing her of the voluntary nature of her responses and the presence of an audiorecording. The study protocol was submitted to the Institutional Review Board for CDC's National Center for Chronic Diseases Prevention and Health Promotion and on June 20, 2008, was classified as a public health practice activity, thereby exempt from further review.
Codebook development and analysis
An inductive (data-driven) thematic approach was primarily used to develop the codes included in the analysis. In addition, several codes were suggested by the existing literature, investigators' interest, and Social Cognitive Theory, 28 the guiding theoretical framework of the Inside Knowledge campaign.
After reviewing the focus group notes and transcripts, the lead investigator drafted the initial codebook and a companion coding form. The codebook listed code names (e.g., ovarian cancer may be asymptomatic in early stages; Pap test screens for multiple conditions; family history is a risk factor), definitions, and coding instructions. The coding form provided a template for extracting and organizing coded text to facilitate intercoder agreement comparisons and data analysis. A participant utterance or unbroken unit of speech was the coding segmentation unit used in this study. Participant utterances were numbered in sequential order in each transcript. Moderator utterances were not numbered or coded but were included on the coding form if needed for context.
The initial codebook and coding form were reviewed and revised by eight investigators. Then, these documents were refined through an iterative process in which four investigators, two primary coders and two consultants, independently coded the same three transcripts, compared coding forms, clarified code definitions, and revised the codebook accordingly. Finally, the two primary coders independently coded the 12 remaining transcripts. After coding each transcript, they reconciled all coding discrepancies, the vast majority of which (94%) were the result of one coder missing a valid code detected by the other. In rare instances in which the primary coders could not resolve a coding dispute (<1% of coding discrepancies), one of the consultants was asked to cast the deciding vote. The primary coders continued to refine the codebook; however, no new codes were added, nor were other major revisions made to the codebook after the seventh transcript. The final codebook included 75 codes related to the topics reported here. The coding forms of the first six transcripts coded were updated to reflect the final codebook. Intercoder reliability among the two primary coders was evaluated using Cohen's kappa 29 for the 12 transcripts coded after the initial codebook development process and ranged from 0.84 (standard error [SE]=0.0062) to 0.95 (SE=0.0069), with a median of 0.91.
The data were analyzed by reviewing individual codes across all 15 groups. The relative prominence of responses was assessed by evaluating both the number of participants and the number of groups associated with a code. As focus group results are not generalizable and not every participant voices an opinion on every point discussed, the frequency of responses is described using modifiers such as many, some, and few, rather than actual counts or percentages. 30
Group cohesion
Cohesion refers to participants' comfort and engagement in group activities 31 and is an indicator of focus group data quality. 32 Five interaction scenarios were tracked as proxies of cohesion: all instances of (1) a participant sharing personal health experiences with the group and up to two instances per group of (2) a participant agreeing with another, (3) a participant disagreeing with another, (4) a participant interacting directly with another (without the moderator's involvement), and (5) a participant referring to another by name. In every focus group, the majority of participants shared personal health experiences and two or more agreed with the opinions of another, disagreed with the opinions of another, and interacted with each other directly without the moderator intervening. In 14 of the 15 groups, one or more participants referred to another by name.
Consistency of findings across cities
No differences in gynecologic cancer awareness or knowledge of risk factors and symptoms were noted among participants in Miami, New York City, Chicago, and Los Angeles. However, participants in Chicago seemed somewhat more knowledgeable about screening. For instance, the extended Pap test screening interval was mentioned only by women in Chicago. No explanation for this difference was apparent; Chicago participants did not differ significantly from participants in other cities in terms of their demographic characteristics.
Results
Gynecologic cancer awareness
Most participants reported that they had heard of cervical, ovarian, and uterine cancers (Table 3). Generally participants used the terms “cervical cancer,” “ovarian cancer,” and “uterine cancer” before they were mentioned by the moderator. Conversely, most participants had not heard of vaginal, vulvar, or gynecologic cancer. Not only were they unfamiliar with vulvar cancer, but also many reported they were unsure of the anatomical location of the vulva. A few participants confused the vulva with the clitoris.
Modifiers, such as most, many, some, few, and very few are used to describe the relative frequency of responses in focus groups rather than actual counts or percentages. 30 (Kruegar). The focus group method supports inventorying the range of responses related to the discussion topics and their relative prominence in qualitative terms, but it cannot provide firm numbers to characterize response volume.
The term “gynecologic cancer” was not familiar to the majority of participants; however they generally were able to infer its definition as cancer specific to women or cancer related to the female reproductive organs based on their familiarity with the words “gynecology” and “gynecologist.” When asked what cancers they would classify as gynecologic, many correctly said cervical, ovarian, and uterine. Participants in five groups said breast cancer was a gynecologic cancer, but in two of these groups, other participants disagreed with this assertion, one of them explaining: “Because breast cancer could be for men or women. These are only women. So I don't think breast cancer is in that.” In three groups, participants named gastrointestinal cancers, including stomach, colon, intestinal, and anal cancers, and no participants disputed that these were gynecologic cancers. One participant reasoned that the colo in “gynecology” was related to the word “colon.”
Participants attributed their awareness of various gynecologic cancers to several sources, such as health experiences of others, medical professionals, advertisements, and news and entertainment media. The source mentioned most often by participants was health experiences of others, including people they knew directly (family members, friends, and co-workers), those they heard about through interpersonal sources (e.g., mother-in-law of a friend), and celebrities whose health experiences were discussed in the mass media. The moderator did not directly ask groups if they knew of any celebrities who had been diagnosed with cancer, but participants mentioned many celebrities with a cancer link during the groups, including Christina Applegate (breast), Lance Armstrong (testicular), “Katie Couric's husband” (colon), Sheryl Crow (breast), Fran Drescher (uterine), Farrah Fawcett (anal), Jade Goody (cervical), and Gilda Radner (ovarian). However, participants did not always correctly identify the type of cancer associated with a celebrity.
Screening
The benefit of early cancer detection was deeply ingrained in many participants. For example, one woman said, “If you get it detected early and you get treated, pretty much you'll beat it.” All participants were familiar with the Pap test, and many commented that it could be lifesaving. However, there was a great deal of misinformation about its scope. Participants who correctly reported that the Pap test is used exclusively to screen for cervical cancer were in the minority. Most thought that the Pap test screened for multiple conditions, and a few said they were unsure for what the Pap test screens.
The strength of the belief that the Pap test screens exclusively for cervical cancer varied among those participants who expressed this view. A few participants were firm believers, whereas others were not as certain and hedged their belief with qualifiers: “I thought cervical. I'm not sure.” Also, a few were swayed by the dominant opinion; in one group, two participants who initially reported that the Pap test screened exclusively for cervical cancer later agreed with those who said it screened for uterine cancer as well.
Participants who thought that the Pap test screened for a variety of conditions described it as an “all-inclusive” or “catch-all” test. Most of these women said the Pap test screened for multiple gynecologic cancers, primarily cervical, ovarian, and uterine cancers. A few said it screened for STDs, fibroids, yeast infections, and menopause, in addition to various cancers. Others did not specify conditions for which the Pap test screens, giving broad answers, such as “everything,” “abnormalities, no matter what they are,” or “cancer, diseases, and stuff like that.” A few participants explained that the Pap test was able to screen for multiple cancers because of the close proximity of the female reproductive organs.
Among the focus group participants, there was widespread belief that the best defense against all or most forms of gynecologic cancer was regular Pap test screening. Some participants described the peace of mind that regular Pap screening afforded them: “I don't think about it [any form of gynecologic cancer] that much. But then again, I always get my Pap smear.” A few participants remarked that regular Pap testing was especially vital to detect ovarian cancer, which they said may not cause symptoms in the early stages. When the scope of the Pap test was divulged, many participants responded with concern; they frequently asked what screening tests were available for the other gynecologic cancers: “What about the other four [gynecologic cancers]? Where's the test for that?…Test me for the other four now.”
Participants generally were familiar with an annual Pap screening interval. Some also mentioned a shorter screening interval for women with prior abnormal Pap test results or other risk factors. Participants in two groups mentioned an extended Pap screening interval but were unaware of the specific characteristics of potential candidates, reporting it was an option for “older women” or “after a certain age.” A few participants in these groups reacted negatively to the idea of waiting 3 years between Pap tests, saying they believed this interval would not constitute regular screening.
As the Pap test is the only screening test recommended for population-based detection of any form of gynecologic cancer (cervical), no other screening modalities were included in the discussion guide. In the natural course of discussion, however, pelvic examination was mentioned in five groups, transvaginal ultrasound (TVU) was mentioned in six groups; and CA-125 was mentioned in three groups. A few participants reported that the pelvic examination was an important supplement to the Pap test and believed that it could detect a variety of abnormalities, including ovarian cancer, vulvar cancer, and fibroids. Also, a few women reported having been screened for cervical or ovarian cancer with a TVU or CA-125 blood test. A few others were familiar with these tests but had not had them.
Risk factors
Many participants knew that a sexually transmitted virus increases gynecologic cancer risk, although some struggled to remember that it is HPV. Some specified that HPV caused cervical cancer; others just said “cancer.” Participants generally knew that an HPV vaccine existed and that it is offered to young women. A few participants voiced concern that they were not candidates for the vaccine: “… I'm over 26, what the heck do I do? Initially when they first started talking about the vaccine for HPV, the first thing I did when I went to my next annual exam was ask the physician for that particular vaccine and was told I can't have it. I was too old.” Many participants mentioned learning about HPV through direct-to-consumer advertisements for the vaccine. Others said they learned about it as a result of their daughters or daughters of family members and friends being offered the HPV vaccine or being diagnosed with HPV.
Some participants also mentioned that family history increased gynecologic cancer risk. A few participants said that increasing age and being African American were associated with increased risk but did not specify for which types of cancer. Also, one participant mentioned that oral contraceptive use reduced the risk of ovarian cancer, and another said that giving birth reduced risk, but she did not say for which cancer. A few participants said a healthy diet reduced risk, but none specified for which type of cancer.
Symptoms
At the beginning of each group, participants were given a list of symptoms and asked what might cause them (Table 2). Although the symptoms all were related to gynecologic cancer, participants generally did not associate any of them with a particular gynecologic cancer, with the exception of unexplained vaginal bleeding, which some participants recognized as a possible symptom of cervical cancer. When participants were shown campaign messages, which included the statement, “Gynecologic cancers have warning signs,” one or more participants in most groups commented that they did not know the warning signs of gynecologic cancers. During discussion about the various forms of gynecologic cancer, some women commented that ovarian cancer may not cause symptoms in the early stages, and a few said that early-stage ovarian cancer never causes symptoms and that once symptoms emerge, “it's too late.”
A few participants said they learned about gynecologic cancer symptoms through friends diagnosed with these diseases. Several participants reported that they learned about the asymptomatic nature of ovarian cancer as a result of public awareness raised by Gilda Radner's death.
Campaign messages
During each focus groups, participants were asked to react to four core messages of the Inside Knowledge campaign: (1) pay attention to your body and know what is normal for you; gynecologic cancers have warning signs, (2) when gynecologic cancers are found early, treatment is most effective, (3) see a doctor right away if you notice any vaginal bleeding that is not normal for you or you have any other unexplained gynecologic cancer signs or symptoms that persist for 2 weeks or longer, and (4) get a Pap test regularly to screen for cervical cancer.
Participants generally reacted favorably to these messages, and their concerns generally reflected knowledge gaps. First, several questioned whether ovarian cancer caused symptoms in the early stages, saying that ovarian cancer was asymptomatic until death was imminent and unavoidable. Second, most participants said that 2 weeks was too long to wait to seek care for symptoms. However, participants generally were unfamiliar with the symptoms of gynecologic cancer. Also, the few participants who had experienced symptoms typically reported they did not seek care promptly unless they experienced severe pain. Many participants questioned why the statement “Get a Pap test regularly to screen for cervical cancer” only mentioned cervical cancer, saying they thought the Pap test screened for multiple cancers. Finally, some participants who were unaware of alternative Pap screening intervals specifically mentioned that “Get a Pap test regularly” should be changed to “Get a Pap test annually,” as they thought “annually” and “regularly” were synonymous.
Discussion
Many women in our study exhibited knowledge gaps about the purpose and frequency of the Pap test and the risk factors and symptoms associated with various gynecologic cancers. It is important for women to understand that the only gynecologic cancer for which the Pap test is recommended is cervical cancer. Similarly, as no population-based screening is recommended for the other gynecologic cancers, it is all the more important for women to be familiar with signs and symptoms associated with all types of gynecologic cancer. Although the clinical utility of gynecologic cancer symptom recognition has been found to be low in the case of ovarian cancer, 33 the failure to recognize that gynecologic cancer symptoms can signal a serious underlying condition has been found to contribute to delays in care seeking among women who are later diagnosed with these cancers. 34 Thus, educating women about gynecologic cancer symptoms may result in care seeking and diagnosis at an earlier stage, when treatment may be more effective and potentially save a greater number of lives. Further, increasing awareness about recommended Pap test screening intervals may support the appropriate use of resources, by facilitating patients' understanding that an extended (up to 3 years) screening interval does in fact constitute regular screening for some women.
In addition to focusing on knowledge gaps, we also looked at how and where participants reported learning about the discussion topics. They frequently mentioned acquiring their information through personal experience, the experiences of others (including celebrities), healthcare providers, advertisements, and news and entertainment media. The reported influence of the mass media supports the viability of multimedia educational strategies. The power of the mass media to educate consumers is illustrated by the reported impact of celebrity cancer experiences, an interesting phenomenon, given how long these stories are retained by consumers. The celebrities whose names and experiences were recounted most often in our focus groups were Fran Drescher, who was diagnosed with uterine cancer in 2000, and Gilda Radner, who died of ovarian cancer in 1989. Several women mentioned that Drescher consulted multiple doctors before her cancer was diagnosed: “She was very sick and went to different doctors…[they] couldn't find anything wrong. She could have been dead.” Participants also remember details about Radner's case, such as an impassioned televised appeal made by Gene Wilder, Radner's husband, in which he described the early warning signs of ovarian cancer. However, participants more often reported that they learned about the asymptomatic nature of ovarian cancer through coverage of Radner's cancer. In any case, these examples illustrate that celebrity cancer stories can impart health information that remains in the public memory for an extended period of time, even a decade or more.
The focus group method used in this study produced rich, open-ended data and allowed for the exploration of nuances that might be obscured in a survey. For this reason, qualitative research is a useful precursor to quantitative efforts, and the results of this study were used to refine national survey items. Although focus groups support inventorying the range of responses related to the discussion topics and a qualitative assessment of their relative prominence, they cannot provide firm numbers to characterize response volume. Some readers may be frustrated by the proscribed use of such qualifiers as few, some, and many instead of counts or percentages. In any case, this study is one of the few to investigate knowledge about gynecologic cancer among middle-aged U.S. women in the general population, and it is the only study known to us that investigates awareness and knowledge of vaginal and vulvar cancers. As with all qualitative research, the results are not generalizable and should not be projected beyond those women who took part in the study. Nevertheless, the results of this study are generally consistent with prior quantitative studies with a variety of populations.
The Inside Knowledge campaign uses a variety of communication strategies to educate women about gynecologic cancer. Among these, print and broadcast public service announcements (PSAs) have the widest reach, although the amount of information that can be communicated through PSAs is limited by length and space constraints. Thus, Inside Knowledge PSAs are all designed to prompt women to seek additional information both at the campaign's website and from healthcare providers. Although there are many competing priorities for limited counseling time in primary care and obstetrics/gynecology practices, Pap tests and routine examinations offer ideal opportunities for healthcare providers to educate women about gynecologic cancer symptoms, risk factors, and the purpose and frequency of the Pap test. To support physicians' efforts to educate their patients, a variety of free materials is available at the Inside Knowledge website (
Footnotes
Acknowledgments
Funding for this study was provided by the Division of Cancer Prevention and Control, National Center for Chronic Disease Prevention and Health Promotion, Centers for Disease Control and Prevention. The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.
We thank Wendy Child for moderating the focus groups and Jennifer Chu, Margo Gillman, Lauren Grella, Stephanie Mui, and Jennifer Wayman for their assistance in developing the codebook and coding procedure.
Disclosure Statement
The authors have no conflicts of interest to report.
