Take Test
Abstract
Background: The vast majority of cases of ovarian cancer are diagnosed at stage III or IV, and five-year survival rates after diagnosis at these stages are 71% and 31%, respectively. Although a consensus among researchers on the signs and symptoms of ovarian cancer has evolved over time, whether women themselves know them isn't clear.
Objective: To assess how well informed women ages 40 and older are of ovarian cancer symptoms and risk factors.
Methods: In 2006 the National Ovarian Cancer Coalition developed an online survey with a private research firm that asked respondents about their familiarity with ovarian cancer symptoms and risk factors. Women were also asked whether they thought the Papanicolaou test diagnosed ovarian cancer (a common misconception) and whether they had discussed ovarian cancer with a physician. If they had discussed the issue, they were asked who had initiated the conversation. Data from a convenience sample of 1,235 responses to the online survey were analyzed, using descriptive and comparative statistics. Respondents were categorized by age, education level, race or ethnicity, and whether or not they knew someone with ovarian cancer. Comparisons were made to determine whether demographic factors were associated with women's knowledge of specific symptoms and risk factors associated with ovarian cancer.
Results: Only 15% of respondents were familiar with ovarian cancer symptoms, and more than two-thirds incorrectly believed that the Papanicolaou test diagnoses the disease. Four out of five had never had a conversation with a physician about symptoms and risk factors; among these, more than half assumed that because their physician had not initiated such a discussion, ovarian cancer was "not an issue." Of the 19% of women who'd had such discussions, two-thirds had initiated them themselves. Respondents were more knowledgeable about risk factors; 59% correctly identified personal or family history of breast, ovarian, or colon cancer, and half of respondents correctly identified genetic predisposition, as risk factors.
Conclusions: Awareness of ovarian cancer symptoms and risk factors among women in the general population is low. Ovarian cancer is often diagnosed at late stages, when cure is difficult; consequently, heightening women's awareness of risk factors and symptoms might help to reduce delays in diagnosis. Nurses should provide women with specific information on symptoms and risk factors in educating them on ovarian cancer.
Accounting for only 3% of all cancers among women, ovarian cancer can be thought of as a low-prevalence, high-consequence disease: it's estimated that 21,550 new cases will occur in 2009, with 14,600 deaths.1 Patients diagnosed in the early stages of the disease have a five-year survival rate of 93%. But only 19% of cases are diagnosed at early stages. The vast majority are diagnosed at stage III (spread beyond the ovaries within the abdomen) and stage IV (metastasis beyond the abdomen), when the chances of cure are limited. Five-year survival rates are 71% for those diagnosed at stage III and 31% for those diagnosed at stage IV.1, 2
Risk factors. The risk of developing ovarian cancer increases with age and peaks in women in their late 70s. Family history is the single greatest risk factor, but it explains only 5% to 15% of cases of ovarian cancer.3, 4 Women who've had breast cancer or have a family history of breast or ovarian cancer are at increased risk for ovarian cancer.5, 6 Two heritable conditions, breast cancer caused by a BRCA1 or BRCA2 mutation and hereditary nonpolyposis colon cancer, are associated with an elevated risk of ovarian cancer and account for 90% of familial cases.3, 4, 7, 8
A high body mass index (BMI, weight in kilograms divided by the square of height in meters) may also increase the risk of ovarian cancer, most likely because of elevated levels of circulating estrogens. Fairfield and colleagues analyzed longitudinal cohort data from 109,445 participants in the Nurses' Health Study and found that women with a BMI of 25 or higher at age 18 (compared with those with a BMI of less than 20 at age 18) had an increased risk of premenopausal ovarian cancer.9 In a case–control study of more than 2,000 women, Greer and colleagues found that recent gains in both weight and BMI were associated with an increased risk of ovarian cancer in nulliparous women but not in parous women.10 In a longitudinal cohort study of more than 97,000 women, long-term use (more than 10 years) of unopposed estrogen and estrogen with sequential progestin (estrogen each day and progestin roughly 10 days monthly to mimic hormone cycles) was associated with an increased risk, compared with no postmenopausal hormone therapy.11
According to one metaanalysis, in women with an increased genetic risk of developing ovarian cancer, bilateral salpingo-oophorectomy (removal of both ovaries and fallopian tubes) is associated with an 80% reduction in the risk of ovarian and fallopian cancers.12 Another factor in risk reduction is oral contraceptive use. In the most definitive research to date, the Collaborative Group on Epidemiological Studies of Ovarian Cancer conducted a metaanalysis of 45 studies (including 23,257 women with ovarian cancer and 87,303 controls) and found that every five years of oral contraceptive use was associated with a 20% reduction in the risk of ovarian cancer; those who had used oral contraceptives for 15 years had less than half the risk of developing ovarian cancer that those who had never used them had.13 This risk-reducing effect of oral contraceptives persists for up to 30 years but decreases somewhat over time. The authors estimate that oral contraceptives may have prevented as many as 200,000 ovarian cancers to date.
Screening for ovarian cancer has become a focus of research in the past decade, especially the development of diagnostic protocols that might detect ovarian cancer in its early stages. To date, there's no available test with sufficient accuracy in identifying early-stage ovarian cancer. Multiple systematic reviews have confirmed that neither transvaginal ultrasonography nor the cancer antigen–125 (CA-125) blood test, both of which are widely used to track response to treatment and recurrence in women after a diagnosis of ovarian cancer, is effective in identifying early-stage cancers and thereby in reducing mortality.14-16 A common misconception is that a Papanicolaou test is useful in screening for ovarian cancer; however, it's useful only in cervical cancer screening.
The Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial (known as the PLCO trial) randomized 78,000 women over age 55 to a control group or a screening group using several screening methods (annual physical examination, the CA-125 test, and transvaginal ultrasonography).2 The initial report showed that of 1,706 women who had an abnormal CA-125 result or an abnormal transvaginal ultrasound or both, 570 underwent a surgical procedure and 541 had no neoplasm.17 Twenty-nine malignant ovarian neoplasms were identified. Because of the high number of false positives in the trial, the investigators advise against routine CA-125 tests and ultrasonography for screening in the general population. Enrollment in the PLCO trial has been completed and data analysis continues.
Signs and symptoms. Some women diagnosed with ovarian cancer report having had symptoms before the diagnosis, which seems to contradict the common description of ovarian cancer as a "silent killer." The challenge is to distinguish the symptoms of ovarian cancer from those of other conditions, such as irritable bowel syndrome. Several studies have helped to clarify these distinctions.
In a retrospective study conducted in 2000 by Goff and colleagues, 95% of women with ovarian cancer who responded to a survey said they'd had symptoms before diagnosis: 77% reported abdominal symptoms (bloating, pain, increased size); 70%, gastrointestinal symptoms (indigestion, constipation, nausea); 58%, symptoms involving pain (abdominal pain, pain with intercourse, back pain); 50%, constitutional symptoms (fatigue, anorexia, weight loss); 34%, urinary symptoms (frequency or incontinence); and 26%, pelvic symptoms (bleeding, a palpable mass).18 It might be expected, because of the ovaries' location, that pelvic complaints would be common; interestingly, they were the least reported of the major groups of symptoms. In another study of women with early-stage disease, the most commonly reported symptom was bloating, followed by gastrointestinal disturbances.19 When comparing the symptoms of women who had ovarian cancer with controls, the researchers found that patients with cancer reported more bloating, lack of appetite, abdominal pain, fatigue, urinary frequency, and constipation.
A benchmark prospective study attempted to distinguish ovarian cancer symptoms from benign pelvic symptoms.20 The study included 1,709 women who visited two primary care clinics and completed an anonymous survey about their symptoms over a one-year period. Results showed that the two groups' symptoms were distinguished by type, severity, frequency, and duration. Women subsequently diagnosed with ovarian cancer reported more severe abdominal and pelvic pain, abdominal bloating, and constipation than women with benign pelvic problems. Also, women subsequently diagnosed with ovarian cancer said their symptom occurred 15 to 30 times a month; women with benign pelvic problems said their symptom occurred two to three times a month. Women presenting with ovarian cancer had been experiencing symptoms for a shorter duration of time (three to six months), compared with those presenting with irritable bowel syndrome (12 months to two years). Finally, women subsequently diagnosed with ovarian cancer reported a higher median number of symptoms (eight), compared with the benign pelvic problem group (four).
As a result of these studies, the Gynecologic Cancer Foundation, the Society of Gynecologic Oncologists, and the American Cancer Society authored a consensus statement, along with 34 endorsing organizations, that included the following21
Women who have these symptoms [bloating, pelvic or abdominal pain, difficulty eating or feeling full quickly, urinary urgency or frequency] almost daily for more than a few weeks should see their doctor, preferably a gynecologist. Prompt medical evaluation may lead to detection at the earliest possible stage of the disease. Early stage diagnosis is associated with an improved prognosis.
To date, there have been no published studies reporting knowledge of the signs and symptoms of ovarian cancer among women in the general population. Therefore, the purpose of this study was to
* describe the knowledge of women in the general public regarding ovarian cancer.
* evaluate whether that knowledge differs according to age, race or ethnicity, education, or knowing someone with the disease.
METHODS
Design and sample. The National Ovarian Cancer Coalition (NOCC), a nonprofit advocacy organization with more than 50 chapters across the United States, engaged a national survey research firm, Impulse Research, to conduct a descriptive, cross-sectional survey of a random sample of U.S. women ages 40 and older. (The mission of the NOCC is "to raise awareness and promote education about ovarian cancer. The coalition is committed to improving the survival rate and quality of life for women with ovarian cancer." See www.ovarian.org . Impulse Research maintains a panel of approximately 1 million members selected to represent a cross section of the U.S. population with an overall sampling error of +/- 3% at the 95% level of confidence [e-mail communication, Varda Novick, June 18, 2009]. See www.impulseresearch.com .) Survey questions were developed by the NOCC in consultation with Impulse Research.
In March 2006 the firm e-mailed all female members of the panel who were 40 years of age or older, which included those age groups known to have a higher risk of ovarian cancer, inviting them to follow a link to the Impulse Research Web site to complete the survey. Participants could read a confidentiality statement on the Web site; no identifiable information was collected in the survey. In March and April 2006, the responses of the first 1,235 women who completed the survey were tabulated.
Measures. Respondents were asked to provide basic demographic information (age, race or ethnicity, and education; see Table 1). They were also asked whether they or anyone they knew had ever had ovarian cancer. The following areas were addressed in the survey to assess ovarian cancer knowledge:
 | Table 1. Demographic Data (N = 1,235) |
1. familiarity with ovarian cancer symptoms, which was assessed with a single item: "How familiar are you with the symptoms of ovarian cancer?" (See Table 2.)
 | Table 2. Survey Questions and Responses (N = 1,235) |
2. knowledge of specific symptoms associated with ovarian cancer, which was assessed by asking respondents to identify which of seven listed items could be symptoms of ovarian cancer (see Table 3). Respondents were instructed to "check all that apply" but were also given the options "None of the above" or "Not sure about the symptoms of ovarian cancer." (All but one of the symptoms ["continuous fever"] are possible symptoms of ovarian cancer.)
 | Table 3. Knowledge of Ovarian Cancer Symptoms: Comparison Between Women Who Know Someone with Ovarian Cancer and Those Who Don't |
3. familiarity with risk factors for ovarian cancer, which was assessed with a list of six possible risk factors (see Table 4). Respondents were asked to "check all that apply" but were also given the option "Not sure about the risk factors of ovarian cancer." All risk factors listed are true risk factors, except for prolonged use of birth control pills. In fact, use of birth control pills reduces ovarian cancer risk.
 | Table 4. Knowledge of Ovarian Cancer Risk Factors: Comparison Between Women Who Know Someone with Ovarian Cancer and Those Who Don't |
4. perceptions of whether Papanicolaou tests are effective in diagnosing ovarian cancer, which was assessed with a single item. (See Table 2.)
5. communication with a physician about ovarian cancer symptoms and risks, which was evaluated by inquiring whether the respondents had ever talked with a physician about the symptoms and risks (yes or no) and who (the respondent or physician) initiated the discussion. (See Table 2.)
For comparison purposes, the survey included two items that assessed respondents' familiarity with breast cancer symptoms and whether they had talked with their physician about the symptoms and risk factors associated with breast cancer. (See Table 2.)
After the NOCC received the report from Impulse Research and integrated its findings into educational materials and outreach programs, the NOCC provided the anonymous dataset to the authors for analysis and dissemination to the nursing community.
Statistical analysis. A single [chi]2 analysis was conducted for each of the knowledge-related questions (symptoms and risk factors), comparing results according to each of the following designations: those who knew a woman with ovarian cancer and those who did not; age (40 to 49 years old, 50 to 59 years old, and 60 years old and older); educational level ("high school graduate or less," "some college," "college graduate," and "postgraduate study or degree"); and race or ethnicity ("African-American," "Caucasian," "Asian," "Latino or Hispanic," and "other"; note that these racial and ethnic designations were the terms as they appeared on the survey instrument). If the overall [chi]2 analysis was significant (for example, if respondents' knowledge of symptoms differed significantly according to their educational level), pairwise comparisons of column proportions were conducted using a two-sided proportion z-test. In order to control for type I error associated with multiple comparisons, we used Bonferroni adjustments at [alpha] divided by the number of pairwise comparisons.
RESULTS
Of the 1,211 women surveyed who did not have a history of ovarian cancer, more than three-quarters (79%) did not know anyone with ovarian cancer. Of the 256 women who knew someone with ovarian cancer, 106 (41%) had a family member with ovarian cancer, 85 (33%) had a friend or coworker with ovarian cancer, and 83 (32%) had an acquaintance with ovarian cancer. Only 24 (2%) of the total 1,235 respondents had been diagnosed with ovarian cancer.
Only 15% of respondents reported that they were either familiar or very familiar with the symptoms of ovarian cancer, whereas 59% reported that they were either not very or not at all familiar with the symptoms (See Table 2). (Interestingly, more than three times as many respondents (52%) were familiar or very familiar with the symptoms of breast cancer.) More than two-thirds (67%) of the women surveyed incorrectly believed that the Papanicolaou test is effective in diagnosing ovarian cancer.
Four out of five respondents (81%) reported that they had never had a conversation with their physician about the symptoms and risk factors associated with ovarian cancer (by comparison, 60% had discussed breast cancer with their physician). Of the 239 women who had discussed ovarian cancer with their physician, 151 (63%) had initiated the conversation themselves. Of the 996 women who did not discuss ovarian cancer with their physician, 534 (54%) "did not think [ovarian cancer] was an issue" because their physician had not brought it up.
Knowledge of specific symptoms. Fewer than half of respondents correctly identified each of the symptoms associated with ovarian cancer (see Table 3). The symptoms most likely to be recognized included "pelvic and/or abdominal swelling, bloating and/or feeling of fullness" (47%), "ongoing unusual fatigue" (36%), and "unexplained weight gain or weight loss" (36%). The least-recognized symptoms were "unexplained changes in bowel habits" (28%) and "frequency and/or urgency of urination in the absence of an infection" (26%).
Knowledge of symptoms was compared across categories of respondents. Overall, successful identification of symptoms differed according to whether the respondent knew someone with ovarian cancer ([chi]2(9) = 206.23; P < 0.001). And for each symptom, the pairwise comparison showed that women who knew someone with ovarian cancer were more likely (P < 0.006) to identify it as a symptom of ovarian cancer than were those who didn't know someone with the disease. However, women who knew someone with ovarian cancer were also more likely (P < 0.006) to identify, incorrectly, continuous fever as a symptom of the disease.
Knowledge of symptoms also differed according to education level [chi]2(27) = 214.51; P < 0.001). When looking at pairwise comparisons, women with a college degree or who had engaged in postgraduate studies were significantly more likely (P < 0.0009) to identify each of the correct symptoms than were those with a high school education or less (see Figure 1). Knowledge of symptoms did not differ according to age ([chi]2(18) = 20.2; P = 0.321) or race or ethnicity ([chi]2(36) = 36.89; P = 0.428). (Because the overall [chi]2 value was not significant, we did not conduct further analysis of knowledge-related results as they pertained to respondents' race or ethnicity or age.)
 | Figure 1. Percentage of Women, by Education Level, Who Identified Each Item as a Symptom of Ovarian Cancer |
Knowledge of specific risk factors. Thirty-nine percent of respondents reported that they were unsure about the risk factors associated with ovarian cancer (see Table 4). Three out of five respondents (60%) correctly identified personal or family history of breast, ovarian, or colon cancer and 51% correctly identified genetic predisposition as risk factors for developing ovarian cancer. About one-quarter (26%) of respondents incorrectly identified the extended use of birth control pills as a risk factor for ovarian cancer.
Knowledge of risk factors was compared across categories of respondents. Identification of risk factors differed according to whether the respondent knew someone with ovarian cancer ([chi]2(7) = 177.42; P < 0.001). For each risk factor listed (including the extended use of birth control pills, which doesn't increase the risk of ovarian cancer), pairwise comparisons showed that women who knew a woman with ovarian cancer were more likely to identify it as a risk factor for ovarian cancer. Knowledge of risk factors also differed by age ([chi]2(14) = 44.05; P < 0.001), with pairwise comparisons showing that women ages 60 and older were less knowledgeable (P < 0.002) about the true risk factors for ovarian cancer than were women in the younger two age categories.
Finally, knowledge of risk factors differed by level of education ([chi]2(21) = 159.91; P < 0.001). In pairwise comparisons, women with a college degree or who had engaged in postgraduate studies were significantly more likely (P < 0.001) to identify each of the true risk factors than were those with a high school education or less (See Figure 2). Knowledge of risk factors did not differ by race or ethnicity ([chi]2(28) = 25.02; P = 0.626).
 | Figure 2. Percentage of Women, by Education Level, Who Identified Each Item as a Risk Factor for Ovarian Cancer |
Respondents who knew a woman with ovarian cancer certainly had heightened awareness of the disease. Interestingly, their knowledge wasn't always more accurate than that of those who didn't know a woman with the disease. Respondents who knew a woman with ovarian cancer broadly endorsed all symptoms and risk factors at a higher frequency, without distinguishing between the real symptoms and risk factors and the false ones.
DISCUSSION
There are no ovarian cancer screening tools with sufficient accuracy to be recommended for use in the general population. Therefore, researchers have focused on identifying symptoms associated with ovarian cancer whose appearance could trigger focused diagnostic screening and, ultimately, increase the percentage of women diagnosed at early stages of the disease. Goff and colleagues have identified a set of possible symptoms and their characteristic pattern: a new onset of bloating, pelvic or abdominal pain (or both), difficulty eating or feeling full quickly, and urinary symptoms (urgency or frequency) that occur almost daily and persist for more than three weeks.20, 21 Currently, it isn't known whether implementing the recommendations of the Gynecologic Cancer Foundation consensus statement will reduce delays in ovarian cancer diagnosis or result in an increase in early-stage diagnoses. This is an important area for future research.
Unfortunately, the results of our study demonstrate that women are not knowledgeable about the risk factors and symptoms associated with ovarian cancer. In order to reduce delays in diagnosis, women should be encouraged to seek medical examination (from a gynecologist or gynecology or women's health NP) after experiencing a new onset of pelvic or abdominal symptoms. Health care providers should consider ovarian cancer when performing a differential diagnosis in women presenting with these symptoms.
The number of respondents who believed that a Papanicolaou test was effective for ovarian cancer screening also indicates that patient education on the purpose of Papanicolaou testing is insufficient. Furthermore, the results show that many women do not discuss ovarian cancer with their physician, suggesting that nurses have a significant opportunity to educate women, to help them learn about their personal risk factors, recognize symptoms of ovarian cancer, and have a dialogue with their physician or health care provider about all aspects of gynecologic health care. Of particular note is the finding that women with lower levels of education are less knowledgeable regarding symptoms and risk factors of ovarian cancer.
As patient advocates, nurses may have many opportunities to effect early detection of ovarian cancer by offering educational programs to groups in settings such as community centers, retirement centers, churches, and various nonprofit agencies. Based on the findings from this study, nurses should target women with lower education levels, in particular, to reduce disparities in knowledge. Ovarian cancer advocacy groups such as the NOCC and the Ovarian Cancer National Alliance, as well as the Gynecologic Cancer Foundation, can provide literature and other resources that can be used in the development of audience-appropriate programs.
Limitations. There were several limitations related to the study's design and methodology. The study was conducted outside of a traditional research setting by a professional agency that assists organizations in online data collection. The sample was randomly selected from across the United States; however, respondents were members of a survey research panel. Although Impulse Research carefully selects members to be representative of the general population, the demographics of this sample show that minorities and older adults were underrepresented. Limiting results to the first 1,235 respondents may have introduced a systematic selection bias (those most interested in ovarian cancer), as does the requirement of Internet access. Because of anonymity filters and programming used for distribution of the survey, the total number of women who met participation criteria is also unknown, making it impossible to calculate the actual response rate. In addition, the NOCC-developed survey didn't have established validity. This is not a significant limitation, however, given the use of the items simply to describe knowledge of established symptoms and risk factors.
An exclusive focus on communication with physicians is a limitation: the survey asked whether the woman talked with her "physician" about symptoms and risks for ovarian cancer. This ignores the fact that women often engage with health care providers other than physicians, including NPs and office nurses. Consequently, the survey may underestimate the extent to which women talk with a health care provider about ovarian cancer. However, the same question structure was used for the questions about breast cancer, and in comparison, women discussed ovarian cancer much less frequently.
Further research should aim to identify the barriers—both individual and systemic—to discussion and timely diagnosis of ovarian cancer. Clinical tools to streamline the assessment of indicators listed in the recent consensus statement (bloating, pelvic or abdominal pain, difficulty eating or feeling full quickly, and urinary symptoms) should be developed. Prompts on intake or patient history forms could facilitate earlier diagnosis and improved outcomes.
Tests currently available (the CA-125 blood test, transvaginal ultrasound, or a combination) lack the sensitivity to support screening of the general population. Studies to identify tumor markers and diagnostics specific to ovarian cancer are in various stages of development. Although a screening test for ovarian cancer may be available in the future, until that time it's imperative that health care providers clarify misconceptions and educate women regarding the risks and symptoms of ovarian cancer.
For more than 15 additional continuing nursing education articles related to the topic of cancer, go to www.nursingcenter.com/ce .
REFERENCES
1. American Cancer Society. Cancer facts and figures 2009. Atlanta; 2009. www.cancer.org/downloads/STT/500809web.pdf . [Context Link]
2. National Cancer Institute. Ovarian cancer research results from the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial: fact sheet. 2005. http://www.cancer.gov/cancertopics/factsheet/Detection/PLCOOvarianFactSheet . [Context Link]
3. Russo A, et al. Hereditary ovarian cancer. Crit Rev Oncol Hematol 2009;69(1):28–44. [Context Link]
4. South SA, et al. Consideration of hereditary nonpolyposis colorectal cancer in BRCA mutation–negative familial ovarian cancers. Cancer 2009;115(2):324–33. [Context Link]
5. Lynch HT, et al. Hereditary carcinoma of the ovary and associated cancers: a study of two families. Gynecol Oncol 1990;36(1):48–55. [Context Link]
6. Prat J, et al. Hereditary ovarian cancer. Hum Pathol 2005;36(8):861–70. [Context Link]
7. Lynch HT, et al. Hereditary ovarian cancer. Heterogeneity in age at onset. Cancer 1993;71(2 Suppl):573–81. [Context Link]
8. Piver MS, et al. Characteristics of familial ovarian cancer: a report of the first 1,000 families in the Gilda Radner Familial Ovarian Cancer Registry. Eur J Gynaecol Oncol 1996;17(3):169–76. [Context Link]
9. Fairfield KM, et al. Obesity, weight gain, and ovarian cancer. Obstet Gynecol 2002;100(2):288–96. [Context Link]
10. Greer JB, et al. Anthropometry and the risk of epithelial ovarian cancer. Cancer 2006;106(10):2247–57. [Context Link]
11. Lacey JV Jr, et al. Menopausal hormone therapy and ovarian cancer risk in the National Institutes of Health–AARP Diet and Health Study Cohort. J Natl Cancer Inst 2006;98(19):1397–405. [Context Link]
12. Rebbeck TR, et al. Meta-analysis of risk reduction estimates associated with risk-reducing salpingo-oophorectomy in BRCA1 or BRCA2 mutation carriers. J Natl Cancer Inst 2009;101(2):80–7. [Context Link]
13. Collaborative Group on Epidemiological Studies of Ovarian Cancer. Ovarian cancer and oral contraceptives: collaborative reanalysis of data from 45 epidemiological studies including 23,257 women with ovarian cancer and 87,303 controls. Lancet 2008;371(9609):303–14. [Context Link]
14. Bell R, et al. The performance of screening tests for ovarian cancer: results of a systematic review. Br J Obstet Gynaecol 1998;105(11):1136–47. [Context Link]
15. Fung MF, et al. Screening postmenopausal women for ovarian cancer: a systematic review. J Obstet Gynaecol Can 2004;26(8):717–28. [Context Link]
16. U.S. Preventive Services Task Force. Screening for ovarian cancer: recommendation statement. Rockville, MD: Agency for Healthcare Research and Quality; 2004 May. http://www.ahrq.gov/clinic/3rduspstf/ovariancan/ovcanrs.htm . [Context Link]
17. Buys SS, et al. Ovarian cancer screening in the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial: findings from the initial screen of a randomized trial. Am J Obstet Gynecol 2005;193(5):1630–9. [Context Link]
18. Goff BA, et al. Ovarian carcinoma diagnosis. Cancer 2000;89(10):2068–75. [Context Link]
19. Olson SH, et al. Symptoms of ovarian cancer. Obstet Gynecol 2001;98(2):212–7. [Context Link]
20. Goff BA, et al. Frequency of symptoms of ovarian cancer in women presenting to primary care clinics. JAMA 2004;291(22):2705–12. [Context Link]
21. Women's Cancer Network. Ovarian cancer symptoms consensus statement. Gynecologic Cancer Foundation. 2007. http://www.wcn.org/articles/types_of_cancer/ovarian/symptoms/concensus_statement.html . [Context Link]