Objective
We evaluated baseline knowledge of ovarian cancer risk and perceptions toward ovarian cancer screening (OCS) by initiating the normal risk ovarian screening study.
Study Design
Average-risk, postmenopausal women were enrolled between 2001 and 2011 as they entered the normal risk ovarian screening study. Participants completed baseline surveys of risk perception, cancer worry (Cancer Worry Scale), anxiety (State-Trait Anxiety Inventory), health and well-being survey (SF-36 HEALTH SURVEY), and acceptability of OCS.
Results
Of the 1242 women who were enrolled, 925 women (74.5%) completed surveys. The respondents estimated a mean lifetime risk of ovarian cancer of 29.9%, which is much higher than the actual risk of 1.4% for women in the United States. Only 2.8% of participants correctly estimated their risk; 35.4% of the participants reported their lifetime risk to be ≥50%. Cancer worry was low, with a median Cancer Worry Scale score of 7 of 24. Anxiety was comparable with published norms for women in this age group, with median STAI-State and STAI-Trait scores of 30 and 29 of 80, respectively. Overall, women reported good physical and mental well-being. In terms of OCS acceptability, 97.2% of respondents agreed or strongly agreed that “the benefits of screening outweigh the difficulties.” Very few women were reluctant to undergo OCS because of time constraints (1.1%), pain (2.0%), or embarrassment (1.9%).
Conclusion
Average-risk women who underwent OCS highly overestimated their risk of ovarian cancer. Despite this, participants reported low cancer worry and anxiety. The discrepancy between knowledge of and attitudes toward ovarian cancer risk highlights the need for educational efforts in this area.
Although not common, ovarian cancer is the most deadly gynecologic malignancy; there were 22,000 new cases and >14,000 deaths anticipated in 2013. Although stage I disease is associated with a 90% 5-year survival rate, the prognosis for advanced stages is poor. Because of the nonspecific symptoms of early ovarian cancer, 80% of women are stage III or IV at the time of diagnosis. One strategy to decrease the mortality rate is to establish an effective method for the detection of early-stage ovarian cancer. To date, however, potential screening methods for ovarian cancer have had unacceptably high false-positive rates without any demonstration of a significant reduction in ovarian cancer deaths, which has led to recommendations against routine ovarian cancer screening for the general population.
A single-arm, multiinstitutional prospective study of ovarian cancer screening that used the Risk of Ovarian Cancer Algorithm, known as the Normal Risk Ovarian Screening Study (NROSS), is currently underway in the United States. This study involves annual CA125 measurements in postmenopausal women who have an average (eg, normal, population-based) risk of the development of ovarian cancer. The screening algorithm incorporates a woman’s age and the change in her CA125 measurement over time to estimate her risk of having undetected ovarian cancer. Women deemed to be at intermediate risk (between 1 in 2000 and 1 in 500) are triaged to undergo a repeat CA125 measurement in 3 months, although women who are found to be at an elevated risk are triaged to undergo a transvaginal ultrasound scan and referral to a gynecologic oncologist. Recently published results demonstrate that ovarian cancer screening with the Risk of Ovarian Cancer Algorithm is feasible, with a positive predictive value of 40% and a specificity of 99.9%.
There have been few published studies regarding ovarian cancer knowledge and risk perception among average-risk women who undergo ovarian cancer screening. Baseline analysis of participants in the UK Collaborative Trial of Ovarian Cancer Screening (UKCTOCS), which is a prospective trial of ovarian cancer screening in the United Kingdom, found that, although >40% of women correctly identified their risk, approximately one-half of respondents underestimated their ovarian cancer risk. In contrast, studies in the United States have found that most women tend to overestimate their ovarian cancer risk. This lack of knowledge regarding personal cancer risk is not surprising given that multiple studies have demonstrated that women generally are unaware of ovarian cancer risk factors. Little is known about ovarian cancer worry or acceptance of screening among average-risk women who undergo ovarian cancer screening.
The present study was designed as a companion to the NROSS. We aimed prospectively to assess ovarian cancer risk perception, ovarian cancer worry, and acceptability of ovarian cancer screening among women who initiate an ovarian cancer screening trial that uses the Risk of Ovarian Cancer Algorithm.
Materials and Methods
Between July 2001 and October 2012, 1457 women were enrolled prospectively in the NROSS at MD Anderson Cancer Center. Most of these women were recruited to the trial from MD Anderson’s Cancer Prevention Clinic, which is a clinic that provides care to patients without active cancer. Other participants found out about the trial through word-of-mouth from women who already were enrolled in the trial. Women who were eligible for the NROSS were postmenopausal, from 50-75 years old, had at least 1 ovary, had no active cancer and no history of ovarian cancer, and were willing to undergo annual blood draws, follow-up CA125 measurements, and transvaginal ultrasound scans as recommended by the NROSS protocol. Women were ineligible if they were at high risk for the development of ovarian cancer. High-risk was defined as a patient with at least 1 of the following: (1) BRCA1 or BRCA2 mutation, (2) known or suspected Lynch syndrome, (3) a first- or second-degree male relative with breast cancer, (4) Ashkenazi Jewish descent with premenopausal breast cancer or a family history of ovarian cancer or premenopausal breast cancer, or (5) 2 first- or second-degree relatives with ovarian cancer, premenopausal breast cancer, or both. The full details of the NROSS protocol have been published.
After informed consent was obtained for NROSS, women who could read and speak English were offered participation in the questionnaire portion of the trial. Because of financial constraints, only NROSS participants at MD Anderson Cancer Center were asked to enroll in the questionnaire study. Of the 1457 women who were eligible for the survey, 1242 women (85.2%) provided separate written, informed consent to participate before study enrollment. This study was approved by the Institutional Review Board of The University of Texas MD Anderson Cancer Center.
On entry into the study and before undergoing ovarian cancer screening, participants were given the study questionnaires. Patient demographics and medical history were collected prospectively. The survey assessed ovarian cancer risk perception and the acceptability of ovarian cancer screening, ovarian cancer worry with the use of a modified Lerman breast cancer worry scale, anxiety with the use of the Spielberger State/Trait Anxiety Inventory (STAI), and general quality of life with the SF-36 Health Survey. No educational materials were provided as part of this study.
The SF-36 Health Survey, STAI, and the Cancer Worry Scale were scored according to the respective scoring mechanism for each instrument. Baseline demographic information was summarized with descriptive statistics that included medians, means, standard deviations, ranges, and frequencies. Mann-Whitney U and χ 2 tests were used to compare differences between groups. IBM SPSS Statistics for Windows (version 19.0; IBM Corp, Armonk, NY) was used for statistical analysis. A 2-sided probability value of < .05 was considered statistically significant.
Results
Of the 1242 women who were enrolled in the study, 925 women (74.5%) completed the surveys. Table 1 shows a comparison of the demographics of women who completed surveys (responders) to those who did not (nonresponders). Overall, the groups were similar, although responders were slightly older and more likely to be white than were nonresponders.
| Variable | Responders (n = 925) | Nonresponders (n = 317) | P value |
|---|---|---|---|
| Age, y | .004 | ||
| Median, range | 59 (50–74) | 58 (50–74) | |
| Mean | 60.1 | 59 | |
| Race, n (%) | < .001 | ||
| White | 792 (85.6) | 236 (74.4) | |
| African American | 43 (4.6) | 29 (9.1) | |
| Hispanic | 43 (4.6) | 30 (9.5) | |
| Asian | 46 (5.0) | 21 (6.6) | |
| Other | 1 (0.1) | 1 (0.3) | |
| Parity, n (%) | .325 | ||
| Parous | 745 (80.5) | 257 (81.1) | |
| Nulliparous | 179 (19.4) | 52 (16.4) | |
| Unknown | 1 (0.1) | 8 (2.5) | |
| Cancer history, n (%) | .101 | ||
| None | 402 (43.5) | 145 (45.8) | |
| Breast cancer | 234 (25.3) | 80 (25.2) | |
| Other cancer | 289 (31.2) | 92 (29.0) | |
| Family history, n (%) | |||
| Breast cancer | 325 (35.1) | 99 (31.2) | .185 |
| Ovarian cancer | 71 (7.7) | 26 (8.2) | .782 |
In general, survey participants highly overestimated their risk of ovarian cancer. Respondents estimated their mean lifetime risk to be 29.9% ( Figure ). Approximately one-fifth of the participants stated that their risk of ovarian cancer was 25-49%, and more than one-third of them believed their risk to be ≥50%. Only 2.8% of women correctly identified their risk as 1-2% ( Table 2 ). When asked what their chances of getting ovarian cancer were compared with “other women your age,” 45.3% of the respondents said they were “about the same.” However, 22.2% of the women believed their ovarian cancer risk was “a little higher” or “much higher” than other women. Interestingly, only 26.4% of the women who responded noted that they were “fairly certain” or “very certain” about their opinions on ovarian cancer risk. Of note, 55% of respondents said that they received their information about ovarian cancer from television, newspapers, or magazines, although just over one-quarter of them relied on their healthcare provider for this information.
| SF-36 Health Survey domains | Median survey participant score | 50th percentile for women in the United States | 75th percentile for women in the United States |
|---|---|---|---|
| Physical functioning | 90.0 | 90.0 | 100.0 |
| Role-physical | 100.0 | 100.0 | 100.0 |
| Bodily pain | 84.0 | 74.0 | 100.0 |
| General health | 82.0 | 72.0 | 85.0 |
| Vitality | 70.0 | 60.0 | 75.0 |
| Social functioning | 100.0 | 87.5 | 100.0 |
| Role-emotional | 100.0 | 100.0 | 100.0 |
| Mental health | 88.0 | 80.0 | 88.0 |
Although survey participants tended to overestimate their risk of ovarian cancer, they were optimistic regarding the ability to screen for the disease. Of the respondents, 93.6% “agreed” or “strongly agreed” that ovarian cancer could be cured when detected early. Furthermore, 91.5% of the women believed that transvaginal ultrasonography could detect ovarian cancer at a stage when it could be cured; 86.4% of them thought the same of a single CA125 value. Additionally, in spite of perceiving themselves to be at high risk for ovarian cancer, few women (12%) were afraid that their ovarian cancer screening results would be abnormal. Likewise, only 4.4% of respondents reported that they often thought about ovarian cancer; 0.6% of them stated that thoughts about ovarian cancer affected their mood, and 0.3% of them reported that thoughts about ovarian cancer interfered with their ability to perform their daily activities.
A modified cancer worry scale and the STAI were used to measure objectively study participants’ anxiety and ovarian cancer worry. As expected, given participants’ responses regarding frequency of ovarian cancer thoughts, cancer worry was low, with a median score of 7.0 (range, 6.0–20.0) of 24. The STAI also found anxiety to be low overall. The median STAI-State (STAI-S) score, which measures anxiety about an event, was 30 of 80. The median STAI-Trait (STAI-T) score, which measures one’s baseline anxiety, was 29 of 80. This is comparable with the published norms of postmenopausal women for these scales of 32.2 and 31.79, respectively.
As evidenced by the SF-36 Health Survey scores, study participants demonstrated good physical and mental health. Table 3 lists participants’ median scores for each of the 8 scales on the SF-36 Health Survey as well as the 50th and 75th percentiles of SF-36 Health Survey scores for women in the general population in the United States. Survey participants were also queried regarding acceptability of ovarian cancer screening ( Table 3 ). The overwhelming majority of women “agreed” or “strongly agreed” that there were more benefits to ovarian cancer screening than difficulties. Very few women said pain, embarrassment, or time constraints were reasons that they would be reluctant to undergo screening. The most commonly cited reason to not be screened was fear of insurance not covering the cost of the tests.
