Current cervical cancer screening recommendations for women aged 30–65 years by the U.S. Preventive Services Task Force (2018) and the American College of Obstetricians and Gynecologists (2021) include Pap test alone, human papillomavirus (HPV) testing alone, or both (cotesting).^1-3 In 2020, the American Cancer Society recommended moving toward HPV testing alone as the preferred screening test.⁴ A recent study in a nationwide convenience sample of women with private health insurance from 2013 to 2019 showed that cotesting increased in women aged 30–65 years, whereas HPV testing alone was rarely used.⁵ National health surveys are critical sources for monitoring the nation’s cancer screening coverage and guiding Healthy People objectives.⁶ However, self-reported surveys may be limited in providing the most accurate information and may not yield reliable estimates of cervical cancer screening, especially if women are not aware of their testing status.⁷ In this study, we estimate the prevalence of women’s responses to HPV testing awareness status questions and identify the demographic characteristics associated with these responses.

Self-reported data on women were analyzed using the 2020 Behavioral Risk Factor Surveillance System (BRFSS) (n=80,648, aged 30–64 years),⁸ the 2019 National Health Interview Survey (NHIS) (n=7,062, aged 30–65 years),⁹ and the 2017–2019 National Survey of Family Growth (NSFG) (n=2,973, aged 30–49 years).¹⁰ Overall response rates were 47.9% for BRFSS, 59.1% for NHIS, and 65.3% for NSFG. Excluded from the analyses were women reporting hysterectomy (18.1% [BRFSS], 17.7% [NHIS], and 8.4% [NSFG]) and women having had cervical cancer without hysterectomy (<1%) in NHIS and NSFG. Questions about women’s Pap and HPV testing experiences are in Appendix Table 1 (available online).

In 2022, unadjusted and adjusted associations between HPV testing awareness status (answers yes, no, or don’t know) and demographic characteristics were examined. The analyses varied by survey questions, sample sizes, survey years (Appendix Tables 1 and 2, available online), and available characteristics (age, race or ethnicity, education, region, health insurance, and family income as a ratio to poverty threshold). Therefore, there was no intention to compare the surveys with one another. Percentages were weighted to each survey’s study population.

Associations for the adjusted analyses were performed with generalized multinomial logistic model to account for the 3-level HPV testing awareness status outcome. Adjusted prevalence ratio (APR) for each subcategory was generated,¹¹ and adjusted risk differences (pairwise comparisons) were calculated through predictive margins proportions. Statistical significance was assessed with t-test, and p-values were presented only for the response don’t know. All analyses were performed with SAS-callable SUDAAN, Version 10 (Research Triangle Institute, Research Triangle Park, NC), to account for the complex structure of the data and nonresponse.

More than 94% of women in all surveys responded that they had had a Pap test. However, only 49.3%, 36.6%, and 56.8% of women in BRFSS, NHIS, and NSFG, respectively, responded that they had received HPV testing (Appendix Table 2, available online). Moreover, 21.8% in BRFSS and 19.5% in NHIS (representing >12 million and 10 million U.S. women, respectively) and 9.4% in NSFG were unaware of their HPV testing status. The percentage of don’t know ranged from 16.2% for women aged 30–34 years to 29.2% for women aged 60–64 years in BRFSS (Table 1) and from 16.2% for women aged 30–34 years to 22.5% for those aged 50–54 years in NHIS (Table 2). The percentages of don’t know in BRFSS varied from 13.0% among non-Hispanic (NH) Native Hawaiian/Pacific Islander, 15.2% among Hispanic women, and 15.6% among NH Black women to 25.6% among NH White women. In NHIS, only the percentage of NH Black women (14.5%) was lower than that of NH White women (20.9%). In BRFSS, the West had the lowest percentage (18.8 %) of women responding don’t know than all other regions. Controlling for all other variables in each survey’s model, the analyses showed that women aged 40–44 years or older (APR=1.17, 95% CI=1.04, 1.30) in BRFSS and those 50–54 years or older (APR=1.35, 95% CI=1.11, 1.66) in NHIS were significantly more likely than women aged 30–34 years to respond don’t know to the HPV screening awareness status question (p<0.01 for each survey). NH Native Hawaiian/Pacific Islander women (APR=0.60, 95% CI=0.39, 0.91), NH Black women (APR=0.63, 95% CI=0.56, 0.70), Hispanic women (APR=0.64, 95% CI=0.57, 0.72), and NH Asian women (APR=0.79, 95% CI=0.66, 0.95) in BRFSS and NH Black women in NHIS (APR=0.72, 95% CI=0.58, 0.88) had significantly lower don’t know percentages than NH White women (p<0.01). Education level was not associated with don’t know. In BRFSS, women in the West were less likely to respond don’t know to the HPV testing awareness status question than women in all other regions (p<0.01). No significant associations with don’t know response of HPV testing awareness and demographic characteristics were found in NSFG (Appendix Table 3, available online).

Almost all women in the 3 surveys were aware of their Pap test status; in contrast, about 1 in 5 or >12 million U.S. women represented in BRFSS and >10 million represented in NHIS were unaware of their HPV testing status. Despite the increasing adoption of HPV testing for cervical cancer screening, national gaps in women’s awareness should be communicated to researchers using national health survey data to estimate HPV testing uptake for cervical cancer screening.

Interestingly, race/ethnicity minority women were less likely than NH White women to respond Don’t know to the HPV testing awareness status question. This finding might partly explain the difference in women’s responses between the West and all other regions. Additional BRFSS data analysis of race or ethnicity by region showed that the West had the lowest percentage of NH White women (49.3%) and the highest total percentage (45.5%) of Hispanic, NH Asian, and NH Black women (Appendix Table 4, available online).

Self-reported responses could be influenced by recall bias, especially among older women, and by low health literacy and social desirability.¹² In addition, perhaps low HPV testing awareness occurs because Pap and HPV test specimen collections are similar. A recent study with >230,000 women participants revealed the challenge of reporting cervical cancer screening estimates where authors report having to exclude >25% of the participants who were unsure about receiving up-to-date HPV tests from the analysis.⁷

Possible interventions for improving HPV testing awareness could be clearer communication between provider and patient about the test administered. For example, patient reminders that have been found to increase clinic-level rates of cervical, breast, and colorectal cancer screening¹³ might also provide opportunities to improve patients’ health literacy by addressing the purpose and importance of the HPV test. National and local cervical cancer screening campaigns could include education about HPV testing to increase women’s awareness and knowledge about the test,¹⁴ providing an informative statement related to HPV infection and HPV testing in surveys may help women answer related questions, and future studies need to consider the large proportion of women who are not aware of having had HPV testing.

These results from 2 large national surveys reveal a national problem in estimating HPV testing uptake in the U.S. One in 5 U.S. women eligible for cervical cancer screening was not aware of her past HPV testing status, and awareness was lower among older and NH White women. Future studies of HPV test uptake using self-reported surveys need to take the don’t know answer into consideration because ignoring it may lead to unreliable estimates.