Abstract
Purpose.
The 2010 Affordable Care Act (ACA) provided millions of people with health insurance coverage and facilitated routine cancer screening by requiring insurers to cover preventive services without cost-sharing. Despite greater access to affordable cancer screening, Pap testing has declined over time. The aim of this study is to assess participation in Pap test and HPV vaccination, and adherence to guidelines as outlined by the American Cancer Society (ACS) from the 2010 ACA provision eliminating cost-sharing for preventive services.
Methods.
Using multi-year responses from the Behavioral Risk Factor Surveillance System, we examined the association between the ACA and participation in and adherence to Pap testing and HPV vaccination behaviors as set by the ACS. The sample included women aged 21–29 who completed the survey between 2008 and 2018 (every other year) and who live in 24 U.S. States (N = 37,893).
Results.
Results showed significant decreases in Pap testing rates but increases in the uptake of the HPV vaccine series for all age groups and across all demographics. Post-ACA year significantly predicted increases in HPV + Pap co-testing participation and adherence. Women with health insurance coverage were more likely to engage in both behaviors.
Conclusion.
Findings raise concerns around declines in the proportion of women receiving and adhering to Pap testing guidelines. A need exists for research to examine the role of increases in HPV vaccination uptake on decreases in Pap testing. Moreover, effective strategies should target increases in cervical cancer screening uptake among women vaccinated against HPV.
Keywords: Affordable Care Act, Cost-sharing, Human Papilloma Virus, Cervical Cancer, Cancer Screening
Introduction
According to estimates from the most recent data available from the United States (U.S.), 606,880 Americans will die from cancer (all types) during this current year alone [1]. While cervical cancer incidence rates have decreased substantially since the 1950s development of the Papanicolaou smear, or “Pap test,” cancer of the cervix is the second leading cause of cancer deaths among American women ages 20 to 29 [1]. Human Papilloma Virus (HPV) is a well-established cause of cervical cancer and the most common sexually transmitted infection in the U.S [2]. In 2006, U.S. health officials introduced a vaccine to inoculate against HPV infections and prevent several types of HPV-associated cancers. Despite the longstanding release of the HPV vaccine, U.S. physicians diagnose approximately 12,000 preventable cervical cancer cases annually [3].
Current American Cancer Society (ACS) guidelines recommend that women between ages 21 and 29 get a pap test every three years and women 30 years and older get a pap test every five years [18]. Prior to the establishment of these guidelines in 2012, the ACS recommended pap tests every year [18]. Recommendations for routine HPV vaccination of women under age 26 consist of three vaccine doses, and these guidelines have remained unchanged since the approval of the vaccine by the US Food and Drug Administration in 2006 [19]. Men were included in HPV vaccination recommendations in 2009 [19]. Adherence to guidelines for routine Pap testing combined with a scale-up of HPV vaccination rates can eliminate cervical cancer globally [4]. However, temporal trends in cervical cancer screening through Pap testing and HPV + Pap co-testing show declines over time and fall short of the Healthy People 2020 target of 93% [5–6]. Women least likely to receive a Pap test include those in their 20s, who lack a permanent source of medical care, who are un- or under-insured, and who had not visited a doctor in the last twelve months [5]. Despite decreases in Pap test screening, HPV vaccination rates have increased consistently since the introduction of the vaccine in 2006 [7]. Yet, uninsured women are less likely to receive the HPV vaccination compared to their insured peers as are women with less than a high school education [8–9].
Women’s health was a focus of the 2010 Patient Protection and Affordable Care Act (ACA), which required that most health insurers cover preventive services such as cervical cancer screening and HPV vaccination without cost-sharing [10–11]. In addition to removing cost barriers related to preventive health, the ACA drastically reduced the number of uninsured people in the U.S [10]. Since the ACA rollout, the rate of uninsured Americans continued to sharply decrease until stabilizing in the last few years for all ages, races/ethnicities, and income levels [12].
Despite post-ACA improvements in access to care and on self-reported health [13–14], existing research finds that the removal of cost-sharing did not significantly increase Pap testing rates [15–16]. Instead, research links the ACA rollout to increases in HPV vaccination rates [17]. Studies examining post-ACA changes in both Pap testing and HPV vaccination behaviors remain limited. Zhang and Sun described temporal changes in Pap test and HPV vaccination rates between 2007 and 2016 using Behavioral Risk Factor Surveillance System (BRFSS) data [18]. The researchers found that the rate of Pap tests declined while HPV vaccination rates increased. However, the association between the ACA with these inversely related health promotive behaviors remains unclear. To this day, no other research has examined adherence to the American Cancer Society’s (ACS) recommendations around Pap test and HPV vaccination independently and together.
The current article aims to fill in knowledge gaps found in the literature by analyzing annual responses about cervical cancer preventive behaviors collected by the BRFSS between 2008 and 2018 (every other year). We examined the 2010 ACA coverage mandate of preventive services among women who fall within the age range encompassing both the recommended start of routine Pap testing and completion of the HPV vaccination series, as set by the ACS. Using pre- and post-ACA data for women aged 21–29, we assessed 1) participation in Pap testing (ever had the Pap test?) and in HPV vaccination (ever had the HPV vaccine?) independently, 2) adherence to ACS guidelines (had a pap test in the last three years; had all three doses of the HPV vaccine) for both behaviors independently, and 3) concurrent participation in and adherence to both behaviors.
Methods
Participants
The current sample included BRFSS responses from women (N = 37,893) aged 21–29 living in one of 24 U.S. states between 2008 and 2018. Table 1 shows a listing of states used in this analysis by BRFSS year.
Table 1.
Listing of states opting into as king HPV vaccination questions by BRFSS year.
| 2008 | 2010 | 2012 | 2014 | 2016 | 2018 | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| State | N | State | N | State | N | State | N | State | N | State | N |
| DE | 168 | CT | 165 | AL | 314 | AL | 296 | AL | 278 | AL | 190 |
| MN | 133 | MA | 416 | AZ | 190 | DE | 155 | CT | 223 | CT | 239 |
| OK | 354 | RI | 159 | CT | 309 | GA | 153 | HI | 276 | HI | 290 |
| PA | 231 | WV | 126 | DE | 235 | IN | 303 | MO | 160 | MS | 246 |
| TX | 437 | WY | 144 | ME | 144 | MA | 355 | NE | 559 | MO | 150 |
| WV | 154 | MA | 702 | MN | 607 | NC | 236 | N.T | 94 | ||
| TX | 179 | RI | 162 | SC | 338 | TN | 173 | ||||
| WV | 205 | WY | 91 | SD | 161 | TX | 371 | ||||
| 272 | |||||||||||
Study Design and Setting
This cross-sectional study used responses from the Centers for Disease Control and Prevention (CDC) annual Behavioral Risk Factor Surveillance System (BRFSS) survey to assess cervical cancer prevention behaviors. The BRFSS is an annual telephone-based survey of non-institutionalized adults (aged 18 years or over) living in the U.S. or associated territories that collects data about health risk behaviors and use of preventive services. Health researchers have increasingly used the survey system’s state and local level data as a tool for surveillance of health conditions and for the building of health promotive behaviors. The BRFSS collected responses about cervical cancer prevention behaviors (i.e., Pap test and HPV vaccine) in 2008, 2010, 2012, 2014, 2016, and 2018.
All states collected responses for Pap test screening behaviors as part of the core module, but only a subset collected HPV vaccination responses as part of the optional HPV module. Across all years of data collection, 3,655 respondents (3.6%) answered both HPV vaccination questions (described below). Twenty-four of 50 states collected HPV vaccination responses for at least one of the six BRFSS years examined (see Table 1). Thus, unless otherwise noted, all analyses presented involve responses from the 24 states to allow for appropriate comparisons of both preventive behaviors, separately and together.
Dependent Variables
The dependent variables consisted of responses to the following four BRFSS questions about cervical cancer prevention behaviors: 1) “A Pap test is a test for cancer of the cervix. Have you ever had a Pap test?” 2) “How long has it been since you had your last Pap test?” 3) “Have you ever had the Adult Human Papilloma Virus (HPV) vaccination?” 4) “How many shots did you receive?” Questions about ever having had a Pap or the HPV vaccine consisted of binary (yes/no) responses. Time since last Pap test and number of HPV shots were recoded to binary variables to indicate adherence to American Cancer Society (ACS) recommendations. Pap test recommendations by the ACS changed during our study period (2008 – 2018) in 2012 from once a year to every three years for women ages 21 to 29. However, we defined adherence to Pap test guidelines as having a Pap test within the last three years (regardless of HPV vaccination status) and adherence to HPV vaccination guidelines as receiving at least three shots before the age of 26. These represent the 2012 updated ACS guidelines for Pap test screening that resulted in lower recommended frequency from once a year to once every three years [18]. HPV vaccination guidelines have remained consistent since 2006 [19]. Thus, the guidelines outlined apply to all women in our sample regardless of BRFSS data collection year. We examined Pap test and HPV vaccination participation behaviors separately and jointly.
Independent Variables
Since the requirement to cover cervical cancer preventive services without cost-sharing went into effect in 2010, pre-ACA consisted of the BRFSS data collection year of 2008 and post-ACA consisted of every other year inclusive of 2010 through 2018. Health insurance coverage involved a binary response (yes/no). Covariates included age, race, income, education, employment status, marital status, and Metropolitan Statistical Area (MSA) status code.
Statistical Methods
We conducted all analyses using IBM SPSS statistics (version 25), a popular statistical software package able to analyze complex survey data. To account for the complex survey design of the BRFSS, we utilized the CDC weights provided in the datasets to assume representativeness of the sample. Variables were weighted using the BRFSS weighting factor (_LLCPWT) specific for each state and survey year. Chi-square tests compared pre-ACA and post-ACA cervical cancer preventive behaviors. We used Odds Ratios (ORs) calculated from binomial logistic regression models to evaluate the association between ACA timepoint, health insurance coverage, and the dependent variables of interest—Pap test and HPV vaccination participation and adherence. For all logistic regression models, we reported all ORs with (adjusted) and without (unadjusted) covariates.
Results
In that all states collected responses for Pap test behaviors but only 24 collected HPV vaccination information, we examined potential differences by comparing proportions of participation in/adherence to ACS Pap testing guidelines using responses from the 24 states and responses from the other 26 states and the District of Columbia (D.C.). Although chi-square tests comparing Pap testing participation and adherence to guidelines were statistically significant (both χ2s > 33, ps < .001), results yielded similar proportions of participation (88% for 24 states vs. 87% for 26 states + D.C.) and adherence (80% for 24 states vs. 82% for 26 states + D.C.). In fact, the direction of the pre-/post-ACA trends for Pap testing participation remained consistent when using the 24 states (reductions from pre- to post-ACA participation from 93.1 to 85.6) and when using responses from the other states and D.C. (reductions between pre-/post-ACA Pap test participation from 93.7 to 86.9). Similar results occurred between pre-/post-ACA adherence to Pap testing guidelines using the 24 states (88.8 to 78.4) compared to the other 26 states plus D.C. (89.9 to 80.5). Thus, all other results reported include responses from the 24 states that collected responses about both preventive behaviors.
Table 2 shows weighted percentages for overall sample demographics (first column) and for Pap test screening/vaccination status by pre-/post-ACA time points and across multiple demographic characteristics. As the table shows, respondents reporting ever having had the Pap test decreased between pre- and post-ACA time points for all demographic categories except for those reporting their marital status as widowed or divorced. Conversely, the proportions of respondents reporting ever having had the HPV vaccine increased between pre-/post-ACA comparisons for all descriptive variables. Respondents reporting having engaged in both preventive behaviors also increased between both time points for all demographic variables. Younger age predicted greater reductions in Pap test participation and in the conjoint participation of Pap test + HPV vaccination. Hispanic/Latina and/or non-White respondents reported greater reductions in pre-/post-ACA Pap testing and in Pap + HPV vaccination participation, and lower post-ACA increases in HPV vaccination rates than non- Hispanic White women. Single or cohabitating status, less children, higher education, active employment or student status, higher income, and urban area of residence were associated with higher post-ACA rates of participation in HPV vaccination and lower post-ACA decreases in Pap test participation.
Table 2.
Weighted Percentages of Overall Demographics and of Respondents Reporting Ever Having Had a Pap and/or HPV Vaccine by ACA
| Overall demographics | Pap Test | HPV Vac. | Pap + HPV Vac. | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Pre-ACA | Post-ACA | sign. | Pre-ACA | Post-ACA | sign. | Pre-ACA | Post-ACA | sign. | |||
| Age (Mdn = 25) | 21 | 11.3 | 78.9 | 61.9 | * | 21.3 | 50.3 | * | 7.0 | 9.3 | * |
| 22 | 11.2 | 89.5 | 68.9 | * | 14.8 | 47.5 | * | 5.0 | 9.6 | * | |
| 23 | 11.9 | 85.3 | 78.7 | * | 13.5 | 43.3 | * | 4.8 | 11.0 | * | |
| 24 | 12.2 | 93.7 | 84.5 | * | 11.2 | 41.4 | * | 4.5 | 11.2 | * | |
| 25 | 10.3 | 93.6 | 88.0 | * | 7.8 | 40.1 | * | 2.5 | 10.2 | * | |
| 26 | 9.9 | 96.0 | 91.5 | * | 3.0 | 34.8 | * | 0.9 | 9.2 | * | |
| 27 | 10.8 | 96.3 | 92.6 | * | 5.6 | 32.3 | * | 2.0 | 7.4 | * | |
| 28 | 11.1 | 96.9 | 91.2 | * | 0.8 | 30.4 | * | 0.3 | 8.7 | * | |
| 29 | 11.4 | 96.3 | 93.5 | * | 2.8 | 23.4 | * | 1.0 | 6.6 | * | |
| Hispanic/Latina | Yes | 19.7 | 93.9 | 79.1 | * | 8.1 | 30.4 | * | 4.7 | 7.9 | * |
| No | 79.8 | 91.7 | 84.2 | * | 9.2 | 41.1 | * | 2.7 | 9.6 | * | |
| Race | White | 68.0 | 92.8 | 85.5 | * | 8.9 | 40.4 | * | 2.8 | 10.1 | * |
| Black | 16.5 | 93.8 | 84.9 | * | 9.0 | 36.2 | * | 2.3 | 8.7 | * | |
| Asian | 2.8 | 69.4 | 66.4 | * | 6.5 | 27.6 | * | 3.0 | 4.7 | * | |
| Native Hawaiian | 3.1 | 71.6 | 50.2 | * | 0.0 | 46.1 | * | 0.0 | 6.1 | * | |
| American Indian | 1.2 | 94.7 | 80.3 | * | 17.8 | 30.3 | * | 6.9 | 5.5 | * | |
| Other race | 4.6 | 93.1 | 79.3 | * | 9.3 | 39.4 | * | 5.2 | 10.4 | * | |
| Marital Status | Married | 32.9 | 96.6 | 92.7 | * | 6.8 | 29.5 | * | 2.6 | 6.9 | * |
| Divorced | 3.1 | 96.2 | 96.4 | * | 2.8 | 27.2 | * | 1.1 | 7.6 | * | |
| Widowed | 0.2 | 69.4 | 92.2 | * | N/A | 38.4 | * | N/A | 11.6 | * | |
| Separated | 2.6 | 96.3 | 88.7 | * | 5.9 | 30.8 | * | 2.7 | 8.3 | * | |
| Never married | 51.0 | 85.5 | 75.8 | * | 12.6 | 43.3 | * | 3.9 | 10.5 | * | |
| Cohabitating | 9.6 | 96.9 | 87.9 | * | 7.7 | 41.3 | * | 2.8 | 11.1 | * | |
| # of children | 0 | 44.3 | 86.9 | 77.5 | * | 11.8 | 46.3 | * | 3.6 | 10.9 | * |
| 1 | 23.4 | 94.3 | 87.0 | * | 7.9 | 36.6 | * | 2.3 | 9.2 | * | |
| 2 | 19.4 | 95.6 | 90.0 | * | 6.1 | 29.1 | * | 2.5 | 7.4 | * | |
| 3+ | 12.3 | 94.4 | 87.5 | * | 9.5 | 26.1 | * | 4.2 | 6.4 | * | |
| Education | High school or less | 36.1 | 92.1 | 82.0 | * | 6.8 | 28.8 | * | 27.0 | 7.0 | * |
| Some college or more | 63.5 | 92.1 | 83.7 | * | 10.6 | 44.0 | * | 3.4 | 10.6 | * | |
| Employment | Employed for wages | 54.4 | 93.8 | 85.6 | * | 9.7 | 40.4 | * | 3.0 | 10.3 | * |
| Self-employed | 3.9 | 88.2 | 82.5 | * | 10.9 | 35.0 | * | 4.4 | 8.6 | * | |
| Unemployed | 9.7 | 91.4 | 81.8 | * | 10.9 | 34.8 | * | 3.0 | 7.6 | * | |
| Homemaker | 13.3 | 93.6 | 91.2 | * | 6.6 | 22.8 | * | 3.4 | 5.5 | * | |
| Student | 14.5 | 82.7 | 68.5 | * | 9.6 | 48.4 | * | 2.9 | 10.5 | * | |
| Unable to work | 3.1 | 91.9 | 84.9 | * | 4.8 | 28.1 | * | 1.6 | 7.4 | * | |
| Income level | < $25K | 33.2 | 92.4 | 82.7 | * | 6.1 | 34.7 | * | 2.4 | 8.1 | * |
| < $50K | 23.3 | 93.4 | 86.0 | * | 11.5 | 40.4 | * | 3.8 | 9.5 | * | |
| < $75K | 11.8 | 94.2 | 87.6 | * | 7.9 | 40.1 | * | 2.8 | 10.0 | * | |
| $75K or > | 14.3 | 92.2 | 86.8 | * | 11.8 | 50.2 | * | 3.7 | 14.1 | * | |
| Metro Status | Center city (CC) | 17.0 | 89.7 | 85.2 | * | 10.8 | 38.2 | * | 4.9 | 6.6 | * |
| Outside CC but in metro | 10.2 | 91.4 | 86.9 | * | 7.1 | 31.3 | * | 2.1 | 9.8 | * | |
| Suburb | 6.7 | 94.5 | 86.9 | * | 13.6 | 41.1 | * | 2.4 | 6.6 | * | |
| Non-metro area | 9.7 | 95.4 | 87.4 | * | 3.2 | 30.2 | * | 1.1 | 4.8 | * | |
Note.
= chi-square p-value < .001.
We calculated weighted percentages of the uptake of both preventive behaviors for women with health insurance coverage who also regularly see a doctor (routine check-up within last year). This subset of women (N = 20,798) had high access to both preventive behaviors and, therefore, represent a group with highest potential for seeking cervical cancer screening and HPV vaccination uptake due to the ACA. A calculation of differences between weighted percentages of Pap test participation between 2008 (pre-ACA baseline year) and each subsequent year resulted in an average difference of 5.4%, with the largest downward discrepancy between 2008 (95%) and post-ACA year 2016 (84%) at −11%. An inverse pattern of differences occurred for HPV vaccine participation, with the average difference at 15% and the largest discrepancy between 2008 (13.9%) and 2014 (48.6%) at 35%.
Figure 1 shows the pre-/post-ACA trends by participation in Pap testing and in HPV vaccination and adherence to established ACS guidelines set forth for both behaviors. As the figure shows, participation in and adherence to the behaviors decreased post-ACA for Pap testing but increased for HPV vaccination uptake. We compared levels of participation in and adherence to both behaviors using chi-square analyses examining differences between 2008 and each subsequent post-ACA year (see Table 3). Results show substantial reductions in Pap participation/adherence between pre- and post-ACA time points, with the downward trends becoming larger for later years. Results show the opposite trend for HPV vaccination rates (significant increases in participation and adherence over time).
Figure 1:
Pap test and HPV vaccination participation and adherence by ACA time point.
Table 3.
Comparisons of Cervical Cancer Screening Behaviors by BRFSS Year.
| Pap participation |
HPV vac. participation |
||||||
|---|---|---|---|---|---|---|---|
| Year | Yes N(%) | No7V(%) | P | Yes N(%) | No7V(%) | P | |
| Baseline | 2008 | 6123 (92) | 382(8) | 102 (8.9) | 1375 (91.1) | ||
| Comparison | 2010 | 4695 (90.2) | 389(9.8) | < .001 | 219(25.7) | 791 (74.3) | < .001 |
| 2012 | 7516(86.6) | 887 (13.4) | < .001 | 847 (29.7) | 1431 (70.3) | < .001 | |
| 2014 | 6237 (82.6) | 984(17.4) | < .001 | 993 (43.2) | 1129(56.8) | < .001 | |
| 2016 | 5387 (78.5) | 1120(21.5) | < .001 | 1049 (40.4) | 1454(59.6) | < .001 | |
| 2018 | 1845 (77) | 405 (23) | < .001 | 844 (43.5) | 909 (56.5) | < .001 | |
| Pap adherence |
HPV vac. adherence |
||||||
| Year | Yes N(%) | No7V(%) | P | Yes N(%) | No7V(%) | P | |
| Baseline | 2008 | 5873 (88.7) | 632(11.3) | 35 (2.4) | 1416(97.6) | ||
| Comparison | 2010 | 4485 (85.4) | 599(14.6) | < .001 | 158 (18.7) | 845 (81.3) | < .001 |
| 2012 | 7134(81.7) | 1269(18.3) | < .001 | 543 (19.6) | 1629 (80.4) | < .001 | |
| 2014 | 5856(77.7) | 1365 (22.3) | < .001 | 674 (29) | 1383 (71) | < .001 | |
| 2016 | 4988 (72.3) | 1519(27.7) | < .001 | 639 (22.1) | 1763 (77.9) | < .001 | |
| 2018 | 1730(63.4) | 725 (36.6) | < .001 | 497 (57.4) | 347 (42.6) | < .001 | |
Note . Chi-square tests were used to assess statistical significance.
Note . Percentages in parentheses are weighted using the BRFSS final design weighting variable.
Table 4 shows results from six logistic regression models for Pap test and HPV vaccine participation and adherence (separately), and Pap testing + HPV vaccination participation and adherence (combined). Post-ACA year (compared to 2008 as the pre-ACA baseline) and health insurance coverage were included as the main predictors. We report Odds Ratios (ORs) for models that included multiple demographic variables as covariates (adjusted OR) and for models excluding the covariates (unadjusted OR). Fit indices for all models showed adequate fit. First, whereas health insurance coverage (adjusted OR = 1.8) was positively associated with Pap test participation, post-ACA year (compared to baseline) negatively predicted Pap test participation for all years (all adjusted ORs between .95 and .31). Similar findings resulted from the model examining Pap test adherence in that health insurance coverage (adjusted OR = 2.3) and post-ACA year (except 2010; all adjusted ORs between .65 and .24) positively and negatively predicted adherence to Pap test screening recommendations, respectively. For HPV vaccine participation/adherence, health insurance coverage (participation adjusted OR = 2.0; adherence adjusted OR = 2.1) and post-ACA year (all adjusted ORs between 3.0 and 123.8) were positively associated with both outcome variables. Similarly, participation in and adherence to both preventive behaviors, combined, were positively associated with health insurance coverage (participation adjusted OR = 1.7; adherence adjusted OR = 2.2) and post-ACA year (all adjusted ORs between 2.6 and 71.6). To note, the combined Pap test + HPV vaccine participation OR for the 2010 post-ACA year was 0.89, but the association remained positive and statistically significant. This result suggests that the reduction in odds of conjoint participation in both behaviors in 2010 may have been the result of the BRFSS weighting factor suppressing the OR.
Table 4.
Binomial Logistic Regression models of Pop Test Participation and Adherence by Independent Variables of Interest
| Wald’s | Unadjusted | Adjusted | Omnibus X2 | |||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Model | Predictor | b | SE b | X2 | df | OR | OR | 95% CI | −2 LL | (df, p-value) |
| 1. Pap participation | Constant | 1.7 | 0.2 | 106.0 | 1 | 8.46 | 5.36 | 6522.5 | 1098.2 | |
| Year 2008 | 108.1 | 5 | (37, <.001) | |||||||
| 2010 | −0.2 | 0.1 | 3.3 | 1 | 0.81 | 0.95 | (0.94, 0.96) | |||
| 2012 | −0.5 | 0.1 | 24.0 | 1 | 0.56 | 0.61 | (0.60, 0.61) | |||
| 2014 | −0.8 | 0.1 | 55.0 | 1 | 0.41 | 0.39 | (0.38, 0.39) | |||
| 2016 | −1.0 | 0.1 | 60.8 | 1 | 0.30 | 0.31 | (0.30, 0.31) | |||
| 2018 | −1.1 | 0.3 | 15.1 | 1 | 0.28 | 0.32 | (0.31, 0.32) | |||
| Health insurance coverage | 0.7 | 0.1 | 78.6 | 1 | 1.61 | 1.81 | (1.80, 1.82) | |||
| 2. Pap adherence | Constant | 1.0 | 0.1 | 49.6 | 1 | 4.93 | 2.81 | 9618.0 | 1020.1 | |
| Year 2008 | 163.3 | 5 | (37, <.001) | |||||||
| 2010 | −0.1 | 0.1 | 2.7 | 1 | 0.76 | 0.80 | (0.79, 0.80) | |||
| 2012 | −0.4 | 0.1 | 22.1 | 1 | 0.57 | 0.65 | (0.65, 0.65) | |||
| 2014 | −0.7 | 0.1 | 69.9 | 1 | 0.43 | 0.42 | (0.42, 0.42) | |||
| 2016 | −1.1 | 0.1 | 103.6 | 1 | 0.30 | 0.24 | (0.24, 0.25) | |||
| 2018 | −1.1 | 0.2 | 19.4 | 1 | 0.21 | 0.39 | (0.39, 0.40) | |||
| Health insurance coverage | 1.0 | 0.1 | 273.3 | 1 | 2.03 | 2.33 | (2.32, 2.34) | |||
| 3. Vaccine participation | Constant | −3.1 | 0.3 | 141.7 | 1 | 0.06 | 0.05 | 3530.1 | 924.2 | |
| Year 2008 | 343.9 | 5 | (37, <.001) | |||||||
| 2010 | 1.1 | 0.1 | 58.2 | 1 | 3.11 | 3.05 | (3.02, 3.07) | |||
| 2012 | 1.9 | 0.1 | 187.9 | 1 | 4.16 | 5.27 | (5.22, 5.32) | |||
| 2014 | 2.4 | 0.2 | 249.6 | 1 | 7.10 | 8.57 | (8.49, 8.65) | |||
| 2016 | 2.2 | 0.2 | 160.1 | 1 | 6.42 | 9.95 | (9.84, 10.06) | |||
| 2018 | 2.8 | 0.2 | 133.5 | 1 | 7.51 | 13.58 | (13.41, 13.74) | |||
| Health insurance coverage | 0.7 | 0.1 | 22.9 | 1 | 2.21 | 2.00 | (1.98, 2.02) | |||
| 4. Vaccine adherence | Constant | −4.6 | 0.3 | 179.0 | 1 | 0.01 | 0.00 | 2644.8 | 825.8 | |
| Year 2008 | 239.1 | 5 | (37, <.001) | |||||||
| 2010 | 1.8 | 0.2 | 66.4 | 1 | 7.96 | 7.01 | (6.91, 7.10) | |||
| 2012 | 2.5 | 0.2 | 139.5 | 1 | 9.37 | 11.84 | (11.69, 12) | |||
| 2014 | 2.9 | 0.2 | 168.9 | 1 | 14.75 | 20.02 | (19.74, 20.30) | |||
| 2016 | 2.7 | 0.2 | 128.7 | 1 | 10.37 | 14.83 | (14.59, 15.07) | |||
| 2018 | 4.1 | 0.4 | 130.0 | 1 | 50.07 | 123.75 | (121.36, 126.18) | |||
| Health insurance coverage | 0.6 | 0.2 | 12.2 | 1 | 2.63 | 2.11 | (2.09, 2.14) | |||
| 5. Pap + Vaccine participation | Constant | −5.3 | 0.2 | 463.7 | 1 | 0.02 | 0.02 | 5644.1 | 1253.1 | |
| Year 2008 | 579.6 | 5 | (37, <.001) | |||||||
| 2010 | 1.1 | 0.1 | 68.6 | 1 | 0.85 | 0.89 | (0.89, 0.90) | |||
| 2012 | 2.1 | 0.1 | 268.6 | 1 | 2.33 | 3.57 | (3.54, 3.60) | |||
| 2014 | 2.5 | 0.1 | 328.8 | 1 | 2.97 | 4.79 | (4.75, 4.83) | |||
| 2016 | 2.6 | 0.2 | 258.7 | 1 | 3.52 | 6.38 | (6.32, 6.45) | |||
| 2018 | 4.1 | 0.2 | 295.5 | 1 | 7.55 | 26.16 | (25.87, 26.46) | |||
| Health insurance coverage | 0.8 | 0.1 | 37.8 | 1 | 1.94 | 1.66 | (1.65, 1.68) | |||
| 6. Pap + Vaccine adherence | Constant | −6.9 | 0.3 | 403.6 | 1 | 0.00 | 0.00 | 3912.4 | 1013.1 | |
| Year 2008 | 323.4 | 5 | (37, <.001) | |||||||
| 2010 | 1.8 | 0.2 | 76.1 | 1 | 2.25 | 2.60 | (2.57, 2.63) | |||
| 2012 | 2.7 | 0.2 | 181.1 | 1 | 5.28 | 8.68 | (8.57, 8.79) | |||
| 2014 | 3.0 | 0.2 | 196.3 | 1 | 6.61 | 11.22 | (11.07, 11.36) | |||
| 2016 | 3.0 | 0.2 | 170.8 | 1 | 6.24 | 10.62 | (10.45, 10.78) | |||
| 2018 | 4.2 | 0.3 | 189.8 | 1 | 15.17 | 71.60 | (70.51, 72.70) | |||
| Health insurance coverage | 0.9 | 0.2 | 27.0 | 1 | 2.73 | 2.15 | (2.12, 2.18) | |||
Note. 2008 represents baseline year compared to all others.
Note. The following variables were included in the model but not reported in main table because they were treated as covariates: are, race, income, education, employment status, marital status, metropolitan status code.
Discussion
The aim of this study was to assess participation in Pap testing and in HPV vaccination independently and together among a nationally representative sample of U.S. women aged 21–29 before and after the 2010 implementation of the ACA. We also examined adherence to the ACS guidelines about the uptake of these two preventive behaviors, both separately and jointly. Through the analysis of six years of BRFSS survey data (pre- ACA year: 2008; post-ACA years: 2010, 2012, 2014, 2016, 2018), we found that health insurance coverage significantly predicted participation in and adherence to both behaviors. Results also showed that the ACA negatively predicted Pap testing rates (lower participation and adherence). In fact, comparisons between the baseline year (2008) and each subsequent post-ACA year yielded consistently lower odds of engaging in Pap testing despite post-ACA increases in the number of insured people. The inverse was true for HPV vaccination uptake and adherence in that post-ACA years were associated with higher odds of getting vaccinated and adhering to guidelines. Analysis of both behaviors, together, showed that post-ACA years were associated with higher participation in/adherence to both behaviors. However, steep increases in the uptake of HPV vaccination drove increases in the conjoined engagement of Pap testing and HPV vaccination.
Findings provide evidence that participation and adherence to both cervical cancer preventive behaviors increased after the removal of cost-sharing under the ACA among U.S. women aged 21–29. However, these findings also raise concerns around the decline in the proportion of women receiving Pap testing. The ACS recommends that women in our sample seek Pap testing every three years regardless of HPV vaccination uptake. Compared to the pre-ACA year (2008), post-ACA years significantly predicted lower likelihood of Pap test participation and adherence to ACS guidelines around Pap test frequency after controlling for health insurance coverage and other demographic variables. This finding is further supported by other research that has shown declines in Pap testing after women receive the HPV vaccine [20–21]. Future research should explore the reasons why rates of Pap testing have fallen recently as HPV vaccination rates have risen sharply. Our findings underscore the need for active measures to ensure greater uptake of all cervical cancer screening/preventive measures (i.e., Pap testing and HPV vaccination).
We found that greater access to health insurance coverage or the removal of cost barriers to preventive services by the ACA did not sufficiently increase cervical cancer screening as intended. Moreover, all ethnic/racial groups consistently exhibited reductions in Pap test participation and increases in HPV vaccination rates, but some groups reported lower participation in both behaviors because of the ACA. Particularly, Hispanic women and racial minorities exhibited lower rates of participation in both preventive behaviors after the rollout of the ACA. These findings suggest that racial/ethnic disparities exist in the participation of preventive services despite greater access to health insurance from the ACA and the elimination of cost-sharing for preventive services. Research in the future should further explore barriers faced by racial/ethnic minority groups that promote such disparities.
Findings in this study are bounded by several limitations. First, this is a cross-sectional study. Thus, we cannot analyze cervical cancer preventive behaviors longitudinally or interpret causation. Second, current findings do not account for more than a binary measurement of health insurance coverage (yes/no) and cannot differentiate between public versus private health insurance coverage or account for access issues to care. Third, while our findings are based on a large phone-based health survey that included women 21–29 across the U.S., respondents must have had a telephone to participate (i.e., landline and/or cell phone from 2011 onward), which may have biased the sample. Fourth, Pap testing and HPV vaccination information was self-reported and subject to recall bias and under- or over-reporting. The data cannot be verified with medical records and/or health care professionals. Fifth, data are limited to responses from 24 states that elected to ask the optional HPV vaccination question module in at least one year included in analyses, therefore findings might not be nationally representative. However, it is important to note that the states included are geographically and politically diverse. Sixth, only a small subset of states opted into asking the HPV module, so findings about HPV vaccination rates may not be nationally representative.
This study evaluated the association between the removal of cost-sharing under the ACA in 2010 and Pap test and HPV vaccination participation and guideline adherence. It is possible that other factors may explain changes in cervical cancer prevention behaviors. For example, it is possible that increases in HPV vaccination are attributable to increases in public acceptance and uptake over time as demonstrated with other vaccines (e.g., Hepatitis B) [22]. It is also possible that other provisions of ACA legislation, such as dependent coverage extension up to age 26 (effective in 2010) and Medicaid expansion (rolled out 2014) influenced pap testing and HPV vaccinations. In regard to general acceptance of the HPV vaccine explaining increases over time, cost or insurance coverage of the vaccine was a commonly cited barrier to HPV vaccination among young women prior to the implementation of the ACA [23–24]. Previous studies have shown that the ACA significantly increased HPV vaccination rates [25]. Regarding the other provisions of the ACA, it is possible that policies beyond the removal of cost-sharing in 2010 resulted in increases in HPV vaccinations since the number of insured Americans increased over time alongside greater access to care [26]. However, such explanation does not clarify the observed declining trend in pap testing participation and adherence since we expected to see increases because of greater insurance coverage and access to care.
Conclusions
The removal of cost-sharing under the ACA was associated with an increase in HPV vaccination participation and adherence, a decrease in Pap testing participation and adherence, and an increase in participation and adherence of both behaviors together. Findings raise concerns around declines in the proportion of women receiving and adhering to Pap testing guidelines. A need exists for research to examine the role of increases in HPV vaccination uptake on decreases in Pap testing. Future research should also focus on effective strategies for increasing cervical cancer screening uptake among women vaccinated against HPV. Other studies should explore factors associated with both HPV vaccine and pap testing guideline adherence given that the goal is to ensure that women receive the HPV vaccine and pap testing together.
Acknowledgments
Funding
This work was supported by National Cancer Institute Grant T32CA078447 for all authors.
Footnotes
Ethics approval
The data presented on this article represents publicly available, retrospectively collected survey responses that were de-identified. The Internal Review Board of the University of Arizona confirmed that no ethics approval is required.
Consent
Not applicable in that this research is based on publicly available, retrospective data.
Data and/or code availability
The annual survey data can be obtained from the Center of Disease Control’s Behavioral Risk Factor Surveillance System’s website found through the following link: https://www.cdc.gov/brfss/annual_data/annual_data.htm
This Author Accepted Manuscript is a PDF file of an unedited peer-reviewed manuscript that has been accepted for publication but has not been copyedited or corrected. The official version of record that is published in the journal is kept up to date and so may therefore differ from this version.
Conflicts of interest/competing interests
The authors declare no potential conflicts of interest or competing interests.
References
- 1.Siegel RL, Miller KD, Jemal A (2019) Cancer statistics, 2019. CA Cancer J Clin 69:7–34. doi: 10.3322/caac.21551 [DOI] [PubMed] [Google Scholar]
- 2.Bosch FX, Lorincz A, Munoz N, Meijer CJ, Shah KV (2002) The causal relation between human papillomavirus and cervical cancer. J Clin Pathol 55:244–65. doi: 10.1136/jcp.55.4.244 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Center for Disease Control and Prevention (2014) Cervical cancer is preventable. https://www.cdc.gov/vitalsigns/cervical-cancer/index.html. Accessed 9 March 2020
- 4.Simms KT, Steinberg J, Caruana M, et al. (2019) Impact of scaled up human papillomavirus vaccination and cervical screening and the potential for global elimination of cervical cancer in 181 countries, 2020–99: a modelling study. Lancet Oncol 20:394–407. doi: 10.1016/S1470-2045(18)30836-2 [DOI] [PubMed] [Google Scholar]
- 5.Hall IJ, Tangka FKL, Sabatino SA, Thompson TD, Graubard BI, Breen N (2018) Patterns and Trends in Cancer Screening in the United States. Prev Chronic Dis 15:E97. doi: 10.5888/pcd15.170465 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Watson M, Benard V, Flagg EW (2018) Assessment of trends in cervical cancer screening rates using healthcare claims data: United States, 2003–2014. Prev Med Rep 9:124–30. doi: 10.1016/j.pmedr.2018.01.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Walker TY, Elam-Evans LD, Singleton JA, et al. (2017) National, regional, state, and selected local area vaccination coverage among adolescents aged 13–17 Years - United States. MMWR Morb Mortal Wkly Rep 66:874–82. doi: 10.15585/mmwr.mm6833a2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Lu PJ, Yankey D, Jeyarajah J, et al. (2018) Association of health insurance status and vaccination coverage among adolescents 13–17 years of age. J Pediatr 195:256–62 e1. doi: 10.1016/j.jpeds.2017.12.024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Sauer AG, Siegel RL, Jemal A, Fedewa SA (2017) Updated review of prevalence of major risk factors and use of screening tests for cancer in the United States. Cancer Epidemiol Biomarkers Prev 26:1192–208. doi: 10.1158/1055-9965.EPI-18-1169 [DOI] [PubMed] [Google Scholar]
- 10.The Henry J Kaiser Family Foundation (2013) Summary of the Affordable Care Act. https://www.kff.org/health-reform/fact-sheet/summary-of-the-affordable-care-act/. Accessed 9 March 2020 [Google Scholar]
- 11.French MT, Homer J, Gumus G, Hickling L (2016) Key provisions of the patient protection and Affordable Care Act (ACA): A systematic review and presentation of early research findings. Health Serv Res 51:1735–71. doi: 10.1111/1475-6773.12511 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Berchick ER, Hood E, Barnett JC (2018) Health insurance coverage in the United States. U.S. Government Printing Office; https://www.census.gov/content/dam/Census/library/publications/2018/demo/p60-264.pdf Accessed 9 March 2020 [Google Scholar]
- 13.Alcala HE, Chen J, Langellier BA, Roby DH, Ortega AN (2017) Impact of the Affordable Care Act on health care access and utilization among Latinos. J Am Board Fam Med 30:52–62. doi: 10.3122/jabfm.2017.01.160208 [DOI] [PubMed] [Google Scholar]
- 14.Sommers BD, Gunja MZ, Finegold K, Musco T (2015) Changes in self-reported insurance coverage, access to care, and health under the Affordable Care Act. JAMA 314:366–74. doi: 10.1001/jama.2015.8421 [DOI] [PubMed] [Google Scholar]
- 15.Alharbi A, Khan MM, Horner R, Brandt H, Chapman C (2019) Impact of removing cost sharing under the affordable care act (ACA) on mammography and pap test use. BMC Public Health 19:370. doi: 10.1186/s12889-019-6665-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Han X, Robin Yabroff K, Guy GP, Zheng Z, Jemal A (2015) Has recommended preventive service use increased after elimination of cost-sharing as part of the Affordable Care Act in the United States? Prev Med 78:85–91. doi: 10.1016/j.ypmed.2015.07.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Corriero R, Gay JL, Robb SW, Stowe EW (2018) Human papillomavirus vaccination uptake before and after the Affordable Care Act: Variation according to insurance status, race, and education (NHANES 2006–2014). J Pediatr Adolesc Gynecol 31:23–7. doi: 10.1016/j.jpag.2017.07.002 [DOI] [PubMed] [Google Scholar]
- 18.American Cancer Society (2016) The American Cancer Society guidelines for the prevention and early detection of cervical cancer. https://www.cancer.org/cancer/cervical-cancer/detection-diagnosis-staging/cervical-cancer-screening-guidelines.html. Accessed 9 March 2020
- 19.Saslow D, Andrews KS, Manassaram-Baptiste D, et al. (2016) Human papillomavirus vaccination guideline update: American Cancer Society guideline endorsement. CA Cancer J Clin 66:375–85. doi: 10.3322/caac.21355 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Terk D, Bonham A, Caveglia S, Li D, Prast B, Day S (2017) Adherence to recommended cervical cancer screening based on human papillomavirus vaccination status. Obstet Gynecol 129:6S. doi: 10.1097/01 [DOI] [Google Scholar]
- 21.Paynter CA, Van Treeck BJ, Verdenius I, et al. (2015) Adherence to cervical cancer screening varies by human papillomavirus vaccination status in a high-risk population. Prev Med Rep 2:711–6. doi: 10.1016/j.pmedr.2015.07.011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Center for Disease Control and Prevention (2002) Achievements in public health: Hepatitis B vaccination--United States, 1982–2002. MMWR Morb Mortal Wkly Rep 51:549–552, 563. [PubMed] [Google Scholar]
- 23.Conroy K, Rosenthal SL, Zimet GD, et al. (2009) Human papillomavirus vaccine uptake, predictors of vaccination, and self-reported barriers to vaccination. J Womens Health (Larchmt) 18:1679–86. doi: 10.1089/jwh.2008.1329 [DOI] [PubMed] [Google Scholar]
- 24.Zimet GD, Weiss TW, Rosenthal SL, Good MB, Vichnin MD (2010) Reasons for non- vaccination against HPV and future vaccination intentions among 19–26 year-old women. BMC Womens Health 10:27. doi: 10.1186/1472-6874-10-27 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Lipton BJ, Decker SL (2015) ACA provisions associated with increase in percentage of young adult women initiating and completing the HPV vaccine. Health Aff (Millwood) 34:757–64. doi: 10.1377/hlthaff.2014.1302 [DOI] [PMC free article] [PubMed] [Google Scholar]