Abstract
Background
Health reform, including Medicaid expansion, is increasing insurance coverage and financial access to breast and cervical cancer screening for low-income women, although services for low-income uninsured women are still needed.
Methods
American Community Survey and administrative data about Medicaid and health insurance enrollment are used to estimate the number of low-income women who will be uninsured in 2017, focusing on the age ranges 21 to 64, 40 to 64, and 50 to 64.
Results
Assuming that 29 states expand Medicaid (as of June 2015), the national percentage of low-income women 21 to 64 who are uninsured will fall from 32.2% in 2013 to 14.6% by 2017. Among Medicaid-expanding states, the percentage of uninsured will decrease from 28.7% to 8.0%, whereas in non-expanding states, the level will decrease from 36.9% to 23.3%. About 5.7 million women 21 to 64 and 2.6 million women 40 to 64 will remain uninsured in 2017. The size of the uninsured low-income population will remain much larger than the 659,000 women who have previously received Pap tests and 548,000 obtaining mammograms under the National Breast and Cervical Cancer Early Detection Program in 2013.
Discussion
Even before 2014, women living in states that are not expanding Medicaid were less likely to get mammograms and Pap tests than women in expanding states. Affordable Care Act–related insurance expansions will lower financial barriers to screening and should boost overall screening rates. But disparities in insurance coverage and cancer screening across Medicaid-expanding and non-expanding states could widen.
Conclusions
Programs to support cancer screening for low-income uninsured women will still be needed.
Over the past decade cervical cancer screening rates in the United States decreased and breast cancer screening rates remained flat (
National Cancer Institute, 2015
). Implementation of the Affordable Care Act (ACA) is decreasing the number of uninsured Americans and should substantially expand access to cancer screening. The number of uninsured Americans decreased sharply in 2014, after health insurance marketplaces and Medicaid expansions began, with greater reductions in states that have expanded Medicaid (National Cancer Institute. (2015). Early detection. Cancer trends progress report. Available: http://progressreport.cancer.gov/detection. Accessed: September 23, 2015.
Cohen and Martinez, 2015
, Smith and Medalia, 2015
). The ACA also required most insurance plans to cover breast and cervical cancer screening without cost sharing. These changes can reduce substantially financial barriers and increase the demand for cancer screening.These changes should increase early detection and treatment of cancers and could lead to improved outcomes (
Council of Economic Advisors, the White House, 2015
). Studies of insurance expansions in Oregon and Massachusetts found increased breast and cervical cancer screening as a consequence of insurance expansions (Council of Economic Advisors, the White House. (2015). Missed opportunities: The consequences of state decisions not to expand Medicaid. June 2015. Washington, DC: White House. Retrieved from https://www.whitehouse.gov/sites/default/files/docs/missed_opportunities_medicaid.pdf.
Baicker et al., 2013
, Sabik and Bradley, 2015
, - Sabik L.M.
- Bradley C.J.
The impact of near-universal insurance coverage on breast and cervical cancer screening: Evidence from Massachusetts.
Health Economics. 2015; ([Epub ahead of print])https://doi.org/10.1002/hec. 3159
Finkelstein et al., 2011
). Cancer patients residing in counties with fewer uninsured had earlier detection and longer survival times (Smith et al., 2013
).Even so, millions of low-income women will remain uninsured and face financial barriers to screening. The
Congressional Budget Office, 2015
estimates that the ACA will lower the number of uninsured Americans by 24 million by 2017, but 27 million people will remain uninsured and without an affordable health care coverage option for a variety of reasons, including that many states are not expanding Medicaid, some eligible people do not participate in Medicaid or health insurance marketplaces, and some, such as undocumented immigrants, are not eligible for assistance. Many low-income people are exempt from the mandate to have health insurance or are not even aware of it (Congressional Budget Office. (2015). Insurance coverage provisions of the Affordable Care Act—CBO's March 2015 baseline table. Available: www.cbo.gov. Accessed: March 2015.
Karpman et al., 2015
). Of course, insurance coverage is no guarantor that people will seek or receive screening: they may not be aware of the importance of screening, not receive recommendations or referrals from health professionals, lack transportation, or encounter language barriers.A key public health program to improve cancer screening is the Centers for Disease Control and Prevention's (CDC's) National Breast and Cervical Cancer Early Detection Program (NBCCEDP), which supports screening and diagnostic services for low-income uninsured or underinsured women, as well as outreach, education, and navigation services. Targeting to the uninsured is particularly important since being uninsured is a leading barrier to screening (
American Cancer Society, 2015
). Evidence indicates NBCCEDP contributes to reduced breast cancer death rates (Howard et al., 2010
), decreases time from cancer diagnosis to Medicaid enrollment, expands women's treatment options (Adams et al., 2009
), and improves the timing of diagnosis and treatment of cancer (Lantz and Soliman, 2009
, Richardson et al., 2010
).In this paper, we estimate the number of low-income women who will gain insurance coverage by 2017 and those who will remain uninsured depending on whether a state expands Medicaid or not. “Low-income” is defined as family income at or below 250% of the federal poverty level (FPL), which is the federal income criterion for NBCCEDP. The target population for cervical cancer screening is women 21 to 64 and for breast cancer screening is women 40 to 64, with women 50 to 64 considered a priority population. Virtually all women 65 or older are insured, at least through Medicare.
This report updates an earlier paper that estimated health reform-related changes in insurance coverage for women in 2014 (
Levy et al., 2012
). A key difference is that this paper includes estimates related to whether a state decides to expand Medicaid. The earlier analysis was developed before the Supreme Court's 2012 ruling in NFIB v Sebelius (Rosenbaum and Westmoreland, 2012
), which gave states the option to expand Medicaid. It assumed the Medicaid expansion was required in all states, based on the original intent of the ACA. As of June 2015, 29 states were expanding Medicaid; the rest were not or were still considering the issue (Kaiser Family Foundation, 2015
). States may alter their plans in the future; more may choose to expand Medicaid and some that have already expanded could reverse course. Other difference in this update are technical in nature, using more recent data.Kaiser Family Foundation. (2015) Status of state action on the Medicaid expansion decision, as of March 6, 2015. Available: http://kff.org/health-reform/state-indicator/state-activity-around-expanding-medicaid-under-the-affordable-care-act/. Accessed: March 15, 2015.
This paper estimates the number of women who will be uninsured in 2017, both with and without a Medicaid expansion. The insights about changes in insurance coverage under health care reform can provide insights into the changes in the demand for cancer screening in the near future and improve policy planning to help ensure that CDC's program is addressing current and future needs (
Plescia et al., 2014
).Methods
The estimation approach in this report is adapted from the methods described in
Levy et al., 2012
and August et al., 2016
with some important modifications. A key concept in these papers is that the ACA was largely modeled on Massachusetts' 2006 health reform law (Gruber, 2011
, Holtz-Eakin, 2011
), so recent coverage for Massachusetts residents can be used to estimate coverage for residents of other states.This model uses data about women 18 to 64 from the 2013 American Community Survey Public Use Microdata Sample, which surveys about 3 million people with response rates of greater than 90% (
Census Bureau, 2015
). Our models included data about health insurance status, race/ethnicity, marital status, having children, employment status, industry of employment, poverty status, citizenship status, disability, and education.Census Bureau. (2015) American Community Survey. Data: response rates. Available: http://www.census.gov/acs/www/methodology/sample-size-and-data-quality/response-rates/index.php. Accessed: March 25, 2015.
We constructed weighted multivariate logit models of health insurance status in Massachusetts. We then applied the model coefficients from Massachusetts to demographic and economic characteristics of the American Community Survey Public Use Microdata Sample respondents in all 50 states and the District of Columbia to predict individual-level probabilities of being insured under health reform. We project forward from 2013 to 2017 by modifying survey weights to account for expected population growth and shifts in the age distribution between 2013 and 2017, based on Census projections (
Census Bureau, Population Division, 2014
).We recognize that states differ from Massachusetts in many ways and calibrated estimates to account for state-specific differences in policies, ACA implementation efforts, market characteristics, and other state traits. We adjusted our estimates to incorporate more recent data available from administrative sources. We used state counts of persons receiving tax credits in health insurance marketplaces as of February 2015 (
Office of the Assistant Secretary for Planning and Evaluation, HHS, 2015
) and changes in Medicaid enrollment between late 2013 and December 2014 (Centers for Medicare and Medicaid Services, 2015
). These were adjusted to account for the estimated share of marketplace and Medicaid enrollees who were non-elderly adults with incomes at or below 250% FPL. Other adjustments account for the share of Medicaid and marketplace enrollees who might otherwise have been insured privately and expected growth in the number of marketplace and Medicaid enrollees by 2017. Overall, the average calibration adjustment is modest; our final estimates of the number of uninsured women is 4.9% lower than estimated by the base model. But the difference varies by state; revised estimates are higher than the base model increase in some states, but lower in others. (Other details about the methodology are available from the authors.)For every state we estimate scenarios of 1) Medicaid expansion to at least 138% of poverty by 2017 and 2) no expansion, using state-specific eligibility levels in the absence of an expansion. (For expanding states, we use Medicaid or similar state program eligibility, including income level and categorical eligibility status in 2013; for non-expanding states, we used January 2015 eligibility criteria [
Brooks et al., 2015
, Kaiser Family Foundation, 2015
]). For scenarios without an expansion model, results are modified based on the expected changes in the uninsured population from 2013 to 2017, but with no gain in insurance coverage for those whose incomes fall between the state's current Medicaid eligibility level and 100% of poverty, who are eligible for neither Medicaid nor tax credits to purchase coverage through the marketplace. Those with incomes between 100% and 138% FPL in a non-expanding state are eligible for premium assistance in the marketplaces, but we reduce the expected participation rates by one-third based on prior research (Kaiser Family Foundation. (2015) Status of state action on the Medicaid expansion decision, as of March 6, 2015. Available: http://kff.org/health-reform/state-indicator/state-activity-around-expanding-medicaid-under-the-affordable-care-act/. Accessed: March 15, 2015.
Ku and Coughlin, 1999
, Liu and Chollet, 2006
).These analyses are supplemented with data from two other sources. Data from the 2012 Behavioral Risk Factor Surveillance System are used to estimate the percent of women who received mammograms in the past 2 years or Pap tests in the past 3 years. We also used programmatic data about the number of women who received mammograms or Pap tests from NBCCEDP.
Results
In 2013, 12.4 million women 21 to 64 with incomes at or below 250% FPL were uninsured (32.2%). With 29 states expanding Medicaid (as of June 2015), we estimate that 5.7 million low-income women (14.6%) will remain uninsured in 2017 (Table 1). To demonstrate the potential effect of Medicaid expansions, if all states expand Medicaid, 3.9 million women (9.9%) would be uninsured, compared with 7.2 million (18.4%) uninsured if no states expanded their Medicaid programs. In all scenarios, other ACA policies, like health insurance marketplaces and the individual insurance mandate, are in effect and contribute to insurance gains, but insurance coverage improves further when states expand their Medicaid eligibility. Similar changes in insurance coverage are expected among the target population for mammography services—low-income women 40 to 64 or 50 to 64.
Table 1Estimated Changes in Insurance Coverage for Women at or Below 250% of Poverty Under Alternative Medicaid Expansion Scenarios
Source: 2013 American Community Survey data and GW simulation model.
| Target Population (At or Below 250% FPL) | Actual Uninsured 2013 | Estimated Uninsured 2017 | ||
|---|---|---|---|---|
| 29 States Expand, Based on April 2015 Policies | No States Expand Medicaid | All States Expand Medicaid | ||
| Women 21–64 (cervical cancer screening) | ||||
| Thousands uninsured | 12,389.4 | 5,699.9 | 7,190.0 | 3,880.2 |
| Uninsured as % of low income | 32.2 | 14.6 | 18.4 | 9.9 |
| Women 40–64 (breast cancer screening) | ||||
| Thousands uninsured | 5,799.1 | 2,584.8 | 3,349.5 | 1,705.0 |
| Uninsured as % of low income | 31.1 | 13.5 | 17.5 | 8.9 |
| Women 50–64 (breast cancer screening) | ||||
| Thousands uninsured | 3,029.6 | 1,336.6 | 1,773.2 | 864.2 |
| Uninsured as % of low income | 28.7 | 12.2 | 16.2 | 7.9 |
Table 2 compares changes in insurance status in the 29 Medicaid-expanding states versus the 22 non-expanding states. Even before the expansions were implemented, expanding states had fewer uninsured women (28.7% of women ages 21–64) in 2013 than non-expanding states (36.9%). By 2017, 8.0% of the women in Medicaid-expanding states will be uninsured versus 23.3% in non-expansion states. Non-expanding states generally began with more uninsured women and disparities between expanding and non-expanding states will widen further by 2017.
Table 2Estimated Changes in Insurance Coverage of Women at or Below 250% FPL, by Medicaid Expansion Policies as of June 2015
| Target Population (At or Below 250% FPL) | 29 States (Including DC) Expanding Medicaid | 22 States Not Expanding Medicaid | ||
|---|---|---|---|---|
| Uninsured 2013 | Uninsured 2017 | Uninsured 2013 | Uninsured 2017 | |
| Women 21–64 (cervical cancer screening) | ||||
| Thousands uninsured | 6,287.3 | 1,783.0 | 6,102.1 | 3,916.9 |
| Uninsured as % of low income | 28.7 | 8.0 | 36.9 | 23.3 |
| Women 40–64 (breast cancer screening) | ||||
| Thousands uninsured | 2,970.7 | 779.4 | 2,828.4 | 1,810.5 |
| Uninsured as % of low income | 27.9 | 7.1 | 35.3 | 22.0 |
| Women 50–64 (breast cancer screening) | ||||
| Thousands uninsured | 1,556.0 | 392.4 | 1,473.6 | 944.2 |
| Uninsured as % of low income | 26.0 | 6.3 | 32.2 | 20.0 |
Source: 2013 American Community Survey data and GW simulation model.
Before insurance expansions, women living in states that are not expanding Medicaid were less likely to be screened for cancer than women living in states that expand. Our analyses of 2012 Behavioral Risk Factor Surveillance System data found that 83% of women 21 to 64 had a Pap test in the past 3 years in non-expansion states versus 86% in Medicaid expansion states (p < .001). Similarly, 72% of women 40 to 64 had a mammogram in the past 2 years in non-expansion states, compared with 76% in Medicaid expanding states (p < .001). Given the relation of insurance status to cancer screening, it seems likely that widening disparities in insurance status will also lead to widening disparities in cancer screening.
Sabik et al., 2015
also found that women's cancer screening rates were lower in non-expansion states.To provide a more detailed review of the changes at the state level, Table 3 indicates the expected changes in insurance status of women 21 to 64 from 2013 to 2017, with and without Medicaid expansions, grouped by expansion status.
Table 3Estimated Changes in Insurance Coverage for Women 21 to 64 at or Below 250% FPL Based on Medicaid Expansion Policies, by State
Source: 2013 American Community Survey data and GW simulation model.
| 2013 | 2017 | |||||
|---|---|---|---|---|---|---|
| Thousands Uninsured | % Uninsured | Without Medicaid Expansion | With Medicaid Expansion | |||
| Thousands Uninsured | % Uninsured | Thousands Uninsured | % Uninsured | |||
| 29 states expanding Medicaid as of June 2015 | ||||||
| Arizona | 286.5 | 33.4 | 203.8 | 23.4 | 116.6 | 13.4 |
| Arkansas | 144.9 | 34.0 | 30.7 | 7.1 | 30.7 | 7.1 |
| California | 1,725.9 | 35.7 | 435.1 | 8.8 | 435.1 | 8.8 |
| Colorado | 176.7 | 30.8 | 96.5 | 16.5 | 32.7 | 5.6 |
| Connecticut | 71.9 | 22.7 | 30.6 | 9.5 | 15.3 | 4.8 |
| Delaware | 21.7 | 21.2 | 11.5 | 11.0 | 10.5 | 10.1 |
| District of Columbia | 7.6 | 9.2 | 5.5 | 6.5 | 5.5 | 6.5 |
| Hawaii | 23.7 | 17.9 | 13.1 | 9.7 | 13.1 | 9.7 |
| Illinois | 419.9 | 29.0 | 305.7 | 20.7 | 169.3 | 11.5 |
| Indiana | 261.7 | 31.7 | 165.4 | 19.7 | 91.0 | 10.8 |
| Iowa | 65.8 | 20.1 | 49.5 | 14.9 | 26.9 | 8.1 |
| Kentucky | 200.1 | 32.7 | 136.3 | 21.9 | 21.6 | 3.5 |
| Maryland | 130.9 | 24.7 | 83.0 | 15.4 | 28.4 | 5.3 |
| Massachusetts | 44.4 | 7.1 | 32.4 | 5.1 | 32.4 | 5.1 |
| Michigan | 313.2 | 25.2 | 241.5 | 19.1 | 118.4 | 9.4 |
| Minnesota | 97.6 | 18.8 | 8.5 | 1.6 | 8.5 | 1.6 |
| Nevada | 151.5 | 40.8 | 103.0 | 27.2 | 46.0 | 12.2 |
| New Hampshire | 33.7 | 28.1 | 22.4 | 18.4 | 10.1 | 8.3 |
| New Jersey | 295.7 | 35.9 | 151.6 | 18.1 | 93.4 | 11.1 |
| New Mexico | 112.9 | 38.0 | 73.5 | 24.3 | 30.5 | 10.1 |
| New York | 472.0 | 20.6 | 233.0 | 10.0 | 204.4 | 8.8 |
| North Dakota | 18.3 | 25.5 | 14.8 | 20.5 | 7.3 | 10.1 |
| Ohio | 356.0 | 25.3 | 247.6 | 17.3 | 104.8 | 7.3 |
| Oregon | 160.2 | 31.4 | 113.1 | 21.8 | 19.4 | 3.7 |
| Pennsylvania | 319.7 | 23.3 | 214.0 | 15.3 | 36.7 | 2.6 |
| Rhode Island | 26.0 | 23.6 | 16.9 | 15.1 | 3.8 | 3.4 |
| Vermont | 7.3 | 10.5 | 3.7 | 5.2 | 3.7 | 5.2 |
| Washington | 256.8 | 33.8 | 173.7 | 22.5 | 52.8 | 6.8 |
| West Virginia | 84.8 | 32.8 | 56.5 | 21.4 | 14.3 | 5.4 |
| 22 states not expanding Medicaid | ||||||
| Alabama | 213.1 | 31.3 | 141.1 | 20.3 | 66.4 | 9.6 |
| Alaska | 25.4 | 43.5 | 16.2 | 27.4 | 11.6 | 19.6 |
| Florida | 1,059.6 | 40.0 | 623.9 | 23.1 | 280.1 | 10.4 |
| Georgia | 564.9 | 40.5 | 357.8 | 25.2 | 153.0 | 10.8 |
| Idaho | 72.9 | 35.1 | 39.9 | 18.9 | 13.2 | 6.3 |
| Kansas | 105.8 | 32.8 | 71.3 | 21.7 | 40.3 | 12.3 |
| Louisiana | 249.7 | 37.4 | 177.2 | 26.1 | 102.9 | 15.2 |
| Maine | 35.1 | 21.3 | 20.1 | 11.9 | 13.9 | 8.3 |
| Mississippi | 161.5 | 33.9 | 108.0 | 22.3 | 40.3 | 8.3 |
| Missouri | 233.6 | 30.6 | 162.9 | 21.0 | 89.8 | 11.6 |
| Montana | 40.2 | 32.7 | 24.6 | 19.6 | 10.3 | 8.2 |
| Nebraska | 58.0 | 29.2 | 38.7 | 19.2 | 26.3 | 13.0 |
| North Carolina | 477.3 | 35.4 | 281.5 | 20.5 | 96.5 | 7.0 |
| Oklahoma | 192.1 | 37.3 | 137.1 | 26.2 | 95.5 | 18.2 |
| South Carolina | 227.5 | 33.4 | 134.5 | 19.4 | 71.5 | 10.3 |
| South Dakota | 33.9 | 33.5 | 24.5 | 23.8 | 15.1 | 14.7 |
| Tennessee | 262.1 | 29.2 | 150.7 | 16.5 | 81.6 | 8.9 |
| Texas | 1,583.5 | 46.4 | 1,095.7 | 31.6 | 677.7 | 19.5 |
| Utah | 93.5 | 29.5 | 59.3 | 18.5 | 37.0 | 11.5 |
| Virginia | 273.1 | 32.3 | 181.1 | 21.0 | 108.0 | 12.5 |
| Wisconsin | 121.8 | 19.0 | 59.3 | 9.1 | 59.3 | 9.1 |
| Wyoming | 17.5 | 32.1 | 11.4 | 20.6 | 6.9 | 12.5 |
The NBCCEDP supports cancer screening for women who remain uninsured, but its capacity is limited by grant funding levels. Over the 2011 to 2013 period, 658,611 women received Pap tests under NBCCEDP, and 548,386 women 40 to 64 and 467,617 women 50 to 64 received breast cancer screening in 2012 to 2013 (3- and 2-year periods are used based on the recommended testing periodicity). If the same caseloads could be served in 2017 and only 29 states expand Medicaid, 11.6% of 21- to 64-year-old uninsured women eligible for cervical cancer screening could be served, 21.2% of 40- to 64-year-old women eligible for breast cancer screening and 37.8% of the priority population of women 50 to 64 years old could be served (Table 4). Even if all states expand Medicaid, the number of eligible uninsured women still outstrips the current NBCCEDP caseloads.
Table 4Comparison of Number of Women Screened Under NBCCEDP in 2011 to 2013 to the Estimated Number of Uninsured Women at or Below 250% FPL in 2017, Under Alternative Scenarios
Source: CDC data and GW simulation model.
| Target Population | Thousands of Women Screened by NBCCEDP | As % of Uninsured Women if 29 State Expand Medicaid, 2017 | As % of Uninsured Women if No States Expand Medicaid, 2017 | As % of Uninsured Women if all States Expand Medicaid, 2017 |
|---|---|---|---|---|
| Women 21–64 receiving pap tests, 2011–2013 | 658.6 | 11.6 | 9.2 | 17.0 |
| Women 40–64 receiving mammograms, 2012–2013 | 548.4 | 21.2 | 16.4 | 32.2 |
| Women 50–64 receiving mammograms, 2012–2013 | 467.6 | 37.8 | 26.4 | 54.1 |
Abbreviations: NBCCEDP, National Breast and Cervical Cancer Early Detection Program.
Discussion
The ACA has created a golden opportunity to increase cancer screening in the coming years. Although the overall national prospects are enhanced by the ACA, differences in Medicaid expansion decisions will also affect outcomes. In 2013, the risk a low-income woman in a non-expanding state was uninsured (36.9%) was about one-third higher than the risk for women in Medicaid expanding states (28.7%). Based on current expansion plans, by 2017, about three times as many women in non-expanding states will be uninsured (23.3%), compared with women in states that expand Medicaid (8.0%). These findings are in accord with the most recent census data for insurance coverage changes between 2013 and 2014 (
Smith and Medalia, 2015
). At the national level, Current Population Survey data indicate that the percent of low-income (under 250% of the poverty line) women 40 to 64 who were uninsured decreased by almost one-third between 2013 and 2014, from 28.1% to 19.6% (author's tabulations of Current Population Survey data). As noted, women living in non-expanding states were less likely to have had mammograms or Pap tests even before the insurance expansions began. The increasing differences in insurance coverage could widen disparities in women's cancer screening as well. This may, in turn, contribute to greater disparities in cancer outcomes between the states.Health insurance expansions will increase the demand for services to bolster cancer screening rates. After the first year of a randomized expansion of Medicaid in Oregon, the percent of women who had a mammogram or Pap test in the past year was about 18 to 19 percentage points higher than women in the comparison group (
Finkelstein et al., 2011
). Analyses of the effects of Massachusetts health reform also found significant increases in breast and cervical cancer screening, particularly for low-income women and screening rates continued to increase over time (Sabik and Bradley, 2015
). On the other hand, - Sabik L.M.
- Bradley C.J.
The impact of near-universal insurance coverage on breast and cervical cancer screening: Evidence from Massachusetts.
Health Economics. 2015; ([Epub ahead of print])https://doi.org/10.1002/hec. 3159
Han et al. (2015)
failed to find changes in cancer screenings between 2009 and 2011 to 2012 just after the ACA required that preventive services be offered without cost sharing, although they found increased use of other preventive services, such as cholesterol checks and flu vaccinations. Because breast and cervical cancer screenings are recommended only every 2 or 3 years, it may take more time for improvements to be detected.As a result of the ACA, the number of uninsured women will decrease and the number of women eligible for the NBCCEDP will decline; however, millions will remain uninsured and struggle to afford timely cancer screenings. Our analyses indicate that the number of women who could be served with current funding levels will remain far below the number of uninsured eligible women: roughly 12% to 38% of the remaining low-income uninsured women could be served at current levels. This gap would make it harder for the nation to attain the Healthy People 2020 objectives that 93% of women targeted for screening under recent guidelines receive cervical cancer screens and 81% of women targeted to receive mammograms (
Office of Disease Prevention and Health Promotion, HHS, 2015
). The number of women who can be served by the program will greatly depend on future federal funding and federal budget restrictions (e.g., sequestration). Should current funding levels be maintained and the budget restrictions continue, funding levels will still be too low to serve all low-income uninsured women in the future. Programs providing cancer control and prevention services to low-income and uninsured women will still be needed after health reform. Availability of NBCCEDP services is associated with higher mammography use (Office of Disease Prevention and Health Promotion, HHS. (2015). Healthy People 2020 cancer objectives. Available: www.healthypeople.gov/2020/topics-objectives/topic/cancer/objectives. Accessed: September 23, 2015.
Kouroukian et al., 2015
). Our estimates are conservative because they only consider those who are uninsured. However, the NBCCEDP also provides services to underinsured individuals, which includes women who are insured but require diagnostic testing rather than preventive screening whose deductibles or cost-sharing render care unaffordable. (Decisions about how to determine whether a woman is underinsured is a local programmatic option.)This analysis has both technical and policy limitations. The earlier report described some of the technical limitations (
Levy et al., 2012
). Our estimates are largely based on analyses of insurance coverage in Massachusetts, which assumes this experience can be used to model effects of the ACA in other states, although we incorporate state-specific adjustments based on Medicaid and health insurance marketplace enrollments in 2014 and 2015 to account for differences across states. Finally, both self-reported data about insurance coverage and other characteristics and administrative data may be subject to reporting error. Nonetheless, our results accord with early 2014 data showing that insurance coverage of non-elderly adults is rising, particularly in states expanding Medicaid.Many policy issues remain in flux. States may continue to change their Medicaid expansion policies and as a result we provide estimates for every state with and without a Medicaid expansion to indicate the potential impact of changes. While the Supreme Court decision in King v Burwell has upheld the tax subsidies for health insurance marketplaces in all states (
Barnes, 2015
), other political and legal challenges to the ACA remain. Scientific recommendations about cancer screening may change over time. After the U.S. Preventive Services Task Force changed its recommendations for the age range for cervical cancer screening from 18 to 64 to 21 to 64, the CDC changed the target ages for NBCCEDP. The Task Force (Siu et al., 2016
) recently released a final recommendation to recommend mammograms every 2 years for women ages 50 to 64. As a result, the demand for mammograms may change and CDC could change the eligibility age for NBCCEDP. However, in December 2015, Congress added a provision (Section 229) in the Consolidated Appropriations Act of 2016 that places a moratorium on the implementation of any new USPSTF mammography recommendations until January 2018 (American College of Radiology, 2015
). Regardless of federal policy changes, however, revisions in cancer screening guidelines could also affect physician and patient behaviors and use of services. For example, use of screening mammography by Medicare women decreased after 2009 after Task Force recommendations changed, even though older women were still recommended for screening (American College of Radiology. (2015). Press release: MPPR rollback and mammography protections in Consolidated Appropriations Act: A victory for patients and providers. Available: www.newswise.com/articles/mppr-rollback-and-mammography-protections-in-consolidated-appropriations-act-a-victory-for-patients-and-providers. Accessed: December 18, 2015.
Sharpe et al., 2013
).These analyses indicate how the ACA insurance expansions could increase women's cancer screening after a decade of flat or declining screening rates. But it also demonstrates that progress will also rely on state policy decisions about Medicaid expansion. If insurance disparities widen between expanding and non-expanding states, existing disparities in cancer screening rates could widen as well. The public health need for NBCCEDP to support cancer screening for low-income uninsured women will continue, although the Department of Health and Human Services and its grantees should continue to seek to improve the program's effectiveness.
Implications for Policy and/or Practice
Health insurance expansions under the ACA will substantially increase the number of low-income women who have financial access to breast and cervical cancer screening, which should also lead to earlier detection and improved outcomes. The level of increases will depend in large measure on whether women live in states with Medicaid expansions, although insurance coverage will grow in both Medicaid expanding and non-expanding states. Because women living in states without expansions already had lower rates of cancer screening in earlier years than those in expanding states, decisions about Medicaid expansions could increase the disparities in access to cancer screening. Non-expanding states could decrease the number of uninsured women and increase cancer screening for low-income women if they opt to expand Medicaid and enrollment in Medicaid and health insurance exchanges should be strongly promoted. Nonetheless, millions of low-income women will remain uninsured and will continue to need assistance from the CDC's NBCCEDP.
Acknowledgments
This project was funded by the American Cancer Society and American Cancer Society Cancer Action Network, grant 21618. Staff of the Centers for Disease Control and Prevention provided data about NBCCEDP. The authors acknowledge helpful comments from Citseko Staples of ACS CAN and from Jacqueline Miller, Faye Wong and Janet Royalty of CDC.
References
- The Breast and Cervical Cancer Prevention and Treatment Act in Georgia: Effects on time to Medicaid enrollment.Cancer. 2009; 115: 1300-1309
- Cancer prevention & early detection facts & figures: 2015-16.American Cancer Society, Atlanta2015
American College of Radiology. (2015). Press release: MPPR rollback and mammography protections in Consolidated Appropriations Act: A victory for patients and providers. Available: www.newswise.com/articles/mppr-rollback-and-mammography-protections-in-consolidated-appropriations-act-a-victory-for-patients-and-providers. Accessed: December 18, 2015.
- Projecting the unmet need and costs for contraception services after health care reform.American Journal of Public Health. 2016; 106: 334-341
- The Oregon experiment – Effects of Medicaid on clinical outcomes.New England Journal of Medicine. 2013; 368: 1713-1722
Barnes, R. (2015). Affordable Care Act survives Supreme Court challenge. Washington Post, June 25, 2015.
- Modern era Medicaid: Findings from a state survey of eligibility, enrollment, renewal, and cost-sharing policies in Medicaid and CHIP as of January 2015.Kaiser Commission on Medicaid and the Uninsured, Washington, DC2015
- Table 3. Projections of the population by sex and selected age groups for the United States: 2015 to 2060 (NP2014–T3).U.S. Census Bureau., Washington, DC2014
Census Bureau. (2015) American Community Survey. Data: response rates. Available: http://www.census.gov/acs/www/methodology/sample-size-and-data-quality/response-rates/index.php. Accessed: March 25, 2015.
- Medicaid & CHIP: December 2014 monthly applications, eligibility determinations and enrollment report. Feb. 23, 2015.Department of Health and Human Services., Baltimore, MD2015
- Health insurance coverage: Early release of estimates from the National Health Interview Survey, 2014.National Center for Health Statistics, Atlanta2015
Congressional Budget Office. (2015). Insurance coverage provisions of the Affordable Care Act—CBO's March 2015 baseline table. Available: www.cbo.gov. Accessed: March 2015.
Council of Economic Advisors, the White House. (2015). Missed opportunities: The consequences of state decisions not to expand Medicaid. June 2015. Washington, DC: White House. Retrieved from https://www.whitehouse.gov/sites/default/files/docs/missed_opportunities_medicaid.pdf.
Finkelstein, A., Taubman, S., Wright, B., Bernstein, M., Gruber, J., Newhouse, JP, … Baicker, K. (2011). The Oregon health insurance experiment: Evidence from the first year. NBER Working Paper 17190 July 2011.
- Massachusetts points the way to successful health care reform.Journal of Policy Analysis and Management. 2011; 30: 184-192
- Has recommended preventive service use increased after elimination of cost-sharing as part of the Affordable Care Act in the United States?.Preventive Medicine. 2015; 78: 85-91
- Does Massachusetts points the way to success with national reform?.Journal of Policy Analysis and Management. 2011; 30: 178-184
- The impact of a national program to provide free mammograms to low-income, uninsured women on breast cancer mortality rates.Cancer. 2010; 116: 4456-4462
Kaiser Family Foundation. (2015) Status of state action on the Medicaid expansion decision, as of March 6, 2015. Available: http://kff.org/health-reform/state-indicator/state-activity-around-expanding-medicaid-under-the-affordable-care-act/. Accessed: March 15, 2015.
- QuickTake: As of December, many uninsured adults were not aware of tax penalties for not having coverage, the marketplaces, or the open enrollment deadline.Urban Institute, Washington, DC2015
- Lasting effects of the Breast and Cervical Cancer Early Detection Program on Breast Cancer Detection and Outcomes, Ohio 2000-2009.Preventing Chronic Disease. 2015; 23: E116
- Sliding-scale premium health insurance programs: Four states' experiences.Inquiry. 1999; 36: 471-480
- An evaluation of a Medicaid expansion for cancer care: The Breast and Cervical Cancer Prevention and Treatment Act of 2000.Women's Health Issues. 2009; 19: 221-231
- Health care reform and women’s insurance coverage for breast and cervical cancer screening.Preventing Chronic Disease. 2012; 9: 120069
- Price and income elasticity of the demand for health insurance and health care services: A critical review of the literature.Mathematica Policy Research, Washington, DC2006
National Cancer Institute. (2015). Early detection. Cancer trends progress report. Available: http://progressreport.cancer.gov/detection. Accessed: September 23, 2015.
- Health insurance marketplaces 2015 open enrollment period: March enrollment report.Department of Health and Human Services, Washington, DC2015
Office of Disease Prevention and Health Promotion, HHS. (2015). Healthy People 2020 cancer objectives. Available: www.healthypeople.gov/2020/topics-objectives/topic/cancer/objectives. Accessed: September 23, 2015.
- The National Breast and Cervical Cancer Early Detection Program in the era of health reform: A vision forward.Cancer. 2014; 120: 2620-2624
- Timeliness of breast cancer diagnosis and initiation of treatment in the National Breast and Cervical Cancer Early Detection Program, 1996 – 2005.American Journal of Public Health. 2010; 100: 1769-1776
- The Supreme Court's surprising decision on the Medicaid expansion: How will the federal government and states proceed?.Health Affairs. 2012; 31: 1663-1672
- The impact of near-universal insurance coverage on breast and cervical cancer screening: Evidence from Massachusetts.Health Economics. 2015; ([Epub ahead of print])https://doi.org/10.1002/hec. 3159
- State Medicaid expansion decisions and disparities in women's cancer screening.American Journal of Preventive Medicine. 2015; 48: 98-103
- The effect of the controversial US Preventive Services Task Force recommendations on the use of screening mammography.Journal of the American College of Radiology. 2013; 10: 21-24
- Screening for Breast Cancer: The U.S. Preventive Services Task Force Recommendation.Annals of Internal Medicine. 2016; (2015 Jan 12. http://dx.doi.org/10.7326/M15-2886. [Epub ahead of print])
- U.S. Census Bureau, Current Population Reports, P60–253, health insurance coverage in the United States: 2014.U.S. Government Printing Office, Washington, DC2015
- Does increasing insurance improve outcomes for US cancer patients?.Journal of Surgical Research. 2013; 185: 15-20
Biography
Leighton Ku, PhD, MPH, is Professor, Health Policy and Management in the Department of Health Policy and Management, Milken Institute School of Public Health, George Washington University; Director of the George Washington Center for Health Policy Research; Interim Department Chair.
Tyler Bysshe, MPH, is a Senior Research Assistant in Department of Health Policy and Management at George Washington University.
Erika Steinmetz, MBA is a Senior Research Scientist in Department of Health Policy and Management at George Washington University.
Brian K. Bruen, MS is a Lead Research Scientist and Lecturer in Department of Health Policy and Management at George Washington University. He is also a PhD candidate in health policy at the Trachtenberg School of Public Policy and Public Administration.
Article info
Publication history
Published online: February 23, 2016
Accepted:
January 8,
2016
Received in revised form:
December 18,
2015
Received:
June 29,
2015
Footnotes
Conflict of Interest: This project was supported by the American Cancer Society and American Cancer Society Cancer Action Network grant 21618. The American Cancer Society reviewed a draft of the manuscript, but was not involved in the study design or execution and had no editorial control over the manuscript nor approval rights. Staff of the Centers for Disease Control and Prevention provided some data for the project but was not involved in the study design nor execution and reviewed a draft of the manuscript, but had no editorial control nor approval rights.
Financial Disclosures: Leighton Ku, Tyler Bysshe, Erika Steinmetz, and Brian Bruen have no financial disclosures.
Identification
Copyright
© 2016 Jacobs Institute of Women's Health. Published by Elsevier Inc.
