. Author manuscript; available in PMC: 2023 Sep 1.
Abstract Introduction:The U.S. Preventive Services Task Force recommends breast, cervical and colorectal cancer (CRC) screening to reduce mortality from these cancers, but screening use has been below national targets. The purpose of this study is to examine the proportion of screening-eligible adults up-to-date with these screenings, and how screening use compares to Healthy People 2020 (HP2020) targets.
Methods:Data from the 2019 National Health Interview Survey (NHIS) were used to examine percentages up-to-date with breast cancer screening among women aged 50‒74 years without prior breast cancer; cervical cancer screening among women aged 21‒65 years without prior cervical cancer or hysterectomy; and CRC screening among adults aged 50‒75 years without prior CRC. Estimates are presented by sociodemographic characteristics and healthcare access factors. Analyses were conducted in 2021.
Results:Percentages of adults up-to-date were 76.2% (95%CI 75.0–77.5) for breast cancer screening, 76.4% (95%CI 75.2%−77.6%) for cervical cancer screening, and 68.3% (95%CI 67.3%−69.3%) for CRC screening. Although some population subgroups met breast and CRC screening targets (81.1% and 70.5%, respectively), many did not, and cervical cancer screening was below the target for all examined subgroups. Lower education and income, non-metropolitan county of residence (which included rural counties), no usual source of care or health insurance coverage, and Medicaid coverage were associated with lower screening test use.
Conclusions:Estimated use of breast, cervical and CRC screening tests based on the 2019 NHIS were below national targets. Continued monitoring may allow examination of screening trends, inform interventions, and track progress in eliminating disparities.
INTRODUCTIONThe U.S. Preventive Services Task Force (USPSTF) recommends breast, cervical, and colorectal cancer (CRC) screening.1 Screening use has been below national targets, with disparities among some population groups.2,3 This analysis uses the most recent National Health Interview Survey (NHIS) data to examine use of these screenings. Findings are compared with Healthy People 2020 (HP2020) targets.4 The USPSTF recommends lung cancer screening for some adults;1 however, this information was not collected in 2019.
METHODS Study SampleData came from the 2019 NHIS, a survey of a nationally representative sample of the civilian, noninstitutionalized U.S. population.5 NHIS underwent survey redesign in 2019, including changes in some questions and response options, imputation methods, weighting and other changes.5 Screening questions (Table 1) were asked of 1 randomly selected adult from each household (final response rate 59.1%). For each screening type, screening-eligible adults per USPSTF recommendations were included (breast: n=7,289; cervical: n=11,763; CRC: n=13,989) (Table 2).1 Respondents with personal or unknown history of that cancer (breast: n=447; cervical: n=171; CRC: n=138), unknown screening status (breast: n=76; cervical: n=530; CRC: n=414), and for cervical screening previous or unknown hysterectomy (n=1,787) were excluded.
Table 1.2019 National Health Interview Survey (NHIS) Questions on Breast, Cervical, and Colorectal Cancer Screening
Cancer screening/NHIS survey universe Survey question Breast cancer Females aged ≥30 years Have you ever had a mammogram? If yes: About how long has it been since your most recent mammogram? Cervical cancer Females aged ≥18 years Have you ever had a test to check for cervical cancer? If yes: When did you have your most recent test to check for cervical cancer? At your most recent cervical cancer screening, did you have a Pap test? At your most recent cervical cancer screening, did you have an HPV test? Colorectal cancer Adults aged ≥40 years Colonoscopy and sigmoidoscopy are exams to check for colon cancer. Have you ever had either of these exams? If yes: Have you had a colonoscopy, a sigmoidoscopy, or both? Adults aged ≥40 years who ever had sigmoidoscopy or both colonoscopy and sigmoidoscopy When was your most recent sigmoidoscopy? Adults aged ≥40 years who ever had colonoscopy or both colonoscopy and sigmoidoscopy About how long has it been since your most recent colonoscopy? Adults aged ≥40 years who ever had colonoscopy or sigmoidoscopy but don’t know which type When was your most recent colonoscopy or sigmoidoscopy? Adults aged ≥40 years Have you ever had any other kind of test for colorectal cancer, such as virtual colonoscopy, CT colonography, blood stool test, FIT-DNA or Cologuard test? Adults aged ≥40 years who had a test other than colonoscopy or sigmoidoscopy Have you ever had a CT colonography or virtual colonoscopy? If yes: When was your most recent CT colonography or virtual colonoscopy? Have you ever had a blood stool or FIT test, using a home test kit? If yes: When was your most recent blood stool or FIT test, using home test kit? Adults aged ≥40 years who ever had a home blood stool or FIT test Have you ever had a Cologuard test? If yes: Was the blood stool or FIT test you reported earlier conducted as part of a Cologuard test? When did you have your most recent Cologuard test? (not released) Table 2.Definitions of Up-to-Date With Cancer Screening, by Cancer Screening Type
Screening type Age and sex eligibility criteria Definition of up-to-date with screening Breast cancer Women aged 50‒74 years Mammography within 2 years Cervical cancer Women aged 21‒65 years Papanicolaou (Pap) test within 3 years among women aged 21‒65 years, or human papillomavirus (HPV) test only or with Pap (co-test) within 5 years among ages 30‒65 years Colorectal cancer Adults aged 50‒75 years Colonoscopy within 10 years, sigmoidoscopy or CT colonography within 5 years, home blood stool (FOBT) or fecal immunochemical (FIT) test within 1 year, or FIT-DNAa within 3 years MeasuresUp-to-date screening included tests for any reason within USPSTF-recommended intervals. For CRC tests, responses about time since most recent fecal immunochemical test (FIT) with stool deoxyribonucleic acid tests (FIT-DNA) were not released by the National Center for Health Statistics (NCHS). Therefore, timing of home blood stool (FOBT) or FIT test was used to classify FIT-DNA status (Table 2).
Estimates are presented by sociodemographic characteristics and healthcare access factors (Tables 3‒4).5,6 Ethnicity includes Hispanic and non-Hispanic groups. Income is presented as percentage of poverty thresholds. County metropolitan status in NHIS includes 4 groups based on the 2013 NCHS Urban-Rural Classification Scheme for Counties.5 Usual source of care includes places respondents usually go when sick or needing healthcare. No usual source included no place, no one place most often, emergency rooms, urgent care centers, or grocery or drug store clinics.
Table 3.Percentage of Women Up-to-Date With Breast and Cervical Cancer Screening ‒ U.S., 2019
Characteristics Breast cancer screening Cervical cancer screening Sample n Weighted % (95%CI) p-valuea HP2020 target Sample n Weighted % (95%CI) p-valuea HP2020 target 81.1% 93.0% Overall ‒ unadjusted 6,766 76.2 (75.0, 77.5) 9,275 76.4 (75.2, 77.6) Overall ‒ age-standardizedb 6,766 76.2 (74.9, 77.5) 9,275 76.8 (75.6, 77.9) Age, years 0.122 <0.001 21‒30 2,017 71.1 (68.3, 73.8) 31‒40 2,466 81.9 (79.9, 83.8) 41‒50 1,896 79.3 (77.0, 81.5) 51‒65 2,896 74.4 (72.4, 76.3) 50‒64 4,085 75.5 (73.9, 77.2) 65‒74 2,681 77.7 (75.5, 79.7) Race 0.109 0.001 AIAN 104 64.3 (52.2, 75.2) 183 75.6 (63.6, 85.2) Asian only 266 72.3 (64.7, 79.0) 620 67.3 (62.4, 71.9) Black/African American only 788 79.0 (74.9, 82.6) 1,206 77.8 (74.7, 80.7) White only 5,293 76.0 (74.5, 77.4) 6,569 77.9 (76.5, 79.2) Other single and multiple race 60 80.8 (68.2, 90.0) 145 70.6 (60.2, 79.6) Missing/Unknown 255 80.6 (74.0, 86.2) 552 70.8 (65.5, 75.7) Ethnicityc 0.308 <0.001 Non-Hispanic 6,103 76.0 (74.6, 77.3) 7,744 78.0 (76.7, 79.2) Hispanic 663 78.1 (74.0, 81.8) 1,531 69.9 (66.9, 72.9) Mexican/Mexican American 341 77.3 (71.4, 82.5) 876 70.2 (66.0, 74.1) All other Hispanic groupsd 311 79.0 (72.9–84.2) 642 69.4 (64.9–73.5) Education <0.001 <0.001 <High school 580 68.1 (62.9, 72.9) 610 59.3 (54.1, 64.2) High school/GED 1,710 73.2 (70.7, 75.7) 1,965 68.9 (66.2, 71.5) Some college 2,151 75.9 (73.7, 78.0) 2,743 77.3 (75.3, 79.2) College degree 2,304 82.5 (80.6, 84.3) 3,933 85.1 (83.7, 86.4) Missing/Unknown 21 h 24 h Income, % poverty threshold <0.001 <0.001 ≤138% 1,225 66.9 (63.4, 70.3) 1,808 64.6 (61.6, 67.6) >138%‒250% 1,297 70.7 (67.3, 73.8) 1,752 71.2 (68.3, 73.9) >250%‒400% 1,320 75.1 (71.9, 78.0) 1,947 77.8 (75.1, 80.2) >400% 2,923 82.9 (81.2, 84.4) 3,768 84.3 (82.8, 85.7) Duration of U.S. residence 0.088 <0.001 <10 years 58 h 372 56.4 (49.9, 62.7) ≥10 years 869 77.4 (73.7, 80.8) 1,335 73.4 (70.5, 76.2) Born in U.S. 5,709 76.6 (75.2, 78.0) 7,364 78.6 (77.3, 79.9) Missing/Unknown 130 65.7 (54.8, 75.5) 204 64.4 (55.4, 72.7) County metropolitan statuse 0.002 0.033 Large central metropolitan 1,794 77.3 (74.8, 79.7) 2,994 77.1 (75.1, 79.0) Large fringe metropolitan 1,592 78.5 (75.9, 80.9) 2,146 78.6 (76.3, 80.7) Medium/small metropolitan 2,182 76.6 (74.3, 78.7) 2,910 75.3 (73.1, 77.4) Nonmetropolitan 1,198 70.5 (67.0, 73.8) 1,225 72.9 (69.2, 76.4) Sexual orientation 0.797 0.381 Lesbian or gay 61 72.1 (56.0, 84.9) 138 72.3 (61.6, 81.4) Straight 6,476 76.5 (75.2, 77.8) 8,581 77.2 (76.0, 78.4) Bisexual 37 h 244 72.3 (64.3, 79.3) Other 16 h 48 h Missing/Unknown 176 68.0 (58.9, 76.1) 264 58.6 (50.8, 66.0) Usual source of care <0.001 <0.001 Yes 6,136 79.3 (78.0, 80.6) 7,582 79.6 (78.3, 80.8) No 628 47.4 (42.3, 52.6) 1,688 63.4 (60.5, 66.2) Missing/Unknown 2 h 5 h Insurancef Age <65 years <0.001 <0.001 Private 2,888 80.0 (78.2, 81.7) 6,273 81.5 (80.2, 82.7) Medicaid/Other Public 550 74.6 (70.0, 78.8) 1,346 70.9 (67.6, 74.0) Other coverage 282 78.5 (72.3, 83.9) 396 78.1 (72.1, 83.3) Uninsured 360 43.2 (37.0, 49.5) 1,039 57.4 (53.6, 61.1) Missing/Unknown 5 h 12 h Age ≥65 yearsg <0.001 0.799 Private 1,175 79.8 (76.5, 82.9) 94 67.4 (55.0, 78.3) Medicare + Medicaid 202 66.3 (57.1, 74.7) 15 h Medicare Advantage 789 83.5 (80.1, 86.6) 52 69.0 (54.0, 81.5) Medicare only 365 66.5 (60.1, 72.6) 29 h Other coverage 117 76.5 (66.5, 84.7) 12 h Uninsured 26 h 7 h Missing/Unknown 7 h 0 h Table 4.Percentage of Adults Aged 50‒75 Years Up-to-Date With Colorectal Cancer Screening ‒ U.S., 2019
Colorectal cancer screening Characteristics Sample n Weighted % (95% CI) p-valuea HP2020 target Overall ‒ unadjusted 13,437 68.3 (67.3, 69.3) 70.5% Overall ‒ age-standardizedb 13,437 67.9 (66.9, 68.9) Age, years <0.001 50‒64 7,979 62.2 (60.9, 63.6) 65‒75 5,458 79.7 (78.3, 81.0) Sex 0.002 Male 6,202 66.7 (65.2, 68.2) Female 7,235 69.8 (68.4, 71.0) Race <0.001 AIAN 216 62.8 (53.8, 71.1) Asian only 524 57.6 (51.9, 63.3) Black/African American only 1,461 69.5 (66.6, 72.2) White only 10,677 69.8 (68.7, 70.9) Other single and multiple race 97 62.6 (51.1, 73.1) Missing/Unknown 462 56.0 (50.3, 61.6) Ethnicityc <0.001 Non-Hispanic 12,202 70.3 (69.3, 71.3) Hispanic 1,235 53.8 (50.1, 57.4) Mexican/Mexican American 627 50.0 (45.3, 54.6) All other Hispanic groupsd 585 57.6 (52.1, 63.0) Education <0.001 <High school 1,235 52.2 (48.4, 56.0) High school/GED 3,496 64.5 (62.7, 66.3) Some college 4,034 71.0 (69.2, 72.7) College degree 4,616 76.3 (74.7, 77.9) Missing/Unknown 56 g Income, % poverty threshold <0.001 ≤138% 2,253 53.7 (50.8, 56.5) >138%‒250% 2,454 62.1 (59.5, 64.7) >250%‒400% 2,615 68.7 (66.2, 71.1) >400% 6,115 75.8 (74.3, 77.2) Duration of U.S. residence <0.001 <10 years 107 29.0 (19.7, 39.8) ≥10 years 1,703 57.8 (54.9, 60.7) Born in U.S. 11,341 71.4 (70.4, 72.4) Missing/Unknown 286 57.5 (49.9, 64.9) County metropolitan statuse 0.002 Large central metropolitan 3,564 65.7 (63.6, 67.8) Large fringe metropolitan 3,233 70.5 (68.5, 72.4) Medium/small metropolitan 4,262 69.8 (67.9, 71.5) Nonmetropolitan 2,378 66.6 (63.9, 69.1) Sexual orientation 0.041 Lesbian or gay 172 77.4 (69.5, 84.1) Straight 12,778 68.5 (67.5, 69.6) Bisexual 67 g Other 32 g Missing/Unknown 388 57.5 (51.5, 63.4) Usual source of care <0.001 Yes 12,029 71.9 (70.9, 72.9) No 1,404 39.4 (36.4, 42.4) Missing/Unknown 4 g Insurancef Age <65 years <0.001 Private 5,600 67.4 (65.9, 68.9) Medicaid/Other public 950 54.7 (50.4, 58.9) Other coverage 659 70.3 (66.0, 74.3) Uninsured 758 31.1 (27.1, 35.2) Missing/Unknown 12 g Age ≥65 years <0.001 Private 2,350 85.4 (83.6, 87.0) Medicare + Medicaid 369 64.9 (57.9, 71.4) Medicare Advantage 1,506 82.2 (79.7, 84.6) Medicare only 714 68.8 (64.7, 72.7) Other coverage 461 79.1 (73.9, 83.8) Uninsured 47 g Missing/Unknown 11 g Statistical AnalysisEstimates are presented as percentages with Korn-Graubard CIs. Overall percentages were age-standardized to the 2000 U.S. standard population. Wald F tests were used to test differences across groups. Design variables and survey weights were used to account for the complex sample design. Estimates not meeting NCHS standards for reliability were suppressed.7 SAS (version 9.4) and SUDAAN (version 11.0.03) were used in analyses conducted in 2021.
RESULTSFor breast cancer screening (Table 3), 76.2% of women were up-to-date (age-adjusted 76.2%), below the HP2020 target (81.1%). Lower use was associated with lower educational attainment and income, nonmetropolitan county, and no usual source of care or health insurance coverage. Use was generally lowest among women without a usual source of care (47.4%) or aged <65 years without health insurance coverage (43.2%).
For cervical cancer screening, 76.4% of women were up-to-date (age-adjusted 76.8%), below the HP2020 target (93.0%). Use was generally lowest among women without a high school education (59.3%), with U.S. residence <10 years (56.4%), and aged <65 years without health insurance coverage (57.4%).
For CRC screening (Table 4), 68.3% of adults were up-to-date (age-adjusted 67.9%), below the HP2020 target (70.5%). Use was generally lowest among those with U.S. residence <10 years (29.0%), no usual source of care (39.4 %) and aged <65 years without health insurance coverage (31.1%).
DISCUSSIONBased on 2019 NHIS estimates, three-quarters of women eligible for breast and cervical cancer screening and two-thirds of those eligible for CRC screening were up-to-date. Each was below HP2020 targets, although CRC test use approached its target. Some population subgroups exceeded breast and CRC targets while many did not. Screening disparities were observed by education, income, health insurance coverage, and U.S. residence duration, as previously reported.2,3 Nonmetropolitan counties had lower test use, a finding relevant to rural health. Lack of healthcare access has been associated with lower cancer screening uptake.8‒12 Consistent with this, lacking health insurance coverage or a usual source of care were strongly associated with lower test use. Although the disparities in the current analysis are not new,2,3 their persistence and the continued failure to meet national targets for screening test use are important. Such results can help inform efforts to implement strategies to address screening barriers and healthcare inequities. As the population grows and ages, more adults will be at risk for cancer13 and in need of screening, further underscoring efforts around promoting and facilitating recommended screening.
The 2019 NHIS redesign5 has implications for interpreting findings. The extent to which changes in questions and methods may have influenced estimates is uncertain, and changes limit direct comparison with prior years.14 Differences between 2019 estimates for breast and cervical cancer screening and those based on the 2018 NHIS (72.4% and 82.9%, respectively)2 could reflect changes in test use, survey changes, or both. Little change was reported in these 2 screenings from 2005 to 2018.2 In 2018, the USPSTF added human papillomavirus (HPV) testing alone as a cervical cancer screening option for women aged 30‒65 years.15 This option was included in this analysis and not the 2018 analysis,2 but this unlikely explains the lower proportion up-to-date in 2019. In general, increasing guideline complexity presents greater opportunity for missing or incomplete self-reported information about tests; differences in handling missing information could result in different estimates across studies. For CRC screening, differences between 2019 estimates and those using 2018 data (66.9%)2 were smaller. CRC screening use has increased over time,2 and findings could reflect that trend or survey changes. Although the redesign limits comparisons with earlier years, examining trends in future years will enable monitoring of screening use.
In March 2020, the WHO declared COVID-19 a pandemic,16 and subsequent reports documented reductions in cancer screening,17‒20 leading to concerns about the effects of delayed or canceled screenings on health outcomes.21,22 Evidence suggests at least a partial recovery in screening use over time.17,19,20 These 2019 findings can serve as a pre-pandemic baseline to compare with future estimates to assess recovery and growth in screening use.
LimitationsFindings are subject to limitations. Information is self-reported and not verified using medical records. Previous research demonstrated generally good validity for self-reported screening for breast, cervical and CRC, although some over-reporting has been noted.23‒25 Less is known about self-reported HPV tests for cervical cancer screening, suggesting an area for future research. Some variables had missing information for some respondents; thus, caution may be warranted in interpretating some subgroup estimates. The final sample adult response rate was 59.1%; therefore, nonresponse bias may be present despite survey weight adjustments. Up-to-date screening was defined as a relevant test for any reason among eligible adults within USPSTF-recommended intervals, consistent with HP2020 measures and earlier studies.2‒4,8,26 Furthermore, having a diagnostic test might result in being considered screened in effect. In an analysis of 2018 NHIS data, 95% of women who reported a recent mammogram said it was part of a “routine exam”,2 suggesting relatively few may be diagnostic. Similarly, an NHIS analysis examining CRC screening suggested most respondents indicated tests were done for screening purposes.8
CONCLUSIONSEstimated breast, cervical and CRC screening test use based on the 2019 NHIS remained below national targets. Continued monitoring will help to examine progress in screening uptake and eliminating disparities, and track recovery from changes in use resulting from the pandemic.
ACKNOWLEDGMENTSThe findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention or the National Cancer Institute. Use of trade names and commercial sources is for identification only and does not imply endorsement by the U.S. Department of Health and Human Services. The National Health Interview Survey was funded by the U.S. government. All authors are federal government employees. Dr. Croswell reports spousal stock ownership in Johnson and Johnson. The authors report no other conflicts of interest.
FootnotesCredit Statement
Susan Sabatino: Conceptualization, Methodology, Formal Analysis, Writing – Original Draft, Writing – Review & Editing, Project Administration. Trevor Thompson: Methodology, Software, Formal analysis, Data Curation, Writing – Review & Editing, Visualization. Mary White: Conceptualization, Methodology, Writing – Review & Editing. Jean Shapiro: Conceptualization, Methodology, Formal Analysis, Writing – Review & Editing. Tainya Clarke: Software, Validation, Writing – Review & Editing. Jennifer Croswell: Conceptualization, Writing – Review & Editing. Lisa Richardson: Writing - Review & Editing.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
REFERENCESRetroSearch is an open source project built by @garambo | Open a GitHub Issue
Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo
HTML:
3.2
| Encoding:
UTF-8
| Version:
0.7.3