Abstract
Objectives. This study assessed the relationship between risk factors for cervical cancer and Papanicolaou (Pap) test use within the past year among reproductive-age women.
Methods. The 1995 National Survey of Family Growth, a demographic and reproductive health survey of 10 847 women aged 15 to 44, was analyzed with multiple logistic regression.
Results. Of the women, 62% reported having had a Pap test within the past year. Use was significantly higher among women with risk factors and among African American women. Use was significantly lower among uninsured, poor, and foreign-born women and among women with lower educational attainment and of “other” race/ethnicity.
Conclusions. Strategies to improve Pap test use include (1) educational campaigns that inform women of cervical cancer risk factors and encourage screening and (2) increased support for programs that expand access to Pap tests.
Most cases of cervical cancer are caused by sexually transmitted infections, principally certain types of the human papillomavirus.1,2 All sexually active women are at risk for cervical cancer, but the disease is more common among women with certain risk factors, such as early initiation of sexual intercourse, a history of multiple sexual partners (or partners with multiple sexual partners), or a history of sexually transmitted infections.3–5
Cervical cancer is one of the most preventable cancers because a precancerous condition can be identified early through the Papanicolaou (Pap) screening test.6,7 The incidence of precancerous lesions identified by the Pap test is highest among reproductive-age women.8 Of the 12 900 new US cases of cervical cancer in 2001, 44% occured among women aged 18 to 44 years.3,9 In the United States, most organizations recommend annual Pap tests once a woman has become sexually active (or has reached age 18), with some recommending less frequent screening following 3 normal test results.10
The objectives of this study were to describe the prevalence of selected risk factors for cervical cancer among reproductive-age women and to assess how the presence of risk factors is associated with Pap test use, controlling for sociodemographic characteristics and health insurance status.
METHODS
Data Sources
The National Survey of Family Growth is a demographic and reproductive health survey conducted by the National Center for Health Statistics. In 1995, 10 847 women aged 15 to 44 years were interviewed in their homes regarding their pregnancy and birth history, marriage and cohabitation history, sexual partner history, contraceptive use, diseases related to fertility, sex behaviors, and use of Pap tests in the past year.11,12 The National Survey of Family Growth response rate was 79%.
Statistical Analyses
The National Survey of Family Growth has a complex survey design involving stratification, clustering, and disproportionate sampling. All proportions and population counts presented here were weighted to provide national estimates. Variance estimates for proportions and logistic regression model odds ratios (ORs) were calculated by use of the Taylor series approximation technique, taking into account the complex design of the survey.13
RESULTS
Risk Factors for Cervical Cancer
In 1995, 43.2% of the women reported at least 1 sexual practice or a reproductive health history that increased the risk for cervical cancer. The most frequently reported risk factors were initiating sex at age 15 or younger (25.9%) and having had 10 or more sexual partners (15.9%) (Table 1 ▶).
TABLE 1.
—Distribution of Risk Factors Associated With Cervical Cancer and Papanicolaou (Pap) Test Use Within Risk Groups Among Reproductive-Age Women: National Survey of Family Growth, 1995
| Risk Factor (Sample Size) | National Estimate, in 1000s | % (95% CI) | % Having Pap Test in Past Year (95% CI)a |
| All women (10 847) | 60 201 | 100.0 | 61.9 (60.7, 63.1) |
| Composite riskb (10 847) | |||
| Any risk factors | 26 035 | 43.2 (42.0, 44.4) | 68.1 (66.5, 69.8) |
| No risk factors | 34 166 | 56.8 (55.6, 57.9) | 57.2 (55.7, 58.6) |
| Age at initiation of sex, y (9904) | |||
| ≤ 15 | 13 944 | 25.9 (24.9, 26.9) | 64.5 (62.4, 66.6) |
| 16–17 | 17 504 | 32.5 (31.4, 33.7) | 70.0 (68.1, 71.8) |
| 18–19 | 11 700 | 21.7 (20.8, 22.7) | 67.8 (65.7, 70.0) |
| 20–24 | 8909 | 16.6 (15.7, 17.4) | 66.7 (63.9, 69.5) |
| 25–29 | 1435 | 2.7 (2.3, 3.1) | 61.4 (55.7, 67.1) |
| 30–44 | 309 | 0.6 (0.4, 0.8) | 55.0 (40.7, 69.3) |
| No. of sexual partners in lifetime (10 847) | |||
| 0 | 6009 | 10.0 (9.2, 10.7) | 16.4 (13.8, 19.1) |
| 1 | 13 978 | 23.2 (22.2, 24.2) | 59.9 (57.7, 62.2) |
| 2 | 7480 | 12.4 (11.7, 13.1) | 65.2 (62.4, 68.0) |
| 3 | 5925 | 9.8 (9.2, 10.5) | 68.7 (65.7, 71.8) |
| 4 | 5002 | 8.3 (7.8, 8.8) | 70.3 (67.0, 73.7) |
| 5 | 4881 | 8.1 (7.5, 8.7) | 69.6 (66.2, 72.9) |
| 6 | 3041 | 5.0 (4.6, 5.5) | 72.8 (68.4, 77.2) |
| 7 | 2071 | 3.4 (3.1, 3.8) | 66.7 (61.2, 72.2) |
| 8 | 1504 | 2.5 (2.2, 2.8) | 69.6 (63.8, 75.4) |
| 9 | 724 | 1.2 (1.0, 1.4) | 72.1 (63.0, 81.1) |
| ≥ 10 | 9585 | 15.9 (15.0, 16.8) | 71.7 (69.3, 74.1) |
| History of pelvic inflammatory disease (10 844) | 4561 | 7.6 (7.0, 8.1) | 73.4 (70.1, 76.8) |
| History of sexually transmitted disease (STD)c | |||
| Chlamydia (9888) | 2557 | 4.8 (4.2, 5.3) | 76.5 (72.7, 80.4) |
| Genital herpes (9895) | 1236 | 2.3 (1.9, 2.7) | 76.6 (69.5, 83.6) |
| Gonorrhea (9895) | 1054 | 2.0 (1.6, 2.3) | 62.2 (55.2, 69.3) |
| Genital warts (9893) | 2431 | 4.5 (4.0, 5.1) | 75.9 (71.7, 80.2) |
| Syphilis (9897) | 191 | 0.4 (0.2, 0.5) | 72.9 (55.0, 90.8) |
| Any STD (10 847) | 6218 | 10.3 (9.5, 11.2) | 75.2 (72.5, 77.9) |
| Male sexual partner having sex with other female partners (9130)c | 7095 | 14.3 (13.4, 15.1) | 71.1 (68.4, 73.8) |
Note. CI = confidence interval.
aWomen were shown a card listing several medical services and asked if they had received any of them from a doctor or other medical care provider. Pap test was listed and described as a “sample or test for cancers of the cervix or uterus.”
bWomen with at least 1 of the following 5 risk factors: age at initiation of sex of 15 or younger, 10 or more lifetime sexual partners, history of pelvic inflammatory disease, history of sexually transmitted disease, and male sexual partner in past 12 months having sex with other female partners around the same time.
cQuestions pertaining to sexually transmitted disease, numbers of sexual partners, and the sexual practices of male sexual partners were asked via audio-CASI (computer-assisted self-interview). The principal author obtained permission from the National Center for Health Statistics to use the audio-CASI portion of the interview, which is stored in the “Omitted Items File.”
Pap Test Use
The majority (61.9%) of the reproductive-age women reported having had a Pap test within the past year (Table 2 ▶). The finding in bivariate analyses that Pap test use was significantly higher among women with at least 1 risk factor, compared with women without risk factors (68.1% vs 57.2%), was confirmed in multivariate analyses (OR = 1.63) (Table 2 ▶). Pap test use was significantly lower among women who were uninsured (OR = 0.54), and was comparable among women who had Medicaid coverage (OR = 1.10), relative to privately insured women. Women who were poor (i.e., resided in family with incomes below 150% of the poverty level) had lower rates of Pap test use (OR = 0.63) than did women residing in families with incomes at or above 300% of the poverty level. Relative to women with at least a college degree, women with less education had lower rates of Pap test use (e.g., less than high school graduation, OR = 0.49).
TABLE 2.
—Number and Distribution of Reproductive-Age Women, Percentage Who Reported Receipt of a Papanicolaou (Pap) Test in Past Year and Results of Multivariate Logistic Regression Model, by Health and Sociodemographic Characteristics: National Survey of Family Growth, 1995
| National Estimate, in 1000s | % Distribution (95% CI) | % Using Pap Test (95% CI)a | OR (95% CI) | |
| All womenb | 60 201 | 100.0 | 61.9 (60.7, 63.1) | . . . |
| High riskc | ||||
| Yes | 26 035 | 43.2 (42.0, 44.4) | 68.1 (66.5, 69.8) | 1.63 (1.46, 1.81) |
| No | 34 166 | 56.8 (55.6, 57.9) | 57.2 (55.7, 58.6) | . . . |
| Insurance | ||||
| Medicaid only | 7259 | 12.1 (11.2, 12.9) | 61.1 (58.2, 64.1) | 1.10 (0.93, 1.29) |
| No insurance | 7011 | 11.6 (10.9, 12.4) | 46.6 (43.3, 49.8) | 0.54 (0.46, 0.64) |
| At least some private insurance | 45 498 | 75.6 (74.4, 76.7) | 64.6 (63.2, 66.0) | . . . |
| Age at interview, y | ||||
| ≤ 17 | 5496 | 9.1 (8.4, 9.8) | 23.0 (20.2, 25.9) | 0.43 (0.33, 0.56) |
| 18–19 | 3573 | 5.9 (5.4, 6.5) | 50.5 (45.9, 55.2) | 1.13 (0.88, 1.46) |
| 20–24 | 8946 | 14.9 (14.1, 15.7) | 69.2 (66.5, 72.0) | 2.06 (1.73, 2.45) |
| 25–29 | 9794 | 16.3 (15.6, 17.0) | 70.6 (68.1, 73.2) | 1.72 (1.46, 2.01) |
| 30–34 | 10 982 | 18.3 (17.5, 19.0) | 69.4 (67.0, 71.9) | 1.51 (1.29, 1.77) |
| 35–39 | 11 297 | 18.8 (18.1, 19.5) | 63.2 (60.6, 65.8) | 1.06 (0.91, 1.24) |
| ≥ 40 | 10 015 | 16.7 (16.1, 17.5) | 62.4 (60.2, 64.7) | . . . |
| Marital status | ||||
| Never married | 22 679 | 37.7 (36.5, 38.9) | 52.1 (50.0, 54.3) | 0.61 (0.52, 0.71) |
| Currently married | 29 673 | 49.3 (48.1, 50.4) | 68.6 (67.1, 70.1) | . . . |
| Formerly married | 7849 | 13.0 (12.3, 13.8) | 64.8 (62.4, 67.3) | 0.87 (0.76, 1.00) |
| Educational attainmentd | ||||
| < High school graduation | 5396 | 11.3 (10.4, 12.2) | 51.7 (48.5, 55.0) | 0.49 (0.41, 0.58) |
| High school graduation or general equivalency diploma | 18 177 | 38.1 (36.8, 39.5) | 65.8 (64.0, 67.7) | 0.71 (0.62, 0.82) |
| Some college, no degree | 8772 | 18.4 (17.5, 19.3) | 68.2 (65.6, 70.7) | 0.75 (0.64, 0.86) |
| ≥ College degree | 15 332 | 32.2 (30.8, 33.5) | 73.5 (71.4, 75.5) | . . . |
| Poverty-level income,e % | ||||
| 0–149 | 13 586 | 22.6 (21.5, 23.6) | 55.3 (52.9, 57.8) | 0.63 (0.54, 0.73) |
| 150–299 | 19 618 | 32.6 (31.5, 33.7) | 57.2 (55.3, 59.0) | 0.69 (0.61, 0.77) |
| ≥ 300 | 26 995 | 44.8 (43.5, 46.2) | 68.6 (67.1, 70.2) | . . . |
| Race/ethnicity | ||||
| Hispanic | 6702 | 11.1 (9.9, 12.4) | 52.3 (49.6, 54.9) | 0.92 (0.80, 1.07) |
| Non-Hispanic White | 42 522 | 70.6 (69.0, 72.3) | 63.3 (61.9, 64.6) | . . . |
| Non-Hispanic Black | 8210 | 13.6 (12.4, 14.8) | 67.6 (65.1, 70.1) | 1.63 (1.42, 1.87) |
| Non-Hispanic other | 2767 | 4.6 (3.8, 5.4) | 47.6 (40.8, 54.5) | 0.66 (0.48, 0.90) |
| Birthplace | ||||
| United States | 54 419 | 90.4 (89.5, 91.3) | 63.0 (61.8, 64.2) | . . . |
| Outside of United States | 5782 | 9.6 (8.7, 10.5) | 51.5 (48.0, 55.0) | 0.79 (0.66, 0.96) |
| Language of interview | ||||
| Spanish | 1390 | 2.3 (1.9, 2.7) | 43.4 (37.2, 49.5) | 0.96 (0.69, 1.35) |
| English | 58 811 | 97.7 (97.3, 98.1) | 62.3 (61.2, 63.5) | . . . |
Note. Ellipses indicate referent category for the multiple logistic regression model (model includes only variables shown in table).
aWomen were shown a card listing several medical services and asked if they had received any of them from a doctor or other medical care provider. Pap test was listed and described as a “sample or test for cancers of the cervix or uterus.”
bSample size was 10 847. Estimates of Pap test use include women reporting hysterectomy (5% of the women). Some women with hysterectomies have a uterine cervix and are candidates for Pap tests (presence of cervix was not determined in the National Survey of Family Growth).
cWomen reporting at least 1 of the following 5 risk factors: age at initiation of sex of 15 or younger, 10 or more lifetime sexual partners, history of pelvic inflammatory disease, history of sexually transmitted disease, and male sexual partner in past 12 months having sex with other female partners around the same time.
dDescriptive statistics for educational attainment were limited to women aged 22 years and older at time of interview (sample size was 8868).
ePoverty-level income was based on the respondents' combined family income from all sources in the 12 months before the survey divided by the 1994 poverty thresholds established by the US Bureau of the Census.
Relative to non-Hispanic White women, non-Hispanic Black women reported significantly higher rates of Pap test use (OR = 1.63), whereas women classified as being of non-Hispanic “other” race/ethnicity (e.g., Asian and Pacific Islander, Native American descent) had significantly lower Pap test use (OR = 0.66). Hispanic women's Pap test use was not significantly different from that of non-Hispanic White women (OR = 0.92). Women born outside of the United States had significantly lower Pap test use than did native-born women (OR = 0.79).
Location of Pap Tests
Pap tests were most often conducted at private doctors' offices or health maintenance organizations (78.9%; 95% confidence interval [CI] = 77.5%, 80.3%) and less often conducted in clinic settings (8.0%; 95% CI = 7.1%, 8.9% at publicly funded Title X family planning clinics; 10.2%; 95% CI = 8.9%, 11.5% at other clinics) or at hospitals, schools, or other settings (2.8%; 95% CI = 2.4%, 3.3%). Compared with women reporting no risk factors, women reporting at least 1 risk factor were more likely to have received their Pap tests at a clinic funded through the Title X program (10.1%; 95% CI = 8.8%, 11.4% vs 6.1%; 95% CI = 5.1%, 7.1%) and less likely to have been tested at a private doctor's office or at a health maintenance organization (74.6%; 95% CI = 72.9%, 76.4% vs 82.8%; 95% CI = 81.0%, 84.6%).
DISCUSSION
Results of these analyses suggested that the presence of cervical cancer risk factors was associated with higher Pap test use; that African American and Hispanic women have achieved rates of Pap test use comparable to or greater than those of other women; and that lack of health insurance, low educational attainment, poverty, and being of “other” race/ethnicity or foreign born were associated with lower rates of Pap test use. The sociodemographic correlates of Pap test use in these analyses were consistent with those identified in other studies.14,15
Higher-risk women might have enhanced opportunities for screening because of greater exposure to health care providers—nearly half (47.4%; 95% CI = 45.7%, 49.1%) of the women reporting risk factors said that they had used a reproductive health service (e.g., family planning services, prenatal care) in the previous year, compared with just over a third (34.9%; 95% CI = 33.6%, 36.2%) of the women who did not report any risk factors. Almost all (86.5%; 95% CI = 85.5%, 87.6%) women who had received a reproductive health service reported having received a Pap test, suggesting that the test is routinely provided during family planning and pregnancy-related visits.
Despite higher rates of Pap test use among women with risk factors, nearly one third of the higher-risk women reported that they did not have a Pap test in the past year, leaving much room for improvement. Strategies to improve Pap test use include implementation of educational campaigns that inform women of cervical cancer risk factors and encourage screening and provision of increased support for programs that expand access to Pap tests. The largest program to promote cancer screening among low-income and underserved women is the National Breast and Cervical Cancer Early Detection Program, which operates in all states with support from the Centers for Disease Control and Prevention. From 1991 to 1997, more than a million Pap tests were performed as part of the program, but estimates are that fewer than 15% of the women eligible for the program are served.16–18 Family planning clinics that offer services on a free and reduced-fee basis (e.g., Title X clinics) also provide opportunities to increase the use of Pap tests, especially among women at higher risk for cervical cancer.19
One caution to interpreting results of cancer screening behavior from surveys is the problem of respondents misrepresenting their actual behavior.20–23 An inability to recall events, the desire on the part of respondents to provide socially desirable answers, or the failure to correctly date events in memory can all contribute to misrepresenting Pap use in surveys.24 Despite these shortcomings, the 1995 National Survey of Family Growth provided valuable information on the determinants of Pap tests and descriptive information on where tests are conducted.
Acknowledgments
We would like to acknowledge the assistance of Anjani Chandra, PhD, Demographer, National Survey of Family Growth, National Center for Health Statistics. She provided the National Survey of Family Growth confidential data and answered questions about the data and their use.
Note. The analysis, opinions, and assertions contained herein are those of the authors and are not to be construed as reflecting the views or position of the National Academy of Sciences, the Institute of Medicine, or the National Research Council.
M. Hewitt conducted analyses of the National Survey of Family Growth and drafted the paper. S. Devesa provided tabulations from the Surveillance, Epidemiology, and End Results Program on the epidemiology of cervical cancer and assisted in the analysis plan and the writing of the paper. N. Breen reviewed background literature and assisted in the analysis plan and the writing of the paper.
Peer Reviewed
References
- 1.Eng TR, Butler WT, eds. The Hidden Epidemic: Confronting Sexually Transmitted Diseases. Washington, DC: National Academy Press; 1997. [PubMed]
- 2.Cain JM, Howett MK. Preventing cervical cancer. Science. 2000;288:1753–1754. [DOI] [PubMed] [Google Scholar]
- 3.American Cancer Society. Cancer Facts & Figures—2001. Atlanta, Ga: American Cancer Society; 2001.
- 4.American Cancer Society. Cancer Risk Report: Prevention and Control. Atlanta, Ga: American Cancer Society; 1998.
- 5.Devesa SS. Cancer in women. In: Goldman MB, Hatch MC, eds. Women and Health. San Diego, Calif: Academic Press; 2000:863–870.
- 6.National Institutes of Health Consensus Program. Cervical cancer. NIH Consensus Statement. 1996;14(1):1–38. [PubMed] [Google Scholar]
- 7.Hunter RD. Carcinoma of the cervix. In: Peckham M, Pinedo H, Veronesi U, eds. Oxford Textbook of Oncology. Vol 2. Oxford, England: Oxford University Press; 1995:1324–1337.
- 8.Surveillance, Epidemiology, and End Results (SEER) Program of the National Cancer Institute. SEER*Stat 41. Available at: http://seer.cancer.gov/ScientificSystems/SEERStat/ [special tabulation]. Accessed October 24, 2000.
- 9.Ries LAG, Eisner MP, Kosary CL, eds. SEER Cancer Statistics Review, 1973–1997. Bethesda, Md: National Cancer Institute; 2000.
- 10.US Preventive Services Task Force. Guide to Clinical Preventive Services. 2nd ed. Baltimore, Md: Williams & Wilkins; 1996.
- 11.Abma JD, Chandra A, Mosher WD, Peterson LS, Piccinino LJ. Fertility, Family Planning, and Women's Health: New Data From the 1995 National Survey of Family Growth. Hyattsville, Md: National Center for Health Statistics; 1997. [PubMed]
- 12.National Center for Health Statistics. National Survey of Family Growth, Cycle V 1995 [CD-ROM Series 23 (no. 3)]. Hyattsville, Md: National Center for Health Statistics; 1997.
- 13.Stata Statistical Software: Release 6.0 [computer program]. College Station, Tex: Stata Corp; 1999.
- 14.Martin LM, Wingo PA, Calle EE, Heath CW. Comparison of mammography and Pap test use from the 1987 and 1992 National Health Interview Surveys: are we closing the gaps? Am J Prev Med. 1996;12:82–90. [PubMed] [Google Scholar]
- 15.Potosky AL, Breen N, Graubard BI, Parsons PE. The association between health care coverage and the use of cancer screening tests: results from the 1992 National Health Interview Survey. Med Care. 1998;36:257–270. [DOI] [PubMed] [Google Scholar]
- 16.The National Breast and Cervical Cancer Early Detection Program: At-a-Glance 2000. Atlanta, Ga: Centers for Disease Control and Prevention; 2000.
- 17.Blackman DK, Bennett EM, Miller DS. Trends in self-reported use of mammograms (1989–1997) and Papanicolaou tests (1991–1997)—Behavioral Risk Factor Surveillance System. MMWR Morb Mortal Wkly Rep CDC Surveill Summ. 1999;48(SS-6):1–22. [PubMed] [Google Scholar]
- 18.Lawson HW, Lee NC, Thames SF, Henson R, Miller DS. Cervical cancer screening among low-income women: results of a national screening program, 1991–1995. Obstet Gynecol. 1998;92:745–752. [DOI] [PubMed] [Google Scholar]
- 19.Frost JF. Family planning clinic services in the United States, 1994. Fam Plann Perspect. 1996;28:92–100. [PubMed] [Google Scholar]
- 20.Sudman S, Warnecke R, Johnson T, O'Rourke D, Davis AM. Cognitive aspects of reporting cancer prevention examinations and tests. Vital Health Stat 6. 1994;No. 7.
- 21.Bowman JA, Sanson-Fisher R, Redman S. The accuracy of self-reported Pap smear utilization. Soc Sci Med. 1997;44:969–976. [DOI] [PubMed] [Google Scholar]
- 22.Montano DE, Phillips WR. Cancer screening by primary care physicians: a comparison of rates obtained from physician self-report, patient survey, and chart audit. Am J Public Health. 1995;85:795–800. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Suarez L, Goldman DA, Weiss NS. Validity of Pap smear and mammogram self-reports in a low-income Hispanic population. Am J Prev Med. 1995;11:94–98. [PubMed] [Google Scholar]
- 24.Groves RM. Survey Errors and Survey Costs. New York, NY: John Wiley & Sons; 1989.