Cervical Cancer Screening Rates in a Chart Review of Adolescent Patients at an Academic Institution before and after the Publication of the 2009 American College of Obstetricians and Gynecologists' Recommendations

National Trends in Cervical Cancer Screening in Adolescents

STUDY OBJECTIVE

To characterize patterns of cervical cancer screening among adolescents ages 14-20 years before and after the 2009 American College of Obstetrics and Gynecology recommendations.

DESIGN

Retrospective cohort study.

SETTING

National Ambulatory Medical Care Survey 2005-2016 data.

PARTICIPANTS

Female adolescents and young women ages 14-26 years.

INTERVENTIONS

Multivariable logistic regression models identified independent predictors of unindicated cervical cancer screening at office-based visits among adolescents ages 14-20 years.

MAIN OUTCOME MEASURES

Cervical cancer screening in women ages 14-20 and 21-26 years in 2 time periods: 2005-2008 and 2009-2016.

RESULTS

Between 2005 and 2016, 11,768 office visits were among adolescents ages 14-20 years. Overall, the cervical cancer screening rate for young women ages 14-20 years was 4.0%, which decreased from 4.5% to 0.4% (P = .008) during the study period. Adolescents who received cervical cancer screening during an office visit were older (18-20 years: 24.1% vs 14-17 years: 8.2%; P < .001), had a preventive care visit (preventive care: 79.7% vs other visit types: 20.3%; P < .001), and saw an obstetrician/gynecologist (obstetrician/gynecologist: 74.81% vs other specialties: 25.1%; P < .001). After adjusting for age, year, period, insurance status, region, and provider type, screening for cervical cancer was associated with living in the Southern region of the United States (adjusted odds ratio, 1.88; 95% confidence interval, 1.09-3.25; P = .02) and public insurance (adjusted odds ratio, 0.47, 95% confidence interval, 0.34-0.64; P < .001).

CONCLUSION

Despite recommendations, cervical cancer screening continued to occur in the adolescent population-especially older adolescents-creating unnecessary costs and potential harms. These findings show a slow uptake of guidelines nationally and the need for initiatives that encourage implementation of performance measures for providers, including ongoing provider and patient education.

Prevalence of and Factors Associated With Receiving a Pelvic Examination or Pap Testing Under the Age of 21 Years

PURPOSE

Current guidelines recommend that individuals receive their first Pap test at age 21 years and only receive a pelvic examination before age 21 years for clinical indications. We sought to determine the prevalence and associated covariates of receiving a pelvic examination or Pap test before 21 years of age.

METHODS

We analyzed the 2013-2015 National Survey of Family Growth. We conducted bivariate analyses comparing individuals who had and had not had a pelvic examination or Pap test and multivariable logistic regression to identify factors associated with having a pelvic examination or Pap test under 21 years.

RESULTS

This study included 1,170 individuals. Of respondents, 30.8% received a pelvic examination and 25.1% received a Pap test before 21 years of age. Receiving a pelvic examination was associated with being sexually active (adjusted odds ratio [aOR]: 6.6, 95% confidence interval [CI]: 3.8-11.7), having ever taken contraceptive pills (aOR: 2.6, 95% CI: 1.6-4.1) compared with no contraceptive method, and being screened for sexually transmitted infections in the past 12 months (aOR: 12.6, 95% CI: 7.3-21.8). Receiving a Pap test was also associated being sexually active (aOR: 7.2, 95% CI: 3.7-14.0), having ever taken contraceptive pills (aOR: 3.0, 95% CI: 1.9-4.7) compared with no contraceptive method, and being screened for sexually transmitted infections in the past 12 months (aOR: 8.94, 95% CI: 5.12-15.61).

CONCLUSIONS

Contrary to contemporary guidelines, a notable proportion of individuals under the age of 21 years continues to receive pelvic examinations and Pap testing.

Prevalence of Cervical Cancer Overscreening: Review of a Wellness Registry

Appropriately matching preventive health services and screenings with patient risk is an important quality indicator. Adherence by both providers and patients to cervical cancer screening guidelines has been inconsistent, resulting in overscreening and increased costs. This study examined the prevalence of cervical cancer overscreening following changes in screening guidelines in a wellness registry database. Cervical cancer overscreening after guideline implementation decreased for 18- to 20-year-old patients from 26.8% to 24.8% (P < .001) and increased for those aged 65 years and older from 11.1% to 12.5% (P = .0005). Black race, Hispanic ethnicity, Medicaid insurance, and the presence of a personal health record were associated with overscreening. Reliability and accuracy of data are a concern when data intended for one purpose, such as clinical care, are used for research. Correctly identifying screening tests in the electronic health record is important so that appropriate screening can be reliably assessed. In this study on the prevalence of cervical cancer overscreening, we used a focused chart review to identify whether screening Pap tests were accurately identified in the electronic medical record. Pap tests were correctly identified as screening in 85% of those aged 18 to 20, and in 74% of those aged 65 and older.

Impact of Significant Decreases Over Time in the Proportion of Sexually Active Medicaid Women Who Had Papanicolaou Testing or Were Pregnant on Trends of Overall Chlamydia Testing Rates

We examined Medicaid claims data during 2004 to 2013. The proportion of sexually active females aged 15 to 25 years who had Papanicolaou testing or were pregnant significantly decreased during 2004 to 2013 (67.0% to 43.9%, P < 0.05), resulting in a slowed increasing trend in overall chlamydia screening rates. Structural-level interventions for improving chlamydia screening are urgently needed.

Reducing overuse of cervical cancer screening: A systematic review

Overuse of clinical preventive services increases healthcare costs and may deprive underserved patients of necessary care. Up to 45% of cervical cancer screening is overuse. We conducted a systematic review of correlates of overuse of cervical cancer screening and interventions to reduce overuse. The search identified 25 studies (20 observational; 5 intervention). Correlates varied by the type of overuse measured (i.e., too frequent, before/after recommended age to start or stop screening, after hysterectomy), the most common correlates of overuse related to patient age (n = 7), OBGYN practice or provider (n = 5), location (n = 4), and marital status (n = 4). Six observational studies reported a decrease in overuse over time. Screening overuse decreased in all intervention studies, which used before-after designs with no control or comparison groups. Observational studies suggest potential targets for de-escalating overuse. Randomized clinical trials are needed to establish best practices for reducing overuse.

Cervical Cancer Screening and Follow-Up Procedures in Women Age <21 Years Following New Screening Guidelines

PURPOSE

The 2009 American College of Obstetricians and Gynecologists guidelines recommended no cervical cancer screening before age 21 years. We examined changes in screening, diagnostic, and treatment procedures for cervical dysplasia after guideline introduction, and cost implications.

METHODS

We studied Davidson County women aged 18-20 years, enrolled in Tennessee Medicaid, 2006-2014. We identified those with at least one Papanicolaou (Pap) test, human papillomavirus detection test, colposcopy, or excisional dysplasia treatment annually via Current Procedural Terminology coding. We used rate ratios with 95% confidence intervals to compare annual changes in procedure and treatment rates from 2014 to 2006. We counted total outcomes to estimate annual costs based on 2014 average procedural costs.

RESULTS

From 2006 to 2014, about 3,800 Davidson County women aged 18-20 years were enrolled in Medicaid annually. From 2006 to 2014, there were declines in Pap tests from 55.6 to 15.2 per 100 women (rate ratio .27, 95% confidence interval .25-.3); human papillomavirus tests from 13.8 to 5.9 per 100 (.42, .36-.5); colposcopy from 9.4 to 1.1 per 100 (.12, .08-.17); and dysplasia treatment from 1.1 to 0 per 100. The estimated cost of screening and procedures fell from $53 to $8 per enrolled woman, not accounting for changes in visits or complications associated with these procedures.

CONCLUSIONS

The 2009 screening guidelines were associated with major declines in screening, diagnostic, and treatment procedures for cervical dysplasia. Minimum estimated procedure and treatment costs saved were $45 per enrolled woman age 18-20 years.

The Impact of the American College of Obstetricians and Gynecologists Guideline Changes in Pap Tests on Annual Chlamydia Test Rates

PURPOSE

To assess impact of the 2009 American College of Obstetricians and Gynecologists (ACOG) Pap guideline changes on chlamydia testing rates among sexually active young women.

METHODS

The study included sexually active women aged 15-25 years enrolled in commercial health plans from 2005 to 2014. We identified sexually active women by diagnosis, procedure, and drug codes in inpatient, outpatient, and drug claims databases. We identified Pap tests and chlamydia tests among sexually active adolescents (15-20 years) and young adults (21-25 years) over time. Using piecewise regression models, we compared the change in chlamydia testing rates before and after 2009 ACOG guidelines.

RESULTS

From 2005 to 2014, chlamydia testing rates in sexually active women increased from 23% to 37% among adolescents and from 24% to 43% among young adults. Although the overall increase in chlamydia testing was positive, the annual rate of change in chlamydia testing (slope) decreased significantly after the 2009 ACOG guideline change from 1.9% before to 1.0% after for adolescents (p < .05) and from 2.5% to 1.7% for young adults (p < .05).

CONCLUSIONS

Although chlamydia test rates are increasing among sexually active women aged 15-25 years from 2005 to 2014, the slower growth in chlamydia testing rate after 2009 may relate to the change in the Pap testing guidelines. Our finding that more than half of sexually active women aged 15-25 years did not have chlamydia testing and that the rate of increased chlamydia testing slowed after 2009 suggests that interventions to improve chlamydia testing apart from combining with Pap testing are still needed.