Assessing Management of Abnormal Cervical Cancer Screening Results and Concordance with Guideline Recommendations in Three US Healthcare Settings

Abstract

Background: Follow-up of abnormal results is essential to cervical cancer screening, but data on adherence to follow-up are limited. We describe patterns of follow-up after screening abnormalities and identify predictors of guideline-concordant follow-up.

Methods: We identified the index screening abnormality (positive human papillomavirus test or atypical squamous cells of undetermined significance or more severe cytology) among women of ages 25 to 65 years at three US healthcare systems during 2010 to 2019. We estimated the cumulative incidence of surveillance testing, colposcopy, or treatment after the index abnormality and initial colposcopy. Logistic regressions were fit to identify predictors of guideline-concordant follow-up according to contemporaneous guidelines.

Results: Among 43,007 patients with an index abnormality, the cumulative incidence of any follow-up was 49.6% by 4 years for those with atypical squamous cells of undetermined significance/human papillomavirus-negative and higher for abnormalities warranting immediate colposcopy. The 1-year cumulative incidence of any follow-up after colposcopy was 70% for patients with normal results or cervical intraepithelial neoplasia I and 90% for patients with cervical intraepithelial neoplasia II+. Rates of concordant follow-up after screening and colposcopy were 52% and 47%, respectively. Discordant follow-up was associated with factors including age, race/ethnicity, overweight/obese body mass index, and specific types of public payor coverage or being uninsured.

Conclusions: Adherence to the recommended follow-up of cytologic and histopathologic abnormalities is inconsistent in clinical practice. Concordance was poor for mild abnormalities and improved, although suboptimal, for more severe abnormalities.

Impact: There remain gaps in the cervical cancer screening process in clinical practice. Further study is needed to understand the barriers to the appropriate management of cervical abnormalities.

Figure 1:. Cumulative incidence of any follow-up, repeat testing, immediate colposcopy, and immediate treatment after an index ASC-US/HPV-negative (Panel A) and index NILM/HPV-positive, LSIL-HPV-negative, or ASC-US alone result (Panel B), overall and by healthcare system.

Each sub-panel shows the cumulative incidence of overall (“Any follow-up”) and specific types of follow-up (“Repeat testing”, “Immediate colposcopy”, “Immediate treatment”), among the total sample (black) and stratified by patients at MGB (blue), PH (red), and KPWA (green). The left set of four sub-panels (Panel A) shows cumulative incidences after an index ASC-US/HPV-negative screening result, and the right set (Panel B) shows cumulative incidences after an index NILM/HPV-positive, LSIL-HPV-negative, or ASC-US alone screening result. Cumulative incidences are estimated based on the time from index abnormality to the first of either a repeat test, colposcopy, or treatment procedure that occurs thereafter, considering occurrence of another type of follow-up as a competing event. Repeat testing refers to having both cytology and HPV test within 14 days of each other, without simultaneously receiving a colposcopy or treatment procedure. Colposcopy events include colposcopy with or without biopsy, endocervical curettage, or biopsy without a record of colposcopy. Treatment events include loop electrosurgical excision procedure, cold knife conization, cryotherapy, laser ablation, total or radical hysterectomy, trachelectomy, or excisional procedure not otherwise specified. Any follow-up refers to a composite event consisting of either repeat testing, colposcopy, or treatment. The number of patients at risk at each time point along the x-axis is reported by healthcare system below each figure. The recommended follow-up in parentheses of titles are based on the 2006 and 2012 ASCCP guidelines for the corresponding index abnormalities.

Figure 2:. Cumulative incidence of any follow-up, repeat testing, immediate colposcopy, and immediate treatment after an index HPV16/18+, ASCUS/HPV-positive, LSIL/HPV-positive, LSIL on cytology alone, HSIL, ASC-H, AGC, adenocarcinoma in situ (AIS), or suspicious cancer result, overall and by healthcare system.

Each sub-panel shows the cumulative incidence of overall (“Any follow-up”) and specific types of follow-up (“Repeat testing”, “Immediate colposcopy”, “Immediate treatment”), among the total sample (black) and stratified by patients at MGB (blue), PH (red), and KPWA (green). This set of four sub-panels shows cumulative incidences after HPV16/18+, ASC/HPV-negative, LSIL/HPV-positive, LSIL on cytology alone, HSIL, ASC-H, AGC, adenocarcinoma in situ (AIS), or suspicious cancer results. Cumulative incidences are estimated based on time from index abnormality to the first of either a repeat test, colposcopy, or treatment procedure that occurs after the index abnormality, considering occurrence of another type of follow-up as a competing event. Repeat testing refers to having both cytology and HPV test within 14 days of each other, without simultaneously receiving a colposcopy or treatment procedure. Colposcopy events include colposcopy with or without biopsy, endocervical curettage, or biopsy without a record of colposcopy. Treatment events include loop electrosurgical excision procedure, cold knife conization, cryotherapy, laser ablation, total or radical hysterectomy, trachelectomy, or excisional procedure not otherwise specified. Any follow-up refers to a composite event consisting of either repeat testing, colposcopy, or treatment. The number of patients at risk at each time point along the x-axis is reported by site below each figure. The recommended follow-up in parentheses of titles are based on the 2006 and 2012 ASCCP guidelines for the corresponding index abnormalities.

Figure 3:. Cumulative incidence of any follow-up, repeat testing, immediate colposcopy, and immediate treatment after an initial colposcopy with a normal or CIN I result (Panel A), or CIN II or CIN III or more severe result (Panel B), overall and by healthcare system.

Each sub-panel shows the cumulative incidence of overall (“Any follow-up”) and specific types of follow-up (“Repeat testing”, “Immediate colposcopy”, “Immediate treatment”), among the total sample (black) and stratified by patients at MGB (blue), PH (red), and KPWA (green). The left set of four sub-panels (Panel A) shows cumulative incidences after an initial colposcopy with a normal or CIN I result, and the right set (Panel B) shows cumulative incidences after a CIN II or CIN III or more severe result. Cumulative incidences are estimated based on time from index abnormality to the first of either a repeat test, colposcopy, or treatment procedure that occurs after the index abnormality, considering occurrence of another type of follow-up as a competing event. Repeat testing refers to having both cytology and HPV test within 14 days of each other, without simultaneously receiving a colposcopy or treatment procedure. Colposcopy events include colposcopy with or without biopsy, endocervical curettage, or biopsy without a record of colposcopy. Treatment events include loop electrosurgical excision procedure, cold knife conization, cryotherapy, laser ablation, total or radical hysterectomy, trachelectomy, or excisional procedure not otherwise specified. Any follow-up refers to a composite event consisting of either repeat testing, colposcopy, or treatment. The number of patients at risk at each time point along the x-axis is reported by site below each figure. The recommended follow-up in parentheses of titles are based on the 2006 and 2012 ASCCP guidelines for the corresponding index abnormalities.