Factors associated with loss to follow up after abnormal cervical cancer screening in pregnancy.

OBJECTIVE: To assess risk factors associated with loss to follow up in patients referred for colposcopy after abnormal cervical cytology during pregnancy in a Southern safety net hospital population. METHODS: An urban colposcopy center was queried for patients referred for follow up of abnormal cervical cytology during pregnancy and the postpartum period. Patients were identified through a standardized referral code in the electronic medical record. Multivariable logistic regression was used to compare patient characteristics between those who followed up for colposcopy and those lost to follow up. Independent risk factors assessed included age, parity, race, insurance, HIV status, history of mental illness, BMI, gestational age and trimester at screening, cytology at colposcopy referral, interval days until colposcopy, and biopsy histology. RESULTS: 1063 patients were identified, with 40.8% of patients who completed referred colposcopy. Patient characteristics predictive for colposcopy follow up included: maternal age at referral cervical cytology >30 years (1.67; 1.27-2.20; < 0.003), gestational age < 18 weeks at abnormal cervical cytology (1.57; 1.23-2.01; <0.0002), maternal race non-African American (2.20; 1.32-3.65; <0.0024) and with high grade cervical cytology (2.42; 1.81-3.24; <0.0001). CONCLUSION: In this population, inadequate follow up for abnormal cervical cytology during pregnancy is prominent, especially among those with younger maternal age, African American (AA) race, cervical cytology completed at later gestational ages of pregnancy, and low-grade initial cytology. Higher no-show rate among AA patients supports well-documented health disparities and need for further investigation and protocols to identify those at risk for loss to follow up.

An Alternative to Cytology in Triaging Cisgender Men and Transgender Women With HIV for High-Resolution Anoscopy.

OBJECTIVES: To evaluate high-risk human papillomavirus testing (hrHPV) as an alternative for anal cytology in screening for high-grade anal neoplasia (AIN2-3) among males with HIV. To identify predictive risk factors for AIN2-3 and develop a clinical tool to triage males with HIV for high-resolution anoscopy (HRA) without cytology. DESIGN: Retrospective cohort study of 199 adult cisgender men and transgender women with HIV referred to an anal neoplasia clinic in the Southeastern United States between January 2018 and March 2021. METHODS: Each subject underwent cytology, hrHPV, and HRA. Clinical and sociodemographic risk factors were collected for each subject. Significant risk factors for AIN2-3 were identified using logistic regression, and a triage tool incorporating these factors was developed. Screening test characteristics were calculated for cytology with and without adjunct hrHPV, hrHPV alone, and the triage tool. RESULTS: In multivariate analysis, significant predictors of AIN2-3 were hrHPV positivity (odds ratio [OR] = 11.98, CI = 5.58-25.69) and low CD4 count (OR = 2.70, CI = 1.20-6.11). There was no significant difference in positive or negative predictive values among the tool, stand-alone hrHPV, and anal cytology with adjunct hrHPV. Sensitivity and specificity were not significantly different for stand-alone or adjunctive hrHPV testing. Compared with cytology, stand-alone hrHPV and the novel triage tool reduced unnecessary HRA referrals by 65% and 30%, respectively. CONCLUSIONS: Stand-alone hrHPV would have missed 11 of 74 AIN2-3 and generated 74 fewer unnecessary HRAs than current cytology-based screening patterns, which led to 115 unnecessary HRAs in our cohort. We propose triaging those with low CD4 count, hrHPV positivity, and/or smoking history for HRA.