Acceptability of High-Resolution Anoscopy for Anal Cancer Screening in HIV-infected Patients

JO Lam; GM Barnell; M Merchant; CG Ellis; MJ Silverberg


HIV Medicine. 2018;19(10):716-723. 

In This Article


Study Population and Data Collection

This study used data for HIV-infected patients who had a first-time HRA at Kaiser Permanente Northern California (KPNC), an integrated health care system serving over 3.9 million patients, representing 30% of insured Northern Californians.[24] HRA screening was offered at two facilities within KPNC – San Francisco and Oakland. The San Francisco HRA programme was located within an HIV primary care clinic and offered HRA screening to all HIV-infected patients. The Oakland HRA programme was in a general surgery department and served as an HRA referral clinic for the rest of KPNC. The two KPNC HRA clinics used different treatment protocols and modalities. Following HRA for evaluation and biopsy, patients at the San Francisco clinic were scheduled for a follow-up appointment for treatment with infrared coagulation if the biopsy was found to be a high-grade squamous intraepithelial lesion (HSIL). The Oakland clinic used HRA for evaluation and biopsy, and immediately treated lesions with electrocautery if clinically suspicious for HSIL. Procedures were performed by six providers, using recommended techniques previously described.[13] Patients were recommended to return for a repeat HRA in 3, 6 or 12 months, depending on extent of disease and pathological diagnosis.

The acceptability of HRA as a screening approach for detection of anal dysplasia was assessed via: (1) an EHR-based cross-sectional study of all HIV-infected KPNC patients who received their first KPNC HRA between July 2008 and December 2013, (2) an EHR-based analysis of patients who received their first KPNC HRA between July 2008 and December 2013 but did not return for a repeat HRA (a proxy for HRA acceptability) and (3) a survey mailed to all HIV-infected KPNC patients who completed their first HRA at the Kaiser Oakland facility between January 2014 and August 2014. To protect the identity of patients with HIV infection, the survey was mailed to all patients completing HRA at KPNC Oakland and responses from HIV-infected patients were included in this study. All patients were offered repeat HRA; however, to allow adequate observation time of repeat HRA, our analyses of repeat HRAs included only patients who had an initial HRA by December 2011, followed by at least 21 months of KPNC membership.

For the cross-sectional study, lack of acceptability of HRA was defined by any of the following noted in the patient's EHR: receipt of HRA with sedation, dispensation of opioid analgaesia, and visit to an emergency room (ER) or an urgent care department within 1 week of HRA. The presence of these indicators would suggest that the procedure was poorly tolerated and would therefore be unacceptably invasive for anal cancer screening. Potential demographic and clinical predictors of poor tolerance of HRA were also extracted from the EHR. The presence of a pre-existing anal condition – including wart, skin tag, polyp, bleeding, abscess, ulcer and/or pain – either as a presenting complaint at first HRA visit or documented within 90 days before HRA was considered a primary predictor of lack of HRA acceptability. Other demographic and health factors considered included the following: age, race/ethnicity, sexual risk (i.e. MSM, male non-MSM or female), history of smoking, CD4 count, HIV RNA level, antiretroviral therapy (ART) and prior clinical AIDS diagnosis. These same factors were assessed in the analysis of patients who did not return for a recommended repeat HRA, with lack of acceptability of initial HRA considered as a primary predictor.

For the survey, patients were asked to recall their experience with their first HRA, including: level of pain (Likert scale from 0 to 10; 0 = no pain and 10 = worst pain imaginable) during HRA, 2–3 days after the procedure and 1 week after the procedure; bleeding; impact of adverse effects on lifestyle/normal activities; and willingness to return for follow-up screening. For the measures of pain and bleeding, patients were asked whether the amount experienced was greater, equal to or less than the amount expected, and whether the amount experienced was acceptable to them. All patients surveyed had their HRA performed by the same practitioner (author GMB).

Statistical Analyses

Descriptive statistics were used to summarize patients' characteristics. For the analyses using EHR data, predictors of lack of HRA acceptability and likelihood of not returning for a repeat HRA were modelled using logistic regression. Statistical analyses were performed using sas 9.3 software (SAS Institute, Cary, NC).

For analyses of survey results, χ 2 tests were used to evaluate binary outcomes, Wilcoxon rank sum tests were used for comparisons of HRA tolerability across independent patient categories, Wilcoxon matched-pairs signed-rank tests were used for comparisons of within-individual (i.e. before-after) responses and Kruskall-Wallis tests were used to evaluate Likert scale responses. Statistical analyses were performed using stata 12 software (StataCorp, College Station, TX). Statistical significance was defined by a two-sided P–value < 0.05 for all analyses.

The study was approved by the KPNC Institutional Review Board, including waivers of written informed consent for the cross-sectional analysis. For the survey portion of the study, return of the survey served as documentation of informed consent.