Abstract
BACKGROUND: While anal cancer screening is now recommended in several countries for high-incidence populations, barriers impede screening uptake. It is well known that patient recognition of symptoms increases cancer screening uptake, and we previously demonstrated that patients can detect small masses at the perianus and in the anal canal. In this current analysis, we aimed to longitudinally assess the effect of repetition and practice on patient self-recognition of small masses. METHODS: Individuals and couples of sexual minority men and transgender women were taught to conduct an anal self-examination or anal companion examination (ASE/ACE) at visit 1 and then randomized to a practice or control condition. Six months later, at visit 2, the effect of practice and repetition of the ASE/ACE on detection of abnormalities was assessed by comparing the lay exam results to those of a clinician. Concordance, κ, and area under the receiver operating characteristic curves (AUC) was used to quantify the comparison. FINDINGS: Concordance between lay exam and clinician exam increased from visit 1 (73%, 524/714) to visit 2 (95%, 535/561) (κ = 0.87, 95% CI 0.82-0.92). Overall AUC at visit 2 was 0.93 (95% CI 0.90-0.96). Although there was no difference in concordance between the practice (95%, 267/281) and control arms (96%, 268/280) (p = 0.69), concordance increased with ASE/ACE repetition (p(trend) < 0.001) and was 98% (354/363) for individuals performing the ASE/ACE ≥2 times between visits. For individuals with incident abnormalities at visit 2, concordance was 100% (27/27). Results did not differ by age or HIV status. False positive and false negative results by the ASE/ACE were 2% (10/561) and 3% (16/561), respectively. INTERPRETATION: Because lay individuals can detect anal abnormalities, clinicians conducting an anal examination may suggest that the patient's own lay exams may detect early invasive anal cancer. FUNDING: National Cancer Institute.