Recorded presentation

Ana Maria Lopez, MD, MPH, FACP¹, Sarah Kurker, MSW¹, Jose Guillen, PhD¹, Janet Major, BS², Kenneth Hatch, MD¹, Francisco, A.R. Garcia, MD, MPH³, Bel Barker, MD⁴, Sydney Lazarus, MPH¹, DeeDee Avery, MS, MPH¹, and Ronald S. Weinstein, MD². (1) Arizona Cancer Center, University of Arizona, 1515 N. Campbell Ave., PO Box 245024, Tucson, AZ 85724-5024, 520-626-2271, alopez@azcc.arizona.edu, (2) Arizona Health Sciences Center, University of Arizona, PO Box 245032, Tucson, AZ 85721, (3) Department of Obstetrics & Gynecology, University of Arizona Health Science Center, P.O. Box 245078, Tucson, AZ 85724-5078, (4) OB/Gyn, Kaiser Permanente, 1188 N. Euclid Ave., 4th floor, Anaheim, CA 92801

Background: Although the mortality from invasive cervical cancer in the US has declined over 50% during the past 40 years, the decrease has not been as dramatic among minority and/or clinically underserved populations. Although colposcopy is considered the standard of care after an abnormal Pap smear, for women who live in underserved or rural areas, access to a skilled colposcopist may not be readily available. The ability to conduct a colposcopy via telemedicine, known as telecolposcopy, allows the medically underserved woman equal access to specialty or expert care that would not otherwise be readily available, if at all. Telemedicine technology can be used to transmit colposcopic images to an observer located at a distance from the patient. The colposcopist is then able to review saved images at a future time (store-forward). Although telecolposcopy technology is commercially available, the data on its clinical efficacy are limited. Objective: This study examined the efficacy and diagnostic accuracy of utilizing telecolposcopy by determining its sensitivity and specificity in a tertiary care setting. Methods: All participants received a clinical exam, an in-person exam, and a store-forward telecolposcopy exam. Three different examiners, experienced in colposcopy, assessed the store-forward images. Statistical measures included comparison of the clinical and telemedicine diagnoses, comparison of the clinical and telemedicine diagnoses with the biopsy results, inter-observer correlations, intra-observer correlations, time to assess the image, satisfaction with the image, confidence in the diagnosis, and patient acceptance. Results: One-hundred and ten women were consented for this study, 93 sets of cervical images were reviewed, and 64 biopsies were obtained. Results indicate that the sensitivity of telecolposcopy was significantly better than chance (p < 0.001). The positive predictive value of the colposcopic impression was highest for no CIN and high grade or invasive lesions. The positive predictive value of the telecolposcopic impression was highest in high grade or invasive lesions. The in-person exam diagnosis was consistent with the pathology diagnosis 62.12% with kappa= 0.4192. The telecolposcopy diagnoses were consistent with the pathological diagnosis 58.4%/kappa=0.3675, 58.49%/kappa=0.3435, 44.26%/kappa=0.1552. Reviewers found the quality of the images to be adequate and confidence in the diagnoses was high. Patients accepted the technology well and did not express discomfort that images would be transmitted. Conclusion: Telecolposcopy diagnosis agrees well with colposcopy in-person diagnosis. Telecolposcopy shows promising results in reducing health disparities in equal access to specialized care for rural/underserved women.

Learning Objectives: