Procedural and temporal compliance to cervical cancer screening guidelines in a CDC-funded program in Alaska

To estimate changes in clinical practice in a cervical cancer screening program after the American Society for Colposcopy and Cervical Pathology's issuance of 2006 consensus guidelines (2006CG) for managing abnormal Pap tests.

Screening and diagnostic procedure records were extracted from the operational database of a Centers for Disease Control and Prevention-funded cervical cancer screening program in Alaska. New software was developed to compare these records to the American Society for Colposcopy and Cervical Pathology algorithms. Of the 16 algorithms, 11 were encoded, and results of patients with abnormal Pap results in 2 study windows (cohort A: 2005-2006: n = 1,431; cohort B: October 2007 to September 2008: n = 1,369) were compared with 2006CG recommendations. Comparisons between the practices in the 2 study windows were made using the Fisher exact test.

Among adolescents with low-grade squamous intraepithelial lesion or atypical squamous cells of undetermined significance Pap results, 117 (48%) of 243 in cohort A were given colposcopies compared with 28 (23%) of 192 in cohort B (p < .0001). There was no difference in the rate of human papillomavirus testing between cohorts A and B (23% vs 26%, p = .449). Among adults with low-grade squamous intraepithelial lesion, 63 (16%) of 402 in cohort A were given repeat Pap tests instead of colposcopies compared with 25 (7%) of 367 in cohort B (p < .0001). Temporal compliance was significantly improved (p = .025), with 75% in cohort A and 91% in cohort B being graded as timely or early.

Clinical practice has moved toward 2006CG recommendations in this program, although gaps remain. This evaluation will be used to inform a compliance improvement program. The software can be easily modified as 2006CG evolves and can be ported to other algorithmic practice guidelines.