Appropriate management of borderline nd low grade abnormalities on Papanicolaou (pap) smears is high controversial. The goal of screening is to prevent invasive cervical cancers while minimizing physical and psychological morbidity as well as unnecessary use of resources. This project will construct a cost-effectiveness model of the management of patients with borderline and low grade abnormal pap smears, contrasting a strategy of immediate colposcopy and biopsy wit a more conservative strategy of follow-up with repeat pap smears. Data to be included in the model will be derived from a structured literature review and meta- analysis, and from data collection conducted in collaboration with Planned Parenthood of Sacramento Valley (PPSV). Data collection will include retrospective estimation of patient adherence to recommended follow-up for colposcopy or repeat pap smears, correlation of cytology with histology to determine the probability of a low grade or borderline pap smear resulting from a high grade lesion on the cervix and evaluation of the clinical usefulness of repeating the pap smear at the time of colposcopy. Costs for the alternative strategies will be evaluated from both a fee for service and managed care perspective, and data on patient preferences will be collected prospectively. Two hundred women patients at PPSV clinics will be interviewed, and patient preferences will be formally assessed using direct scaling and time trade-off methods. Probabilities and costs for the models will be derived from the meta- analysis and original studies as described. Probabilities which cannot be determined from any of these methods will be estimated by an expert panel using a modified Delphi approach. Model outcomes will include the development of invasive cervical cancer, progression to high grade cervical neoplasia, morbidity related to diagnostic evaluations and treatment, and quality adjusted life years. The cost per cervical cancer or high grade lesion avoided, and per quality adjusted life year gained will be calculated. The marginal cost per cancer or high grade lesion avoided and quality adjusted life year gained for the more effective strategy will be calculated.
