Search

BackgroundAccurate identification of new diagnoses of human papillomavirus–associated cancers and precancers is an important step toward the development of strategies that optimize the use of human papillomavirus vaccines. The diagnosis of human papillomavirus cancers hinges on a histopathologic report, which is typically stored in electronic medical records as free-form, or unstructured, narrative text. Previous efforts to perform surveillance for human papillomavirus cancers have relied on the manual review of pathology reports to extract diagnostic information, a process that is both labor- and resource-intensive. Natural language processing can be used to automate the structuring and extraction of clinical data from unstructured narrative text in medical records and may provide a practical and effective method for identifying patients with vaccine-preventable human papillomavirus disease for surveillance and research. ObjectiveThis study's objective was to develop and assess the accuracy of a natural language processing algorithm for the identification of individuals with cancer or precancer of the cervix and anus. MethodsA pipeline-based natural language processing algorithm was developed, which incorporated machine learning and rule-based methods to extract diagnostic elements from the narrative pathology reports. To test the algorithm’s classification accuracy, we used a split-validation study design. Full-length cervical and anal pathology reports were randomly selected from 4 clinical pathology laboratories. Two study team members, blinded to the classifications produced by the natural language processing algorithm, manually and independently reviewed all reports and classified them at the document level according to 2 domains (diagnosis and human papillomavirus testing results). Using the manual review as the gold standard, the algorithm’s performance was evaluated using standard measurements of accuracy, recall, precision, and F-measure. ResultsThe natural language processing algorithm’s performance was validated on 949 pathology reports. The algorithm demonstrated accurate identification of abnormal cytology, histology, and positive human papillomavirus tests with accuracies greater than 0.91. Precision was lowest for anal histology reports (0.87, 95% CI 0.59-0.98) and highest for cervical cytology (0.98, 95% CI 0.95-0.99). The natural language processing algorithm missed 2 out of the 15 abnormal anal histology reports, which led to a relatively low recall (0.68, 95% CI 0.43-0.87). ConclusionsThis study outlines the development and validation of a freely available and easily implementable natural language processing algorithm that can automate the extraction and classification of clinical data from cervical and anal cytology and histology.

BackgroundAscertaining history of prior immunization with human papillomavirus (HPV) vaccine can be challenging and resource-intensive. Computer-assisted self-interviewing instruments have the potential to address some of the challenges of self-reporting, and may also reduce the time, costs, and efforts associated with ascertaining immunization status. ObjectiveThis study assesses both the feasibility and the accuracy of a computer-assisted self-interviewing instrument to ascertain a patient’s history of immunization with the HPV vaccine. MethodsWe developed both a survey and a Web-based data collection system using computer-assisted self-interviewing to ascertain self-reported HPV vaccine immunization history. We implemented the instrument in a sample of adult women enrolled in an ongoing study of the HPV vaccine. Vaccine records from prior sources of care were reviewed to verify reported immunization history. ResultsAmong the 312 participants who provided HPV vaccine immunization history by self-report, almost all (99%) were able to do so using the computer-assisted self-interviewing instrument. The median survey completion time was 10 minutes (IQR 7-17). The accuracy of self-report was 84%, sensitivity was 89%, specificity was 80%, and the negative predictive value was 92%. ConclusionsWe found that it is feasible to collect a history of immunization with the HPV vaccine using a computer-assisted self-interviewing instrument. This approach is likely to be acceptable to adult women and is reasonably accurate in a clinical research setting.

Background: In 2019, the CDC expanded their recommendations for human papillomavirus (HPV) vaccination beyond age 26 years to include shared clinical decision-making (SCDM) among adults aged 27–45 years ("mid-adults"). The purpose of this study was to describe HPV vaccination status among mid-adult women before the implementation of SCDM for HPV vaccination. Methods: A cross-sectional survey was conducted during 2016–2019 in Connecticut, United States, and enrolled women born in 1981 or later (birth cohorts eligible for HPV vaccination). This analysis was restricted to participants aged 27 years and older at the time of the survey. Correlates of vaccination status, sources of vaccine information, and reasons for not receiving the vaccine were examined. Results: Among 298 participants, 64.4% had not received HPV vaccine. Other than age (younger age was associated with being vaccinated), no other demographic or behavioral correlates were associated with vaccination. Compared with unvaccinated women, vaccinated women were more likely to have heard about the HPV vaccine from a doctor (odds ratio [OR] = 3.45, 95% confidence interval [CI]: 2.00–5.88) and less likely to have heard about it from television (OR = 0.23, 95% CI: 0.13–0.41). The main reasons for not being vaccinated were "vaccine not offered" (48%) and "too old" (40%). Conclusions: A majority of mid-adult women in this study were not previously vaccinated against HPV, signaling the large opportunity for SCDM with this population. This may be facilitated by ensuring health care providers and mid-adult women know about the availability and potential benefits of HPV vaccination to inform decision making. [ABSTRACT FROM AUTHOR]