Benefits of a novel in-line decision support and ordering platform to support genomic test ordering.

Abstract

431 Background: For many cancer patients, identifying the best treatment options requires timely genomic and molecular testing. However, only a small fraction of eligible patients receive timely access to evidence-based testing. Testing barriers include the clinician efforts to stay abreast of the vast number of genomic tests, rapidly evolving recommendations, and the variety of vendors offering these costly tests. We piloted a genomic test ordering solution that is integrated directly with the EMR and evaluated its benefits. Methods: We worked with experts from multiple specialties to finalize simple, context-based decision support and used standard EMR functionality to enable providers to order appropriate tests directly from the decision support webpage, seamlessly queuing orders in the EMR. We evaluated a set of metrics based on expected benefits of this new solution. We focused our analysis on lung cancer patients, and on five genomic tests: blood-based NGS, tissue-based NGS, EGFR PCR, PDL1 28-8, PDL1 22C3. Regional oncologists and nurse navigators were surveyed before and after the pilot period to ascertain ordering length of time and experience. Aggregate claims data were reviewed to assess adherence rates to recommended testing guidance and the ordering tool. The pilot launched July 2021, and six-month pre- and post-pilot data were used to evaluate its impact. Results: We received 46 and 53 pre- and post-pilot survey responses, respectively. There was an increase in the proportion of respondents who reported that ordering took ‘Less than 5 minutes’ (39% (pre) to 72% (post)), and that the ordering process was ‘Easy’ (28% (pre) to 74% (post)). In pre-pilot survey comments, ‘knowing the right test’ and ‘how to order’ comprised 30% of 40 comments. These issues were not raised in the 37 post-pilot survey comments we received. Approximately 35% of post-pilot comments revealed no issues with the new process, though 24% of comments expressed issues finding test guidance and the order button. Adherence to recommended guidance and the ordering tool grew from 0% to 44% from July to November 2021 but maintained a steady pace since then to April 2022. Conclusions: Connecting genomic test decision support to ordering capability can lead to significant benefits for providers and patients. The pilot was successful in saving ordering time, improving the ordering experience, and improving adherence to guideline concordant treatments. Success factors include elimination of a cumbersome ordering interface, development of individually orderable tests, implementation of a central ordering location, and guidance to preferred vendors. There are opportunities to improve clinician onboarding, boost adherence rates, and assess other impacts to outcomes such as cost savings and time to treatment. Additional analyses are required to understand the broader impact to these measures.