Variation in performance status assignment following the interoperability and patient access rule.

608 Background: Karnofsky Performance Status (KPS) is a scale utilized to classify a patient’s functional status. It ranges from 0-100 and is scored in ten-point increments, where 100 denotes no evidence of disease, while a lower score denotes reduced ability to care for themselves and perform daily activities. First introduced in 1949, KPS has since evolved into a crucial clinical tool in oncology, informing treatment planning, estimating survival for certain subtypes, and shaping eligibility for clinical trials. Despite its utility, KPS can be subject to significant variation, potentially stemming from subjectivity or inconsistent scoring practices. We hypothesize that both inter- and intra- provider variation in KPS assignment exist across four-week encounter periods for the same patient. We further hypothesize that this variation is observed on an annual scale, especially after the 2020 Interoperability and Patient Access rule (CMS-9115-F), which improved patient access to electronic health record notes, inadvertently supporting homogeneity and alignment with previous KPS assignment. Methods: We extracted encounter data from all patients at a single institution from 2013–2024 and constructed four-week intervals (±14 days around each reference visit), retaining only intervals with ≥2 encounters. For each interval, we recorded if multiple providers were involved, the mean and standard deviation of KPS, and simple difference in KPS. Intervals were further stratified into pre-policy (2013–2021) (n = 943) and post-policy (2022–2024) (n = 8,096) groups. We compared mean simple differences using Welch’s t-test. To address group sample-size imbalance, we performed 1,000 iterations of random subsampling from the post-policy group and repeated Welch’s t-test on each draw. Results: Our extraction and cleaning resulted in 30,683 unique patients and 73,002 total encounters. Among 9,039 total intervals, 3,384 (37.4%) involved multiple providers. Overall mean KPS was 84.93 with a standard deviation of 12.36. The mean simple difference across the entire sample of intervals was -0.87, and the difference between the post and pre-policy groups was -0.81 (95% CI -1.27 to -0.35, p = 0.001). Further subsampling analysis of these groups revealed a median p value of 0.009 while 86.1% of iterations remained significant (p < 0.05). Conclusions: Within the context of an exploratory study, we observed a modest yet statistically significant difference in KPS assignment following the July 2021 policy implementation deadline. However, given that KPS is scored in 10-point increments, the magnitude of this difference may not reflect clinically meaningful assignment changes. Future analyses will focus on provider-level scoring variability and examining shifts around clinically meaningful thresholds (eg. KPS 60) to better understand drivers of KPS assignment.