Performance characteristics of CMS' OP-35 quality measure: Sensitive but not specific.

421 Background: OP-35 is a measure that the Center for Medicare and Medicaid Services (CMS) uses to evaluate quality, intended to measure rates of potentially preventable complications (PPCs) of chemotherapy. The tool assesses the rate of emergency department visits and admissions (EDV/As) for patients (pts) receiving outpatient intravenous (IV) systemic anti-cancer therapy (SACT) and focuses on the presence of ≥1 of 10 qualifying diagnoses (QD): anemia, dehydration, diarrhea, emesis, fever, nausea, neutropenia, pain, pneumonia, or sepsis. To qualify for OP-35, the metric uses a strict definition of pts with a primary diagnosis (dx) including 1 of the QDs or cancer as the primary dx and 1 of the QDs as a secondary dx. However, the ability of this strict definition to define PPCs is unknown. Methods: We conducted a retrospective review of pts who received outpatient IV SACT (the denominator for OP-35) at the University of Texas MD Anderson Cancer Center in 2023. All pts who had an EDV/A were assessed to capture the primary and secondary diagnoses associated with their encounters. Pts were divided into 3 groups: (1) Strict OP-35 if they had 1 of the 10 QDs as a primary dx or cancer as primary and a QD as a secondary dx; (2) Other OP-35 if they did not meet the strict definition but had ≥1 QD (e.g. QD as a secondary dx but not cancer as the primary dx); or (3) Non-OP-35 if they did not have any of the 10 QDs. Results: The total number of pts included in the population who received outpatient IV SACT was 10,353. Of these, 3,245 (31.3%) had an EDV/A within 30 days of receiving outpatient IV SACT. Of pts with an EDV/A, 87% of pts had a QD. Strict OP-35 represented 1,562 (48%), Other OP-35 was 1,249 (38%); Non-OP-35 was 434 (13%). The median length of stay (LOS) for all admitted pts (N=1,839) was 5 days. Of all OP-35 pts discharged from the ED (N=787), 47% of Strict OP-35 stayed ≤5 hours vs 49% of Other OP-35 (P=0.62). Of all pts discharged from the ED, 76% of Non-OP-35 stayed ≤5 hours vs 48% of all OP-35 (OR 9.2, P<0.0001). Of pts admitted, 33% of Strict OP-35 stayed ≤3 days vs 30% of Other OP-35 (P=0.12). Analyzing all OP-35 pts, 52% of pts had an EDV/A between 0-14 days vs 48% 15-30 days. Admitted pts were more likely to have pneumonia (27%) and sepsis (18%) than those discharged from the ED (3% and 0.2%, respectively), OR 12.3 and 96.3 (P<0.0001 for both). Conclusions: The goal of OP-35 is to measure PPCs. However, Strict OP-35 pts had no significant difference in time to ED discharge or admission LOS compared to Other OP-35 pts. The strict OP-35 definition does not appear to help differentiate between all pts and ones with PPCs. Non-OP-35 pts had the shortest ED visits. Nearly half the pts had an EDV/A >2 weeks from their SACT, which are unlikely to represent PPCs. Pts with sepsis or pneumonia were much more likely to be admitted rather than discharged from the ED and likely do not reflect PPCs. OP-35 appears to be sensitive to all-cause EDV/A but not specific to PPCs.