Despite its potential benefits, access to continuous glucose monitors (CGMs) is often restricted based on blood glucose monitoring frequency. We aim to study if CGM efficacy depends on blood test strip (TS) usage prior to CGM initiation. IBM MarketScan™ Commercial Claims and Medicare Supplemental databases were used in this retrospective, observational analysis. MarketScan contains insurance billing claims for inpatient, outpatient and pharmacy expenses. Cohort requirements included: de novo CGM purchase in 2017 Q1-2018 Q2, type 1 or type 2 diabetes, ≥18 years old, fast- or short-acting insulin, ≥ 6-months pre-CGM observation time. Blood TS usage was determined based on purchases 6 months pre-CGM. Primary outcome was acute diabetes events (ADE): hospitalization with hyper-/hypoglycemia as the primary diagnosis or outpatient emergency associated with a code of hyper-/hypoglycemia. Andersen-Gill Cox regression compared event rates 6-months pre-/post-CGM. The cohort (n=12,521, age 43.8±15.2 years, 50.8% male, 77% type 1) experienced a reduction in ADE from 0.245 to 0.132 events/patient-year (HR: 0.54 [0.49 0.59]; P:<0.001). A similar reduction was seen in patients using <4 and ≥4 TS per day. A reduction was seen in all TS usage groups. Initiation of CGM use was associated with reduced ADE regardless of baseline blood TS usage.