Abstract Introduction Administrative claims data are increasingly leveraged in sleep research. Accurate cohort identification and selection of an appropriate “index date” are essential for valid assessment of downstream health outcomes. Idiopathic hypersomnolence (IH) is a chronic sleep disorder associated with several long-term comorbidities. To estimate the impact of newly diagnosed IH on important comorbidities including cardiovascular diseases, it is essential to establish an accurate index date. We hypothesize that defining the index date using additional sleep testing rather than relying solely on diagnostic claims, will more accurately identify newly diagnosed patients, as demonstrated by their differences in baseline characteristics. Methods We created a cohort using Marketscan®, a large national claims database, from 2005-2022. Patients were enrolled at the index date when they were diagnosed with IH. To differentiate newly diagnosed cohort from prevalent diagnosed cohort, we additionally required a confirmative test (including multiple sleep latency test, polysomnography, or a home sleep apnea test) within 60 days prior to the index date. Then, we excluded patients who were diagnosed with narcolepsy, experienced major adverse cardiovascular events, subclinical cardiovascular disease within 1-year prior to index date or did not have 1-year continuous enrollment before the index date for both cohorts. We reported the baseline characteristics, and conducted chi-square tests of these two cohorts. Results We identified 5,472 and 10,866 patients who were newly and prevalent diagnosed with IH, respectively. Those newly diagnosed were statistically younger (≤25 years old: 26.1% vs 24.3%; 26-45 years old: 49.3% vs. 49.0%; 46-65 years old: 24.6% vs 26.7%), statistically more likely to experience anxiety (19.9% vs. 18.5%), depression (24.0% vs 22.2%), mood disorder (21.4% vs. 20.0%), obesity (9.8% vs. 8.3%), periodic limb movement disorder (3.0% vs. 2.4%), and sleep apnea (34.0% vs. 27.0%). They were statistically less likely to receive IH medication including oxybate (0.3% vs. 0.7%), stimulant (9.5% vs. 14.4%), and wake-promoting agent (10.7% vs. 21.5%). Conclusion A recent sleep test serves as a useful indicator for identifying patients with newly diagnosed IH patients and provides a more accurate index date for outcomes research. Support (if any) Sleep Research Society Foundation (Grant #: 25-FRA-001).
Jiao et al. (Fri,) studied this question.