Abstract Title Effect of Metformin on N3 and REM Sleep of Medicare Patients Compared to Non-Medicare Patients

Background During N3, or slow-wave sleep, the body repairs itself, while REM sleep is associated with memory consolidation and brain health. Metformin, a first line treatment for type 2 diabetes mellitus (T2DM), has been reported to cause insomnia and sleep disturbances. In the US, Medicare and Medicaid are forms of public health insurance available to patients who are elderly, disabled, or financially limited. Our study examines the impact of Metformin on sleep architecture in patients with public and private health insurance.

Methods Overnight polysomnography was performed on 27 diabetic patients taking Metformin. Patients underwent six channel EEG, Electrooculogram, Electromyogram EKG, pulse oximetry, respiratory and flow monitoring. Sleep staging was performed using AASM guidelines, verified by a physician. Age, gender, BMI, insurance, and medication data were collected. N1, N2, N3 and REM sleep stage data were analyzed, and total sleep time (TST) was calculated as the sum duration of N1, N2, N3, and REM sleep. 22% of patients on Metformin had Medicare (Medicare), 22% had Medicaid (Medicaid), and 56% had private insurance (PI).

Results Among Medicare, the mean duration of N3 was 6.28% of TST (+ 0.362% SD) and the mean duration of REM sleep was 1.57% of TST (+ 0.134% SD). Among PI and Medicaid, the N3 sleep duration was 17.5% of TST (+ 1.35% SD), and REM sleep duration was 8.84% of TST (+ 0.417% SD). Both N3 (p<0.00201) and REM (p<0.00081) sleep duration were significantly longer in non-Medicare patients. Among Medicare and Medicaid, the duration of N3 was 10.6% of TST (+ 0.698% SD), and the duration of REM sleep was 6.42% of TST (+ 0.528% SD). Among PI, N3 sleep duration was 18.4% of TST (+ 1.64% SD), and REM sleep duration was 7.87% of TST (+ 0.394% SD). N3 (p<0.0343) sleep duration was significantly longer in PI.

Conclusion Medicare diabetic patients taking Metformin exhibit significant declines in both REM and N3 duration when compared to non-Medicare diabetic patients on Metformin. When comparing sleep quality of Medicare and Medicaid patients taking Metformin with that of patients on private insurance on Metformin, the significant decrease in N3 duration persists. Our study is limited by sample size as well as patient demographics in Southern California. Nevertheless, our results suggest a link between Metformin usage, sleep quality, and healthcare accessibility, prompting future research.