OBJECTIVES: Previous research has shown that untreated insomnia within the elderly has significant economic burden. The lifetime prevalence of major depressive disorder (MDD) is more than 17%, of which up to 80% have comorbid insomnia. The purpose of this study was to estimate the economic burden of untreated insomnia within an elderly comorbid depressed population. METHODS: A large managed care claims database was used to identify elderly MDD subjects between January 1, 1998 to November 30, 2007 using MDD-specific ICD-9-CM codes. Depressed insomniacs were selected from this cohort using insomnia-specific ICD-9-CM codes/hypnotic-use. Direct costs were compared with a 1:1 matched control group that had depression and no insomnia vs. depressed-insomniacs twelve-months prior to initiating insomnia treatment/diagnosis. Total direct costs included inpatient, outpatient, ER and drug costs. Multivariate regression analyses were conducted to control for factors that were different even after matching. Cost outcomes were log-transformed and then retransformed to their original scale after applying smearing factor. All statistical tests were conducted using SAS 9.1. RESULTS: A total of 2900 controls were matched to 2900 depressed-insomniacs on demographic characteristics, index-date, antidepressant use, plan type, and region using propensity-scoring technique. Depressed-insomniacs had higher (p<0.001) total outpatient visits 11.20 (SD=10.32) vs control 8.71 (SD = 7.32); depression-related visits 4.05 (SD=8.41) vs control 1.70 (SD = 3.27); antidepressant-use 3.39 (SD=5.18) vs control 2.87 (SD = 4.51). Overall total direct costs for depressed-insomniacs were significantly (p<0.001) higher than controls ($9835 vs. $8384) after matching and controlling for differences in age, location, plan-type, index year, and comorbidity patterns. The $1451 difference in average medical expenditure per-subject reflected the estimate of direct cost of untreated insomnia in an elderly depressed population. CONCLUSIONS: Untreated insomnia within an elderly comorbid depressed population is associated with significant direct costs. Future research needs to determine prevalence and cost-effectiveness of treating comorbid insomnia within this population.