One More Step Towards Justifying Targeted Treatment for Insomnia

In the early 1980s, as sleep medicine was coming into its own as a medical subspecialty, there was no rallying cry as popular as “Insomnia is a symptom, not a disorder.” After more than three decades, it is interesting to find that insomnia is once again considered a distinct nosological entity. The best example of the change in the prevailing zeitgeist is that the DSM-V will likely drop the distinction between Primary Insomnia and Secondary Insomnia in favor of the classification “Insomnia Disorder.” This, along with a preponderance of data regarding the efficacy and effectiveness of the existing treatments for insomnia,^1–4 clearly suggests that insomnia should be conceptualized—and treated—as an independent disorder. These developments, however, will not in and of themselves guarantee that the more than 30 million individuals with Insomnia Disorder will receive treatment.

One of the remaining barriers to making the regular treatment of insomnia standard practice is the need for medical economic data. That is, data that speak to the (1) costs of untreated insomnia to society, (2) costs of untreated insomnia to the individual, and (3) cost savings to society and the individual when the insomnia is treated. To date, there are a variety of studies that speak to the first issue (cost to society), only two studies conducted in the United States that speak to the second issue (cost to the individual), and only one study that speaks to the third issue (the cost savings of treatment).

With respect to societal costs, there are a number of studies that suggest that the direct and indirect costs of insomnia are approximately $100 billion per year.⁵ The bulk of this expense is related to lost work place productivity ($77-$92 billion). While such a number seems unfathomably high, it is possible to align the figure with a common-sense metric. For example, taking the data in 2006 as a point of departure, a 10% loss in productivity translates into a number that is roughly equivalent to those of the published studies. That is, if 10% of the population had insomnia in 2006 (30 million) and there was a 10% loss of productivity (calculated as 10% of the 2006 median income of $26,000), then the loss in productivity would be approximately $2,600 per individual and approximately $75 billion to society.

With respect to the cost to the individual, the first study conducted in the United States to address this issue was conducted by Ozminkowski et al. in 2007.⁶ This investigation found that the indirect and direct costs of insomnia were approximately $2,400 more per individual (per annum) than individuals without insomnia ($1,253 more for subjects 18-64, and $1,143 more for subjects ≥ 65, over 6 months). This estimate, while consistent with the above findings, may underestimate the economic impact of insomnia, owing to the failure to account for lost work productivity (presenteeism). This limitation served as a point of departure for the second study featured in this issue of SLEEP by Sarsour and colleagues.⁷ In this study, it was found that: (1) lost productivity and health care costs were incrementally higher moving from no insomnia, to sub-threshold insomnia, to moderate and severe insomnia; (2) participants with moderate and severe insomnia reported approximately 9% more lost productivity (including absenteeism, disability and presenteeism) as compared to the no insomnia group (no Insomnia = 14.8% and moderate and severe insomnia = 23.5%); and (3) the direct and indirect costs were increased as compared to the no-insomnia group. Specifically, the group with moderate and severe insomnia was found to have healthcare costs that were 75% larger (a difference of $566 per annum) and lost productivity costs that were 72% larger (a difference of $726 per annum) than the no insomnia group. The total cost difference being $1,292 per annum. Interestingly, this estimate is about half of the annualized amount reported by Ozminkowski et al.⁶ This finding is surprising given that the present study accounted for presenteeism as well as absenteeism and disability costs. While it is possible, as the authors suggest, that this discrepancy is related to differences in the populations evaluated, the reported 8.7% loss in productivity suggests that the given estimate should have been higher ($3,480 per annum), given the median income for their sample ($41,495 [approximately $20/hr]) and the estimated 174 hours of lost work productivity (8.7% of 2,000 hours). This concern notwithstanding, both studies serve to demonstrate that insomnia is associated with significant costs to the individual.

Finally, there is the issue of the cost savings when the insomnia is treated. In many ways, this type of analysis represents the strongest test of whether treatment is justified on medical economic grounds. That is, the two prior approaches, while informative, require that one assume that all the costs associated with insomnia are actually due to insomnia (as opposed to other unmeasured variables), and that one infer that treatment for insomnia necessarily results in the full recovery of all insomnia-related expenses. In contrast, the assessment of economic gains in the context of clinical efficacy trials allows for the direct assessment of treatment related savings. To date, only one such study has been undertaken. Snedecor et al.⁸ found in an analysis of data from a randomized controlled trial of eszopiclone in 824 adults with Primary Insomnia that treatment with eszopiclone resulted in a gain of 0.0137 quality-adjusted life years (QALYs) relative to placebo, with an additional cost savings of $67 per 6 months (a cost of about $5,000 per QALYs). This value is within the range that is considered to demonstrate cost-effectiveness.

In sum, the work by Sarsour and colleagues⁷ adds to the growing literature suggesting that there is sufficient evidence on medical economic grounds to justify the expense of treatments that are already justified on the basis of efficacy and safety. What remains is to expand the present findings to include medical economic analyses with other types of insomnia treatments.

DISCLOSURE STATEMENT

The authors have indicated no financial conflicts of interest.