METHODS

To estimate the number of sleep laboratories and overnight sleep studies performed per year in the United States, we first mailed a brief questionnaire to all American Academy of Sleep Medicine (AASM) accredited sleep centers in the US, excluding Puerto Rico, and all individual members of the AASM. This questionnaire was mailed in March 2001 with all utilized responses received by June 1, 2001. We then selected three states in which to undertake a more precise assessment of the number of laboratories/sleep studies performed (Massachusetts, Oregon, and Louisiana). We selected these states because they are approximately equal in population, but represent a diversity of socioeconomic status and geography. To more precisely determine the number of sleep studies performed, we identified phone numbers and addresses of all known AASM members and sleep laboratories in these states (from AASM listings and telephone directories). We then contacted every AASM member (or accredited center) who did not respond to the original survey by telephone or by email to obtain information concerning the number of polysomnograms performed per week. We also asked about other laboratories in their area and contacted these physicians/clinicians as well. Once we were confident that we knew precisely the number of sleep laboratories and polysomnograms conducted in these three states, we could then determine the percentage of laboratories/studies identified by the original survey:[actual number of sleep studies (or labs)/number from the original mail survey]. This ratio was then used to extrapolate the actual number of laboratories/studies conducted in the other 47 states. The number of sleep studies per capita was calculated by dividing the number of sleep studies performed per year by the country’s or state’s population (from the US Census year 1999).