Insomnia, the subjective sense of having inadequate quantity or quality of sleep, occurs in 20% to 35% of the general population, depending on the inclusion criteria and patient groups studied (Hatoum et al., 1998; Kuppermann et al., 1995; Weyerer and Dilling, 1991; National Sleep Foundation, 1995). Not surprisingly, many insomniacs describe a variety of difficulties in their personal and professional lives (National Sleep Foundation, 1995). As a group, they score lower on formal quality-of-life inventories (Hatoum et al., 1998b).
Recently we have had an opportunity to examine another aspect of the daytime lives of insomniacs: how they interact with the resources of managed care organizations (Hatoum et al., 1998a). We observed that the insomniacs rarely went to seek care for their sleep disturbance and infrequently received sedatives or hypnotics, but they had much higher rates of interactions with the caregiving organizations on a wide range of measures. Perhaps not surprisingly, we also found that roughly half of them had histories suggestive of depressive disorders.
Our subjects came from a multisite study of managed care programs in five large medical programs: Lovelace Health System (Albuquerque, N.M.), Lewis-Gale Clinic (Salem, Va.), Cleveland Clinic (Cleveland), Sutter/CNS (Sacramento, Calif.) and Geisinger Clinic (Danville, Pa.). Patients from each facility completed a two-page health status questionnaire including the SF-36 Health Survey (Ware et al., 1993) and a sleep habit questionnaire, as well as questions regarding medication use, sleep disturbance and utilization of health care during the previous eight weeks (Hatoum et al., 1998a; Hatoum et al, 1998b). In addition, patients were given three questions that served as a screen for depression (Table 1). In a multicenter validation trial against the Diagnostic Interview Schedule, a two-item depression screener from which this questionnaire was adapted had a sensitivity of 83% to 94% and a specificity greater than 90% for detecting major depression or dysthymia (Rost et al., 1993).
Data from the study came from 1,740 mail responses and 1,707 responses from patients in the waiting rooms of primary care clinics (some data are based on the latter group only and will be referred to as the outpatient data set).
The sleep section of the questionnaire contained three categories of questions dealing with sleep loss, daytime sleepiness and nighttime disturbances. Using responses to the five questions in the sleep loss category, three levels of insomnia were defined: no insomnia, difficulty initiating or maintaining sleep without daytime dysfunction (which we will refer to as level 1 insomnia) and insomnia accompanied by daytime dysfunction (level 2 insomnia). People with level 1 or 2 insomnia were prompted to answer 20 additional questions about potential causes or behaviors associated with sleep disturbance.
The demographic profile of the entire sample revealed that a majority had an income greater than $30,000 per year, were female, had education greater than high school, were married and were Caucasian (Hatoum et al., 1998a). The most common age group was 41 to 55 years (32.3%), while 8.2% were under age 25 and 9.1% were over 70. People who reported sleep problems were most likely to be non-Caucasian females less than 25 or more than 70 years old, taking prescription medication for one or more disorders (primarily circulatory system, brain and nervous system, and/or musculoskeletal system), and having a lower income and lower educational level than their counterparts.
In the group as a whole, nearly half (46%) reported sleep problems and approximately one-third (32.5%) reported level 2 insomnia; that is, they believed their sleep problems resulted in daytime dysfunction. We will focus on the level 2 group for most of the analysis. Many of their features were determined by a more detailed questionnaire (Table 2).
Several aspects of the level 2 insomniacs became apparent. They were much more likely to have recently used the health care system, only a small minority were receiving treatment for their sleep difficulties and many had symptoms suggestive of major depression.
Among the entire sample, insomniacs with daytime dysfunction were much more likely to have received some form of health care during the previous eight weeks. Specifically, they were more likely to have gone to the emergency room, had an office visit or a phone call to their physician, had a laboratory test, or taken an over-the-counter medication compared with subjects who did not complain of sleep disturbance (Hatoum et al., 1998b). This led us to examine whether they had more comorbid medical conditions, which we estimated from the nature of other prescription medications they were receiving. On this basis, we found that level 2 insomniacs were more likely to have circulatory or nervous system conditions or to have multiple medical conditions compared to subjects with no sleep complaint.
Only 0.9% of patients had gone to a physician specifically about their sleep difficulties. Of those insomniacs who did not complain of daytime dysfunction (level 1 insomnia), 5.5% were taking a prescription hypnotic, while 11.2% were taking an over-the-counter medication (Hatoum et al., 1998a). Comparable values for insomniacs with daytime dysfunction (level 2 insomniacs) were 11.6% and 21.4%, respectively.
Many patients with depression, of course, complain of poor sleep, and we wondered to what degree unrecognized depression underlies the sleep disturbance in our patients. In order to assess this we examined the three screening questions previously validated against the Diagnostic Interview Schedule (Rost et al., 1993). As might be expected, about half the number of people who screened positive for depression in the past year qualified as being level 2 insomniacs. Conversely, among the level 2 insomniacs, 48.6% had histories suggestive of what we operationally referred to as short-term depression, while 32.6% to 38.5% appeared at risk for long-term depression. Comparable rates for subjects with no sleep complaint were significantly lower (18.1%, 7.8% and 13.5%, respectively).
In summary, we have found that poor sleep accompanied by disturbance in daytime functioning is found in approximately one-third of primary care patients. They are more likely to have comorbid medical conditions and to utilize the health care system compared with patients who sleep well. This suggests that, in many cases, addressing an associated medical condition is an important aspect of dealing with the sleep disturbance.
Approximately one-third of patients with insomnia and daytime dysfunction appeared to be at risk for major depression, suggesting that an underlying affective disorder should be considered in patients who complain of insomnia; of course, if it is found, specific treatments for depression rather than sedative or hypnotic medication would be appropriate. Finally, less than 1% of patients went to see their doctor specifically for their sleep complaints, and only 12% of insomniacs with daytime dysfunction were receiving a prescription medication for sleep. This suggests the importance of being aware of the possibility of sleep disturbance in patients who may present with a variety of problems in a primary care setting. When sleep disturbance is found, it can often be managed by dealing with underlying discomfort from medical conditions, treating associated psychiatric disorders including depression and by considering the possibility that other medications the patient is receiving may be disturbing sleep. When these techniques do not suffice, referral to a sleep disorders center may be appropriate.
Hatoum HT, Kania CM, Kong SX et al. (1998a), Prevalence of insomnia: a survey of the enrollees at five managed care organizations. Am J Man Care 4(1):79-86.
Hatoum HT, Kong SX, Kania CM et al. (1998b), Insomnia, health-related quality of life and health care resources consumption-a study of managed care organization enrollees. Pharmacoeconomics 14:629-637.
Kuppermann M, Lubeck DP, Mazonson PD et al. (1995), Sleep problems and their correlates in a working population. J Gen Intern Med 10(1):25-32.
National Sleep Foundation (1995), Gallup Poll on Sleep in America. Princeton, NJ.
Rost K, Burnam MA, Smith GR (1993), Development of screeners for depressive disorders and substance disorder history. Med Care 31(3):189-200.
Ware JE, Snow KK, Kosinski M, Gandek B (1993), SF-36 Health Survey Manual and Interpretation Guide. Boston: New England Medical Center.
Weyerer S, Dilling H (1991), Prevalence and treatment of insomnia in the community: results from the Upper Bavarian Field Study. Sleep 14(5):392-398.