[Congressional Record Volume 140, Number 143 (Wednesday, October 5, 1994)]
[Senate]
[Page S]
From the Congressional Record Online through the Government Printing Office [www.gpo.gov]


[Congressional Record: October 5, 1994]
From the Congressional Record Online via GPO Access [wais.access.gpo.gov]

 
                       THE ECONOMICS OF INSOMNIA

  Mr. HATFIELD. Mr. President, I rise today to update my colleagues on 
an issue that should be a wake up call to every person in America. In 
an article to be published this month by Melissa Stoller of the 
University of Chicago, in the medical journal Clinical Therapeutics, it 
is estimated that the annual economic cost of insomnia due to reduced 
productivity, accidents, and medical problems is between $92.5 and 
$107.5 billion. This figure does not even begin to include the toll 
that insomnia takes in terms of human suffering, decreased quality of 
life or deteriorated personal relationships.
  This is not a problem that affects only the few. Research 
consistently shows that in any given year 30 to 40 percent of the U.S. 
population suffers from insomnia. The majority of these cases are not 
associated with a psychiatric or medical problem.
  The average person is absent from work 1 day per month. However, the 
average workers suffering from insomnia misses 2.8 additional days per 
month. The estimated cost of absenteeism to a single organization is 
more than $4,800 per year. The most obvious result of nighttime 
insomnia is daytime sleepiness. This daytime impairment is associated 
with more sleep during work breaks, markedly reduced productivity, and 
dissatisfaction with one's job. The estimated annual cost of 
performance impairment due to insomnia is $41.1 billion.
  In addition, both work-related and motor-vehicle accidents are more 
likely committed by someone suffering from insomnia than by a well 
rested individual. Insomniacs have about 1.5 times as many work-related 
accidents as the rest of the population and have auto accidents 2 to 3 
times more often. It is estimated that sleep-related accidents cost $43 
to $56 billion annually; $29 to $38 billion for motor vehicle 
accidents; $10 to $14 billion for work-related accidents; and $2 to $3 
billion for at-home accidents.

  It has also been found that insomnia is directly linked with heart 
disease, high blood pressure, diabetes, stroke, and depression. Persons 
with poor sleep see the doctor more often and have more health problems 
than those who sleep well. Insomnia has also been related to higher 
levels of depressive illness and alcoholism. People who sleep less than 
6 hours per day have a 30 percent higher death rate than those who 
sleep 7-8 hours.
  In 1993 the National Commission on Sleep Disorders Research reported 
to Congress on the need to establish a National Center for Sleep 
Disorders Research. I introduced legislation and in the 1993 NIH 
reauthorization the National Center was established and housed within 
the National Heart, Lung and Blood Institute. In the few months that it 
has existed, the National Center has done a tremendous job in starting 
a national public awareness campaign on sleep disorders while 
coordinating their activities with other Federal agencies. However, 
there is still a long way to go and the National Center needs the 
participation and cooperation of all branches of the government in 
order to make the public and health professionals aware of the 
seriousness of insomnia and other sleep disorders.
  As you can see, insomnia is a problem that has an astronomical 
economic impact on our society. With such grave consequences, insomnia 
can no longer be thought of as simply an irritating but inevitable part 
of modern life. It must be viewed as a potentially life-threatening 
condition that can and should be treated.
  I ask unanimous consent that the article be included in the Record.
  There being no objection, the material was ordered to be printed in 
the Record, as follows:

                   [From Clinical Therapeutics, 1994]

                      Economic Effects of Insomnia

                    (By Melissa Kaleta Stoller, MA)


                                abstract

       Insomnia affects up to 40% of the general population yearly 
     and is a significant cause of morbidity and mortality. The 
     direct and indirect costs of insomnia place a tremendous 
     economic burden on society and employers. In addition to the 
     cost of the medical treatment and drugs, measurable costs of 
     insomnia include reduced productivity, in creased 
     absenteeism, accidents, and hospitalization, as well as 
     medical costs due to increased morbidity and mortality, 
     depression due to insomnia, and increased alcohol 
     consumption. This article reviews the literature on the 
     economic costs and effects associated with insomnia. Based on 
     the data reviewed, a conservative estimate of the total cost 
     of insomnia was calculated at $92.5 to $107.5 billion. Early 
     recognition and treatment of insomnia can reduce the costs 
     associated with the condition, as well as possibly prevent 
     other illnesses.


                              introduction

       Insomnia, or difficulty falling or staying asleep, is one 
     of the most pervasive problems affecting human health. 
     Research and surveys consistently have reported that 30% to 
     40% of the general population suffer from insomnia in a given 
     year.1-7 Similar insomnia rates, averaging 32.4% are 
     reported for Europe and Australia,5,6 suggesting that 
     insomnia is a global problem. About one half of the cases are 
     classified as moderate to severe. The majority of persons 
     suffering from insomnia have primary insomnia;\8\ they do not 
     have a psychiatric or medical problem that accounts for their 
     insomnia. Estimates for the current prevalence of insomnia 
     (i.e., the portion of the population that is actually 
     suffering at any one time) range from 13.4% to 48%.4,9 
     largest US study reports a current prevalence of 32% to 
     33%,\10\ which is the estimate used throughout this article.
---------------------------------------------------------------------------
     Footnotes at end of article.
---------------------------------------------------------------------------
       Insomnia carries an incalculable cost in terms of human 
     suffering and deterioration in personal and professional 
     relationships. Sufferers report reduced satisfaction with 
     life; loss of opportunity; deteriorated relationships with 
     children, spouses, and coworkers; reduced ability to cope; 
     and reduced enjoyment of life.3,7,11,14 Family and 
     friends also may be affected and their lives disrupted.\7\ 
     These costs are not considered here. What will be assessed 
     are the quantifiable economic costs of insomnia to society 
     and employers.
       The objectives of this article are to review the literature 
     on the economic costs and effects of insomnia, provide a 
     current summary of recent research addressing these issues, 
     compile data reflecting the economic costs of insomnia, and 
     calculate and summarize them as precisely as possible.
       Both direct (medical treatment and drugs that address the 
     complaint of insomnia) and indirect costs are involved. The 
     indirect quantifiable costs of insomnia are summarized in 
     Table I. Fortunately, good data exist to establish a range of 
     values for many of these costs. Some factors, such as 
     increased mortality associated with insomnia, are discussed, 
     but no dollar figures are presented. An attempt was made to 
     present the most recent cost estimates and relevant data for 
     all categories. With one exception noted in the text, dollar 
     figures were not adjusted for inflation. Although the 
     analysis is not precise, a reasonable estimate of the overall 
     cost of insomnia in the United States can be calculated.

                  Table I.--Indirect Costs of Insomnia

       Measurable work loss due to reduced productivity and 
     increased absenteeism.
       Accident costs (death, disability, property damage, medical 
     expenses) resulting from significantly higher accident rate.
       Hospitalization and medical costs related to increased 
     morbidity and disproportionate utilization of primary care 
     resources.
       Depression related to chronic insomnia.
       Increased mortality associated with habitually short sleep.
       Self-treatment with alcohol.


                          loss of productivity

       Many persons with insomnia know intuitively what can be 
     quantitatively demonstrated--the most devastating cost of 
     insomnia both to individuals and society may be the reduction 
     in productivity. Work performance is compromised in two ways: 
     increased absenteeism (loss of time from work) and reduced 
     effectiveness (loss of productive ability).15,18

                              Absenteeism

       Insomnia was a powerful predictor of absenteeism (Figure 
     1)* in a large, cross-sectional national study that assessed 
     37 employee and job attributes. It was a more powerful 
     predictor of absenteeism than even age or job satisfaction. 
     The only employee attribute that correlated more strongly 
     with absenteeism was being a mother of small children.
       The negative consequences of increased absenteeism run 
     throughout an organization. For the worker who is frequently 
     absent, there is loss of pay and stature, while coworkers 
     suffer from increased work load and increased work 
     coordination problems, as well as decreased productivity.\19\ 
     There also may be an increased accident rate, both for 
     coworkers and the worker replacing the absentee who are 
     forced to perform additional or unfamiliar work.19,21
       It is estimated that more than 400 million workdays are 
     lost to absenteeism each year. A 1977 study estimated that 
     the cost of absenteeism among nonmanagerial personnel was 
     about $66 per day.\22\ This figure included replacement or 
     overtime, fringe benefits, overhead, productivity loss, and 
     accident and grievance costs. Adjusted by the increase in the 
     employer's costs for employee compensation, which have 
     roughly doubled from $7.43 to $16.14 since 1977,23,24 
     the cost per absent day can now be estimated at $143.22 for 
     nonmanagerial workers, or more than $57 billion per year. 
     Whereas an average worker is absent about 1 day per month, a 
     worker suffering from insomnia is absent approximately 2.8 
     additional days per month,\15\ costing an organization more 
     than $4800 per year.

                         Performance Impairment

       Nighttime insomnia goes hand in hand with impaired daytime 
     functioning. Laboratory studies of task performance by 
     individuals with insomnia have concluded that this group 
     demonstrates impaired daytime performance.\25\ A study of 691 
     persons with untreated insomnia showed that they recognize 
     their impaired daytime functioning: 83% reported being 
     ``easily upset, irritated, or annoyed,'' 78% reported being 
     ``too tired to do things,'' 59% reported having ``more 
     trouble remembering,'' and 43% reported being ``confused in 
     their thinking.''\14\
       Lavie's\1\ large detailed study of the lifestyle, health, 
     sleep, and work habits of 1502 employees concluded that sleep 
     habits directly affect the workplace. Daytime fatigue, a 
     common result of insomnia, was associated with significantly 
     more sleeping during work breaks (14.2% vs 3.5%, P < 0.001), 
     significantly higher frequency of stopping work to take short 
     naps (16.8% vs 1.4%, P < 0.0001), and significantly less 
     satisfaction with one's work (P < 0.03). The association 
     between insomnia and reduced efficiency was supported by the 
     results of a 1992 survey, which recorded two to three times 
     as many days of poor productivity and concentration in 
     individuals with insomnia as in good sleepers.\26\ A third 
     study that matched insomniacs with good sleepers found that 
     good sleepers spent twice as much time working, studying, or 
     communicating compared with insomniacs (Figure 2).\12\* Poor 
     sleepers were twice as likely to be relaxing during the day.
       To complete an economic analysis of insomnia the following 
     questions must be asked: What is the effect of these 
     complaints and habits on overall productivity? What are the 
     economic consequences of insomnia over the course of a career 
     and within an organization? This analysis will limit itself 
     to workplace productivity, although insomnia is 
     unquestionably related to reduced household productivity as 
     well.
       The standard technique used in cost of illness studies is 
     to equate wages with productivity.27,28 In an efficient 
     market, persons will be paid a wage equal to the value of 
     their output.\27\ Diminished efficiency, therefore, should be 
     reflected by lower earnings in subjects with insomnia 
     compared with good sleepers when other variables are held 
     constant. An extensive longitudinal study of 2929 subjects 
     documented the career consequences and overall performance 
     decrement associated with insomnia.\18\ Tracking Navy 
     servicemen who entered the service at the same level, the 
     study found that impaired sleepers received significantly 
     fewer promotions (Figure 3),* remained in lower pay grades, 
     received fewer positive recommendations, and had higher 
     attrition rates compared with good sleepers. This study 
     concluded: ``In all measures used as indices of Navy 
     performance, poor sleepers performed significantly less 
     effectively.''\18\
       Estimates of reduced workplace productivity due to insomnia 
     also can be derived from studies of school performance by 
     individuals with and without insomnia. Schoolwork generally 
     is graded numerically, allowing for quantitative comparisons 
     of retention and output among different quality of sleep 
     categories. Results from a 1990 study showed that insomnia 
     was the most powerful predictor of school failure, more 
     significant than parental education and profession; the rate 
     of failure among insomniacs was twice that of 
     noninsomniacs.\17\ Similarly, a long-term study of medical 
     school students demonstrated that quality of academic 
     performance varied directly with perceived quality of 
     sleep.\29\

          Dollar Cost of Reduced Productivity Due to Insomnia

       A calculation of the actual dollar cost of insomnia related 
     to reduced productivity follows. The calculation is based on 
     data from the Johnson and Spinweber\18\ study of Navy 
     servicemen.

                                 Method

       Earnings are used to represent productivity, as summarized 
     by the Department of Health and Human Services: ``To estimate 
     the value of losses due to reduced productivity, the method 
     used is to take the difference in earnings or income between 
     [affected] and [nonaffected] groups. . . . Attempts are made 
     to account for other factors such as age, education, family 
     structure, that undoubtedly influence earnings.''\27\
       Johnson and Spinweber\18\ present wage data from two 
     populations of insomniacs and noninsomniacs entering the work 
     force at the same level; the data were controlled for sex, 
     age, and education (Table II). Using both population samples 
     and wage data from the 1994 Navy Times pay chart,\30\ the 
     performance decrement associated with insomnia can be 
     estimated at 4%. This can be compared with an estimated 6.63% 
     inefficiency rate for alcoholic individuals (calculated with 
     labor force data only).\27\

    TABLE II.--HIGHEST PAY GRADE ATTAINED IN TWO POPULATIONS OF NAVY    
            SERVICEMEN CLASSIFIED AS POOR OR GOOD SLEEPERS.*            
------------------------------------------------------------------------
                                   Percent at Pay Grade in 1981*        
                          ----------------------------------------------
                             E1     E2     E3     E4     E5    E6    E7 
------------------------------------------------------------------------
  Population 1 (n=1043)                                                 
                                                                        
Poor sleepers............   9.17  11.01  13.76  40.37  22.02  2.75  0.92
Good sleepers............   2.77   6.13  11.86  45.85  32.61  0.79  0   
                                                                        
  Population 2 (n=1186)                                                 
                                                                        
Poor sleepers............  16.49  15.96  38.83  27.66   1.06  0     0   
Good sleepers............   8.50  17.19  31.74  40.14   2.44  0     0   
------------------------------------------------------------------------
*Initial data collection periods: 1976 and 1977 for population 1; 1978  
  and 1979 for population 2.                                            

       Employing the cost-analysis approach, the cost to society 
     of this reduced productivity is calculated from the following 
     equation:

                  LOSS $=(POPPREV)(bY)

     where POP = number of labor force participants, or 127 
     million\23\; PREV = prevalence rate, or 33%; b = percentage 
     income loss for afflicted individuals, or $5; and Y = average 
     income for those without the disorder, or $24,575, the mean 
     annual pay for 1991.\23\
       The following assumptions will be made: (1) the entire loss 
     of income associated with insomnia is due to insomnia and not 
     to another uncontrolled variable; and (2) the performance 
     decrement associated with insomnia does not differ by sex, 
     although performance impairment data are available only for 
     men.
       In addition, several factors will result in an 
     underestimation of the cost of diminished productivity in 
     persons with insomnia: 1. Because the sample controls for 
     educational and occupational variables (students entering a 
     training program at the same level), true income loss will be 
     underestimated. Insomnia affects not only current income but 
     also occupational and educational attainment; that is, this 
     calculation will not measure the cost of the insomniac's 
     inability to enter this sample population due to past 
     performance impairment.
       2. Average earnings are not adjusted for insomnia. A 
     precise estimate would consider mean earnings for the 
     nonafflicted population, not for the general population, not 
     for the general population. These data generally are not 
     available, and cost-of-illness studies commonly substitute 
     general population data.\27\ However, this may result in a 
     significant underestimation of the lost production costs 
     attributable to insomnia as it is so prevalent.
       3. The performance impairment data are based on military 
     data. The military probably is less of a true market than the 
     civilian workplace. Income in the military may correlate more 
     strongly with tenure than with achievement.
       4. We will assume that there is no increase in unemployment 
     due to insomina; that is, the loss of efficiency associated 
     with insomnia does not result in job loss. However, some 
     persons with severe insomnia presumably may become to 
     impaired that they are unable to work.
       5. Additional costs related to the increased absenteeism of 
     the insomniac may be partially reflected in the wage data and 
     in the costs associated with increased morbidity. Therefore, 
     to avoid double counting, costs associated with increased 
     absenteeism are not included in the final productivity 
     calculation

                                Results

       A conservative estimate of the cost of performance 
     impairment due to insomnia is $41.1 billion per year, based 
     on a 4% reduction in productivity among the 42 million 
     working Americans suffering from insomnia. The estimate is 
     conservative because it considers only loss of work 
     productivity among those earning an income. It does not 
     include measures of increased unemployment among persons 
     suffering from insomnia, lost opportunity costs due to 
     insomnia-related academic failure, or lost household 
     productivity.

                      Special Productivity Issues

       Of particular concern to employers is the possibility that 
     insomnia is not only associated with impaired daytime 
     functioning generally but also with failure to respond 
     appropriately to challenge or emergency situations.\31\ In 
     other words, the individual with insomnia shows marked 
     deterioration in performance under the high stress or 
     ``deadline'' conditions commonly associated with certain 
     lines of work. This type of performance breakdown may explain 
     the finding that highly intelligent poor sleepers who 
     performed successfully in college were significantly less 
     successful than good sleepers once they entered the more 
     stressful and competitive atmosphere of medical school.\29\ 
     Similarly, Spinweber and Bellune\32\ concluded from their 
     analysis of the performance of participants in an extremely 
     stressful training program--``hell week'' of the elite 
     special force Navy SEAL teams--that those who developed 
     insomnia would not succeed. Despite its being significant, 
     particularly in situations were training is intense and 
     expensive, the economic impact of this type of insomnia-
     related critical performance failure cannot be assessed.


                   cost of insomnia-related accidents

                    Accidents Related to Sleepiness

       An increasing amount of attention is being given to the 
     role of fatigue as a cause of accidents. It has long been 
     recognized that, in industries and occupations in which 
     constant vigilance is required, accidents do not occur at 
     random times--they peak during the hours the workers are most 
     likely to be sleepy.\13\ As early as 1970, the US Bureau of 
     Motor Carrier Safety concluded that 30% of truck accidents 
     involved a sleeping driver.\33\ Other groups have found that 
     fatigue was the primary cause of 41% of truck accidents and a 
     secondary cause in an additional 18%.\24\ Leger\35\ 
     calculated that 41% to 54% of motor vehicle accidents were 
     fatigue related. A similar figure for commercial vehicle 
     accidents was presented by the Arizona Department of Public 
     Safety, which concluded that 42% to 49% were due to driver 
     sleepiness or inattention.\35\ The National Transportation 
     Safety Board (NTSB)\36\ found fatigue was the cause of 57% of 
     fatal truck accidents, based on reconstruction of the 
     accident site and driver sleeping history.
       The somewhat higher rate of fatigue-related accidents found 
     by the NTSB can be explained by the fact that it considered 
     only fatal commercial vehicle accidents, whereas the Arizona 
     Department of Public Safety figure considered all commercial 
     vehicle accidents. Fatigue-related motor vehicle accidents 
     tend to result in disproportionately more severe injury and 
     property damage, with drivers often falling asleep on 
     highways and hitting barriers or crossing the 
     midline.31,37 In one study, 27% of drivers who lost 
     consciousness while driving had fallen asleep; that 27% 
     accounted for 83% of the fatalities.\37\
       A recent study calculated the cost of sleep-related 
     accidents for the year 1988.\35\ The Human Capital Approach 
     method, which uses wages to represent output, was used. Costs 
     for different accident categories were divided into direct 
     and indirect costs. Direct costs included medical and 
     treatment expenses; indirect costs included loss of 
     productivity due to missed work or premature fatality. The 
     study concluded that the total cost of sleep-related 
     accidents was $43 to $56 billion, including estimates of $29 
     to $38 billion for sleep-related motor vehicle accidents, $10 
     to $14 billion for fatigue-caused work-related accidents, $2 
     to $3 billion for home-based fatigue-caused accidents, and $1 
     to $2 billion for public accidents caused by fatigue.

              Increased Rate of Accidents Among Insomniacs

       Because sleepiness is implicated as the cause of up to 50% 
     of certain types of accidents,\35\ it is not surprising that 
     persons with insomnia have a higher overall accident rate 
     compared with the general population.
       Comparing general accident rates in more than 5000 adults, 
     Balter and Uhlenhuth\38\ calculated an accident rate for 
     chronic insomniacs that was 3.5 to 4.5 times that of the 
     control group (Figure 4).* The authors concluded that a high 
     accident rate is one of the unexamined consequences of 
     insomnia and signals a need for greater physician 
     intervention. Several other studies1,7,26,40 
     specifically investigated the rate of motor vehicle accidents 
     and the rate of work-related accidents among insomniacs. Data 
     from these studies can be used to estimate the total cost of 
     insomnia-related accidents.

                        Motor Vehicle Accidents

       Individuals with insomnia are reported to have auto 
     accidents at a rate two to three times higher than the 
     general population.* A 1991 Gallup Poll found chronic 
     insomniacs reported 2.5 times as many fatigue-related car 
     accidents compared with good sleepers. Aldrick\40\ found that 
     29% of men and 15% of women with disorders of excessive 
     daytime sleepiness had had fatigue-related car accidents, 
     compared with 11% and 6% in the male and female control 
     groups, respectively (both groups, P <0.01 vs controls). The 
     increased car accident rate among insomniacs may be a result 
     of daytime fatigue, as the frequency of drowsy driving 
     doubles or triples in poor sleepers compared with good 
     sleepers (1.9 vs 0.8, P <0.05, Figure 5).\26\

                         Work-Related Accidents

       Lavie\1\ found that the most striking significant 
     difference between workers with excessive daytime sleepiness 
     and the rest of the population was the percentage that had 
     work accidents (52.1% vs 35.6%, P <0.0005). The rate of work-
     related accidents among individuals with insomnia can be 
     estimated at 1.5 times that of the general population.
       Why do insomniacs suffer two to three times the number of 
     auto accidents but only1.5 times the number of work-related 
     accidents? The possible discrepancy is explained by 
     Leger's\35\ findings that although 52.5% of work accidents 
     might reasonably be associated with sleepiness and human 
     error, the vast majority of motor vehicle accidents are 
     caused by human error. Taking these statistics into account, 
     the numbers are in remarkably good agreement. In both cases, 
     insomniacs cause two to three times as many accidents as 
     noninsomniacs.

               Cost of Insomnia-Related Accidents Methods

       The total cost of insomnia-related accidents can be 
     estimated by calculating the difference between accident 
     costs in insomniacs and accident costs in a group of the same 
     size from the general population:

                 $ ACCIDENT = (PREVI RISK n) + PREVG n)

     solve for n, then:

                  $$ COST = (PREVI RISK n) - (PREVI n)

     where $ ACCIDENT = cost of all accidents of the type being 
     considered in a given year ($70.2 billion for motor 
     vehicle\41\; $47.1 billion for work related\41\, $9.86 
     billion for home and public accidents--falls and 
     transportation based\35\); $$ COST = cost of insomnia-related 
     accidents; PREVI = prevalence of insomnia, or 33%; RISK = 
     rate of accidents of the type being considered in insomniacs 
     compared with that in the general population, estimated at 2 
     to 3 for motor vehicle\26\ and 1.5 for work related\1\, and 
     PREVG = prevalence of good sleepers, or 67%. Each category of 
     accident is considered separately.
       No specific data are available on the rate of home-based or 
     public accidents in insomniacs. Home-based or public 
     accidents account for only $28.3 billion of the $143.4 
     billion total cost of accidents.\40\ For the purpose of 
     estimating the total cost of these accidents due to insomnia, 
     the accident rate will be estimated as being two to three 
     times that in noninsomniacs. It seems reasonable to assume 
     that the increased risk for home-based or public accidents is 
     similar to the increased risk for auto-and work-related 
     accidents. As with work-related accidents, not all home or 
     public accidents are potentially related to fatigue. 
     Therefore, only falls and transportation-based public 
     accidents, which account for $9.86 billion of the total $28.3 
     billion, will be considered.\35\
       A second method can be used to calculate the cost of work-
     related accidents due to insomnia. The method multiplies the 
     cost of work-related accidents calculated by Leger\35\ to be 
     possibly sleep related ($24.7 billion) by the standard risk 
     factor for insomniacs--2 or 3--and the frequency of 
     insomnia--0.33. The difference between this figure and the 
     expected costs for a noninsomniac group calculated as above 
     yields an estimated cost of $6.13 to $9.82 billion for work-
     related accidents due to insomnia. thus both methods yield 
     cost estimate that are in good agreement.

                                Results

       The economic cost of the high accident rate among 
     insomniacs is staggering. The total cost of insomnia-related 
     accidents is estimated at $26.42 to $38.43 billion per year 
     compared with the total costs of accidents in 1988 of $143.4 
     billion (Table III). These estimates are based on 1988 
     accident statistics and should be considered low, due to 
     inflation and increased health care costs. Furthermore, these 
     figures do not include the cost of time lost by people not 
     directly involved in the accident, lawsuits directly related 
     to accidents, or catastrophic accidents caused by impaired 
     performance in the insomniac.35 These costs may not be 
     inconsequential; litigation and settlement costs of a serious 
     trucking accident exceeded $8 million in a case reported by 
     the National Commission on Sleep Disorders.3

  TABLE III.--TOTAL COST OF ACCIDENTS IN 1988 AND THE COST OF INSOMNIA- 
                            RELATED ACCIDENTS                           
------------------------------------------------------------------------
                                              Dollar cost in billions   
              Accident type              -------------------------------
                                             Total      Insomnia-Related
------------------------------------------------------------------------
Motor vehicle...........................         70.2     17.41 to 27.91
Work-related............................         47.1               6.60
Home and public.........................         28.3       2.44 to 3.92
All accidents...........................        143.4     26.42 to 38.43
------------------------------------------------------------------------

                         Catastrophic Accidents

       It has been observed that insomniacs who perform adequately 
     under normal conditions may exhibit a marked decrease in 
     performance under high-stress conditions.29 The 
     Association of Professional Sleep Societies' Committee of 
     Catastrophes, Sleep and Public Policy concluded, after 
     examining several industrial catastrophes: ``Sleep loss 
     combined with a period of stress, such as is faced by working 
     groups before production deadlines and launch deadlines, can 
     lead to personality change and irrational behavior.''31
       Although there are many catastrophes in which sleepiness, 
     sleep disorders, and fatigue are clearly implicated, it is 
     not possible to estimate the fiscal impact of insomnia-
     related catastrophic accidents.42,43 Despite post hoc 
     attempts to reconstruct chains of events, decision making, 
     and actions that unfolded during a crisis and to correlate 
     these with sleep impairment, consumption of alcohol, or any 
     other lifestyle issue, it essentially is impossible to 
     attribute a catastrophe to one single cause, such as 
     insomnia.31,42,43 Nevertheless, based on reconstructed 
     sleep/wake histories of key personnel, fatigue has been 
     implicated as a cause or a confounder of many catastrophes 
     (Table IV).

               TABLE IV.--SLEEP-RELATED CATASTROPHES31,42               
------------------------------------------------------------------------
              Accident                           Cause/Damage           
------------------------------------------------------------------------
Nuclear power:                                                          
    Three Mile Island..............  Coolant loss ignored by tired      
                                      worker.                           
    Chemobyl.......................  Reactor meltdown.                  
    David Besse reactor............  Safety feature overridden by       
                                      fatigued worker.                  
    Rancho Seco reactor............  Control system power loss.         
Space/air travel:                                                       
    Space Shuttle Challenger         Errors by fatigued managers.       
     explosion.                                                         
    Columbia Launch abortion.......  Loss of fuel overlooked by fatigued
                                      operators.                        
    China Airlines Flight 006......  Loss of control by fatigued        
                                      captain.                          
Rail:                                                                   
    Burlington Northern head-on      Engineer asleep/estimated damage:  
     collision (Wiggins, CO).         $3,891,428.                       
    Burlington Northern derailment   Engineer asleep/estimated damage:  
     (Newcastle, WY).                 $1,358,993.                       
Sea:                                                                    
    Exxon Valdez grounding.........  Inexperienced and fatigued third   
                                      mate.                             
    A. Regina grounding............  Estimated damage: $5.2 billion.    
                                     Master of ferry insomniac/damage:  
                                      $5 million.                       
------------------------------------------------------------------------

       The economic impact of catastrophes may run as high as $5.2 
     billion for the grounding of the Exxon Valdez. Such estimates 
     may be too low, however, as the real cost of a catastrophe 
     may be the loss of public trust.\43\ After the Three Mile 
     Island nuclear accident, no nuclear reactors were built in 
     the United States. What price can be put on the virtual 
     dissolution of an industry? Nonetheless, despite the 
     potential magnitude of this factor, the absence of 
     quantitative data and the inability to measure the degree to 
     which insomnia-related fatigue is a factor makes it 
     impossible to obtain an accurate economic analysis of the 
     cost of insomnia-related catastrophes.


                        MORBIDITY AND MORTALITY

       Long-term epidemiologic studies have shown that insomnia 
     and habitually short sleep, less than 6 or 7 hours per night, 
     are directly related to the development of heart disease, 
     high blood pressure, diabetes, stroke, and clinical 
     depression.44-48 Several very large long-term studies 
     have reported a significant increased risk of death among 
     those who complain of poor sleep.44,49,50 Numerous 
     studies have attempted to assess the degree to which 
     insomniacs consume a disproportionate amount of health care 
     resources and burden the health care system. This section 
     will assess the cost of this morbidity and dependence on 
     health care by individuals with insomnia.

                       Insomnia-Related Morbidity

       Habitual short sleepers and isomniacs are at higher risk of 
     becoming ill than are good sleepers. More than 50% of those 
     suffering from insomnia reported two or more health problems 
     during the past year, compared with approximately one third 
     of the entire population.\2\ In a study of 5419 Finnish men, 
     short sleepers had significantly more symptoms of corony 
     heart disease. This correlation held after controlling for 
     possible risk factors of coronary heart disease and 
     confounders.46,49 The American Cancer Society study on 
     more than 1 million adults documented a higher rate of fatal 
     coronary heart disease, stroke, and cancer in persons with 
     habitually short sleep.44,50 Shift workers who exhibit 
     high rates of insomnia have a higher incidence of 
     gastrointestinal disorders and heart disease.45,51 
     Industrial workers who suffered from excessive daytime 
     sleepiness were found to have significantly higher rates of 
     asthma, high blood pressure, arthritis, and ulcers compared 
     with satisfied sleepers.\1\
       The extent to which the association between insomnia and 
     poor sleep reflects compromised recovery or actual 
     debilitation in the insomniac due to sleep deprivation is 
     unclear. It is likely that insomnia is both a cause and 
     effect of poor health; the person with insomnia becomes 
     trapped in a cycle of pain or illness that interrupts sleep, 
     and simultaneously, the lack of sleep compounds the 
     disability.\3\ Until more data are collected on the role of 
     sleep in promoting good physical health and recovery and the 
     degree to which insomnia actually causes rather than is 
     associated with the development of illness, the cost of 
     insomnia-related morbidity cannot be estimated. This does not 
     mean that such costs should be ignored, however. Other 
     conditions that may compromise resistance and recovery, such 
     as alcohol abuse, escalate health care costs by billions of 
     dollar.\27\
       A different, intriguing perspective should be considered. 
     To the extent that it is an indicator of poor health2,46 
     or a risk factor for the development of future 
     disease,44,47 insomnia theoretically could save society 
     and employers money. This would be the case if persons 
     suffering from insomnia sought treatment, and physicians, 
     recognizing insomnia as a warning signal, screened for and 
     prevented future somatic and psychiatric diseases. This 
     scenario does not seem likely at this time because of the 
     widespread reluctance of insomniacs to seek treatment and the 
     tendency for physicians to trivialize the complaint of 
     insomnia.39,52

                       Insomnia-Caused Depression

       For many years, insomnia has been related to higher levels 
     of depressive illness.2,5,47,53 As many as 70% of 
     depressed patients report suffering from insomnia.\5\ Only 
     recently was it recognized that insomnia may precede the 
     development of depression and may be a causal factor rather 
     than a sequela. A study of 7946 adults, characterized as 
     ``probably one of the most scientifically rigorous 
     epidemiologic investigations of sleep disturbance and 
     psychopathology ever reported,''\53\ found a dramatically 
     lower incidence of new major depression in patients whose 
     insomnia has resolved (Figure 6).* By considering only the 
     development of new depression in the insomniac and 
     noninsomniac population, this study clarified the role of 
     insomnia in causing depression, not just the previously 
     documented association between insomnia and depression. The 
     study showed that individuals whose insomnia had resolved had 
     a slightly higher, though statistically insignificant, risk 
     (1.6) for developing major depression compared with 
     noninsomniacs. The risk of depression in unresolved 
     insomniacs skyrocketed to 39.8 times that of the 
     noninsomniacs. The incidence of developing major depression 
     was 0.4% for noninsomniacs, 0.6% for resolved insomniacs, and 
     14% for unresolved insomniacs. The investigators concluded 
     that early recognition and treatment of sleep disorders could 
     prevent future psychiatric disorders.
       Depression and mental health disorders are among the most 
     costly and destructive illnesses. An estimated 12 to 20 
     million Americans suffer from depression with an annual 
     direct cost to society of $10 billion.3,8 If up to 95% 
     of new cases of depression in chronic insomniacs could be 
     prevented through resolution of the insomnia, the direct 
     savings in medical care alone could be billions of dollars.

       Disproportionate Utilization of Health Care by Insomniacs

       Insomnia is associated with approximately a twofold 
     increase in the rate of hospitalization\11\ and a two- to 
     three-fold increase in office consultations.\6\ Elderly 
     persons with insomnia have a higher rate of 
     institutionalization in nursing homes than do elderly 
     noninsomniacs.\54\ The increased dependence on medical care 
     may reflect both increased morbidity and mortality associated 
     with insomnia, as well as an increased rate of serious 
     accidents.
       In a study comparing two large insomniac groups with 
     controls, Kales et al\11\ documented that adult chronic 
     insomniacs had a rate of hospitalization twice that of 
     noninsomniacs. Patients suffering from chronic insomnia had 
     been hospitalized a mean of 2.7 times compared with 1.4 
     hospitalizations for the control group (P < 0.01). Similar 
     results were presented in a study of more than 1500 adults 
     that examined the extent to which persons suffering from 
     insomnia used general hospital services: 21.9% of adults with 
     moderate-to-severe insomnia had had a nonpsychiatric 
     admission in the last year compared with 12.2% of controls (P 
     < 0.001).\6\ Bixler et al\10\ reported a somewhat lower but 
     significant rate of hospitalization among adults with 
     insomnia: 15.7%, compared with 11.6% of the total sample (P < 
     0.01). However, because 42.5% of the total sample was 
     classified as having insomnia, the hospitalization rate of 
     the noninsomniac population presumably was significantly 
     lower than that of the total sample.
       Of particular interest to employers are two studies that 
     compared the hospitalization rate of employees with and 
     without insomnia. Johnson and Spinweber's\18\ major 
     longitudinal study reported that 53.2% of poor sleepers had 
     been hospitalized one or more times compared with 39.9% of 
     good sleepers (P<0.02) and that 24.8% of poor sleepers had 
     been hospitalized two or more times compared with 14.2% of 
     good sleepers (P<0.01). Lavie's\1\ study also found a 
     significantly higher rate of hospitalization among those 
     complaining of poor sleep and excessive daytime sleepiness 
     (63% vs 52.4%, P<0.03).
       The economic costs related to an increased hospitalization 
     rate are substantial. Using an average of 1.3 additional 
     hospitalizations per person with insomnia\11\ and estimating 
     the average cost of a single hospital admission at 
     $5947.27,\55\ the additional hospital admissions in each of 
     the approximately 32 million adults with severe chronic 
     insomnia\3\ would cost society more than $25 billion.
       Insomnia also constitutes a major burden for the primary 
     care physician.6,47,56 Some data document a doubling of 
     primary care consultations due to insomnia, from a mean of 
     5.25 consultations per year in good sleepers, to 10.61 per 
     year in persons with mild insomnia, and 12.87 per year in 
     persons with moderate-to-severe insomnia (Figure 7).* 
     Extrapolating these data and the frequency estimates for 
     insomnia to the 179 million Americans 20 years of age or 
     older,\23\ insomniacs would generate an estimated additional 
     328 million primary care consultations over the expected 
     number. Of these, mild insomniacs are responsible for 143 
     million extra consultations, and moderate-to-severe 
     insomniacs for 184 million. The cost of these consultations, 
     using the 1990 mean office fee of $39.87,\55\ is $13.08 
     billion.
       Insomnia figures prominently in the decision to place an 
     elderly adult in a nursing home. A study of men older than 65 
     years concluded that, of all the lifestyle and health factors 
     considered, insomnia carried the highest hazard for nursing 
     home placement, higher than age, poor health, or cognitive 
     impairment (Figure 8).* Physicians have speculated that 
     nighttime sleep disruption becomes unbearable for caregivers 
     of elderly insomniacs.\39\
       The annual cost of nursing home care is approximately 
     $25,000 per patient, for a total annual cost of $59.9 billion 
     per year.\57\ For every day that insomnia hastens the 
     institutionalization of an elderly person, the medical cost 
     to society is about $80.3,55

                   Cost of Insomnia-Related Morbidity

       The National Commission on Sleep Disorders Research report 
     to Congress estimated that the direct medical expense of 
     insomnia was $15.4 billion in 1980.\3\ Despite clear evidence 
     that insomnia is associated with increased morbidity, it is 
     not possible to calculate the total indirect economic costs 
     of insomnia-caused illness. Research is just beginning to 
     address the issues of causation, covariance, and confounding 
     factors in correlating insomnia with illness. The frequency 
     with which insomnia occurs with other diagnoses creates a 
     major measuring difficulty.\47\
       In the preceding section, however, some calculations were 
     made to give an idea of the magnitude of the indirect costs 
     associated with insomnia-related morbidity. The cost of the 
     additional physician consultations associated with insomnia 
     was estimated at $13.08 billion per year. The cost of 
     additional hospitalizations over the lifetime of chronic 
     insomniacs was estimated at more than $25 billion. The 
     increased risk of nursing home placement due to insomnia is 
     documented, but associated costs were not estimated. None of 
     these estimated costs are included in this report's final 
     calculation due to the causation and correlation problems 
     discussed above.


                        insomnia and alcoholism

       Insomnia may be a causal factor in the development of 
     alcohol abuse and thus contribute to alcohol-related 
     morbidity and mortality. It has been long recognized that 
     there is a significant association between alcohol use and 
     insomnia (Figure 9).* The rate of alcoholism among insomniacs 
     is twice that of good sleepers.10,11,47
       Alcohol abusers frequently claim that they use alcohol as a 
     hypnotic; 60% report using alcohol to self-treat a sleep 
     disturbance.\59\ Despite the fact that 28% of insomniacs 
     report using alcohol to promote sleep,\7\ there may be some 
     doubt as to the legitimacy of the alcoholic's claim that his 
     or her drinking is insomnia related.
       However, two recent studies support a prodromal role for 
     insomnia in the development of alcoholism.47,59 In one 
     study that examined sleep complaints and use of alcohol, 60% 
     of alcoholic individuals claimed to use alcohol as a sleep 
     aid.\59\ Of these, the past sleeping history in 15.7% 
     indicated that alcohol abuse developed after the sleeping 
     disorder. From these data, it is reasonable to estimate that 
     9% to 10% of all alcoholism is the consequence of insomnia.
       A very large epidemiologic study confirmed not only that 
     insomnia may precede the development of alcohol abuse but 
     also suggested that treatment of insomnia may reduce the risk 
     of developing alcohol abuse.\47\ The incidence of new cases 
     of alcohol abuse was compared among good sleepers, those who 
     initially had reported insomnia but whose insomnia had 
     resolved, and those who had unresolved, and those who had 
     unresolved insomnia. By comparing only new cases of 
     alcoholism among those who were already suffering or had 
     suffered from insomnia, the extent to which insomnia is a 
     causal, not just an associated, factor in alcohol abuse was 
     clarified. Unresolved insomniacs had 2.4 times the risk of 
     developing alcoholism compared with good sleepers (Figure 
     10.* Those with resolved insomnia had a slightly higher risk 
     of developing alcohol abuse (1.4 times) than did good 
     sleepers. Considering only the approximately 15% of the 
     population with severe chronic insomnia\3\ and estimating 
     that the odds of developing alcohol abuse could be reduced 
     from 2.4 to 1.4\47\ if the insomnia were resolved, we again 
     reach the estimate that approximately 10% of all costs of 
     alcohol abuse could be attributed to insomnia:

                      (0.85 1) + (0.15 2.4) = 1.21

           (0.85 1) + (0.15 1.4) = 1.06 ( a reduction of 12%)

       Several economic studies have suggested that alcohol abuse 
     may be one of the single most costly health problems in 
     America.3, 27 Even the 10% estimated as being 
     attributable to insomnia is a staggering figure. Four recent 
     major studies, including that of the US Department of Health 
     and Human Services, estimated the cost of alcohol abuse at 
     $85 to $116 billion per year.\27\ These estimates include the 
     direct treatment costs of alcoholism, the costs of increased 
     morbidity and mortality associated with alcoholism, reduced 
     productivity and increased unemployment of the alcoholic 
     subject, and expenses related to increased accident and crime 
     rates for the alcoholic subject.\27\ Using these estimates of 
     the cost of alcoholism and the percentage of insomnia-related 
     alcoholism, the annual cost of insomnia-related alcoholism 
     may be between $8.5 and $11.6 billion. In addition, it has 
     been demonstrated that alcohol and sleepiness interact to 
     heighten the effect of each.60-62 Thus insomnacs may be 
     responsible for a disproportionate amount of the cost 
     associated with alcoholism. When more precise data 
     quantifying this interactive factor are collected, any 
     prevalence-based estimate of the cost of insomnia related to 
     alcoholism would have to be adjusted upward.

                  Increased Mortality Among Insomniacs

       Despite the fact that insomnia carries a higher death risk 
     than other factors, such as high blood pressure, which have 
     received significant public attention, mortality associated 
     with insomnia has received little attention.\63\
       Several large studies have shown that insomnia, or 
     habitually short sleep, is a powerful predictor of death and 
     that this correlation remains strong even after other factors 
     are controlled for, such as physical health. 
     \44\,\64\,\65\ The mortality risks associated with 
     a series of health-related behaviors in 6925 adults aged 30 
     to 69 years were calculated in a 9-year study that controlled 
     for age, sex, physical health, and many social and behavioral 
     factors.\64\,\65\ Sleeping fewer than 6 hours a day 
     carried the same mortality risk as physical inactivity and 
     high alcohol consumption; only cigarette smoking carried a 
     higher risk (Figure 11).* Persons sleeping fewer than 6 hours 
     a day had a 30% higher death rate than did those sleeping 7 
     or 8 hours (P<0.01).\65\ The American Cancer Society study 
     determined that habitually short sleep is associated with 
     increased mortality in every age group, even after 
     controlling for physical health.\44\ Men 30 years of age or 
     older who slept 4 or fewer hours per night were 2.8 times 
     more likely to die within 6 years than were those who slept 7 
     or 8 hours; women who slept poorly were 1.5 times more likely 
     to die.\50\ This correlation was independent of age or 
     medical history. Kripke et al\63\ calculated a particularly 
     high death rate among those persons habitually sleeping fewer 
     than 5 hours. Sleep duration was a better predictor of 
     mortality than a history of diabetes, heart disease, stroke, 
     or high blood pressure. Another study that examined mortality 
     in elderly adults calculated that sleeping fewer than 7 hours 
     a night increased mortality risk 1.34 times in women and 1.6 
     times in men.\66\
       The mechanism by which sleep duration or insomnia affects 
     mortality is unknown. The American Cancer Society study 
     suggested, ``It is commonly hypothesized that sleep serves as 
     yet unknown neurobiologic or restorative functions. Possibly, 
     insufficiency or excess of these functions can impair 
     longevity.''\44\ The direct correlation between sleep 
     duration and mortality is further supported by data 
     suggesting that treatment of insomnia reduces morbidity.\38\
       Premature death carries a high psychosocial cost in terms 
     of grief, pain, and suffering. The value of these cannot be 
     estimated. However, each death has a concrete economic cost 
     that can be estimated. For every premature death, society is 
     denied the productive contribution that person could have 
     made. Death takes a person out of the workplace and the loss 
     to society is the loss of his or her future productivity, 
     measured by his or her estimated future earnings. The value 
     of a life is calculated as the arithmetic sum of predicted 
     future lifetime earnings discounted by an adjustment rate, 
     generally 6%. Expected future earnings are calculated from 
     tables of annual mean earnings with a 1% per year rise 
     assumed. Although this method has been criticized as an 
     incomplete measure, it is straight-forward and measures an 
     important aspect of the cost of disease--the reduction in 
     economic resources caused by premature death. For these 
     reasons, it has become the method of choice in cost-of-
     illness studies.\27\,\28\
       Estimates of the value of a single life using the Human 
     Capital Approach method range from $568,546 for those aged 25 
     to 29 years, to $101,085 in those aged 60 to 64 years, the 
     age at which earnings generally cease.\27\ Because insomnia 
     was associated with 3.8% of all deaths in the American Cancer 
     Society study and was significant in all age groups (from age 
     30 to older than 90 years),\44\ it can be safely assumed that 
     billions of dollars are lost each year due to premature death 
     associated with insomnia. Estimation of the true figure would 
     involve a summation of the costs of the deaths directly 
     attributable to insomnia in each of the age and sex 
     categories based on wage data, a project beyond the scope of 
     this article.


                               CONCLUSION

       The direct and indirect costs attributable to insomnia are 
     staggering. Data limitations prevent calculation of cost 
     estimates for all factors that result in losses due to 
     insomnia, and several factors, such as pain and suffering, 
     marital problems, and an increased likelihood of catastrophic 
     accidents due to insomnia, cannot be quantified. Nonetheless, 
     the costs of insomnia may well reach more than $100 billion 
     per year. Table V summarizes the estimated costs calculated 
     in this article. Other cost estimates that were not included 
     in the overall total due to possible overlap between cost 
     categories and indirect costs that cannot be calculated are 
     summarized in Table VI. Such costs may add substantially to 
     the overall cost of insomnia to society.

         Table V.--Summary of the total annual cost of insomnia


        Related Cost                     Annual Dollar Cost in Billions
Loss productivity..................................................41.1
Direct medical cost of treatment...................................15.4
Insomnia-related depression.........................................1.0
Insomnia-related alcohol abuse..............................8.5 to 11.6
Accidents:
  Motor vehicle..........................................17.41 to 27.91
  Work-related......................................................6.6
  Home and public..........................................2.44 to 3.92
    Total...............................................92.45 to 107.53

               Table VI.--Other costs related to insomnia

     Costs that can be estimated:
       Increase in employee absenteeism, $4800 yearly per 
     individual with insomnia.
       Increase in hospitalization rate, $7731 per individual with 
     chronic insomnia.
       Increase in primary care consultations, $13.08 billion 
     yearly.
     Indirect costs that cannot be estimated:
       Catastrophes
       Increase in morbidity
       Increase in mortality
       Lost job opportunities
       Academic failure
       Nursing home care

       A silver lining may exist, however. New research suggests 
     that insomnia may precede the expression of psychiatric and 
     somatic disorders, such as depression and coronary heart 
     disease. This finding raises the possibility that insomnia 
     may serve as a valuable warning flag such that treatment of 
     insomnia may have value in preventing other 
     illnesses.44,47 Unfortunately, many primary care 
     physicians remain unresponsive to complaints of insomnia, and 
     many persons suffering from insomnia trivialize their own 
     conditions.\52\ Thus this opportunity for interventon remains 
     theoretical.


                             ACKNOWLEDGMENT

       This project was funded through a grant provided by G.D. 
     Searle & Co., Chicago, Illinois.


                               references

       1. Lavie P. Sleep habits and sleep disturbances in 
     industrial workers in Israel: Main findings and some 
     characteristics of workers complaining of excessive daytime 
     sleepiness. Sleep 1891;4:147-158.
       2. Mellinger GD, Balter MB, Uhlenhuth EH. Insomnia and its 
     treatment--prevalence and correlates. Arch Gen Psychiatry. 
     1985;42:225-232.
       3. Wake Up America: A National Sleep Alert. Vol. 1. 
     Executive Summary and Executive Report. Report of the 
     National Commission on Sleep Disorders Research. Washington, 
     DC: National Institutes of Health, US Government Printing 
     Office; 1993.
       4. Karacan I. Thornby J. Willams RL. Sleep disturbance: A 
     community survey. In: Guilleminault C. Lugaresi E, eds. 
     Sleep/Wake Disorders: Natural History, Epidemiology, and 
     Long-term Evolution. New York Raven Press; 1983:37-60.
       5. Hohagen F. Rink K, Kappler C, et al. Prevalence and 
     treatment of insomnia in general practice--a longitudinal 
     study. Eur Arch Psych Clin Neurosci, 1983; 242:329-336.
       6. Weyerer S, Dilling H. Prevalence and treatment of 
     insomnia in the community: Results from the Upper Bavarian 
     field study. Sleep. 1991;14:392-398.
       7. Sleep in America: A National Survey of U.S. Adults. A 
     report prepared by the Gallup Organization for the National 
     Sleep Foundation. Los Angeles, Ca: National Sleep Foundation; 
     1991.
       8. Kaplan HI, Sadock BJ. Pocket Handbook of Clinical 
     Psychiatry. London: Williams & Wilkins; 1990:81-95,132-133.
       9. Lugaresi E, Cirignotta F, Zucconi M, et al. Good and 
     poor sleepers: An epidemiological survey of the San Marino 
     population. In: Guilleminault C, Lugaresi E, eds. Sleep/Wake 
     Disorders: Natural History, Epidemiology, and Long-term 
     Evolution. New York: Raven Press; 1983:1-12.
       10. Bixler EO, Kales A, Soldatos C, et al. Prevalence of 
     sleep disorders in the Los Angeles metropolitan area. Am J 
     Psychiatry. 1979; 136:1257-1262.
       11. Kales JD, Kales A, Bixler EO, et al. 
     Biopsychobehavioral correlates of insomnia. V. Clinical 
     characteristics and behavioral correlates. Am J Psychiatry. 
     1948;141: 1371-1376.
       12. Marchini EJ, Coates TJ, Magistad JG, Waldum SJ. What do 
     insomniacs do, think, and feel during the day? A preliminary 
     study, Sleep. 1983;6:147-155.
       13. Dement WC, Mitler MM. It's time to wake up to the 
     importance of sleep disorders, JAMA. 1993;269:1548-1550.
       14. Balter MB, Uhlenhuth EH. The beneficial and adverse 
     effects of hypnotics, J Clin Psychiatry, 1991;52(Supp):16-23.
       15. Leigh P. Employee and job attributes as predictors of 
     absenteeism in a national sample of workers: The importance 
     of health and dangerous working conditions, Soc Sci Med. 
     1991;33:127-137.
       16. Buxrud EG, Bjorndal A. Why are female physicians more 
     often absent from work due to sickness than their male 
     colleagues? Tidsskr Nor Laegeforen. 1991;111:1132-1135.
       17. Blum D, Kahn A, Mozin MJ, et al. Relation between 
     chronic insomnia and school failure in preadolescents. Sleep 
     Res. 1990;18:1. Abstract.
       18. Johnson LC, Spinweber CL. Quality of sleep and 
     performance in the Navy: A longitudinal study of good and 
     poor sleepers. In: Guilleminault C, Lugaresi E, eds. Sleep/
     Wake Disorders: Natural History, Epidemiology, and Long-term 
     Evolution. New York: Raven Press; 1983:13-28.
       19. Gooodman PS, Atkin RS. Effects of absenteeism on 
     individuals and organizations. In: Goodman PS, Atkin RS, et 
     al, eds. Absenteeism--New Approaches to Understanding, 
     Measuring and Managing Employee Absence. San Francisco: 
     Jossey-Bass; 1984:276-321.
       20. Katz HC, Kochan TA, Gobeille KR. Industrial relations 
     performance, economic performance and quality of working life 
     efforts: An interplant analysis, Data on file, Massachusetts 
     Institute of Technology, 1982.
       21. Katz HC. Kochan TA. Weber MR. Assessing the effects of 
     industrial relations and quality of working life efforts on 
     organizational effectiveness. Data on file, Massachusetts 
     Institute of Technology, 1982.
       22. Mirvis PH, Lawler E. Measuring the financial impact of 
     employee attitude. J Appl Psychol. 1977;62:1-8.
       23. Statistical Abstract of the United States, 19. The 
     National Data Book 113th ed. Washington, DC: US Department of 
     Commerce, Economics and Statistics Administration; 
     1993:16,393,425,430.
       24. Statistical Abstract of the United States, 19. The 
     National Data Book 101st ed. Washington, DC: US Department of 
     Commerce, Economics and Statistics Administration; 1980:427.
       25. Mendelson WB, Gamett D, Linnoila M. Do insomniacs have 
     impaired daytime functioning? Biol Psychiatry. 1984;19:1261-
     1264.
       26. Schweitzer PK, Engelhardt CL, Hilliker NA, et al. 
     Consequences of reported poor sleep. Sleep Res. 1992;21:2. 
     Abstract.
       27. The Economic Costs of Alcohol and Drug Abuse and Mental 
     Illness: 19, Hyattsville, Md: US Department of Health and 
     Human Services; DHHS Pub. No. (ADM) 90-1694; 1990.
       28. Cooper B, Rice DP. The economic cost of illness 
     revisited. Soc Sec Bull. 1976;Feb:21-36.
       29. John MW, Dudley HAF, Masterson JP. The sleep habits, 
     personality and academic performance of medical students. Med 
     Educ. 1976;10:158-162.
       30. The 1994 Naval Times Pay Chart. Washington, DC: 
     Department of Defense Office of Compensation; 1994.
       31. Mitler MM, Carskadon MA, Czeisler CA, et al. 
     Catastrophes, sleep and public policy: Consensus report. 
     Sleep. 1988;11:100-109.
       32. Spinweber CL, Bellune J. Development of insomnia is 
     associated with failure in SEAL team training. Sleep Res. 
     1988;17:2. Abstract.
       33. US Bureau of Motor Carrier Safety. Analysis and Summary 
     of Accident Investigations, 19. Washington, DC: Department of 
     Transportation; 1970.
       34. Transportation Research and Marketing. A Report on the 
     Determination and Evaluation of the Role of Fatigue in Heavy 
     Truck Accidents. Falls Church, Va; American Automobile 
     Association Foundation in cooperation with the Commercial 
     Vehicle Safety Alliance; 1985.
       35. Leger D. The cost of sleep-related accidents: A report 
     for the National Commission on Sleep Disorders Research. 
     Sleep. 1994;17:84-93.
       36. Fatigue, Alcohol, Other Drugs and Medical Factors in 
     Fatal to the Driver Heavy Truck Crashes. Washington, DC: 
     National Transportation Safety Board; Vol. 1.PB90-9170. NTSB/
     SS-90/01; 1990.
       37. Parsons M. Fits and other causes of loss of 
     consciousness while driving. Q J Med. 1986;58:295-303.
       38. Balter MB, Uhlenhuth EH. New epidemiologic findings 
     about insomnia and its treatment. J. Clin Psychiatry. 
     1992;53(Suppl): 34-39.
       39. Shapiro CM, Dement WC. Impact and epidemiology of sleep 
     disorders. BMJ. 1993;306:1604-1607.
       40. Aldrich MS. Automobile accidents in patients with sleep 
     disorders. Sleep. 1989;12:487-494.
       41. Accident Facts. Chicago: National Safety Council; 1989.
       42. Lauber JK, Kayten P. Sleepiness, circadian dysrhythmia, 
     and fatigue in transportation system accidents. Sleep. 
     1988;11:503-512.
       43. Mitler MM, Dinges DF, Dement WC. Sleep medicine, public 
     policy and public health. In: Kreiger M, Roth T, Dement W, 
     eds. Principles and Practice of Sleep Medicine. Philadelphia: 
     WB Saunders; 1984:453-462.
       44. Hammond EC. Some preliminary findings on physical 
     complaints from a prospective study of 1,064,004 men and 
     women. Am J. Public Health. 1964;54:11-23.
       45. Knutsson A, Akerstedt T, Jonsson BG, Orth-Gomer K. 
     Increased risk of ischaemic heart disease in shift workers. 
     Lancet 1986;2:89-92.
       46. Urponen H, Vuori I, Hasan J, Partinen M. Sleeping 
     habits, sleep quality and sleep disorders in urban working 
     aged people. J Soc Med. 1988;25:338-351. In Finnish
       47. Ford DE, Kamerow DB. Epidemiologic study of sleep 
     disturbances and psychiatric disorders--an opportunity for 
     prevention? JAMA. 1989;262:1479--1484.
       48. Report of the Presidential Commission on the Space 
     Shuttle Challenger Accident. Washington, DC: National 
     Aeronautics and Space Administration; 1986. Appendix G.
       49. Partinen M, Putkonen TS, Kaprio J, et al. Sleep 
     disorders in relation to coronary heart disease. Acta Med 
     Scand (Suppl.) 1982;660:69-83.
       50. Kripke DF, Simons, RN, Garfinkel L, Hammond EC. Short 
     and long sleep and sleeping pills--is increase mortality 
     associated? Arch Gen Psychiatry. 1979;36:103-116.
       51. Koller M. Health risks related to shift work: An 
     example of time contingent effects on long-term stress. Int 
     Arch Occup Environ Health. 1983;53:59-75.
       52. Dement WC. The proper use of sleeping pills in the 
     primary care setting. J Clin Psychiatry. 1992;52(Suppl):50-56
       53. Reynolds CF. The implications of sleep disturbance 
     epidemiology. JAMA. 1989; 262:1514
       54. Pollak CP, Perlick D. Linsner J, et al. Sleep problems 
     in the community elderly as predictors of death and nursing 
     home placement. J Commun Health. 1990;15:123-135.
       55. Feldstein P. Health Care Economics. Albany, NY: Delamar 
     Publishers; 1993: 219,513.
       56. Weyerer S. Insomnia and its treatment in the general 
     population. Sleep Res. 1991;20A: 406.Abstract
       57. Health United States 1992 and Healthy People 2000 
     Review. Hyattsville, Md: US Department of Health and Human 
     Services; 1992: DHHS Pub. No. (PHS) 93-1232.
       58. Braz S, Masur J, Formigoni ML, et al. Alcohol 
     consumption and sleep complaints in the city of Sao Paulo: An 
     epidemiological survey. Sleep Res. 1991;20:3. Abstract.
       59. Mamdani MB, Hollyfield RL, Ravi SD, et al. Sleep 
     complaints and recidivism in alcoholics reporting use of 
     alcohol as a hypnotic. Sleep Res. 1988;17:2. Abstract
       60. Lumley M, Roehrs T, Asker D, et al. Ethanol and 
     caffeine effects on daytime sleepiness/alertness. Sleep. 
     1987; 10:306-312.
       61. Poppy J. Sleep Walking. Esquire. 1990;114: 101-103.
       62. Roehrs T, Zwyghuizen-Doorenbos A, Timms V. et al. Sleep 
     extension, enhanced alertness and the sedating effects of 
     ethanol. Pharmacol Biochem Behav. 1989;34:321-324.
       63. Kripke DF, Ancoli-Israel S, Mason W, Messin S. Sleep 
     related mortality and morbidity in the aged. In: Chase M, 
     Weitzman ED, eds. Sleep Disorders--Basic and Clinical 
     Research. New York, NY: Spectrum; 1983:415-429
       64. Wingard DL, Berkman LF, Brand R. A multivariate 
     analysis of health-related practices--a nine-year mortality 
     followup of the Alameda County study. Am J Epidemiol. 
     1982;116:765-775.
       65. Wingard DL, Berkman LF. Mortality risk associated with 
     sleeping patterns among adults. Sleep. 1983;6:102-107.
       66. Branch LG, Jette AM. Personal health practices and 
     mortality among the elderly. AmJ Public Health. 1984;74:1126-
     1129.
       *Illustrations not reproducible in the Record.

                          ____________________