American Journal of Medical Genetics Part C (Semin. Med. Genet.) 127C:42 –50 (2004) A R T I C L E Cost of Fetal Alcohol Spectrum Disorders CHUCK LUPTON,* LARRY BURD, AND RICK HARWOOD Fetal alcohol syndrome (FAS) is a common identifiable teratogenic cause of mental retardation, neurological deficit, mental disorders, and developmental disabilities. Accurate estimates of the cost of care for persons with FAS are essential for appropriate funding of health care, developmental disabilities services, special education, and other service systems, as well as prioritizing funding of public health prevention efforts. The cost of care for individuals with FAS can be conceptualized as the annual cost of care for one person or a population, or as the lifetime cost of care for an individual. Annual cost estimates for the United States range from $75 million in 1984 to $4.0 billion in 1998. Estimates of lifetime cost vary from $596,000 in 1980 to $1.4 million in 1988. After adjustments for changes in inflation and population, 2002 estimates of total annual cost and lifetime cost are higher. FAS is increasingly being recognized as a large public health problem with high potential for the prevention of future cases and for the prevention of excess disability and premature mortality in persons who are affected. Each day, from 6–22 infants with FAS are born in the United States, and as many as 87–103 more are born with other impairments resulting from prenatal alcohol exposure. Updated and improved cost data on FAS should be a research priority. ß 2004 Wiley-Liss, Inc. KEY WORDS: fetal alcohol syndrome; cost; annual cost of care; lifetime cost of care; prevention INTRODUCTION In this paper we provide an overview of fetal alcohol spectrum disorders (FASDs), of which fetal alcohol syndrome (FAS) is the most well-known consequence, and issues pertinent to the estimation of costs. Since cost information only exists for FAS, our review presents data on FAS for the annual cost of care for a population, the lifetime cost of care per case, and the annual cost of care per case. We conclude with sugges- Mr. Lupton is the Deputy Project Director for Operations of the Substance Abuse and Mental Health Services Administration (SAMHSA) Fetal Alcohol Spectrum Disorders (FASD) Center for Excellence, Rockville, Maryland. Dr. Burd is the Director of the North Dakota Fetal Alcohol Syndrome Center at the University of North Dakota School of Medicine and Health Sciences, Grand Forks, North Dakota. Mr. Harwood is Vice President, The Lewin Group, Falls Church, Virginia. Sponsor: SAMHSA FASD Center for Excellence, U.S. Department of Health and Human Services; Contract number: 277-01-6068. *Correspondence to: Chuck Lupton, Deputy Project Director for Operations, SAMHSA FASD Center for Excellence, 1700 Research Blvd., Suite 400, Rockville, MD 20850. E-mail: firstname.lastname@example.org DOI 10.1002/ajmg.c.30015 ß 2004 Wiley-Liss, Inc. tions for further research on the costs for all FASDs. Each year, more than 500,000 fetuses (about 13% of all births) in the United States are exposed to alcohol during pregnancy, most at very low levels and only early in pregnancy [Centers for Disease Control, 2002]. The prevalence of alcohol use decreases during pregnancy [Stratton et al., 1996; Morse et al., 1997]. However, 3.3% of women continue to drink throughout their pregnancy on either a frequent(sevenor more drinksper week) or binge-drinking (five or more drinks on any occasion) basis. Annually, about 130,000 pregnant women in the United States expose their fetuses to high levels of alcohol and thus put them at risk for FAS or other impairments resulting from prenatal alcohol exposure Annually, about 130,000 pregnant women in the United States expose their fetuses to high levels of alcohol and thus put them at risk for FAS or other impairments resulting from prenatal alcohol exposure. (i.e., fetal alcohol effects (FAE)). Many of these women also smoke heavily, creating additional risks for their fetuses [Stratton et al., 1996; Abel, 1998]. Alcohol use during pregnancy can have adverse effects on fetal development [Burd and Martsolf, 1989; Abel, 1998]. Severe outcomes include mortality and FAS. The diagnosis of FAS and some related conditions requires a documented history of prenatal alcohol exposure. Yet, there are no conclusive data on the dose, timing, or duration of exposure that is likely or necessary to cause FAS or other related impairments [Abel, 1984; Day et al., 2002; Olney et al., 2002]. The outcome of exposure has been difficult to assess due in part to difficulty in sorting out the respective contributions of the exposure parameters of dose and timing, genetic factors influencing susceptibility, protective factors, and lack of agreement on diagnostic criteria [Stratton et al., 1996; Abel, 1998; Day et al., 2002; Olney et al., 2002]. Prenatal alcohol exposure is a necessary cause of FAS and FAE, but by itself is not always sufficient to cause FAS [Burd and Martsolf, 1989; Burd and Wentz, 1997; Abel, 1998]. Theoretical models of adverse behaviors and factors contributing to FAS have been ARTICLE described as a multielement causal chain of interacting factors [Burd and Wentz, 1997; Abel, 1998; Burd et al., 2003]. Risk factors commonly, but not invariably, include smoking, poor diet, heavy drinking, binge alcohol use, unmarried status, physical abuse, increased maternal age, and increased parity [Burd et al., 1996b, 2003; Abel, 1998]. Prevalence of FAS FAS was first named in 1973, and the disorder came to be accepted by the mid-1980s as the leading identifiable cause of mental retardation and neurologic deficit in the Western world [Abel and Sokol, 1986]. A recent review of the FAS literature concludes that about 10 in 1,000 births (or about 40,000 births per year) have adverse effects due to maternal alcohol use, of whom 0.5–2 per 1,000 (from 2,000–8,000) per year have FAS [May and Gossage, 2001]. FAS prevalence rates have been found to vary widely in some minority populations [Burd and Moffatt, 1994; Famy et al., 1998; May and Gossage, 2001]. When all FASDs (FAS and FAE) are combined, the prevalence rate has been estimated to be in the range of 9–10 cases per 1,000 live births When all FASDs (FAS and FAE) are combined, the prevalence rate has been estimated to be in the range of 9–10 cases per 1,000 live births. [Sampson et al., 1997; May and Gossage, 2001; Astley, 2002]. These prevalence rates may be conservative since they have generally been based on passive, rather than active, surveillance. The Phenotype of FAS The diagnosis of FAS is most easily made between 4 and 14 years of age, when the diagnostic signs are most evident. Attempts to identify FAS in the new- AMERICAN JOURNAL OF MEDICAL GENETICS (SEMIN. MED. GENET.) born period have been problematic [Little et al., 1990]. Many of the dysmorphic signs of FAS (e.g., short palpebral fissures, thin upper lip, absent/ indistinct philtrum) may become less distinct or disappear in late adolescence and adult life [Streissguth et al., 1991; Spohr and Steinhausen, 1996]. Thus, in adolescents or adults, the diagnosis of FAS can be difficult when it is based on a physical examination without supporting history, childhood pictures, or other developmental information. The difficulty of diagnostic certainty reduces the accuracy of prevalence estimates, and FAS and FAE may be underdiagnosed because of concerns about stigma for the mother. The Cognitive, Neurologic, and Behavioral Manifestations of FAS and FAE The delineation of FAS and FAE has been especially problematic [Abel and Sokol, 1986; Burd and Martsolf, 1989; Streissguth et al., 1991; Steinhausen et al., 1994; Sampson et al., 1997; Famy et al., 1998; Burd et al., 2003]. An important area of current emphasis is to determine if prenatal alcohol exposure results in a unique or clearly definable behavioral or cognitive pattern of manifestations, which may result in a distinctive behavioral phenotype. Several neurocognitive features frequently associated with FAS and FAE are learning disabilities, speech and language delays, problems with reasoning and judgment, impairments in abstract thinking, diminished impulse control, volatile emotions, poor social skills, and vulnerability to peer pressure. Increasing attention is also being given to the role of prenatal alcohol exposure as an etiologic factor in a number of other developmental disorders and mental illnesses [Streissguth et al., 1991; Burd et al., 1996a, 2003; Famy et al., 1998]. Previous research has found that over 90% of persons with FAS have mental health problems and other indicators of poor fit with societal expectations [Streissguth et al., 1991]. High rates of mental health problems have been reported in a sample of 43 affected children and adolescents from Germany [Steinhausen et al., 1994]. In this sample the prevalence of these problems seemed to increase with age. In a study of 78 patients with FAS or FAE, 92% received an Axis I diagnosis of drug or alcohol dependence [Famy et al., 1998]. Forty-four percent had depression, 20% had bipolar disorder, and 40% reported other psychiatric symptoms [Famy et al., 1998]. Individuals with FAS and FAE frequently have inattention, hyperactivity, and/or impulsivity problems that are often diagnosed as attention-deficit/hyperactivity disorder (ADHD). Vision, hearing, and speech/ language issues are other common problems for children with FAS. METHODS AND RESULTS What Is Known about the Cost of FASD? We reviewed the literature to determine the major elements of cost for individuals with FAS. Published studies using well-documented approaches fall into two categories: the total annual cost of FAS to the nation, and the lifetime cost of caring for each child born with FAS. Several studies have been done at the state level, however, and no estimates were found for countries other than the United States. Annual U.S. Cost Estimates Over the past 18 years, four welldocumented studies to estimate the annual cost of FAS in the United States have been published. They identify costs by category, such as health care costs, residential and support services, and productivity losses. All three of the lower cost estimates were done by Abel and Sokol. The lowest estimate by Abel and Sokol [1991a] was $75 million for 1984. This was based on the lowest prevalence rate of all the studies: 0.33 babies per 1,000 live births. This estimate included medical treatment for children with FAS up to age 21 and residential care due to mental retardation up to age 21. 44 AMERICAN JOURNAL OF MEDICAL GENETICS (SEMIN. MED. GENET.) Residential care accounted for 77% of the total cost. Two other studies by Abel and Sokol [1987, 1991b] used a prevalence rate of 1.9 cases per 1,000 live births. In their second study, they estimated the total cost to the United States to be $321 million in 1984 for medical treatment and residential care, with much of the difference being related to the use of a prevalence rate six times larger. In their third study, they estimated the total 1987 annual cost at $250 million. Abel and Sokol’s third study included corrections for background rates of low birth weight and costs normally incurred for housing and food, regardless of whether an individual required institutionalization. None of the studies by Abel and Sokol included costs of lost productivity, semi-independent support services (e.g., persons living in community settings with ambulatory care and special education services), or residential services due to mental retardation after age 21. The higher range of cost estimates came from studies by Rice et al. , Rice , Harwood et al. [1984, 1998], and Harwood . Rice et al. estimated 1985 annual costs to be $1.6 billion, using a prevalence rate of 1.9 per 1,000 live births. This cost included neonatal intensive care services and other treatment and care services up to age 21. The estimate was based on the Abel and Sokol approach used for their $321 million estimate. The 1985 estimate by Rice et al. also included residential care for mental retardation for persons over age 21, which accounted for 80% of the total annual costs. Later, Rice projected that the total 1990 cost had risen to $2.1 billion, based on growth in population and increases in health care costs between 1985 and 1990. Another estimate comes from a National Institute on Drug Abuse (NIDA) and National Institute on Alcohol Abuse and Alcoholism (NIAAA) study by Harwood et al., who estimated the 1992 annual cost to be $2.9 billion, based on the basic approach used by Harwood and Napolitano , but using more current data to develop improved estimates [Harwood et al., 1984, 1998; Harwood, 2000]. This study used a prevalence rate of 2.0 cases per 1,000 live births for recent birth cohorts and a rate of 1.0 for earlier birth cohorts. It included treatment and care services to age 21, home and residential care services for moderate and severe cases of mental retardation to age 65, special education services, and lost productivity. A projection by Harwood  from the NIDA/NIAAA study estimated that costs had risen to $4.0 billion by 1998. This updated estimate adjusted for the change in national health care expenditures and the consumer price index for medical services. It also adjusted for changes in the adult population in the United States and in the hourly compensation index for lost productivity. No adjustment was made for FAS-specific trends. Harwood and Napolitano  estimated 1980 annual costs at $3.2 billion, using a prevalence rate of 1.67 cases per 1,000 live births [see also Harwood et al., 1984]. Their estimate included the cost of medical treatment, home and residential care, special education services, and lost productivity for all ages. Table I summarizes the estimates of annual costs, ranging from $75 million at the low end to $4.0 billion at the high end. These diverse estimates reflect the effect of very different assumptions about prevalence rates of FAS, application of FAS prevalence rates to recent and older birth cohorts, and included cost components. This table also provides adjusted 2002 estimates of the annual cost based on changes in population growth and inflation in the cost of medical care services, and exclusion of lost productivity costs. Even with these adjustments, great variation in estimates persists, ranging from $0.2 billion to $9.3 billion. Finally, adjustments are made for prevalence rates and residential care over age 21, with the result being a much smaller difference among studies. These final adjustments show that five of the six studies have estimated costs from $2.3 billion to $6.0 billion for the year 2002, with only the earliest study being significantly higher than the others. The median of these adjusted cost estimates is $3.6 billion. ARTICLE The median of the adjusted cost estimates is $3.6 billion. Variance in annual cost estimates Six major factors account for the variances in annual cost estimates: 1. Prevalence rates substantially affect overall costs when costs are based on a fixed cost per subject. Cost estimates based on low prevalence rates, such as Abel and Sokol’s [1991a] 0.33 cases per 1,000 live births, will be proportionately less than estimates based on prevalence rates of 1.9–2 cases per 1,000. A key factor is both a description of the inclusion criteria for cases and how comorbidity is included in the cost estimates. 2. Differences exist in the estimates of the types, utilization rates, and costs of medical care, residential services, and other services used by individuals with FAS. 3. Studies that estimate residential and support services for affected individuals with mental retardation to age 65 reflect much higher costs than studies that estimate such costs only to age 21. 4. Inflation accounts for some differences among estimates, although we have attempted to adjust for this. 5. The knowledge base for developing cost estimates rapidly increased over time. Thus, we believe the more recent estimates should be given more weight since they have learned from and built on prior estimates. 6. Estimates including the cost of lost productivity typically exceed estimates that do not include productivity costs. What Is the Total Lifetime Cost of FAS? The literature contains only two welldocumented estimates of the total lifetime cost for a person with FAS (Table II). Harwood and Napolitano ARTICLE AMERICAN JOURNAL OF MEDICAL GENETICS (SEMIN. MED. GENET.) 45 TABLE I. Original and Adjusted Estimates of the Total U.S. Annual Cost of Fetal Alcohol Syndrome in Billions of Dollars Ranked by Prevalence Rate per Thousand* Original annual cost estimate 2002 estimate adjusted for population and inflation with productivity losses excludeda 2002 estimate adjusted for prevalence rate and residential care over age 21 with productivity losses excludeda Source of estimate Year Prevalence rate per thousand live births Abel and Sokol [Abel and Sokol, 1991a] Harwood and Napolitano [Harwood et al., 1984] Abel and Sokol [Abel and Sokol, 1987] Abel and Sokol [Abel and Sokol, 1991b] Rice et al. [Rice et al., 1990] Rice; update of 1985 study [Rice, 1993] Harwood et al. [Harwood et al., 1998] Harwood; update of 1992 study [Harwood, 2000] 1984 0.33 $0.075 $0.18 $4.75 1980 1.67 $3.235 $9.26 $11.09 1984 1.9 $0.321 $0.85 $3.57 1987 1.9 $0.250 $0.55 $2.34 1985 1.9 $1.611 $3.42 $3.60 1990 1.9 $2.089 Not calculated Not calculated 1992 2.0/1.0b $2.934 $2.97 $5.95 1998 2.0/1.0b $4.022 Not calculated Not calculated *Source: SAMHSA FASD Center for Excellence, 2003. a These columns exclude costs associated with productivity losses for comparison purposes. The far right column shows the results from adjusting the prevalence rate for all estimates to 2 per 1,000 live births, and adjusting the Abel and Sokol estimates to include residential care for persons over the age of 21. b The first value was used for population cohorts since 1971, and the second value was used for cohorts born prior to that date. This rudimentary adjustment was due to the increase in drinking by females during the last 75 years.  estimated lifetime cost at $596,000 in 1980. If this estimate is adjusted based on the change in the cost of medical care services and lost productivity, accounting for inflation, the adjusted 2002 cost becomes $2.0 million for each individual with FAS. This is an average for all persons with FAS; persons with profound mental retardation have much higher costs than those who do not. This $2 million figure is made up of $1.6 million of medical care services and $0.4 million for productivity losses. Figure 1 shows how cumulative lifetime costs of $2 million for one case of FAS are spread out over the person’s lifetime. One prevented case of FAS saves almost $130,000 in the first 5 years, $360,000 in 10 years, $587,000 in 15 years, and more than $1 million in 30 years. In February 1989, the Senate Advisory Council of the Alaska State Legislature estimated the 1988 lifetime cost of each baby born with FAS at $1.4 million TABLE II. Original and Adjusted Estimates of the Total U.S. Lifetime Cost for Each Child Born With Fetal Alcohol Syndrome (FAS*) Source of estimate Date Original estimate of total U.S. lifetime cost of child born with FAS Harwood and Napolitano [Harwood et al., 1985] Alaska State Legislature [Weeks, 1989] 1980 $596,000 $2,010,000 $931,742 1988 $1,373,836 $2,909,785 $1,466,875 *Source: SAMHSA FASD Center for Excellence, 2003. Adjusted 2002 estimate of total U.S. lifetime cost based on inflation Discounted estimate of total 2002 U.S. lifetime costs after adjustment for inflation 46 AMERICAN JOURNAL OF MEDICAL GENETICS (SEMIN. MED. GENET.) Figure 1. [Weeks, 1989]. This estimate adapted the general approach used by Harwood and Napolitano  and based medical care costs on studies by Abel and Sokol [1991a] and Harwood and Napolitano . It also included services for affected children (e.g., developmental disability services), special education, social service costs, adult vocational services, and institutional care for mental retardation to age 65. The lifetime cost estimate from Alaska is higher than the Harwood and Napolitano estimate for at least three reasons: inflation in the cost of medical care services from 1980–1988, the much higher costs for medical care services in Alaska, and the inclusion of additional service categories, e.g., child and youth services, developmental disability services, and adult vocational services. The cost estimation formula for Alaska was considered conservative because of excluded costs: some medical care and physician services up to age 1, welfare payments, mental health services, criminal justice costs, services for mild physical problems and learning disabilities, or lost productivityof caregivers and persons with FAS. Because costs in Alaska are generally higher than national costs, ARTICLE Cumulative cost for one case of FAS to age 65. this estimate is not appropriate for use as a reliable national figure. When updated based on the medical services price index, the adjusted 2002 total cost per individual is $2.9 million. When updated based on the medical services price index, the adjusted 2002 total cost per individual is $2.9 million. Is State Cost Information Available? Limited cost information is available at the state level. However, four states have made estimates of the cost of FAS, which are described below. South Dakota. The South Dakota Department of Health estimated that the yearly cost of taking care of an individual with FAS ranged from $10,000 to $30,000 [Stanage et al., 1983] in 1982. Using incidence rates of 1 and 2 cases of FAS per 1,000 live births and 12,839 births per year, the 1982 annual cost was estimated to range from $120,000 to $360,000 (at 1 case per 1,000 ¼ 12 persons) to $240,000 to $720,000 (at 2 cases per 1,000 ¼ 24 persons). These costs include expenses only for children born with full FAS. Based on a 60-year life expectancy, the estimated costs for FAS range from $7.2 to $43.2 million for each year’s birth cohort. For instance, if the cost per year is $10,000 and life expectancy is 60 years, the lifetime cost per person is $600,000. If there are 12 births during the year, the total lifetime cost for this birth cohort becomes $7.2 million. Other expected lifetime cost combinations (undiscounted) of prevalence rates and cost per year are shown below. These cost combinations do not include costs for FAS births from prior years. Prevalence Rate Cost of care/ person One case/ thousand Two cases/ thousand $10,000/year $7.2 million $14.4 million $30,000/year $21.6 million $43.2 million ARTICLE Alaska. Alaska estimated its cost of FAS in 1999. The research team started with a 1995 estimate of $1.4 million lifetime cost per child with FAS from the Federal Health Professions Education Partnership Act of 1998; with adjustments for inflation, the 1999 lifetime cost was estimated at $1.5 million [Advisory Board of Alcoholism and Drug Abuse, 2001]. Based on prevalence rates of 1.4–2.8 per 1,000 live births, Alaska’s costs are estimated to range from $21 million to $42 million for one year’s birth cohort. These undiscounted lifetime cost estimates include medical costs, behavior management, and residential services. Residential services include special education, home care, speech therapy, and institutional care. Minnesota. The cost of FAS in Minnesota was estimated by taking the state’s proportion of 1985 national costs ($1.6 billion) as estimated by Rice et al.  and then adjusting it for inflation of health care costs to 1991. Population data were used to prorate costs. The overall annual cost of treating persons with full FAS in Minnesota was estimated to be $45 million in 1991 [Minnesota Department of Health, 1995]. This annual cost estimate for the state includes medical treatment and residential services for persons with mental retardation. It does not include social support and education, foster care, nonsurgical and nonintensive medical care, juvenile and criminal justice, lost productivity, and caregiver costs. AMERICAN JOURNAL OF MEDICAL GENETICS (SEMIN. MED. GENET.) North Dakota. In North Dakota, Klug and Burd  used a health claims database to estimate annual cost of health care for people from birth through 21 years of age with FAS in North Dakota. All private and public payers in North Dakota supply this information to the Health Department Claims Database. Information is not available on persons who are uninsured or on some persons served by the Indian Health Service, which covers about 3% of North Dakota residents. Children were classified as having FAS if they had an International Classification of Diseases, 9th edition (ICD-9) code of 760.71 [Practice Management Information Corporation, 1996]. Annual costs of health care by diagnosis were calculated by averaging the total yearly cost of all cases over the two-year period (Table III). Diagnosis-specific annual costs were calculated by subtracting the average yearly cost of children without FAS from the cost of children with FAS. Cumulative cost savings were defined as a series of costs summed over consecutive years. The mean annual cost of health care was $2,842 (n ¼ 45), which was $2,342 per capita more than the annual average cost of care for children in North Dakota (n ¼ 98,991) who did not have FAS ($500 per year). Since this cost includes only medical costs, it is much lower than other estimates that include special education and housing costs, which make up a large percentage of most other cost estimates. If one new case of FAS is prevented each year, $2,342 in health care costs would be saved the first year [Klug and Burd, 2003]. If the program continues for a second year, the cost savings is now $4,684 (for two years that one child does not have FAS) plus $2,342 for a new child. If the program continues for 10 years, the undiscounted cost saving accumulates to $128,810 (10 years of cost savings plus 9 years of cost savings plus 8 years of cost savings, etc.). After 20 years, the cost savings for preventing one case of FAS each year is nearly half a million dollars (Table III). Prevention programs can also reduce the occurrence of comorbid conditions associated with FAS. Table III also shows the undiscounted cost savings at 1, 10, and 20 years for five common comorbid conditions. Cost savings are greatest from preventing cases of developmental disability or seizures in children with FAS. FAS Prevalence and Cost Calculator Cost estimates for other states can be determined by using the FAS Prevalence and Cost Calculator available online (http://www.online-clinic.com; select ‘‘fetal alcohol syndrome’’). This calculator allows states and provinces to estimate their prevalence of FAS, FAE, and alcohol-related birth defects (ARBDs) and associated costs. This tool uses current published prevalence estimates for FAS, FAE, and comorbid conditions (ADHD, heart defects, etc.). TABLE III. Costs of Health Care for FAS and Related Comorbid Conditions for North Dakota Children From Birth Through Age 21 Condition Fetal alcohol syndrome Attention deficit/hyperactivity disorder Learning disabilities Developmental disabilities Oppositional defiant disorder Seizures Source: Klug and Burd, 2003. 47 Cumulative cost savings from preventing one case per year Average yearly cost Extra yearly cost versus child without the condition After 10 years After 20 years $2,842 $649 $1,302 $2,286 $1,377 $2,181 $2,342 $154 $806 $1,797 $883 $1,689 $128,810 $8,470 $44,330 $98,835 $48,565 $92,895 $491,820 $32,340 $169,260 $377,370 $185,430 $354,690 48 AMERICAN JOURNAL OF MEDICAL GENETICS (SEMIN. MED. GENET.) The costs utilized in the calculator are based on North Dakota studies of the cost of special education, juvenile justice services, health care costs, adult corrections, and other service delivery systems (e.g., foster care and residential care). What are the Public Policy Perspectives? In discussing the costs of FAS, Harwood and Napolitano  indicate that estimation of lifetime costs is central to the evaluation of the benefits to society of prevention programs. Bloss  notes that annual costs are helpful to describe the extent of the problem. However, lifetime costs are more useful from a public policy perspective. Further, Harwood and Napolitano, as well as Bloss, note that it is important to calculate the present discounted value (PDV) of the stream of lifetime costs in order to have an appropriate dollar figure when making public policy decisions about whether to expand prevention services. The PDV can be viewed as the amount of funds that need to be deposited today at a given interest rate (exactly equal to the discount rate) so that principal and interest will cover the expected stream of costs over time. A higher discount rate produces a lower PDV and vice versa. Thus, a PDV calculation converts estimated lifetime cost into a present dollar amount needed at the time of the child’s birth. Using the 1980 estimated lifetime cost of $596,000 per child with FAS and applying a discount rate of 6%, Harwood and Napolitano  determined that the PDV lifetime cost of FAS is $163,000. Bloss  noted that if the cost to prevent a birth with FAS was less than the PDV lifetime cost for that birth, then expanding prevention services yielded a net economic gain. However, the most widely used and recommended discount rate is now 3% [Gold et al., 1996]. Based on a 3% discount rate and the 2002 inflation-adjusted lifetime cost of $2 million, the PDV lifetime cost is $932,000. Both of these calculations demonstrate that programs to prevent births of children with FAS can be justified since the cost savings from prevention are much higher than originally estimated. Based on a 3% discount rate and the 2002 inflation-adjusted lifetime cost of $2 million, the PDV lifetime cost is $932,000. Both of these calculations demonstrate that programs to prevent births of children with FAS can be justified since the cost savings from prevention are much higher than originally estimated. What are the Gaps in Knowledge About the Costs of FAS? Much of the work to estimate the costs of FAS recognizes that full FAS is simply the one disorder in a group of conditions caused by prenatal alcohol exposure. It represents the tip of the iceberg. Other babies born with other FASD conditions make up a much larger group, accounting for as many as five times the number of FAS cases [Astley, 2002]. A recent review and synthesis of the FAS literature by May and Gossage  estimated that births with FAS may account for 0.5–2 cases per 1,000 live births, while all alcohol-affected births are estimated to be 10 cases per 1,000 [Streissguth et al., 1991]. Sokol et al.  estimated that the number of babies born with FAE could be 15 times the number of babies born with FAS. However, costs for these FAE individuals have not been included in the cost estimates reviewed, in part because there have been little firm data on prevalence of FAE or its impacts/disabilities, much less the required services and costs. Thus, a major need is to determine the costs associated with this larger group of alcohol-affected individuals. A number of specific cost categories have not been included in annual and ARTICLE lifetime costs in the past due to the absence of good data; more recent research appears to provide a better foundation for developing such estimates now. Inclusion of these costs would provide a more accurate picture. Examples include better information on the costs of special education and inclusion of costs for juvenile and criminal justice, substance abuse treatment, mental health care, and vocational services. For example, individuals with FAS often become involved with the juvenile and criminal justice systems. A recent study found that 60% of subjects with FAS or FAE had been in trouble with the authorities, charged with a crime, or convicted of a crime [Streissguth et al., 1996]. However, most cost estimates do not include law enforcement costs. National data indicate that the criminal justice system spent $146 billion in 1999 [U.S. Department of Justice, 1999]: $65 billion for police/law enforcement, almost $50 billion for corrections, and $31 billion for judicial services. Information on the costs of the juvenile and criminal justice system of police, court, juvenile detention, prison, and parole/probation services for individuals with FAS/FAE are not available. Costs cannot be estimated because there are no widely used screening and diagnostic tools to identify the number of affected persons within these systems. Yet the costs associated with law enforcement are significant: it can cost almost $30,000 per year to house an inmate in federal prison [Hawk, 1995]; some state prisons spend even more [Keaton, 2001]. Many individuals with FAS also have substance abuse problems of their own. Individuals with FAS and FAE need and use special education and vocational support services. Thus, future studies are expected to show higher costs when these cost categories are included. Finally, the annual and lifetime cost estimates are based on national statistics about the incidence and prevalence of FAS. Costs may be more variable in some racial and ethnic groups, with wide variation in prevalence. The difficulty in estimating the cost of services for individuals with FAS is further compounded by two situations. ARTICLE First, in many communities, these individuals do not get appropriate services and thus cycle through service systems and incur more costs than necessary in service systems, including criminal justice, mental health, and physical health systems. In other cases, they do not receive needed services, such as special education and vocational services. Thus, they do not incur costs of service provision, but constitute a major burden on the justice and social welfare systems. In either case, calculating true costs becomes difficult. FAS appears to be a predisposing, if not causal, agent for mental illness [Steinhausen et al., 1994; Spohr and Steinhausen, 1996; Stratton et al., 1996; Burd et al., 1996a, 2003; Abel, 1998]. Caregiver burden/costs are also quite large. In a North Dakota study of children with severe emotional disorders, some of whom had FAS, Burd et al.  reported that the unreimbursed cost for families was more than $17,000 per year per child. This is a largely unstudied cost, but the magnitude describes the severe adverse effects on families. In future studies of costs of FAS, this important cost variable should be included. DISCUSSION By most measures, the costs of FAS to society and for each alcohol-affected individual are very high. The most recent estimate from NIAAA is that FAS cost the United States more than $4 billion in 1998. The lifetime cost for each child with FAS is $2 million. The most recent estimate from NIAAA is that FAS cost the United States more than $4 billion in 1998. The lifetime cost for each child with FAS is $2 million. The extremely high costs justify major prevention efforts that have demon- AMERICAN JOURNAL OF MEDICAL GENETICS (SEMIN. MED. GENET.) strated their effectiveness. When the costs associated with individuals with FAE are included, the wisdom of investing in prevention efforts will be even clearer. Future studies of the cost of FAS and related conditions should: 1. Determine the number of children born with FAE as a ratio to those born with FAS so that the magnitude of the problem can be better understood. Develop better cost estimates of both FAS and FAE so the entire scope of the problem can be better understood and described. 2. Update or identify the cost of each service system element used by individuals with FAS and other alcoholaffected conditions and ensure that costs are representative of the United States. Improve the FAS cost estimates and develop the first cost estimates of FAE. Address other issues, such as epidemiology, phenotype definition, health economics, necessary services for individuals who are affected, whether mortality rates are different for persons with FAS, and unreimbursed costs of parents or other caregivers. More details on the published cost studies referenced in this article and numerous unpublished cost estimates can be found at the website of the FASD Center for Excellence (http://fascenter. samhsa.gov). REFERENCES Abel EL. 1984. Fetal alcohol syndrome and fetal alcohol effects. New York: Plenum Press. Abel EL. 1998. Fetal alcohol abuse syndrome. New York: Plenum Press. Abel EL, Sokol RJ. 1986. Fetal alcohol syndrome is now leading cause of mental retardation. Lancet 2:1222. Abel EL, Sokol RJ. 1987. Incidence of fetal alcohol syndrome and economic impact of FAS-related anomalies. Drug Alcohol Depend 19:51–70. Abel EL, Sokol RJ. 1991a. A revised conservative estimate of the incidence of FAS and its economic impact. Alcohol Clin Exp Res 15:514–524. Abel EL, Sokol RJ. 1991b. A revised estimate of the economic impact of fetal alcohol syndrome. Recent Dev Alcohol 9:117–125. 49 Advisory Board of Alcoholism and Drug Abuse. 2001. Alaska Department of Health and Social Services. Economic costs of alcohol and other drug abuse in Alaska, phase two. Juneau: McDowell Group. Astley S. 2002. Seventh annual report summary of progress from July 1, 1995 through June 30, 2002. Seattle: Washington State FAS Diagnostic and Prevention Network, University of Washington. Bloss G. 1994. The economic cost of FAS. Alcohol Health Res World 18:53–54. Burd L, Martsolf JT. 1989. Fetal alcohol syndrome: diagnosis and syndromal variability. Physiol Behav 46:39–43. Burd L, Moffatt ME. 1994. Epidemiology of fetal alcohol syndrome in American Indians, Alaskan Natives, and Canadian Aboriginal peoples: a review of the literature. Public Health Rep 109:688–693. Burd L, Wentz T. 1997. Fetal alcohol syndrome: the North Dakota experience. North Dakota J Hum Serv 1:14–19. Burd L, Martsolf JT, Kerbeshian J, Mohr T, Mohr P, Ebertowski M. 1996a. Psychiatric diagnosis and management of fetal alcohol syndrome. Clin Adv Treat Psychiatr Disord 10:1–8. Burd L, Martsolf JT, Klug MG. 1996b. Children with fetal alcohol syndrome in North Dakota: a case control study utilizing birth certificate data. Addict Biol 1:181– 189. Burd L, Klug MG, Hoesel J, Moum K. 2001. Development of a comprehensive cost capture for children with severe emotional disorders. In: Newman CC, Liberton CJ, Kutash K, Friedman RM, editors. 13th annual research conference proceedings: a system of care for children’s mental health: expanding the research base. Tampa, FL: The Research and Training Center for Children’s Mental Health, University of South Florida. p 277–280. Burd L, Klug M, Martsolf J, Kerbeshian J. 2003. Fetal alcohol syndrome: neuropsychiatric phenomics. Neurotoxicol Teratol 25:697– 705. Centers for Disease Control. 2002. Alcohol use among women of childbearing age—United States, 1991–1999. MMWR Morb Mortal Wkly Rep 51:273–276. Day NL, Leech SL, Richardson GA, Cornelius MD, Robles N, Larkby C. 2002. Prenatal alcohol exposure predicts continued deficits in offspring size at 14 years of age. Alcohol Clin Exp Res 26:1584–1591. Famy C, Streissguth AP, Unis AS. 1998. Mental illness in adults with fetal alcohol syndrome or fetal alcohol effects. Am J Psychiatry 155:552–554. Gold MR, Siegel JE, Russell LB, Weinstein M. 1996. Cost-effectiveness in health and medicine. Oxford: Oxford University Press. p 230–233. Harwood H. 2000. Updating estimates of the economic costs of alcohol abuse in the United States: estimates, update methods and data. Report prepared by the Lewin Group. Bethesda, MD: National Institute on Alcohol Abuse and Alcoholism. Harwood HJ, Napolitano DM. 1985. Economic implications of the fetal alcohol syndrome. Alcohol Health Res World 10:38–75. 50 AMERICAN JOURNAL OF MEDICAL GENETICS (SEMIN. MED. GENET.) Harwood HJ, Napolitano DM, Kristiansen PL. 1984. Economic costs to society of alcohol and drug abuse and mental illness: 1980. Rockville, MD: Alcohol Drug Abuse and Mental Health Administration. Harwood H, Fountain D, Livermore G. 1998. The economic costs of alcohol and drug abuse in the United States, 1992. Washington, DC: National Institute on Drug Abuse and National Institute on Alcohol Abuse and Alcoholism. Hawk KM. 1995. Extrapolated from 1995 figure in a letter from Kathleen M. Hawk, Director, U.S. Department of Justice, Federal Bureau of Prisons, to the Honorable Myron H. Bright, based on a yearly inflation rate of 4%. Keaton C. 2001. Fiscal affairs and government operations committee report on adult corrections systems. Atlanta: Southern Legislative Conference, Council of State Governments. Klug MG, Burd L. 2003. Fetal alcohol syndrome prevention: annual and cumulative cost savings. Neurotoxicol Teratol 25:763–765. Little BB, Snell LM, Rosenfeld CR, Gilstrap LC III, Gant NF. 1990. Failure to recognize fetal alcohol syndrome in newborn infants. Am J Dis Child 144:1142–1146. May PA, Gossage JP. 2001. Estimating the prevalence of fetal alcohol syndrome. A summary. Alcohol Res Health 25:159–167. Minnesota Department of Health. 1995. Alcohol use in Minnesota: extent and cost. Minneapolis: Minnesota Department of Health. Morse B, Gehshan S, Hutchins E. 1997. Screening for substance abuse during pregnancy: improving care, improving health. Arlington, VA: National Center for Education in Maternal and Child Health. Olney JW, Wozniak DF, Farber NB, JevtovicTodorovic V, Bittigau P, Ikonomidou C. 2002. The enigma of fetal alcohol neurotoxicity. Ann Med 34:109–119. Practice Management Information Corporation. 1996. ICD-9-CM international classification of diseases, 9th revision, clinical modification, fifth edition, color coded, 1997. Los Angeles, CA: PMIC, Practice Management Information Corp. Rice DP. 1993. The economic costs of alcohol abuse and alcohol dependence: 1990. Alcohol Health Res World 17:10–11. Rice DP, Kelman S, Miller LS, Dunmeyer S. 1990. The economic costs of alcohol and drug abuse and mental illness: 1985. DHHS publication (ADM) 90-1694. Rockville, MD: U.S. Department of Health and Human Services. Sampson PD, Streissguth AP, Bookstein FL, Little RE, Clarren SK, Dehaene P, Hanson JW, Graham JM Jr. 1997. Incidence of fetal alcohol syndrome and prevalence of alcohol-related neurodevelopmental disorder. Teratology 56:317–326. Sokol RJ, Miller SI, Reed G. 1980. Alcohol abuse during pregnancy: an epidemiologic study. Alcohol Clin Exp Res 4:135–145. Spohr HL, Steinhausen HC. 1996. Alcohol, pregnancy, and the developing child. ARTICLE Cambridge, England: Cambridge University Press. Stanage WF, Gregg JB, Massa LJ. 1983. Fetal alcohol syndrome—intrauterine child abuse. S D J Med 36:35. Steinhausen HC, Willms J, Spohr HL. 1994. Correlates of psychopathology and intelligence in children with fetal alcohol syndrome. J Child Psychol Psychiatry 35:323– 331. Stratton KR, Howe CJ, Battaglia FC, Institute of Medicine. 1996. Fetal alcohol syndromediagnosis, epidemiology, prevention, and treatment. Washington, DC: National Academy Press. Streissguth AP, Aase JM, Clarren SK, Randels SP, LaDue RA, Smith DF. 1991. Fetal alcohol syndrome in adolescents and adults. JAMA 265:1961–1967. Streissguth AP, Barr HM, Koga J, Bookstein FL. 1996. Understanding the occurrence of secondary disabilities in clients with FAS and FAE (96-0). Seattle: University of Washington Fetal Alcohol and Drug Unit. U.S. Department of Justice, Bureau of Justice Statistics. 1999. Justice expenditure and employment extracts, NCJ 190642 (online). Washington, DC: U.S. Department of Justice. Weeks M. 1989. Economic impact of fetal alcohol syndrome, IR 89-100015. Memorandum to Senator Johne Binkeley, February 17. Juneau: Senate Advisory Council, Alaska State Legislature.