Intergovernmental committee on Drugs working party on Fetal Alcohol Spectrum Disorders
Fetal Alcohol Spectrum Disorders in Australia: An Update
10.1 Costing FASD: Approaches taken and international evidence
10.1.1 Cost conceptsThe costs of morbidities can be defined and estimated in various ways. Costing FASD is no exception. Four main issues were identified when considering costing FASD.
1. Individual and aggregate community costs.The costs of FASD may be estimated for individuals in each diagnostic group of FASD (FAS, ARBD, ARND) or for all individuals with FASD in the community. Community costs are essentially an aggregation of the costs incurred in individual cases. Accordingly, we focus on how to estimate government and non-government costs of FASD for individual cases for different components of FASD. These costs can be aggregated according to the number of FASD cases in each group.
2. Annual or lifetime costs for individuals with FASD.The costs of FASD may be estimated on a per annum basis or a lifetime basis (from age 0 to death). When lifetime costs are estimated, these may be expressed as a simple total or on a present value basis by discounting all future costs back to a present date.
3. Aggregate community costs may be estimated on an incremental flow or as a prevalence cost.We define an incremental cost as the cost associated with incident FASD cases in a given period. However, this would include any lifetime costs associated with these births. Thus if there are 1000 new FASD cases in a year, the cost of FASD in that year could be calculated as the cost of these 1,000 cases over their lifetime discounted back to that year.
On the other hand, the costs of FASD in any one year could be estimated (in principle at least) as the cost of all FASD cases alive in that year, regardless of when they were born. This could be viewed as an estimate of the prevalence cost FASD that does not take lifetime costs
For policy purposes the former of these estimates is more useful as policy makers want to know how policies may reduce FASD costs going forward. However, some studies of disease costing estimate prevalence costs as well as incremental lifetime costs for example, Applied Economics, 2005 for hepatitis C (Applied Economics 2005).
4. The elements of FASD costs.These may be considered as follows:
- (a) the direct costs to government of health care and accommodation for FASD cases
- (b) other costs to government including special education and employment support services, community services, income support, and justice services
- (c) costs borne by FASD individuals, including loss of productivity (income), reduced quality of life and reduced longevity.
10.1.2 International evidenceAs shown below and elsewhere (Applied Economics 2008), most studies of the costs of FASD focus on direct costs to government for the health care and accommodation for people with FASD, i.e. the elements discussed in 4 (a) above. A few studies have also included estimates of costs borne by FASD individuals (see 4(c) above), specifically productivity loss (Lupton et al. 2004). These authors were unable to identify any studies that estimated the costs associated with 4(b), that is, other costs to government such as support services; similarly, no studies were found to estimate the other costs borne by people with FASD (as per 4(c) above), such as of reductions in quality of life or reduced longevity. The reason for these exclusions is a lack of data. There are no systemic data on the social costs associated with FASD cases or on longevity.
To date, there have been no Australian studies of the costs of FASD for individuals or for the community. As reported in Applied Economics and summarised below, there have been some international studies of the costs of FASD (Applied Economics 2008). However, most of
these are limited in scope, both in terms of estimating the numbers of FASD cases and in terms of the types of costs included, and are also of limited comparability.
Several studies have been conducted into the cost of fetal alcohol exposure. These have included those conducted by Harwood and Napolitano and by Weeks, both of which considered the health system in the United States and the lifetime cost of FAS (Harwood and Napolitano 1985; Weeks 1989). Another study by Klug and Burd considered the lifetime health costs of FAS, while a further paper by Stade et al. included the patient costs per year of FAS/Fetal Alcohol Effects (Klug and Burd 2003; Stade et al. 2006). Between 1984 and 1991
Abel and Sokol undertook a series of studies on the cost impact of FAS to the United States health system in which they progressively edited and updated their data (Abel and Sokol 1987; Abel and Sokol 1991; Abel and Sokol 1991). A meta analysis of various cost studies
by Lupton and colleagues is summarised in Table 10.1, with an additional study by Stade 2006, (Study H) also included (Stade et al. 2006). These studies may be broadly classified as seeking to measure either annual costs (studies A to H) or lifetime costs (studies I and J).
To help compare the definition of cost in these studies, Table 10.1 adapts the basis of each of their original cost estimates so that (a) incidence in the case of studies A to F is standardised on two per 1,000 live births; and (b) residential care up to the age of 65 is included, but in
some cases only for care for mental retardation, and in study H up to the age of 21 (Stade et al. 2006). All studies exclude costs associated with the loss of productivity, apart from study H, which includes the productivity loss of the child and carer and study J which includes the
productivity loss of the child (Weeks 1989; Stade et al. 2006).
All costs in Table 10.1 are measured in constant 2002 $US, with the exception of study G which uses 2003 $US prices and study H which uses 2006 $CAD (Klug and Burd 2003; Stade et al. 2006)
These are as follows:
- total costs with a standardised incidence (two per 1,000) on a United States population of 301 million with the age up to which total costs are included varying from 21 to 65 years (the exception are study H which relates to Canada) (Stade et al. 2006)
- case costs (total costs divided by the number of cases)
- lifetime costs are calculated with assumptions based on study I regarding the incidence of costs up to the age of 65, as most studies only included costs up to the age of 21 (Harwood and Napolitano 1985).
As further explored by Applied Economics, it is clear that the variation between the population basis of the studies, what the studies are seeking to measure and the assumptions concerning measurement are too great to provide meaningful comparisons (Applied Economics 2008). There is a need for additional work on the costs associated with FASD.
In addition to the studies reported in Table 10.1. There have been four recently published FASD costing studies. Stade and colleagues revised the estimated direct and indirect costs associated with FASD using a cross sectional study of 250 participants. Costs were expressed in 2007 $CAD. The adjusted annual cost per child with FASD was $21642. The study was not designed to estimate the costs at the population level, but given the need to illustrate the cost to the nation, population estimates were calculated using an estimate prevalence of 1 in 100 people. The annual cost of FASD to Canada for people aged 0-53 was estimated to be $5.3 billion 2007 $CAD (Stade et al. 2009).
In another Canadian study the short and long term costs of FASD were estimated for the period 2002-2005 in Alberta using an incidence rate of 3-9 per 1000 live births. The number of live births for each year 2002-2005 inclusive were used in the estimations (ranging from 38 313 in 2002 to 41 355 in 2005). The long term costs included the projected annual costs incurred by a cohort of children born with FASD. The short term costs included the amount of money incurred by people presently living with FASD. The total annual costs were estimated to be $48 to $143 million. The long term costs rose from $130 to $400 million each year. The authors acknowledge this is likely to be significant underestimate as it does not include individuals living in institutions (e.g. disability and justice) or homeless people (Thanh and Jonsson 2009).
A recently published paper from the US reported higher medical expenditure costs than have been previously reported for children with FAS (Amendah et al. 2011). The study examined annual medical expenditure for children with FAS, including those with and without intellectual disability, using paediatric Medicaid data from unidentified states. Children with FAS incurred annual expenditures that were nine times greater than for children without FAS ($16 782 compared to $1859 in 2005 $US). Intellectual disability was more common among
children with FAS than in children without FAS. The annual mean expenditure was 2.8 times greater among children with FAS and intellectual disability than in children with FAS who did not have an intellectual disability.
Another recent study examining health service use by children in the Western Cape of South Africa found that the median number of annual visits to public health care facilities by children with FAS was 8, much greater than the 2.57 annual visits made by all children across all public health care facilities. The average annual cost to society of providing health care to a child with FAS/PFAS was $1039.38 ($US 2009). At the population level the estimated annual health cost is $70 960 053.68 which equates to 5% of the Western Cape Department of Health’s budget for 2010/11) (Crede et al. 2011).