Best Practices : Fetal Alcohol Syndrome/Fetal Alcohol Effects and the Effects of Other Substance Use During Pregnancy
6. Identification of FAS and Related Effects
For the purpose of this report, “Identification” encompasses screening, referral and case management activities by various professionals (e.g., public health nurses, teachers, social workers) as well as expert diagnosis of individuals (newborns, children, adolescents, adults) affected by a mother’s use of alcohol during pregnancy. Effects in children caused by prenatal use of cannabis, opiates, cocaine, or inhalants are much more limited and are most likely to be noted in newborns.
Estimates on the prevalence of FAS and related effects vary widely, depending on the diagnostic criteria used, the method of case ascertainment, and the population surveyed. Obviously, use of more rigorous or narrower criteria, such as counting only cases of FAS and not including partial syndromes, will yield lower prevalence rates. As would be expected, population-based studies yield lower estimates than do studies that have investigated specific communities or groups, as these communities and groups typically choose to participate because of a concern with high rates of substance abuse during pregnancy in the community. Both of these kinds of studies have been conducted in Canada, predominantly with Aboriginal populations.
Prevalence of FAS and FAE is poorly understood in Canada.
The population-based prevalence model uses the pre-existing data of a population to estimate the incidence of FAS. Two such studies have been published in Canada. Asante and Nelms-Matzke (1985) surveyed children in northern British Columbia and the Yukon. The authors estimated the rate of FAS and possible alcohol-related effects at 46 per 1,000 Native Canadian children in the Yukon and 25 per 1,000 in British Columbia. They further estimated that 51% to 66% of all children in special education with learning disorders in the study regions were exposed to alcohol in utero.
Habbick et al. (1996) used a database in the Department of Pediatrics, Royal University Hospital in Saskatoon, to estimate the prevalence in the province of Saskatchewan. This study yielded prevalence rates that were close to worldwide estimates of FAS (i.e., 0.5 cases per 1,000 live births). Habbick et al. (1996) only included cases of full FAS, using the US Institute of Medicine (IOM) criteria for FAS with known exposure history. They acknowledged that the prevalence of the full spectrum of alcohol-related conditions is likely to be 3-5 times higher than that of FAS alone. The problem with this model of ascertainment is that it relies on the assumption that all affected cases will be diagnosed and counted. Thus, this model is prone to under estimating the prevalence of FAS, as undiagnosed cases cannot be counted.
“Captive” community studies in which a community invites researchers into the community to investigate the prevalence of FAS and related effects, have documented very high rates of FAS (up to 1 in 4 pregnancies) (Robinson et al., 1987; Williams et al., 1999; Godel et al., 2000). In this type of study, the researchers visit a small community and examine every child within an age range, making blind diagnoses of affected children. For example, Robinson et al. (1987) closely examined children and mothers in an isolated community in northern British Columbia and found that 14 of 116 children below the age of 18 had FAS and 8 had FAE, representing an unexpectedly high prevalence of 190 per 1,000 for FAS and related effects.
While the results of these studies cannot be generalized to other Canadian communities, they can contribute greatly to a local understanding of need and to appropriate steps being taken. This requires careful thought on the part of the community and investigators, as pointed out by Chudley and Jones (2000). Reports of high rates of FAS and related effects will likely be accompanied by various emotions on the part of community members that will need to be addressed before other steps towards improving community health can be taken. Supporting a broader view of the role of research, Gilchrist (1997) calls for an end to research that tends to objectify Aboriginal communities and a move towards models that develop knowledge useful to self determination. In a similar vein, Kowalsky et al. (1996) have suggested guidelines for researchers working with Aboriginal communities, based on FAS-related studies conducted by the authors in northern Canadian communities.
There is no reliable Canadian prevalence figure available, for FAS, for the spectrum of alcohol-related conditions, or for alcohol and other drug-related conditions together. The only population-based estimate is from Habbick et al. (1996) of 0.5/1,000, but these authors acknowledged this was likely to be an underestimate and unreliable. The studies by Robinson et al. (1987); and Williams et al. (1999) indicate that there are specific Canadian communities where the prevalence is much higher. Like many other complex social problems, the consequences of prenatal exposure to alcohol and other substances are not evenly distributed throughout the country.
In the US, the Centers for Disease Control and Prevention have used newborn birth defect registries to estimate the incidence and prevalence of FAS and related effects (Cordero et al., 1994). While these studies have been useful in documenting the differences in the rates of FAS and related effects across racial groups in the US, they are known to underestimate the rates of the condition, as most cases are not diagnosed at birth (Little et al., 1990).
Abel (1998b) points out that prospective/active surveillance systems yield the most accurate estimates of the prevalence of all birth defects, including FAS. He reviewed 29 studies from around the world that used this methodology to document the incidence of FAS. From these studies he extrapolated a worldwide incidence figure of 0.97/1,000 of FAS, which ranged from a low of 0/1,000 from several countries to a high of 3.9/1,000 in Detroit. This figure was only for cases of FAS and did not include cases that did not meet the criteria for the full syndrome. This estimate is comparable to the figure of 0.5/1,000 generated by Habbick et al. (1996), adding credence to their figure as a minimal population-based estimate for a Canadian province.
A number of limitations to this approach pertain to difficulties in diagnosis, and include under-diagnosis, a lack of recording of the diagnosis in medical records, the possibility of stereotyping populations, inclusion of a limited population subset (for example, only infants), and a restricted case definition (Cordero et al., 1994; Little et al., 1990). To illustrate the challenge, some affected newborns may have only subtle facial abnormalities, unapparent central nervous system deficits and normal birth weight, and will be difficult to identify using a restricted case definition (Cordero et al., 1994).
FAS was first reported by Lemoine et al. (1968) in France and then independently identified by Jones and Smith (1973) in Seattle. Jones and Smith coined the term FAS to emphasize the preventable nature of this birth defect. Although the term has enjoyed widespread acceptance, it is not without its critics. Abel (1998c) has recently proposed the term fetal alcohol abuse syndrome to replace FAS in order to emphasize the fact that alcohol abuse (i.e., either heavy per-occasion drinking, or frequent drinking) rather than alcohol consumption causes FAS.
In 1996, the US Institute of Medicine (IOM) (Stratton et al. [eds.], 1996) proposed a revision of the diagnostic criteria for FAS and other alcohol-related effects to reflect current knowledge of the field. The committee recommended five major diagnostic categories:
- Fetal Alcohol Syndrome (FAS) with a confirmed history of maternal alcohol exposure requiring evidence of facial dysmorphology, growth retardation, and central nervous system (CNS) dysfunction;
- FAS without confirmed maternal exposure requiring evidence of facial dysmorphology, growth retardation, and CNS dysfunction;
- Partial FAS (pFAS) requiring a confirmed history of prenatal alcohol exposure, facial dysmorphology, and either growth retardation or CNS abnormalities;
- Alcohol-related birth defects (ARBD) to denote the presence of congenital anomalies (e.g., cardiac, skeletal, renal, ocular, auditory) known to be associated with a history of prenatal alcohol exposure;
- Alcohol-related neurodevelopmental disorder (ARND) requiring a confirmed history of prenatal alcohol exposure and evidence of CNS abnormalities.
The changes were proposed to differentiate between those cases where maternal exposure could be confirmed and those for whom the exposure history was unclear. For children in the foster care system, foreign adopted children (Aronson, 2000), and for fostered or adopted adults, information regarding prenatal exposure is often uncertain or unavailable. By providing a category of FAS without maternal history, the diagnostic dilemma that occurred when the history was not available was addressed (Stratton et al. [eds.], 1996). This classification scheme is intended to stimulate further research, both animal and human, into the teratogenic effects of alcohol, while adding clarity to the field. These diagnostic criteria are intended to be applied across the life-span. There are no published guidelines that specify changes to the diagnostic criteria by age. As yet, there have been no independent evaluations of these criteria.
One important change in the Institute of Medicine’s 1996 nomenclature is the elimination of the term fetal alcohol effects (FAE). This term has been widely criticized (Aase, 1994; Abel, 1998b) for lacking specificity and sensitivity. The IOM proposed the term “partial FAS” to further delineate the diagnosis of those with confirmed alcohol exposure history and some, but not all, of the anomalies. One of the expectations was that this delineation of a diagnosable condition would lead to eligibility to receive services, such as special education and welfare in the US. Abel (1998a) has been critical of this approach, fearing that it will lead to over-diagnosis of any child whose mother drank at any level during pregnancy. Abel (1998a) also warns that over-diagnosis of partial FAS may inhibit diagnosis of other potentially treatable syndromes. What remains a difficult and controversial issue is the identification of partial FAS without confirmation of maternal drinking.
Astley and Clarren (2000) also criticize current clinical guidelines, suggesting that terms such as FAE, ARBD and ARND inappropriately imply alcohol exposure to be the sole cause of anomalies in a given individual. With the likely exception of facial phenotype, the various physical and cognitive-behavioural anomalies associated with alcohol exposure are not caused only by alcohol exposure. They call for new diagnostic terms that more finely document the variability in exposure and outcomes, and do not imply alcohol as the sole causal agent.
Attempting to address this need for more clinical precision to fetal alcohol-related diagnoses, particularly in cases where the typical anomalies are not extreme, Astley and Clarren (1999) have proposed a more elaborate system that introduces objective, quantitative measures to assist in diagnosis. The system uses the numbers 1 to 4 to convey the magnitude of expression in each of the four key diagnostic domains: growth; dysmorphology; central nervous system dysfunction; and alcohol exposure; and yielding 256 possible 4-digit diagnostic codes and corresponding clinical names.
The authors claim that the approach better characterizes the full spectrum of disabilities of alcohol-exposed individuals, and documents the presence of alcohol exposure without judging its causal role. This system can be added to the IOM system to yield a more precise analysis. Data on the utility of the approach for different age and population groups are not yet available, so it remains to be seen whether this more complex system will be effective.
Many authors (e.g., Streissguth, 1997; Abel, 1998b; Gardner, 1997) have emphasized the need for early identification and diagnosis in order to prevent future affected pregnancies and to refer the infant to appropriate services quickly. However, newborn diagnosis is difficult because the specific facial features characteristic of the syndrome may not be apparent until the second year and beyond, and behavioural deficits are more difficult to measure at this point (Gardner, 1997; Aase, 1994). Aase (1994) also points out research that suggests only 70% to 75% of children eventually diagnosed with FAS will experience prenatal growth retardation.
Nevertheless, identification in infancy is possible through attention to growth measures and to common manifestations (i.e., feeding problems, irritability, and unpredictable patterns of sleeping and eating) that may be due to withdrawal from alcohol and make the baby hard to care for (Aase, 1994; Stratton et al. [eds.], 1996). A susceptibility to other common birth defects, for example, congenital heart disease, cleft lip and palate, anomalies of the urinary tract and genitals, spina bifida and ocular anomalies, may support (rather than define) a diagnosis (Aase, 1994; Stromland, 1990). Breast-feeding by a currently drinking mother can produce heightened activity in the infant that would need to be distinguished from the effects of prenatal use (Mennella and Gerrish, 1998). Identification of FAS or related effects in an infant is greatly facilitated by knowledge of the mother’s alcohol use history, which is best gained through routine, prenatal screening of the mother for alcohol use (Hankin and Sokol, 1995; Russell et al., 1996; Cherpitel, 1997). Meconium (the stool passed by a newborn) testing may be helpful in supporting prenatal screening. While cocaine and other drugs of abuse can be measured in neonatal hair, alcohol cannot. A recently developed meconium test detects by-products of alcohol — the fatty acid ethyl esters — and they reflect maternal history of drinking in late pregnancy (G. Koren, pers. com., 2000).
The facial features associated with FAS are thought to become more recognizable in the pre- and early-school period. In addition, alcohol-exposed children present a number of behaviour patterns that can support identification. These children are affectionate, but very active, flighty, distractible, having poor memory and short attention span – traits that are comparable to children with attention-deficit hyperactivity disorder (ADHD) (Nanson and Hiscock, 1990; Streissguth, 1997; Coles et al., 1997). Important manifestations of central nervous system dysfunction are mental retardation and/or learning disabilities (Conry, 1990; Kerns et al., 1997; Mattson and Riley, 1998).
Children with FAS appear to have deficits in language and speech; however, findings in this area have been mixed. While vocabulary may be adequate, higher-order skills such as comprehension and understanding of semantics and syntax are less developed (Stratton et al. [eds.], 1996; Streissguth et al., 1994). One reviewer termed this pattern “social but dysfunctional communicative interaction” (Abkarian, 1992).
As the child moves into adolescence, identification based on dysmorphology and growth is not straightforward, because many of the physical characteristics of FAS, such as the facial features, become less prominent and more difficult to recognize (Gardner, 1997). There is evidence of catch-up growth for some with earlier slenderness giving way to obesity, particularly in affected girls (Jacobson, 1998; Streissguth, et al., 1991). However, some authors (e.g., Habbick et al., 1998; Day et al., 1999) have presented data to suggest that the growth retardation persists over time.
While physical features evolve, cognitive and behavioural deficits generally endure with age (Mitchell et al., 1995). In fact, behavioural, emotional and social problems can become more pronounced, even when the home environment changes for the better (Jacobson, 1998). Cognitive-behavioural deficits manifest themselves in difficulties with abstract thinking, judgment and problem solving, which in turn give rise to learning, academic and social problems (Mattson et al., 1996; Stratton et al. [eds.], 1996; Aase, 1994). Steinhausen et al. (1993) in a longitudinal study with a large sample, found a number of psychological and emotional problems to persist from pre-school to late school stages among children with FAS, even in the absence of intellectual handicap. Streissguth et al. (1991) suggest that adaptive behaviour and social judgment are impaired to a greater extent than intellectual functioning. Deterioration in these areas may, to some extent, be due to “secondary disabilities” that are not specifically caused by the teratological effects of alcohol, but arise from difficult experiences and inappropriate care during childhood (Streissguth, 1997).
Identification in adults is particularly challenging. It is less clear how the physical markers of the syndrome change with age. If catch-up growth does occur in some individuals, then the establishment of a diagnosis will be problematic after the adolescent growth spurt has been completed. Similar issues exist with diagnosis by way of dysmorphology in adults. Streissguth et al. (1991) published sequential photographs of the same individual in infancy, childhood, and adulthood, which suggested that the facial characteristics of FAS are more subtle and difficult to detect in adults. Astley and Clarren (1999) include growth norms up to 18 years of age and psychometric tests that are applicable to adults.
Broadly understood and applied diagnostic criteria are basic to advancing FAS-related activity in this country. Yet, diagnosis is not straightforward for a number of societal as well as clinical reasons. Alcohol use by women, particularly pregnant women, remains stigmatized. The stigma and consequent bias among some medical professionals may result in a fear of labelling the mother and child, with diagnosis not even being considered in some populations, or in a hesitation to apply the label, believing that the various manifestations of FAS can be dealt with just as effectively without it (Cordero et al., 1994; Gardner, 1997). For the same reason, pregnant, alcohol-using women are unlikely to report their use accurately.
Clinically, many of the anomalies associated with prenatal alcohol exposure are not unique to FAS and related effects and can be confused with other disorders or conditions, particularly among children with partial FAS, and Alcohol-Related Effects (i.e., ARBD and ARND). Examples of conditions often easily confused with FAS include Aarskog syndrome, fragile-x syndrome, fetal hydantoin syndrome and Noonan syndrome (Astley and Clarren, 1999). Presentation of FAS anomalies varies widely among individuals, with growth retardation seen in most, but not all, diagnosed children, intelligence scores ranging from severely disabled to normal, and a range of behavioural difficulties that generally but do not always co-occur (Aase, 1994; Stratton et al. [eds.], 1996). Consequently, Aase acknowledged that diagnosing FAS in a specific patient is often difficult even for an experienced clinician because the diagnosis depends on recognition of a consistent pattern of minor, often subtle physical anomalies, generalized but disproportionate growth retardation, and non-specific developmental and behavioural problems – some of which change with time and in severity among individuals.
Some contend that the developmental and behavioural patterns are, in fact, quite specific and await more sophisticated testing. Jacobson (1998) suggests that an initial deficit (or deficits) in pre-school children follow a distinct pattern rather than diffuse or generalized impairments.
Supporting this, Janzen et al. (1995), in a quasi-experimental study with a small sample, found that FAS children showed deficits in verbal, performance, language, behaviour and motor skills, but failed to show deficits in visual perception, quantitative or memory skills. However, findings have been at times conflicting on these matters and further investigation is required (Stratton et al. [eds.], 1996).
It also needs to be noted that a diagnosis of FAS does not preclude other problems that may or may not be related to prenatal alcohol use. For example, Nanson (1992) studied children who had both autism and FAS diagnoses and concluded that it was important to account for both in order to intervene appropriately.
Complicating accurate diagnosis still further is the fact that many children who may have experienced prenatal exposure to alcohol may also be victims of abuse and neglect and suffer behaviour problems for these reasons, making it difficult to determine whether the behaviours are due to the child’s living environment, the result of the prenatal alcohol exposure, or some combination of the two (Stratton et al. [eds.], 1996).
There may be issues in identification with respect to particular Canadian populations. The pattern of facial features proposed by the US Institute of Medicine as markers for FAS, namely short palprebral fissures (short eye slits), flat upper lip, flattened philtrum (the groove between the nose and upper lip), and flat midface, may overlap with racial features, particularly in Aboriginal populations (Aase, 1994). The norms for the measurement of markers such as palprebral fissure length were developed in the US, and may not adequately distinguish facial features found among some in the Canadian population. From studying the effects of prenatal exposure to alcohol among children in a northern Canadian community, Godel et al. (2000) concluded that diagnostic standards need to be established for each ethnic group because of the variance of facial proportions between races. In the meantime, diagnosticians need to guard against over-diagnosis by being aware of local variants in facial features that can occur, and avoid using these alone to support a diagnosis of FAS.
Similarly, the standard growth charts used to assess growth retardation have been developed in the United States. Data are lacking on the applicability of these charts to Canadians. For example, Chudley et al. (1999) reported that normal children in an isolated Manitoba community were typically taller and heavier than standard growth curves would predict. Thus, a child from this community with growth retardation due to prenatal alcohol exposure may be small relative to the other children in the community, but may not be seen as growth retarded on a standard growth chart. There are other anthropological data showing that certain groups of Aboriginal children in BC have a typically larger head circumference compared to North American averages (J. Conry, pers. com. 1999).
Finally, while some of the CNS criteria are objective, such as seizures and structural brain malformations, others may be more subjective, such as the diagnosis of learning disabilities. Psychological tests which have been developed for use with mainstream groups may not be appropriate for children who do not speak mainstream languages or who have not been educated in mainstream cultures. In carrying out an assessment, it is important to ensure that the psychometric tests being used have been evaluated for use with the population group of the child or adult being assessed, and that the examiner is familiar with the cultural characteristics of that community, such as response styles and other behaviours or circumstances that may influence test results.
The health care system is in a strong position to provide early identification of alcohol-exposed children during infancy and early childhood. However, in most cases, health care systems do not identify all or most FAS or related cases (Little et al., 1990). Consequently, while this does not generally occur either, others with whom the mother and child have contact (such as child care workers, social workers, addiction workers, speech and language pathologists, educators and correctional workers) with an understanding of FAS and related effects, and knowledge of the availability of diagnostic services, have an important role to play in screening and referring for diagnosis, and in supporting the diagnosis (Hess and Kenner, 1998; Niccols, 1994; Conry et al., 1997; Jenkins and Culbertson, 1996). Screening for FAS and related effects is particularly called for in some sectors such as the criminal justice system, where there are indications that affected youth and adults are over-represented (Conry et al., 1997).
Currently, there are no fully validated screening tools in common usage to support screening for FAS and related effects. Streissguth et al. (1998) reported on a series of tests of a fetal alcohol behaviour scale to be completed by a person who knows the affected person well. The instrument demonstrated adequate test-retest reliability and produced results that were independent of age, sex, race, IQ and alcohol-related diagnosis. Because it can be administered in a brief period of time, it lends itself to a screening function; however, further study is needed to determine its usefulness with various populations. While this behaviour scale distinguishes the problems of those with FAS from the general population, it has not been shown to distinguish the problems found in FAS with those in other neurological disorders.
Non-specialist physicians and nurses can be of assistance in screening and referral when they obtain information on prenatal exposure history, sequential growth measurements, and help to provide photographs of an affected individual over time (S. Clarren, pers. com. 1999). To support non-specialist medical identification, Astley and Clarren (1995) tested a potentially versatile screening tool based on facial phenotype on the rationale that it is, unlike CNS dysfunction and growth deficiency, specific to FAS. The facial phenotype is characterized by a cluster of minor facial anomalies that include small palpebral fissures, a flat midface, a smooth philtrum, and a thin upper lip. When tested with two racially mixed samples of children, ranging in age from 2 to 10 years, it was found to have high sensitivity and specificity. This screening tool requires further testing of inter-rater reliability and replication in other settings.
Parents’ groups have published screening measures (DeVries [ed.], 1999; DeVries et al., 1998), in many cases over the Internet, but these have not been subject to adequate evaluation. Use of these screening instruments can result in confusion because they cite common behaviour problems in children (e.g., lying) that are not related to the diagnosis or occur in normal development and/or in other conditions as well.
Actual diagnosis of FAS and related effects requires a multidisciplinary focus
(Stratton et al. [eds.], 1996). Various Canadian and US diagnostic clinics described in the literature have employed multidisciplinary teams, using a physician – typically a pediatrician or geneticist to assess growth and dysmorphology – and a psychologist to assess the behaviours that would support the finding of central nervous system anomalies, such as mental retardation, learning disabilities or adjustment problems (Ridd, 1999; Adrian and Fisher, 1997; Astley and Clarren, 1995, 1999; Li and Pearson, 1996). In some cases, more prominent central nervous system anomalies such as microcephaly or seizures will permit all of the diagnostic categories to be assessed within a medical examination.
Active case finding and screening programs using multidisciplinary clinics serve both to minimize the harmful effects associated with FAS and to raise community awareness. In Saskatchewan, a model of traveling clinics with a specialist team that visits remote communities to assist with diagnosis and with the development of local resources has been used. The FAS team consists of a pediatrician, psychologist, social worker, physiotherapist, occupational therapist, and speech therapist. Local professionals, such as nurses and dieticians, participate as much as possible.
These traveling clinics have been independently evaluated (Adrian and Fisher, 1997). High levels of client and community satisfaction with the model were reported, although some parents reported enjoying the opportunity to leave isolated northern communities to attend appointments in a larger centre, particularly when they have extended families to visit there. Concerns were expressed with the limited numbers of families who can receive services at a clinic and the limited length of the appointments. A similar outreach service delivery model is offered in BC through the Children and Women’s Hospital. Patients are seen by a developmental pediatrician, psychologist, geneticist and psychiatrist (C. Loock, pers. com., 1999).
Astley and Clarren (1995, 1999) have described a clinic in Washington DC, emphasizing the need to provide families with information in four broad areas: medical; psychiatric/psychological; educational; and social. The clinic provides diagnosis for all ages, but the majority of patients are between 5 and 15 years of age. This clinic has now been replicated in six other sites in Washington State, and is being replicated in several other states in the US (S. Clarren, pers. com., 1999).
One of these is the Alaskan government, which has conducted and reported on a series of scheduled diagnostic clinics in the larger cities in the state. Children were referred to the clinics from a wide variety of sources: state and community agencies who may have documentation of a child being exposed to alcohol in utero; the state’s early childhood intervention program; Head Start; native health corporations; the public health nursing program; local school districts; adoption agencies; and mental health resources, pediatricians and family practitioners. Newspaper and radio announcements also encouraged parents to bring children who might meet the screening criteria. Evidence of prenatal alcohol exposure and either growth deficiency or central nervous system impairment were the screening criteria. An expert diagnostician examined those referred. Those diagnosed with FAS then had a comprehensive care plan developed through an interdisciplinary team.
The care plan included attention to the parents and child and the involvement of a care coordinator to support the plan. The clinics diagnosed few new cases of FAS, but identified a number of children and youth with alcohol-related effects that received care plans addressing a number of needs not previously considered (Li and Pearson, 1996).
Also in the US, clinics offer diagnosis of affected children available for adoption from Eastern Europe and Russia, using videotapes supplied by the orphanages where the children live (Aronson, 2000). How these videotapes can be used to assess growth is unclear. Neither is it clear if any independent evaluation of the children is provided if they are adopted in the US. The fate of children rejected because of a video-diagnosis of FAS is also unknown.
In Manitoba, teleconference facilities that link specialists in Winnipeg with physicians and other health care providers in remote communities are being used to develop and distribute the diagnostic resources across the province (Ridd, 1999). As these clinics are just beginning, there are no available data on effectiveness, costs, or client and professional satisfaction.
|
