Eleven years ago the adoption of my 17-month-old son from Russia started an unexpected journey into international adoption research. His initial attempts to learn English were perplexing. Words were often unintelligible with a unique phonological rule system that didn't match any textbook patterns. Although his language comprehension developed rapidly, his expressive language was unusual. Despite having more than 150 words in his vocabulary, he rarely combined them.
As a concerned mother I combed the literature for information about speech and language development in internationally adopted children. Back then most of the "evidence" was a collection of anecdotal reports that painted a less-than-rosy, and at times quite scary, picture of his future. Evidence-based information was limited to studies of children adopted from Romania that were conducted by Eleanor Ames and her colleagues, and Michael Rutter and his colleagues.
Although the information was important, their studies did not specifically examine speech and language. Thus my son's issues planted the seeds that led to a line of research focused on speech and language development in internationally adopted children (Glennen, 2005; Glennen 2007; Glennen in press; Glennen & Masters, 2002). Since then, other colleagues including Jennifer Roberts, Kathleen Scott, Deborah Hwa-Froelich, Karen Pollock, Rena Krakow, and Jennifer Windsor have added to the growing body of research about speech and language development and disorders in internationally adopted children. Many professionals, however, aren't familiar with this information.
In our house the Discovery Channel series "Mythbusters" is a family favorite. Each episode takes popular myths based in history, legend, movies, or news and uses evidence to prove the myths as true, plausible, or busted. With 10 years of research regarding the speech and language abilities of internationally adopted children, it's time to do some "myth-busting" of our own. What follows is a list of common myths about international adoption and speech and language.
I confess that I once believed most of them, provided professional advice based on them, and wrote about some of them in early publications (Glennen, 2002). However, it's time to set the record straight. We need to use evidence, not myths, to make clinical decisions about speech and language for internationally adopted children.
Myth 1: Internationally adopted children need many years to fully "catch up" in English language acquisition.
Evidence: The majority of internationally adopted children have rapid language-learning that begins within a few days of arriving home. After one year home, children adopted under the age of 24 months develop English language comprehension, production, and articulation abilities that are well within normal limits using standard norms (Glennen, 2007; see Figure 1 [PDF]). This rate of progress does not mean it takes only one year to develop full language potential; in fact, skills keep improving during the preschool years, especially in the area of expressive syntax. However, the initial surge of language "catch-up" occurs rapidly within the first year home.
We know less about children adopted at older ages, but preliminary data indicate equally rapid rates of language learning. Within one year of adoption, most children adopted as 2-year-olds score within normal limits on English language tests of comprehension and expression (Glennen, in press; see Figure 2 [PDF]). Children adopted as 3- and 4-year-olds also score within normal limits on English language comprehension measures after one year, but take more time to develop expressive language abilities fully in English. More research is needed to fully understand these issues; significant English language delays after the first two years home, however, appear to be rare and should be treated as true language or speech disorders.
Myth 2: Early environmental deprivation results in severe, lifelong language-learning disorders.
Evidence: Adoption helps to counteract the effects of orphanage care. Orphanages are not good places to raise children. Children who remain in orphanages have significant language delays and the length of stay correlates highly with poor cognitive and language abilities (Johnson, 2000; O'Connor et al., 2000; Rutter et al., 1998; Miller, 2005; Windsor, Glaze, Koga, & the Bucharest Early Intervention Project Core Group, 2007). At birth the range of potential language abilities for all children falls along the normal curve. Some children have the potential to develop exceptional language skills; others have less potential. When children enter orphanages, environmental and nutritional deprivation gradually erodes their potential.
The poorer the level of care, and the longer the length of institutionalization, the more potential in the area of language ability is lost. The result is a group of children who still have language abilities arrayed into a normal curve, but the curve has shifted downward. While many children will still fall within the "normal range," lost potential translates into proportionately more children falling below average. Once potential is lost, the more important question is whether it can be regained when the environment improves.
Research indicates that many children raised in orphanages have permanent neurobiological changes related to chronic stress within their environment (Gunnar & Quevedo, 2007; Miller, 2005). Chronic abnormal stress reactions lead to overproduction of cortisol and high levels of glucocorticoids in the brain. Prolonged exposure to glucocorticoids leads to structural changes in the brain, primarily the hippocampus, which is important for memory storage and retrieval. Other areas of the brain affected by abnormal regulation of glucocorticoids include the frontal lobe, responsible for executive function and abstract thinking; the cingulate gyrus, responsible for attention and self-control; and the amygdala, responsible for processing emotions. However, some children are genetically more resilient to the neurobiological effects of stress and are not as affected by the orphanage environment (Gunnar & Quevedo). In addition, the neurobiology of stress reactions responds to improvements in the environment, such as adoption into a nurturing home with consistent caregivers (Gunnar & Quevedo).
Adoption goes a long way to counteract lost potential resulting from environmental deprivation. The majority of infants and toddlers adopted into American homes have mild to low-average delays when they first arrive, but make incredible progress during the first years at home (Glennen, 2005; Glennen 2007; Roberts et al., 2005). The incidence of speech and language disorders in children adopted before age 2 is 22% (Glennen, 2007). Although this rate is higher than the 2%–8% reported in the general preschool population (Law, Boyle, Harris, Harkness, & Nye, 2000), the overwhelming majority of internationally adopted children who are adopted before age 2 have normal English language abilities after one year home (Glennen, 2007; Roberts et al., 2005). Internationally adopted children who do not meet this benchmark have true disorders, and need to be diagnosed and treated.
Myth 3:The child's first language will affect aspects of learning the new adopted language.
Evidence: Studies of internationally adopted children under age 2 have found that the first language has no inhibitory or facilitory effect on learning a new language or its phonology. Young children transitioning from Russian to English learned English-language morphological structures in the same developmental sequence as children who spoke English their entire lives (Glennen, Rosinsky-Grunhut, & Tracy, 2005). Children transitioning from Mandarin to English developed the English sound system similarly (Pollack & Price, 2005). Finally, children adopted from China learned vocabulary in patterns typical for children who spoke English (Snedecker, Geren, & Shafto, 2007). The children in these studies were all adopted under the age of 2 and likely did not have a well-developed first-language base to affect English-language learning.
Based on my clinical experience, children adopted at older ages do show signs of interference and facilitation between the birth and adoptive languages. However, it is unknown whether they follow the same patterns as bilingual children or if they transition differently from one language to another. Functional MRI studies of adults who were internationally adopted as children confirm that adult adoptees no longer recognize nor understand their first language, even those who were adopted at school age (Pallier et al., 2003). However, the same fMRI studies also confirm that internationally adopted adults process their new adopted language using different areas of the brain than those of native-language speakers.
Myth 4: Internationally adopted children are bilingual and should be treated like other second-language learners.
Evidence: Internationally adopted children are bilingual only for a short period of time after adoption. Because most adoptive parents do not speak the child's birth language (L1), children quickly lose their abilities in that language. According to Gindis (2003), children adopted at ages 3–4 lose most expressive use of L1 within six to 12 weeks of adoption; receptive abilities are lost within 16–22 weeks. At that point, internationally adopted children are monolingual in English, but the language is not yet fully acquired.
Children who are adopted at older ages are especially affected because they begin school soon after arriving home. If they struggle academically, it is difficult to assess speech or language disorders validly until English develops further. By then, valuable intervention time is lost. Parents adopting children older than 3 years are advised to gather information about their child's speech and language development during the adoption process. If there are reported concerns in the birth country, the educational team should consider conducting an assessment and begin to provide supports within the classroom soon after the child arrives home.
Myth 5: Most internationally adopted children do well at young ages but have language-related academic difficulties in the elementary grades.
Evidence: Most children adopted before age 2 have normal language abilities during the preschool years, and continue to have normal literacy and academic language abilities at school age; data on speech and language are mixed in school-age internationally adopted children. Initial teacher survey data by Dalen and Rygvold (2006) found that internationally adopted children from poor countries with third-world health care systems had worse "academic language" than nonadopted children. In contrast, children adopted from countries with good health care and economic resources were equal to—if not better than—nonadopted children.
Recent assessments of children adopted at young ages from China indicate that by school age, most are performing at average to above-average levels on literacy measures (Scott, Roberts, & Krakow, 2008). Similarly, parent-reported data on children adopted at young ages from Eastern Europe found that by school age, 80% were in regular education classrooms without accommodations (Glennen& Bright, 2005). However, the same parents reported that 27% of the children were receiving speech and language intervention.
One crucial factor across all studies of school-age children is the child's age at adoption. Parent-reported data on children adopted at 3–6 years of age from Eastern Europe found that after five years home, 57% of girls and 82% of boys were diagnosed with communication disorders (Beverly, McGuiness, & Blanton, 2008). Similarly, Dalen and Rygvold (2006) found that children adopted from Colombia at older ages were more likely to have poor academic language abilities at school age than children adopted at younger ages.
It is clear that children adopted at older ages are more susceptible to risk factors that impede speech and language development; more longitudinal data will help determine whether those risk factors cause a temporary gap in speech and language development or indicate a real difference in language abilities.
The important fact is that most children adopted before age 2 have normal language abilities during the preschool years, and continue to have normal literacy and academic language abilities at school age. Internationally adopted children who experience difficulty with higher-level academic language tasks should be assessed and provided with appropriate supports and services based on assessment results.
Myth 6: Internationally adopted children require unique speech and language diagnosis and intervention methods.
Evidence: When children first arrive home, they should be assessed using methods based on guidelines developed for internationally adopted children (Glennen, 2007). Newly arrived infants and toddlers can be assessed reliably using measures of prelinguistic abilities such as vocalizations, gestures, and social pragmatic abilities (see Table 1 [PDF]). Children adopted at 12–24 months of age who initially scored within normal limits on the Communication and Symbolic Behavior Scales-Developmental Profile (Wetherby&Prizant, 2002) did well when reassessed one year later. Conversely, all but one of the children who initially scored below normal limits continued to have poor language and speech abilities one year later.
The rate at which a child learns to comprehend new words is also important to consider when assessing newly arrived children (Glennen, 2007). Children who learn to comprehend new words rapidly do better than children who learn new words at a slower rate. Surprisingly, the rate of learning to express new words is not as predictive when children first arrive home.
The guidelines in Table 1 [PDF] are useful for newly arrived children; however, after one year at home children adopted before age 2 can be assessed using standard English language procedures. The exception is measures of expressive syntax such as mean length of utterance (MLU). We have found that expressive syntax and morphology require additional time to develop in internationally adopted children and are not reliable measures of language abilities until children are age 4 (see Figure 1 [PDF]). Although the children catch up quickly in vocabulary—including expressive vocabulary—measures of MLU, sentence repetition, and morphology elicitation tasks indicate that expressive syntax and morphology take longer to develop to English-language norm levels. We can rule out processing difficulties as the cause of these delays because the same children score well on tasks that assess comprehension of complex directions and that test syntax and morphological comprehension.
In contrast to infants and toddlers, children adopted at older ages are difficult to assess when they first arrive. Although solid guidelines for younger children help differentiate English language-learning issues from speech-language disorders, there are no similar guidelines for older children. The spoken language abilities of older children can't be validly assessed because there is no proficient language. However, preliminary evidence based on a small number of children who were adopted between the ages of 2 and 4 indicates they can be assessed using most standard English-language comprehension measures after one year home (Glennen, in press). Expressive language emerges more slowly; children adopted at ages 3 and 4 can be assessed using expressive language measures two years after adoption (see Figure 2 [PDF]).
If it is determined that a child has a speech or language delay, the process of making a diagnosis and developing intervention plans is identical to that for any other child. The internationally adopted children I follow (Glennen, 2005, 2007) provide an example of this process. One year after adoption, 22% of the children had delays in speech, language, or both. They had a variety of diagnoses including global developmental delays, expressive language delay, receptive language delay, autism, and phonological disorders. In summary, there was no unique "international adoption speech and language disorder" or unique intervention. Speech and language intervention should target each child's diagnosis and symptoms, not the adoption status.
Myth 7: Now that evidence-based information is available, professionals are making better decisions about speech and language in internationally adopted children.
Evidence: Some SLPs are unaware of current research and do not use it in making treatment decisions. From 54% to 68% of internationally adopted children are referred for speech and language assessments, and 35%–50% receive intervention (Glennen, 2007; Glennen& Masters, 2002; Mason &Narad, 2005), a rate higher than the incidence of disorders in this population (22%). During the first year home, 17 of 27 (68%) newly adopted toddlers followed in my research were assessed for speech and language by early intervention teams (Glennen, 2007). Thirteen of the children (48%) were then seen for speech and language intervention. This group included five of the six children who were later diagnosed with language and speech delays—but it also included eight children who developed normal language. According to prelinguistic language assessments conducted when the children were first adopted, these eight did not need treatment; however, their parents sought services anyway. Although those children surely benefited from the intervention, many of them were functioning at the top of their peer group when they were first adopted, and continued to develop language at a rate that surpassed their peers.
Reports about services for older internationally adopted children also cause concern. One child in my longitudinal study was adopted from Eastern Europe at age 4. By second grade she was struggling academically and her parents requested a school-based assessment. School officials insisted the child be tested in her birth language.
The parents protested this decision, as the child had neither heard nor spoken the language for four years. They noted that the Individuals with Disabilities Education Act requires assessment in the child's primary language and maintained that the birth language was no longer primary. Their protest delayed the assessment process until the case was finally brought to the attention of the head of English as a second language services, who asked the school to proceed with English-language testing.
My son had a true expressive language and phonological disorder that required treatment, and he received the early language intervention he needed. However, some internationally adopted children are put on "wait and see" protocols for extended periods of time, or referred to programs for non-English speakers that fail to meet their extensive language-learning needs. Conversely, other children receive services even when they excel on every language measure given (Glennen, 2007).
It's time to stop providing services based on anecdotes and myths, and instead to make clinical decisions for internationally adopted children based on research.
Sharon Glennen, PhD, CCC-SLP, is professor and chair of the Department of Audiology, Speech Language Pathology and Deaf Studies at Towson University. She adopted two children from Russia.
Originally published on ASHA.
