ACQ Vol 13 no 3 2011

people the same is what is meant by “not discriminate[ing]”, then their position is indeed supported. However, cultural competence and social justice principles are consistent in their presentation that “equity” does not mean “treat all people the same” but rather “treat all according to their need”, and that “respect” and “culturally appropriate” means going beyond our own cultural understandings (Martin, 2009). Interestingly, when reviewing the code of ethics of the New Zealand Speech-Language Therapists’ Association (NZSTA) you find that they have taken the discussion a little further. For example, articles 1.6–1.8 indicate that SPs “advocate that all clients irrespective of age, ethnic background, … should have access to speech language therapy services … acknowledge and allow for individuality of clients including race, age, religion, culture”, and “respect the rights of and be sensitive to factors such as a client’s race, age, religion, culture”, and, at article 3.5, “offer these services to clients in a manner which does not discriminate on the grounds of race, religion, culture” and further: “4.4 Members shall ensure that the research is in accord with the Treaty of Waitangi”. The accreditation standards for programs that train SPs in New Zealand have an entire section on working with the Treaty of Waitangi, which is a legislated requirement. Terms such as “advocate”, “be sensitive to”, and “respect the rights of” are somewhat stronger in their implications. You feel that drawing upon these principles, a SP does have an ethical responsibility to consider and address the sociocultural needs of the individual clients that they are working with in practice. Looking for the research evidence So far then, these documents can support the idea that practices in multicultural and multilingual contexts should be different to those of monolingual ones. While you are aware of expert opinion in the field that supports this view (Battle, 2002; Kohnert, 2010; Roseberry-McKibbin, 2007), you know that you will require more evidence to advocate the need for this shift in thinking to your colleagues. General points regarding the need to adapt services can certainly be made. But the arguments put forward suggest there are some specific questions that require answers. They are: 1. Does the evidence indicate that bilingualism is “harder” for children with language difficulties than monolingualism? 2. Does the evidence indicate that SPs can – or can’t – effectively or competently offer any services in languages that they do not speak well? 3. Is there evidence that supporting multilingualism has any benefits for learners, particularly those with language difficulties? To answer these questions, the process of searching the research literature was undertaken. Popular databases for speech pathology intervention research were first used in attempt to find high quality bilingual or multilingual treatment studies. Review articles and expert opinion on the topic were also sourced as outlined below. Databases The Language and Linguistic Behaviour Abstracts (LLBA), PsychInfo and CINAHL databases were searched, and in order to find any systematic reviews or previously appraised articles on the topic, the Cochrane library and speechBite TM databases were also used. The search terms are listed in Table 1. The search was limited to oral language, excluding studies on speech and literacy.

The LLBA search yielded 114 results of which 19 proved to be about intervention, ranging from 2010 to 1982. Three were dissertations and not available, three were reviews, one was in a foreign language, leaving six identified as research studies on this topic. Studies conducted before 1990 were excluded. PsychInfo yielded six results, which ranged from 1983 to 2009. Only one was on intervention and that was not a research study. Only two of the six were also found by the LLBA search and the others were not relevant. CINHAHL replaced “bilingual” with “multilingualism” and “language impairment” with “language disorders”. It brought up 50 hits from 2010–1997, 18 of which were not in the LLBA search but 13 of these were non-research papers or other sources such as book reviews or magazine articles not reported in LLBA. The remaining five were not relevant to intervention. This left six research articles to review, which are listed in Table 2 with a summary of the purpose of the study, the key findings and level of evidence. The Cochrane database did not bring up any relevant articles. However, speechBite TM brought up 10 under “bilingual child” of which four proved relevant. Where possible, the speechBite TM database provides a rating for the quality of the studies listed as seen in Table 3. None of the relevant studies rated more than 3 out of 10 suggesting low methodological quality overall. It should be noted that while the speechBITE TM database recognises the clinical value inherent in well-designed single case experimental designs, these are currently not rated. This comprehensive search for relevant literature highlights that a limited number of high quality intervention studies have been conducted in the field. To look further at the quality of this research, Table 4 provides an example of a critically appraised article for one of the studies to provide more in-depth insight into the nature of research in this area following the EBP guidelines. Review articles Relevant review articles on the topic were used to expand the search strategy and to explore expert opinion in the field. In 2010, two journals had issues devoted to the topic, Applied Psycholinguistics 31, and the Journal of Communication Disorders 43. The second contained two important reviews of the research evidence, one by Thordardottir (2010) and one by Kohnert (2010), along with a number of other relevant research articles. A 2006 edition of Topics in Language Disorders was also devoted to this topic, and included another review by Goldstein (2006). Please see Table 5 for a list of the review articles. The review articles all confirm the paucity of research in the field and that the available evidence tends to be from the weaker Table 1. Concept map to generate keywords Client group Intervention Comparative Outcomes search terms search terms Intervention Possible Possible No particular No particular search terms: search terms search terms search terms • child* • language used. used. • bilingual* intervention Interested in Interested in • language • bilingual effect of the any outcome. impair* or intervention intervention language only. disorder* * Indicates a truncated string (will pick up “child”, “child’s”, “children”, “children’s” etc.)

149

ACQ Volume 13, Number 3 2011

www.speechpathologyaustralia.org.au