1.1

Introduction

Anomia is the most frequent and persisting symptom of aphasia . Anomia occurs secondary to a neurological event (e.g., stroke, brain injury) and inhibits the ability to formulate language, even at the most basic word level. Anomia is a defining feature of aphasia; it extends across multiple subtypes of aphasia and is observed for all grammatical word forms (e.g., nouns, verbs) . Word naming deficits negatively affect people’s ability to communicate their wants and needs and engage in important social settings and activities. In fact, prior research indicates that people with aphasia (PWA) tend to be more troubled by impairments in speaking than in reading, writing, or listening comprehension and impairments in speaking have important effects on how PWA are regarded by others in daily life . Therefore, identifying an effective treatment for the improvement of naming deficits in PWA is critical.

There are a variety of treatment approaches focused on improving expressive language abilities, specifically anomia, in PWA. It is believed that anomia typically results from overarching semantic impairments and reflects an insufficient engagement of the critical features that distinguish concepts from one another . Typically, semantic approaches to treatment are used to target anomia as opposed to phonological approaches. Examples include circumlocution-induced naming , personalized cueing , and semantic feature analysis (SFA) . The focus of the current review is the clinical effectiveness of SFA for the treatment of anomia in adults with neurological injury.

SFA was first introduced by Ylvisaker and Szekeres and later refined by Massaro and Tompkins . It is a commonly used treatment to improve naming and expressive language abilities of PWA by providing an organized method of activating semantic networks . SFA uses a systematic cueing technique whereby PWA are asked to produce words semantically related to the target word they cannot recall . For example, if the target word is “cup”, the cues might involve questions related to its use (e.g., What do you do with it?), its properties (e.g., What does it look like?), where it might be used (e.g., Where do you find it?), what category it belongs to, and what might be associated with it (e.g., What are other things that are similar to it?). Because it is suggested that anomia results from an impaired semantic network, the goal of therapy is to alter the semantic network connectivity through refinement of the damaged network . Hypothetically, SFA improves the retrieval of conceptual information by accessing and refining semantic networks . Increased activation of the semantic network surrounding the target word elevates the likelihood the word will be retrieved and may also aid to repair the damaged semantic system .

A recent review by Boyle examined the effectiveness of SFA. The review included seven studies, and all studies reported positive outcomes for the effectiveness of SFA for improving anomia for individuals with neurological impairments. However, only three reported the magnitude of the treatment effect . A significant limitation of Boyle’s review is that there was no attempt to calculate magnitude of effect using data reported in the included studies. The absence of effect size calculations in the remaining studies made it difficult to conclude the effectiveness of the treatment. Therefore, the aim of the current study is two-fold. First, we extend Boyle’s review by including new research. Second, we apply statistical methods to investigate the magnitude of treatment effect for the included studies in an effort to answer the following clinical question: For patients with non-degenerative aphasia, does semantic feature analysis improve confrontational naming abilities?

1.2

Methods

A search of the literature was conducted to identify studies that investigated SFA as the primary intervention for anomia for PWA. Seven electronic databases were searched through June 2013: Academic Search Premier, AgeLine, CINAHL, ERIC, Medline, PsycInfo, and Linguistics and Language Behavior Abstracts. An additional search was performed within the American Speech-Language and Hearing Association (ASHA) journals, and references from all relevant articles were examined to identify any other applicable studies. A combination of search terms included: aphasia , semantic feature analysis , language disorder , semantic cues , anomia , language treatment , and naming treatment .

Eighty-eight articles were identified through the search process and the first author subjected these to a title and abstract review, which eliminated 70 articles. A study was excluded from the review if it was not experimental in nature, did not address SFA as a treatment and did not include adults with neurological damage. Studies were considered for review if they were written in English and published in a peer-reviewed journal between 1980 and June 2013. The first author reviewed the remaining 18 articles and an additional 7 were excluded as they combined SFA with other treatment methods (e.g., group therapy or semantic priming therapy) or used a hybrid approach to SFA. To be true to the objective of this review, only those studies that used SFA as initially designed by Massaro and Tompkins were included. The process for identifying articles to be included in the evidence-based systematic review is displayed in Fig. 1 .

Process for identification of included studies.

Two certified and licensed speech-language pathologists critically evaluated the remaining studies for methodological quality. Ten studies were evaluated using the Single Case Experimental Design (SCED) scale . The Marcotte and Ansaldo study was not evaluated using the SCED as it was not a single subject research design study. The SCED is an 11-point scale used to evaluate the methodological quality of single case experimental studies. A perfectly designed and executed study would receive a score of 11. Each independent rater provided a score of 1 if the study adequately addressed the specified quality item or a 0 if it was poorly addressed or not addressed at all. Following independent scoring, inter-rater reliability was calculated using Cohen’s weighted kappa statistic . The weighted kappa is the proportion of agreement beyond chance and takes into consideration the degree of disagreement between two independent raters. The weighted kappa score was .656, which is indicative of good inter-rater reliability . When disagreements were present, an average score was calculated. The first author randomly selected three studies, and re-calculated SCED scores to determine intra-rater reliability. To reduce bias and ensure ratings were not dependent upon one another, re-scoring was completed two weeks after the initial scoring . The weighted kappa score for intra-rater reliability was 1.0, indicating perfect intra-rater reliability.

It is well understood that clinical effectiveness is a critical variable to consider for any treatment. At the same time, it is important to interpret clinical effectiveness in the context of a given study. To strengthen the interpretation of the result of this systematic review, levels of evidence, based on the American Speech and Hearing Association’s (ASHA) levels of evidence hierarchy, were also assigned. The ASHA hierarchy was chosen due to its applicability to the effectiveness of interventions for individuals with communication disorders. Two independent raters evaluated each study and assigned a level of evidence. Following independent rating, the weighted kappa was calculated to assess reliability for assigning levels. The first author randomly chose 30% of the studies for reevaluation, to calculate intra-rater reliability. Inter- and intra-rater reliability were excellent ( k = 1.0).

To determine the clinical efficacy of SFA as a treatment method for aphasia, effect sizes were calculated in those studies that reported sufficient data. Prior to calculation, it was necessary to determine the individual values for the pre-treatment and post-treatment phases for each set of trained items. Therefore, effect size was calculated when response data were presented in whole number increments using a variation of Cohen’s d statistic as described by Busk and Serlin . The following formula was used:

When it was not possible to calculate the d statistic, the percent of non-overlapping data (PND) was calculated. PND is the most widely used method of calculating effect size in single case experimental designs . PND is the percentage of phase B data points (the treatment phase) that do not overlap (in a favorable direction) with phase A data points (baseline or no treatment). To determine the magnitude of effect, benchmarks put forth by Scruggs et al. were used. A PND greater than 90% was interpreted to reflect a highly effective treatment; a PND of 70–90% was considered a moderate treatment outcome; and, a PND of 50–70% was considered a questionable effect. Any PND of less than 50% is interpreted as an ineffective intervention since performance during intervention has not affected behavior beyond baseline performance.

1.3

Results

Ten studies included in the review were single subject research design studies and one was a pre- and post-treatment study. Twenty-four participants were represented in the review. Demographic variables of interest for all participants are presented in Table 1 . There was significant heterogeneity in age and time post onset of participants. Participants ranged in age from 24 to 85 years, with a mean age of 60.75 (S.D. = 14.29). Time post onset ranged from 4 to 187 months, with a mean of 53.74 months (S.D. = 50.44). There were 15 males (63%) and 9 females (37%) represented. Non-fluent aphasia was the most represented subtype in this review ( n = 17; 71%) and seven participants (29%) had fluent aphasia. Twenty-one participants had sustained a single cerebrovascular accident. The remaining three participants presented with aphasia following traumatic brain injury. Of the 11 studies included in the review, aphasia severity was reported in only five studies.

Scores on the Single Case Experimental Design (SCED) ranged from 8.0 to 10.5 with an average score of 9.3 out of 11 ( Table 2 ). Following SCED scoring, studies were assigned levels of evidence . Ten studies were considered well-designed, quasi-experimental studies and received a level IIb rating. The Marcotte and Ansaldo study was determined to be an observational controlled study and assigned a level III rating. The prevalence of high SCED scores and IIb evidence level ratings indicate strong methodological quality and rigor for the reviewed studies.

Three studies self-reported effect sizes for participants . Effect sizes were calculated for five participants ( Table 3 ). When data was collected and reported on two or more trials of the dependent variable, an average effect size was calculated . A large effect size was present for two participants ( d = 18.76; d = 10.56). A medium effect was present for six participants ( d = 7.95; d = 9.28; d = 7.48; d = 7.79; d = 7.32; d = 9.617). A small effect size ( d = 1.79; d = 3.86; d = 5.54; d = 2.97; d = 1.26; d = 5.01; d = 5.32; d = 6.35) was present for eight participants. The high prevalence of small effect sizes indicates that the treatment was not effective in improving anomia for the majority of participants as it was for these eight participants. Effect size and PND could not be calculated for one participant from the Marcotte and Ansaldo study.

PND was calculated for seven participants. A large treatment effect (PND > 90%) was present for six participants and a moderate treatment effect was present for one participant (PND = 85%). When examining clinical effectiveness using PND, treatment was highly effective for the majority of participants. None of the participants evidenced PND scores consistent with ineffective treatment.

1.4

Discussion

Results indicate that SFA is an effective intervention for improving confrontational naming of items trained in therapy for individuals with non-degenerative aphasia. In addition to self-reported positive outcomes, effect size calculations indicated that there was a medium to large treatment effect for the majority of participants represented in the review. Although findings suggest that treatment was effective for improving confrontational naming of trained items, limited generalization to untrained items and connected speech were reported in many of the included studies. This finding warrants further examination to determine factors that may influence generalization, such as intensity or dosage. There are several limitations to this review, which will be discussed in detail.

Disclosure of interest

The authors declare that they have no conflicts of interest concerning this article.

2.1

Introduction

L’anomie est le symptôme le plus fréquent et le plus persistant de l’aphasie . L’anomie survient secondairement à un événement neurologique (AVC, lésion cérébrale…) et inhibe la capacité de formulation de langage, même au niveau des mots les plus élémentaires. L’anomie est une caractéristique qui définit l’aphasie ; elle recouvre de multiples sous-types de celle-ci et concerne toutes les formes grammaticales des mots (noms, verbes…) . Les déficits lexicaux nominatifs déteignent négativement sur la capacité des personnes atteintes de communiquer leurs désirs et besoins et de s’engager et s’intégrer à des activités sociales. Certaines études indiquent que des personnes avec aphasie (PAA) ont tendance à être plus dérangées par leur handicap en expression orale que par leurs handicaps en matière de lecture, d’écriture et de compréhension orale. Qui plus est, leurs difficultés d’expression retentissent sur la façon dont les PAA sont vues par les autres dans la vie de tous les jours . L’identification d’un traitement efficace susceptible de résorber les déficits nominatifs chez les PAA revêt une importance critique.

Il existe déjà un certain nombre d’approches au traitement centrées sur l’amélioration des capacités linguistiques expressives chez les PAA atteintes d’anomie. Elles partent de l’idée que de manière typique, l’anomie est le résultat de handicaps sémantiques globaux et qu’elle est le reflet d’une implication insuffisante des aspects critiques qui distinguent les concepts les uns des autres . De manière générale, dans les traitements ciblant l’anomie les approches sémantiques sont préférées à des approches phonologiques. Parmi les exemples figurent la nomination induite par des circonlocutions , le repérage personnalisé ( personalized cueing ) et l’analyse des traits sémantiques (ATS) . Dans ce recensement, c’est en mettant l’accent sur l’efficacité clinque que nous allons aborder le traitement de l’anomie chez des adultes souffrant d’atteintes neurologiques.

L’ATS fut initialement présentée par Ylvisaker et Szekeres ; par la suite, elle a été développée par Massaro et Tompkins . Il s’agit d’un traitement assez répandu visant en cas d’anomie à améliorer les capacités langagières nominatives et expressives en proposant une méthode organisée d’activation des réseaux sémantiques . L’analyse emploie une technique de repérage dans laquelle on demande aux PAA de produire des mots sémantiquement associés au mot cible qu’ils n’arrivent pas à se rappeler . Par exemple, supposons que le mot cible est « tasse ». Les tentatives de repérage ( cues ) pourraient comporter des questions ayant trait à son utilisation (Qu’est-ce que vous faites avec ?), à ses propriétés (À quoi cet objet ressemble-t-il ?), aux lieux possibles de son utilisation (À quel endroit le trouvez-vous ?), à la catégorie dont il relève et aux associations possibles (À quels autres objets est-il similaire ?). Puisqu’il a été proposé que l’anomie résulte d’un réseau sémantique entravé, le but de la thérapie consiste à modifier la connectivité du réseau sémantique en développent de nouveau le réseau endommagé . De manière hypothétique, l’ATS améliorerait la récupération d’informations conceptuelles en accédant aux réseaux sémantiques qu’elle développerait . L’activation renforcée du réseau sémantique entourant le mot cible augmenterait les chances de récupération du mot et pourrait contribuer à la réparation du système sémantique endommagé .

Une revue récente de Boyle a évalué l’efficacité de l’ATS. Cet auteur a recensé sept études différentes, qui ont toutes rapporté des résultats positifs soulignant l’efficacité de l’ATS en atténuant l’anomie chez les personnes avec des handicaps neurologiques. Cela dit, l’importance ou l’ampleur de l’effet traitement n’est considérée que dans trois des sept études . Une limitation significative de la revue de Boyle concerne l’absence de toute tentative de calculer l’ampleur de l’effet en utilisant les données rapportées dans les études incluses. À cause de l’absence de ce calcul dans les autres études publiées, on peut difficilement conclure à l’efficacité du traitement. Par conséquent, nous avons établi deux objectifs. Premièrement, nous allons élargir la revue de Boyle en évoquant des recherches nouvelles. Deuxièmement, nous allons appliquer des méthodes statistiques en vue de déterminer l’ampleur de l’effet des traitements répertoriés dans les études incluses et de répondre de manière satisfaisante à la question clinique suivante: Dans des patients souffrant d’aphasie non dégénérative, est-ce que l’analyse des traits sémantiques améliore les capacités de désignation par rapport au mot cible ?

2.2

Méthodes

Une revue de la littérature a été effectuée afin d’identifier des études enquêtant sur l’ATS comme intervention primaire en cas d’anomie chez les PAA. Jusque dans leurs versions de juin 2013, sept bases de données électroniques étaient explorées : Academic Search Premier, AgeLine, CINAHL, ERIC, Medline, PsycInfo et Linguistics and Language Behavior Abstracts. Une revue additionnelle a eu lieu en passant au peigne fin les journaux de l’American Speech-Language and Hearing Association, et les références à tous les articles pertinents étaient également épluchées afin d’identifier encore d’autres études utilisables. Parmi les mots-clés retenus figuraient : aphasie , analyse des traits sémantiques , désordres , trouble du langage , repérages sémantiques , anomie , traitement du langage et traitement nominatif .

Lors de la revue 88 articles ont été identifiés, mais dès que le premier auteur a passé en revue les titres et les résumés, 70 articles ont été éliminés. Une étude a été exclue de la revue :

• •
  

  si elle n’était pas expérimentale ;

  
  
• •
  

  si elle n’abordait pas l’ATS comme traitement ;

  
  
• •
  

  si elle ne concernait pas des adultes souffrant d’atteintes neurologiques.

Par contre, dans un premier temps les études étaient retenues :

• •
  

  si elles étaient rédigées en anglais ;

  
  
• •
  

  si elles avaient fait l’objet d’une publication entre 1980 et juin 2013 dans un journal avec comité de lecture.

Ensuite, le premier auteur a passé en revue les 18 articles restants dont 7 étaient exclus parce qu’ils associaient l’ATS à d’autres méthodes de traitement (thérapie de groupe ou thérapie de stimulation sémantique) ou parce qu’ils employaient une approche hybride à l’ATS. Afin de rester en conformité avec l’objectif de la revue, seules les études abordant l’ATS telle qu’elle avait été initialement conçue par Massaro et Tompkins ont été incluses de manière définitive. Le processus par lequel les articles à inclure ont été identifiés dans le cadre de notre revue systématique fondée sur des preuves est résumé sur la Fig. 1 .

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

Crenobalneotherapy (spa therapy) in patients with knee and generalized osteoarthritis: A post-hoc subgroup analysis of a large multicentre randomized trial Diagnostic value of clinical tests for degenerative rotator cuff disease in medical practice Piriformis syndrome diagnosis: On two professional cyclists The relation between repeated 6-minute walk test performance and outcome in patients with chronic heart failure Parent–Professional Partnership The Bedside Examination in Dysphagia

Stay updated, free articles. Join our Telegram channel

Join