Investigating the adequacy of intervention descriptions in recent speech-language pathology literature: Is evidence from randomized trials useable?

Publication Type:
Journal Article
Citation:
American Journal of Speech-Language Pathology, 2017, 26 (2), pp. 443 - 455
Issue Date:
2017-05-01
Metrics:
Full metadata record
Files in This Item:
Filename Description Size
Ludemann_et_al-2017-American_Journal_of_Speech-Language_Pathology.pdfPublished Version507.39 kB
Adobe PDF
© 2017 American Speech-Language-Hearing Association. Purpose: To evaluate the completeness of intervention descriptions in recent randomized controlled trials of speech-language pathology treatments. Method: A consecutive sample of entries on the speechBITE database yielded 129 articles and 162 interventions. Interventions were rated using the Template for Intervention Description and Replication (TIDieR) checklist. Rating occurred at 3 stages: interventions as published in the primary article, secondary locations referred to by the article (e.g., protocol papers, websites), and contact with corresponding authors. Results: No interventions were completely described in primary publications or after analyzing information from secondary locations. After information was added from correspondence with authors, a total of 28% of interventions was rated as complete. The intervention elements with the most information missing in the primary publications were tailoring and modification of interventions (in 25% and 13% of articles, respectively) and intervention materials and where they could be accessed (18%). Elements that were adequately described in most articles were intervention names (in 100% of articles); rationale (96%); and details of the frequency, session duration, and length of interventions (69%). Conclusions: Clinicians and researchers are restricted in the usability of evidence from speech-language pathology randomized trials because of poor reporting of elements essential to the replication of interventions.
Please use this identifier to cite or link to this item: