Karin Brunnegård1, Emilie Hagberg2,3, Christina Havstam4,5, Åsa Okhiria6,7, Kristina Klintö8,9. 1. Department of Clinical Sciences/Speech and Language Pathology, Umeå University, Umeå, Sweden. 2. Karolinska University Hospital, Stockholm Craniofacial Team and Functional Area Speech & Language Pathology, Stockholm, Sweden. 3. Division of Speech and Language Pathology, Karolinska Institutet, CLINTEC, Stockholm, Sweden. 4. Department of Otorhinolaryngology/Speech-Language Pathology, Sahlgrenska University Hospital, Gothenburg, Sweden. 5. Institute of Neuroscience and Physiology/Speech-Language Pathology Unit, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden. 6. Department of Speech Language Pathology, Uppsala University Hospital, Uppsala, Sweden. 7. Department of Surgical Sciences, Plastic Surgery, Uppsala University, Uppsala, Sweden. 8. Division of Speech and Language Pathology, Department of Otorhinolaryngology, Skåne University Hospital, Malmö, Sweden. 9. Department of Clinical Sciences, Lund University, Malmö, Sweden.
Abstract
OBJECTIVE: To assess the reliability of speech variables and speech-related quality indicators in the Swedish quality registry for cleft lip and palate (CLP). DESIGN: Retrospective study. SETTING: Primary care university hospitals. PARTICIPANTS: Fifty-two 5-year-old children with unilateral CLP and 41 with bilateral CLP. MAIN OUTCOME MEASURES: Registry data for "percent nonoral errors" and "perceived velopharyngeal competence" (VPC) were compared to reassessments by 4 independent judges based on audio recordings. Interjudge agreement for "percent consonants correct" (PCC) and the reliability of 3 quality indicators were also assessed. Agreement was calculated with single measures intraclass correlation coefficient (ICC) for articulation outcomes, quadratic weighted κ and ICC for VPC, and percentage agreement and κ for quality indicators. RESULTS: When the agreement between registry data and the judges' reassessments was assessed, the ICC was 0.79 for percent nonoral errors. For VPC, the κ coefficient was 0.66 to 0.75 and the ICC was 0.73. Interjudge agreement for PCC calculated with ICC was 0.85. For the quality indicator "proportion of children with ≥86% correct consonants," all 4 judges were in agreement for 72% of the cases. For "proportion of children without nonoral speech errors" and "proportion of children with competent or marginally incompetent velopharyngeal function," the agreement between registry data and the 4 judges was 89% and 85%, respectively. CONCLUSIONS: The results indicate that registry data on PCC, percent nonoral errors, VPC, and the quality indicators "proportion of children without nonoral speech errors" and "proportion of children with competent or marginally incompetent velopharyngeal function" are reliable.
OBJECTIVE: To assess the reliability of speech variables and speech-related quality indicators in the Swedish quality registry for cleft lip and palate (CLP). DESIGN: Retrospective study. SETTING: Primary care university hospitals. PARTICIPANTS: Fifty-two 5-year-old children with unilateral CLP and 41 with bilateral CLP. MAIN OUTCOME MEASURES: Registry data for "percent nonoral errors" and "perceived velopharyngeal competence" (VPC) were compared to reassessments by 4 independent judges based on audio recordings. Interjudge agreement for "percent consonants correct" (PCC) and the reliability of 3 quality indicators were also assessed. Agreement was calculated with single measures intraclass correlation coefficient (ICC) for articulation outcomes, quadratic weighted κ and ICC for VPC, and percentage agreement and κ for quality indicators. RESULTS: When the agreement between registry data and the judges' reassessments was assessed, the ICC was 0.79 for percent nonoral errors. For VPC, the κ coefficient was 0.66 to 0.75 and the ICC was 0.73. Interjudge agreement for PCC calculated with ICC was 0.85. For the quality indicator "proportion of children with ≥86% correct consonants," all 4 judges were in agreement for 72% of the cases. For "proportion of children without nonoral speech errors" and "proportion of children with competent or marginally incompetent velopharyngeal function," the agreement between registry data and the 4 judges was 89% and 85%, respectively. CONCLUSIONS: The results indicate that registry data on PCC, percent nonoral errors, VPC, and the quality indicators "proportion of children without nonoral speech errors" and "proportion of children with competent or marginally incompetent velopharyngeal function" are reliable.
Authors: Kristina Klintö; Agneta Karsten; Agneta Marcusson; Anna Paganini; Sara Rizell; Jenny Cajander; Karin Brunnegård; Malin Hakelius; Åsa Okhiria; Petra Peterson; Avni Abdiu; Christina Havstam; Hans Mark; Emilie Hagberg; Lena Björnström; Anna-Paulina Wiedel; Magnus Becker Journal: BMC Health Serv Res Date: 2020-06-11 Impact factor: 2.655