J M Riordan1, M Koehn. 1. School of Nursing, Wichita State University, KS 67260-0041, USA. jriordan@feist.com
Abstract
OBJECTIVE: This study examined validity and reliability of three clinical instruments that assess feedings at the breast. DESIGN: A descriptive correlational design testing the validity and interrater and test-retest reliability of instruments. SETTING: Hospital rooms and the participants' homes. SUBJECTS: Eleven breastfeeding women and their neonates were videotaped in 23 breastfeeding observations. INTERVENTIONS: The Infant Breastfeeding Assessment Tool (IBFAT), the Mother Baby Assessment Tool (MBA), and the LATCH assessment tool were scored by three nurse raters using videotapes of breastfeedings. Instruments were completed twice by each rater with a 6-month period between administration. MAIN OUTCOME MEASURES: To test validity, test-retest, and interrater reliability, Spearman correlation coefficients among raters' breastfeeding assessment scores, among scores of each instrument, and between test and retest scores of raters. Percent of agreement among raters for each of the items in the three tools. RESULTS: Reliability coefficients for all three assessment tools are below acceptable levels for clinical decisions. Spearman rank coefficients of pairwise interrater correlations were .57, .27, and .69 for the IBFAT: .66, .64, and .33 for the MBA; and .11, .46, and .48 for the LATCH assessment tool. Spearman rank coefficients among instrument scores were .69, .78, and .68. Test-retest correlations were .88, .78, and .64. Percent of agreement among raters for each of the items in the three tools was highly variable, ranging from 37.0 to 97.2. CONCLUSION: The IBFAT, MBA, and LATCH as tools to measure breastfeeding effectiveness are not sufficiently reliable at this stage in their development; thus, these tools cannot be valid for clinical use. These tools need to be revised and retested before use in clinical practice to identify breastfeeding mother-infant pairs who need intervention.
OBJECTIVE: This study examined validity and reliability of three clinical instruments that assess feedings at the breast. DESIGN: A descriptive correlational design testing the validity and interrater and test-retest reliability of instruments. SETTING: Hospital rooms and the participants' homes. SUBJECTS: Eleven breastfeeding women and their neonates were videotaped in 23 breastfeeding observations. INTERVENTIONS: The Infant Breastfeeding Assessment Tool (IBFAT), the Mother Baby Assessment Tool (MBA), and the LATCH assessment tool were scored by three nurse raters using videotapes of breastfeedings. Instruments were completed twice by each rater with a 6-month period between administration. MAIN OUTCOME MEASURES: To test validity, test-retest, and interrater reliability, Spearman correlation coefficients among raters' breastfeeding assessment scores, among scores of each instrument, and between test and retest scores of raters. Percent of agreement among raters for each of the items in the three tools. RESULTS: Reliability coefficients for all three assessment tools are below acceptable levels for clinical decisions. Spearman rank coefficients of pairwise interrater correlations were .57, .27, and .69 for the IBFAT: .66, .64, and .33 for the MBA; and .11, .46, and .48 for the LATCH assessment tool. Spearman rank coefficients among instrument scores were .69, .78, and .68. Test-retest correlations were .88, .78, and .64. Percent of agreement among raters for each of the items in the three tools was highly variable, ranging from 37.0 to 97.2. CONCLUSION: The IBFAT, MBA, and LATCH as tools to measure breastfeeding effectiveness are not sufficiently reliable at this stage in their development; thus, these tools cannot be valid for clinical use. These tools need to be revised and retested before use in clinical practice to identify breastfeeding mother-infant pairs who need intervention.