Determining Clinically Relevant Cutoff Scores for the Iowa Infant Feeding Attitude Scales Among Prenatal Women in Canada

Abstract

Background:

The original 17-item Iowa Infant Feeding Attitude Scale (IIFAS) has been validated and widely used to assess attitudes toward breastfeeding. A reduced 13-item version of the IIFAS was recently validated in a Canadian setting. However, cutoff scores for categorization of infant feeding attitudes on both scales have not yet been established.

Research Aim:

The aim of this study was to determine optimal cut-ff scores predicting infant feeding attitudes and outcomes for the original and reduced IIFASs.

Methods:

A population-based prospective cohort study was undertaken in the Canadian province of Newfoundland and Labrador. A sample of 658 pregnant women were followed up to 1 month postpartum. The receiver operating curve and Youden index were assessed to identify the sensitivity and specificity of cutoff scores. The magnitude at which these scores predicted postpartum feeding outcomes was evaluated using linear regression.

Results:

Scores of ≤60 (sensitivity = 0.81, specificity = 0.87) and ≤45 (sensitivity = 0.84, specificity = 0.83) for the 17-item and 13-item IIFASs, respectively, were found to be optimal cutoff scores for predicting negative breastfeeding attitudes. The cutoff score for the reduced IIFAS version maintained its ability to predict women who formula-fed at 1 month postpartum (adjusted odds ratio = 6.32, 95% confidence interval = 1.84-11.61) compared with the original scale (adjusted odds ratio = 4.62, 95% confidence interval = 2.42-16.52).

Conclusion:

The proposed cutoff scores for the original and reduced IIFASs have excellent predictive ability to determine infant feeding attitudes and outcomes. The classification of scores enhances the use and applicability of the IIFAS.