Abstract
Purpose
To determine the diagnostic accuracy and the cutoff point of the variables conicity index, waist to height ratio and fat percentage to detect urinary incontinence in physically active older women.
Method
A total of 152 women were analyzed. The instruments used were the International Physical Activity Questionnaire (IPAQ [Area 4]) to check the level of physical activity, and the Diagnostic Form to obtain sociodemographic data and presence of urinary incontinence. To calculate the conicity index, waist to height ratio and fat percentage, body mass, height and waist circumference were measured. Descriptive and inferential statistics were used. Cutoff points, sensitivity (S) and specificity (SP) were determined by receiver operating characteristic (ROC) curves. A 5% significance level was adopted.
Results
The prevalence of urinary incontinence was of 32.2%. The cutoff point with better sensitivity and specificity for the conicity index was 1.23 (S =87.8; SP =35.9); for the waist to height ratio, it was 0.57 (S =79.6; SP =45.6); and for the fat percentage, it was 39.71 (S =89.8; SP =42.7). The area under the ROC curve was 0.666 for the conicity index, 0.653 for the waist to height ratio, and 0.660 for the fat percentage.
Conclusions
The cutoff points for the anthropometric measurements conicity index, waist to height ratio and fat percentage indicate that these measures can be used to predict urinary incontinence in physically active older women. Furthermore, fat percentage seemed to be the best measure for this population.
Keywords:
anthropometry; ROC curve; urinary incontinence
