regression models for ordinal responses a review of methods

Midline episiotomy carried with it a relative risk of

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: ncision at the midline directed obliquely away from the rectum.17 The response variable, laceration, is coded on a fivepoint ordinal scale, classified as ‘1°’ (least severe) to ‘4°’ (most severe), the classification based on the amount of tissue damage involvement, and a fifth group consisting of women free of any laceration.17,18 For the purpose of this paper, we will restrict our analysis to midline episiotomy, coded as ‘0’ indicating the absence of midline episiotomy and ‘1’ if the procedure was performed. The proportional odds (PO) and continuation-ratio (CR) models (models 2 and 4, respectively) were fit to the data described in Table 3, and their results summarized in Table 4. Midline episiotomy carried with it a relative risk of 2.1 (95% CI : 1.8–2.5) compared to no episiotomy for ‘any laceration’ by the PO model, and a relative risk of 1.4 (95% CI : 1.3–1.5) by the CR model. It is important to note that the assumptions of the underlying models differ: the PO model ass...
View Full Document

This document was uploaded on 02/25/2014.

Ask a homework question - tutors are online