BACKGROUND: Labor pain is often described as the worst pain in a woman's life, but the experience is highly variable. Although many factors have been linked to labor pain, it has been difficult to assess the individual effects of these factors because labor is a dynamic process and pain intensity changes over the course of labor. Previous studies have used average pain scores. The aim of this study was to develop and validate a model that would allow for the statistical analysis of factors that affect pain throughout labor. METHODS: We conducted this study with a retrospective database drawn from the medical records of 200 consecutive nulliparous parturients who delivered at New York Presbyterian Hospital between October 2006 and January 2007. Numerical rating scale scores for pain with contractions (0-10 scale), cervical dilation, and oxytocin use before analgesia request were recorded. Nonlinear effects modeling with a sigmoid equation was used to describe the relationship between reported pain and cervical dilation. The modeling technique was developed with data from 91 parturients and validated with an independent set of 95 parturients (all parturients with pain scores more than zero). The resulting model was used to analyze the effect of a sample covariate, oxytocin administration, on reported pain in the entire data set. RESULTS: The model derived from our training set was predictive of the data from our validation set (P < 0.001). Predicted pain scores were on average two numerical rating scale points above or below measured pain scores. Analyzing oxytocin as a covariant showed that women treated with oxytocin reported 48% more pain at the start of labor but did not have a significantly more rapid increase in pain or higher maximal pain when compared with women not treated with oxytocin. Women treated with oxytocin had slower early labor and more rapid late labor. CONCLUSION: We have developed and validated a model for describing pain over the course of labor. Our model is suited to the statistical analysis of covariance and could potentially be used to compare the effects of covariants on labor pain and the rate of change of pain.
BACKGROUND:Labor pain is often described as the worst pain in a woman's life, but the experience is highly variable. Although many factors have been linked to labor pain, it has been difficult to assess the individual effects of these factors because labor is a dynamic process and pain intensity changes over the course of labor. Previous studies have used average pain scores. The aim of this study was to develop and validate a model that would allow for the statistical analysis of factors that affect pain throughout labor. METHODS: We conducted this study with a retrospective database drawn from the medical records of 200 consecutive nulliparous parturients who delivered at New York Presbyterian Hospital between October 2006 and January 2007. Numerical rating scale scores for pain with contractions (0-10 scale), cervical dilation, and oxytocin use before analgesia request were recorded. Nonlinear effects modeling with a sigmoid equation was used to describe the relationship between reported pain and cervical dilation. The modeling technique was developed with data from 91 parturients and validated with an independent set of 95 parturients (all parturients with pain scores more than zero). The resulting model was used to analyze the effect of a sample covariate, oxytocin administration, on reported pain in the entire data set. RESULTS: The model derived from our training set was predictive of the data from our validation set (P < 0.001). Predicted pain scores were on average two numerical rating scale points above or below measured pain scores. Analyzing oxytocin as a covariant showed that women treated with oxytocin reported 48% more pain at the start of labor but did not have a significantly more rapid increase in pain or higher maximal pain when compared with women not treated with oxytocin. Women treated with oxytocin had slower early labor and more rapid late labor. CONCLUSION: We have developed and validated a model for describing pain over the course of labor. Our model is suited to the statistical analysis of covariance and could potentially be used to compare the effects of covariants on labor pain and the rate of change of pain.
Authors: Elena Reitman; Jessamyn Conell-Price; Jennifer Evansmith; Luke Olson; Sofia Drosinos; Nancy Jasper; Paula Randolph; Richard M Smiley; Steven Shafer; Pamela Flood Journal: Anesthesiology Date: 2011-04 Impact factor: 7.892