Tonya J Roberts1, Sandra E Ward. 1. Center for Patient-Centered Interventions, School of Nursing, University of Wisconsin-Madison, USA. tjbeal@wisc.edu
Abstract
BACKGROUND: Latent transition analysis is a method of modeling change over time in categorical variables. It has been used in the social sciences for many years, but not in nursing research. OBJECTIVE: The purposes of this study were to illustrate the utility of latent transition analysis for nursing research by presenting a case example (a secondary analysis of data from a previously conducted randomized control trial testing the effectiveness of a tailored psychoeducational intervention to decrease patient-related attitudinal barriers to cancer pain management) and to understand for whom and in what direction the tailored intervention resulted in change with respect to attitudinal barriers and pain symptoms. METHODS: The model was developed by (a) defining a class structure on the basis of individuals' barrier patterns, (b) adding demographic predictors and distal pain outcomes, and (c) modeling and testing transitions across classes. RESULTS: There were two classes of individuals: Low Barriers and High Barriers. Older, less educated individuals were more likely to be in the High Barriers class at Time 1. Individuals in either class did not have different pain outcomes at the end of the study. Of those individuals that transitioned across classes, those who received the intervention were statistically more likely to move in a favorable direction (to the Low Barriers class). Furthermore, there is evidence that some individuals in the control group had unfavorable outcomes. DISCUSSION: The results from the example provide useful information about for whom and in what direction the intervention resulted in change. Latent transition analysis is a valuable procedure for nurse researchers because it collapses large arrays of categorical data into meaningful patterns. It is a flexible modeling procedure with extensions allowing further understanding of a change process.
RCT Entities:
BACKGROUND: Latent transition analysis is a method of modeling change over time in categorical variables. It has been used in the social sciences for many years, but not in nursing research. OBJECTIVE: The purposes of this study were to illustrate the utility of latent transition analysis for nursing research by presenting a case example (a secondary analysis of data from a previously conducted randomized control trial testing the effectiveness of a tailored psychoeducational intervention to decrease patient-related attitudinal barriers to cancer pain management) and to understand for whom and in what direction the tailored intervention resulted in change with respect to attitudinal barriers and pain symptoms. METHODS: The model was developed by (a) defining a class structure on the basis of individuals' barrier patterns, (b) adding demographic predictors and distal pain outcomes, and (c) modeling and testing transitions across classes. RESULTS: There were two classes of individuals: Low Barriers and High Barriers. Older, less educated individuals were more likely to be in the High Barriers class at Time 1. Individuals in either class did not have different pain outcomes at the end of the study. Of those individuals that transitioned across classes, those who received the intervention were statistically more likely to move in a favorable direction (to the Low Barriers class). Furthermore, there is evidence that some individuals in the control group had unfavorable outcomes. DISCUSSION: The results from the example provide useful information about for whom and in what direction the intervention resulted in change. Latent transition analysis is a valuable procedure for nurse researchers because it collapses large arrays of categorical data into meaningful patterns. It is a flexible modeling procedure with extensions allowing further understanding of a change process.
Authors: Mahmood R Gohari; Joel A Dubin; Richard J Cook; Scott T Leatherdale Journal: Health Promot Chronic Dis Prev Can Date: 2019-09 Impact factor: 3.240