Open AccessFramework for improving outcome prediction for acute to chronic low back pain transitions
Author(s) -
Steven Z. George,
Trevor A. Lentz,
Jason M. Beneciuk,
Nrupen A. Bhavsar,
Jennifer M. Mundt,
Jeff Boissoneault
Publication year - 2020
Publication title -
pain reports
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.113
H-Index - 15
ISSN - 2471-2531
DOI - 10.1097/pr9.0000000000000809
Subject(s) - chronic pain , health care , outcome (game theory) , low back pain , clinical practice , medicine , predictive modelling , physical therapy , computer science , alternative medicine , machine learning , political science , mathematics , mathematical economics , pathology , law
Clinical practice guidelines and the Federal Pain Research Strategy (United States) have recently highlighted research priorities to lessen the public health impact of low back pain (LBP). It may be necessary to improve existing predictive approaches to meet these research priorities for the transition from acute to chronic LBP. In this article, we first present a mapping review of previous studies investigating this transition and, from the characterization of the mapping review, present a predictive framework that accounts for limitations in the identified studies. Potential advantages of implementing this predictive framework are further considered. These advantages include (1) leveraging routinely collected health care data to improve prediction of the development of chronic LBP and (2) facilitating use of advanced analytical approaches that may improve prediction accuracy. Furthermore, successful implementation of this predictive framework in the electronic health record would allow for widespread testing of accuracy resulting in validated clinical decision aids for predicting chronic LBP development.