Session Title: Virtual Poster Hall
Session Time: None. Available on demand.
Disclosures: Thomas Kienbacher, MD: No financial relationships or conflicts of interest
Objective: The authors of this study recently published research on a computer generated machine learning algorithm that automatically identifies patients with impairment in activity and participation categories of the brief ICF Core set for low back pain from routinely gathered patient reported outcome data and on patients` specific limitations according to different age and gender subgroups (Tuechler K et al: accepted for publication EUR J Phys Med Rehabil 2020; Fehrmann E et al: Disabil Rehabil 2018). This study investigated improvements in these limitations along with a stepped rehabilitation program that is widely covered by social security in Europe.
Design: Cohort study Setting : Outpatient referral rehabilitation clinic Participants : 2551 (1637 females), aged 21 to 76 years (mean 50,1 years) chronic low back pain patients
Interventions: Patients completed the Roland Morris Disability Questionnaire and Pain Disability Index prior to and at the end of a six months comprehensive rehabilitation program.
Main Outcome Measures: Percentage of patients with limitations in their subgroup specific (below and above 50 years of age, male, and female) activity and participation categories before and after rehabilitation.
Results: Linear mixed models revealed that older men had highest impairment in “self-care” (d540: 63%) and young men in “employment” (d845: 75%, d850: 80%) whereas young women reported most impairment in “mobility” (d415: 84%, d430: 71%) and “housework” (d640: 71%, d859: 77%) categories before rehabilitation. All limitations significantly improved in all subgroups but specific differences persisted until the assessment after the intervention (d540: 46%; d845: 55%; d850: 61%; d415: 66%; d430: 53%; d640: 46%; d859: 53%). Thus the rehabilitation program provided did not specifically address the patient needs. Conclusions: The novel computer generated algorithm should be considered in intervention planning for matched rehabilitation programs focusing on improvements in individual functional health deficits according to the ICF.
Level of Evidence: Level II
To cite this abstract in AMA style:Kienbacher T. Gender and Age Associated Goal Achievement in Automatically Predicted Activity and Participation Categories Along with Comprehensive Back Pain Rehabilitation [abstract]. PM R. 2020; 12(S1)(suppl 1). https://pmrjabstracts.org/abstract/gender-and-age-associated-goal-achievement-in-automatically-predicted-activity-and-participation-categories-along-with-comprehensive-back-pain-rehabilitation/. Accessed July 30, 2021.
« Back to AAPM&R Annual Assembly 2020
PM&R Meeting Abstracts - https://pmrjabstracts.org/abstract/gender-and-age-associated-goal-achievement-in-automatically-predicted-activity-and-participation-categories-along-with-comprehensive-back-pain-rehabilitation/