Related Experiment Video
Updated: Aug 14, 2025
Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack
Published on: May 15, 2020
Design and validation of a predictive model for determining the risk of developing fibromyalgia
Narly Benachi Sandoval1, Joaquim Fernández Solà2, Andrea Guaita Mateo3
1Dept. of Public Health, Mental Health and Maternal and Child Health Nursing, University of Barcelona, Spain; Research Group "Health Care (recognised by Colciencias)", Universidad Santiago de Cali, Colombia; Consorci d'Atenció Primària de Salut Barcelona Esquerra, Barcelona, Spain; Working group of the Central Sensitivity Syndrome Unit, Àrea Integral de Salut Barcelona Esquerra (AISBE), Barcelona, Spain. nbenachi@clinic.cat, nbenachi@ub.edu.
Objectives:
Fibromyalgia is a prevalent disease of unknown aetiology and is difficult to diagnose. Despite the availability of the American College of Rheumatology criteria for diagnosis, it continues to be a challenge in the field of primary health care in terms of identifying individuals with susceptibility to developing the disease. The aim of this study is to design and validate a predictive model of fibromyalgia in subjects with a history of chronic pain.
Methods:
This multicentre observational retrospective cohort study was performed on patients aged >18 years, who visited four primary health centres between 2017 and 2020, with a diagnosis of fibromyalgia or arthritis. The Bootstrapping resampling method was used for the validation of the model.
Results:
A total of 198 subjects with fibromyalgia (93 with osteoarthritis, 20 with other types of arthritis, 4 with rheumatoid arthritis) and 120 without fibromyalgia (116 with osteoarthritis, 23 with other types of arthritis, 7 with rheumatoid arthritis) participated in the study. The predictive factors of the final model were self-reported age at onset of symptoms, first-line family history of neurological diseases, exposure to levels of stress, history of post-traumatic acute emotional stress, and personal history of chronic widespread pain prior to diagnosis, comorbidity, and pharmacological prescription during the year of diagnostic confirmation. The predictive capacity adjusted by Bootstrapping was 0.972 (95% CI: 0.955-0.986).
Conclusions:
The proposed model showed an excellent predictive capacity. The risk calculator designed from the predictive model allows health professionals to have a useful tool to identify subjects at risk of developing fibromyalgia.
