New Predictive Model Enables Better Diagnosis of Fibromyalgia, Study Suggests

New Predictive Model Enables Better Diagnosis of Fibromyalgia, Study Suggests

A new model may help diagnose fibromyalgia earlier and improve treatment, a study by ProCare Pain Solutions and Pfizer suggests.

The research, “Clinical Characteristics of Fibromyalgia in a Chronic Pain Population,” was published in the journal Pain Practice. It is the first in a series of three articles sponsored by Pfizer.

Fibromyalgia is a chronic disorder characterized by widespread musculoskeletal pain, joint stiffness, and fatigue. Clinical diagnosis and classification of its severity rely on the American College of Rheumatology (ACR) 2010 criteria, which includes the Widespread Pain Index (WPI) and the Symptom Severity Index (SSI).

However, most community-based physicians do not routinely use the established ACR 2010 criteria. Furthermore, although physicians increasingly use electronic health records (EHRs), they do not include the WPI and SSI criteria. Therefore, alternatives that may be more practical through EHRs are needed to better diagnose fibromyalgia patients.

Researchers collected EHR data from 1999 to 2015 from the ProCare network of clinics to compare fibromyalgia characteristics in a total of 82,445 chronic pain patients diagnosed either via a traditional model or a new predictive model.

Results showed that several comorbidities are significantly associated with the diagnosis of fibromyalgia. These comorbidities include muscle spasms, fasciitis (inflammation in the tissue surrounding muscles and nerves), cervicalgia (pain in the neck and upper back), thoracic pain, shoulder pain, arthritis, cervical disorders, and latex allergy.

Patients with fibromyalgia also underwent more musculoskeletal procedures — including physical therapy, joint injections, and selective nerve root, among others — and experienced greater treatment intensity.

“The identification of multiple comorbidities, diagnoses, and musculoskeletal procedures that were significantly associated with FM may facilitate differentiation of FM patients from other conditions characterized by chronic widespread pain,” the researchers wrote.

The new predictive system enabled the identification of 2,444 fibromyalgia patients, compared to 1,069 using the traditional method.

“The predictive model allowed a broader selection of [fibromyalgia] patients and, by doing so, may assist in more accurately identifying patients who may otherwise go undiagnosed, enabling early and appropriate treatment,” the researchers wrote.

“This article and the research done at ProCare Pain Solutions highlights our commitment to provide the best care possible for our patients as well as educate our colleagues, and advance the profession of pain medicine,” Fred Davis, MD, one of the study’s lead authors from ProCare, said in a press release.

ProCare has been conducting additional work to create an algorithm to be incorporated into EHRs using the variables found in this study to support clinical decision-making and to help in the diagnosis of fibromyalgia. Of note, these artificial intelligence tools can be applied to other EHR fields toward better patient care.