Lindsey C. McKernan
Patients with fibromyalgia without any known suicide attempts had spent up to 40-fold more time with providers annually than those who did attempt suicide, according to data published in Arthritis Care & Research.
The researchers determined this while testing a model based on machine learning that was able to reliably predict both suicide ideation and attempts among patients with fibromyalgia, they wrote.
“Fibromyalgia takes a great deal to manage on a week-to-week basis, with patients doing what they can to take care of their pain through self-management techniques at home, along with other aspects of treatment such as getting the recommended exercise, engaging with physical therapy, attending to mental health needs, and maintaining contact with specialists and primary care physicians,” Lindsey C. McKernan, PhD, of Vanderbilt University Medical Center, told Healio Rheumatology. “Putting it all together, it takes lot of contact and time, and may provide different avenues we can help people connect to who may be less engaged.”
Patients with fibromyalgia without any known suicide attempts had spent up to 40-fold more time with providers annually than those who did attempt suicide, according to data.
To test whether a previously published model could be used to detect both suicide ideation and attempts among patients with fibromyalgia, and to identify risk and protective factors unique to the disease, McKernan and colleagues conducted a case-control study using largescale electronic health record data at the medical center from 1998 to 2017. According to the researchers, EHR data included diagnoses, demographics, clinical text, laboratory values and more. In all, 8,879 patients with fibromyalgia who had more than three visits to the medical center during the previous 6 months were included in the analysis.
The model, which is based on an algorithm intended to predict the risk for attempted suicide, was developed using 3,250 cases of documented self-injury, compared with 12,695 control patients from the medical center’s general population. It had previously achieved an internal validation c-statistic as high as 0.92 to predict suicide attempt in 7 days. McKernan and colleagues measured the model’s performance using sensitivity, specificity, area under the receiver operating characteristic (AUC), and calibration plots. In addition, the researchers used L1penalized regression with bootstrapping to select risk factors.
According to the researchers, there were 34 known suicide attempts, and 96 documented cases of suicidal ideation among the 8,879 patients included in the study. The researchers determined the external validity to be good for both suicidal ideation (AUC = 0.8) and attempts (AUC = 0.82), with excellent calibration. Risk factors for suicidal ideation among the patients included fatigue (OR = 1.29; 95%CI, 1.251.32), dizziness (OR = 1.25; 95% CI, 1.221.28) and weakness (OR = 1.17; 95%CI, 1.151.19). Risk factors for suicide attempt included obesity (OR = 1.18; 95% CI, 1.11.27) and drug dependence (OR = 1.15; 95% CI, 1.121.18).
The researchers also found that patients with fibromyalgia and no suicidal ideation spent 3.5 times more time in followup annually. In addition, those without documented suicide attempts spent more than 40 times more time in facetoface visits with providers per year.
“This may suggest a path that we can intervene with patients — finding those who may not be as engaged with their providers, and getting established, or re-connecting with services,” McKernan said. “It is very important to develop relationships with providers, especially in managing a chronic condition like fibromyalgia. This could mean getting established with mental health services, or providing assistance with other aspects of care as well.” – by Jason Laday
Disclosure: McKernan reports no relevant financial disclosures. Please see the study for all other authors’ relevant financial disclosures.