Physician documented fatigue among adults and children with inflammatory bowel disease: a nationwide routinely collected health data study from the epi-IIRN
Gili Focht, Rachel Buchuk, Menachem Sokolik, Shlomi Cohen, Henit Yanai, Amir Ben-Tov, Galia Zacay, Eran Matz, Shir Zekaria, Iris Dotan, Ora Paltiel, Dan TurnerAbstract
Background
We aimed to assess the rates of physician-documented fatigue according to diagnostic codes among patients with inflammatory bowel diseases (IBDs) in a nationwide cohort and to identify clinical factors associated with its documentation.
Methods
We used the nationwide epi- Israeli IBD Research Nucleus cohort derived from 4 HMOs covering approximately 98% of the Israeli population, matched to non-IBD controls. Physician-documented fatigue was defined using routinely collected ICD9 and textual codes and explored in association with demographic, laboratory-derived disease severity clusters, comorbidities, medications, healthcare utilization, and sickness absence. Predictors for fatigue documentation were explored using multivariable Cox models incorporating time-varying covariates.
Results
Fatigue was documented in 16.4% of the 31 945 patients with IBD and 11.0% of the 73 106 matched controls (odds ratio [OR] = 1.60, 95% confidence interval [CI] 1.54-1.66). Fatigue was more prevalent in IBD than matched controls across age groups and comorbidity strata. Among patients with IBD, there was no difference in fatigue rate between those with Crohn’s disease and ulcerative colitis. Time-to-fatigue was shorter among females (P < .001). Across adult males and females and pediatric boys and girls, anemia, anxiety, and ≥3 healthcare contacts were independently associated with higher rates of fatigue documentation, with the strongest associations observed for frequent healthcare contact (HRs ranging from 4.21 [95% CI 1.49-11.70] to 10.22 [95% CI 7.47-13.90]). Longer disease duration was consistently associated with a lower hazard, while additional subgroup-specific associations were observed for sleep disorders, steroid dependency, and IBD-related medications.
Conclusions
Fatigue is more frequently documented among patients with IBD than matched non-IBD controls, especially in females. Several risk factors were associated with fatigue which should be considered in clinical practice. The lower frequency of documented fatigue compared with patient-reported estimates suggests that routine clinical coding captures only part of the overall fatigue burden.