A self-administered questionnaire of ulnar neuropathy at the elbow

We report a new self-administered questionnaire for assessment of symptom severity of ulnar neuropathy at the elbow (UNE). The new UNE and Levine's questionnaires were administered to a sample of UNE subjects and for comparison also to a sample of subjects with carpal tunnel syndrome (CTS). We enrolled 89 consecutive patients (32 women, 57 men, mean age 52.3 years) with UNE and 203 consecutive patients (157 women and 46 men, mean age 53.7 years) with CTS. The protocol of the study consisted in self-administration of the new UNE and Levine's questionnaires, as well as scoring of clinical and electrophysiological severity of entrapment syndromes with ordinal scales. The UNE questionnaire (UNEQ) includes nine questions and considers numbness and tingling in the fourth and fifth fingers, elbow pain and modification of pain and paraesthesia with elbow position. A score from 1 (absence of symptom) to 5 (most severe) is assigned for each question. The overall score is calculated as the mean of the nine scores. Test–retest reliability, internal consistency and validity were assessed. Responsiveness was also tested in a sample of patients undergoing conservative treatment. The UNEQ was reproducible. Spearman's correlation coefficient between scores at successive observations (test–retest reliability), assessed in the first 44 patients, was 0.97 and Cohen coefficients κ for single items were between 0.64 and 0.81. Internal consistency was high: Cronbach's α, which summarises interitem correlations among all items of UNEQ, was 0.87. Validity was demonstrated by a direct correlation with UNE clinical and electrophysiological severity scores (0.65 and 0.35). On the contrary, Spearman's correlation coefficients between UNEQ and clinical and electrophysiological CTS severity scores were low (0.11 and 0.02, respectively). Responsiveness was calculated at 6–8 months follow-up in 25 cases. The effect size was 0.46. The Wilcoxon rank-test showed significant improvement between basal and follow-up UNEQ scores (Z=–2.39, p=0.017), but not Boston Questionnaire scores. There was also significant correlation between UNEQ changes and an arbitrary scale of patient satisfaction at follow-up (r=0.85, p<0.001). The UNEQ is reproducible, internally consistent and valid. Although further studies are required to test its responsiveness to clinical changes, UNEQ may be also considered responsive. UNEQ can be used to measure subjective discomfort in UNE patients.