Innovations In Clinical Neuroscience

NOV-DEC 2017

A peer-reviewed, evidence-based journal for clinicians in the field of neuroscience

Issue link:

Contents of this Issue


Page 48 of 83

49 ICNS INNOVATIONS IN CLINICAL NEUROSCIENCE November-December 2017 • Volume 14 • Number 11–12 O R I G I N A L R E S E A R C H Relieved and non-Relieved subjects was 27 percent. When instead setting the threshold for Relief symptoms at 3 or less, similar to the Remission criteria, roughly 37 percent of all subjects were considered to satisfy criteria for Relief. The overlap coefficient was 32 percent. However, only one in three of these subjects was in the Relief range for the latent trait θ. Although there were similar percentages of subjects flagged with the Remission criteria, the subjects identified by Relief and Remission were different. The Cohen's kappa for the Relief criteria and Remission criteria when using the same threshold was 0.69, as shown in Table 4. When using different thresholds (Relief ≤2), the Cohen's kappa value was 0.23. Collectively, this suggests that these item subsets differ substantially in the type of patients they identify. The Relief item subset can be set more liberally to include more patients, but the patients that the Relief criteria did identify were, on average, still less ill in general severity than were the individuals identified using Remission criteria, with greater separation between those who did and did not meet the criteria. DISCUSSION The overarching objectives of this research were to use a bifactor model, both factor FIGURE 6. The distributions of the remitted and non-remitted patients showed a large overlap (43.5%) in general severity. The Relief criteria had less overlap between patients who did and did not qualify, for all thresholds assessed. TABLE 4. Relief and remission criteria THRESHOLD KAPPA (RELIEF CRITERIA, REMISSION CRITERIA) % OF SUBJECTS MEETING CRITERIA % OF SUBJECTS WHO MEET CRITERIA WITH SYMPTOMS ABOVE -1 OVERLAP COEFFICIENT FOR SUBJECTS WHO DO AND DO NOT MEET CRITERIA MEAN THETA OF SUBJECTS WHO MEET CRITERIA SD THETA OF SUBJECTS WHO MEET CRITERIA MEAN THETA OF SUBJECTS WHO DO NOT MEET CRITERIA SD THETA OF SUBJECTS WHO DO NOT MEET CRITERIA Relief: all items ≤2 0.23 0.08 0.19 0.27 -1.71 0.78 0.15 0.91 Relief: 5 of 8 items ≤2 0.51 0.31 0.55 0.32 -1.03 0.72 0.46 0.79 Relief: all items ≤3 0.69 0.37 0.66 0.40 -0.84 0.75 0.50 0.84 Remission criteria - 0.47 0.66 0.44 -0.80 0.77 0.47 0.86 NOTE: When setting the Relief and Remission criteria at the same threshold (≤3), the subjects identified by both measures had an overlap value of kappa=0.69. The Relief criteria had a less ill group than the Remitted group, with less overlap between subjects who did and did not meet criteria. SD: standard deviation

Articles in this issue

Archives of this issue

view archives of Innovations In Clinical Neuroscience - NOV-DEC 2017