Discriminating between Anxious and Non-Anxious Subjects Using the Toronto Hospital Alertness Test

Abstract

BackgroundAlertness designates the internal feeling of wakefulness or arousal and is often described to be linked to the level of anxiety. An adequate level of anxiety favoring the alertness needed to deal with a faced specific situation efficiently; too much anxiety can result in failure to process information and respond appropriately. Thus, it would be of interest to verify if different alertness profiles can be observed depending on anxiety level. The Toronto Hospital Alertness Test (THAT) is a test designed to measure alertness. The present survey’s aim is to verify if the THAT allows observing different alertness profile between self-described anxious and non-anxious subjects.MethodsSubjects >18 years were selected from online databases in three countries (Canada, USA, and UK). All respondents filled in a Hospital Anxiety Depression Scale questionnaire, and only those self-classified as anxious or non-anxious (HAD-A ≥11 or ≤7, respectively) took part to the survey and were asked to complete the THAT.ResultsAmong 616 respondents retained in the survey, 414 were self-assessed as anxious and 202 as non-anxious. The mean THAT score for anxious and non-anxious subjects was 21.4 and 38.9, respectively. A receiver operating characteristic (ROC) curve of THAT scores indicated that a threshold score of 30 was required to achieve good sensitivity (86.7%) and specificity (88.6%), with good discriminatory power [an area under the curve (AUC) of 0.938]. As age was determined to be a potential confounder, subjects were age-matched giving a ROC with an AUC of 0.931, with good sensitivity (88.5%) and specificity (89.3%), and the threshold remaining at 30. The internal reliability of THAT in anxious subjects was good (Cronbach alpha = 0.84).LimitationsNo independent verification of anxious or non-anxious status or other eligibility criteria was done.ConclusionThe alertness profiles of self-defined anxious and non-anxious subjects observed on THAT are different. Based on a subject’s alertness profile, it is possible to discriminate between self-defined anxious and non-anxious, using THAT, with good specificity and sensitivity at a threshold score of 30.

Psychometric evaluation of the Chinese version of the Toronto Hospital Alertness Test

“…However, the last two items, items 9 and 10, had low factor loadings that reflect their low association with the other items. This is consistent with a previous Canadian study [16]. This may be due to the questionnaire format, as the last two items are negatively worded, whereas the others are positively worded [16].…”

“…This is consistent with a previous Canadian study [16]. This may be due to the questionnaire format, as the last two items are negatively worded, whereas the others are positively worded [16]. Individuals usually tend to agree more with positively worded items than negatively worded items [38].…”

“…Acceptable convergent validity was also shown, with a significant correlation with ZOGIM-A (r = 0.37, P < 0.01) [6]. Although THAT is relatively new, it has already been used to evaluate the effect of interventions [20,21] and also could distinguish selfreported anxious from non-anxious patients [16].…”

“…Impaired alertness may be associated with fatigue, low energy, drowsiness, reduced attention, and decreased concentration [9]. These symptoms and feelings can induce automobile accidents [15], psychological symptoms [16], physical symptoms [17], and decreased quality of life [9], as well as the higher likelihood of disability and increased risk of mortality [18]. In adolescents, daytime impaired alertness may induce depressed mood and impact academic performance, while depressed mood can reduce academic performance further [19].…”

“…A standardized measurement tool specific for alertness is needed to guide treatment and facilitate the development of psychopathology [6,16]. Two instruments have been developed for assessing alertness, namely the Toronto Hospital Alertness Test (THAT) and the ZOGIM Alertness (ZOGIM-A) Scale [6].…”