The first Qmci subtest, orientation, asks five questions and includes tests of orientation in time (What year, month, day, and date?) and place (What country?). It is more heavily weighted towards orientation in time, which is useful in identifying those who warrant more detailed assessment [7] and as a predictor of overall cognitive decline when compared to questions testing orientation to place [8]. Two points are given for the correct answer, one point for wrong answers and zero points for no answer or a conceptually unrelated answer. The timing allows for a maximum of 10 s for each answer to a total time of 30 s. The maximum score is 10 points. Compared with the ABCS 135, the weighting of this subtest was reduced by a factor of 2.5 (from 25 to 10 points) and it now represents just 10 % of the total score i.e. 10 points from a total of 100. Orientation is a poor predictor of MCI with significant ceiling effects [2, 6, 9], and was retained to prevent floor effects so as to allow the instrument to monitor progression in advancing cognitive impairment.

The second subtest is word registration. It is composed of five items to be repeated back immediately. Three validated alternative word sets are provided [4]. One point is scored for each word recalled after the first reading. If a subject recalls all five, the five items are repeated once before proceeding to the next subtest. If a subject does not repeat all five, the five items are repeated until the subject correctly recalls all items or for a maximum of three trials. The second and third trials do not count towards the score and are there to help the person learn in preparation for the delayed recall subtest. Ten seconds are allowed for recall. The maximum score is five points. Following analysis of the ABCS 135 subtests, registration was reduced by a factor of 5, from, 25 to 5 points.

The third Qmci subtest is a 1-min clock drawing test (CDT). Clock drawing is a popular short screening test for dementia, in both community [10] and hospital settings [11], and can be scored reliably by both trained and untrained raters [12]. The CDT is a moderately sensitive and specific CSI in its own right (see Chap. 5). The CDT assesses several cognitive domains including visuospatial [13, 14] and executive function [15, 16]. There are several methods of scoring the CDT [15]. The Qmci CDT scoring method, based on the technique developed for the ABCS 135, has relatively complex scoring instructions compared to other short CSI that also incorporate the CDT [17, 18]. The Qmci CDT scoring instructions are reliable and valid compared to other scoring techniques [19]. Indeed, the increased complexity arguably increases the utility of the subtest [3].

To accommodate the CDT within the Qmci, its scoring structure was reduced, by a factor of 2 from 30 points in the ABCS 135 to a new maximum total of 15 points, and the scoring instructions simplified. A blank circle or ‘clock face’ and transparent scoring template, to be placed over the circle of the completed clock, were provided with the ABCS 135. To simplify scoring for the Qmci, new instructions were developed. The subject is still provided with the blank ‘clock face’, found on the reverse of the two-sided scoring sheet, instructed to ‘use the circle provided over page to draw a clock face’ and to set the time to ‘ten past eleven’. One point is given for each number (1–12), for each hand and for the pivot correctly placed at or close to their ideal location (as denoted on the visual scoring aid accompanying the blank clock face e.g. one point is given for each hand placed between the dashed lines). A single point is lost for each number duplicated or greater than 12, e.g. a 15 or 45, i.e. errors. This provides a total of 15 points. The subject is allowed 1 min.

The fourth subtest, five-word delayed recall, tests episodic memory and is also valid as a stand-alone test in dementia [20, 21]. Episodic memory loss occurs early in most dementia subtypes. The Qmci’s delayed recall task is based on the five words used in the registration subtest with the CDT functioning as an interval distractor task. The subject is asked to remember the five words, which may be recalled in any order. The Qmci’s delayed recall subtest is timed at 30 s with a maximum score of 20 points. Five-word delayed recall adds to the sensitivity of CSIs for MCI, particularly amnestic MCI and is associated with hippocampal atrophy and burden of neurofibrillary tangles in patients with Alzheimer’s pathology [22].

The fifth subtest assesses verbal fluency. Verbal fluency facilitates memory retrieval and can be presented as categorical (i.e. semantic, e.g. naming of animals within 1 min) or letter (i.e. phonemic, e.g. naming of words beginning with a designated letter) fluency. Tests of verbal fluency also involve executive control [23]. In the Qmci, categorical fluency is assessed with subjects requested to name as many words as possible relating to a named category within 60 s. A half a point is given for each word named to a maximum of 40 words. The final score is rounded up. Words with different suffixes are not counted twice (e.g. fish/fishes, mouse/mice, etc.) but alternate species (e.g. blue jay, robin, sparrow, duck, etc.) are accepted. Alternate validated forms include animals, fruits and vegetables, and cities and towns [4]. The maximum score is 20 points. Compared to the ABCS 135 verbal fluency had its total score reduced from 30 to 20 points, although its overall weighting increased. Patients with Alzheimer’s dementia perform less well with categorical fluency than letter fluency, which influenced the decision to include this type of verbal fluency testing within the Qmci [24], though both types are abnormal in MCI [25, 26].

The sixth and final subtest is logical memory, a linguistic test of episodic memory consisting of immediate verbal recall of a short story [27]. Logical memory is a highly sensitive and specific test to differentiate normal cognition from MCI [4] and is relatively unaffected by age or education [28]. For the Qmci version, logical memory is tested using a short story consisting of four sentences which, though not directly connected, provide a coherent ‘logical’ story. Two points are given for each correct word item recalled verbatim. Only bolded words within each section of the short story need be recalled to score two points. Otherwise the subject scores zero for that word. Each story includes 15 bolded words to provide a maximum score of 30 points. Although no paraphrasing is allowed, recall may be in any order. In total, 30 s are allowed for administration and 30 s for response. Again validated alternatives are available [4].

The Qmci, like the ABCS 135, was originally developed in a Canadian population. The index validation compared the Qmci with its predecessor, the ABCS 135, and the Standardized Mini-Mental State Examination (SMMSE) [29, 30] in 965 patients and their caregivers (normal controls) attending four memory clinics in Ontario, Canada [1]. The study showed that the Qmci has greater accuracy in differentiating MCI from normal controls than the SMMSE with an area under the receiver operating characteristic (ROC) curve (AUC) of 0.86 versus 0.67 (p < 0.001) respectively [1]. It also showed that the Qmci has greater accuracy than the ABCS 135 (AUC of 0.83, p = 0.05), while all three instruments accurately separated MCI from dementia including mild dementia when this was separated out from those with moderate to severe stage disease [1]. Tables 12.2 and 12.3 present the characteristics of studies validating the Qmci and the psychometric properties demonstrated by the instrument in each study, respectively.