(723 days)
PKQ
Not Found
No
The description focuses on presenting visual stimuli and scoring based on timing and accuracy, with no mention of AI or ML in the device's operation or data analysis beyond standard statistical methods for determining cut-off values and agreement.
No
The device is described as an "adjunctive tool for evaluating perceptual and memory function" and a "screening aid in the assessment of adults" for identifying cognitive decline, referring for further testing, and monitoring changes. Its purpose is diagnostic/assessment, not therapeutic.
Yes
The device is indicated as an "adjunctive tool for evaluating perceptual and memory function" and provides "objective measurements of cognitive function as a screening aid in the assessment of adults...for the purpose of identifying a potential decline in cognitive function," which directly describes a diagnostic purpose.
No
The device description explicitly mentions a "rotatory mouse manipulandum" used by the patient to interact with the visual stimuli. This indicates a hardware component is part of the device, making it not software-only.
Based on the provided information, this device is not an In Vitro Diagnostic (IVD).
Here's why:
- IVDs analyze specimens from the human body. The description of Cognivue clearly states that it presents visual stimuli and the patient interacts with a cursor using a rotatory mouse. It measures cognitive function based on the timing and accuracy of responses to these visual stimuli. There is no mention of collecting or analyzing any biological specimens (like blood, urine, tissue, etc.).
- The intended use and device description focus on behavioral and perceptual responses. The device is described as an "adjunctive tool for evaluating perceptual and memory function" and provides "objective measurements of cognitive function." This is done through interaction with visual stimuli, not through the analysis of biological samples.
Therefore, while Cognivue is a medical device used in a healthcare setting for diagnostic purposes (screening for cognitive decline), it does not meet the definition of an In Vitro Diagnostic device.
N/A
Intended Use / Indications for Use
Cognivue testing is indicated as an adjunctive tool for evaluating perceptual and memory function in individuals aged 55-95 years old.
Product codes
PKQ
Device Description
Cognivue provides clinicians in a healthcare setting with objective measurements of cognitive function as a screening aid in the assessment of adults 55 years of age and older. This is done for the purpose of identifying a potential decline in cognitive function relative to baseline test performance of other age-normal adults, referring those adults for further testing where warranted, and monitoring changes in cognitive function over time.
Cognivue presents a series of visual stimuli and a wedge-shaped cursor. The display presents stimuli at varying signal strengths with a moving domain-specific target. The patient identifies the target location by moving the cursor using the rotatory mouse manipulandum. The patient is scored based on the timing and accuracy of the responses. Repeated trials of varying difficulty characterize the subject's performance in each of the tested functional domains.
Mentions image processing
Not Found
Mentions AI, DNN, or ML
Not Found
Input Imaging Modality
Not Found
Anatomical Site
Not Found
Indicated Patient Age Range
55-95 years old.
Intended User / Care Setting
The Cognivue is intended be used by medical professionals qualified to interpret the results of a cognitive assessment examination.
The device is not intended to be used as a stand-alone diagnostic device.
The device is not intended to identify the presence or absence of clinical diagnoses.
Description of the training set, sample size, data source, and annotation protocol
Not Found
Description of the test set, sample size, data source, and annotation protocol
The sponsor conducted two separate studies. The first study was conducted to determine the cut-off values for Cognivue (e.g., impaired, intermediate, and unimpaired cognitive function) by comparing the performance of Cognivue against a reference standard, the St. Louis University Mental Status (SLUMS) Examination. The data obtained from 92 subjects were utilized to determine the cut-off data.
The second study was a clinical validation study which examined the agreement between the Cognivue classifications and the SLUMS classifications. The clinical validation study also examined the test - retest reliability study of Cognivue, and the determination of the construct validity of Cognivue via comparison with traditional paper and pencil neuropsychological tests. 401 subjects were enrolled at 13 older adult independent living communities. These sites consist of senior independent living facilities and an adult program for independent living seniors who are living at home. These sites enroll clients 55-95 years of age and who are cared for by primary care practitioners. Subjects were stratified in to three roughly equal groups, based on SLUMS scores: Normal, Mildly impaired, Impaired.
Summary of Performance Studies (study type, sample size, AUC, MRMC, standalone performance, key results)
Study Type: Comparative study, using a paired design.
Sample Size:
- Cut-off determination: 92 subjects.
- Clinical Validation and Agreement Analysis: 401 subjects (first test session).
- Test-Retest Reliability: 358 subjects (from the 401, due to some unavailable for repeat testing or deadline constraints).
Key Results:
Cognivue Cut-Offs Analysis (based on 92 subjects): - Cognivue scores ≤50 are consistent with a conservative standard for impairment.
- Cognivue scores ≥75 are consistent with a conservative standard for un-impairment.
- Scores between 50 and 75 represent an intermediate result that has no significance. The original SLUMS cut-offs were: >26 unimpaired, 74 is unimpaired, 50-75 intermediate.
- SLUMS classifications based on education.
- PPA (including intermediate as unimpaired for calculation) = 56% (95% Wilson interval: 0.47 to 0.65)
- NPA (including intermediate as unimpaired for calculation) = 93% (95% Wilson interval: 0.91 to 0.97)
- Quadratic weighted kappa score = 0.57 (Wald 95% CI 0.50 to 0.63), suggesting a clear categorical relationship.
- Classification agreement omitting all intermediate groups:
- PPA = 82% (95% Wilson interval: 0.72 to 0.89)
- NPA = 98% (95% Wilson interval: 0.93 to 0.99)
- Regression Analysis: Cognivue = -29.98 + 4.465 SLUMS (r = 0.71)
Analysis of the Test-Retest Reliability (based on 358 subjects):
-
Cognivue Test-Retest Regression Analysis:
- Cognivue 2 = 9.06 + 0.936 Cognivue 1 (r = 0.90), demonstrating substantial and significant agreement.
- PPA for impaired = 89% (95% Wilson interval: 0.77 to 0.95)
- NPA for all other categories = 93% (95% Wilson interval: 0.90 to 0.96)
- Percent agreement for impaired: 89%
- Percent agreement for mildly impaired: 57%
- Percent agreement for unimpaired: 87%
- Fisher-Freeman-Halton exact test (SPSS v22) yields 3.62 (df = 2), p = 0.161. Cognivue classifications for impairment, mild impairment, and unimpairment do not differ significantly by the time of the test.
-
SLUMS Test-Retest Regression Analysis:
- SLUMS 2 = 4.15 + 0.90 SLUMS 1 (r = 0.82), demonstrating substantial and significant agreement.
- PPA for impaired = 87% (95% Wilson interval: 0.76 to 0.94)
- NPA for all other categories = 87% (95% Wilson interval: 0.83 to 0.90)
- Fisher-Freeman-Halton exact test (SPSS v22) yields 24.89 (df = 2), p = 21; Cognivue >= 50):
- Negative Percent Agreement (NPA) = 50%
- Positive Percent Agreement (PPA) = 86%
- Accuracy (ACC) = 86%
- Error Symmetry = 0.01
-
Unimpaired cut-off (SLUMS >= 27; Cognivue >= 75):
- Negative Percent Agreement (NPA) = 67%
- Positive Percent Agreement (PPA) = 66%
- Accuracy (ACC) = 66%
- Error Symmetry = 0.05
Agreement analysis between SLUMS and Cognivue (401 subjects, including intermediate):
- PPA = 56% (95% Wilson interval: 0.47 to 0.65)
- NPA = 93% (95% Wilson interval: 0.91 to 0.97)
Agreement analysis between SLUMS and Cognivue (401 subjects, omitting intermediate):
- PPA = 82% (95% Wilson interval: 0.72 to 0.89)
- NPA = 98% (95% Wilson interval: 0.93 to 0.99)
Test-retest reliability of Cognivue (358 subjects):
- PPA (Impaired category) = 89% (95% Wilson interval: 0.77 to 0.95)
- NPA (combined intermediate, unimpaired with respect to impaired) = 93% (95% Wilson interval: 0.90 to 0.96)
- Percent agreement for impaired: 89%
- Percent agreement for mildly impaired: 57%
- Percent agreement for unimpaired: 87%
Test-retest reliability of SLUMS (358 subjects):
- PPA (Impaired category) = 87% (95% Wilson interval: 0.76 to 0.94)
- NPA (combined intermediate, unimpaired with respect to impaired) = 87% (95% Wilson interval: 0.83 to 0.90)
- Percent agreement for impaired: 87%
- Percent agreement for mildly impaired: 55%
- Percent agreement for unimpaired: 90%
Predicate Device(s)
Not Found
Reference Device(s)
Not Found
Predetermined Change Control Plan (PCCP) - All Relevant Information
Not Found
§ 882.1470 Computerized cognitive assessment aid.
(a)
Identification. The computerized cognitive assessment aid is a prescription device that uses an individual's score(s) on a battery of cognitive tasks to provide an interpretation of the current level of cognitive function. The computerized cognitive assessment aid is used only as an assessment aid to determine level of cognitive functioning for which there exists other valid methods of cognitive assessment and does not identify the presence or absence of clinical diagnoses. The computerized cognitive assessment aid is not intended as a stand-alone or adjunctive diagnostic device.(b)
Classification. Class II (special controls). Except when the computerized cognitive assessment aid is intended for diagnostic assessment of specific diseases or conditions and relies on inputs from visual cues, auditory cues, and/or functional use of the hand, the device is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to the limitations in § 882.9. The special control(s) for this device are:(1) The technical parameters of the device's hardware and software must be fully characterized and be accompanied by appropriate non-clinical testing:
(i) Hardware specifications must be provided. Appropriate verification, validation, and hazard analysis must be performed.
(ii) Software, including any proprietary algorithm(s) used by the device to arrive at its interpretation of the patient's cognitive function, must be described in detail in the software requirements specification (SRS) and software design specification (SDS). Appropriate software verification, validation, and hazard analysis must be performed.
(2) The device must be designed and tested for electrical safety.
(3) The labeling must include:
(i) A summary of any testing conducted to demonstrate how the device functions as an interpretation of the current level of cognitive function. The summary of testing must include the following, if available: Any expected or observed adverse events and complications; any performance measurements including sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) per the devices intended use; a description of the repeatability of measurements; a description of how the cut-off values for categorization of measurements were determined; and a description of the construct validity of the device.
(ii) A warning that the device does not identify the presence or absence of clinical diagnoses.
(iii) A warning that the device is not a stand-alone diagnostic.
(iv) The intended use population and the intended use environment.
(v) Any instructions technicians must convey to patients regarding the administration of the test and collection of cognitive test data.
0
DE NOVO CLASSIFICATION REQUEST FOR COGNIVUE
REGULATORY INFORMATION
FDA identifies this generic type of device as:
Computerized Cognitive Assessment Aid. The Computerized Cognitive Assessment Aid is a prescription device that uses an individual's score(s) on a battery of cognitive tasks to provide an interpretation of the current level of cognitive function. The Computerized Cognitive Assessment Aid is used only as an assessment aid to determine level of cognitive functioning for which there exists other valid methods of cognitive assessment and does not identify the presence or absence of clinical diagnoses. The Computerized Cognitive Assessment Aid is not intended as a stand-alone or adjunctive diagnostic device.
NEW REGULATION NUMBER: 882.1470
CLASSIFICATION: CLASS II
PRODUCT CODE: PKQ
BACKGROUND
DEVICE NAME: COGNIVUE
SUBMISSION NUMBER: DEN130033
DATE OF DE NOVO: JUNE 24, 2013
- CONTACT: CEREBRAL ASSESSMENT SYSTEMS, INC. CHARLES J. DUFFY, MD, PHD, FOUNDER AND CEO 2850 Clover Street PITTSFORD, NY 14534
REQUESTER'S RECOMMENDED CLASSIFICATION: CLASS II
INDICATIONS FOR USE
Cognivue testing is indicated as an adjunctive tool for evaluating perceptual and memory function in individuals aged 55-95 years old.
LIMITATIONS
For prescription use only.
The safety and effectiveness of the Cognivue for individuals with less than 12 years of education has not been established.
1
The Cognivue is intended be used by medical professionals qualified to interpret the results of a cognitive assessment examination.
The device is not intended to be used as a stand-alone diagnostic device.
The device is not intended to identify the presence or absence of clinical diagnoses.
DEVICE DESCRIPTION
Cognivue provides clinicians in a healthcare setting with objective measurements of cognitive function as a screening aid in the assessment of adults 55 years of age and older. This is done for the purpose of identifying a potential decline in cognitive function relative to baseline test performance of other age-normal adults, referring those adults for further testing where warranted, and monitoring changes in cognitive function over time.
Cognivue presents a series of visual stimuli and a wedge-shaped cursor. The display presents stimuli at varying signal strengths with a moving domain-specific target. The patient identifies the target location by moving the cursor using the rotatory mouse manipulandum. The patient is scored based on the timing and accuracy of the responses. Repeated trials of varying difficulty characterize the subject's performance in each of the tested functional domains.
Cognivue Hardware Components:
- Commercially available 4-wheel cart with seated height table top, slide-● out keyboard tray, upper shelf, and lower shelf
- . Plastic sheet connected to cart upright posts to enclose table-top testing area
- . Commercially available printer adhered to upper shelf
- Commercially available mini-pc adhered to upper shelf ●
- Commercially available LCD flat screen display adhered to table top .
- . Rotatory mouse manipulandum adhered to cart table top for patient input
- Commercially available, wet/dry keyboard with integrated touch pad, located on slide-out keyboard tray.
- Commercially available, power supply adhered to lower shelf ●
- Commercially available power and interconnection cables
Cognivue Software Components:
- Software to support the graphical user interface (GUI) ●
- Software to interface with the rotatory mouse manipulandum ●
- . Software to administer a battery of motor, visual, perceptual and memory tests (an overview of which is provided below)
- . Software to support real-time test report generation, printing, and archiving
2
The Cognivue device computer, monitor, rotatory manipulandum, printer and mouse/keyboard are built from commercially available components and do not incorporate any unique materials other than those typical of desktop computers (Figure 1). The Cognivue cart is built out of steel, with four attached casters to provide stable mobility, and two shelves made of industrial-grade, laminated particle board. The monitor shroud is fabricated from a rigid non-translucent material secured to the cart. The Cognivue device cart, mini-pc, display screen, and printer do not incorporate any unique or proprietary materials. The Cognivue software is pre-installed.
Image /page/2/Figure/1 description: This image shows a medical computer cart with various components and dimensions labeled. The cart includes a monitor, mini PC, printer, keyboard with touchpad, and a rotary mouse manipulandum. The dimensions of the cart are also labeled, including a height of 28 inches from the floor to the keyboard, 20 inches from the keyboard to the monitor, and 18.5 inches from the monitor to the top of the cart.
Figure 1. Cognivue Components
CAS's Cognivue technology enables the automated characterization of aspects of the perceptual and memory functions linked to human cortical information processing. Cognivue uses software running on a personal computer. The results are used to detect brain functional impairments in older adults (ages 55-95). This is accomplished by recording the subject's ability to respond to ten variations of visual stimuli. In all ten tests, the patient tracks the movement of a cursor-width target wedge (pie slice depicted in Figure 2) in a circular display by using a hand controller (rotatory manipulandum) to place the green cursor over the target wedge.
The CAS Cognivue test is rapid (~10 mins), extensive (including many brain functional domains), and non-invasive (patient contact only is made by the manipulandum).
3
Figure 2. Target Wedge Screen Example
Image /page/3/Picture/1 description: The image shows a gray ring on a black background. A section of the ring is filled with a pattern of black and white circles of varying sizes. The filled section is outlined in green, highlighting it as a distinct area within the ring.
Cognivue presents ten brain function tests in ten minutes. The battery is organized into three sub-batteries, with each sub-test preceded by transitional guidance that facilitates the test subject's engagement with minimal supervision:
Transitional guidance and textual instructions are presented at the start of each subtest, as described below:
- Sub-Battery A: Two tests that measure adaptive motor control and dynamic visual contrast sensitivity, the results being used to adjust the remaining sub-tests to the basic response characteristics of the subject.
- o Adaptive Motor Control: Patient must follow a series of dots as they move around the circle and change direction and speed.
- Visual Saliency Testing: Dots from previous test begin to fade (decreasing o contrast) as the subject tracks them around the circle, while changing speed and direction.
- Sub-Battery B: Four perceptual tests measuring letter, word, shape, and motion ● processing ability:
- o Letter discrimination. Subject must correctly identify a real English letter.
- Word discrimination. Subject must correctly identity a real English word. O
- Shape discrimination. Subject must correctly identify which shape is O different.
- Motion discrimination. Subject must identify which dot pattern is о different.
- Sub-Battery C: Four memory tests measuring: ●
- Letter memory. Subject must remember what letter was shown as a pre- O cue, and identify that letter when presented in a circle populated with other letters.
- Word memory. Subject must remember what word was shown as a pre- O cue, and identify that word when presented in a circle populated with other words.
- Shape memory. Subject must remember what shape was shown as a pre- O cue, and identify that shape when presented in a circle populated with other shapes.
4
- Motion memory. Subject must remember what direction of motion was o shown as a pre-cue, and identify that direction when presented in a circle populated with other directions.
Cognivue was validated against the St. Louis Mental Status (SLUMS) examination. The SLUMS is a cognitive screening tool which detects early cognitive problems, specifically mild cognitive impairment that is often missed by other mental status screening tools. The SLUMS is a psychometric scale with validated cut-off scores based upon the subject's education level. SLUMS scores are classified as: If subject is a high school graduate: >26 is unimpaired, 24 is unimpaired, 75 is consistent with being unimpaired.
4 errors in reading the introductory paragraph of the consent form aloud while holding the text at the most comfortable viewing distance (typically 12-18", printed in Gulim sans serif, size 14, character spacing expanded by 1 point and line spacing printed at 1.5) while wearing their customary corrective lenses. The text is as follows:
"This consent form describes what to expect if you decide to take part in this study. Please listen carefully as the consent form is read to you. Ask any questions you have before deciding to take part. We suggest that you use the bathroom before the testing begins."
- iv. Inability to answer three questions about testing after consent is read to the subject:
-
- How long does each session take? (Correct response is fewer than 90 mins per session.)
-
- How many sessions are there? (Correct response is a total of two sessions.)
-
- How much will you be getting paid? (Correct response is twenty dollars per session.)
-
9
8. Clinical Performance Results.
- a. Cognivue Cut-Offs Analysis. Scores on Cognivue and the SLUMS were compared in a sample of assisted and independent living adults. ages 55-95 years old, to find Cognivue scores that yield correspondence with SLUMS patient classification as: impaired, unimpaired, and intermediate. The data obtained from 92 subjects were utilized to determine the cut-off data. SLUMS scores ranged from 0 - 30 and classified as: >26 unimpaired, =27), the sponsor obtained the cut-off values for Cognivue scores that minimize the objective function. In the second method, the sponsor repeated the same procedure but optimized another objective function, where 1-ACC is inaccuracy, (1- [TP + TN / total) and BGI is error bias ( contrast ratio [ difference/sum] of FPs & FNs.
The results from these analyses indicated that Cognivue scores ≤50 are consistent with a conservative standard for impairment and that Cognivue scores ≥75 are consistent with a conservative standard for un-impairment. Scores between 50 and 75 represent an intermediate result that has no significance.
Image /page/9/Figure/4 description: The image contains a confusion matrix and a scatter plot. The confusion matrix shows the relationship between SLUMS and -CAS, with categories Imp, Mild, and Unimp. The scatter plot shows the relationship between Slums and -CAS, with N=92. The plot includes a regression line and horizontal lines at approximately -50 and -75, with an R-squared value of -0.364.
The scatter plot representing the scores for the 92 subjects in the Cut-Offs Study:
The table to the left of the scatter plot is a key for relating the plot to subject classification. The table enclosed in the scatter plot shows the results of analyses of classification:
Top Line: Impairment cut-off values of SLUMS ≥ 21; Cognivue ≥ 50 yield the following:
10
Negative Percent Agreement (NPA, 100 x [TP/TP+FP]) = 50 Positive Percent Agreement (PPA, 100 x [TP/TP+FP]) = 86 Accuracy (ACC, [True Positives + True Negatives]/total sample) = .86 Error Symmetry ([False Positives - False Negatives]/total) = . 01.
Bottom Line: Unimpaired cut-off values of SLUMS ≥ 27 and Cognivue ≥ 75 yield the following: Negative Percent Agreement (NPA. 100 x [TP/TP+FP]) = 67
Positive Percent Agreement (PPA, 100 x [TP/TP+FP]) = 66 Accuracy (ACC, [True Positives + True Negatives]/total sample) = .66
Error Symmetry ([False Positives - False Negatives]/total) = . 05.
Summary: Cognivue cut-off scores between 55 and 64 provide an acceptable adjunctive tool for recognizing patients who would score 0-20 on SLUMS, a range of SLUMS scores that suggests the presence of impairment. Cognivue cut-off scores between 74 and 79 provide an acceptable adjunctive tool for recognizing patients who would score 27-30 on SLUMS, a range of SLUMS scores that suggests the absence of impairment. Cognivue scores between the impairment and unimpairment ranges described above would provide an acceptable adjunctive tool for recognizing patients who should score 21-26 on SLUMS, a range of SLUMS scores that suggests an intermediate condition between the presence and absence of impairment.
Cognivue scores ≤ 50 are consistent with a conservative standard for impairment. That is, an impairment standard that avoids errors in which Cognivue might imprudently support the impression that a patient is impaired. Cognivue scores > 75 are consistent with a conservative standard for unimpairment. That is, an unimpairment standard that avoids errors in which Cognivue might imprudently support the impression that a patient is not unimpaired.
- b. Agreement analyses (SLUMS and Cognivue). This analysis was an agreement analysis between the subiect device. Cognivue score (range from 0-100). and the non-reference standard, the St. Louis University Mental Status Examination (SLUMS, range from 0-30). The sponsor also did the agreement analysis on the classification results of Cognivue and SLUMS, impaired, unimpaired, and intermediate. The data obtained from the first test session of 401 subjects was utilized to determine the agreement between SLUMS and Cognivue. Subjects were classified as follows:
- SLUMS: SLUMS scores are classified as: If subject is a high school . graduate: >26 is unimpaired, 24 is unimpaired, 74 is unimpaired, PPA = 100% X 60 / (60 + 13) = 82% (95% Wilson interval: .72 to .89) NPA = 100% X 110 / (2 + 110) = 98% (95% Wilson interval: .93 to .99)
This suggests a categorical relationship between Cognivue and SLUMS outcomes for patients who are either impaired or unimpaired.
Summary: These findings confirm agreement between SLUMS and Cognivue. Cognivue can inform an impression that a patient is unimpaired or is impaired. Cognivue sets a conservative standard for supporting the impression that a mildly impaired patient should be presumed to be impaired.
- c. Analysis of the Test-Retest Reliability of Cognivue Classification and SLUMS Scores. The purpose of these analyses was to compare scores from the repeated administration of Cognivue, and from the repeated administration of SLUMS, to assess the re-test reliability of Cognivue, and to compare the re-test reliability of Cognivue to the re-test reliability of SLUMS. Subjects underwent Cognivue and SLUMS testing in two sessions 1-2 weeks apart without feedback, minimizing the impact of learning effects.
Cognivue and SLUMS subject classifications from the first and second tests are in 3 X 3 tables. Scatterplots are presented to represent data related to regression analyses. PPAs, NPAs assess test-retest reliability along with weighted kappa scores at a 95% CI. Changes in SLUMs (>3) and GDS scores (>2) will filter-out subjects who may show non-measurement re-test changes. Intraclass correlation coefficients describe relations between Cognivue and SLUMS scores, and their respective retest reliability using one-way random effects analyses of variance with subjects as the
15
random effect. Bland-Altman plots of the Day 2 - Day 1 difference (Y) vs. the mean over the 2 days (X) describe these findings.
The data obtained for the test-retest reliability analyses included 358 subjects reduced from 401 who were unavailable for repeat testing or among the last tested who could only be tested once due to deadline time constraint. The group of 43 subjects who were only tested once are those who did not sign-up for a second session within the period of testing at their site.
- i. Cognivue test-retest regression analysis. The following is the scatterplot of of Cognivue first test scores (abscissa, full-scale 0-100) and Cognivue second test scores (ordinate, full-scale 0-100) co-plotted with 45 degree line:
Image /page/15/Figure/3 description: The image is a scatter plot that shows the relationship between Cognivue scores from Test 1 and Test 2. The x-axis represents the Cognivue score from Test 1, and the y-axis represents the Cognivue score from Test 2. The data points are scattered around a diagonal line, indicating a positive correlation between the two tests. Most of the data points are clustered in the upper right corner of the plot, suggesting that most participants scored relatively high on both tests.
Deming regression (unweighted) calculated in Analyse-it yields the line:
Cognivue 2 = 9.06 + 0.936 Cognivue 1 Intercept 95% CI = 4.27 to 13.84, SE = 2.433, p=.0002 Slope 95% CI = 0.880 to 0.993, SE = 0.0285, p=.0264) r = 0.90.
These findings demonstrate substantial and significant agreement between Cognivue first test and Cognivue second test scores.
- ii. SLUMS test-retest regression analysis. The following is a scatterplot of SLUMS first test scores (abscissa, full-scale 0-30) and SLUMS second test scores (ordinate, full-scale 0-30) co-plotted with 45 degree line:
16
Image /page/16/Figure/0 description: This image is a scatter plot that compares SLUMS scores from two tests. The x-axis represents the SLUMS score from Test 1, and the y-axis represents the SLUMS score from Test 2. The scores range from 0 to 30 on both axes. There is a diagonal line going from the bottom left to the top right of the plot, and the data points are scattered around this line.
Deming regression (unweighted) calculated in Analyse-it yields the line: SLUMS 2 = 4.15 + 0.90 SLUMS 1 Intercept 95% CI = 2.24 to 6.06, SE = 0.970, pSLUMS
2nd Test | | Impaired | Intermediate | Unimpaired | Total |
| | Impaired | a 48 | b 35 | c 5 | 88 |
| | Intermediate | d 7 | e 70 | f 80 | 157 |
| | Unimpaired | g 0 | h 23 | i 90 | 113 |
| | Total | 55 | 128 | 175 | 358 |
PPA = 100% × a / (a + d + g) PPA = 100% × 48 / (48 + 7 + 0) = 87% (95% Wilson interval: .76 to .94)
NPA = 100% × (e + f + h + i) / (b + c + e + f + h + i) NPA = 100% × (70 + 80 + 23 + 90) / (35 + 5 + 70 + 80 + 23 + 90) = 87% (95% Wilson interval: .83 to .90)
- Percent Agreement for Cognivue Cut-Points (Three Categories): V.
| Cognivue 1st Test | |||||||
|---|---|---|---|---|---|---|---|
| Cognivue | |||||||
| 2nd Test | Impaired | Mildly Impaired | Unimpaired | Total | |||
| Impaired | a 42 | b 21 | c 0 | 63 | |||
| Mildly Impaired | d 5 | e 41 | f 32 | 78 | |||
| Unimpaired | g 0 | h 10 | i 207 | 217 | |||
| Total | 47 | 72 | 239 | 358 |
Percent agreement for impaired: tests 1 and 2 = 100.0% *42/(42+5+0) = 89%
Percent agreement for mildly impaired: tests 1 and 2 = 100.0%*41/(21+41+10) = 57%
18
Percent agreement for unimpaired: tests 1 and 2 = 100.0% * 207/(0+32+207) = 87%
| SLUMS 1st Test | |||||
|---|---|---|---|---|---|
| SLUMS | |||||
| 2nd Test | Impaired | Mildly Impaired | Unimpaired | Total | |
| Impaired | a 48 | b 35 | c 5 | 88 | |
| Mildly Impaired | d 7 | e 70 | f 80 | 157 | |
| Unimpaired | g 0 | h 23 | i 90 | 113 | |
| Total | 55 | 128 | 175 | 358 |
-
vi. Percent Agreement for SLUMS Cut-Points (Three Categories):
Percent agreement for impaired: tests 1 and 2 = 100.0% * 48/(48+7+0) = 87% Percent agreement for mildly impaired: tests 1 and 2 = 100.0% * 70/(35+70+23)=55% Percent agreement for unimpaired: tests 1 and 2 = 100.0% * 90/(5+80+90) = રાજ -
vii. Impairment Classification Agreement for Cognivue Test 1 to Test 2:
| | Impaired | Mildly
Impaired | Unimpaired | Total |
|------------|----------|--------------------|------------|-------|
| Cognivue 1 | 47 | 72 | 239 | 358 |
| Cognivue 2 | 63 | 78 | 217 | 358 |
| Total | 110 | 150 | 456 | 716 |
The Fisher-Freeman-Halton exact test (SPSS v22) yields a value of 3.62 (df = 2), p = 0.161. The Cognivue classifications for impairment, mild impairment, and unimpairment do not differ significantly by the time of the test in the tested population.
- viii. Impairment Classification Agreement for SLUMS Test 1 to Test 2:
| | Impaired | Mildly
Impaired | Unimpaired | Total |
|---------|----------|--------------------|------------|-------|
| SLUMS 1 | 55 | 128 | 175 | 358 |
| SLUMS 2 | 88 | 157 | 113 | 358 |
| Total | 143 | 285 | 288 | 716 |
19
The Fisher-Freeman-Halton exact test (SPSS v22) yields a value of 24.89 (df = 2), p 75 is consistent with being unimpaired.