Informant Questionnaire on Cognitive Decline in the Elderly

Transient Ischemic Attack (TIA)/Stroke: (van Nieuwkerk et al., 2021; n = 1068; mean age = 72.9 (12.3), male = 52.3%; 16-item IQCODE)

• The optimal cut-off score of IQCODE for assessing prestroke dementia was > 3.48 (sensitivity 89.7%, specificity 84.2%).
  • Optimal cut-off score was nonsignificantly higher for major stroke (NIH Stroke Scale score >= 3) than minor stroke (> 3.85 vs. > 3.47), but was similar for patients with first-ever stroke (> 3.48; sensitivity 85.7%, specificity 84.8%).

Acute Stroke: (Tang et al., 2003; n = 189; mean age = 74.2 (10.4) years; female = 52.9%; mean National Institute of Health Stroke Scale score = 11.9 (10.6); time since stroke 14.3 (1.4) weeks; Chinese sample and translation of 26-item IQCODE).

• The optimal cut-off point of the IQCODE for assessing post-stroke dementia was ≤ 3.40 (sensitivity 88%; specificity 75%) 

Acute Stroke: (Hénon, Pasquier, Durieu, Godefroy, Lucas, Lebert, Leys, 1997; n = 202, median age = 75 years; age range = 42-101; female = 52.0%; French sample and translation of 26-item IQCODE).

• Dementia cut-off IQCODE score >= 104

Transient Ischemic Attack (TIA)/Stroke: (van Nieuwkerk et al., 2021)

• Mean item score for all IQCODES = 3.23 (0.55) with median (IQR) = 3.13 (3.00 – 3.38)
  • Mean item score for participants identified with pre-event dementia was 4.34 (0.59) vs. 3.15 (0.46) for those with no dementia (p < 0.0001)

Acute Stroke: (Tang et al., 2003; n = 189).

• Excellent: Cronbach's alpha = 0.95*

Post-Stroke: (Othman et al., 2015; n = 50, mean age = 42.7 (11.6) years, age range = 21-68 years, female = 64%, caregivers of elderly (age > 60) post-stroke outpatients, Malay sample and translation of 16-item IQCODE).

• Excellent:  Cronbach’s alpha = 0.94*

*Scores higher than 0.9 may indicate redundancy in the scale questions.

Predictive validity:

Transient Ischemic Attack (TIA)/Stroke: (van Nieuwkerk et al., 2021)

• Excellent predictive ability of IQCODE for detecting pre-event dementia (AUC = 0.94, 95% C.I. 0.90-0.97, p < 0.001)
  • Significant differences among all 16 IQCODE items for patients with vs. without pre-event dementia, both overall and after stratification for event severity (p < 0.001)

Acute Stroke: (Tang et al., 2003; n = 189).

• Adequate predictive validity (AUC = .88)
• Low positive predictive value (33%) for detecting post-stroke dementia

Alzheimer’s Disease: (Ding et al., 2018; n = 394; mean age = 70.01 (9.77) years; age range = 43-91 years; female = 53.2%; 16-item IQCODE).

• Cut-off between mild to moderate dementia: 65 (sensitivity 66.1%, specificity 59.8%)
• Cut-off between moderate to severe dementia: 75 (sensitivity 73.9%, specificity 67.7%)  

Progressive Dementia: (Perroco, et al., 2008; n = 49; mean age = 70.5 years; female = 49%; 9 classified as Clinical Dementia Rating (CDR) of 0, 11 as CDR = 0.5, and 29 as CDR >= 1; Brazilian sample and translation of 16-item IQCODE).

• Cut-off score between participants with mild cognitive impairment and dementia: 4.0 – 4.1

Progressive Dementia: (Perroco, et al., 2008; n = 49; Brazilian sample and translation of 16-item IQCODE).

Mean scores on the 16-item IQCODE according to Clinical Dementia Rating (CDR) classification

Alzheimer’s Disease: (Ding et al., 2018; n  = 394).

• Excellent inter-rater reliability (k = 0.894, p < 0.001).

Convergent validity:

Alzheimer’s Disease: (Ding et al., 2018; n = 394).

• Poor area under the curve (AUC) for the IQCODE discriminating between mild and moderate dementia (AUC = 0.666, 95% CI: 0.601-0.732)
• Adequate area under the curve (AUC) for the IQCODE discriminating between moderate and severe dementia (AUC = 0.768, 95% CI: 0.715–0.822)
• Adequate convergent validity between IQCODE scores and neuropsychological measures for those with moderate dementia (CDR = 2) and for all dementia groups:
  • Mini-Mental State Exam (MMSE) (r = -0.409 for moderate, -0.528 for all)
  • Mattis Dementia Rating Scale (r = -0.324 for moderate, -0.436 for all)
  • Alzheimer’s Disease Assessment Scale – Cognitive subscale (r = -0.370 for moderate, -0.477 for all)
  • The above findings suggest that IQCODE is more accurate for assessing moderate dementia than for assessing mild or severe dementia (p. 149).

Progressive Dementia: (Perroco et al., 2008; n = 49).

• Excellent convergent validity between IQCODE-S scores and Clinical Dementia Rating (r = 0.65, p < 0.001)
• Excellent convergent validity between IQCODE-S scores and Clinical Dementia Rating controlled by age and education (r = 0.61, p < 0.001)

Discriminant validity:

Progressive Dementia: (Perroco et al., 2008; n = 49).

• Adequate discriminant validity between IQCODE-S scores and years of schooling (r = -0.33, p = 0.021).

Alzheimer’s Disease: (Ding et al., 2018; n = 394).

• Dementia of different severities had significantly different IQCODE scores for items, except the third item
• The total scores from the mild, moderate and severe groups were 63.38 (±9.49), 68.39 (±8.73) and 75.96 (±3.52), and the differences were significant (p < 0.01).

Older Adults: (Jansen et al., 2008; n = 4823; female = 50.5%; mean age = 76 (6.6) years; Netherlands sample; Dutch language translation of self-report 16-item IQCODE-SR).

• Cut-off score for dementia screening = 3.6

Older Adults: (Phung et al., 2015; n = 236; age = 65 years or older and their informants; Arabic-speaking sample and translation of 16-item IQCODE).

• Cut-off score = 3.34 (sensitivity 92.5, specificity 94.4)

Elderly: (Lourenço & Sanchez, 2014; n  = 417; female = 71.9%; mean age = 79.3 (7.3) years; Brazilian sample and translation of IQCODE into IQCODE-BR)

• Cut-off score = 3.26 (sensitivity 89%, specificity 72%) 

Elderly: (Foroughan et al., 2019; n = 95; age = 60 or greater; Iranian sample; Farsi translation of 16-item IQCODE).

• Acceptable test-retest reliability: (ICC = 0.81)

Elderly: (Jorm & Jacomb, 1989; n = 309 informants who were current caregivers; mean age of subjects = 74 (8.0), female = 61%; 26-item IQCODE)

• Adequate test-retest reliability: (ICC = 0.75) 

Older Adults: (Jansen et al., 2008; n = 4823).

• Excellent: Cronbach’s alpha = .94
• Range of Item-total correlations: r = 0.62 – 0.72
• Range of item-item correlations: r = 0.36 – 0.71 

Older Adults: (Phung et al., 2015; n = 236).

• Excellent:  Cronbach’s alpha = .97

Elderly: (Foroughan et al., 2019; n = 95).

• Excellent: Cronbach’s alpha = .93

Elderly: (Jorm & Jacomb, 1989; n = 613 informants from general population)

• Excellent Internal consistency: Cronbach’s alpha = .95

Elderly: (Foroughan et al., 2019; n = 95).

• To evaluate the content validity and appropriateness of the Farsi version of the IQCODE, an expert panel of seven persons -- three linguists, two psychiatrists, a psychometrist, and a neuropsychologist -- supervised all the steps of translation and adaption of the test, first individually and then in group meetings.
• The finalized version all panel members agreed upon was used for the fieldwork.

Older Adults: (Phung et al., 2015; n = 236).

• IQCODE could effectively discriminate between normal cognition, mild dementia, and moderate dementia (p < 0.0005). 

Elderly: (Lourenço & Sanchez, 2014; n = 417).

• Multivariate analysis showed that there was no statistically significant association between the IQCODE-BR scores and the informants’ mental state, presence of depressive symptoms, as well as living in the same household and being the primary caregiver.
• IQCODE was accurate in assessing the cognitive ability of elderly participants, comparing present and past cognition and functionality

Systemic Lupus Erythematosus (SLE): (Chalhoub & Luggen, 2019; SLE patients fulfilling the updated 1997 American College of Rheumatology Classification Criteria; n  = 78, mean age = 45.9 (11.2); female = 91%; mean disease duration = 10.4 (6.9) years; 16-item IQCODE)

• At the standard cutoff score of >= 3.2 to indicate greater cognitive decline, 26% were identified by the IQCODE as being impaired
• >= 3.1 was the best cutoff for normal controls across all definitions of cognitive dysfunction (CD) from CD1 to CD4 (sensitivity 53%, specificity 52%)
• >= 3.1 was also the best cutoff for rheumatoid arthritis patients (sensitivity 32%, specificity 47%) for CD1 only (number of patients meeting the criteria for CD2, CD3, and CD4 were too low to make reliable assessments. 

Systemic Lupus Erythematosus (SLE): (Chalhoub & Luggen, 2019)

• Poor concurrent validity between total throughput score (TTS) of the Automated Neuropsychologic Assessment Metrics (ANAM) and IQCODE scores (r = -0.034, p = 0.8152)
• No Significant correlations were found between IQCODE scores and any of the individual ANAM subtests. 

Down Syndrome (DS): (Mattar, et al., 2022; n = 92, mean age = 42.43 (8.48) years, age range = 30 – 64), female = 34 (36.9%); inclusion criteria = diagnosed with DS according to ICD-10, code Q90; age >= 30, in daily contact with informant for at least 10 years; diagnostic assessment of dementia made using Cambridge Examination for Mental Disorders of Older People with Down’s Syndrome and Others with Intellectual Disabilities (CAMDEX-DS), with classification into stable condition (n  = 62), mild cognitive impairment (prodromal dementia) (n = 17), and AD (dementia) (n = 13); Portuguese translation of 26-item IQCODE)

• >= 3.14 was the optimal cutoff score for dementia vs. stable condition (sensitivity 100%, specificity 96.8%)
• >=3.11 was the optimal cutoff score for prodromal dementia + dementia vs. stable condition (sensitivity 93.3%, specificity 91.9%)

Predictive validity:

Down Syndrome (DS): (Mattar, et al., 2022; n = 92; Portuguese translation of 26-item IQCODE)

• Excellent predictive ability of IQCODE for dementia vs. stable condition (AUC = 0.993, p < 0.001)
• Excellent predictive ability of IQCODE for prodromal dementia + dementia vs. stable condition (AUC = 0.975, p < 0.001)

Concurrent validity:

Down Syndrome (DS): (Mattar, et al., 2022; n = 92; Portuguese translation of 26-item IQCODE)

• Adequate correlation of IQCODE Score with Cambridge Cognitive Examination adapted for individuals with DS (CAMCOG-DS) total score and domains:
  • Total score (r = -0.333, p = 0.002)
  • Memory domain (r = -0.374, p < 0.001)
  • Attention domain (r = -0.302, p = 0.005)
  • Perception domain (r = -0.338, p = 0.001)
• Poor correlation of IQCODE Score with Cambridge Cognitive Examination adapted for individuals with DS (CAMCOG-DS) domains:
  • Orientation domain (r = -0.214, p = 0.048)
  • Language domain (r = -0.297, p = 0.006)
  • Praxis domain (r = -0.295, p = 0.006)
  • Abstract thinking domain (r = -0.104, p = 0.341)