Informant-based screening tools for dementia: an overview of systematic reviews

Informant accuracy of IQCODE, AD8 and GPCOGi for diagnosis of dementia: does your friend know best?

BACKGROUND

Increasing numbers of people require evaluation for possible dementia. However, research on the accuracy of informant questionnaires in primary care remains limited.

METHODS

This study assessed the diagnostic accuracy of IQCODE, AD8, and GPCOGi based on the informant's relationship to the patient. We recruited 240 participants from 21 general practices in South West England. The reference standard for a diagnosis of dementia was made by a specialist clinician using ICD-10 criteria. A threshold of greater than 3.3 on IQCODE, greater or equal to 2 on AD8 and less than 5 on the informant component of GPCOG was used to indicate an abnormal test.

RESULTS

Of 238 participants with informant data, 131 had dementia, 60 had CIND, and 47 had normal cognition. Median informant age was 70 years (IQR 60 years to 78 years). 71% of informants were female and 56% were spouses. On all three questionnaires, compared to spouses, adult descendants tended to score participants more cognitively impaired, whereas friends scored participants less cognitively impaired. However, there was little evidence of difference by informant type once fully adjusted. Sensitivity by informant type ranged from 91 to 100% for IQCODE, 94-100% for AD8 and 99% to100% for GPCOGi. There was no significant difference in sensitivity by informant type. Specificity by informant type ranged from 25 to 79% for IQCODE, 13-75% for AD8 and 17-38% for GPCOGi. Adult descendants tended to have the lowest specificity at 25% (95% CI 10-47%) for IQCODE, 13% (95% CI 3-32%) for AD8 and 17% (95% CI 5-37%) for GPCOGi. Friends tended to have the highest specificity at 79% (95% CI 49-95%) for IQCODE, 75% (95% CI 48-93%) for AD8 and 38% (95% CI 15-64%) for GPCOGi.

CONCLUSIONS

An informant of any relationship type, using IQCODE, AD8 or GPCOGi may be useful for ruling out dementia but not for ruling it in. We found no evidence of difference between spouse or adult descendants but friends performed significantly better overall on IQCODE and AD8.

An overview of systematic reviews on imaging tests for diagnosis of pulmonary embolism applying different network meta-analytic methods

PURPOSE

This study aimed to apply different methods of diagnostic test accuracy network meta-analysis (DTA-NMA) for studies reporting results of five imaging tests for the diagnosis of suspected pulmonary embolism (PE): pulmonary angiography (PA), computed tomography angiography (CTPA), magnetic resonance angiography (MRA), planar ventilation/perfusion (V/Q) scintigraphy and single-photon emission computed tomography ventilation/perfusion (SPECT V/Q).

METHODS

We searched four databases (MEDLINE [via PubMed], Cochrane CENTRAL, Scopus, and Epistemonikos) from inception until June 2, 2022 to identify systematic reviews (SRs) describing diagnostic accuracy of PA, CTPA, MRA, V/Q scan and SPECT V/Q for suspected PE. Study-level data were extracted and pooled using a hierarchical summary receiver operating characteristic (HSROC) meta-regression approach and two DTA-NMA models to compare accuracy estimates of different imaging tests. Risk of bias was assessed using the QUADAS-2 (Quality Assessment of Diagnostic Accuracy Studies-2) tool and certainty of evidence using the GRADE (Grading of Recommendations Assessment, Development and Evaluation) framework.

RESULTS

We identified 13 SRs, synthesizing data from 33 primary studies and for four imaging tests (PA, CTPA, MRA and V/Q scan). The HSROC meta-regression model using PA as the reference standard showed that MRA had the best overall diagnostic performance with sensitivity of 0.93 (95% confidence interval [CI]: 0.76, 1.00) and specificity of 0.94 (95% CI: 0.84, 0.99). However, DTA-NMA models indicated that V/Q scan had the highest sensitivity, while CTPA was most specific.

CONCLUSION

Selecting a different DTA-NMA method to assess multiple diagnostic tests can affect estimates of diagnostic accuracy. There is no established method, but the choice depends on the data and familiarity with Bayesian statistics.

Informant-Administered Cognitive Screening as a Component of Memory Clinic Assessment: Overcoming Ethical and Practical Challenges

Possin and Rosen provide a robust commentary exploring the challenges of using caregivers as cognitive testers. Informants have an important and often overlooked role in diagnosing cognitive impairment. O'Caoimh et al. show they can support cognitive screening in advance of clinic, suggesting new research avenues including the potential for home-monitoring. Although concerns testing may engender bias, introduce practice effects, and impact patient autonomy are valid and require examination, these should be viewed in light of patient preference, clinical need, and the broader ethics of assessing dementia. The importance of distinguishing concerns over accuracy and ethical appropriateness is also discussed.

Comparative validity of informant tools for assessing pre-stroke cognitive impairment

OBJECTIVES

Various informant-based questionnaires are used in clinical practice to screen for pre-stroke cognitive problems. However, there is no guidance on which tool should be preferred. We compared the validity of the two most commonly used informant-based tools.

METHODS

We recruited consecutively admitted stroke patients. Patients' informants completed the Informant Questionnaire for Cognitive Decline in the Elderly Short Form (IQCODE-SF, 16-item) and Ascertain Dementia 8 (AD8). We assessed construct validity (accuracy) against a semi-structured clinical interview for dementia or mild cognitive impairment (MCI), describing test accuracy metrics and comparing area under ROC curves (AUROC). We described criterion validity by evaluating associations between test scores and neuroimaging markers of dementia and overall 'brain frailty'. Finally, we described prognostic validity comparing ROC curves for 18-month clinical outcomes of dementia, death, stroke, and disability.

RESULTS

One-hundred-thirty-seven patient-informant dyads were recruited. At usual clinical cut-points, the IQCODE-SF had comparable sensitivity to the AD8 (both = 92%) for pre-stroke dementia, but superior specificity (IQCODE-SF: 82% vs. AD8: 58%). Youden index suggested that the optimal AD8 threshold for diagnosis of dementia is ≥4. The IQCODE-SF demonstrated stronger associations with markers of generalised and medial-temporal lobe atrophy, neurovascular disease, and overall brain frailty. IQCODE-SF also demonstrated greater accuracy for predicting future dementia (IQCODE-SF AUROC = 0.903, 95% CI = 0.798-1.00; AD8 AUROC = 0.821, 95% CI = 0.664-0.977).

CONCLUSIONS

Both IQCODE-SF and AD8 are valid measures of pre-stroke dementia. Higher cut points for AD8 may improve performance in the acute stroke setting. Based on consistent superiority across a range of validity analyses, IQCODE-SF may be preferable to AD8 for pre-stroke dementia screening.