Rhodius‐Meester 2016.

Study characteristics
Patient sampling	Primary objectives: evaluate how a clinical decision support systems such as the PredictAD tool can aid clinicians to integrate biomarker evidence to support AD diagnosis Study population: patients with MCI from the Amsterdam Dementia Cohort, who had visited the Alzheimer center at the VU University Medical Center (VUMC) between 2000 and 2012. Selection criteria: MCI were included if a MMSE score was present, if both MRI and CSF biomarkers were available, if a follow‐up of at least 2 years was conducted. Exclusion criteria not reported in the published article. According to the Amsterdam Dementia Cohort protocol, all MCI were assessed in order to identify a potential neurodegenerative disease. In Amsterdam Dementia Cohort vascular contribution to dementia conversion was considered (Van der Flier 2014). 23 people with MCI that progressed to another dementia were excluded from the study, 40 cases without MRI were excluded from the study Study design: prospective longitudinal study (VUMC cohort)
Patient characteristics and setting	Clinical presentations: MCI was diagnosed using Petersen's criteria (Petersen 2004); in addition all participants fulfilled the core clinical criteria of the NIA‐AA for MCI (Albert 2011) Age mean (SD): MCI who progressed to AD: 72 ± 7; stable MCI: 68 ± 6 Gender (% men): MCI who progressed to AD: 45%; stable MCI: 69% Education years mean (SD): MCI who progressed to AD: 5 ± 1; stable MCI: 5 ± 1 ApoE4 carriers (%): not stated Neuropsychological tests: employed; MMSE mean (SD): MCI who progressed to AD: 26 ± 3; stable MCI 27 ± 2 Clinical stroke excluded: unclear: infarcts were permitted but it is not specified if clinical or radiological. On FLAIR MRI white matter hyperintensities were rated using Fazekas scale, lacunes were counted and defined as deep lesions with low signal on T1‐weighted sequences and high signal on T2‐weighted sequences. Microbleeds were counted on T2 star sequences. Co‐morbidities: not specified Number enrolled: 171 Number available for analysis: 171 Setting: Amsterdam Dementia Cohort from VUMC Country: Netherlands Period: 2000‐2012 Language: English
Index tests	Index test: MRI visual method for estimation of MTA Manufacturer: Siemens Magnetom Impact and Sonata, GE Healthcare Signa HDXT Tesla strength: 1.0 or 1.5 Assessment methods: all scans were visually rated. Visual rating of MTA was performed on coronal T1‐weighted images according to Scheltens 1997. The PredictAD tool was also performed to judge the combined biomarkers as indicative of AD pathophysiology. Description of positive cases definition by index test as reported: MTA averaged left and right score ≥ 1.5 was considered pathologic (Van de Pol 2014). Examiners: a trained rater evaluated all scans. Images were evaluated again in a consensus meeting with an experienced neuroradiologist. Interobserver variability: inter‐ and intra‐rater weighted kappa's of at least 0.80 for MTA was required.
Target condition and reference standard(s)	Target condition: AD Prevalence of AD in the sample: 104/171 (61% of participants included in the analysis) Stable MCI or converted to other dementia: 67 (39%) stable MCI Reference standards: NINCDS‐ADRDA criteria (McKhann 1984) and NIA‐AA core clinical criteria (McKhaan 2011) Median clinical follow‐up: 3 years
Flow and timing	Withdrawals explained and losses to follow‐up: none Uninterpretable MRI results have not been reported
Comparative
Key conclusions by the authors	The ability of the PredictAD tool to identify AD pathophysiology was comparable to individual biomarkers. The PredicAD toll has the advantage that assigns likelihood to all participants, regardless of missing or conflicting data, allowing clinicians to integrate biomarker data in daily practice
Conflict of interests	Study authors' disclosures available online (www.j‐alz.com/manuscript‐disclosures/15‐0548r1).
Notes	Source of funding: VUMC Alzheimer centre is supported by Alzheimer Nederland and Stichting VUMC fonds. The clinical database structure was developed with funding from Stichting Dioraphte. Other grants for the project: grant no. 733050201. grant agreement 611005, grant agreements601055 (VPH‐DARE@IT) and 224328 (PredictAD) 2 x 2 table: data from the published article; we only considered the MTA score for this review.
Methodological quality
Item	Authors' judgement	Risk of bias	Applicability concerns
DOMAIN 1: Patient Selection
Was a consecutive or random sample of patients enrolled?	No
Was a case‐control design avoided?	Yes
Did the study avoid inappropriate exclusions?	No
		High	Low
DOMAIN 2: Index Test All tests
Were the index test results interpreted without knowledge of the results of the reference standard?	Unclear
Did the study provide a clear pre‐specified definition of what was considered to be a "positive" result of the index test?	Yes
Was the index test performed by a single operator or interpreted by consensus in a joint session?	Yes
		Unclear	Low
DOMAIN 3: Reference Standard
Is the reference standards likely to correctly classify the target condition?	Yes
Were the reference standard results interpreted without knowledge of the results of the index tests?	Yes
		Low	Low
DOMAIN 4: Flow and Timing
Was there an appropriate interval between index test and reference standard?	Yes
Did all patients receive the same reference standard?	Yes
Were all patients included in the analysis?	Yes
		Low