Brain scans & tests

• Sensitivity 92%
• Specificity 100%
• Results in Label: 10-minute PET images starting approximately 30-50 minutes after injection. We found no source for how long the report takes.
• Availability: Limited
• Radiation: Effective dose 19 microSv/MBq; ~7 mSv for 370 MBq (label Table 1). Additional dose from CT if PET/CT.

What it measures

Estimates amyloid-beta neuritic plaque density in the brain (label: 'positron emission tomography (PET) of the brain to estimate amyloid beta neuritic plaque density in adults with cognitive impairment').

When it is used

Label indications: (1) evaluation of Alzheimer's disease and other causes of cognitive decline; (2) selection of patients indicated for amyloid-beta-directed therapy. Negative scan = sparse-to-no neuritic plaques, inconsistent with AD at time of imaging.

Medicare coverage

NCD 220.6.20 (effective 2013-09-27, implemented 2014-07-07) covered 'one PET Abeta scan per patient' only under coverage with evidence development (CED) in CMS-approved studies. NCA CAG-00431R decision memo dated 2023-10-13: CMS is 'removing the national coverage determination (NCD) at 220.6.20, ending coverage with evidence development (CED) for positron emission tomography (PET) beta amyloid imaging' and allowing coverage determinations 'to be made by the Medicare Administrative Contractors (MACs)', which 'may include covering more than one scan per patient's lifetime'. NCD retired effective 2023-10-13 (NCD 356 ver 2); transmittal R12364NCD issued 2023-11-16, implementation 2023-12-19 (CR 13429). No LCD was found; coverage is at MAC discretion under SSA 1862(a)(1)(A). Hospital outpatient: A9586 is separately payable (OPPS status indicator K) since the CY2025 diagnostic-radiopharmaceutical policy.

Guidelines say

Rabinovici et al. 2025 updated AUC (Alzheimer's Association / SNMMI; J Nucl Med 2025;66(Suppl 2):S5-S31, doi 10.2967/jnumed.124.268756; also Alzheimers Dement doi 10.1002/alz.14338): 7 amyloid-PET scenarios rated APPROPRIATE (MCI/dementia <65 y with suspected AD; amnestic MCI/dementia >=65 y; MCI/dementia with atypical features; equivocal CSF biomarkers; prognosis in MCI due to suspected AD; eligibility for amyloid-targeting therapy; monitoring response to amyloid-targeting therapy), 2 uncertain, 8 RARELY APPROPRIATE (incl. cognitively unimpaired persons whether or not at increased risk by age/APOE4/family history; in lieu of genotyping for suspected autosomal-dominant carriers). AA DETeCD-ADRD primary-care guideline 2025 (Alzheimers Dement 21(6):e14333) Rec 18: amyloid PET 'if diagnostic uncertainty still exists after obtaining structural imaging with or without FDG PET and/or CSF' (Strength B). Jack et al. 2024 revised criteria (Alzheimers Dement 20(8):5143-5169) list amyloid PET as a Core 1 biomarker: 'An abnormal Core 1 biomarker result is sufficient to establish a diagnosis of AD'.

Sensitivity 92% (95% CI 78-98), specificity 100% (95% CI 80-100), majority read of in-person-trained readers. Inter-reader Fleiss kappa 0.83 (0.78-0.88) in Study 2. Original NDA approved 2012-04-06 (Drugs@FDA ORIG-1); efficacy supplement S-046 approved 2025-06-23 (label effective 2025-12-11). COST: the typical cost is a SUM of two published CY2026 OPPS national rates (Addendum B, January 2026): CPT 78811 PET limited area $1,322.69 (APC 5593) + HCPCS A9586 florbetapir $1,825.64 (SI K); using 78814 PET/CT limited instead of 78811 gives $1,460.92 + $1,825.64 = $3,286.56. 2024 Medicare Part B national averages (physician/supplier claims, non-facility): A9586 avg allowed $3,153.92 (10,099 services, 10,069 beneficiaries); 78811 avg allowed $989.93 (office) / $73.92 professional component (facility). Availability 'limited' is a judgement from the AUC 2025 statement that coverage 'remains uncertain for many patients' and IDEAS (595 US sites 2016-17); no scanner-count source was loaded.

Source: api.fda.gov · 2012

Cost source: cms.gov

• Sensitivity 93%
• Specificity 84%
• Results in Label: images starting approximately 60-120 minutes after injection. We found no source for how long the report takes.
• Availability: Limited
• Radiation: Whole-body effective dose 5.9 mSv for 185 MBq (label).

What it measures

Estimates amyloid-beta neuritic plaque density in adults with cognitive impairment.

When it is used

Evaluation of AD and other causes of cognitive decline; selection of patients for amyloid-beta-directed therapy (label indications as for Amyvid).

Medicare coverage

NCD 220.6.20 (effective 2013-09-27, implemented 2014-07-07) covered 'one PET Abeta scan per patient' only under coverage with evidence development (CED) in CMS-approved studies. NCA CAG-00431R decision memo dated 2023-10-13: CMS is 'removing the national coverage determination (NCD) at 220.6.20, ending coverage with evidence development (CED) for positron emission tomography (PET) beta amyloid imaging' and allowing coverage determinations 'to be made by the Medicare Administrative Contractors (MACs)', which 'may include covering more than one scan per patient's lifetime'. NCD retired effective 2023-10-13 (NCD 356 ver 2); transmittal R12364NCD issued 2023-11-16, implementation 2023-12-19 (CR 13429). No LCD was found; coverage is at MAC discretion under SSA 1862(a)(1)(A). Hospital outpatient: HCPCS Q9982 'Flutemetamol f18 diagnostic' separately payable (SI K) in January 2026 Addendum B.

Guidelines say

Rabinovici et al. 2025 updated AUC (Alzheimer's Association / SNMMI; J Nucl Med 2025;66(Suppl 2):S5-S31, doi 10.2967/jnumed.124.268756; also Alzheimers Dement doi 10.1002/alz.14338): 7 amyloid-PET scenarios rated APPROPRIATE (MCI/dementia <65 y with suspected AD; amnestic MCI/dementia >=65 y; MCI/dementia with atypical features; equivocal CSF biomarkers; prognosis in MCI due to suspected AD; eligibility for amyloid-targeting therapy; monitoring response to amyloid-targeting therapy), 2 uncertain, 8 RARELY APPROPRIATE (incl. cognitively unimpaired persons whether or not at increased risk by age/APOE4/family history; in lieu of genotyping for suspected autosomal-dominant carriers). AA DETeCD-ADRD primary-care guideline 2025 (Alzheimers Dement 21(6):e14333) Rec 18: amyloid PET 'if diagnostic uncertainty still exists after obtaining structural imaging with or without FDG PET and/or CSF' (Strength B). Jack et al. 2024 revised criteria (Alzheimers Dement 20(8):5143-5169) list amyloid PET as a Core 1 biomarker: 'An abnormal Core 1 biomarker result is sufficient to establish a diagnosis of AD'.

Sensitivity 93% (95% CI 81-99), specificity 84% (95% CI 64-96), majority read of electronic-media-trained readers. Fleiss kappa 0.83 (0.79-0.86). NDA approved 2013-10-25 (Drugs@FDA); efficacy supplement S-024 approved 2025-06-23; label effective 2026-04-06. COST: SUM of CY2026 OPPS rates: 78811 $1,322.69 + Q9982 $1,944.01 = $3,266.70. No 2024 Part B claims rows exist for Q9982/A9599 in the physician dataset (tracer was billed under NOC code A9599 before Q9982; A9599 is absent from the January 2026 Addendum B).

Source: api.fda.gov · 2013

Cost source: cms.gov

• Sensitivity 96.2%
• Specificity 80%
• Results in Label: images starting approximately 45-130 minutes after injection. We found no source for how long the report takes.
• Availability: Limited
• Radiation: Effective dose 19 microSv/MBq; 5.8 mSv for 300 MBq (label Table 1).

What it measures

Estimates amyloid-beta neuritic plaque density in adults with cognitive impairment.

When it is used

Evaluation of AD and other causes of cognitive decline; selection of patients for amyloid-beta-directed therapy.

Medicare coverage

NCD 220.6.20 (effective 2013-09-27, implemented 2014-07-07) covered 'one PET Abeta scan per patient' only under coverage with evidence development (CED) in CMS-approved studies. NCA CAG-00431R decision memo dated 2023-10-13: CMS is 'removing the national coverage determination (NCD) at 220.6.20, ending coverage with evidence development (CED) for positron emission tomography (PET) beta amyloid imaging' and allowing coverage determinations 'to be made by the Medicare Administrative Contractors (MACs)', which 'may include covering more than one scan per patient's lifetime'. NCD retired effective 2023-10-13 (NCD 356 ver 2); transmittal R12364NCD issued 2023-11-16, implementation 2023-12-19 (CR 13429). No LCD was found; coverage is at MAC discretion under SSA 1862(a)(1)(A). Hospital outpatient: HCPCS Q9983 'Florbetaben f18 diagnostic' separately payable (SI K) in January 2026 Addendum B.

Guidelines say

Rabinovici et al. 2025 updated AUC (Alzheimer's Association / SNMMI; J Nucl Med 2025;66(Suppl 2):S5-S31, doi 10.2967/jnumed.124.268756; also Alzheimers Dement doi 10.1002/alz.14338): 7 amyloid-PET scenarios rated APPROPRIATE (MCI/dementia <65 y with suspected AD; amnestic MCI/dementia >=65 y; MCI/dementia with atypical features; equivocal CSF biomarkers; prognosis in MCI due to suspected AD; eligibility for amyloid-targeting therapy; monitoring response to amyloid-targeting therapy), 2 uncertain, 8 RARELY APPROPRIATE (incl. cognitively unimpaired persons whether or not at increased risk by age/APOE4/family history; in lieu of genotyping for suspected autosomal-dominant carriers). AA DETeCD-ADRD primary-care guideline 2025 (Alzheimers Dement 21(6):e14333) Rec 18: amyloid PET 'if diagnostic uncertainty still exists after obtaining structural imaging with or without FDG PET and/or CSF' (Strength B). Jack et al. 2024 revised criteria (Alzheimers Dement 20(8):5143-5169) list amyloid PET as a Core 1 biomarker: 'An abnormal Core 1 biomarker result is sufficient to establish a diagnosis of AD'.

In-person-trained readers (3) sensitivity 96.2% (95% CI 86.8-99.5), specificity 80.0% (61.4-92.3); electronic-media-trained readers (5) sensitivity 96.2%, specificity 76.7% (57.7-90.1). Fleiss kappa 0.79 (0.77-0.83). NDA approved 2014-03-19 (Drugs@FDA, sponsor Lantheus); efficacy supplement S-024 approved 2025-06-23; label effective 2026-07-24. COST: SUM of CY2026 OPPS rates: 78811 $1,322.69 + Q9983 $1,584.67 = $2,907.36.

Source: api.fda.gov · 2014

Cost source: cms.gov

• Availability: Limited

What it measures

Change in clinical management within 90 days of amyloid PET in Medicare beneficiaries with MCI or dementia of uncertain cause.

When it is used

Evidence for clinical utility used in the 2023 CMS reconsideration.

Medicare coverage

Scans in IDEAS were covered under the 2013-2023 CED policy (one scan per lifetime in CMS-approved studies). See amyloid-pet-medicare-coverage-2013-2023.

(Rabinovici et al., JAMA 2019;321(13):1286-1294; PMID 30938796): 946 dementia specialists at 595 US sites enrolled 16,008 patients Feb 2016-Sep 2017. Composite management end point changed in 60.2% (95% CI 59.1-61.4) of 6,905 MCI patients and 63.5% (62.1-64.9) of 4,504 dementia patients. Amyloid PET positive in 55.3% of MCI and 70.1% of dementia patients. Single-arm, no control group; management change is not an outcome measure.

Source: doi.org · 2019

• Availability: Limited

What it measures

Whether cognitively unimpaired people with both amyloid and tau PET positivity go on to decline, and how fast, relative to A+T- and A-T- groups.

When it is used

Prognostic interpretation of dual-biomarker PET in preclinical AD; relevant to prevention-trial selection and to counselling after a positive amyloid scan.

Medicare coverage

Research-cohort study; not a coverage source.

Guidelines say

Authors conclude that abnormal amyloid plus tau PET 'is strongly associated with short-term (that is, 3-5 years) cognitive decline in cognitively unimpaired individuals and is therefore of high clinical relevance'.

(Ossenkoppele R, Pichet Binette A, Groot C, et al. Nat Med 2022;28(11):2381-2387; PMID 36357681; PMC9671808): n=1,325 cognitively unimpaired. Cox models for progression to MCI vs A-T- reference: A+T(NEO)+ HR 19.2 (95% CI 10.9-33.7); A+T(MTL)+ HR 14.6 (8.1-26.4); A+T- HR 2.4 (1.4-4.3). Versus A+T-: A+T(MTL)+ HR 6.0 (3.4-10.6), A+T(NEO)+ HR 7.9 (4.7-13.5). Linear mixed models: faster global cognitive decline in A+T(NEO)+ (beta -0.056 +/- 0.005), A+T(MTL)+ (-0.024 +/- 0.005) and A+T- (-0.008 +/- 0.002) vs A-T- (all P<0.001). Research volunteers; wide confidence intervals reflect small A+T+ subgroups.

Source: doi.org · 2022

• Sensitivity 89%
• Specificity 58%
• Availability: Limited

What it measures

Diagnostic test accuracy of a florbetapir scan for predicting which people with MCI will progress to Alzheimer's dementia or any dementia.

When it is used

Prognosis in MCI; the review concluded routine use for this purpose could not be recommended.

Medicare coverage

See amyloid-pet-florbetapir-amyvid for label and coverage; this record is a Cochrane prognosis review.

Guidelines say

Cochrane authors' conclusion: 'we cannot recommend routine use of 18F-florbetapir PET in clinical practice to predict the progression from MCI to ADD' or to any form of dementia, citing poor specificity, limited data and high financial cost.

Blank/negative finding, measured per included study (no pooling possible; Martinez G, Vernooij RW, Fuentes Padilla P, Zamora J, Bonfill Cosp X, Flicker L. Cochrane Database Syst Rev 2017;11:CD012216; PMID 29164603; PMC6486090; search current to May 2017): 3 studies. MCI to AD dementia at 1 to <2 years follow-up (n=401, 1 study): visual read sensitivity 89% (95% CI 78-95), specificity 58% (53-64); SUVR quantification sensitivity 87% (76-94), specificity 51% (45-56). At 2 to <4 years (n=47, 1 study): sensitivity 67% (30-93), specificity 71% (54-85). MCI to any dementia (n=5): sensitivity 67% (9-99), specificity 50% (1-99). 61 of 401 (15.2%) converted at 1.6 years. Two studies at high risk of bias for flow and timing; applicability concerns for the reference standard in all three. The sensitivity/specificity fields carry the largest-study visual-read values.

Source: doi.org · 2017

• Sensitivity 92%
• Specificity 52%
• Results in Label: 20-minute PET image starting approximately 80 minutes after injection. We found no source for how long the report takes.
• Availability: Specialist-centres
• Radiation: Effective dose 8.7 mSv for 370 MBq in a 70-kg adult (label).

What it measures

Density and distribution of aggregated tau neurofibrillary tangles (Braak stage V/VI pattern) in adults with cognitive impairment being evaluated for AD.

When it is used

Evaluation for AD; AUC 2025 rates tau PET appropriate for MCI/dementia <65 y with suspected AD, atypical presentations, prognosis in MCI or dementia due to suspected AD, and eligibility for amyloid-targeting therapy (5 scenarios), 6 uncertain, 6 rarely appropriate (incl. cognitively unimpaired).

Medicare coverage

No NCD exists for tau PET. AUC 2025 (PMC12794165): 'CMS did not issue a noncoverage policy for tau PET; thus, it is covered by CMS under the discretion of the local Medicare Administrative Contractors'. HCPCS A9601 was added to CMS billing article A59049 Group 1 effective 2022-07-01. In the January 2026 OPPS Addendum B A9601 has pass-through status (SI G) at $371.00 per millicurie. No 2024 Part B physician-office claims rows for A9601 exist in the CMS national dataset.

Guidelines say

Rabinovici 2025 AUC (first AUC for tau PET). Jack 2024 revised criteria: tau PET is a 'Core 2' biomarker that 'can provide prognostic information, and when abnormal, will increase confidence that AD is contributing to symptoms'. Not mentioned in the DETeCD-ADRD 2025 primary-care summary.

Sensitivity by reader 92%-100%, specificity 52%-92% for Braak V/VI tau (per-reader table: 97/68, 92/92, 92/88, 92/76, 100/52). Recorded the sensitivity and specificity figures are the LOWER bounds of the reader ranges (convention as in our biomarker table). In the label's analysis restricted to the indicated population sensitivity ranged 68%-86% and specificity 63%-100%. Fleiss kappa 0.87 (0.83-0.91) across 241 patients. Corroborated by Fleisher et al. JAMA Neurol 2020 (doi 10.1001/jamaneurol.2020.0528, PMID 32338734; n=64 primary cohort): sensitivity 92.3%-100%, specificity 52.0%-92.0% for B3 tau; high ADNC sensitivity 94.7%-100%, specificity 50.0%-92.3%. NDA approved 2020-05-28 (Drugs@FDA); label effective 2025-12-30. COST: SUM of CY2026 OPPS rates: 78811 $1,322.69 + A9601 $371.00/mCi x 10 mCi labelled dose = $3,710.00 -> $5,032.69 (the x10 multiplication is mine; per-mCi rate is CMS-published).

Source: api.fda.gov · 2020

Cost source: cms.gov

• Sensitivity 89.9%
• Specificity 90.6%
• Availability: Limited

What it measures

Paired-helical-filament tau (neurofibrillary tangle) burden, compared against MRI atrophy measures for separating Alzheimer's dementia from other neurodegenerative disorders.

When it is used

Differential diagnosis of AD dementia vs non-AD neurodegenerative disorders in memory clinics; the reference standard was clinical diagnosis, not autopsy.

Medicare coverage

See tau-pet-flortaucipir-tauvid and amyloid-tau-pet-opps-radiopharmaceutical-payment-2025-2026 for FDA approval (2020) and CMS payment status.

Guidelines say

Authors: accuracy 'in clinically more representative populations' requires further research; not a guideline statement.

(Ossenkoppele R, Rabinovici GD, Smith R, et al. JAMA 2018;320(11):1151-1162; PMID 30326496; PMC6233630): cross-sectional, 719 participants from 3 dementia centres (South Korea, Sweden, US), June 2014-Nov 2017: 160 cognitively normal, 126 MCI (65.9% amyloid-positive), 179 AD dementia, 254 non-AD neurodegenerative disorders. AD dementia vs all non-AD disorders: medial-basal/lateral temporal ROI sensitivity 89.9% (95% CI 84.6-93.9), specificity 90.6% (86.3-93.9) at SUVR 1.34; Youden cut-off SUVR 1.27 gave sensitivity 96.8% (92.0-99.1), specificity 87.9% (81.9-92.4). AUC for the 5 tau-PET ROIs 0.92-0.95 versus 0.63-0.75 for the 3 volumetric MRI measures (all P<.001). Performance was lower in MCI due to AD (AUC 0.75-0.84). Auc left blank because the abstract reports a range across ROIs rather than a single primary value.

Source: doi.org · 2018

• Availability: Research-only

What it measures

Neurofibrillary tangle deposition by Braak stage, including early transentorhinal (Braak I) tau that first-generation tracers detected poorly.

When it is used

Research staging of tau pathology in living people; discriminating AD dementia from MCI and frontotemporal dementia.

Medicare coverage

No FDA approval or CMS coverage source for MK-6240 was found; FDA approval left blank rather than asserted.

(Pascoal TA, Therriault J, Benedet AL, et al. Brain 2020;143(9):2818-2830; PMID 32671408): 301 individuals (30 cognitively unimpaired young, 138 cognitively unimpaired elderly, 67 MCI, 54 AD dementia, 12 FTD). MK-6240 discriminated AD dementia from MCI and FTD with ~85-100% accuracy (abstract gives the range only, so sensitivity/specificity left blank). Braak-consistent topographic patterns in 98% of the population; cognition and amyloid status explained most Braak-stage variance (R2=0.75). 68% of amyloid-positive and 37% of amyloid-negative cognitively unimpaired elderly showed tau at least in transentorhinal cortex (Braak I). Context from the abstract: first-generation tau tracers detected Braak I-II tau in only ~20% of cognitively unimpaired elderly versus ~80-90% in post-mortem cohorts. Single-centre research cohort.

Source: doi.org · 2020

• AUC 0.8
• Availability: Limited
• Radiation: None.

What it measures

Regional brain volumes vs normative database; longitudinal atrophy.

When it is used

Adjunct to visual MRI reading in cognitive impairment; NeuroQuant PET (2026) extends to PET quantification.

Medicare coverage

Billed with CPT 0865T/0866T (see mri-volumetry-quantitative-cpt-0865t-0866t).

FDA 510(k) clearances (openFDA): K061855 NEUROQUANT 2006-08-10 (product code LLZ); K170981 2017-09-07; K241098 2024-08-22 (product code QIH 'Automated Radiological Image Processing Software'); K261916 'NeuroQuant PET' 2026-08-12 (LLZ). Accuracy (Brain Behav 2024, doi 10.1002/brb3.3397, PMID 38600026; n=86: 41 subjective/mild cognitive impairment vs 45 dementia): NeuroQuant hippocampal volume AUC 0.80 vs visual MTA rating AUC 0.69 for separating dementia from non-dementia; other regions comparable between methods. Small single-centre sample; reference standard = clinical diagnosis. Cost recorded = OPPS 0865T/0866T rate, not a vendor price.

Source: api.fda.gov · 2006

Cost source: cms.gov

• Availability: Research-only

What it measures

Subtle AD-pattern cortical thinning in asymptomatic adults, as a marker of early neurodegeneration years before dementia.

When it is used

Research prognostic marker; illustrates what structural MRI can detect before symptoms and how small the absolute differences are (~0.2 mm).

Medicare coverage

Research measure; standard MRI brain acquisition (see mri-structural-brain-70551) but the cortical-thickness analysis itself is not a billed service in this record.

SMALL SAMPLES (Dickerson BC, Stoub TR, Shah RC, et al. Neurology 2011;76(16):1395-1402; PMID 21490323; PMC3087406): two independent samples of cognitively normal adults. Sample 1: 8 converters to AD dementia (mean 11.1 years later) vs 25 stable; sample 2: 7 converters (mean 7.1 years) vs 25 stable. AD-signature thinning in converters was ~0.2 mm in both samples (p<0.05) with Cohen d >1. Of 11 people with baseline thickness >=1 SD below the cohort mean, 55% developed AD dementia over nearly a decade; none of 9 with thickness >=1 SD above the mean did. Hazard ratio 3.4 per SD of thinning (p<0.0005). Total converters across both samples = 15, so estimates are imprecise.

Source: doi.org · 2011

• Sensitivity 64%
• Specificity 92%
• Results in Automated result at end of test (duration not sourced).
• Availability: Widespread
• Radiation: None.

What it measures

Objective cognitive function as a screening aid; classification: FDA product code PKQ 'Computerized Cognitive Assessment Aid' (21 CFR 882.1470, Class II): 'objective measurements of cognitive function as a screening aid in the assessment of adults 55 years of age and older ... 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'.

When it is used

Primary-care cognitive screening; annual wellness visit; trial pre-screening (Bio-Hermes).

Medicare coverage

No NCD/LCD loaded. Billing via CPT 96146 / 99483 (see notes).

Guidelines say

AA DETeCD-ADRD 2025 (Alzheimers Dement 21(1):e14335) provides a summary of validated instruments for office cognitive testing; the abstract loaded does not name specific digital devices.

FDA De Novo DEN130033 granted 2015-06-05 (first PKQ device; openFDA). Validation (World J Psychiatry 2020, doi 10.5498/wjp.v10.i1.1, PMID 31956523; n=401, 358 retested 1-2 wk): vs SLUMS <21 (impairment) Cognivue cut-off 54.5 gave NPA 0.92 and PPA 0.64 (recorded as specificity/sensitivity equivalents; comparator is a screening test, not biomarkers or clinical diagnosis); test-retest r=0.90. Biomarker cohort (Bio-Hermes; Sci Rep 2024;14:24519, PMID 39424626; n=887 with amyloid PET: 388 CN, 282 MCI, 217 probable AD): LR+ 2.17, LR- 0.29, diagnostic odds ratio 7.48; sensitivity/specificity not reported in abstract. COST: Medicare has no device-specific payment; digital tests are typically billed as CPT 96146 ('Administration of psychological or neuropsychological test by single standardized instrument via electronic platform with automated result only'): CY2026 OPPS $29.55 (APC 5731); 2024 Medicare Part B national average allowed $2.23 office (8,782 services) / $2.14 facility. Neuropsychological test evaluation 96132: OPPS $381.24; 2024 avg allowed $124.09 office. Cognitive assessment & care plan 99483: OPPS $103.79; 2024 avg allowed $255.96 office (130,107 services). Vendor subscription prices were not obtained.

Source: api.fda.gov · 2015

Cost source: cms.gov

• Sensitivity 77%
• Specificity 74%
• AUC 0.8
• Results in Automated result at end of test (duration not sourced).
• Availability: Limited
• Radiation: None.

What it measures

Visuospatial episodic memory (hippocampal-dependent) as a screening aid (product code PKQ).

When it is used

Memory screening in primary care / memory clinics.

Medicare coverage

No NCD/LCD loaded. Billing via CPT 96146 / 99483.

FDA 510(k) K161328 'Cantab Mobile' cleared 2017-01-13 (PKQ; openFDA). Accuracy (Acta Psychol 2026;269:107597, PMID 42567065): n=65, 15 with MoCA-defined MCI (MoCA <=25); PAL mean-errors-to-success AUC 0.80 (95% CI 0.66-0.95), optimum cut-off 5.8: sensitivity 77%, specificity 74%. Very small sample; reference standard is MoCA, not clinical or biomarker diagnosis. COST: Medicare has no device-specific payment; digital tests are typically billed as CPT 96146 ('Administration of psychological or neuropsychological test by single standardized instrument via electronic platform with automated result only'): CY2026 OPPS $29.55 (APC 5731); 2024 Medicare Part B national average allowed $2.23 office (8,782 services) / $2.14 facility. Neuropsychological test evaluation 96132: OPPS $381.24; 2024 avg allowed $124.09 office. Cognitive assessment & care plan 99483: OPPS $103.79; 2024 avg allowed $255.96 office (130,107 services). Vendor subscription prices were not obtained.

Source: api.fda.gov · 2017

Cost source: cms.gov

• Sensitivity 81%
• Specificity 94%
• Results in Automated result at end of test (duration not sourced).
• Availability: Limited
• Radiation: None.

What it measures

Multi-domain cognitive performance vs normative data.

When it is used

Cognitive screening and care-planning support in primary care.

Medicare coverage

No NCD/LCD loaded. Billing via CPT 96146 / 99483.

FDA status: searches of the openFDA 510(k) database by applicant and device name ('BrainCheck') returned no clearance; FDA approval left blank (company marketing claims not used). (Groppell et al., JMIR Aging 2019;2(1):e12615, PMID 31518280; n=583 community volunteers >49 y, Houston): BrainCheck Memory composite ROC sensitivity 81%, specificity 94% against physician diagnosis / screening tests (physician-diagnosed cohort only n=18; SLUMS cohort n=84). Later agreement study (Front Psychol 2024;15:1428560, PMID 39291174; n=55): correlation with MoCA r=0.77; PPA 84.1%/NPA 81.8% for dementia-or-MCI vs normal. COST: Medicare has no device-specific payment; digital tests are typically billed as CPT 96146 ('Administration of psychological or neuropsychological test by single standardized instrument via electronic platform with automated result only'): CY2026 OPPS $29.55 (APC 5731); 2024 Medicare Part B national average allowed $2.23 office (8,782 services) / $2.14 facility. Neuropsychological test evaluation 96132: OPPS $381.24; 2024 avg allowed $124.09 office. Cognitive assessment & care plan 99483: OPPS $103.79; 2024 avg allowed $255.96 office (130,107 services). Vendor subscription prices were not obtained.

Source: doi.org · 2019

Cost source: cms.gov

• Availability: Widespread

When it is used

The Medicare billing vehicles through which office cognitive testing (including digital instruments) is paid.

Medicare coverage

Covered Part B services; 2024 utilisation: 99483 130,107 office + 9,958 facility services (127k beneficiaries); 96146 8,782 office services.

2024 Medicare Part B national average allowed amounts (physician/supplier): 99483 $255.96 office / $180.23 facility (avg submitted charge $485.68 office); 96146 $2.23 office; 96132 $124.09 office; 96138 $33.08 office. CY2026 OPPS: 99483 $103.79 (APC 5822), 96146 $29.55, 96132 $381.24, 96138 $456.40.

Source: data.cms.gov · 2024

• Sensitivity 77.7%
• Specificity 90.4%
• Availability: Limited
• Radiation: Label: effective dose 21.3 microSv/MBq; 3.94 mSv for a 185 MBq (5 mCi) dose in an adult.

What it measures

Striatal dopamine transporter density; low uptake indicates nigrostriatal degeneration seen in dementia with Lewy bodies (DLB) and Parkinsonian syndromes but not in Alzheimer's disease.

When it is used

Label: adjunct to other diagnostic evaluations in adults with suspected Parkinsonian syndromes or suspected DLB. Used to separate DLB from AD, which imaging of amyloid or tau does not do.

Medicare coverage

No NCD located for ioflupane SPECT in this session; A9584 is separately payable in the hospital outpatient setting (OPPS status indicator K, APC 0781) per the CY2026 January Addendum B. Coverage determined by MACs.

Guidelines say

Abstract of the phase III study: the revised International Consensus Criteria for DLB recommended that low DAT uptake in the basal ganglia on SPECT or PET be a 'suggestive feature' for diagnosis.

(McKeith I, O'Brien J, Walker Z, et al. Lancet Neurol 2007;6(4):305-313; PMID 17362834): phase III multicentre study of 326 patients with consensus-panel clinical diagnoses (94 probable DLB, 57 possible DLB, 147 non-DLB dementia, mostly AD; 28 undiagnosed). Three blinded readers: mean sensitivity 77.7% for probable DLB, specificity 90.4% for excluding non-DLB dementia, overall accuracy 85.7%, PPV 82.4%, NPV 87.5%, inter-reader kappa 0.87. Reference standard was clinical consensus, not autopsy. LABEL data from openFDA (DaTscan SPL effective 2026-04-24, NDA022454): indication, dose, timing and dosimetry as stated above. COST: the typical cost is the SUM of CY2026 OPPS national rates (January 2026 Addendum B): CPT 78803 SPECT single area $554.73 (APC 5592) + HCPCS A9584 ioflupane $1,387.74 (SI K); using 78830 SPECT/CT instead gives $1,322.69 + $1,387.74 = $2,710.43. Physician-fee-schedule and Part B measured averages were not loaded.

Source: doi.org · 2007

Cost source: cms.gov