Mining for Lost Memories: A Best Practice Approach for Alzheimer s Disease Diagnosis

Transcription

1 Mining for Lost Memories: A Best Practice Approach for Alzheimer s Disease Diagnosis Michael Rosenbloom, MD and Terry Barclay, PhD HealthPartners Center for Dementia and Alzheimer s Care

2 Presentation Outline I. Diagnostic Challenges in Dementia II. Rational for Early Detection III. Practice Tips I. Typical and atypical dementia work-up IV. Sensitivity/Specificity of Popular Cognitive Screening Tools V. HealthPartners 6 th Vital Sign Project VI. Clinician Toolbox

3 Alzheimer s Diagnosis: Current Limitations Epidemiological Concerns Growing Epidemic Expected to Impact 13M Americans by 2050 and consume 40% of Healthcare Spending AD Diagnostic Limitations Diagnostic Accuracy 50% for AD Diagnosis Delayed by 6 Years Impairment of IADLs at Diagnosis

4 Beyond mountains, there are mountains. Haitian Proverb

5 Diagnostic Challenges Cultural Ageism and defining normal aging Fear of delivering bad news Associated stigma Implications for physician/patient relationship Medical Heterogeneous baseline Time No lab test Wrong diagnosis Treatment options/efficacy

6 Rational for Early Detection 1) Improve quality of life 2) Identify reversible causes 3) Maximize treatment response 4) Prevent diagnosis during crises 5) Identify strategies for medication adherence 6) Promote independence, lifestyle changes 7) Reduce cost of care by decreasing hospitalizations and ER visits

7 AD Diagnosis History Baseline premorbid cognitive state First symptom quality and onset Address memory, language, visuospatial, and executive function Evaluate AD risk factors Assess functional impairments Supplemental history from family/friend Neurological Exam Basic mental status Cranial nerve, motor, coordination, and gait testing to address pyramidal, extrapyramidal, or cerebellar tract findings

8 General Practice Tips Raise your expectation of the older patient Clinical interview Let patient answer questions without help Remember: Social skills remain intact Easy to be fooled by a sense of humor, irritability, reliance on old memories, or quiet/affable demeanor Subjective interviews FAIL to detect dementia in early stages

9 General Practice Tips Red flags Repetition (not normal in 7-10 min conversation) Tangential, circumstantial responses Losing track of conversation Frequently deferring to family Unexplained weight loss or failure to thrive Inattentive to appearance Unable to adapt to stressful circumstances (hospitalization)

10 General Practice Tips Family observations: ANY instances whatsoever of getting lost while driving, trouble following a recipe, asking same question repeatedly, mistakes paying bills Ask: Let s suppose your family member was alone on a domestic flight across the country and the trip required a layover with a gate change. Would he/she be able to manage that kind of mental task on his/her own?

11 Mental Status Exam Intact older adult should be able to: Describe 2 current events in some detail Describe what happened on 9/11, New Orleans disaster Name the current President and 2 immediate predecessors Describe medical history and names of some medications

12 AD Diagnosis Laboratory Studies CBC, BMP, LFTs, TSH, B12, Ca2+ Neuroimaging Brain MRI/CT

13 Dementia with Lewy Bodies Clinical Features Parkinsonism, hallucinations, cognitive fluctuations Neuropsychological Testing Predominant visuospatial dysfunction with relative sparing of verbal memory Neuroimaging Occipital hypometabolism on FDG-PET Non-specific MRI atrophy pattern

14 Frontotemporal Dementia Clinical Features Typical onset prior to age 65 Behavioral symptoms Disinhibition, apathy, loss of empathy, repetitive stereotyped movements, hyperorality Language symptoms Expressive aphasia, surface dyslexia Neuropsychological Testing Impairments on executive function/language with relative sparing of episodic memory and visuospatial function Neuroimaging Atrophy of frontal and anterior temporal cortex

15 Cognitive Screening

16 Objective Screening Measures Initial considerations Research findings re: early detection Balance b/w time and sensitivity/specificity How will your practice incorporate screening? Who will administer tests? MDs, Nurses, social workers, allied health professionals What happens once screen is positive?

17 Objective Screening Measures Wide range of options Mini-Cog Mini-Mental State Exam (MMSE) St. Louis University Mental Status Exam (SLUMS) Montreal Cognitive Assessment (MoCA) All but MMSE free online, in public domain AD Association Physician Toolkit

18 Objective Screening Measures Do NOT Allow patient to give up prematurely or skip questions Deviate from standardized instructions Offer multiple choice answers Bias score by coaching Be soft on scoring Score ranges already padded for normal errors Deduct points where necessary be strict

19 Mini-Cog Contents Verbal Recall (3 points) Clock Draw (2 points) Subject asked to recall 3 words Leader, Season, Table +3 Advantages Quick (2-3 min) Easy High yield (executive function, memory, visuospatial function) Subject asked to draw clock, set hands to 10 past 11 +2

20 Mini-Cog Pros Takes only 2-3 minutes to administer Clock drawing sensitive to both visuospatial & executive dysfunction Simple scoring and interpretation Cons Not considered as sensitive for MCI or early dementia when compared to longer screens Brevity means less information to interpret

21 Mini-Cog Performance unaffected by education or language Borson Int J Geriatr Psychiatry 2000 Sensitivity and Specificity similar to MMSE (76% vs. 79%; 89% vs. 88%) Borson JAGS 2003 Does not disrupt workflow & increases rate of diagnosis in primary care Borson JGIM 2007 Failure associated with inability to fill pillbox Anderson et al Am Soc Consult Pharmacists 2008

22 MMSE

23 MMSE Pros Widely accepted and validated tool for dementia screening 30-point scale well known and score is easily interpretable Measures orientation, working memory, recall, language, praxis Cons Scale developed 40 years ago, before MCI criteria and when early dementia less well understood Lacks sensitivity to MCI and early dementia Takes 7 min. to administer Copyright issues

24 SLUMS

25 SLUMS Pros More measures of executive functioning Good balance between easy and difficult items More sensitive than MMSE in detecting MCI and early dementia 30-point scale similar to MMSE Score range for MCI and dementia Free online Cons Takes 10 min. to administer Slightly more complex directions than MMSE Less name recognition than MMSE

26 MoCA

27 MoCA Pros Much more sensitive than MMSE in detecting MCI and early dementia More content tapping higher level executive functioning 30-point scale similar to MMSE Translations available in 35+ languages Free online Cons Takes min. to administer More complex administration and directions than MMSE

28 Screening Tool Selection Montreal Cognitive Assessment (MoCA) Sensitivity: 90% for MCI, 100% for dementia Specificity: 87% St. Louis University Mental Status (SLUMS) Sensitivity: 92% for MCI, 100% for dementia Specificity: 81% Mini-Mental Status Exam (MMSE) Sensitivity: 18% for MCI, 78% for dementia Specificity: 100% Larner et al Int Psychogeriatr 2012; Nasreddine et al J Am Geriatr Soc 2005; Tariq et al Am J Geriatr Psychiatry 2006; Ismail et al Int J Geriatr Psychiatry 2010

29 What about cognitive screening in patients without a history of cognitive disorder?

30 The 6 th Vital Sign Project HealthPartners Clinical Initiative to Address the Following: 1) Delays in AD diagnosis and growing geriatric population 2) Impaired management of chronic disease & medications in this population HTN Diabetes Hyperlipidemia Anticoagulation management

31 Lessons from Chronic Disease

32 MiniCog: Prior Studies Borson and colleagues administered MC to 524 patients 65 in primary care setting Screening did not disrupt clinic flow 18% screen failure rate (MC score<4) Only 17% of providers took appropriate action with screen fails» Borson et al. J. Gen. Intern. Med 2007 McCarten and colleagues administered MC to 8,342 patients aged 70 in VA setting Screen well-accepted by older veterans Testing completed between 1-3 minutes 25.8% failure rate among asymptomatic population» McCarten et al J Am Geriatr Soc

33 6 th Vital Sign Project: Clinical Setting HealthPartners Medical Group 640 physician multispecialty care delivery organization Health plan includes 750,000 members Tertiary care hospital serving an inner city population

34 6 th Vital Sign Project: Methods Rooming nurse staff Identifies patients aged 70 without history of MCI or dementia in chart review Administers MC to patients Alert provider for screen failures Physician provider Assesses patient for cognitive impairment Decides upon further diagnostic evaluation and/or referral

35

36 6 th Vital Sign Project: Preliminary Data Demographics Total Screened (246) Mean Age (78.5 yrs) %Male/Female (42.1%/57.9%) Results #Passes (151) #Fails (95) %Failures (38.6%)

37 6 th Vital Sign Project: Future Directions Clinical Goals Expand to ophthalmology in March 2012 Expand to a primary care clinic in April 2012 Research Goals Investigate screen feasibility and failure rate Retrospectively compare chronic disease management between screen passes and fails Prospectively compare markers of chronic disease management pre- and post-mc screen among screen failures

38 6 th Vital Sign Project: Research Plan Development of Screening Process Pilot Test Screen in Specialty Care Clinic >70 yr Pilot Test Screen in Primary Care Clinic >70 yr Expand Screen to Primary Care Throughout Primary Care Analysis 1. Measure prevalence of Mini-Cog screen fails and subsequent diagnosis of dementia 2. Measure feasibility of Mini-Cog as a screen in clinical care and acceptance of screen by nurses, providers, and leaders in each clinic 3. Identify impact of undiagnosed cognitive impairment on management of chronic disease and cost of care in each patient population by comparing 1 yr retrospective data from patients passing versus failing the screen 4. Identify potential benefits of screening on clinical outcomes and cost of care in each patient population by comparing 1 yr retrospective with 1 yr prospective data in screen fails

39 6 th Vital Sign Project: Limitations and Concerns Is the MC the Appropriate Screen? Does the MC favorably impact clinical care? What does one do with an abnormal MC screening test? How does one build the infrastructure to effectively handle patients failing the MC?

40 Clinician Toolbox Cognitive Impairment Detection Algorithm

41

42 Clinician Toolbox Standard Memory Loss Work-Up

43 Dementia Work-Up Provider Checklist Follow these diagnostic guidelines in response to patient failure on cognitive screening (e.g., Mini-cog) or other signs of possible cognitive impairment. History and Physical Review onset, course, and nature of memory and cognitive deficits (Alzheimer s Association Family Questionnaire may assist) and any associated behavioral, medical or psychosocial issues Assess ADLs and IADLs, including driving and possible medication and financial mismanagement Conduct structured mental status exam (e.g., MoCA, SLUMS, MMSE) 1 Assess mental health (consider depression, anxiety, chemical dependency) Perform neurological exam focusing on focal/lateralizing signs, vision, including visual fields and extraocular movements, hearing, speech, gait, coordination, and evidence of involuntary or impaired movements Diagnostics 1. Routine lab tests CBC, lytes, BUN, Cr, Ca, LFTs, Glucose 2. Neuroimaging CT or MRI when clinically indicated Dementia screening labs: TSH, B12 Contingent labs (per patient history) RPR or MHA-TP, HIV, heavy metals 3. Neuropsychological testing Indicated in cases of early or mild symptom presentation, for differential diagnosis, determination of nature and severity of cognitive functioning, and/or development of appropriate treatment plan Typically maximally beneficial in the following score ranges: MoCA 19-27; SLUMS 18-27; MMSE Diagnosis Mild Cognitive Impairment Mild deficit in 1 cognitive function: memory, executive, visuospatial, language, attention Intact ADLs and IADLs; does not meet criteria for dementia Alzheimer s disease Most common type of dementia (60-80% of cases) Memory loss, confusion, disorientation, dysnomia, impaired judgment/behavior, apathy/depression Dementia with Lewy Bodies / Parkinson s dementia Second most common type of dementia (up to 30% of cases) Hallmark symptoms include visual hallucinations, parkinsonism, and fluctuations in cognition Frontotemporal dementia Third most common type of dementia primarily affecting individuals in their 50s and 60s EITHER marked changes in behavior/personality OR language variant (difficulty with speech production or loss of word meaning) Vascular dementia Relatively rare in pure form (6-10% of cases) Symptoms often overlap with those of AD; frequently there is relative sparing of recognition memory Family Meeting Include family care partners Review diagnosis and intervention checklist (attached) Refer to Alzheimer s Association ( or

44 Clinician Toolbox Intervention Checklist

45

46 Acknowledgements N. Bud Grossman Center for Memory Research and Care (UMN) Karen Ashe Kathleen Zahs Kamakshi Lakshminarayan Melanie Kiihn Minneapolis VA GRECC J. Riley McCarten James Cleary Maurice Dyken

47 The Center for Dementia and Alzheimer s Care Questions?