Alzheimer’s Drug Therapy Initiative: Seniors Medication Study (SMS) Ging-Yuek Robin Hsiung, MD MHSc FRCPC FACP Assistant Professor Division of Neurology,

Alzheimer’s Drug Therapy Initiative: Seniors Medication Study (SMS)

Ging-Yuek Robin Hsiung, MD MHSc FRCPC FACPAssistant Professor

Division of Neurology, Dept. of Med,

Malcolm Maclure, ScD, ProfessorDept. Anesthesiology, Pharmacology & Therapeutics

University of British Columbia

Disclosure

• Dr. Hsiung have received research support from Baxter, Bristol-Myers-Squibbs, Elan, Janssen-AI, Pfizer, Hoffman-La Roche, and Genentech as an Alzheimer clinical trial centre at UBC

• None of the medications discussed today are related to any of the clinical trials sponsored by the above

Drug Payer Policy: Coverage with Evidence Development

“In April 2005, the Centers for Medicare and Medicaid Services (CMS) posted on its Web site a draft guidance document describing a new approach to coverage policy called Coverage with Evidence Development (CED). CED offers an option for coverage of promising drugs, biologics, devices, diagnostics, and procedures that would not otherwise meet Medicare’s evidentiary standards for being ‘reasonable and necessary’. ”

Alzheimer’s Drug Therapy Initiative

“Provides eligible B.C. residents with PharmaCare coverage of medications used in the treatment of Alzheimer’s disease while gathering evidence on their effectiveness.”

Original RCT Trial Design of SMS: Continuing vs. Stopping vs. Switching

Initial Estimated N of 3000Positive Response

Indeterminate Response

Non-Response

Continue

Stop

Switch

RandomizationDx 6 mo

Continuing users

Stoppers due to intolerance

30,000

2,000

10,000

Unwilling

12,000

New AD cases

10,000

Existing AD, new users

10,0003000

3000

3000

1,000

1,000

1,000

Seniors Medication Study – stumbling blocks page 1

• No incentives for patients or their doctors to participate in trial. Coverage available to all who fit clinical criteria.

• One ethics committee argued it would be coercive for PharmaCare to limit coverage of drug costs to patients who agreed to be in the trial.

• Another ethics committee judged it unethical to do a small trial that had insufficient power to rule out a small but clinically important effect. A drug plan might interpret an inadequately powered study that is ‘not significant’ as meaning ‘evidence of no effect’.

Seniors Medication Study – stumbling blocks page 2

• We took 1 year to develop and test tools and protocol for randomization in busy general practictioners’ office

• BC Medical Association opposed Special Authority form without financial compensation

• Only a few patients were randomized after 12 months.• A Medical Services Plan billing code was created: fees for

referrals and Clock Drawing Tests. These had only a small impact on randomization, but increased enrolment into the plan.

• Fair Pharmacare is not “free”, but based on income tax• SMS became an observational cohort study with

validation/reproducibility study of Standardized Mini-Mental State Exam and various scales.

Current Observational Study (SMS)N ~ 200

Positive Response

Indeterminate Response

Non-Response

Continue

Stop

Switch

Doctor & Patient Choice

Initiation of treatment to one of three ChEIs (doctor & patient choice)

6 mo

Eligible AD cases

Seniors Medication Study (n~200)

Start ofCoverage 6 mo 12 mo 18 moSMMSE SMMSE SMMSE SMMSE(10-26) OPAR & GDS OPAR & GDS OPAR & GDSGDS Clock* Clock* Clock* TICS* TICS** TICS**

*Clock Drawing Test **Telephone Interview of Cognitive StatusSMMSE = standardized Mini Mental State ExaminationOPAR = Overall Patient Assessment Rating GDS = Global Deterioration Scale

Average SMMSE vs. yrs of schooling at baseline/enrolment

SMMSE vs TICS (6 mon)

SMMSE

TICS

Correlation = 0.59

Low SMMSE means low TICS

(i.e. good negative predictive value)

Correlation of SMMSE with Components of the Telephone

Interview of Cognitive Status (TICS)

TICS (n=111)SMMS

EOrien-tation

Lan-guage Recall

Compre-hension

Orientation 0.49

Language 0.34 0.30

Recall 0.47 0.41 0.31

Comprehension 0.47 0.42 0.22 0.50 Attention, Calculation 0.42 0.48 0.36 0.39 0.38

SMMSE vs Clock Drawing Test

SMMSE

Clock Drawing Test

Correlation = 0.58

Clock Drawing Test vs. TICS

Clock Drawing

Test

TICS

Correlation = 0.57

SMMSE vs GDS: low correlation

SMMSE

GDS

Correlation = - 0.22

5

4

TICS vs GDS: low correlation

GDS

TICSCorrelation

= - 0.18

Clock Test vs GDS: low correlation

Clock Drawing Test

GDS

Correlation = - 0.15

Correlations among measures

SMMSE

GDS

TICS

Clock Drawing

n = 122

n = 122

n = 111

n = 74

0.59

0.58

- 0.220.57

- 0.15

- 0.18

Lessons learned• RCT in “phase 4” setting is not feasible – but can do observational

study for effectiveness analysis• Limited data collection in real life doctor’s office – can only be a

brief test; need to compromise between efficiency & comprehensiveness

• Clock Drawing Tests conducted by GPs, faxed to researchers for scoring, enabled patients to be assessed without confounding by financial incentives or telephone/hearing problems.

• Good correlation between SMMSE & TICS, SMMSE & CDT, CDT vs TICS, (validate use of SMMSE in doctor’s office setting)

• Low correlation between GDS vs TICS, and GDS vs CDT: the test measure different domain – Functional vs Cognitive

• Limitation of using SMMSE only as the criterion for eligibility for drug funding and outcome measure

Further Studies• Await long term outcome (12 & 18 months) of the

study• Compare trajectory of changes in each outcome

measure by different ChEIs• SMMSE can be applied in real life with good

validity, but limited domains• Will need to include multiple eligibility criteria

(SMMSE, CDT, GDS) and multiple outcome measurements