Multiple Errands Test

Last Updated

July 26, 2023

Purpose

The Multiple Errands Test measures how impairments in executive performance affect functioning in natural contexts.

Link to Instrument

Instrument Details

Acronym MET

Area of Assessment

Executive Functioning
Activities of Daily Living
Attention & Working Memory
Occupational Performance
Life Participation

Assessment Type

Performance Measure

Cost

Free

Cost Description

VMET requires GestureTek® IREX™ Systems, approximately $15,000

CDE Status

Not a CDE—last searched 8/25/2023.

Neurological Disorders

Stroke

Parkinson's Disease

Alzheimer's Disease and Progressive Dementia

Mixed Populations

Additional versions of the MET have been created in addition to the original, including:
--Baycrest MET Revised Version (BMET-R, 2007)
--Multiple Errands Test Revised (MET-R, 2013)
--VR Version (VMET, 2014)
--Multiple Errands Test - Contextualized Version (MET-CV, 2018)
--Home Version (MET-Home, 2019)
--Big Store Version (2019)
--Baycrest MET Version (BMET) or Chinese-MET (2020)
In all versions, at least 12 tasks must be completed. The MET-R requires task set-up prior to test administration—tutorial videos guide the selection of site-specific tasks. The original version was developed by Shallice and Burgess in 1991. Its complexity necessitated its use with high functioning patients. The MET-R and MET-HV were adapted to be used in a controlled environment, such as hospital grounds or a shopping mall. The VMET is done virtually. Therefore, it can be used with clients who have motor impairments that preclude them from the extensive ambulation needed to complete the assessment. It utilizes a video capture virtual supermarket programmed into a virtual reality system which must be purchased.
In all versions, tasks are broken down into sets:
--Errands (purchasing items, using a telephone and mailing a letter)
--Obtaining information (closing time of local library, price of a candy bar, finding a specified flyer and keeping it with them until the BMET-R is completed),
--Meeting the assessor at a designated time and place (meet at gift shop 10 minutes after starting)
--Informing the assessor when task is completed
The participant is required to complete these tasks while adhering to 6-8 rules (depending on the version) such as:
--Adhering to spending limits with purchases
--Not purchasing more than a certain number of items
--Not visiting the same area twice
--Avoiding restricted areas and communication with others
The participant is given the rules before beginning. The tester observes the participant and records behavior during performance. For the hospital version, this is done at a distance.
The score is based on the total number of errors recorded and the time it took the participant to complete the assessment. Errors are categorized in several areas: inefficiencies, rule breaks, misinterpretation of task requirements, task failures (not completing any of 12 tasks).
Errors are categorized in several areas: inefficiencies, rule breaks, misinterpretation of task requirements, task failures (not completing any of 12 tasks)
MET Scores Possible Range: Inefficiencies (0-9), Rule Breaks (0-9), Task Failures (0-12), Total errors (0-30)

Number of Items

12 tasks with 6-8 rules

Wristwatch (with GPS tracking for some versions)
Carrier Bag
Money (amount specified by location of test)
Pen
Clipboard
Map
Test instructions
Test packet or Task List
Scoring sheet
Neuro VR Software and Player (for VMET)

45-60 minutes

Required Training

No Training

Required Training Description

No required training but two examiners are required to be present during test administration.

Adolescent

13 - 17

years

Adult

18 - 64

years

Elderly Adult

65 +

years

Instrument Reviewers

Mary Elizabeth Patnaude, MS, OTR/L

Galila Dandridge, OT Student, UIC

Luz Rocha, OT Student, UIC

Maya Soto, OT Student, UIC

Melissa Wen, OT Student, UIC

Susan Magasi, PhD, Associate Professor, Departments of Occupational Therapy and Disability Studies, UIC

Body Part

Head

ICF Domain

Participation
Environment
Activity

Measurement Domain

Participation & Activities
Activities of Daily Living
Cognition

Professional Association Recommendation

None found—last searched 8/25/2023.

Considerations

Executive functioning must be evaluated in real life settings.
There are multiple versions of the Multiple Errands Test.
The test requires that the participant can ambulate through a natural context for 60 minutes. The VMET is done virtually and can be used with individuals who have motor impairments.
There is minimal research supporting its use with clients who have suffered an acute brain injury or acute stroke since they may be too medically fragile to participate in the assessment.
Assessing its reliability and validity has been difficult since the measure must be tailored to the environment in which it is administered.
The system in which the virtual mall is acquired, the GestureTek® IREX™ costs approximately $15,000, and may be cost prohibitive for some settings to acquire.

Interrater/Intrarater Reliability

Acute Stroke (MET-R): (Morrison et al, 2013; mCVA group n = 25, Mean Age = 60 (10.8) during acute inpatient stay; control group n = 21, Mean Age = 59.9 (15.50); Mean time post CVA = 6 months)

Excellent interrater reliability (ICC = 1)

Stroke (MET-Home): (Burns et al., 2019; n = 23; Mean Age (SD) = 56.7 (10.6) years; Mean Time Post CVA: ≥ 90 days post-stroke; Mild to Moderate Severity; English speaking)

Excellent interrater reliability during MET–Home piloting (ICC ≥ .80)
Excellent interrater reliability for subscores (ICC's ranged from .88 to .96)

Internal Consistency

Chronic Stroke (MET-HV): (Knight et al, 2002)

Excellent internal consistency (Cronbach's alpha = 0.77)

Stroke (MET-Home): (Burns et al., 2019)

Acceptable: Cronbach’s alpha ranged from .68 to .74 for each of the items on the MET-Home task list (overall list of 14 items; ⍺ = .73).

Criterion Validity (Predictive/Concurrent)

Concurrent validity:

Chronic Stroke (MET-HV) (Knight et al, 2002)

Adequate correlation with the Behavioral Assessment of Dysexecutive Syndrome Battery (BADS) (r = -.57)

Acute Stroke (MET-R) (Morrison et al, 2013)

Adequate correlation between the MET-R task completion score and the Executive Function Performance Test (EFPT) total score (r = -.55).

Stroke (MET-Home): (Burns et al., 2019)

Adequate correlations between the MET-Home Accurately Completed (r = -0.49, p < 0.05) and Omissions (r = 0.46, p < 0.05) subscores and EFPT Total Score.
Adequate correlations between MET-Home Accurately Completed (r = 0.48, p < 0.05) and Omissions (r = 0.59, p < 0.01) subscales with EFPT Organization subscale.
Adequate correlations between MET-Home Passes (r = 0.48, p < 0.05) and Rule Breaks (r = 0.57, p < 0.01) subscales with EFPT Sequencing subscale.

Predictive validity:

Acute Stroke (MET-HV) (Maier, Kraus & Katz, 2010; n = 30; Mean Age = 53.6 (15.04), Mean Days in Rehabilitation = 54.6 (20.31))

Adequate predictive validity of the MET-HV at acute rehabilitation hospital discharge at predicting Participation Index (M2PI) scores at 180 days post discharge (r = 0.4) (Maier et al, 2010)

Construct Validity

Convergent Validity:

Stroke (MET-Home): (Burns et al., 2019)

Adequate convergent validity between the MET-Home subscores for Accurately Completed (p < .05), Inefficiencies (p < .01), and Rule Breaks (p < .05) and the Symbol Digit Modulation Test (SDMT).
Adequate convergent validity between the MET-Home subscore for Passes and the Delis-Kaplan Executive Function System (D-KEFS) Tower Test – Rule Violations subscale (p < .05).

Discriminant Validity:

Chronic Stroke (MET-SV) (Alderman, 2003; Group 1: n = 46, mean age = 29.2 (8.5); Group 2 (Neurologically impaired): n = 50, mean age 34.6 (12.7), mean time since injury 72.1 (68.4) months.

Adequate construct validity of the MET–SV to differentiate participants with chronic stroke/TBI from healthy participants (r = .44)

Acute Stroke (MET-R): (Morrison et al., 2013).

Significant differences between control and mCVA groups on MET-R component scores for total tasks completed (p < .001), total number of rule breaks (p < .001), and performance efficiency (p < .002)

Interrater/Intrarater Reliability

Traumatic Brain Injury/Chronic Stroke (MET-HV): (Knight, Alderman & Burgess, 2002; n = 22, Mean Age = 35.6 (11.3) years, Mean Time Post-Injury = 80.9 (62.6) months)

Good inter-rater reliability (ICC = .81) for interpretation failures
Excellent inter-rater reliability (ICC = 1) for rule breaks

Construct Validity

Convergent validity

People with Substance Dependence (PWSD): (Valls-Serrano et al., 2018; n = 90; mean age = 35.88 (8.91); PWDS group n = 58; all participants fluent in Spanish; MET-Contextualized Version (CV))

MET-CV scales correlated with traditional neuropsychological tasks in polysubstance users:
- Adequate convergent validity between Task Failures and Letter Number Sequencing Test (r = -.341, p < .001)
- Adequate convergent validity between Rules Breaks and Zoo Map Test (r = -.309, p < .001)
- Poor convergent validity between Interpretation Failures and Stocking of Cambridge Task (r = .273, p < .05)
MET-CV scales correlated with drug use variables:
- Adequate convergent validity between Rule Breaks and abstinence (r = −0.306; p < .001) (Definition of rule breaks = when a specific social rule has been broken; for example, spoke to examiner or argue with staff)
- Adequate convergent validity between Rule Breaks and the Severity of Consumption Index (r = 0.453, p < .001)
- Poor convergent validity between Rule Breaks and Interpretation Failures (r = .269, p < .001) (Definition of interpretation failures = where the requirements of a task have been misunderstood; for example, buy products not ordered)

Content Validity

Chronic Stroke/TBI (VMET): (Rand et al, 2009; n = 9, Mean Age = 64.2 (7.7); Time Post-Stroke = 4-72 months)

Excellent content validity of the MET (r = -.93) and VMET (r = -.87) with the Zoo Map subtest from the Behavioral Assessment of Dysexecutive Functioning Syndrome (BADS) assessment of executive functioning

Face Validity

Chronic Stroke/TBI: (Rand et al, 2009)

Excellent face validity of MET (r = -.76) and VMET (r = -.82) with measure of independence in IADL.

Responsiveness

Scores on modified version of MET significantly improved in the area of Task Failures, after a goal oriented training for executive functioning. This represents minimal evidence of responsiveness (Novakovic-Agopian et al; Poulin et al, 2013). No effect size was reported.

Internal Consistency

Schizophrenia (MET-1991 Version): (Bulzacka et. al., 2016; n = 100; mean age = 30.9 (9.4); mean duration of the illness = 7.7 (6.1) years; schizophrenia group n = 75, schizoaffective disorder group n = 25); 81% male; French Sample)

Adequate: Cronbach's alpha = 0.75 (average inter-item covariance: 13.2)

Criterion Validity (Predictive/Concurrent)

Concurrent validity:

Schizophrenia (MET-1991 Version): (Bulzaka et. al., 2016)

Executive Functioning Tests:
- Adequate correlations with the Wisconsin Card Sorting Test-64 (WCST)
  - WCST total correct: ρ = 0.3, p = 0.003
  - WCST perseverative: ρ = 0.3, p = 0.003
  - WCST categories completed: ρ = 0.3, p = 0.004).
- Adequate correlation association with the Paper version Errands Test (PET; ρ = 0.4, p = 0.0001)
Functional Outcomes:
- Adequate correlations with Social Autonomy Scale (SAS), with the MET composite task score and the global error score being the strongest associations (ρ = 0.39; p = 0.0001 for both)

Standard Error of Measurement (SEM)

Dementia (mild to moderate) population (Chinese-MET): (Lai et al., 2020; n = 160; mean age = 70.23 (8.2); mild to moderate dementia; All participants fluent in Mandarin)

SEM for Dementia group from the Chinese-MET (n = 20; mean age = 69.25 (16.32)): 3.649