Arnadottir OT-ADL Neurobehavioural Evaluation (A-ONE)

Evidence Reviewed as of before: 09-01-2012
Author(s): Annabel McDermott, OT
Editor(s): Nicol Korner-Bitensky, PhD OT

Purpose

The Arnadottir OT-ADL Neurobehavioural Evaluation (A-One) evaluates the impact of neurobehavioural impairment on functional performance of activities of daily living (ADL).

In-Depth Review

Purpose of the measure

The Arnadottir OT-ADL Neurobehavioural Evaluation (A-One) is a standardized, performance-based measure that identifies the impact of neurobehavioural impairment on functional performance of ADL. The measure allows observation of ADL and evaluation of the level of assistance required for ADL performance (Arnadottir et al., 2009). Accordingly, the A-ONE provides the therapist with an ecologically-relevant assessment of the consequences of neurobehavioural impairments through clinical observation of ADL tasks using a ‘top-down’ (occupation-based) approach (Arnadottir et al., 2009; Bottari et al., 2006; Carswell et al., 1992; Cooke et al., 2006).

The A-ONE is comprised of two parts: (a) assessment of the individual’s independence in ADL tasks and the type of assistance required; and (b) identification of the type and severity of neurobehavioural impairment that is limiting the individual’s independence in these tasks (Gardarsdottir & Kaplan, 2002).

The A-ONE can be used to assist therapists in goal setting and treatment planning (Gardarsdottir & Kaplan, 2002).

Available versions

The A-ONE was previously named the Arnadottir Occupational Therapy – ADL (OT-ADL) Neurobehavioural Evaluation.

Features of the measure

Items:

The A-ONE is comprised of 2 scales: the Functional Independence scale, more commonly referred to as the Activities of Daily Living Scale (ADL scale), and the Neurobehavioural Impairment scale (NBI scale).

The ADL scale measures 5 ADL domains (dressing; grooming and hygiene; transfers and mobility; feeding; and communication) using 20 everyday tasks.

1. Dressing

  • i. Put on shirt
  • ii. Put on pants
  • iii. Put on socks
  • iv. Put on shoes
  • v. Manipulate fastenings

2. Grooming and hygiene

  • i. Wash face
  • ii. Comb hair
  • iii. Brush teeth
  • iv. Shave beard/apply cosmetics
  • v. Perform toilet hygiene
  • vi. Bathe

3. Transfers and mobility

  • i. Sit up in bed
  • ii. Transfer from sitting
  • iii. Maneuver around
  • iv. Transfer to toilet
  • v. Transfer to tub

4. Feeding

  • i. Drink from glass/cup
  • ii. Use fingers to bring food to mouth
  • iii. Bring food to mouth by fork or spoon
  • iv. Use knife to cut and spread

5. Communication

The NBI scale consists of items to assist the therapist in identifying the probable site of cortical dysfunction based on the observed neurological behaviours. The NBI scale is comprised of 2 subscales:

1. The Specific Subscale Impairment (NBSIS) subscale

  • i. Motor apraxia
  • ii. Ideational apraxia
  • iii. Unilateral body neglect
  • iv. Spatial relations
  • v. Unilateral spatial neglect
  • vi. Organization and sequencing
  • vii. Perseveration
  • viii. Topographical disorientation (transfers and mobility)
  • ix. Sensory aphasia (communication)
  • x. Anomia (communication)
  • xi. Paraphasia (communication)
  • xii. Expressive aphasia (communication)

2. The Pervasive Subscale Impairment (NBPSI) subscale

  • i. Lability
  • ii. Apathy
  • iii. Depression
  • iv. Irritability
  • v. Frustration
  • vi. Restlessness
  • vii. Insight
  • viii. Judgement
  • ix. Confusion
  • x. Attention
  • xi. Distraction
  • xii. Initiative
  • xiii. Motivation
  • xiv. Performance latency
  • xv. Working memory
  • xvi. Confabulation

Description of tasks:

The therapist observes the patient performing the listed ADL tasks and determines the level of assistance required to complete the tasks (see scoring below). Errors in task performance are an indication of underlying neurobehavioural impairments. Different neurobehavioural impairments manifest as different errors or difficulties in ADL task performance. The therapist observes for the presence and severity of neurobehavioural impairments, according to how much the impairment impacts on the individual’s ability to perform the ADL task independently (Gardarsdottir & Kaplan, 2002; Arnadottir et al., 2009).

What to consider before beginning:

The A-ONE should be performed in the clinical setting (Bottari et al., 2006).

Scoring and Score Interpretation:

The ADL and the Neurobehavioural linear scales were developed as criterion-referenced rating scales of the ordinal type by application of Rasch analysis (Arnadottir & Fisher, 2008; Arnadottir et al., 2010).

The ADL scale measures the individual’s need for assistance to overcome neurobehavioural impairments during ADL task performance. Arnadottir et al. (2008) examined the original 5-point rating scale structure and noted that thresholds were disordered. This disorder was eliminated when score 2 (verbal assistance) and score 3 (supervision) were combined, resulting in a 4-point rating scale:

0 = Full assistance needed
1 = Minimum to considerable physical assistance needed
2 = Verbal assistance/supervision needed
3 = Independent

Scores can be added within each ADL domain, but total ADL scores should not be added. Individuals are not penalized for using assistive devices when performing ADL tasks (Arnadottir et al., 2008).

Scoring of the NBI scale is based on the extent to which the neurobehavioural impairment interferes with ADL task performance, not the severity of the impairment. Most items of the NBSIS are rated several times:

  1. Motor apraxia*
  2. Ideational apraxia*
  3. Unilateral body neglect*
  4. Spatial relations*
  5. Unilateral spatial neglect*
  6. Organization and sequencing*
  7. Perseveration^
  8. Topographical disorientation (transfers and mobility)
  9. Sensory aphasia (communication)
  10. Anomia (communication)
  11. Paraphasia (communication)
  12. Expressive aphasia (communication)

* scored 4 times (during dressing; grooming and hygiene; transfers and mobility; and feeding ADL tasks)

^ scored 5 times (during all ADL tasks)

Items that are rated more than once are scored using a 5-point ordinal rating scale, from 0 = the particular neurobehavioural impairment is not observed, to 4 = the patient is unable to perform the task due to the neurobehavoural impairment. All other items (including communication items from the NBSIS and all items from the NBPSI are rated dichotomously: 0 = absent or 1 = present during ADL task performance (Arnadottir et al., 2009).

The manual includes conceptual and operational definitions for all items as well as standardized instruction and detailed criteria for administration and scoring of the instrument.

Time:

Time taken to administer the A-ONE has not been reported.

Equipment:

Not reported

Alternative forms of the assessment

The author has developed several variations of the NBI scale including 2 global scales and 4 psychometrically-sound diagnostic-specific scales (Arnadottir, 2010):

  • Common global scale (NBI-CVA, 53 items)
  • NBI common short form scale (29 items)
  • Left-hemisphere CVA (NBI-LCVA, 42 items)
  • Right-hemisphere CVA (NBI-RCVA, 51 items)
  • Dementia Alzheimers Type (NBI-DAT, 49 items)
  • Dementia Unspecified (NBI-DU, 40 items).

The scales contain different items and hierarchical structure. These versions can be used across diagnostic groups but should not be used to compare different diagnostic groups (Arnadottir, 2010).

Client suitability

Can be used with:

The A-ONE can be used with patients with dementia and other neurological disorders (Gardarsdottir & Kaplan, 2002).

Should not be used with:

As the A-ONE has been designed for use with adults with neurobehavioural disorders, it is not recommended for use with individuals with other diagnoses or disorders.

Languages of the measure

English, Dutch

Summary

What does the tool measure? The Arnadottir OT-ADL Neurobehavioural Evaluation (A-One) measures the impact of neurobehavioural impairment on functional performance of ADLs. The A-ONE is used to assess (a) independence in ADL tasks and (b) neurobehavioural impairments that limit the individual’s independence in ADL tasks.
What types of clients can the tool be used for? The A-ONE is designed for use with adults with neurological dysfunction of cortical origin, including stroke, dementia, Alzheimer’s disease and other neurological disorders.
Is this a screening or assessment tool? Assessment tool.
Time to administer The A-ONE ADL scale takes approximately 25 minutes to administer.
Versions

There is only 1 version of the ADL scale but there are several versions of the neurobehavioural scale:

  • Common global scale (NBI-CVA)
  • NBI common short form scale
  • Left-hemisphere CVA (NBI-LCVA)
  • Right-hemisphere CVA (NBI-RCVA)
  • Dementia Alzheimers Type (NBI-DAT)
  • Dementia Unspecified (NBI-DU).

There is a Dutch version of the A-ONE.

Other Languages English and Dutch.
Measurement Properties
Reliability

Internal consistency:

  • One study reported adequate internal consistency of the ADL scale (α = 0.75 – 0.79), poor to adequate internal consistency of the NBSIS scale (α= 0.69 – 0.75), and poor internal consistence of the NBPIS scale (α= 0.59 – 0.63) using Cronbach’s alpha.
  • One study reported excellent internal consistency (item separation reliability coefficient = 0.98, item separation index = 8.02; person separation reliability coefficient = 0.90, person separation index = 2.93) using Rasch analysis.

Intra-rater:

  • No studies have reported on intra-rater reliability of the A-ONE.

Inter-rater:

  • Two studies have reported excellent inter-rater reliability for the A-ONE ADL scale (kappa = 0.83; weighted kappa=0.90; ICC=0.98; Kendall’s r=0.92), the NB scale (kappa = 0.85) or the NBSIS scale (ICC=0.93, weighted kappa=0.74).

Test-retest:

  • One study reported excellent one-week test-retest reliability of the A-ONE (agreement of 0.85 or higher for all items).
Validity
  • One study reported logical ordering of ADL items according to difficulty, but noted large gaps in the hierarchy of item difficulty in some NBI scales.
  • One study reported that the ADL scale may not be well targeted to higher functioning individuals.
  • One study reported a moderate inverse relationship between ADL and neurobehavioural impairment scales, using Pearson product moment correlation (r=-0.57).

Content:

  • Content validity of the A-ONE is based on literature review and expert opinion.
  • Internal validity of the A-ONE is determined by examination of goodness of fit for items, logical hierarchical ordering of items, targeting, and PCA analysis.
  • One study reported unidimensionality of the ADL scale and logical hierarchical ordering of ADL items.
  • One study reported unidimensionality of the NBI scales.

Criterion:

Predictive Validity:

  • No studies have reported on the predictive validity of the A-ONE.

Concurrent Validity:

  • One study reported excellent correlations between the A-ONE ADL scale and the Barthel Index (r=0.70), and between the A-ONE and the MMSE (r=0.85).

Construct:

  • One study reported that the ADL scale has 3 factors and the NBSIS scale has 2 factors. A later study reported that the NBSIS scale has a third factor.

Known Groups:

  • One study reported no significant difference in the extent of the impact of neurobehavioural impairment on ADL between patients with right CVA and left CVA.
  • One study reported significant differences between adults with right and left CVA on the ADL scale and the NBSIS scale.

Convergent/Discriminant Validity:

  • No studies have reported on the convergent or discriminant validity of the A-ONE.
Floor/Ceiling Effects The A-ONE ADL scale and neurobehavioural scales demonstrate potential floor/ceiling effects. The ADL scale should be restricted to individuals who are not independent in ADLs. The NBI-CVA should be used clinically with patients with LCVA due to ceiling effects of the NBI-LCVA scale.
Sensitivity/Specificity No studies have reported on the sensitivity/specificity of the A-ONE.
Does the tool detect change in patients? The ADL ordinal rating scale can be used as an interval scale, which allows measurement of change in ADL task performance over time.
Acceptability No studies have reported on the acceptability of the A-ONE.
Feasibility No studies have reported on the feasibility of the A-ONE.
How to obtain the tool? The A-ONE can be found in the textbook The Brain and Behavior: Assessing Cortical Dysfunction Through Activities of Daily Living
For more information email: a-one@islandia.is

Psychometric Properties

Overview

A literature search was conducted to identify all relevant publications on the psychometric properties of the Arnadottir OT-ADL Neurobehavioural Evaluation (A-ONE). Nine articles were reviewed.

Floor/Ceiling Effects

In a study of the psychometric properties of the ADL scale Arnadottir et al. (2008) reported that 9 of 209 participants with left hemisphere stroke (LHS) reached maximum scores on all items, indicating possible ceiling effects of the scale. The authors concluded that use of the A-ONE should be restricted to individuals who are not independent in ADLs (where maximum scores indicate independence in ADLs). However, while a patient may achieve maximum scores on the ADL scale, they may demonstrate neurobehavioural impairments that can be detected using the NBI scale (Arnadottir, 2010).

In a study of the psychometric properties of the neurobehavioural scales Arnadottir et al. (2009) reported potential floor/ceiling effects of the measure. Two patients with LCVA and 6 patients with RCVA attained extreme (minimum) scores. Extreme (maximum) measures were seen on 10 items when used with patients with LCVA and on 4 items when used with patients with RCVA.

Following refinement of the NBI scales and development of additional versions of the scale, further analysis was conducted using Rasch analysis (Arnadottir, 2010). Results indicated that 30 of 422 patients with CVA or dementia achieved maximum or minimum scores on the 29-item NBI-Common scale; 6 of 215 patients with left-hemisphere or right-hemisphere CVA achieved maximum or minimum scores on the 53-item NBI-CVA scale; 9 of 114 patients with LCVA achieved maximum or minimum scores on the 42-item NBI-LCVA scale; and 0 of 108 patients with RCVA achieved maximum or minimum scores on the 51-item NBI-RCVA scale. Arnadottir (2010) recommended the NBI-CVA for clinical use with patients with LCVA due to the possible ceiling effect when using the NBI-LCVA scale.

Reliability

Internal consistency:

Arnadottir (1990) investigated the internal consistency of the A-ONE and reported adequate internal consistency of the ADL scale (α= 0.75 – 0.79), poor to adequate internal consistency of the NBSIS scale (α= 0.69 – 0.75), and poor internal consistency of the NBPIS scale (α= 0.59 – 0.63).

Arnadottir et al. (2008) examined the reliability of the ADL scale by performing a Rasch analysis with retrospective data from 209 patients with neurological conditions (dementia, n=111; CVA, n=95, other, n=3). Item separation reliability was 0.98 and the item separation index was 8.02, indicating reliable differentiation of items into at least 9 strata of difficulty. A separation reliability coefficient of 0.90 and person separation index of 2.93 was found, indicating reliable differentiation of the sample into at least 3 statistically distinct strata of ADL ability.
Note: Item separation reliability is the ratio of the “true” (observed minus error) variance to the obtained variation. The smaller the error, the higher the ratio will be. It ranges from 0.00 to 1.00 and is interpreted the same as the Cronbach’s alpha. A separation index > 2.00 is equivalent to a Cronbach’s alpha of 0.80 or greater (excellent).

Intra-rater:

No studies have reported on intra-rater reliability of the A-ONE.

Inter-rater:

Arnadottir (1990) reported excellent inter-rater reliability for the A-ONE ADL scale (average kappa coefficient = 0.83) and the NB scale (kappa = 0.85).

Further analysis by Arnadottir (2008) reiterated excellent inter-rater reliability for the A-ONE ADL scale (ICC=0.98; Kendall’s r=0.92, weighted kappa=0.90) and the NBSIS scale (ICC=0.93, weighted kappa=0.74).

Test-retest:

Gardarsdottir & Kaplan (2002) reported that one-week test-retest reliability of the A-ONE was excellent (agreement of 0.85 or higher for all items).

Validity

Content:

Internal validation of the A-ONE was performed by examination of goodness of fit for items, logical hierarchical ordering of items, targeting, and PCA analysis (Arnadottir, 2010).

Arnadottir et al. (2008) performed factor analysis of the A-ONE ADL scale using retrospective data from 209 patients with neurological conditions (CVA, n= 95; dementia, n= 111, other diagnosis, n=3). Analysis of all 22 ADL items revealed that the two communication items (expression, comprehension) and one feeding item (‘use knife’) did not demonstrate acceptable goodness of fit (total of 13.6% item misfit). Following removal of the two communication items, the item ‘use knife’ demonstrated substantially reduced misfit to an acceptable rate (≤ 5%), and as such was maintained. With removal of the two communication items, 84% of total variance was explained by the measures, with 3.6% of the unexplained variance accounted for by first contrast. These results support unidimensionality of the ADL scale.

Arnadottir et al. (2009) performed factor analysis of nomotor neurobehavioural items (34 NBSIS items and 16 NBPIS items) using retrospective data from 206 patients with CVA and dementia. After four items (anomia, expressive aphasia, working memory, motivation) were removed due to outfit misfit, 56.8% of variance was explained by the Rasch factor (global measure of neurobehavioural impairments), with 4.9% of the unexplained variance accounted for by the first contrast. These results indicate that the neurobehavioural impairment items can be viewed as unidimensional – i.e. belonging to the same construct. The authors proceeded to conduct a principal component analysis (PCA) of global hierarchies according to diagnosis (LCVA, n=36; RCVA, n=37; dementia, n=111). After removal of misfit items (LCVA group – 2 items; RCVA and dementia groups – 3 items), improved results were seen for all diagnostic groups (Rasch factors: LCVA group = 85.5%, RCVA group = 83.3%, dementia group = 79.2%; unexplained variance in first contrast: LCVA group = 2.4%, RCVA group = 3.4%, dementia group = 1.7%). These results indicate that the hierarchical structure of the dimension varies across diagnostic groups.

Arnadottir (2010) reported on factor analysis conducted in development of the NBI common short form scale. All diagnosis-specific versions of the NBI scales demonstrate unidimensionality, as confirmed by PCA analysis (Arnadottir, 2010). The original NBSIS scale included neurobehavioural motor items that measured left- and right-sided performance, which were collapsed to a singular motor item. A resulting 33 neurobehavoural motor items were common to all 4 diagnostic groups (LCVA, RCVA, Dementia Alzheimers type, Dementia). Four of the 33 items were omitted due to misfit by Rasch analysis and the remaining 29 items demonstrated acceptable goodness of fit. PCA analysis revealed 72.8% of variance was explained by Rasch factor, supporting unidimensionality.

Arnadottir et al. (2008) examined the hierarchical ordering of difficulty of ADL items using retrospective data from individuals with dementia (n=111), CVA (n=95), or other neurological conditions (n=3) and reported logical ordering according to item difficulty. However, Arnadottir (2010) conducted an evaluation of the targeting of person ability to item difficulty and identified that the ADL scale may not be well targeted to higher functioning individuals (discrepancy between mean measures = 1.61 logits).

Arnadottir (2010) reported that some NBI scales have large gaps in the hierarchy of item difficulty as there are few items that evaluate neurobehavioural impairments of higher-functioning individuals (mean person measure = -1.74, SD = 1.34).

Arnadottir et al. (2010) examined the relationship between ADL ability and the impact of neurobehavioural impairments on ADL using retrospective data from 215 patients with stroke. A moderate inverse relationship was found between ADL ability and the extent of neurobehavioural impairment impacting ADL, using Pearson product moment correlation (r=-0.57).

Arnadottir (2010) reported that the NBI-CVA scale demonstrates acceptable goodness of fit statistics for all retained items (MnSq ≤1.4, z < 2) and acceptable PCA.

Criterion:

Predictive:

No studies have reported on the predictive validity of the A-ONE.

Concurrent:

Steultjens (1998) examined the concurrent validity of the A-ONE. Comparison of the A-ONE ADL scale with the Barthel Index and comparison of NB scores with the MMSE revealed excellent correlations (r=0.70 and r=0.85 respectively).

Construct:

Arnadottir (1990) conducted exploratory factor analysis and reported that the ADL scale has 3 factors and the NBSIS scale has 2 factors.

Arnadottir et al. (2009) performance factor analysis of the A-ONE neurobehavioural items and reported an additional factor is formed by neurobehavioural impairments that reflect occupational errors representative of lateralized motor impairments (e.g. tone).

Known Group Validity:

Arnadottir et al. (2010) examined whether patients with right or left CVA differ in the extent to which their neurobehavioural impairments impact performance of ADLs, using retrospective data from 215 patients with stroke. No significant difference in the extent of the impact of neurobehavioural impairment on ADL was seen between patients with right CVA (n=103) and patients with left CVA (n=112).

Gardarsdottir & Kaplan (2002) examined the construct validity of the A-ONE ADL scale and Neurobehavioural Specific Impairment Scale (NBSIS) in adults with right CVA (n=19) and left CVA (n=23). Mann-Whitney U tests identified significant differences between the groups for only 3 of 18 ADL tasks: shave/makeup (p=0.013), comprehension (p=0.005), and speech (p=0.001), whereby patients with left CVA were more dependent than patients with right CVA for these tasks. Mann-Whitney U and chi-square tests revealed significant between-group differences for 13 of 46 neurobehavioural impairments, all within the three NSIS categories of motor apraxia, unilateral body neglect and abnormal tone. Results indicated that patients with left CVA demonstrated greater severity of motor apraxia in dressing (p=0.022), grooming and hygiene (p=0.001) and feeding (p=0.002) than patients with right CVA. Patients with left CVA also demonstrated greater severity of abnormal tone on both sides of the body in grooming and hygiene tasks (p=0.001) than patients with right CVA, whereas patient with right CVA demonstrated greater severity of abnormal tone on both sides of the body during performance of dressing (p=0.001), transfers and mobility (p=0.001) and feeding (p=0.001). Patients with right CVA also demonstrated greater severity of unilateral body neglect in grooming and hygiene tasks (p=0.002) than patients with left CVA.

Convergent/Discriminant Validity:

No studies have reported on the convergent or discriminant validity of the A-ONE.

Responsiveness

Principal component analysis of the ADL domain supported unidimensionality, enabling conversion of the ordinal rating scale to an interval scale, which would allow measurement of change in ADL task performance over time (Arnadottir, 1990).

References

  • Arnadottir, G. (2010). Measuring the impact of body functions on occupational performance: Validation of the ADL-focused Occupation-based Neurobehavioural Evaluation (A-ONE). (Doctoral dissertation). Retrieved from Swedish Dissertations database.
  • Arnadottir, G. (1990). The brain and behavior: Assessing cortical dysfunction through activities of daily living. St. Louis, MO: Mosby.
  • Arnadottir, G. & Fisher, A.G. (2008). Rasch analysis of the ADL scale of the A-ONE. The American Journal of Occupational Therapy, 62, 51-60
  • Arnadottir, G., Fisher, A.G., & Löfgren, B. (2009). Dimensionality of nonmotor neurobehavioural impairments when observed in the natural contexts of ADL task performance. Neurorehabilitation and Neural Repair, 23(6), 579-86.
  • Arnadottir, G., Löfgren, B., & Fisher, A.G. (2010). Difference in impact of neurobehavioural dysfunction on activities of daily living performance between right and left hemisphere stroke. Journal of Rehabilitation Medicine, 42, 903-7.
  • Bottari, C., Dutil, É., Dassa, C., & Rainville, C. (2006). Choosing the most appropriate environment to evaluation independence in everyday activities: Home or clinic? Australian Occupational Therapy Journal, 53, 98-106.
  • Carswell, A., Carson, L.J., Walop, W. & Zgola, J. (1992). A theoretical model of functional performance in persons with Alzheimer disease. Canadian Journal of Occupational Therapy, 59(3), 132-40.
  • Cooke, D.M., McKenna, K. & Fleming, J. (2005). Development of a standardized occupational therapy screening tool for visual perception in adults. Scandinavian Journal of Occupational Therapy, 12, 59-71.
  • Gardarsdottir, S. & Kaplan, S. (2002). Validity of the Árnadottir OT-ADL Neurobehavioral Evaluation (A-ONE): Performance in activities of daily living and neurobehavioural impairments of persons with left and right hemisphere damage. American Journal of Occupational Therapy, 56, 499-508.
  • Steultjens, E.M. (1998). A-ONE: De Nederlands Versie [A-ONE: The Dutch version]. Nederlands Tidskrift for Ergoterapie, 26, 100-4.

See The Measure

How to obtain the assessment?

The A-ONE assessment is in the textbook: The Brain and Behavior: Assessing Cortical Dysfunction Through Activities of Daily Living.

For more information email: a-one@islandia.is

Table of contents
What do you think?