ATOMS Project Technical Report:
The ICF in the Context of Assistive Technology (AT) Interventions and Outcomes

Roger O. Smith, PhD, OT, Carleen Jansen, BS, OTS, Jamie Seitz, BS, OTS, and Kathy Longenecker Rust, MS, OT

Introduction

The International Classification of Functioning, Disability and Health (ICF) was published by the World Health Organization in 2001 with the overall aim “to provide a unified and standard language and framework for the description of health and health-related states” (p. 3).  It is further discussed as, “a meaningful and practical system that can be used by various consumers for health policy, quality assurance and outcome evaluation in different cultures” with applications as a statistical tool, a research tool, a clinical tool, a social policy tool, and as an educational policy tool (p.4).  As part of its comprehensive needs assessment process, the ATOMS (Assistive Technology Outcome Measurement System) Project performed a comprehensive evaluation of the ICF for its ability to serve as an organizing framework for assistive technology outcomes.  This paper presents the results of that review and demonstrates that for multiple reasons, the ICF falls short of serving the needs of AT outcomes researchers and clinicians, thereby limiting the effectiveness of the instrument in serving the consumer with assistive technology (AT) needs.  The 2001 publication of the ICF explores potential future directions for the system.  Two of these include “research into treatment or intervention matching” (p.251), and “further research on environmental factors to provide the necessary detail for use in describing both the standardized and current environment” (p.252).  AT is a significant intervention and an important environmental influence affecting the function and participation of persons with disabilities.  The ICF does not provide a cohesive structure to map the impact of AT.

Background

Essentially, the ICF classifies individuals based on the components of categories: a) Body Functions, b) Body Structures, c) Activities and Participation, d) Environmental Factors, and e) Personal Factors (see table 1).  The classification and coding of Personal Factors will not be discussed in this analysis. 

Instructions in the variety of methods to code the ICF are included with the published document.  Additionally, to facilitate the application of the ICF in clinical settings and research projects, the ICF Checklist (2001, 2003) was developed.  It allows qualification of the magnitude of the functioning problem and the extent to which an environmental factor is a facilitator or barrier (CAS Team, WHO, 2002).  Beta 2 Field Trials test-retest reliability showed high kappa values ranging from 0.7 to 1.0 for the sections: Body Functions, Body Structures, Activities and Participation, and Environmental Factors.  Each of these sections, however, pose problems when considering the use of assistive technology.  As it is claimed that the “ICF has universal application… and encompasses all aspects of human health and some health-relevant components” (WHO, 2001, p.7), then anyone administering the ICF Checklist should be able to assess all areas of health for an individual and be able to apply this to AT interventions and outcomes.  The assessor should be able to identify the exact impairment or group of impairments of an individual and match this with the exact assistive technology device (ATD) that addresses the impairment(s).  However, four challenges of using the ICF with AT are identified that preclude precise and meaningful quantification of this interaction.

Four Challenges

Challenge one

Challenge two

Challenge three

Challenge four

Challenge one

The first challenge is the subjective nature of coding while using the ICF, specifically with regards to the coding and qualifier systems that accompany the structure of the ICF.  The coding process is quite complicated.  The ICF offers a different qualifier system for coding each of the major categories.  While this qualifier system differs with each category it is consistently scored on a scale to indicate the extent of an impairment.  Due to the complicated nature of the coding system, a summary for each section is provided.

The category 'Body Functions' is coded with one qualifier 'extent of the impairments'.

See Table 2.  However, for the next category, 'Body Structures', two qualifiers are presented a) extent of impairment, b) nature of the change.  See Table 3.  Additionally, the ICF document allows for a third possible qualifier for this category, location of impairment.

The third category ‘Activities and Participation” is coded in the ICF Checklist based on two qualifiers a) performance and b) capacity.  See table 4.  The definition of capacity qualifier

includes the criteria “without assistance”.  Does this mean without human assist, or does it mean without ATD assist, or without either?  This lack of clarification will lead to reliability issues when using the ICF Checklist with persons with disabilities.  In the ICF document two additional qualifiers are described, a Capacity qualifier with assistance and Performance qualifier without assistance.

The last category 'Environmental Factors', is coded one way in the ICF Checklist (see table 5), however the ICF document offers three methods: 1) coding separate from the other three categories (as in the ICF Checklist), 2) coding each environmental factor for every other category, e.g.,

'Body functions' ________	E code ________
'Body structures' ________	E code ________
'Activities and Participation'	E code ________

or 3) coding by using the qualifiers 'performance' and 'capacity' for every item, e.g.

Performance qualifier _____________	E code _______
Capacity qualifier __________________	E code _______

(WHO, 2001, p. 225-226)

The first method of coding (Table 5) allows for a simplistic consideration of at devices and allow that they be considered as either a barrier or facilitator to function.  However, the lack of detail provided provides little meaning.  This scoring of an individual who uses a walker as their only ATD would look pretty much the same as the scoring for an individual with multiple disabilities who uses multiple devices.  Within  the two other suggested coding methods there is no specific method provided for coding AT.  However, suggested coding examples (in the book) appear to imply that AT is something that can be added “per user.”  For example, when coding an individual for their ability to walk short distances, the ICF document describes this as “walking for less than a kilometer, such as walking around rooms or hallways, within a building or for short distances outside,” (p.144).  When coding using the performance qualifier the example given is “getting around on foot, in the person’s current environment, such as on different surfaces and condition, with the use of a cane, walker, or other assistive technology, for distances less than one kilometer,” (Italics added) (p. 232).  The process and method of coding this way is up to the evaluator and therefore becomes inconsistent, problematic, and rather arbitrary.

All of these examples demonstrate significant problems in exclusion.  That is, the coding system may leave out important pieces of information, such as the use of ATDs, that provide increased function for an individual.  Additionally, reliability threats are prevalent across all of these examples as lack of detailed instruction as to how AT is considered in the qualifiers may result in different raters coding performance of the same individual very differently.

B. Challenge Two

The second challenge in using the ICF to quantify AT interventions and outcomes is its lack of specificity, i.e., the exact functional limitation is not specified and the direct use of an AT intervention is not specified.  For example, consider the use of Dycem as a non-slip surface during eating.  It is likely being used because the user has difficulty keeping his/her plate, bowl, or glass in place.  Assessing this individual using the ICF classification system, a deficit would be marked in the category of eating.  An individual with burns to both hands or an individual with impaired swallowing ability would also be scored as having a deficit in the area of eating; however, the functional performance (and appropriate interventions) of these individuals is remarkably different.  The use of a button hook for dressing provides a second example. The ICF category is limited to 'putting on clothes'.  It does not break down this category into more detailed components.  Thus, when a deficit is scored here for an individual who cannot button up his/her shirt, viewing the coded deficit in the category of ‘putting on clothes’, it may be assumed that he/she cannot dress at all.  A final example is seen when scoring an individual’s performance when washing oneself.  The ICF identifies 'washing body parts' and 'washing whole body' along with ‘drying oneself’. If a person is unable to wash their lower extremities they would be marked deficient in both areas of washing.  Does this make sense?  It would be unclear from the ICF coding which body part the individual is unable to wash and why.  This results in a great deal of ambiguity.  This lack of detail and specificity within Activities and Participation categories does not provide the specific level at which task performance breaks down and therefore eliminates the possibility of matching that limitation with an AT intervention. If more categories were included in both assessing the person and the AT intervention, simplified matching based on detailed coding would be possible. 

C. Challenge Three

The third challenge experienced when using the ICF for tracking AT interventions is that the system does not adequately consider qualifiers of performance such as safety or task/activity completion time.  Therefore, an individual could be depicted as not having a deficit, when in actuality the person is not functional due to the poor quality of performance.  Alternatively, the person may be depicted by the ICF as having a deficit, but the reason for the scored deficiency in carrying out the task or activity, if due to the impact of one of the qualifiers such as poor safety, is not available from the coding system.

Taking the first case into consideration, assume that an individual is deemed able to independently dress him/herself according to the ICF system.  However, what if it took the individual more than an hour to dress?  Since time is an important variable, it could be incorrect not to categorize this individual as deficient in the area of dressing because the unreasonable time it took reduces the quality of performance making the individual’s performance of dressing not functional.  If this were known, an AT intervention to assist with dressing could be used to reduce the time it takes the individual to dress, making the individual more efficient in completing the task.  Due to the nature of the ICF coding system this variable of time would not have been known.  In this situation, the existing IFC system classifies the individual as functional, when in reality, the individual may be frustrated by the time requirements of the task and the AT interventions to assist that person in reaching a more satisfactory level of performance may never be recognized as needed and never provided.

In the second example, assume an individual is marked as deficient in a certain area but the cause of the deficiency or the extent of the deficiency is an unknown variable.  It is possible that the issue of safety may be a factor and could be inhibiting the individual from completing a specific task or participating in an activity.  For example, an older individual who does not feel safe standing for long periods of time because he/she is afraid of falling may be depicted as deficient in a category within the ICF such as 'preparing simple meals'.  However, it is not that the person is unable to organize, cook, and serve the meal but rather that the person just does not feel safe standing for the length of time required.  The ICF would depict a deficit in the area of 'preparing simple meals' and it would not provide an adequate picture of the individuals functional performance. If safety were known as the primary limitation to functional performance, then specific interventions could be implemented in order to reduce the risk of falling.

D. Challenge Four

The final problem faced when attempting to match AT devices with the ICF system is inherent in the overlap and relationship between body structure and functions with activities and participation.  When considering a specific AT device, on what level is it impacting?  Should an AT device such as a universal cuff be matched to the problematic body functions (muscle power functions or control of body movement functions), should it be matched to the problematic body structure (structure of the nervous system or structures related to movement) or should it be matched to activities and participation (carrying out daily routine, self care, or work and employment)?   Is it possible to consider AT devices as having primary and secondary usages (refer to Appendix A for examples) to fit the ICF system?  Would all AT devices fit a primary and secondary coding classification?  Essentially, the ambiguity due to the overlap and relationship of the major ICF categories precludes mapping of an ATD to one and only one category.  The multiplicity of ICF items for which an AT device might enhance health and activity makes such a matching attempt not only unwieldy, but probably of little value as well.

Conclusion 

Though a reputable classification instrument, the ICF presents problems for the documentation of AT instrument use and outcomes measurement.  These include the ICF’s subjective nature, lack of specificity within categories, inconsistent use of qualifiers of performance, and the ambiguity and overlap within the general ICF classification categories and AT classification categories.  As a result, the specific cause and reason for an individual’s problem requiring the use of AT may not be indicated and therefore not coded.  Primary and secondary AT usages, if possible to determine, do not map well to existing categories that lack specifics.  Moreover, the subjective nature of the ICF poses many problems including the exclusion of important information by the rater as well as its reliability. If these concerns were addressed, an all encompassing coded and quantifiable instrument could emerge with the ability to specify an individual’s impairment as well as the specific AT service or device that would accommodate that impairment, or vise versa.  In its current form, the ICF does not serve to adequately explain or codify the significant impact that assistive technology has on the lives of persons with disabilities.

References

World Health Organization (2001). International classification of functioning, disability and health: ICF. Geneva, Switzerland, World Health Organization.

World Health Organization. (2003). ICF Checklist, Version 2.1a, clinician form for international classification of functioning, disability and health. Retrieved July 2, 2006, from http://www.ibv.liu.se/content/1/c6/04/02/82/icf-checklist.pdf. Now available at http://www.who.int/classifications/icf/training/icfchecklist.pdf.

World Health Organization Classification Assessment Surveys and Terminology Team (CAS). (2002). The ICF Checklist: Development and Application. Retrieved July 6, 2006, from http://www.aihw.gov.au/international/who_hoc/hoc_02_papers/brisbane95.doc.

Appendices

Appendix A

 

Copyright © 2001-2015 ATOMS Project at the University of Wisconsin-Milwaukee