Advertisement

Self-Scoring Templates for Motor and Cognitive Subscales of the FIM Instrument for Persons With Spinal Cord Injury

Published:December 04, 2013DOI:https://doi.org/10.1016/j.apmr.2013.11.009

      Abstract

      Objective

      To provide self-scoring templates for the FIM instrument's motor and cognitive scales that enable clinicians to monitor progress during rehabilitation using equal-interval Rasch-calibrated measures instead of ordinal raw scores.

      Design

      Secondary analysis of a prospective, observational cohort study.

      Setting

      Six geographically dispersed hospital-based rehabilitation centers in the United States.

      Participants

      Subset of consecutively enrolled individuals with new traumatic spinal cord injuries discharged from participating rehabilitation centers (N=1146).

      Intervention

      Not applicable.

      Main Outcome Measures

      Subscores of the FIM instrument, including a 13-item motor scale, a 5-item cognitive scale, an 11-item (without sphincter control items) motor scale, a 3-item transfer scale, a 6-item self-care scale, a 3-item self-care upper extremity scale, and a 3-item self-care lower extremity scale.

      Results

      KeyForms for the FIM instrument scales allow clinicians and investigators to estimate patients' functional status and monitor progress. In cases with no missing data, the look-up tables provide more accurate estimates of patients' functional status.

      Conclusion

      Clinicians can use KeyForms and look-up tables for FIM instrument subscales to monitor patients' progress and communicate improvement in equal-interval units.

      Keywords

      List of abbreviations:

      SCI (spinal cord injury)
      Measurement of functional outcomes is important not only for administrative requirements but also for guiding clinical decisions for individual patients during rehabilitation intervention. Standardized instruments have gained prominence; the most widely used standardized functional outcome instrument is the FIM instrument.
      • Hamilton B.B.
      • Granger C.V.
      • Sherwin F.S.
      • Zielezny M.
      • Tashman J.S.
      A uniform national data system for medical rehabilitation.
      However, many clinicians still do not use standardized instruments, in part because of the burden of administration and difficulty in interpreting results.
      • Jette D.U.
      • Halbert J.
      • Iverson C.
      • Miceli E.
      • Shah P.
      Use of standardized outcome measures in physical therapist practice: perceptions and applications.
      Research on the use of interval measures for the FIM instrument spans the past 20 years when separate difficulty estimates were published for persons with spinal cord injuries (SCIs).
      • Heinemann A.W.
      • Linacre J.M.
      • Wright B.D.
      • Hamilton B.B.
      • Granger C.
      Relationships between impairment and physical disability as measured by the Functional Independence Measure.
      • Linacre J.M.
      • Heinemann A.W.
      • Wright B.D.
      • Granger C.V.
      • Hamilton B.B.
      The structure and stability of the Functional Independence Measure.
      In a recent study,
      • Whiteneck G.
      • Gassaway J.
      • Dijkers M.
      • Jha A.
      New approach to study the contents and outcomes of spinal cord injury rehabilitation: the SCIRehab Project.
      • Whiteneck G.G.
      • Gassaway J.
      SCIRehab uses practice-based evidence methodology to associate patient and treatment characteristics with outcomes.
      • Kozlowski A.J.
      • Heinemann A.W.
      Using individual growth curve models to predict recovery and activities of daily living after spinal cord injury: an SCIRehab project study.
      data from a large sample of patients with SCIs were calibrated to examine the association of patient and treatment characteristics with outcomes and predict functional recovery. The authors used Rasch-calibrated FIM instrument measures for 6 subscales, transformed them to a 0 to 100 scale, and provided conversion tables for situations in which all items in a subscale have been scored. Although the smaller subscales may be easier to score, they come with trade-offs such as a narrower scale range and lower score reliability.
      • Kozlowski A.J.
      • Heinemann A.W.
      Using individual growth curve models to predict recovery and activities of daily living after spinal cord injury: an SCIRehab project study.
      Previously, clinicians were limited to using the ordinal summary ratings from the FIM instrument and other instruments, without guidance on how to interpret the amount of observed change on specific activities. Clinicians could monitor progress on each activity with item-level information gathered throughout their interventions. One benefit of linear scaling is that differences between points on a scale are equal across the range of the scale, allowing for discrimination of functional abilities along the entire continuum. Clinicians may find additional benefit in using the smaller subscales because they will be able to convert raw ratings to Rasch measures for a less burdensome subset of FIM instrument items (eg, 6-item self-care) that are more relevant to their practice or a specific patient.
      The objective of this brief report was to provide self-scoring templates for motor and cognitive scales of the FIM instrument that enable clinicians to monitor progress during rehabilitation using equal-interval, Rasch-calibrated measures instead of ordinal raw scores. Using these FIM instrument SCI KeyForms, clinicians can now record individuals' ratings at admission, document rehabilitation goals in each activity, and monitor progress over the rehabilitation stay and beyond to quantify the extent of functional improvement and goal attainment.

      Methods

       Sample

      A total of 1376 patients with SCI were recruited from 6 SCIRehab study sites, and admission, discharge, and 1-year follow-up FIM instrument ratings were reported by trained clinicians for 1146 patients with data for each. A total of 81% were men, 67% were non-Hispanic white, 38% were married, and 79% had high school or greater education. Further details about the sample are reported in previous SCIRehab articles.
      • Whiteneck G.
      • Gassaway J.
      • Dijkers M.
      • Jha A.
      New approach to study the contents and outcomes of spinal cord injury rehabilitation: the SCIRehab Project.
      • Whiteneck G.G.
      • Gassaway J.
      SCIRehab uses practice-based evidence methodology to associate patient and treatment characteristics with outcomes.
      • Kozlowski A.J.
      • Heinemann A.W.
      Using individual growth curve models to predict recovery and activities of daily living after spinal cord injury: an SCIRehab project study.

       Instruments

      The FIM instrument is a measure of independence consisting of 13 physical and 5 cognitive functioning items rated by trained clinicians at admission and discharge and by self-report via telephone or in-person interview at follow-up. Kozlowski and Heinemann
      • Kozlowski A.J.
      • Heinemann A.W.
      Using individual growth curve models to predict recovery and activities of daily living after spinal cord injury: an SCIRehab project study.
      reported scoring information for a 13-item motor scale, an 11-item (without sphincter control items) motor scale, a 3-item transfer scale, and a 6-item self-care scale, with a breakdown into a 3-item self-care upper extremity scale and a 3-item self-care lower extremity scale.

       Analysis

      We used Rasch calibration to estimate item difficulties and person abilities along a shared metric of functioning for the FIM instrument items. For this study, we selected a random sample of FIM instrument ratings at admission, discharge, and 1-year follow-up such that only 1 record was included for each case. We anchored the items and rating scale steps on the basis of these data and computed FIM instrument subscale measures for all 3 time points. Then, we transformed the resulting measures to a scale that ranged from 0 to 100, although some KeyForms graphs show different extreme measures. We used Winsteps version 3.73a to produce self-scoring FIM instrument SCI KeyForms on which a clinician can record a patient's performance. We provide 1 KeyForm for each subscale reported in the SCIRehab studies. Overall performance on the set of activities in each subscale was used to estimate the ability level of these individuals (horizontal axis); the activities are arranged vertically in the order of their difficulty.
      The FIM instrument SCI KeyForms are designed so that clinicians can easily see performance differences across activities and categories within activities: it is more difficult to improve from “modified independence” to “complete independence” than to improve from any other point lower in the scale. The horizontal alignment and spacing of the 7 rating scale categories reflect an assessment of the difficulty in performing these activities. Gaps in the vertical alignment of these activities indicate differences in the difficulty in performing them, with stairs being the most difficult activity to perform and eating the least difficult. Small differences in this vertical alignment indicate activities with relatively the same level of difficulty for the sample as a whole, although for individuals, there will probably be differences. The size of the horizontal gaps between categories reflects the difference in performance or improvement from one category to the next: small gaps reflect little improvement, whereas large gaps reflect large improvements. Because the size of the gaps is not consistent across categories and activities, the amount varies from one category to the next and from one activity to the next.

      Results

      FIM instrument SCI KeyForms for the 13-item motor (fig 1) and 5-item cognitive (fig 2) scales are presented in this report and look-up tables are available online (supplemental figs S1 and S2, www.archives-pmr.org). KeyForms and look-up tables for the remaining scales are also available online. Separate KeyForms and look-up tables are not available for the walk/wheelchair and stairs activities and sphincter control, which consist of only 2 activities.
      Figure thumbnail gr1
      Fig 1FIM instrument SCI 13-Item Motor KeyForm. Vertical scale: FIM instrument items listed in order of difficulty, with more difficult items at the top and less difficult items at the bottom. Horizontal scales: (1) FIM instrument measures at top and bottom of graph ranging from 0 to 100, with higher values reflecting greater independence. (2) Percentiles ranging from 0 to 99, with higher values reflecting performance better than that percentage of persons in the SCIRehab sample. (3) FIM instrument category labels ranging from 1 to 7, reflecting increasing levels of functional independence for each item.
      Figure thumbnail gr2
      Fig 2FIM instrument SCI 5-item Cognitive KeyForm.

      Discussion

      Clinicians routinely score FIM instrument items for rehabilitation inpatients on admission and discharge
      • Heinemann A.W.
      • Linacre J.M.
      • Wright B.D.
      • Hamilton B.B.
      • Granger C.
      Relationships between impairment and physical disability as measured by the Functional Independence Measure.
      but may not do so during inpatient care or for postdischarge services such as outpatient therapy. More than half of the physical therapists surveyed reported using no standardized instruments due in part to the administrative burden.
      • Jette D.U.
      • Halbert J.
      • Iverson C.
      • Miceli E.
      • Shah P.
      Use of standardized outcome measures in physical therapist practice: perceptions and applications.
      Clinicians could benefit from the ubiquitous nature of the FIM instrument with versions that have lower burden, are more informative, or both. A computer-adaptive test version of the motor FIM instrument could reduce burden
      • Dijkers M.P.
      A computer adaptive testing simulation applied to the FIM instrument motor component.
      but may be difficult to implement because the algorithm results in one of eight 6-item subsets determined by the patient's wheelchair or walking ability and bladder management scores. The FIM instrument SCI KeyForms reduce burden by allowing clinicians to examine change on specific items or on subscales that relate more specifically to their practice.
      Our approach is similar to that of Velozo and Woodbury,
      • Velozo C.A.
      • Woodbury M.L.
      Translating measurement findings into rehabilitation practice: an example using Fugl-Meyer Assessment-Upper Extremity with patients following stroke.
      who provide guidance using a KeyForm to examine item-level Fugl-Meyer scores for persons with stroke. Clinicians can identify transition zone items that fall between the items that patients clearly can and cannot perform. Transition zone items may be most suitable for setting goals. Similarly, FIM instrument SCI KeyForms can facilitate evaluation of change and clinical decisions to continue with or to revise the initial treatment plan during intervention.
      However, the FIM instrument SCI KeyForm items have more rating scale levels for more activities than does the Fugl-Meyer; thus, FIM instrument scores may not form a distinct transition zone. Instead, clinicians may want to select the items that are most relevant to their clinical practice, or one of the smaller subscales, and can use the median item scores to evaluate change on the Rasch measure. Different shaped markers (squares, circles, triangles, diamonds, etc) can differentiate admission, discharge-goal, periodic reassessment, and discharge-actual item scores on the same KeyForm. Periodic reassessment can provide early confirmation of progress (or lack of progress) toward discharge goals. In addition, outcomes for patients who continue to outpatient therapy could be monitored across services using a common metric. We provide an example of this method in the online supplement (see supplemental fig S1 and S2).
      To monitor progress during rehabilitation, clinicians can record therapy goals after admission and readminister items periodically (eg, weekly) to evaluate performance during the rehabilitation stay. At admission, rate the patient according to standardized criteria
      • Heinemann A.W.
      • Linacre J.M.
      • Wright B.D.
      • Hamilton B.B.
      • Granger C.
      Relationships between impairment and physical disability as measured by the Functional Independence Measure.
      and mark the observed level of independence for each activity with a symbol (eg, squares). Set rehabilitation goals by marking the anticipated performance level for each activity with another symbol (eg, triangles). Retain the marked KeyForms and record intermediate or discharge ratings with other symbols (eg, diamonds and circles).
      By using the FIM instrument SCI KeyForms, clinicians can take advantage of the equal-interval spacing of rating scale categories and items to assess changes in performance over time or in comparison to expectations. Various uses include the following: (1) Clinicians could evaluate progress in each activity by comparing the discharge and admission ratings. The equal-interval nature of the new scale provides an illustration of the amount of improvement represented by an increase of 1 point (less in the middle of the scale and more at the extremes). (2) If intermediate ratings were available, clinicians could monitor progress over time and revise goals and therapy plans as needed. KeyForms provide a visual display of progress of all or a selection of items, which may facilitate patient engagement and motivation. However, change in selected items can indicate a pattern of progress, but not true change or clinically important change. (3) At discharge, clinicians could identify activities in which goals were met or exceeded and those in which they were not attained by comparing the discharge and goal ratings.

      Study limitations

      One limitation of the KeyForm is that it is not readily integrated into an electronic health record.

      Conclusions

      This report provides self-scoring templates for clinicians to use, record, and monitor progress during rehabilitation using an equal-interval–level scale. The FIM instrument SCI KeyForms are based on the performance of a large sample of persons with SCIs and provide an accurate illustration of the amount of improvement that would be needed to progress from one level of independence to the next.

      Supplier

      • a.
        Winsteps version 3.73; Rasch Measurement computer program. Available at: www.Winsteps.com.

      Acknowledgments

      We thank Kristin Dams-O'Connor, PhD, and Gale Whiteneck, PhD, for their careful review and invaluable comments on the contents of this report.

      Supplemental Fig S1. Sample FIM SCI Motor KeyForm and its Use

      An example of an FIM SCI 13-item Motor KeyForm is presented below. Reading down the form in this example, someone with an overall ability level (measure) of 50 would be expected to achieve a rating of 2 (maximal assistance) on Stairs, a rating of 6 (modified independence) on Tub/Shower Transfer, Bowel Management, Toileting, Toilet Transfer, Lower Body Dressing, Bladder Management, Bathing, Bed/Chair/Wheelchair Transfer, and Walk/Wheelchair Locomotion, and a rating of 7 (complete independence) on Upper Body Dressing, Grooming, and Eating. Someone with a lower overall measure would be expected to be more dependent in each of these activities and someone with a higher overall measure would be expected to be more independent in each of these activities.
      In this example, symbols are used for each activity rating to illustrate its use in monitoring progress during rehabilitation. Patient performance was rated as 1 (total assistance) in stair climbing and sphincter control at admission (□); goals at discharge (Δ) were set at 3 (moderate assistance) but discharge ratings (O) were 2 (maximal assistance) which represented less than expected improvement. Initial performance at admission and discharge and goals levels set were higher and improvement greater in transfer and walking/wheelchair locomotion activities, and performance at admission and discharge, goal levels set, and the greatest improvement was found in the self-care activities. In this example, discharge status was at or above the goals set for rehabilitation for all activities except sphincter control and would represent areas in possible need of further rehabilitation. To estimate the overall performance level at any administration, a vertical line can be drawn with approximately half of the ratings falling to the left and half falling to the right of the line. Using the admission ratings in this example, the line drawn here (red dashed line) was approximately at a measure of 29. The bottom line of the form shows the percentile that measure would represent; in this example, a measure of 29 is at about the 45th percentile of persons in the SCIRehab population.
      The median Rasch measure for the 8 items on which goals were set was about 40 (green dashed line), which represents the 65th percentile of persons in the SCIRehab population. The median discharge Rasch measure for the subset of goal items is also about 40, indicating that on the whole the goal was met, even though the outcomes for the individual goal items varied, (some item goals were met, some not met, and others exceeded). Also note that the discharge median for the 8 goal items represents the median for discharge scores for all items, which suggests the selected goal items provided a reasonable representation of the discharge FIM score.
      Figure thumbnail fx1

      Supplemental Fig S2. FIM SCI KeyForms and Score-to-Measure Look-up Tables

      Figure thumbnail fx2
      Figure thumbnail fx3
      Figure thumbnail fx4
      Figure thumbnail fx5
      Figure thumbnail fx6
      Figure thumbnail fx7
      Figure thumbnail fx8
      Figure thumbnail fx9
      Figure thumbnail fx10
      Figure thumbnail fx11
      Figure thumbnail fx12
      Figure thumbnail fx13

      References

        • Hamilton B.B.
        • Granger C.V.
        • Sherwin F.S.
        • Zielezny M.
        • Tashman J.S.
        A uniform national data system for medical rehabilitation.
        in: Fuhrer M.J. Rehabilitation outcomes: analysis and measurement. Brookes Publishing Co, Baltimore1987
        • Jette D.U.
        • Halbert J.
        • Iverson C.
        • Miceli E.
        • Shah P.
        Use of standardized outcome measures in physical therapist practice: perceptions and applications.
        Phys Ther. 2009; 89: 125-135
        • Heinemann A.W.
        • Linacre J.M.
        • Wright B.D.
        • Hamilton B.B.
        • Granger C.
        Relationships between impairment and physical disability as measured by the Functional Independence Measure.
        Arch Phys Med Rehabil. 1993; 74: 566-573
        • Linacre J.M.
        • Heinemann A.W.
        • Wright B.D.
        • Granger C.V.
        • Hamilton B.B.
        The structure and stability of the Functional Independence Measure.
        Arch Phys Med Rehabil. 1994; 75: 127-132
        • Whiteneck G.
        • Gassaway J.
        • Dijkers M.
        • Jha A.
        New approach to study the contents and outcomes of spinal cord injury rehabilitation: the SCIRehab Project.
        J Spinal Cord Med. 2009; 32: 251-259
        • Whiteneck G.G.
        • Gassaway J.
        SCIRehab uses practice-based evidence methodology to associate patient and treatment characteristics with outcomes.
        Arch Phys Med Rehabil. 2013; 94: S67-S74
        • Kozlowski A.J.
        • Heinemann A.W.
        Using individual growth curve models to predict recovery and activities of daily living after spinal cord injury: an SCIRehab project study.
        Arch Phys Med Rehabil. 2013; 94: S154-S164
        • Dijkers M.P.
        A computer adaptive testing simulation applied to the FIM instrument motor component.
        Arch Phys Med Rehabil. 2003; 84: 384-393
        • Velozo C.A.
        • Woodbury M.L.
        Translating measurement findings into rehabilitation practice: an example using Fugl-Meyer Assessment-Upper Extremity with patients following stroke.
        J Rehabil Res Dev. 2011; 48: 1211-1222