R E S E A R C H R E P O R T

Outcomes for Students Receiving

School-Based Physical Therapy as
Measured by the School Function
Assessment
Susan K. Effgen, PT, PhD, FAPTA; Sarah Westcott McCoy, PT, PhD, FAPTA; Lisa A. Chiarello, PT, PhD, PCS, FAPTA;
Lynn M. Jeffries, PT, PhD, PCS; Catherine Starnes, BS; Heather M. Bush, PhD
Department of Rehabilitation Sciences (Dr Effgen), University of Kentucky, Lexington; Department of Rehabilitation
Medicine (Dr Westcott McCoy), University of Washington, Seattle; Department of Physical Therapy and Rehabilitation
Sciences (Dr Chiarello), Drexel University, Philadelphia, Pennsylvania; Department of Rehabilitation Sciences
(Dr Jeffries), University of Oklahoma Health Sciences, Oklahoma City; Department of Biostatistics (Ms Starnes and
Dr Bush), College of Public Health, University of Kentucky, Lexington.

Purpose: To describe School Function Assessment (SFA) outcomes after 6 months of school-based physical
therapy and the effects of age and gross motor function on outcomes. Methods: Within 28 states, 109 physical
therapists and 296 of their students with disabilities, ages 5 to 12 years, participated. After training, therapists
completed 10 SFA scales on students near the beginning and end of the school year. Results: Criterion scores
for many students remained stable (46%-59%) or improved (37%-51%) with the most students improving in
Participation and Maintaining/Changing Positions. Students aged 5 to 7 years showed greater change than
8- to 12-year-olds on 5 scales. Students with higher gross motor function (Gross Motor Function Classification
System levels I vs IV/V and II/III vs IV/V) showed greater change on 9 scales. Conclusions: Positive SFA change
was recorded in students receiving school-based physical therapy; however, the SFA is less sensitive for older
students and those with lower functional movement. (Pediatr Phys Ther 2016;28:371–378) Key words: school-
based physical therapy, School Function Assessment, student outcomes

INTRODUCTION Individualized education programs (IEPs), which include

The provision of school-based physical therapy annual objectives, are developed for each student. Parents
became nationwide with the implementation of the and service providers sign the IEP agreement and there is a
Education of All Handicapped Children’s Act of 1975 (PL requirement that the services be provided; however, there
94-142).1 This act and its later amendments provided is no obligation that the objectives be achieved. In the
guidelines for how students with disabilities should be 2004 reauthorization of the Individuals with Disabilities
served in their schools. Physical therapists (PTs) were Education Act,2 there was recognition that although
included as related service providers to assist students student outcomes could not be guaranteed, there should
with disabilities to benefit from special education.1,2 be student assessment and performance appraisal of those
providing the services.
0898-5669/110/2804-0371 There is a dearth of research indicating the outcomes
Pediatric Physical Therapy of school-based physical therapy services in general as
Copyright C 2016 Wolters Kluwer Health, Inc. and Academy of

Pediatric Physical Therapy of the American Physical Therapy

well as the use of specific interventions.3 The largest
Association study related to functional, school-based therapy service
delivery was carried out in Canada and involved 50
Correspondence: Susan Katherine Effgen, PT, PhD, FAPTA, Rehabilita- children, only 13 of which received physical therapy for
tion Sciences, University of Kentucky, 900 S. Limestone, Lexington, KY
40536 (seffgen@[Link]).
mobility problems.4 Statistically and clinically significant
The authors declare no conflicts of interest.
improvement after 6 months of intervention was noted us-
ing goal attainment scaling, 5 scales (Travel, Maintaining
DOI: 10.1097/PEP.0000000000000279
and Changing Positions, Manipulation With Movement,

Pediatric Physical Therapy Student Outcomes Measured by SFA 371

Copyright © 2016 Wolters Kluwer Health, Inc. and the Academy of Pediatric Physical Therapy of the American Physical Therapy Association.
Unauthorized reproduction of this article is prohibited.
Recreational Movement, and Up/Down Stairs) of the School ministered the SFA to each participating student near the
Function Assessment (SFA), and the Vineland Adaptive beginning of the school year. At the end of 6 months, the
Behavior Scales. Stuberg and DeJong5 reviewed objectives PTs again completed the SFA on each student in collab-
at the impairment, functional limitation, and disability oration with other team members as appropriate. Greater
levels in the IEPs and Individualized Family Service Plans details on the methods of the larger study can be found in
for Early Intervention programs of 566 children receiving Effgen et al.10
physical therapy in one Nebraska school system. Their
careful review of these objectives indicated that 91% of
the objectives were scored as met or in progress. Participants
Although studying student outcomes on the basis Sample characteristics of the PTs and students are
of IEP goal achievement is very appropriate, systemwide provided in Table 1. Our final sample size was 109 PTs
and national studies usually include use of a standardized (mean age 46 years, standard deviation 9.2; 96% female),
measure.4,6 The SFA7 is a standardized, criterion- and 296 students with disabilities aged 5 to 12 years (x =
referenced tool, specifically developed to examine the 7.3, standard deviation = 2.02), from 59 school systems
functional performance and participation of children with in 28 states across the nation. The most common diagno-
disabilities in school from kindergarten through grade sis of the students was cerebral palsy followed by Down
6. The SFA can be used to measure skills that promote syndrome.
participation in the natural environmental settings of
general or special education, addressing both individual
and contextual factors. It is a judgment-based test that is Measures
both discriminative (identifies functional limitations) and Gross Motor Function Classification System
evaluative (measures change over time). The SFA contains (GMFCS). The GMFCS11 is a 5-level system designed to
3 sections focused on the student’s level of participation, classify children with cerebral palsy up to 18 years of age
extent of supports provided, and functional activities. It on the basis of performance in daily life. A classification is
assesses the student’s levels of activity, required support, made on the basis of current gross motor function in daily
and performance in daily school routines.7 activities. The GMFCS has evidence of content, construct,
The first purpose of this study was to report on and discriminative validity, and interrater reliability and
the descriptive results of student Participation, assistance is used in most research involving students with cerebral
and adaptation (Task Supports), self-care (Clothing Man- palsy.12-14 Although the GMFCS was developed for chil-
agement, Eating and Drinking, Hygiene), posture/mobility dren with cerebral palsy, we wanted to have an indication
(Maintaining and Changing Positions, Manipulation With of the functional ability level for all students in the study
Movement, Travel), and Recreational Movement outcomes and this was the most appropriate option.
as measured on the SFA after students received 6 months School Function Assessment. The SFA Part I Par-
of school-based physical therapy. Second, we examined ticipation, Part II Task Supports, and 7 of 8 scales of Part
the effects of age and functional mobility level on SFA III Activity Performance Physical Tasks were used in this
outcomes. study. According to the developers, separate scales of
the SFA can be used for targeted assessment.7(p.61) Part I
METHODS Participation exams the student’s level of participation in
6 settings: Classroom, Playground/Recess, Transportation,
Study Design and Procedures Bathroom/Toileting, Transitions, and Mealtime/Snack Time.
A practiced-based evidence8,9 research design was Participation is rated using a 6-point Likert scale from 1
used in the larger study, entitled “PT Related Child (extremely limited participation) to 6 (full participation).
Outcomes in the Schools” (PT COUNTS), to determine The ratings in each setting are summed to provide a partic-
what interventions predict outcomes of school-based phys- ipation total raw score, which can then be converted into
ical therapy. This article describes, via longitudinal re- a criteria score. Part II Task Supports exams the amount
search and comparative analysis, students’ performance of assistance and adaptations the student needs for the
on the SFA with data taken near the beginning of one Physical Tasks of Travel, Maintaining and Changing Posi-
school year and at the end of that school year. During the tion, Recreational Movement, Manipulation With Movement,
first year of the study, Institutional Review Board approvals Using Materials, Setup and Cleanup, Eating and Drinking,
were obtained and PTs were recruited from school districts Hygiene, and Clothing Management. Task Supports are rated
that agreed to participate. The PTs completed training and using a 4-point Likert scale from 1 (extensive assistance or
passed tests on research ethics, administration of the out- adaptations) to 4 (no assistance or adaptations). The assis-
come measures, and using the data collection system to tance and adaptations are rated separately and the separate
increase reliability of the data collection. They then re- raw scores ratings are summed and then converted into
cruited students. All PTs and parents of students signed criteria scores. Part III Activity Performance Physical Tasks
approved consent forms from the appropriate institutions exams the student’s ability to perform common school
before participation. Data collection started in year 2 of the activities on 21 separate scales using a 4 point Likert scale
study in the fall of the 2012 to 2013 school year. PTs ad- from 1 (does not perform) to 4 (consistent performance).8

372 Effgen et al Pediatric Physical Therapy

Copyright © 2016 Wolters Kluwer Health, Inc. and the Academy of Pediatric Physical Therapy of the American Physical Therapy Association.
Unauthorized reproduction of this article is prohibited.
 TABLE 1 TABLE 1
Characteristics of Physical Therapists and Students Characteristics of Physical Therapists and Students (Continued)

Characteristics Value Characteristics Value

Physical therapists, N = 109 Central 82 (28)

Age, mean (SD) 46 y (9.2) Northwest 78 (26)
Sex, n (%) Diagnostic category, n (%)b
Female 105 (96) Cerebral palsy 103 (35)
Male 4 (4) Down syndrome 46 (16)
Race, n (%) Other genetic syndromes 40 (14)
White 105 (96) Global developmental delay 32 (11)
Black 3 (3) Otherc 75 (25)
Other 1 (1) Gross Motor Function Classification
Ethnicity, n (%)a Level, mean age (SD); n (%)
Non-Hispanic or Latino 105 (96) I 6.7 y (1.72) 113 (38)
Hispanic or Latino 2 (2) II/III 7.8 y (2.1) 117 (40)
Professional degree, n (%) IV/V 7.7 y (2.2) 66 (22)
Certificate 2 (2) Manual Ability Classification Level, n (%)a
Baccalaureate 59 (54) I 53 (18)
Masters 34 (32) II/III 173 (59)
DPT 14 (13) IV/V 69 (23)
Highest postprofessional degree, n (%) Communication Function
None 58 (53) Classification Level, n (%)a
PT clinical masters 5 (5) I 48 (16)
PT clinical doctorate 25 (23) II/III 124 (42)
Academic Masters 19 (17) IV/V 123 (42)
Academic doctorate (PhD, EdD, DSc) 2 (2) Other services received at school,
Employment status, n (%) n (%)
Full-time 73 (67) Occupational therapy 256 (86)
Part-time 36 (33) Speech and language therapy 234 (79)
Employment relationship, n (%)a Adaptive physical education 123 (42)
Employed by school 82 (75) At grade level in school, n (%)a 73 (26)
Contract with school 11 (10) Classroom in which student spends
Through an agency 5 (5) most of the day, n (%)a
Other 10 (9) Typical classroom 89 (31)
Years in practice, mean (SD) Special classroom 114 (39)
As a PT 21 (10.4) Combination of both 86 (30)
As a pediatric PT 16 (9.7) Received physical therapy outside of 97 (33)
As a school-based PT 13 (9.1) education system, n (%)a
APTA member, n (%)a
Yes 57 (52) Abbreviations: PT, physical therapist; SD, standard deviation.
a Data were missing for some participants.
No 50 (46)
b Percentages do not equal 100 because of rounding.
Pediatric specialist certification, n (%)
c Other diagnostic categories include autism, learning disability,
Yes, current 9 (8)
Yes, not maintained 1 (1) attention-deficit hyperactivity disorder, speech language disorder, devel-
In process 8 (7) opmental coordination disorder, developmental delay because of health
No 91 (84) conditions, myelomeningocele, vision disorder, hearing disorder, trau-
matic brain injury, and limb deficiency.
Students, N = 296
Age, y, mean (SD) 7.3 (2.02)
Age group, n (%)
5-7 years old 172 (58)
For our Posture and Mobility outcome of interest we used
8-12 years old 124 (42) scales: Travel, Maintaining and Changing Positions, and
Sex, n (%) Manipulation with Movement; for Recreation and Fitness we
Female 130 (44) used the scale: Recreational Movement; and under Self-care
Male 166 (56) we used the scales: Eating and Drinking, Hygiene, and
Race, n (%)a
White 188 (66)
Clothing Management. These scales were chosen as they
Black 34 (12) relate to the adaptive and functional skills PTs commonly
Multiracial 27 (9) focus on when serving children in school.4 As a judgment-
Other 21 (7) based assessment, the PT was encouraged to consult with
Asian 16 (6) colleagues (teachers, aides, and related service providers)
Ethnicity, n (%)
Non-Hispanic or Latino 238 (82)
familiar with the student’s performance to determine
Hispanic or Latino 51 (18) the student’s rating on each item. For both the pre- and
Geographic region, n (%)b post-SFA results, the raw total scores were calculated
Northeast 51 (18) for each scale and then were converted into criterion
Southeast 85 (29) scores. The criterion scores range from 0 to 100, with 100
(continues)
indicating a full range of grade-appropriate functioning

Pediatric Physical Therapy Student Outcomes Measured by SFA 373

Copyright © 2016 Wolters Kluwer Health, Inc. and the Academy of Pediatric Physical Therapy of the American Physical Therapy Association.
Unauthorized reproduction of this article is prohibited.
in that area.7 The items within each scale are reported ANCOVAs are presented in Table 5. All overall ANCOVA
to be unidimensional and organized hierarchically where model analyses were statistically significant; however, no
each item becomes more difficult.8 Because the SFA was interactions between age and GMFCS were significant (not
developed using item response theory, the standard error reported in Table 5); therefore, the single effects of GMFCS
of measurement (SEM) varies by each criterion score for and age were considered and reported in Table 5. All scales
each scale. Psychometric studies suggest the SFA has high except Travel showed significant single-effect differences
internal consistency (0.92 to 0.98) for students receiving in change among the GMFCS level groups, with the ma-
special education and is comprehensive and appropriate jority showing significant differences in the change scores
for elementary students with disabilities.7,15-18 Test-retest between GMFCS levels I and IV/V and between levels II/III
reliability and interrater reliability studies reported good and IV/V. Students with higher gross motor abilities had
stability of the scores (test-retest r values > 0.80).7,18 higher change scores on the SFA. Participation, Maintaining
and Changing Positions, Recreational Movement, Hygiene,
and Clothing Management showed significant single-effect
Data Analyses differences between age groups, with the 5- to 7-year-old
Descriptive statistics were used to present the pre- and children showing higher change scores as compared with
postcriterion scores and the change from pre to post for the the 8- to 12-year-old children.
SFA scales. The sample sizes for each scale vary on the ba-
sis of the availability of completed data for that scale. SFA
scale criterion scores were then divided into 3 categories DISCUSSION
on the basis of the outcomes’ SEM; those with a criterion The majority of students improved on the SFA; how-
change score below −5, −5 to 5 and above 5. Because SEM ever, when the psychometric properties of the SFA are
varies by criterion score and scale, the investigators conser- considered and the SEM accounted for, only 37% to 51%
vatively chose these SEM categories to represent the whole of the students, depending on the scale, showed improve-
sample data. Those with criterion change scores below ment beyond estimated measurement errors. The greatest
−5 were considered to have regressed and those above 5 number of students improved in Participation (51%). Par-
to have improved in that scale. These data were used to ticipation within school settings is vital to a student’s suc-
describe the sample’s progress on SFA across the school cessful integration and functioning in school, and these
year of the study. Comparisons to determine age (5- to 148 students had meaningful improvements in school par-
7-year-old and 8- to 12-year-old groups) and GMFCS (level ticipation and required less supervision and assistance to
I, level II-III, and level IV-V groups) interaction or single participate after 6 months of intervention. This was fol-
effects using the students’ SFA change scores were com- lowed by positive change in Maintaining and Changing Po-
pleted using 2-way analysis of covariances (ANCOVAs), sition (46%), Manipulation With Movement (44%), Travel
controlling for students’ prescores. Post hoc comparisons (43%), and Recreational Movement (42%), areas likely ad-
were used in the event that the overall F-test was found dressed by the PT during the school year and the areas of
to be statistically significant. Alpha was set at 0.05 for all positive change found by King and colleagues.4 Concur-
statistical tests. SAS v9.3 was used for data management rently, the areas with some positive change, but not for
and statistical analyses. as many students, were ADL (Eating and Drinking [37%],
Hygiene [37%], and Clothing Management [38%]), perhaps
not as often addressed directly by the PTs. We encourage
RESULTS PTs to collaborate with other team members to identify
Table 2 provides the mean pre- and postcriterion and implement strategies to optimize students’ ADL.
scores, the number of students who changed in criteria Regression in performance in a scale was relatively
scores above, within, and below an SEM of 5. The majority rare and ranged from 5 (2%) to 24 (8%) students. The scale
of scores (51%-59%) in each scale, except for Participa- with the most regression in performance was Task Sup-
tion, had criterion scores in the −5 to 5 range suggest- ports: Assistance. The majority of students who regressed
ing no change. Scores above an SEM of 5 were found in were reported to have increased health issues, which may
37% to 51% of the students suggesting improvement in have then required more assistance for participation in the
SFA scores. The areas in which most students improved school activities. This was followed by Eating and Drink-
were Participation (51%), Task Support Adaptations (46%), ing, and Hygiene. Perhaps the PTs were not fully aware of
Maintaining and Changing Position (46%), and Manipula- the students’ true performance on ADL at the beginning
tion With Movement (44%). The areas in which the lowest of the school year, and then at the end of the year had a
number of students improved were in activities of daily liv- better appreciation of the students’ abilities, which might
ing (ADL), including Eating and Drinking (39%), Hygiene have been overestimated initially. The limited changes in
(37%), and Clothing Management (38%). From 2% to 8% performance for those with GMFCS levels IV and V are
of the students regressed in a scale. The criterion scores at not surprising on the basis of implications suggested in
pre- and posttesting and the change scores for all students the literature.19 The SFA is probably not sensitive or indi-
by age group are in shown Table 3 and by GMFCS level vidualized enough to detect the subtle changes in perfor-
group in Table 4. Results of the 2-way (GMFCS × age) mance in those students with lower gross motor function.

374 Effgen et al Pediatric Physical Therapy

Copyright © 2016 Wolters Kluwer Health, Inc. and the Academy of Pediatric Physical Therapy of the American Physical Therapy Association.
Unauthorized reproduction of this article is prohibited.
We would encourage use of more individualized measures 95% confidence interval of 48.08-55.92) and a posttest raw
such as goal attainment scaling20 or the Canadian Occupa- score of 47 (criteria score of 55 with an SEM of 2 and a 95%
tional Performance Measure21 or measures that focus on confidence level of 51.08-58.92). Because the confidence
the student’s specific disability such as the Gross Motor intervals overlap, the amount of change could be due to
Function Measure.22 chance or other errors and not indicate functional change.
Because we were conservative in classifying the per- However, consider this student with the raw score of 43
centage of students who made positive changes in perfor- who at the start of the school year receives a score of par-
mance on the basis of the SEM, in actuality, more indi- tial performance (2 points) on Travel item 9: Moves around
vidual students, on the basis of their own criterion scores, room freely with no or infrequent bumping into obstacles or
may have shown improvement above the SEM. In addi- people. Then near the end of the school year, the student
tion, even students who may have not changed above the receives a score of consistent performance (4 points) on
SEM may still have made clinically meaningful advances the same item. Most educators and PTs will consider that
in performance in the areas measured by the SFA. For ex- a clinically significant change in performance in the class-
ample, consider a student who had a pretest raw score of room. Not bumping into obstacles or people is important in
43 on Travel (criteria score of 52 with an SEM of 2 and a a school setting, yet the significance of this improvement

TABLE 2
School Function Assessment Outcomes

Pretest SFA Posttest SFA SFA Criterion Score Change

Mean SEM for Mean SEM for

Criterion Mean Criterion Mean Below −5 −5 to 5 Above 5
Scores Criterion Scores Criterion SEM, SEM, SEM,
SFA Subscale (SD) Score n 95% CI (SD) Score n 95% CI n (%) n (%) n (%)

Participation 54 (23) 4 291 42-65 59 (24) 4 295 47-71 9 (3) 134 (46) 148 (51)
Task Supports: Assistance 48 (26) 5 291 34-62 55 (27) 5 295 40-70 24 (8) 149 (51) 117 (40)
Task Supports: Adaptations 56 (27) 5 290 40-71 61 (28) 5 295 45-77 5 (2) 155 (53) 135 (46)
Travel 64 (25) 3 295 56-73 71 (25) 3 295 61-81 11 (4) 157 (53) 127 (43)
Maintaining and Changing 69 (25) 5 295 52-85 74 (25) 5-6 295 56-93 9 (3) 151 (51) 134 (46)
Positions
Recreational Movement 41 (22) 4 294 29-53 48 (27) 4 295 35-62 14 (5) 155 (53) 123 (42)
Manipulation With 57 (24) 3 295 47-67 63 (27) 3 292 50-75 9 (3) 155 (53) 127 (44)
Movement
Eating and Drinking 65 (28) 4 293 50-81 70 (29) 5 295 53-88 18 (6) 161 (55) 111 (39)
Hygiene 59 (31) 3 292 43-75 63 (32) 4 295 46-80 23 (8) 160 (55) 109 (37)
Clothing Management 53 (26) 3 294 42-64 57 (27) 3 295 46-69 11 (4) 172 (59) 111 (38)

Abbreviations: CI, confidence interval; n, sample size (varies based on data availability); SEM, standard error of the measurement; SFA, School Function
Assessment; SD, standard deviation.

TABLE 3
School Function Assessment Outcomes by Age Group

Pretest SFA Posttest SFA Criterion Change Scores

5-7 y Mean 8-12 y Mean 5-7 y Mean 8-12 y Mean

Criterion Criterion Criterion Criterion 5-7 y Mean 8-12 y Mean All Mean
Score Score Scores Score Change Score Change Score Change Score
SFA Subscale (SD), n (SD), n (SD), n (SD), n (SD), n (SD), n (SD), n

Participation 56 (22), 169 50 (23), 122 62 (24), 171 55 (23), 124 6.7 (9.4), 169 4.5 (6.2), 122 5.8 (8.2), 291
Task Supports: Assistance 50 (25), 169 45 (26), 122 57 (27), 171 51 (28), 124 7.6 (11.8), 169 6.7 (8.4), 122 7.2 (10.5), 291
Task Supports: 60 (27), 168 49 (27), 122 66 (28), 171 54 (27), 124 5.5 (12.8), 168 4.7 (8.5), 122 5.2 (11.2), 290
Adaptations
Travel 66 (24), 171 62 (26), 124 73 (24), 171 68 (26), 124 6.5 (10.2), 171 6.2 (7.9), 124 6.4 (9.3), 295
Maintaining and Changing 71 (25), 171 66 (26), 124 77 (34), 171 71 (26), 124 6.3 (10.7), 171 5.4 (8.5), 124 5.9 (9.8), 295
Positions
Recreational Movement 43 (21), 171 38 (24), 123 53 (27), 171 42 (25), 124 9.5 (12.4), 171 4.1 (6.8), 123 7.2 (10.8), 294
Manipulation With 59 (24), 171 54 (25), 124 66 (27), 170 58 (27), 122 6.2 (10.0), 170 4.5 (7.8), 122 5.5 (9.2), 292
Movement
Eating and Drinking 67 (28), 170 63 (30), 123 72 (29), 171 69 (30), 124 5.3 (10.9), 170 5.3 (10.3), 123 5.3 (10.6), 293
Hygiene 60 (29), 170 57 (34), 122 65 (30), 171 61 (33), 124 4.8 (12.1), 170 4.0 (8.3), 122 4.4 (10.7), 292
Clothing Management 54 (25), 171 52 (28), 123 59 (26), 171 55 (29), 124 4.7 (9.4), 171 3.1 (5.6), 123 4.0 (8.0), 294

Abbreviations: n, sample size (varies based on data availability); SFA, School Function Assessment; SD, standard deviation.

Pediatric Physical Therapy Student Outcomes Measured by SFA 375

Copyright © 2016 Wolters Kluwer Health, Inc. and the Academy of Pediatric Physical Therapy of the American Physical Therapy Association.
Unauthorized reproduction of this article is prohibited.
 TABLE 4
School Function Assessment Outcomes on the Basis of Gross Motor Function Classification System Level
Pretest SFA Criterion Score Posttest SFA Criterion Score SFA Criterion Change Scores
GMFCS GMFCS GMFCS GMFCS GMFCS GMFCS GMFCS GMFCS GMFCS
Level I Level II/III Level IV/V Level I Level II/III Level IV/V Level I Level II/III Level IV/V
Mean Mean Mean Mean Mean Mean Mean Mean Mean
SFA Subscale (SD), n (SD), n (SD), n (SD) n (SD), n (SD), n (SD), n (SD), n (SD), n
Participation 69 (16), 112 53 (14), 116 27 (21), 63 75 (16), 113 59 (16), 117 31 (21), 65 6.1 (9.1), 112 6.3 (8.2), 116 4.3 (6.4), 63
Task Supports: 65 (17), 112 48 (19), 115 16 (18), 64 75 (17), 113 55 (19), 117 20 (19), 65 9.4 (10.4), 112 6.6 (11.7), 115 4.3 (7.4), 64
Assistance
Task Supports: 75 (16), 112 56 (18), 115 21 (22), 63 82 (15), 113 60 (18), 117 25 (22), 65 6.3 (11.9), 112 4.8 (10.7), 115 3.8 (11.0), 63
Adaptations
Travel 81 (13), 113 66 (13), 117 33 (27), 65 87 (11), 113 73 (15), 117 40 (28), 65 5.5 (7.0), 113 6.7 (10.0), 117 7.3 (11.3), 65
Maintaining and 87 (13), 113 70 (16), 117 34 (19), 65 91 (11), 113 78 (17), 117 39 (18), 65 4.6 (8.3), 113 7.7 (11.6), 117 5.0 (8.5), 65
Changing Positions
Recreational Movement 58 (13), 113 41 (15), 116 12 (13), 65 70 (16), 113 46 (18), 117 15 (14), 65 11.9 (10.9), 113 5.5 (10.6), 116 2.3 (7.3), 65
Manipulation With 76 (15), 113 56 (14), 117 26 (19), 65 84 (15), 112 62 (17), 115 27 (20), 65 7.4 (8.5), 112 6.2 (9.8), 115 1.0 (7.6), 65
Movement
Eating and Drinking 80 (18), 112 68 (23), 117 36 (30), 64 85 (16), 113 74 (24), 117 38 (31), 65 5.4 (9.2), 112 6.6 (13.1), 117 2.6 (7.0), 64
Hygiene 77 (20), 112 61 (26), 116 23 (24), 64 81 (18), 113 65 (26), 117 25 (26), 65 5.0 (10.8), 112 4.6 (12.1), 116 3.3 (7.1), 64
Clothing Management 70 (15), 112 54 (20), 117 21 (20), 65 76 (15), 113 58 (21), 117 23 (22), 65 5.1 (6.5), 112 4.0 (9.3), 117 2.2 (7.7), 65

Abbreviations: GMFCS, Gross Motor Function Classification System; n, sample size (varies based on data availability); SD, standard deviation; SFA,
School Function Assessment.

is lost in the summative data and interpretation of SFA Limitations

results. This is a common complication of using standard- There was no control of intervening factors that might
ized measures, which highlights the need to also consider have influenced the outcomes such as service delivery
use of individualized measures. model, amount, frequency, and intensity of direct services
Children in the younger age group (5-7 years old) and services on behalf of the student. Although all the
showed more positive change in Participation, Maintaining PTs completed training and had to pass a test on the SFA,
and Changing Positions, Recreational Movement, Hygiene, there was no assessment of their accuracy of completing
and Clothing Management scores on the SFA as compared the students’ SFA. As already noted, issues regarding the
with older children (8-12 years old). When considering the SEM of the SFA affected our interpretation of the find-
tasks, many of the items are skills teachers and PTs focus ings because the most students did not exceed the SEM,
on during the early elementary years. However, as children which would suggest change in performance beyond mea-
age, the focus shifts to compensatory or adaptive strategies surement error. This is a drawback of this measurement
when the skills have not been achieved at younger ages. tool. Although use of a more sensitive measure of school
In addition, younger children are more likely to make de- function might have shown more positive results, there
velopmental gains because of maturation, whereas with is presently no such tool available. So although use of
the older children gains might be more likely due to in- standardized, norm-referenced measures might be the gold
tervention. The effects of therapy might not be captured standard for outcomes research, the scarcity of adequately
specifically on a standardized measure such as the SFA sensitive functional measures of school performance of stu-
where progressing from inconsistent to consistent perfor- dents with disabilities seriously restricts research efforts.
mance may not be a realistic expectation for older children Finally, we used the GMFCS to classify all children within
with disabilities. the study; however, research has only supported using this
In addition to measuring student outcomes, standard- functional classification measure for children with cerebral
ized measures of student performance are being used in palsy.11
therapist performance appraisal. On the basis of our find-
ings, the SFA would appear to be more appropriate to use
with students younger than 8 years and those at GMFCS CONCLUSIONS
levels I, II, and III. PTs need to reflect how to best measure Most students made positive gains within the scales
and support the progress of students who are older or have of the SFA across 6 months of school-based physical ther-
more limitations. We suggest that therapists specifically apy. Only a few students had regression in performance
consider the addition of individualized measures to cap- of scale items usually reported because of a new medical
ture the student’s performance on a broad array of func- problem. The greatest number of students made progress
tional skills required for school and that they especially in the SFA scales of Participation, Task Supports: Adapta-
promote team collaboration. tions, Maintaining and Changing Position, and Manipulation

376 Effgen et al Pediatric Physical Therapy

Copyright © 2016 Wolters Kluwer Health, Inc. and the Academy of Pediatric Physical Therapy of the American Physical Therapy Association.
Unauthorized reproduction of this article is prohibited.
 TABLE 5
ANCOVA Results: 3 (GMFCS Groups, I, II/III, IV/V) × 2 (Age Groups, 5- to 7-Year-Olds, 8- to 12-Year-Olds)

SFA Subscale Overall ANCOVA Modela GMFCS Main Effectb Age Main Effectc

Participation, n = 266 F = 3.75 F = 5.45 F = 7.67

df = 6, P = .001 df = 2, P = .005 df = 1, P = 0.006
I to IV/V (P = .01)
II/III to IV/V (P = .001)
Adaptations, n = 285 F = 7.23 F = 16.43 NS
df = 6, P < .0001 df = 2, P < .0001
I to II/III P = .001
I to IV/V P < .0001
II/II to IV/V P = .0001
Assistance, n = 266 F = 4.12 F = 10.47 NS
df = 6, P = .001 df = 2, P < .0001
I to II/III P = .01
I to IV/V P < .0001
II/II to IV/V P = .0004
Maintaining and Changing Positions, n = 285 F = 10.54 F = 18.26 F = 3.99
df = 6, P < .0001 df = 2, P < .0001 df = 1, P = .047
I to IV/V P < .0001
II/III to IV/V, P < .0001
Manipulation With Movement, n = 279 F = 5.57 F = 11.74 NS
df = 6, P < .0001 df = 2, P < .0001
I to IV/V P < .0001
II/III to IV/V, P < .0001
Recreational Movement, n = 277 F = 11.22 F = 4.87 F = 16.62
df = 6, P < .0001 df = 2, P = .008 df = 1, P < .0001
I to II/III, P = .005
I to IV/V, P = .005
Travel, n = 284 F = 3.96 NS NS
df = 6, P = .001
Hygiene, n = 269 F = 5.69 F = 10.72 F = 4.24
df = 6, P < .0001 df = 2, P < .0001 df = 1, P = .04
I to IV/V, P = .0001
II/III to IV/V, P < .0001
Clothing Management, n = 283 F = 2.89 F = 3.08 F = 7.86
df = 6, P = .01 df = 2, P = .048 df = 1, P = .005
I to IV/V, P = .03
II/III to IV/V, P = .02
Eating and Drinking, n = 275 F = 5.25 F = 12.17 NS
df = 6, P < .0001 df = 2, P < .0001
I to IV/V, P = .0003
II/III to IV/V, P < .0001

Abbreviations: ANCOVA, analysis of covariance; GMFCS, Gross Motor Function Classification System; n, sample size (varies based on data availability);
NS, nonsignificant; SFA, School Function Assessment.
a All interactions of GMFCS × age groups were nonsignificant.
b For all significant differences between GMFCS level groups, the children with higher gross motor functional ability had higher changes scores, when

pretest scores were controlled for, than children with lower gross motor functional ability.
c For all significant differences between age groups, the 5- to 7-year-old children had higher change scores, when pretest scores were controlled for, than

the 8- to 12-year-old children.

With Movement followed by Travel and Recreational Move- REFERENCES

ment, areas likely addressed by the PTs during the school 1. Public Law 94-142, Education of All Handicapped Children Act
year. Fewer students made gains in the ADL, areas that (1975), 89 Stat. 773-796.
might not have been a principal focus of the physical ther- 2. Public Law 108-446, Individuals with Disabilities Education
apy intervention. Students who were 8-12 years old, and Improvement Act of 2004. [Link]
[Link].
those who showed less functional ability on the basis of the
3. Effgen SK, McEwen I. Review of selected physical therapy inter-
GMFCS (levels IV-V), demonstrated fewer improvements ventions for school-age children with disabilities. Phys Ther Rev.
on the SFA across the year as compared with children who 2008;13(5):297-312.
were 5 to 7 years old and children with greater functional 4. King GA, McDougall J, Tucker MA, et al. An evalua-
ability (GMFCS levels I-III). Therapists need to consider tion of functional, school-based therapy services for chil-
dren with special needs. Phys Occup Ther Pediatr. 1999;19(2):
how and with whom to use the SFA to monitor change
5-29.
across time and with therapy.

Pediatric Physical Therapy Student Outcomes Measured by SFA 377

Copyright © 2016 Wolters Kluwer Health, Inc. and the Academy of Pediatric Physical Therapy of the American Physical Therapy Association.
Unauthorized reproduction of this article is prohibited.
 5. Stuberg W, DeJong SL. Program evaluation of physical therapy as an 14. Wood E, Rosenbaum P. The Gross Motor Function Classification
early intervention and related service in special education. Pediatr System for Cerebral Palsy: a study of reliability and stability over
Phys Ther. 2007;19(2):121-127. time. Dev Med Child Neurol. 2000;42(5):292-296.
6. American Physical Therapy Association. Guide to Physical Thera- 15. Davies PL, Soon PL, Young M, Clausen-Yamaki A. Validity and re-
pist Practice 3.0. Alexandria, VA: American Physical Therapy Associ- liability of the School Function Assessment in elementary school
ation; 2014. [Link] Accessed Septemper students with disabilities. Phys Occup Ther Pediatr. 2004;24(3):23-
22, 2015. 43.
7. Coster W, Denny T, Haltiwanger J, Haley SM. School Function Assess- 16. Gates PE, Otsuka NY, Sanders JO, McGee-Brown J. Relationship be-
ment. San Antonio, TX: The Psychological Corporation; 1998. tween parental PODCI questionnaire and School Function Assess-
8. Gassaway J, Horn SD, DeJong G, Smout RJ, Clark C, James R. Ap- ment in measuring performance in children with CP. Dev Med Child
plying the clinical practice improvement approach to stroke rehabil- Neurol. 2008;50(9):690-695.
itation: methods used and baseline results. Arch Phys Med Rehabil. 17. Hwang LJ, Davies PL. Brief report: Rasch analysis of the School Func-
2005;86(12 suppl 2):S16-s33. tion Assessment provides additional evidence for the internal valid-
9. Teeter L, Gassaway J, Taylor S, et al. Relationship of physical therapy ity of the activity performance scales. Am J Occup Ther. 2009;63:
inpatient rehabilitation interventions and patient characteristics to 369-373.
outcomes following spinal cord injury: the SCIRehab project. J Spinal 18. Hwang J, Davies PL, Taylor MP, Gavin WJ. Validation of School
Cord Med. 2012;35(6):503-526 Function Assessment with elementary school children. Occup Ther J
10. Effgen SK, Westcott McCoy S, Chiarello LA, Jeffries LM, Bush Res. 2002;22(2):48-58.
H. Physical therapy-related child outcomes in school: an exam- 19. Dewar R, Love S, Johnston LM. Exercise interventions improve pos-
ple of practice based evidence methodology. Pediatr Phys Ther. tural control in children with cerebral palsy: a systematic review. Dev
2016;28(1):47-56. doi:/PEP.0000000000000197. Med Child Neurol. 2015;57(6):504-520.
11. Palisano RJ, Rosenbaum P, Bartlett D, Livingston MH. Content valid- 20. McDougal J, King G. Goal Attainment Scaling: Description, Utility, and
ity of the expanded and revised Gross Motor Function Classification Applications in Pediatric Therapy Services. 2nd ed. London, Ontario,
System. Dev Med Child Neurol. 2008;50(10):744-750. Canada: Thames Valley Children’s Centre; 2007.
12. Palisano R, Rosenbaum P, Walter S, Russell D, Wood E, Galuppi 21. Law M, Baptist S, Carswell A, McColl MA A, Polatajko H, Pol-
B. Development and reliability of a system to classify gross mo- lack N. COPM Canadian Occupational Performance Measure. 4th ed.
tor function in children with cerebral palsy. Dev Med Child Neurol. Ottawa, Ontario, Canada: The Canadian Association of Occupational
1997;39(4):214-223. Therapists; 2005.
13. Palisano RJ, Hanna SE, Rosenbaum PL, et al. Validation of a model 22. Russell DJ, Rosebaum PL, Wright M, Avery LM. Gross Motor Function
of gross motor function for children with cerebral palsy. Phys Ther. Measure (GMFM-66 and GMFM-88) User’s Manual. 2nd ed. London:
2000;80(10):974-985. Mac Keith Press; 2013.

378 Effgen et al Pediatric Physical Therapy

Copyright © 2016 Wolters Kluwer Health, Inc. and the Academy of Pediatric Physical Therapy of the American Physical Therapy Association.
Unauthorized reproduction of this article is prohibited.