Performance following ability-focused physical therapy intervention in individuals with severely limited physical and cognitive abilities.

Key Words: Ability-focused physical therapy, Generalization, Goal attainment scaling, Motor learning, Severe physical and cognitive disability, Transfer.

Gross motor abilities that allow individuals to interact with people and their environment are essential to the development of basic cognitive and communication abilities,[1-3] People who do not move and interact with others, or who are excluded from participation in social opportunities (eg, education, meals, work and recreation with peers without motor impairments), may expect social failure and might not try to move or interact, a learned helplessness.[4,5] Although individuals with severely limited physical and cognitive abilities may learn gross motor abilities that are useful to them, these abilities often will not look like, or develop in the same order as, the skills of peers their age without disabilities.[6] Examples of gross motor abilities that may enhance function include (1) arm and hand movements to activate an environmental control unit (eg, to turn a television or lamp on or off); (2) arm and hand movements to contact objects, signal others, or use an alternative communication system; (3) trunk and limb movements to reposition the body for social interaction and comfort; (4) protective limb movements to prevent or slow a fall; and (5) active movements that oppose forces that could lead to the development of deformities. These abilities may help individuals with severely limited cognitive and physical abilities engage in social interactions in addition to reducing the physical assistance required for their care."

Gains in gross motor ability in individuals with severely limited physical and cognitive abilities have been reported following intervention.[7-11] The gains reported include improved head controls,[8-9] improved arm movements[8] and a pattern of sitting,[7] maintenance of a 4-point posture,[8,9] standing up,[11] and maintenance of walking.[10,11] Interventions were administered 2 to 3 times per week, with a duration of 10 to 40 minutes per session. Treatment included high-repetition practice of behavioral objectives,[7-11] knowledge of results,[7-11] systematic fading of prompts,[7-11] behavior shaping,[7-9.11] use of minimal physical assistance,[7,11] and adapted positioning.[8,9]

Individuals with severely limited physical and cognitive abilities require high-repetition training and protracted treatment to obtain meaningful gains.[12] Although long-range goals must be focused on meaningful functional outcomes within the individual's environments, short-term objectives may have to focus on achievable components of those long-range outcomes. In contrast to studies of successful approaches that focused on specific abilities and systematic fading of prompts,[7-11] limited progress has been reported following neurodevelopmental physical therapy that focused on acquisition of motor milestones, quality of movement, muscle tone, and postural reactions.[13,14]

Ability-focused physical therapy, as defined in this report, emphasizes teaching gross motor abilities that are important to both the individual and the individual's caregivers. This approach to therapy is a composite that we developed from the aim-oriented approach of Scrutton[15] and from physical therapy, psychology, and special education procedures that have been reported to enhance the learning of individuals with severely limited physical and cognitive abilities (Appendix 1).[2,5,6,15-49] Ability is defined as the capability to perform an activity. Ability-focused therapy is designed for use with individuals with severely limited physical and cognitive abilities whose potential to learn the skilled actions typical in the daily routines of their peers without disability is extremely limited. We contend that this is an important distinction for individuals who may not learn typical skills but who may acquire useful gross motor abilities.

The emphasis of ability-focused physical therapy is on simple motor behaviors that are meaningful in the individual's current or future environment. Ability-focused physical therapy is not based on adherence to the motor development sequence or on the expectation that persons with severely limited physical and cognitive abilities will move in a manner similar to people without disability. Rather, this approach is characterized by the therapist's (1) acceptance of individually preferred types and levels of stimulation and ways of moving (eg, the patient may not tolerate normal lighting and sound levels), (2) use of behavior management to enhance motivation, and (3) communication of goals and outcomes to the individual, using strategies designed for persons with severely limited cognitive abilities and sensory impairments. Physical impairments are treated only when they are thought to interfere with attainment of motor skills or may result in the long-range loss of motor ability. During practice, the therapist attempts to gradually and systematically reduce the amount of (1) physical assistance and (2) tactile, visual, and verbal guidance provided to the individual. Therapist feedback and reinforcement are directed primarily at leading to an individual's success in achieving a desired goal, rather than on qualitative aspects of movement. In our view, the emphasis of ability-focused physical therapy on potentially useful, but not necessarily skilled, movements helps to avoid a learned expectation of failure that may occur if the therapist requests normal movement patterns that the individual is not capable of achieving.

Limited transfer of motor behaviors to spontaneous use with other persons, in other places, and at other times of the day has been extensively reported in the literature of psychology and special education, which primarily investigated fine motor and communication abilities, for individuals with severely limited cognitive abilities,[8,16,17] In the few studies of physical therapy for individuals with severely limited physical and cognitive abilities in which transfer of new abilities was examined, there was some, but limited, transfer when the therapist prompted the learner in nontreatment settings.[7,10,11,14,18] There are no studies on transfer of gross motor abilities following physical therapy in which the individual was prompted by persons naturally present in those environments (eg, family, friends, caregivers). The primary purpose of our study was to examine whether individuals aged 3 to 30 years with severely limited physical and cognitive abilities will improve their gross motor ability after receiving 18 weeks of ability-focused physical therapy. The secondary purpose of our study was to examine whether abilities initially attained in the clinical setting would be used when prompting is supplied by a person naturally present in the subjects' recess or home setting.

The primary hypothesis was that, after 18 weeks of physical therapy, the subjects' gross motor level for the treatment goals that were practiced would be higher than the level for control goals. The secondary hypotheses were (1) when the subjects first reach their expected gross motor level for goals treated in the physical therapy setting, their level for the treatment goals would be lower when they are prompted in the recess and home settings and (2) after 18 weeks of physical therapy, when the subjects are prompted in the recess and home settings, their gross motor level for treatment goals would not differ from the level for control goals.

Method

Subjects

A single-group pretest-posttest design was used, with subjects serving as their own controls. The subjects were 24 children and young adults who were nonambulatory, had severely limited physical and cognitive abilities, and resided in Custer County, South Dakota. Only 2 individuals received direct physical therapy services just prior to the study. Guardian consent was sought for all 3- to 30-year-old individuals in Custer County, South Dakota, who met the following inclusion criteria:

1. Received special education services under an individualized education program with the Custer Independent School District 16-1 while living in a family home or received special education or workshop services under an individualized habilitation plan in developmentally grouped classrooms, with on-grounds residence, at the South Dakota Developmental Center at Custer.

2. Did not walk or use a wheelchair independently.

3. Scored below an adaptive behavior achievement age of 20 months on the Preschool Attainment Record (PAR)[50] on their last comprehensive assessment. The PAR is an achievement test with high test-retest and interrater reliability[51,52] that is often used by psychologists to assess adaptive behaviors of individuals of all ages with limited cognitive abilities.

4. Scored below a gross motor age equivalent of 13 months on the Gross Motor Scale of the Peabody Developmental Motor Scales[53] (PDMS-GM) on their last comprehensive assessment.

After receiving approval from the institutional review boards, informed consent was obtained from each subject's guardian. Thirty-three children and young adults met the initial inclusion criteria. Three potential subjects were excluded because they had undergone recent major surgery, and their surgeon prevented their participation. Two potential subjects were excluded because their guardians did not provide consent. Three potential subjects, although scoring low enough on their prior PAR or PDMS-GM assessment (done up to 3 years earlier), scored too high on one of those assessments tar study inclusion when retested at the start of the study. Twenty-five subjects started the study. One subject was discontinued due to a femur fracture not associated with physical therapy.

Of the 24 subjects, 10 subjects (42%) were female and 14 subjects (58%) were male. The mean age of the sample was 20.1 years (SD=8.1, range=3-30). The limited abilities of most subjects were attributed to more than one etiology, and only 1 subject had limited ability of unknown etiology. Three subjects (12%) had cerebral palsy, 12 subjects (50%) had microcephaly, 2 subjects (8%) had hydrocephaly, 4 subjects (17%) had other cranial anomalies, 3 subjects (12%) had post-maternal infection, 3 subjects (12%) had post-maternal drug use, 6 subjects (25%) had genetic disorders, 4 subjects (17%) had cerebral anoxia, and 2 subjects (8%) had low birth weight ([is less than] 2,500 g). Seven subjects (29%) were fed by gastrostomy pumps. Although not controlled, no changes occurred in use of pharmaceutical agents during the study. Subjects had a mean score of 10.5 months (SD=4.1, range = 3.4-19.5) on the PAR adaptive behavior assessment and a mean score of 1.5 months (SD=2.8, range=0.0-12.0) on the PDMS-GM.

Selection and Quantification of Goals

Three gross motor goals were selected for each subject. The treating therapist, guardians, caregivers, teachers, and subjects participated in goal selection. In accordance with ability-focused therapy, the goals selected were related to the subjects' needs and preferences instead of therapist-perceived needs. A summary of the types of goals selected is presented in Table 1.

Table 1.
Types of Goals Selected for Subjects

No. of
Goals     Gross Motor Ability Targeted

   27     Head control (11 prone/9 pull-to-sit/5 sitting/2 supine)
   12     Rolling
    8     Propping on forearms and crawl on elbows
    8     Protective arm extension and propping on hands
    5     Reaching and grasping to midline over chest while supine
    5     Sitting (assume and maintain)
    2     Stand from floor
    2     Stand free of support
    2     Walk using walker
    1     Attain 4-point posture
   72     Total (3 goals per individual)

Goals were written in goal attainment scaling (GAS) format. Goal attainment scaling is an individualized client-centered outcome measure that incorporates the components of the behavioral objective and includes an ordinal scale that states possible levels of goal attainment following a therapeutic intervention.[54] The GAS format uses 5 possible outcomes, as opposed to only 1 outcome tar the behavioral objective. For GAS, the expected level of attainment for each goal is stated, as are the 2 possible levels of attainment that are less favorable and the 2 possible levels of attainment that are more favorable. This format allows the physical therapist to determine whether a patient has made progress despite not achieving the desired goal and whether a patient's progress has surpassed the expected outcome. Another useful and unique feature of GAS is the ability to compute a change score for multiple goals, providing an aggregate measure of change. The composite change score can be expressed as a T score with a mean equal to 50 and a standard deviation of 10.

In our study, the -2 level of goal attainment was a description of each subject's current abilities based on some element of his or her PAR and PDMS-GM assessments and the judgment of the interdisciplinary, team.[55] The 0 level of goal attainment reflected the interdisciplinary team's expectation of the most likely outcome after 18 weeks of physical therapy. The +2 level of goal attainment represented the team's judgment of the best possible outcome alter 18 weeks physical therapy. Behaviors representing attainment between these levels were used for the -1 and + 1 scores. To illustrate the method of scaling, the 3 goals approved for subject 10 are shown in Appendix 2, along with his case history and the 100-mm-long visual analog scale lines used to assess the content validity of each item. The goals, in our view, were functionally relevant, based on individual needs, anti are representative of goals developed for individuals with severely limited physical and cognitive abilities. For most of the subjects, the interdisciplinary team did not anticipate that fully functional skills could be attained in only 18 weeks. Most goals, therefore, represented what we think were components of functional tasks.

After each subject's interdisciplinary team approved the 3 goals, a die was cast to randomly select the 1 control goal. The 2 remaining goals were the treatment goals that were practiced in the physical therapy setting. The team was unaware of which goals were practiced.

Procedure for Measurement of Goal Attainment

Each subject's performance for each goal was videotaped on 4 test days in the physical therapy (at the subject's school), recess, and home settings. The test days were at the study start, after 9 and 18 weeks of physical therapy, and when the treating therapist determined that a subject first reached the expected outcome (0 level of goal attainment) for a treatment goal in the physical therapy setting. Subjects did not receive physical therapy on test days. Prior to beginning each therapy session, the treating therapist prompted the subject to perform each goal to determine when the subject first attained a score of 0. This protocol was modified in 2 instances to keep tine number of measurements of each goal constant. If a subject did not achieve the expected outcome for a goal prior to the 16th week of therapy, the subject was videotaped attempting to achieve that goal and the control goal during the 17th week to keep the number of measurements for each goal constant. If a subject first achieved the expected level for a goal in the physical therapy setting after 9 weeks of therapy, a fourth videotaping was not necessary, for that goal.

On each test day, subjects were videotaped first in the physical therapy setting during their therapy time, followed by videotaping in their recess setting and then in their home setting. The order of the settings corresponded to each subject's daily routine. Within each setting, the order of goal prompting on the 3 goals was randomized to minimize goal-order effects. For the subjects in public school, the recess setting was their adapted physical education class. For subjects residing at the state developmental center, the recess setting was their recreation class.

An adult (training technician or caregiver) who routinely interacted with the subject prompted goals in the recess or home setting. After assisting the subject to assume the initial position, prompting included cuing the subject to perform tine specified gross motor activity using environmentally appropriate subject-specific stimuli, signs, or gestures. For example, a recreation technician might use a promise of activation of a favorite game, verbal or tactile encouragement, and praise in front of peers to motivate a subject. The subject was allowed tip to 3 trials per goal per setting, with any spontaneous attempts counted as a trial.

The adult who prompted the subject in each setting remained constant throughout the study. The goals of subjects who attended public school were prompted by tine aides who assisted them in adaptive physical education and the family member who was their most consistent caregiver in the home setting. For subjects who resided at the state developmental center, goals in the recess setting were prompted by their recreation technician, and goals in the home setting were prompted by their evening caregiver.

Individuals with severely limited physical and cognitive abilities do not always perform within their capacity when tested. When a subject did not perform even at the -2 level, a score of -3 was given for that test day. The independent rater scored a -3 level on 13.0% of her ratings (112 out of 864 tests). We suggest that, when using GAS, failure to document an individual's actual attainment may result in underestimation of within-group variance, thereby inflating reported effect sizes.

After data collection, the test-day videotapes of each subject were randomly reordered prior to scoring by an independent rater. The independent rater was an experienced pediatric physical therapist who did not provide services to any of these subjects. The rater's experience included the writing and scoring of behavioral objectives similar to those included in the GAS scales and the provision of direct physical therapy services to individuals with severely limited cognitive and physical abilities.

Content validity of measurements related to goals. To assess how meaningful the goals were to the home setting, caregivers were asked to complete a l00-nnnlong visual analog scale (Appendix 2). Each caregiver completed the visual analog scale at the beginning of the study and again near the end of the study. All persons who served as caregivers for 20 or more of the prior 100 days were asked to complete survey forms. The caregiver was instructed to mark the point on the scale that reflected his or her judgment of the potential usefulness of the goal for the individual's current and future settings. The term "fixture settings" was defined as any future living, work, or recreation environments the individual may have access to in the foreseeable future. Responses were measured to the nearest millimeter. There was a 100% return of visual analog scale forms. For the 22 subjects with more than one caregiver, weighted average scores (using the number of days of service provided by each of their 2 caregivers for the weightings) were used to compute the score.

The caregivers' mean ratings of goal usefulness were 75.8 (SD=16.5) at the start of the study and 72.3 (SD=21.3) at the end of the study. The results suggest that the goals were considered of potential use to the subjects. The low correlation (Pearson matched-pair r=-.09, n=72, P [is greater than] .25) between the start and end of the study ratings, however, suggests that caregivers changed their opinion of which goals had more usefulness after prompting the subjects to achieve the goals during the test sessions.

Interrater reliability of GAS scores. Interrater reliability was examined by having the treating therapist score each subject's levels of goal attainment in the physical therapy setting from direct observation on the test clays. The treating therapist's scores were compared with those of the independent rater who scored the subjects' levels of goal attainment from the videotapes. To examine interrater reliability across all 3 settings, a third physical therapist independently scored the levels of goal attainment of 3 subjects from the videotapes. Those 108 scores (3 subjects X 3 goals X 3 sites X 4 test times) were compared with the independent racer's scores. Pearson r correlation coefficients assess whether scores are associated in a linear pattern. In contrast, the intraclass correlation coefficient (ICC) assesses whether 2 raters obtain similar means. We chose to examine both aspects of reliability.

The Pearson r correlations and interrater ICCs (2,1) between the scores of the treating therapist and the independent raters were r=.84 (P [is less than] .0001, n=360, [r.sup.2]=70.9%) and ICC=1.00 (between raters: df=1, SS=0.01; within raters: df=695, ss=1172.65), respectively. The coefficients between scores of the 2 independent raters were r=.81 (P [is less than] .0001, n=135, [r.sup.2]=66.2%) and ICC=.997 (between raters: (df=1, SS=1.48; within raters: df-245, SS=433.39). The results support acceptable interrater reliability of the scores for the goals in this study.

Treatment Procedure

Each subject received one-on-one direct physical therapy twice a week for 18 weeks. Eighteen weeks was selected based on the first author's clinical experience working with this population, which suggested that a minimum of 4 months was needed to obtain meaningful progress. The therapy sessions lasted 30 minutes, with 10 to 15 minutes per session devoted to practice of each treated goal. The first author provided treatment to all subjects. He had 18 years of experience working with individuals with limited cognitive and physical abilities, he had a master's degree in special education, and he was certified in pediatric physical therapy by the American Board of Physical Therapy Specialties.

The physical therapy intervention consisted of (1) goal-relevant treatment of joint, fascia, and skin contracture in addition to facilitation or inhibition of specific muscle use by application of passive motion, quick or prolonged stretching, tapping, and vibration[19]; (2) preparatory techniques, including positioning, environmental modification, and sensory stimuli to either alert or calm the subject; (3) use of behavior management principles and techniques; (4) communication principles specific to the unique needs of the subject; and (5) high-repetition practice of the 2 targeted gross motor abilities. The specific techniques used are described in Appendix 1. Stretching procedures were done only when tissue tightness or laxity were thought to interfere with an individual's goal-related movement. All prolonged stretching was limited to procedures described by Boehme.[56] The specific procedures and strategies used were based on the therapist's interpretation of each subject's goals and needs.

Individuals with severely limited cognitive ability need high task repetition prior to ability acquisition.[8] Unfortunately, clients, therapists, teachers, aides, and family members may become bored anti lose motivation if the same prompts, materials, and rewards are always used. General case instruction,[20,57,58] in which the relevant antecedent stimuli (prompts or materials) are varied, was used in an effort to help maintain subject interest in the task and to increase the likelihood of ability transfer to the subject's recreation and home settings. Reinforcers (consequent stimuli) that proved successful were alternated to avoid subject boredom and eventual lack of response to single reinforcers. Thus, treatment was characterized by individualized variation in both antecedent (cue) and consequent (reinforcer) stimuli, but it was consistent in regard to which abilities the subjects practiced.

Treatment Consistency

The subjects received 82.6% of the planned sessions ([bar]X= 14.9, SD= 1.1, range=12-17) during the first 9-week period and 84.7% of the planned sessions ([bar]X=15.3, SD=1.1, range=14-18) during the second 9-week period. The missed sessions were due to subject illness and to treatment cancellation on the 4 test days.

Our approach to physical therapy incorporated principles from behavioral psychology and special education. A behavioral psychologist and a special education director observed the therapy sessions of 3 randomly selected subjects and scored checklists on the correct use of behavior management and communication principles, respectively. A second independent physical therapist also observed therapy for 3 randomly selected subjects and scored a checklist related to treatment procedures. All raters were experienced in serving this population. The 3 checklists were developed based on the treatment and management procedures used in this study and were previously published.[59]

Data Analysis

Computerized data entry was done using dBASE IV, version 1.5.(*) Data were analyzed using the Statistical Package for the Social Sciences (SPSS/PC+, version 3.0).([dagger]) The required level of significance for all tests was P=.01, instead of the more common level of P=.05. We used this level of significance to avoid the increased chance of a Type I error that occurs with repeat analysis of data.

Two-goal GAS T scores were computed for treatment goals, whereas a one-goal GAS T score was computed for the control goal. These scores were computed for each preplanned test day (study start, 9 weeks, 18 weeks) and for the at-criterion-level test day, using the formula developed by Kirsuk and Shermann[54]:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

where [W.sub.i] represents the weighting for a particular goal and [X.sub.i] represents the outcome score for each goal (a value of -2 to +2). The r value in the formula reflects the estimated average intercorrelation for the outcome scores of multiple goals. The formula for computing the T score assumes a correlation of .30 between goals. In this study, all goals received equal weight to allow comparison of the mean GAS T scores for the treatment goals with the mean GAS T scores for the control goals. The relationship between the GAS T scores and the GAS raw scores from which they were computed is shown in Table 2.

Table 2.
Goal Attainment Scaling (GAS) T Score Equivalents for 1-Goal
Control and 2-Goal Treatment GAS Raw Scores

GAS T       1-Goal Control   2-Goal Treatment
Score       Raw Score        Raw Scores

12.8                         -3,-3
19.0                         -3,-2
20.0        -3
25.2                         -3,-1 or -2,-2
30.0        -2
31.4                         -3,1 or -2,-1
37.6                         -3,1 or -2,0 or -1,-1
40.0        -1
43.8                         -3,2 or -2,1 or -1,0
50.0         0               -2,2 or -1,1 or 0,0
56.2                         -1,2 or 0,1
60.0         1
62.4                         0,2 or 1,1
68.6                         1,2
70.0         2
74.8                         2,2

Goal attainment scaling was designed to simulate parametric-like data distributions, particularly when larger numbers of goals are combined into subject T scores. We ran both parametric and nonparametric statistics (Friedman 2-way analyses of variance [ANOVAs] and Wilcoxon matched-pairs signed-rank tests). Because both types of statistics gave the same results and the GAS T scores were normally distributed, only the parametric statistics are presented.

A 2-factor (goal type [treatment versus control] X test day) repeated-measures ANOVA was done to determine whether there was a significant goal X test day interaction, indicating a differential change between GAS T scores for treatment and control goals across the initial, 9-week, and 18-week test days. A separate analysis was done for each of the 3 test settings (physical therapy, recess, home). For the settings in which the goal type X test day interactions were significant, a Student paired t test was run on 18-week GAS T scores to test for a treatment effect. The t value was based on the mean sums of squares of the error term from the ANOVA.

The mean GAS T score for the 27 treatment goals in which the subjects were videotaped when they first attained the 0 level of goal attainment in the physical therapy setting was by definition 50; thus, there was no variance. The between-setting difference in mean GAS T scores, therefore, were analyzed using Student t tests for matched pairs, with the standard error of the mean of the non-physical therapy setting score used for the usual standard error of the difference between setting scores.

Results

After 18 weeks of therapy in the physical therapy setting, 9 subjects had a level of goal attainment of 0 or higher for both treatment goals, 10 subjects had a score of 0 or higher for one of their treatment goals, and the remaining 5 subjects had levels of goal attainment of -1 or -2 for either treatment goal. Two of the 5 subjects did not demonstrate any change in level of goal attainment after 18 weeks of therapy in the physical therapy setting. Both subjects died within a year after the study, suggesting to us that their potential for change was limited at the time of the study. One of those individuals also had severe opisthotonos, despite use of muscle relaxant medications. Mean GAS T scores for treatment and control goals are shown in Table 3. The mean GAS T score of 45.6 for treatment goals in the physical therapy setting was close to the expected outcome GAS T score of 50.

Table 3.
Mean Goal Attainment Scaling T Scores by Test Day, Treatment or
Control Goal, and Setting (N=24)

                              Setting
                              Physical Therapy

Test Day and Type of Goal     [bar] X   SD      Range

At study start
  Treatment(a)                27.00     8.1     12.8-50.0
  Control(b)                  30.40     6.9     20.0-50.0
After 9 weeks
  Treatment(a)                37.30     11.8    25.2-62.4
  Control(b)                  32.50     7.9     20.0-60.0
After 18 weeks
  Treatment(a)                45.60     10.5    25.2-68.6
  Control(b)                  34.60     11.8    20.0-60.0

                              Recess
Test Day and Type of Goal     [bar] X   SD      Range

At study start
  Treatment(a)                  28.0     8.0    19.0-43.8
  Control(b)                    32.5     9.4    20.0-50.0
After 9 weeks
  Treatment(a)                  30.6     8.2    12.8-50.0
  Control(b)                    31.7    10.1    20.0-60.0
After 18 weeks
  Treatment(a)                  31.1    10.3    12.8-62.4
  Control(b)                    32.5     9.9    20.0-50.0

                              Home
Test Day and Type of Goal     [bar] X   SD      Range

At study start
  Treatment(a)                27.8       8.8    12.8-56.2
  Control(b)                  30.4       8.1    20.0-50.0
After 9 weeks
  Treatment(a)                30.0       9.7    19.0-50.0
  Control(b)                  32.5       9.9    20.0-60.0
After 18 weeks
  Treatment(a)                35.5       8.7    19.0-62.4
  Control(b)                  35.0      10.6    20.0-50.0

(a) Two-goal T scores.

(b) One-goal T scores.

GAS T Scores for Treatment and Control Goals Measured in Physical Therapy Setting

The subjects' mean GAS T scores for treatment goals ([bar] X=45.6, SD=10.5, range=25.2-62.4) compared with control goals ([bar] X=34.6, SD=11.8, range=20.0-60.0) measured in the physical therapy setting are plotted in the Figure (plot A). The results of the 2-way ANOVA indicate a goal type X test day interaction (F=8; df=2,143; P [is less than] .001). The results of the matched-pairs t test indicated that the mean GAS T score for the treatment goals was higher than the mean T score tar the control goals (t=4.06, df=23, P [is less than] .001). The primary hypothesis was accepted. After 18 weeks of physical therapy, the subjects' progress on treatment goals was greater than their progress on the control goals.

[Figure ILLUSTRATION OMITTED]

GAS T Scores Measured in Recess and Home Settings When Goals Were Attained in Physical Therapy Setting

The independent rater also scored 27 of the treatment goals in the recess and home settings on test days when the therapist stated the subject first achieved the expected level of goal attainment in the physical therapy setting. The results of the t tests to compare mean GAS T scores for each setting are presented in Table 4. When the subjects first reached their expected outcome in the physical therapy setting, they did not perform as well in the recess and home settings, supporting secondary hypothesis 1. The difference in mean T scores between the home setting and the recess setting was not significant.

Table 4.
Matched-Pair t Tests of (GAS) T Scores for Treatment
Goals When Subjects First Reached the Expected Level
of Goal Attainment in Physical Therapy, by Setting

GAS T Scores for Physical Therapy Setting Versus Recess Setting

Setting                                    N      [bar] X

Physical therapy                           27     50.0
Recess                                     27     35.9

Mean difference= 14.1
(SD= 11.5, SE=2.2)
t value(a)=6.35, df=26, 1-tailed
P<.00001, d=2.49

GAS T Scores for Physical Therapy Setting Versus Home Setting

Setting                                    N      [bar] X

Physical therapy                           27     50.0
Home                                       27     42.2

Mean difference=7.8
(SD= 12.2, SE=2.3)
t value(a)=3.31, df=26, 1-tailed
P<.005, d=1.30

GAS T Scores for Recess Setting Versus Home Setting

Setting                                    N      [bar X]

Recess                                     27     35.9
Home                                       27     42.2
Mean difference=6.3
(SD= 14.5, SE=2.8)
t value=2.26, df=26, 2-tailed
P<.01, d=0.89

GAS T Scores for Physical Therapy Setting Versus Recess Setting

Setting                                    SD     SE

Physical therapy                           0      0
Recess                                     11.5   2.2

Mean difference= 14.1
(SD= 11.5, SE=2.2)
t value(a)=6.35, df=26, 1-tailed
P<.00001, d=2.49

GAS T Scores for Physical Therapy Setting Versus Home Setting

Setting                                    SD     SE

Physical therapy                           0      0
Home                                       12.0   2.3

Mean difference=7.8
(SD= 12.2, SE=2.3)
t value(a)=3.31, df=26, 1-tailed
P<.005, d=1.30

GAS T Scores for Recess Setting Versus Home Setting

Setting                                    SD     SE

Recess                                     11.5   2.2
Home                                       12.2   2.3
Mean difference=6.3
(SD= 14.5, SE=2.8)
t value=2.26, df=26, 2-tailed
P<.01, d=0.89

GAS T Scores for Physical Therapy Setting Versus Recess Setting

Setting                                    Range

Physical therapy                           50.0-50.0
Recess                                     20.0-60.0

Mean difference=14.1
(SD= 11.5, SE=2.2)
t value(a)=6.35, df=26, 1-tailed
P<.00001, d=2.49

GAS T Scores for Physical Therapy Setting Versus Home Setting

Setting                                    Range

Physical therapy                           50.0-50.0
Home                                       20.0-60.0

Mean difference=7.8
(SD= 12.2, SE=2.3)
t value(a)=3.31, df=26, 1-tailed
P<.005, d=1.30

GAS T Scores for Recess Setting Versus Home Setting

Setting                                    Range

Recess                                     20.0-60.0
Home                                       20.0-60.0
Mean difference=6.3
(SD= 14.5, SE=2.8)
t value=2.26, df=26, 2-tailed
P<.01, d=0.89

(*) Based on the SE of tile mean in tile non-physical therapy setting.

Comparison of Subjects' GAS T Scores for Recess and Home Settings

Subjects' GAS T scores for the recess and home settings are plotted in the Figure (plots B and C, respectively). The results of the 2-way ANOVAs indicate that there was no goal type X test day interaction for either the recess setting (F=1; df=2,143; P [is greater than] .30) or the home setting (F=2; df=2,143; P [is greater than] .15). Secondary hypothesis 2 was accepted. The subjects' level of goal attainment on treatment versus control goals did not differ across time in the recess and home settings.

We examined post hoc how many subjects demonstrated transfer of gross motor ability from the physical therapy setting to either the recess or home setting after 18 weeks of physical therapy. Compared with their level of goal attainment in the physical therapy setting, the subjects had the same or a higher level of goal attainment for 17 (36%) of their treatment goals in the recess setting and for 22 (46% of their treatment goals in the home setting.

Discussion

Effects of Ability-Focused Physical Therapy

We attributed the subjects' higher GAS T scores for treatment goals compared with control goals to the effects of ability-focused physical therapy. The effect size (d=1.69) was large, and the goal type X test day interaction for the ANOVA indicates a differential change between GAS T scores for treatment versus control goals across the initial, 9-week, and 18-week test days. Additionally, each subject's control goal was randomly selected, which minimized the probability of differences in difficulty between treatment and control goals. Based on our experience, we suggest that the characteristics of the subjects in this pilot study are representative of the population of individuals with severely limited physical and cognitive abilities. Furthermore, treatment was provided by one physical therapist who had extensive experience and skills working with this population. Future research will need to include other subject samples and treating therapists before the efficacy of ability-focused physical therapy can be asserted for individuals with severely limited physical and cognitive abilities.

The results of our study support the findings of 5 other studies of physical therapy interventions,[7-11] suggesting that the techniques included in our ability-focused approach to providing physical therapy were effective in teaching gross motor abilities to individuals with severely limited physical and cognitive abilities even though there was transfer to the home setting for only 46% of the treatment goals. In contrast, 4 other studies that investigated individuals with severe physical and cognitive disabilities demonstrated little or no improvement in gross motor abilities following neurodevelopmental treatments or vestibular stimulation.[13,14,60,61] Those studies focused treatment on quality of movement patterns and postural reactions based on normal development. This approach is different from our approach, where the focus was on initial learning of simple gross motor tasks that might later be combined with other simple movements to enable performance of a task meaningful to the subjects and their caregivers. We emphasized strategies that we believed were specific to the learning needs of individuals with severely limited physical and cognitive abilities. We think the strength of our approach is that physical therapy procedures were incorporated into strategies commonly used by psychologists and educators to enhance the learning of individuals with severe disabilities.

Maturation probably did not contribute to improved attainment on the control goals, because the mean age of the subjects was 20 years. A training effect from testing of control goals was unlikely, as the testing was sparse. Rather, the tendency for the subjects to make small gains on control goals most likely reflected the subjects' positive relationship with, and a sense of knowing how to respond to, their therapist. These observations suggest that social interaction and self-initiation may also be important for initial learning of gross motor skills in individuals with severely limited physical and cognitive abilities.

Our primary objective was to examine whether the subjects would demonstrate a small change in the gross motor abilities that were practiced in the physical therapy setting, yet a change that would be meaningful to their caregivers. The gross motor abilities were thought to be functionally relevant and reflected the judgment of each subject's interdisciplinary, team on the change expected over an 18-week period, Our study was restricted by limited knowledge of 3 questions that are important for outcome-based research. These questions are: (1) How is function defined for this population? (2) What are the changes in gross motor function that individuals with severely limited physical and cognitive abilities are capable of making? and (3) What constitutes change in motor ability that is meaningful to individuals and caregivers? A better understanding of these issues is essential to investigation of the effectiveness of physical therapy for individuals with severely limited physical and cognitive abilities.

Transfer of Gross Motor Ability

As hypothesized, the subjects' improved attainment of goals in the physical therapy setting did not transfer fully to their recess or home settings. Only 4 subjects (17%) demonstrated transfer of gross motor ability to the recess setting for both of their treatment goals. Subjects performed as well in the recess setting as in the physical therapy setting on 36% of the goals practiced in the physical therapy setting. These ,findings are similar to those reported for transfer of cognitive and language abilities in individuals with severely limited cognitive abilities.[16] Little has been reported on transfer from therapy to home settings.[7,10] In our study, only 4 subjects (17%) also demonstrated transfer of gross motor ability to the home setting for both treatment goals. Subjects performed as well in the home setting as in the physical therapy setting on 46% of the goals practiced in the physical therapy setting.

Three factors, related to the study design and procedures, may have contributed to the lack of transfer of gross motor ability. Constraints imposed by the study design required that each subject's interdisciplinary team be unaware of which goals were practiced in therapy. Staff members and caregivers, therefore, were not permitted to observe and participate in physical therapy, a departure from conventional practice. If the staff and caregivers who prompted the subjects to perform the goals in the recess and home settings had knowledge of the treatment goals, they may have inadvertently assisted the subjects in practicing the goals. Such knowledge would have confounded investigation of transfer of gross motor ability and potentially biased how the subjects were prompted on test days.

The lack of transfer of gross motor ability also may be related to the higher amount of distracting visual and auditory stimuli present in the recess and home settings. Individuals with visual impairment may be irritated by what is considered proper lighting. All except 2 of the subjects demonstrated a preference for reduced general stimuli, particularly visual stimuli. Therefore, during treatment the amount of lighting was intentionally reduced, although it was not reduced during videotaping in any setting to maintain a balanced comparison among settings.

The greater percentage of goals in which subjects demonstrated transfer of gross motor ability to the home setting compared with the recess setting may reflect differences among the persons who prompted the subjects to exhibit motor behavior in each setting. Recess at the state developmental center consisted primarily of group activities involving limited, short-duration, one-on-one contacts. Two of the 3 recreation technicians never appeared comfortable prompting subjects during test days and repeatedly requested therapist guidance on how to motivate the subjects. In the home setting, although some caregivers initially were uncomfortable when asked to prompt subjects to perform gross motor tasks, they developed confidence as the study progressed.

Although subjects were able to perform new gross motor tasks when prompted by their therapist in the physical therapy setting, their abilities in performing these tasks often did not consistently transfer (the term "generalization" is used in special education[60]) to recess and home settings. Similar setting-specific findings for communication and fine motor skills have been reported for individuals with severely limited physical and cognitive abilities.[58] The partial transfer of ability to other settings suggests that, for this population, providing physical therapy in an isolated setting is not always effective when the goals are to improve function and to reduce the amount of caregiver assistance needed to perform activities as part of daily routines in home, education, and community settings. Further research is needed to determine whether provision of physical therapy in natural environments and use of planned methods to improve transfer will result in more consistent transfer of abilities to other environments and increase spontaneous, unprompted use of these abilities.

Conclusion

The purpose of our study was to examine whether individuals with severely limited physical and cognitive abilities improve gross motor abilities when given ability-focused physical therapy. After 18 weeks of ability-focused physical therapy provided in an isolated setting, 24 subjects aged 3 to 30 years had higher GAS T scores for treatment goals compared with control goals. Tire results are attributed to the emphasis in physical therapy on subject-preferred methods of moving and incorporation of strategies used by psychologists and educators to enhance learning of individuals with severely limited physical and cognitive abilities. The subjects' level of goal attainment in the therapy setting, however, did not transfer to their recess and home settings. The results suggest that physical therapy provided in an isolated setting is only partly effective when the goals are to improve the ability of individuals with severely limited physical and cognitive abilities to perform motor activities as part of daily routines.

(*) Inprise Carp, 100 Enterprise Way, Scotts Valley, CA 95066.

([dagger]) SPSS Inc, 444 N Michigan Ave, Chicago, IL 60611.

References

[1] Perrin EC, Gerrity PS. Development of children with a chronic illness. Pediatr Clin North Am. 1984;31:19-31.

[2] Piaget J. The Origins of Intelligence in Children. New York, NY: International Universities Press; 1952.

[3] Van Dijk J. Motor Development in the Education of Deaf-Blind Children: Seminar on the Teaching of Deaf-Blind Children. Boston, Mass: Perkins School for the Blind; 1964.

[4] Levy, J. Social reinforcement and knowledge of results as determinants of motor performance among EMR children. American Journal of Mental Deficiency. 1974;78:752-758.

[5] Seligman MEP. Learned helplessness. Annu Rev Med. 1972;23: 407-412.

[6] Campbell PH, McInerney WF, Cooper MA. Therapeutic programming for students with severe handicaps. Am J Occup Ther. 1984;38: 594-602.

[7] Bragg JH, Houser C, Schumaker J. Behavior modification: effects on reverse tailor sitting in children with cerebral palsy. Phys Ther. 1975; 55:860-868.

[8] Horn EM, Warren SF. Facilitating the acquisition of sensorimotor behavior with a microcomputer-mediated teaching system: an expericognitive analysis. Journal of the Association for Persons With Severe Handicaps. 1987;12:205-215.

[9] Horn EM, Warren SF, Reith HJ. Effects of a small-group microcomputer-mediated motor skills instructional package. Journal of the Association for Persons With Severe Handicaps. 1992; 17:133-144.

[10] O'Brien F, Azrin NH, Bugle C. Training profoundly retarded children to stop crawling. J Appl Behav Anal. 1972;5:131-137.

[11] Westervelt VD, Luiselli JK. Establishing standing and walking behavior in a physically handicapped, retarded child: clinical report. Phys Ther. 1975;55:761-765.

[12] Burch MR, Clegg JC, Bailey JS. Automated contingent reinforcement of correct posture. Res Der Disabil. 1987;8:15-20.

[13] Noonan MJ. Teaching postural reactions to students with severe cerebral palsy: an evaluation of theory and technique. Journal of the Association for Persons With Severe Handicaps. 1984;9:111-122.

[14] Wright T, Nicholson J. Physiotherapy for the spastic child: an evaluation. Dev Med Child Neurol. 1973;15:146-163.

[15] Scrutton D. Aim-oriented management. In: Scrutton D, ed. Management of the Motor Disorders of Children With Cerebral Palsy. Philadelphia, Pa: JB Lippincott Co; 1984:49-58.

[16] Rutherford RB Jr, Nelson CM. Generalization and maintenance of treatment effects. In: Witt JC, Elliott SN, Gresham FM, eds. Handbook of Behavior Therapy. New York, NY: Plenum Press; 1988:227-324.

[17] Horner RH. Stimulus control, transfer, and maintenance of upright walking posture with a severely retarded adult. In: Bellamy GT, Horner RH, Inman DP, eds. Habilitation of the Severely and Profoundly Retarded. Eugene, Ore: University of Oregon Press; 1977;2:69-80.

[18] Fowler SA, Rowbury TG, Nordyke NS, Baer DM. Color-matching techniques to train children in the correct use of stairs. Phys Ther. 1976;56:903-910.

[19] Cherry DB. Review of physical therapy alternatives for reducing muscle contracture. Phys Ther. 1980;60:877-881.

[20] Stokes TF, Osnes PG. The developing applied technology of generalization and maintenance. In: Homer RH, Dunlap G, Koegel RL, eds. Generalization and Maintenance: Life Style Changes in Applied Settings. Baltimore, Md: Patti H Brookes Publishing Co; 1988:5-19.

[21] Held JM. Recovery of function after brain damage: theoretical implications for therapeutic intervention. In: Carr JH, Shepherd RB, Gordon J, et al, eds. Movement Science Foundations for Physical Therapy in Rehabilitation. Rockville, Md: Aspen Publications Inc; 1987:155-178.

[22] Threlkeld AJ. The effects of manual therapy on connective tissue. Phys Ther. 1992;72:893-902.

[23] Bishop B. Vibratory stimulation, part 3: possible applications of vibration in treatment of motor dysfunctions. Phys Ther. 1975;55:139-143.

[24] Brunnstrom S. Movement Therapy in Hemiplegia. New York, NY: Harper & Row; 1970.

[25] Ayers J. Sensory Integration and Learning Disorders. Los Angeles, Calif.' Western Psychological Services; 1972:120.

[26] Farber SD, ed. Sensorimotor Evaluation and Treatment Procedures for Allied Health Personnel. 2nd ed. Indianapolis, Ind: Indiana University Foundation; 1974.

[27] Newton MJ, Lehmkuhl D. Muscle spindle response to body heating and localized muscle cooling: implications for relief of spasticity. Phys Ther. 1965;45:91-105.

[28] Kamm K, Thelen E, Jensen JL. A dynamical systems approach to motor development. Phys Ther. 1990;70:763-775.

[29] Biederman GB, Davey VA, Ryder C, Franchi D. The negative effects of positive reinforcement in teaching children with developmental delay. Except Child. 1994;60:458-465.

[30] Ilmer S, Drews J. Differential analysis of selected prompts and neurological variables in motor assessment of moderately cognitively retarded children. American Journal of Mental Deficiency. 1980;84:508-517.

[31] Skinner BF. Science and Human Behavior. New York, NY: Macmillan; 1953.

[32] Dewson MRJ, Whiteley JH. Sensory reinforcement of head turning with nonambulatory, profoundly mentally retarded individuals. Res Dev Disabil. 1987;8:413-426.

[33] Alberto PA, Troutman AC. Applied Behavior Analysis for Teachers. 2nd ed. Columbus, Ohio: Merrill; 1986:220-226.

[34] DeVore MS. Individualized Learning Programs for the Profoundly Retarded. Springfield, Ill: Charles C Thomas, Publisher; 1977.

[35] Munk DD, Repp AC. The relationship between instructional variables and problem behavior: a review. Except Child. 1994;60:390-401.

[36] Donnellan AM, Mirenda PL, Mesaros RA, Fassbender LL. Analyzing the communicative functions of aberrant behavior. Journal of the Association for Persons With Severe Handicaps. 1984;9:201-212.

[37] Cintas HM. The relationship between motor skill level and risk-taking during exploration in toddlers. Pediatric Physical Therapy. 1992; 4:165-170.

[38] Alberto PA, Schofield P. An instructional interaction pattern for the limitedly handicapped. Teaching Exceptional Children. 1979;12:16-21.

[39] Higgins S. Motor skill acquisition. Phys Ther. 1991;71:123-139.

[40] Van Dijk J. The non-verbal deaf-blind child and his world of symbols. In: Vermelde Studies I. Sint Michielsgestel, the Netherlands: Instituut Voor Doven; 1963.

[41] Adler SS, Beckers D, Buck M. PNF in Practice. New York, NY: Springer-Verlag New York Inc; 1993.

[42] Dummer GM. Information processing in the acquisition of motor skills by mentally retarded children. Dissertation Abstracts International. 1978;38:7914584. Doctoral dissertation, University of California at Berkeley.

[43] Shea JB, Zimny ST. Context effects in memory and learning movement information. In: Magill RA, ed. Memory and Control of Action. New York, NY: Elsevier Science Inc; 1983:345-366.

[44] Gentile AM. Skill acquisition: action, movement, and neuromotor processes. In: Cart JH, Shepherd RB, Gordon J, et al, eds. Movement Science: Foundations for Physical Therapy in Rehabilitation. Rockville, Md: Aspen Publishers Inc; 1987:93-154.

[45] Slominski AH. Winthrop Phelps and the Children's Rehabilitation Institute. In: Scrutton D, ed. Management of the Motor Disorders of Children With Cerebral Palsy. Philadelphia, Pa: JB Lippincott Co; 1984:59-74.

[46] Freeman MAR, Wyke B. Articular contributions to limb muscle reflexes: the effects of partial neurectomy of the knee-joint on postural reflexes. Br J Surg. 1966;53:61-68.

[47] Harris FA. Facilitation techniques and technological adjuncts in therapeutic exercise. In: Basmajian JV, ed. Therapeutic Exercise. 4th ed. Baltimore, Md: Williams & Wilkins; 1984:110-178.

[48] Winstein CJ. Motor learning considerations in stroke rehabilitation. In: Duncan PW, Badke MB, eds. Stroke Rehabilitation: The Recovery of Motor Control. Chicago, Ill: Year Book Medical; 1987:109-134.

[49] Winstein CJ. Knowledge of results and motor learning: implications for physical therapy. Phys Ther. 1991;71:140-149.

[50] Doll EA. Preschool Attainment Record: Research Edition. Circle Pines, Minn: American Guidance Service; 1966.

[51] Buros OK, ed. Seventh Mental Measurements Yearbook. Highland Park, NJ: Gryphon Press; 1972.

[52] Owens EP, Bowling DH. Internal consistency and factor structure of the Preschool Attainment Record. American Journal of Mental Deficient. 1970;75:170-171.

[53] Folio MR, Fewell RR. Peabody Developmental Motor Scales and Activity Cards. Hingham, Mass: DLM Teaching Resources; 1983.

[54] Kirsuk TJ, Shermann RE. Goal attainment scaling: a general method for evaluating comprehensive community cognitive health programs. Community Ment Health J. 1968;4:443-453.

[55] Ottenbacher KJ, Cusick A. Goal attainment scaling as a method of clinical service evaluation. Am J Occup Ther. 1990;44:519-525.

[56] Boehme R. Myofascial Release and Its Application to Neuro-Developmental Treatment. Available at: http://www.boehmeworkshops.com. Accessed 1991.

[57] Horner RH, McDonald RS. Comparison of simple instance and general case instruction in teaching a generalized vocational skill. Association for the Severely Handicapped Journal. 1982;8:7-20.

[58] Stokes TF, Osnes PG. Programming the generalization of children's social behavior. In: Strain PS, Guralnick M, Walker H, eds. Children's Social Behavior: Development, Assessment, and Modification. Orlando, Fla: Academic Press Inc; 1986:407-443.

[59] Brown DA. Acquisition and generalization of gross motor ability following physical therapy in persons with significant physical and mental disabilities. Dissertation Abstracts International. 1994;56:95-30472.

[60] Ottenbacher K, Short A, Watson PJ. The effects of a clinically applied program of vestibular stimulation on the neuromotor performance of children with severe developmental disability. Physical & Occupational Therapy in Pediatrics. 1981;1(3):1-28.

[61] Sommerfeld D, Fraser BA, Hensinger RN, Beresford CV. Evaluation of physical therapy service for severely mentally impaired students with cerebral palsy. Phys Ther. 1981;61:338-344.

[62] Billingsley FF, White OR, Liberty KA. Writing objectives for generalization, probing skill use, strategies to improve generalization, and decision rules and procedures for generalization. In: Haring NG, ed. Generalization for Students With Severe Handicaps. Seattle, Wash: University of Washington Press; 1988:73-100.

Appendix 1.

Outline of the Ability-Focused Method of Physical Therapy Used in This Study

Service Delivery Components

* Isolated physical therapy area in individual's school (common practice and required by study design, not part of ability-focused method)

* Staff and family not informed of individual's progress and did not train gross motor goals (not part of ability-focused method, but necessary for control in study)

* Student and family aim-oriented therapy goals[15]

* Task-specific individual physical therapy treatment[21]

Preparatory Physical Therapy Techniques

* Prolong stretching to 2 min per hold[19,22]

* Facilitate low muscle tone by vibrating or tapping the weak agonist muscles[23,24]

* Use reflex-inhibiting movements, slow rocking, or pressure on the tendons to decrease resistance from spastic antagonists to movement[25-27]

* Use environmental changes and positioning to increase the likelihood that the individual will anticipate the varied times and places to use new ability

* Fade above techniques as individual ability and strength improve to avoid a dependence on therapist that will block skill transfer[18]

* Use starting positions that ease the desired motion with gravity or physically block nonuseful options and require the motion[28]

* Use modeling to decrease individual need for verbal guidance[29]

Behavior Management Strategies Used During Practice of Goals

* Use least-to-maximum multimodal antecedent prompts until ability emerges, then decrease verbal and tactile prompts until only natural situational cues signal the need to use the ability[17,30]

* Vary training materials and style of presentation to avoid boredom and satiation

* Reinforce correct responses, improved effort, or approximations of desired responses[31]

* Use individual-specific, age-appropriate reinforcers[32]

* Give continuous reinforcement for approximations of target ability and improved effort until initially learned, then give reinforcement intermittently to maintain ability[33,34]

* Where tasks can be broken into components, practice partial elements and intersperse the component practice with high-probability-of-success, individual-choice-relevant tasks to decrease problem behaviors[35]

Communication Principles Used During Practice of Goals

* Remain sensitive to individual's cues regarding his or her interests and needs[20,36]

* Adjust intensity of sound, lighting, and touch low enough for individual's tolerance, yet high enough to encourage alertness[30,32]

* Set level of challenge between individual's need for both security and risk-taking[5,37]

* Use simple, clear, concise, age-appropriate language

* Allow time for processing of language (eg, respect individual's latency of response by spacing brief requests)[33]

* Use gesture or full-contact deaf-blind signing, as appropriate

* Maintain a positive emphasis; for example, say "Good trying," "Interesting way," or "How clever[33,38] * Use the word "No!" only to warn of immediate danger; use "Try another way" to guide[16]

* Allow individual to learn at own pace, through own trials and experiences, providing feedback (knowledge of results)[37,39]

* Initially align your body in direct contact with the individual to ease learning body parts and imitation of desired movements and posture[2,26]

* Respect individual need for nurturing, followed by looking for individual resonance signals to continue a motion passively, followed by co-active action (parallel imitation), then distance (fade) your prompts[2,40]

Physical Therapy Techniques During Practice of the Goals

* For weak agonists, recruit muscles into synergistic contraction patterns by repetition of approximate movement, shaping performance to the desired ability[41]

* Use spaced, brief-massed practice (high repetition of one ability at a time) with rest periods, as needed[42]

* When needed, suggest what the individual moves, and how, while respecting the individual as an active learner and problem solver[43]

* Use adaptive devices (orthoses or equipment) to bypass disabilities that will not change[44]

* Include strengthening and endurance activities in spiral movements (the "diagnosis" of proprioceptive neuromuscular facilitation literature) matching muscle alignment[42]

* Use weight bearing, approximation of joints, and rhythmic stabilization techniques[42,45]

* Select positions that use gravity to initially help, or later resist, movements[25]

* Use head-position-initiated postural reactions to guide movement, when available, until strength, endurance, and awareness adequate to attempt ability without such guidance[6,46]

* Guide movements minimally, allowing the individual to experience the consequences of self-initiated variation in position, timing, and force[47-49]

Appendix 2.

Case Description For Subject 10 and the Visual Analog Scale Used to Assess His Caregiver's Impression of the Potential Usefulness of the Selected Goals(a)

Chris is a 22-year-old man with Pierre Robin syndrome and global profound developmental delay. He is nonverbal and tends to get skin rashes. Complications from left pelvis surgery resulted in left hip joint dislocation 6 years before. Chris' worsening moderate kyphoscoliosis was a major concern for his father and caregivers. Trunk bracing had not slowed the trend toward increasing trunk deformity, thought to be related to the weakness of Chris' postural muscles. His father was also concerned about Chris' lack of protective reactions because Chris spontaneously assumed a sitting position but did not lower himself to the floor or a mat safely. Chris' standardized test scores were: PAR adaptive behavior age=10 months, PAR ambulatory age=0.5 year, PDMS-GM basal/ceiling levels=0/12-14 months, PDMS-GM age equivalent=4 months, PDMS-GM scaled score=389.

Visual Analog Scales Completed by Student Caregivers Functional Movement and Posture Training Study

The 3 training goals identified by our study for Chris and approved For his IEP are listed below. We are told you provide at least 20% of Chris' care. We want to know how important these objectives are to Chris. We will ask only 4 questions. You will not have to give us your name.

Please place a vertical mark across the line below between the arrows showing how useful Chris might find the ability described, both now and in future settings. Do this for each goal:

Goal L3: Chris will improve his neck and upper back strength to counteract his trunk deformity. When placed on his stomach on a mat with his elbows by his sides and given a novel object to obtain, Chris will, on 1 of 3 trials in physical therapy:

-2 attempt holding his head erect (baseline) -1 raise his head to midline, chin touching mat 0 clear his chin from the mat, head in midline +1 raise his head to vertical midline +2 raise his head to vertical midline, then extend a hand above shoulder level to get object

[ILLUSTRATION OMITTED]

Goal L4: Chris will improve his posture and propping reactions for safety. When placed on his back on a mat with his left side blocked and given a novel object to obtain, Chris will, on 1 of 3 trials in physical therapy:

-2 roll to his left side to sit up, if not blocked (baseline) -1 roll to his right side to attain partial side-sitting, propped on right elbow 0 roll to right side to attain full side-sitting +1 push himself to right side-sitting using both hands +2 roll to right side using counterrotation, prior to coming to sitting

[ILLUSTRATION OMITTED]

Goal L5: Chris will develop an ability to protect himself during falls. When supported by his waist, with his head tipped toward the floor mat, Chris will, on 1 of 3 trials in physical therapy:

-2 over a padded :barrel, take no weight on both hands (baseline) -1 over a padded barrel, actively extend both hands, touching mat 0 over a barrel, maintain weight on hands for 2 seconds +1 when placed in 4-point posture without barrel, maintain weight on hands and knees for 2 seconds +2 when placed in 4-point posture without barrel, maintain weight on hands and knees for 5 seconds

[ILLUSTRATION OMITTED]

On about how many of the upcoming 100 days do you expect to be Chris' primary caregiver?

--days.

Today's Date: --.

Please let us know the names of, and how to contact, any other caregiver(s) who may be doing over 20 days of Chris' care during the next 100 days below:

YOU DO NOT NEED TO SIGN THIS FORM

When completed, please put this questionnaire in the enclosed envelope and drop in the mail. Thank you for your time and help! Subject 10

(a) PAR=Preschool Attainment Record,[50] PDMS-GM=Gross Motor Scale of the Peabody Developmental Motor Scales,[53] IEP=individualized education program.

DA Brown, PhD, PT, PCS, is Assistant Professor, Department of Physical Therapy, University of New England, Marcil Hall, Room 106, 11 Hills Beach Rd, Biddeford, ME 04005-9,599 ([e-mail]) Address all correspondence to Dr Brown.

SK Effgen, PhD, PT, is Professor and Director, Pediatric Physical Therapy Program, Allegheny University of the Health Sciences (formerly The Medical College of Pennsylvania and Hahnemann University), Philadelphia, Pa.

RJ Palisano, ScD, PT, is Professor and Director. Program in Movement Science, Allegheny University, of the Health Sciences. He is also Co-Investigator, Neurodevelopmental Clinical Research Unit, McMaster University, Hamilton, Ontario, Canada.

This research was completed in partial fulfillment of the requirements of Dr Brown's doctoral studies in pediatric physical therapy at The Medical College of Pennsylvania and Hahnemann University.

This work was supported in part by the US Department of Education (OSER No. H029D00005), the South Dakota Department of Human Services, the Foundation for Physical Therapy Inc, and the American Physical Therapy Association Section on Pediatrics.

A portion of this study was presented at the Second Joint Congress of the Canadian Physiotherapy Association and the American Physical Therapy Association, June 5, 1994, Toronto, Ontario, Canada.

This study was approved by the Hahnemann University Human Subjects Committee and the South Dakota Developmental Center at Custer Human Rights Committee.

This article was submitted January 27, 1997, and was accepted February 24, 1998.