The Psychological And Social Factors Related To Successful Prosthetic Training In Juvenile Amputees: A Preliminary Study

Everett I. Campbell, Ph. D. Judith C. Bansavage, M. S. Jean Holland Van Ormer E. Paul Dick

Personnel engaged in the rehabilitation of juvenile amputees have frequently noted that the physical capacities of the child and the mechanical characteristics of the prosthesis alone are insufficient for accurate prediction of the degree of success the child will attain in the utilization of his prosthesis. Steensma (12) has reported: "The problem is not entirely solved by our providing adequate surgery, excellent post-operative care, expertly fitted modern prostheses and the latest training techniques. It would seem that a careful investigation of the psychological and social factors involved is essential."

Siller (11) has suggested the potential value of psychological insights in the prescription, design, and construction of prostheses and in use training. The therapeutic desirability of such insights in the practical management of juvenile amputees and their parents was also stressed. In a study reported in 1960, Siller (11) used 52 upper and lower extremity child amputees of both sexes to study four areas of psychological interest; namely, reactions to disability, parental acceptance, social sensitivity and general adjustment. He reported that in his sample, mechanisms such as denial, compensation and feelings of inferiority were frequently present and of great importance. Further, the juvenile amputees in this group were highly sensitive to social appraisal of themselves or of their injury, had an undesirably high proportion of inadequate adjustments, and had low parental acceptance.

Centers (3) compared 26 juvenile upper-extremity amputees of both sexes with a control group of 26 normal persons on the basis of self-portraits. She reported that the two groups differed significantly, with gross deformation of arms or absence of arms or absence of limbs being the most common differentiating factor. However, she found that in the drawings the two groups differed little in expressions of conflict, anxiety or defensiveness.

Dr. Sidney Fishman included psychosocial variables in studies conducted under his direction by the Research Division, College of Engineering, New York University. Using a nation-wide sample of 159 upper-extremity juvenile amputees, studied between 1953 and 1956, Dr. Fishman (4) concluded that three major types of factors influenced (both positively and negatively) the outcomes of training in the use of a prosthesis; viz., 1) the personality o£ the child, 2) parental influences, and 3) the situational influences during the fitting and post-treatment experiences of the child.

New Study

Since physical and mechanical considerations alone did not accurately predict success in training juvenile amputees in past studies, our Psychology Department sought to determine what other variables are operative. Recently, with the assistance of Miss Jean Holland Van Ormer and Mr. E. Paul Dick, graduate students from the School of Education, Department of Special Education and Rehabilitation, University of Pittsburgh, we attempted to study -- within the context of a partial sample - certain psychological, social and other variables; viz., sex, age at time of training, etiology (whether congenital or traumatic), length of training period, intelligence, body-image distortion, parental attitude, and economic status of the family in relation to success achieved in the prosthetic training program.

Since April 1956, the Home for Crippled Children has provided juvenile amputees with prostheses and with the training to utilize these prostheses adequately. In the period between April 1956 and June 30, 1964, 131 juvenile amputees have been treated. This population is described in Table I and Table II .

In Table II , an interesting statistic is the great preponderance of left arm, be-low-elbow incidence; viz., 34 of the total of 86 arm amputees. We have not found a medical or genetic explanation for this surprising occurrence and would appreciate comments from other clinics.

In the absence of a medical or a genetic explanation, we have concluded that it represents a bias of our sample; that is, children from Pennsylvania (primarily) who were referred to our agency.

A considerable number of these amputees had received treatment prior to the Psychology Department's preparation of adequate research plans. Hence, this report does not include data on the entire sample.

Rating Scale

A Rating Scale was devised by the senior author to measure the degree by which the individual arm or leg amputee succeeded in his prosthetic training program. The five-point scale was as follows:

  1. Failed completely to reach goal. No measurable success.

  2. Made slight progress toward goal. Achieved slight success.

  3. Moderately successful in achieving goal.

  4. Made good progress towards goal. Fell just short of complete success in training and maximum use of prosthesis.

  5. Completely successful. Complete achievement of goal.

The Physical Therapist who conducted the actual training was asked to rate each child at the conclusion of the training period, using the rating scale listed above.* This rating provided the degree of success or failure in training which was then related to the psychological, social and other factors enumerated earlier in our report. In the initial statistical treatment of the data, points 1, 2 and 3 of the rating scale were combined in one category (labelled failure) , and points 4 and 5 were combined in a second category (success). The chi square statistic (7) usually expressed as Equation 1 was applied. Corrections for continuity were made when required.

In general, the 5% level of confidence was employed to determine the significance of differences. By this is meant that only in five times out of a hundred could a difference of the magnitude achieved be attributable to chance alone, assuming adequate sampling. In order to conserve space, individual chi square tables, showing observed and expected frequencies for all factors, are omitted in this report. (They are available on request). In the next sections of our report, the methods used to measure the variables and the results obtained from analysis of the data relevant to leg amputees and arm amputees will be reported separately.

Methods Utilized

A brief description of the methods by which the various factors studied were measured is of interest. Information as to sex of the child, age at time of training, and etiology of amputation was obtained from medical and social records and with the assistance of the Program Coordinators of the Home for Crippled Children.**

Age at time of training was divided into eight categories of two years each, beginning with one-three years; length of training period was measured in days from the date prosthesis was received until the date the patient was discharged. The number of days of training were trichotomized; viz., lower third, middle third and upper third. The Wechsler Intelligence Scale for Children (hereafter abbreviated WISC) was used to measure intelligence for the majority of the sample. It yields both Performance and Verbal I.Q.'s in addition to a Full Scale I. Q. I. Q. scores were divided into Average (ratios from 90 through 109), Below Average (below 90) and Above Average (above 109) for purposes of analysis.

A Human Figure Drawing was utilized to determine body-image distortion. The drawings were rated by the senior author of this study, on a three-point scale; viz.,

  1. Absence or malformation of important body parts. Distorted perception of body-image.

  2. Body-image considered adequate -- no serious distortions.

  3. Lack of malformations or absence of important body parts; patient's perception of body-image considered to be above average.

The Coordinator of the program of each child in the sample was asked to rate the attitude of the parents (towards the child and his handicap) as poor, average or good. The criteria applied were:

  1. Attitude in terms of parents' acceptance of the amputee handicap of a child.

  2. Attitude in terms of parents' confidence in the child's future ability to function adequately.

The occupation of the father was employed as a measure of the economic status of each child's family, this information being obtained from admission records. In only a few instances was the father's occupational level higher than that of skilled worker; consequently, the group was dichotomized. The fathers who were unemployed or were unskilled laborers were placed in the first group. Those who were engaged in semi-skilled, skilled, clerical and professional occupations were assigned to the second group.

Factors Studied Relevant to Success in Leg Amputee Program

The lower extremity sample of 41 patients included 28 males and 13 females. Ages of the patients ranged from one to 18 years. Fourteen were congenital amputees, and 27 had amputations of a traumatic etiology. Twenty-eight were fitted with above-knee, and 13 with below-knee prostheses.

1. We found that the relationship between sex of patient and degree of success in training was not statistically significant. However, 65.2% of the boys were trained successfully but only 53.8% of the girls. Fishman (9) has pointed out that cos-mesis is an important need for the amputee. The need for conformity in appearance and dress is stronger in girls than in boys, who are more likely to value physical prowess and performance. Since a dress cannot cover up an artificial leg, although a pair of trousers can, the training of girls might be expected to show less successful outcome because of ambivalent feelings. In our sample, the results indicated that the boys were successful more frequently than the girls.

2. The relationship between age at time of training and success produced interesting, although not statistically significant results. Only 44.4% of the patients under five years of age and only 53.8% of those above eleven years were successful; whereas, 76.5% of those between the ages of five and eleven were successful. The relatively low level of success (as measured in this study) of those less than five years

of age is probably due to phsiological, intellectual and emotional immaturity. However, this conclusion would require further verification. The emotional turmoil accompanying adolescence is well documented; and the relatively high percentage of training failures in those children above 11 years of age is likely to be of an emotional-psychological causation. In this sample, the age of greatest success (five to eleven years) is roughly analogous to the latent period of psycho-sexual development according to Freud. Blank (1) has stated that on the basis of the Freudian theory, it could be expected that blindness occuring at age nine would be less traumatic to the ego than at age five or 13. On the other hand, according to Dr. Wright (13), "research has simply not shown with any consistency that adjustment to disability is easier or more difficult at one certain age of occurrence than at others". This factor also merits further study.

3. While the relationship between etiology (congenital versus traumatic) and degree of success in training was not statistically significant, important differences should be reported. 70.3% of the traumatic amputees were successful, as contrasted with 42.8% of the congenital amputees. The general question as to whether a congenital or a traumatic amputation is more conducive to successful training in the use of a prosthesis has been argued both ways, but a definite conclusion has never been reached. Dr. Wright (13) has suggested that a patient with a congenital disability does not have to cope with an alteration in self-concept. Conversely, the traumatically disabled child does not have to contend with the stigma of hereditary defect which congenital deficiency connotes in the minds of many people. In the matter of the important variable of motivation (difficult to measure!), it appears that the traumatic amputee has the edge. The difference between the traumatic and congenital groups reported above is undoubtedly magnified in this study, since many of the congenital amputees were less than five years of age at the time of training, the period shown by our data to be least conducive to success.

4. The relationship between the total amount of training and success was found to be just short of statistical significance. Of the third of the sample whose training period was the shortest, 84.57» were successful, while only 38.5% of the third who spent the longest time in training attained this status. Evidently the Law of Diminishing Returns operates in the training situation.

5. Intelligence, or intellectual functioning level, as measured by the WISC, was found to have a significant relationship to success in training. The full scale WISC I.Q.'s of the patients were related to success at better than the 5% level of confidence. Performance Scale I.Q.'s alone approached this degree of significance; but the relationship of Verbal Scale I.Q.'s to success was not statistically significant. Since successful mastery of the operation of a prosthesis requires the ability to learn, these results are not really surprising. This learning is not primarily of a verbal type and it could have been expected that Performance Scale I.Q.'s (which depend more on the ability to attack a new problem) would have a significantly higher relationship to success than WISC Verbal Scale I. Q.'s.

The significance of the relationship between intellectual functioning (as measured by the WISC) and success has direct implications for the training program. Care should be taken that each child understands the instructions given to him; and that methods are as conducive as possible to learning (i.e., spaced rather than massed practice periods, immediate application of reinforcement, etc.). The significant relationship verified by this study makes it almost mandatory that researchers should study the application of learning theory to the training of juvenile amputees.

6. Fishman (8) has reported that the adjustment of an individual amputee to his prosthesis depends to a considerable degree upon self-concept. Bender and Silver (2) have noted the importance of body-image perceptions and distortions in relation to disability. Machover (10), studying the human figure drawings of 20 orthopedically handicapped persons, has reported that important features of the subjects' reactions to their defect were made graphically explicit in most instances. Cruickshank (6) has written that until the child has a coordinated and coherent understanding of the body-image, learning of a socially acceptable nature will not take place, or will be retarded. In this sample, the relationship between successful training and distortion of body-image as determined by a human figure drawing was not statistically significant. Of greater significance, however, was the fact that 21 of the 29 subjects for whom human figure drawings were available reflected distorted perception of their body-image through absence or malformation of important body parts. This finding supports the contention that amputation is a psychologically traumatic experience. A more sophisticated analysis of the current data to investigate this factor may be desirable.

7. Parental attitude in relationship to success in training yielded a statistically insignificant chi square value. This measurement was, admittedly, the weakest in the current study, in that the ratings of parental attitudes were frequently ex post facto; i.e., at varying periods after the discharge of the patient. In the future, these attitudes will be rated at the time of training. Professional persons in the rehabilitation field have stated that the child's attitude toward, and his acceptance of, himself is greatly influenced by parental attitudes.

8. The father's occupation was not related to training success in a statistically significant manner, but a definite trend was noted in the current data. 68.9% of those whose fathers were engaged in semi-skilled, skilled, clerical or professional occupations (29 in our sample) were successfully trained; whereas only 41.6% of the 12 children whose fathers were either unemployed or were employed as unskilled laborers were successful.

Factors Studied Relevant to Success in Arm Amputee Program

The population in this aspect of the study consisted of 27 juvenile upper-extremity amputees, of whom 12 were males and 15 were females. Seven had above-elbow and 20 below-elbow amputations. Twenty-one were congenital amputees, while the amputations of the other six were traumatic in origin. The age range was 1-0 years to 15-10 years, the median age being 5-4 years. The congenital amputees were considerably younger than those whose amputations were the result of trauma. The median age of the former group was 3-9 years ; and of the latter, 11-6 years. It is unfortunate that we did not set up our research instruments earlier in the history of the program. Since we have data (ratings of success in training, etc.) on only 27 out of a total of 86 upper extremity amputees served; i.e., 32%, the sample cannot be considered entirely representative. The psychological and social factors investigated for the Leg Amputee Program were studied in exactly the same manner for the arm amputees treated.

1. Although no statistically significant relationship between sex and success of training was found, it should be noted that 83.3% of the boys were deemed to have been successfully trained as contrasted with 46.6% of the girls. This result reinforced the trend noted with leg amputees. Steensma (12) has contended that teenage girls were more concerned than boys with cosmesis, and, hence, had more difficulties in adjusting to a prosthesis during the training period. One would expect, therefore, that the level of motivation of girls would be raised if arm prostheses had better cosmesis.

Fishman (5), in a Child Prosthetic Studies report issued in 1964, which dealt with the acceptability of a functional-cosmetic artificial hand for younger children, concluded that "girls of all ages for whom the hand is of appropriate size appear to be potentially the best candidates for the hand". It should be noted, however, that this "child artificial hand" was not designed for adolescents.

2. In our data no significant relationship between etiology (congenital versus traumatic) and success in training was found. However, a significant trend was again apparent; namely, that the training of all six of the traumatic amputees was successful, whereas only 52.4% of the congenital amputee population was successfully trained. However, the fact that the traumatic amputees were in an older age range than were those whose amputations were congenital in origin (median age of 11-6 years as compared with a median age of 3-9 years) may have been a contributing factor. It will be recalled that leg amputees in the middle years were more successfully trained than those younger than five years and older than eleven years.

Meng (quoted in 6) has suggested another possibility to account for the lower level of success in training congenital amputees. He contends that the congenitally amputated child experiences a lack of contact with reality through normal play. As a result, he often develops excessive narcissistic tendencies and views himself as different or even unique because of his disability. Meng also believes that the congenital amputee often has ambivalent feelings towards his parents, whom he regards as both the source of, and the solace for, his handicap. Unable to resolve this conflict, he frequently resorts to compensatory mechanisms of a hypochondriacal nature. The close relationship between the body structure and his defect increases ego libido. As the child comes to feel more rejected by his parents and society, this relationship increases. Thus, he may view the introduction of a prosthesis as a threat to this ego defense.

3. The relationship between age (divided into eight two-year groups) and success in training yielded a statistically significant chi square value. However, the data does not reveal any particular trend, and was possibly contaminated by other factors. A more sophisticated type of statistical treatment is required.

4. Although the data failed to reach the statistical significance required for the 5% level of confidence, an important trend was again noted regarding length of training. Arm amputees requiring less than the median time of fourteen days were compared with those requiring more than fourteen days. The trend was again in favor of the shorter training period. 78.6% of the children in the short-term group were rated as having been successfully trained; whereas only 46.2% of those exposed to the longer period of training were successful.

5. There were 19 Binet I.Q.'s and 9 WISC I.Q.'s (with one over-lapping) in this sample of 27 arm amputees. The relationship between the WISC Verbal Scale I.Q.'s and success in training was statistically significant and the relationship between WISC Full Scale I.Q.'s and WISC Performance Scale I. Q.'s approached significance. The relationship between Binet I.Q.'s and success was not found to be statistically significant. This result is difficult to interpret, except for the fact that Binets are commonly given to younger-aged children who, in our data, were found to be less successful in the training program.

6. Distortion of body-image was not found to be significantly related to success in training in this sample. It should be noted, in this context, that Centers (3) interpreted body distortions in amputee drawings as a realistic and non-defensive representation of the self and viewed the omission of the amputee's appropriate body anomalies from his drawing as either denial or the wish for an intact body. It is obvious that this factor requires more intensive research before definitive conclusions can be reached.

7. The relationship between parental attitudes (as determined by a rating scale) and success in training did not yield statistically significant results. We have already noted that our measuring device in this area was inadequate. This instrument will be refined for future research.

8. Finally, no significant relationship between fathers' occupation and success in training was obtained. Measures of socio-economic levels are difficult to establish and it is obvious that the use of the father's occupation was not an adequate measure in our small sample.

Nine of the 41 leg amputees and 10 of the 27 arm amputees had previously worn prostheses. In order to determine if this would introduce a contaminating factor, the relationship between previous prosthetic history and success in training was investigated by the use of the chi square statistic. From a statistical standpoint, this relationship was found to be completely insignificant.


Based on the results reported above, it is interesting to postulate the characteristics of a child for whom the possibility of successful training in the use of a prosthesis would be maximal. The subject would be a boy, between the ages of five and eleven, with a traumatically sustained amputation. He would be average or above in intellectual functioning, show a somewhat greater score on the Performance Scale of the WISC than on the Verbal Scale, and his father would be employed in a semiskilled, skilled, clerical or professional occupation. However, since this "ideal" patient is primarily a statistical concept (even though such children do exist), no contraindication to the treatment of any child is implied.

Training in the use of a prosthesis should embody the most adequate learning techniques at our disposal with cognizance taken of such factors as interest span, spacing of training and practice periods, reinforcement, and the law of diminishing returns.

The psychological factors implicit in consideration of etiology, self-image and parental acceptance posit the requirement of sustained psychological counseling with child and parent.

Analysis of our data indicated the need for more intensive research into the complex relationships between the factors under study as well as for more sophisticated methods of statistical analysis.


Selected psychological and social factors related to degree of success in prosthetic training were studied in a population of 41 leg amputees and 27 arm amputees. These factors were related to a dichotomy of failure versus success in prosthetic training, utilizing a rating scale completed by the physical therapist doing the training. The chi square statistic (X2) was used as the basic analytic tool in this initial study. The most significant relationship was found to be between intelligence and success. Although not reaching statistical significance, results indicated important relationships between age at beginning of training, length of training, and self-image as measured by human figure drawings. Comments have been made concerning the implications of these findings. It is our intention to refine the statistical treatment of our data and to initiate further study of these psychological and social factors. Comments from other Child Amputee Clinics and interested readers of the Inter-Clinic Information Bulletin regarding this initial report will be heartily welcomed.

* The assistance of Miss Virginia Whitfield, Chief, Physical Therapy Department, Home for Crippled Children; Miss Ruth Ann Keen, Senior Staff Therapist, and Miss Betty Mutschler, Senior Staff Therapist, is gratefully acknowledged. Miss Whitfield rated the bulk of the leg amputees, Miss Keen the arm amputees, and Miss Mutschler assisted Miss Whitfield with several of the leg amputees.

** The assistance of Miss Belle Grierson, who acted as Coordinator for the bulk of the children in the sample, Miss Martha Lang, Mrs. Emily Lisi and Mr. Frank Glackin -- Program Coordinators --is gratefully acknowledged.


  1. Blank, H., "Psychoanalysis and Blindness", Psychoanalytic Quarterly. 1957, 26, pp. 1-24.

  2. Bender, L. and Silver, A., "Body-Image Problems of the Brain Injured Child", Journal of Social Issues. 1948, 4, pp. 84-89.

  3. Centers, Louise, "A Comparison of the Body-Image of Amputees and Non-Amputee Children as Revealed in Figure Drawings", Journal of Projective Techniques, 1963, 27, pp. 258-265.

  4. Child Prosthetic Studies, Research Division, College of Engineering, New York University, "The Clinical Treatment of Juvenile Amputees, 1953-1956", 1958.

  5. Child Prosthetic Studies, Research Division, College of Engineering, New York University, "Acceptability of a Functional-Cosmetic Artificial Hand for Young Children", 1964.

  6. Cruickshank, William, Psychology of Exceptional Children and Youth. Engle-wood Cliffs, New Jersey; Prentice-Hall, Inc., 1963.

  7. Edwards, Allen L., Statistical Analysis for Students in Psychology and Education, Rinehart & Co., Inc., New York, 1946, pp. 239-257.

  8. Fishman, Sidney, "Self-concept and Adjustment to Leg Prostheses". Unpublished Doctoral Dissertation, Columbia University, 1949.

  9. Fishman, Sidney, "Amputee Needs, Frustrations and Behavior", Rehabilitation Literature 1959, 20, pp. 322-328.

  10. Machover, K., Personality Projection in the Drawing of the Human Figure. Charles C. Thomas, Springfield, Illinois; 1949.

  11. Siller, J., "Psychological Concomitants of Amputation", Child Development. 1960, 31, pp. 109-120.

  12. Steensma, J., "Problems of the Adolescent Amputee", Journal of Rehabilitation, March-April, 1959, 25, pp. 19-22.

  13. Wright, B., Physical Disability -- A Psychological Approach. Harper & Row, New York; 1960.

Everett Campbell is Chief Psychologist and Judith Bansavage is Staff Psychologist at Home for Crippled Children, Pittsburgh, Pennsylvania

J. H. Van Ormer and E. P. Dick are Internes in Psychology, University of Pennsylvania