Forty-five participants, including 25 children (14 girls) with a mean age of 13.7 years (9.0 to 17.9 years) and 20 parents (16 mothers), were enrolled in Phase 1. Table 1 provides a summary of the participants’ age (children) and gender (parents and children), categorized by the questionnaire version they were administered. Sixteen children had a congenital LD (e.g., fibular hemimelia), 5 acquired (e.g., post-traumatic growth arrest), and 4 developmental (e.g., genu varum). Total leg length discrepancies ranged from zero to 100 mm, and deformity [35] ranged from none to greater than 15 degrees of malalignment. Additional file 2 further details children’s diagnostic characteristics. Of the 25 children enrolled, one opted to review but not complete the GOAL questionnaire and agreed to participate in the cognitive interview. In total, 13 children participated in the cognitive interviews (8 for the GOAL and 5 for the GOAL-LDdraft1). Thirty-one HCPs (81% orthopedic surgeons; 52% North American) completed the Phase 2 survey that reviewed the GOAL-LDdraft2. A demographic summary of HCP respondents is provided in Table 2.
Table 1 Phase 1 participants by questionnaire version administered
|
Children
|
Parents
|
Questionnaire Administered
|
GOAL Completion (n)
|
Mean age (SD)
Range
|
Cognitive Debriefa (n)
|
Mean age (SD)
Range
|
GOAL Completion (n)c
|
GOAL
|
12
(7 girls/5 boys)
|
14.8 years (2.3)
12-17 years
|
8
(5 girls/3 boys)
|
15.7 years (2.6)
13-17
|
11
(9 mothers/2 fathers)
|
GOAL-LDdrft1
|
5
(3 girls/2 boys)
|
11.8 years (2.6)
9-16 years
|
5
(2 girls/3 boys)
|
13.2 (2.7)
10-16
|
3
(2 mothers/1 father)
|
GOAL-LDdrft2
|
7
(4 girls/3 boys)
|
13.4 years (2.4)
9-17 years
|
n/a
|
n/a
|
6
(5 mothers/1 father)
|
Total
|
24b
(14 girls/10 boys)
|
13.7 years (2.6)
9-17 years
|
13
(7 girls/6 boys)
|
14.6 (2.5)
10-17
|
20
(16 mothers/4 fathers)
|
a cognitive debrief sample is a subsample of the children who consented to participate, b total study enrollment was 25 children. One child opted not to complete the questionnaire (GOAL-LDdraft1) but did participate in a cognitive debrief interview, c each parent completed a child-parent dyad
Table 2
Demographics of phase 2 HCP respondents (n = 31)
Demographics | n (out of 31) |
Healthcare Profession |
Pediatric Orthopedic surgeon | 25 |
Physical Therapist | 4 |
Other: Physical Therapy Practitioner, Nurse Practitioner | 2 |
Experience working with children with LDs (years) |
< 5 | 6 |
5 to < 10 | 9 |
10-<15 | 5 |
> 15 | 11 |
Country |
Canada | 5 |
United States | 11 |
United Kingdom | 10 |
Europe | 3 |
Unknown | 2 |
Abbreviations: HCP, health care professional; LD, lower limb difference |
GOAL Content Adaptation
Item Addition
In total, seven new items were added, six derived from children’s and parents’ suggestions. These were retained in the final iteration based on the items’ subsequent scoring performance and the HCPs’ ratings. An item about wearing a prosthesis was introduced based on HCPs’ recommendations and was included to increase the questionnaire’s generalizability.
Item Modification
Three items were modified to increase their level of difficulty (e.g., carrying an object while walking (e.g., toy, doll, book, cellphone) became carrying heavy objects while walking (e.g., grocery bags, several schoolbooks). Other examples of modifications through the iterations include: i) three items that were split to improve their specificity ii) two items that were combined to minimize redundancy, and iii) and 3 items that were moved to a different domain where the development team felt that they fit better conceptually. Six modifications were directly informed by HCP recommendations and involved wording to make items more explicit (e.g., walking on slippery or icy surfaces became walking on wet, slippery or icy surfaces) or to facilitate international utility (e.g., including metric and imperial measurements).
Item Elimination
In total, 12 original GOAL items were eliminated of which six were in the ADL category and removed due to poor item performance in field-testing. Walking for more than 15 minutes was eliminated because in the cognitive interviews, children could not distinguish this item from walking for more than 250 meters. The latter was retained because children were better able to conceptualize 250 meters (“from swimming”, “from track and field”) and it had lower item scores (i.e., considered more difficult). One item was eliminated based on HCP feedback related to perceived redundancy, and the development team agreed that moving quickly when in a hurry overlapped with running fast since both items were speed related. The latter was retained based on an overall lower item score and wider distribution (range) of scores within the GOAL-LDdraft2.
An overview of the content adaptations is provided with full item-by-item details by iteration, in Additional file 3.
Sensibility Evaluation
Comprehensibility
No participant asked for assistance or language clarification during GOAL or GOAL-LDdraft1 administration. During cognitive interviews, all children demonstrated that they could read and understand the meaning of words, phrases, and concepts in the questionnaire.
Clarity of instruction
Five participants indicated they had not read the instructions prior to completing the questionnaire. Further, it became clear that instructions with respect to item importance ratings were problematic. The intended purpose of these ratings was to capture how important a goal it was to improve on the item, allowing identification of items that contribute most to treatment related decision-making. During the cognitive interviews, many respondents had a more generalized interpretation of importance, considering how it related to daily living overall rather than the importance as a goal for improvement. Written instructions were modified in the GOAL-LDdraft2 to improve clarity.
Suitability of the response scale
In the original version of the questionnaire (GOAL), the recall period (in the past 4 weeks) was problematic for the domain of Physical Activities, Sport & Recreation, as some of these items were seasonal and many respondents checked the option “I did not do this in the past 4 weeks”. One child commented “I’ve been living with this a long time, I know how these activities affect me even if I haven’t done them in the past 4 weeks”. Thus, in the GOAL-LDdraft2, the instructions were adapted and the respondents were instructed to consider the past year for the activities in this domain. No further changes were suggested related to the response scale.
Ease of usage
Nine children (69%) regarded the questionnaire as easy to complete. Three (23%) felt the questionnaire was too long. Time required for children to complete the questionnaire versions varied from 12 to 19 minutes.
Overall, 23 of 30 HCP respondents (77%) stated they would consider using the questionnaire in their clinical practice and 27 (90%) would apply for research purposes. Twenty-five HCPs who identified strengths of the GOAL-LDdraft2 based on its content, with 13 (52%) commenting that its comprehensiveness was an asset. However, 15 of 31 HCPs (48%) responded that the amount of time required to administer the GOAL was “too long”.
The Resulting GOAL-LD
The final iteration, the GOAL-LD, reflects the cumulative results of Phases 1 and 2. It contains 45 items organized into six domains and both the child and parent version are available in Additional file 4 and 5 respectively. Seven new items were added, 12 were eliminated, and 26 of the original 50 items were retained, as detailed in Additional file 3.