FACTOR ANALYSIS OF UPPER LIMB PROSTHETIC ACCEPTANCE FROM RETROSPECTIVE PROSTHETIC CLINICIAN SURVEY

Upper limb prosthetic acceptance seems to be relatively unchanged from 1958 where it was measured to be 75% for transradial, 61% for transhumeral, and 35% for shoulder disarticulation levels. A practitioner survey from 2013 by the author found this to be largely unchanged at 79.6%, 57.8%, and 32.8% respectively. An upper limb meta-analysis showed that the most significant factors affecting prosthetic rejection using a median rating were function, comfort, ease of use, weight, heat, lack of sensory feedback, inconvenience, lifestyle, dissatisfaction with technology, irritation, and availability of services. An earlier survey by the author condensed these factors of rejection to amputation level, functional advantage, and comfort, and included confidence of the prosthetist, availability of therapy, and support of the patient context. Also it was speculated that the value of factors influencing rejection of prostheses may not be simply the converse of those accepting the prosthesis but different scales.


Introduction
Upper limb prosthetic acceptance seems to be relatively unchanged from 1958 where it was measured to be 75% for transradial, 61% for transhumeral, and 35% for shoulder disarticulation levels.A practitioner survey from 2013 by the author found this to be largely unchanged at 79.6%, 57.8%, and 32.8% respectively.An upper limb meta-analysis showed that the most significant factors affecting prosthetic rejection using a median rating were function, comfort, ease of use, weight, heat, lack of sensory feedback, inconvenience, lifestyle, dissatisfaction with technology, irritation, and availability of services.An earlier survey by the author condensed these factors of rejection to amputation level, functional advantage, and comfort, and included confidence of the prosthetist, availability of therapy, and support of the patient context.It was speculated that the factors influencing rejection of prostheses may not be the converse of those accepting the prosthesis but different scales.

Methods
A retrospective case survey was created using a third-party web-based survey that was posted from 3/15/17 to 4/1/17.Prosthetists were asked to recall up to their last three upper limb fittings.
There were 75 respondents with 209 retrospective cases.Based on the number of UL fitting, participants were classified: 12 specialists, 20 experts, 27 intermediates, and 16 novices.
The participants were asked to assess various retrospective factors including prosthetic level, type of control, patient gadget tolerance, patient functional expectation, comfort tolerance, patient value of cosmesis, confidence of prosthetist, availability of therapy, experience level of therapist, daily wear time, patient description of prosthesis, and patient assessment of cosmetic quality.By level the distribution was 24 finger/transcarpal, 17 WD, 110 TRl, 4 ED, 47 TH, 6 SD, and 1 IT reflecting a common distribution.Body power control predominated with 84 and external power as 72 with passive at 26.

Results
The clinician self-assessments predominantly showed highly favorable, and perhaps skewed, outcomes with respect to functional expectation, gadget tolerance, prosthetic confidence, contextual support, patient wear time, patient proficiency, cosmetic acceptance, and description of the prosthesis.Areas of more normalized distribution appeared to be upper limb experience for therapists, patient experience, and prosthetic description.Using Pearson's coefficient, there were statistically significant relationships at p (one-tailed) <.01 level between the patient acceptance level and patient experience, gadget tolerance, functional expectation, comfort tolerance, prosthetist confidence, and patient cosmetic description and a negative p<.05 relationship with experience of the therapist.The relationships were true of the number of hours worn.A very strongly predictive relationship of R2 Linear = .564F(10,197) = 25.513 with all of the factors listed above.However a systematic multi-variable reduction found a strong significance, R2Linear =.556, F(3,204) = 85.302, with the patient acceptance level, gadget tolerance, patient experience, which may not have practical clinical relevance.

Conclusion
The skewed distribution of prosthetist self-assessment for functional expectation, gadget tolerance, comfort, prosthetic confidence, contextual support, and patient proficiency, wear time, and prosthetic description th this retrospective may be indicative of a group halo or optimism bias.Also this calls into question the validity and reliability of the sample and relevance to the application.However, the negative relationship between the expertise of therapist and proficiency of the patient may indicate that the therapist may mitigate this bias and influence a clearer understanding of functional outcome.The strong statistically predictive