2019 HSR&D/QUERI National Conference

4087 — Assessing Foot Self-Care Behavior: The Validity and Reliability of a Foot Self-Care Survey in Veterans with Diabetic Neuropathy.

Lead/Presenter: Bernardo Da Silva
All Authors: Da Silva B (VA New York Harbor Healthcare System), Nicholson A (VA New York Harbor Healthcare System), Natarajan S (VA New York Harbor Healthcare System, New York University School of Medicine)

Objectives:
Diabetic neuropathy, the most common complication of diabetes, puts patients at high risk for diabetic foot ulcer (DFU) and amputation. Proper foot self-care is considered crucial to prevent diabetes-related foot complications. This study reports on the validity and reliability of a modified survey to determine patient adherence to foot self-care behaviors.

Methods:
We used data from the PATRIOT trial testing a primary prevention strategy to prevent DFUs among at-risk Veterans (n = 387). All participants completed the 14-item VA Foot Self-Care Survey (validated in patients with prior DFU) during in-person visits. Internal consistency was determined by calculating Cronbach alpha. Convergent validity was assessed from the correlation between the instrument's footcare skills scores and patients' foot self-care proficiency scores from a separate free-recall demonstration in which participants demonstrated their foot-care. Discriminant validity was evaluated by comparing the correlations between patient foot self-care scores and patient height separately. Reliability was assessed by comparison with the control group's scores at the 6-month follow-up visit.

Results:
The 14-item survey demonstrated strong internal consistency (alpha = 0.66). Shortening the survey to include only the 6-item basic foot self-care behaviors further increased internal consistency (alpha = 0.74). Significant correlations between self-care demonstration and basic footcare skill scores (0.30, p = < .0001) and with extended footcare skill scores (0.21, p-value = < .0001) were observed. No correlation between height and basic foot-care scores (-0.071, p-value = .16) or between extended foot-care scores (-0.040, p-value = .43) were noted. Test-retest reliabilities for the 6-item basic (0.67, p-value = < .0001) and the 4-item extended (0.55, p-value = < .0001) foot self-care behaviors were strong.

Implications:
The 14-item foot self-care survey demonstrated satisfactory validity and internal consistency for assessing adherence to foot self-care behaviors among primary prevention diabetic veterans. The 6-item basic measure had better psychometric properties.

Impacts:
The 6-item measure is valid and reliable for efficiently assessing foot self-care in Veterans at risk of developing a DFU. Future implementation studies should incorporate the 6-item measure in usual VA care settings and target poorly adherent patients for tailored care to reduce the risk of incident DFU in such Veterans.