Reproducibility of ACT, A simple-self-administered sCreening Tool for diabetic peripheral neuropathy

Diabetic Peripheral Neuropathy (DPN) is associated with pain and increased risk of foot ulceration and lower-limb amputation. Most people remain undiagnosed due to the lack of a simple, rapid, and reliable screening tool suitable for primary care. We evaluated the reproducibility of a short, self-administered screening tool (ACT) for DPN compared with the Michigan Neuropathy Screening Instrument Questionnaire (MNSI_q) and examination (MNSI_e).

In this observational study, 126 adults with type 1 or type 2 diabetes were recruited from an outpatient endocrinology clinic. Participants completed the ACT_{questionnaire (q) + examination (e)} which can be summed to create a total_{q + e} score of 0-20, and MNSI_{q + e} at baseline and after 14 days. Reproducibility was assessed using intraclass correlation coefficients (ICC) for continuous scores and Cohen's kappa (κ) for categorical reproducibility.

ACT_{q + e} demonstrated good reproducibility with an ICC = 0.85 (ICC = 0.78 for ACT_q; ICC = 0.88 for ACT_e) which was comparable to ICC = 0.82 for MNSI_q; ICC = 0.75 for MNSI_e; ICC = 0.85 for MNSI_{q + e}. Dichotomized classification of the overall ACT showed strong repeatability with κ = 0.83 (κ = 0.61 ACT_q and moderate-to-strong repeatability for ACT_e with κ = 0.78). Symptoms showed variable agreement, with the strongest reproducibility for burning and pins-and-needles sensations. The sensory examination components of ACT_e demonstrated strong agreement (κ = 0.81).

The overall ACT tool demonstrated good 14-day reproducibility compared to the overall MNSI_{q + e}. The brevity, self-administered format, and minimal reliance on clinician expertise, support the utility of ACT as a practical screening instrument for DPN in busy, resource-limited clinical settings.