Analysis of insole footprints can support forensic human identification when footwear is linked to crime scene shoeprints. Various methods are used for analysis, though few have undergone validation to confirm reliability for criminal justice systems. The aim of this research was to investigate validity of one of these methods known as the Overlay method. This included exploring potential for method development. A literature search revealed a lack of research underpinning its foundational validity for forensic podiatry practice. A case study approach was therefore utilised involving three forensic podiatrists applying the method to compare a questioned insole footprint against a sample of reference insole footprints. Validity was analysed using Sensitivity, Specificity, False Positive and Negative Rates, Receiver Operating Characteristic and Cohen’s Weighted Kappa for rater agreement. Findings showed not all raters agreed indicating this method should not be used in isolation in forensic practice. However, due to the small sample size of raters and insoles, results were regarded as preliminary. Exploration of method development to reduce reliance on subjective judgements revealed limitations as an objective tool for practice. Thus, a novel approach was developed and named the Howsam-Frame (HF) method for measurement of inter toeprint distances on insoles. An exploratory pre-study was first initiated to investigate intra and inter insole measurement variation in a sample of pairs of insoles from one person. Although this limited external validity, analysis generated preliminary baseline data of the magnitude of variation in a person. This provided insight into the method’s potential to discriminate. Descriptive analysis showed variation was greatest in measurements containing T5 toeprint. Smallest variations were seen in measurements containing T1 toeprint. Hypothesis testing investigating differences between left versus right insoles showed significant differences in six out of ten measurements with T1 to T4 measurement showing no significant difference. This was selected in combination with a measurement of high variation (T2 to T5) to compare measurements in a pilot study for investigating feasibility for reliability testing of the HF method. This pilot utilised an intra-rater reliability approach including repeated measures to explore feasibility. Although this study was limited to a small sample and one novice participant, reliability analysis using Intraclass Correlation Coefficient, Standard Error of Measurement and Bland and Altman 95% Limits of Agreement plots showed the HF method demonstrated higher reliability compared to a rudimentary Manual method and that T1 to T4 measurement demonstrated greater repeatability compared to T2 to T5 measurement. However, due to the size of this study, results were considered indicative and not confirmatory of reliability. A similar research design was then applied for main reliability testing where three forensic podiatrists collected repeat measurements of T1 to T4 from 32 insole footprints. Although insoles were not associated with actual crime scenes, intra-rater and inter-rater reliability analyses indicated high levels of reliability. Findings provide empirical evidence supporting the HF method as a reliable and valid method for forensic podiatry practice but with some limitations including issues influencing precision of measurements relative to diminished quality and coalescence of toeprint outlines.
| Date of Award | 30 Oct 2025 |
|---|
| Original language | English |
|---|
| Supervisor | Jenny Killey (Main Supervisor) & Sarah Reel (Co-Supervisor) |
|---|