Developing a foot ulcer risk model: what is needed to do this in a real‐world primary care setting?
Bibliographic record
Abstract
AIM: To determine how routinely collected data can inform a risk model to predict de novo foot ulcer presentation in the primary care setting. METHODS: Data were available on 15 727 individuals without foot ulcers and 1125 individuals with new foot ulcers over a 12-year follow-up in UK primary care. We examined known risk factors and added putative risk factors in our logistic model. RESULTS: mmol/mol (63 ± 21 vs 59 ± 19) (p<0.0001) concentration [+0.45 (95% CI 0.33-0.56), creatinine level [+6.9 μmol/L (95% CI 4.1-9.8)] and Townsend score [+0.055 (95% CI 0.033-0.077)]. Absence of monofilament sensation was more common in people with foot ulcers (28% vs 21%; P<0.0001), as was absence of foot pulses (6.4% vs 4.8%; P=0.017). There was no difference between people with or without foot ulcers in smoking status, gender, history of stroke or foot deformity, although foot deformity was extremely rare (0.4% in people with foot ulcers, 0.6% in people without foot ulcers). Combining risk factors in a single logistic regression model gave modest predictive power, with an area under the receiver-operating characteristic curve of 0.65 (95% CI 0.62-0.67). The prevalence of ulceration in the bottom decile of risk was 1.8% and in the top decile it was 13.4% (compared with an overall prevalence of 6.5%); thus, the presence of all six risk factors gave a relative risk of 7.4 for development of a foot ulcer over 12 years. CONCLUSION: We have made some progress towards defining a variable set that can be used to create a foot ulcer prediction model. More accurate determination of foot deformity/pedal circulation in primary care may improve the predictive value of such a future risk model, as will identification of additional risk variables.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.039 | 0.147 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.005 | 0.002 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.006 | 0.004 |
| Open science | 0.004 | 0.002 |
| Research integrity | 0.003 | 0.004 |
| Insufficient payload (model declined to judge) | 0.004 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".