P22. Prognostic Predictive Model for the Development of Osteoarthritis using Electronic Medical Record Data
Bibliographic record
Abstract
Background: As the most common joint disorder worldwide (1), osteoarthritis represents a growing concern for older adults. Prognostic predictive models (statistical models used to predict future disease development (2)) may enable the identification of patients at high risk of developing osteoarthritis, allowing for health and lifestyle modifications aimed at reducing the risk of disease development (3,4).\nMethods: For our project, we accessed the DELPHI (Deliver Primary Healthcare Information) database which contains de-identified electronic medical records of more than 60,000 primary care patients in Ontario (5,6). From these data, we constructed a retrospective cohort examining patients’ risk factors and followed them over time to observe incident cases of osteoarthritis. This retrospective cohort was used to develop and test prognostic predictive models, using methods such as logistic regression, to determine the models’ ability to predict development of osteoarthritis. Models were evaluated, examining both discrimination (AUC) and calibration (calibration plots), using a reserved portion of patient data.\nResults: A logistic regression model was built that predicts the incidence of osteoarthritis based on patient age, sex, Body Mass Index (BMI), osteoporosis status, and leg injury status (AUC: 0.73).\nDiscussion & Conclusion: By creating a prognostic predictive model for osteoarthritis, we aim to support primary health care practitioners in estimating an individual patient’s risk of osteoarthritis; thereby allowing practitioners and patients to create unique plans to address the patient’s personal risk factors.\nInterdisciplinary Reflection: This project is highly interdisciplinary as it spans the fields of epidemiology, statistics, health informatics, primary health care, and computer science.\nReferences:\n1. Lopez AD, Mathers CD, Ezzati M, Jamison DT, Murray CJL. Global and regional burden of disease and risk factors, 2001: systematic analysis of population health data. Lancet (London, England) [Internet]. 2006 May 27 [cited 2016 Feb 13];367(9524):1747–57. Available from: http://www.ncbi.nlm.nih.gov/pubmed/16731270\n2. Hendriksen JMT, Geersing GJ, Moons KGM, de Groot JAH. Diagnostic and prognostic prediction models. J Thromb Haemost [Internet]. 2013 Jun [cited 2016 Aug 10];11 Suppl 1:129–41. Available from: http://www.ncbi.nlm.nih.gov/pubmed/23809117\n3. Felson DT, Zhang Y, Anthony JM, Naimark A, Anderson JJ. Weight loss reduces the risk for symptomatic knee osteoarthritis in women. The Framingham Study. Ann Intern Med [Internet]. 1992 Apr 1 [cited 2016 Jun 23];116(7):535–9. Available from: http://www.ncbi.nlm.nih.gov/pubmed/1543306\n4. Felson DT. Weight and osteoarthritis. Am J Clin Nutr [Internet]. 1996 Mar [cited 2016 Jun 23];63(3 Suppl):430S–432S. Available from: http://www.ncbi.nlm.nih.gov/pubmed/8615335\n5. CPCSSN. DELPHI (Deliver Primary Healthcare Information) Project [Internet]. 2013. Available from: http://cpcssn.ca/regional-networks/delphi-deliver-primary-healthcare-information-project/\n6. Birtwhistle R, Keshavjee K, Lambert-Lanning A, Godwin M, Greiver M, Manca D, et al. Building a pan-Canadian primary care sentinel surveillance network: initial development and moving forward. J Am Board Fam Med [Internet]. 2009 Jan [cited 2016 May 19];22(4):412–22. Available from: http://www.ncbi.nlm.nih.gov/pubmed/19587256
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 | 0.009 | 0.033 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.008 | 0.002 |
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".