Can a supervised algorithmic assessment of men for prostate cancer improve the quality of care? A retrospective evaluation of a prostate assessment pathway in Saskatchewan
Bibliographic record
Abstract
INTRODUCTION: The Saskatoon Prostate Assessment Pathway (SPAP) was developed in 2013 in part to decrease the wait times between physician referral and biopsy for patients with suspected prostate cancer. Using an algorithm carefully designed to optimize appropriate prostate biopsy rates, physicians can directly refer patients for biopsy through the SPAP without seeing a urologist. All other patients are referred to the Saskatoon Urology Associates (SUA). The present study evaluates the performance of the algorithm. METHODS: 971 patients seen at the SUA and 302 patients seen through the SPAP were identified. Information on age, biopsy status and outcome, risk stratification, and time between referral and biopsy was collected. Biopsy wait time data was analyzed using gamma distribution. Association between referral method and biopsy rate, and between referral method and risk stratification, was analyzed using Z-test. RESULTS: The expected wait time from referral to biopsy for patients seen through SUA was 2.63 times longer than those seen through SPAP (34 vs. 91 days). The biopsy rate of patients seen in the SPAP was significantly higher than those by SUA (88% vs. 69%, 95% confidence interval [CI] 0.14-0.26; p<0.00001). There was no significant difference in positive biopsy rates for patients seen through the SPAP vs. SUA (81% vs. 74%, 95% CI -0.011,0.14; p=0.095), for detection of low-risk cancer, (12% vs. 10%, 95% CI -0.034,0.080; p=0.44), or for clinically relevant cancer, i.e., intermediate- and high-risk cancer, for SPAP vs. SUA (56.54% vs. 56.68%, 95% CI -0.091,0.089; p=0.49). CONCLUSIONS: The algorithm used in the SPAP is effective in decreasing wait time to prostate biopsy and has the same cancer/pre-cancer detection rate, but at the cost of a higher biopsy rate. Both referral mechanisms result in few low-risk cancer detection biopsies, finding primarily cases of high- or intermediate-risk cancer.
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 | 0.003 | 0.009 |
| Meta-epidemiology (narrow) | 0.000 | 0.001 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.004 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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".