Lessons learned: Linking patient-reported outcomes data with administrative databases
Bibliographic record
Abstract
IntroductionSince 2007, Cancer Care Ontario (CCO) has systematically collected patient-reported outcomes (PROs) in the form of symptom data, for cancer outpatients visiting regional cancer centres or affiliate institutions. Data are used in real-time to facilitate conversation between clinicians and patients and have recently been combined with provincial administrative databases. Objectives and ApproachCCO collects PROs using the Edmonton Symptom Assessment System (ESAS), which scores 9 symptoms on a scale of 0 (no symptoms) to 10 (worst symptom severity). Data were imported from CCO in 2015 and linked to a cancer cohort at ICES. We investigated differences between patients who completed $\geq$1 ESAS record and patients who did not, as well as the number of records, timing of data collection and missingness. We describe our experience linking and using the PRO data to administrative data, including presenting trajectories of symptoms over time and combining scores into composite indices. Results120,745 cancer patients had 729,861 symptom records between 2007 and 2014. Not all patients with a cancer diagnosis had $\geq$1 ESAS record and this varied by patient, disease and system level factors. Because implementation occurred from a clinical perspective, data collection was irregular within and across patients and depended on treatment and other factors; the number of records per patient varied, as well the number of contributing patients in each time period following diagnosis. Attempts were made to create meaningful composite indices by combining all symptom scores as well as combining multiple high scores for each individual symptom. As a result, selecting the best statistical analysis to use these PRO data as an exposure or outcome is still uncertain. Conclusion/ImplicationsPRO data linked to provincial, administrative data holdings represent a new frontier for population-based cancer research, both in their challenging structure as well as their implications for clinical practice and health system. These lessons learned will hopefully support other researchers rigorous use of these data in the future.
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 | 0.210 | 0.557 |
| Meta-epidemiology (narrow) | 0.003 | 0.002 |
| Meta-epidemiology (broad) | 0.002 | 0.003 |
| Bibliometrics | 0.011 | 0.020 |
| Science and technology studies | 0.002 | 0.004 |
| Scholarly communication | 0.018 | 0.022 |
| Open science | 0.010 | 0.010 |
| Research integrity | 0.005 | 0.009 |
| Insufficient payload (model declined to judge) | 0.005 | 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; the direct Gemma label and the distilled Codex classifier agree on what is shown here.
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".