Measuring quality outcomes across hospital systems: Using a claims data model for risk adjustment of mortality rates
Bibliographic record
Abstract
Healthcare delivery systems around the world are designing care through value-based models where value is defined as a function of quality of care outcomes and cost. Mortality is a sentinel outcome measure of quality of care, of fundamental importance to patients and providers. Discovery Health (DH), an administrative funder of healthcare in South Africa (SA), uses service claims data of client medical schemes to examine standardised mortality rates (SMRs) at condition level across hospital systems for the purpose of healthcare system improvement. To accurately examine and contrast variation in condition-level SMRs across acute hospital systems, this outcome metric needs to be risk-adjusted for patient characteristics that make mortality more, or less, likely to occur. This article describes and evaluates the validity of risk-adjustment methods applied to service claims data to accurately determine SMRs across hospital systems. While service claims data may have limitations regarding case risk adjustment, it is important that we do not lose the important opportunity to use claims data as a reliable proxy to comment on the quality of care within healthcare systems. This methodology is robust in its demonstration of variation of performance on mortality outcomes across hospital systems. For the measurement period January 2014 - December 2016, the average risk-adjusted SMRs across hospital systems where DH members were hospitalised for acute myocardial infarction, stroke, pneumonia and coronary artery bypass graft procedures were 9.7%, 8.0%, 5.3% and 3.2%, respectively. This exercise of transparently examining variation in SMRs at hospital system level is the first of its kind in SA's private sector. Our methodological exercise is used to establish a local pattern of variation of SMRs in the private sector as the base off which to scrutinise reasons for variation and off which to build quality of care improvement strategies. High-performing healthcare systems must seek out opportunities for learning and continuous improvement such as those offered by examining important quality of care outcome measures across hospitals.
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 | 0.022 | 0.050 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.003 |
| Bibliometrics | 0.002 | 0.004 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.005 | 0.003 |
| Open science | 0.004 | 0.003 |
| Research integrity | 0.003 | 0.004 |
| Insufficient payload (model declined to judge) | 0.002 | 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".