Development and validation of data quality rules in administrative health data using association rule mining
Bibliographic record
Abstract
IntroductionData quality assessment is a challenging facet for researches using coded administrative health data. Our previous study had demonstrated the potentials of association rule mining to assess data quality. The objective of this study is to develop and validate a set of coding association rules for data quality assessment. Objectives and ApproachWe used the Canadian reabstracted hospital discharge abstract data (DAD) with clinical diagnosis coded in International Classification of Disease – 10th revision, Canada (ICD-10-CA) codes for rule development. The DAD data were divided into 5 age groups. Association rule mining were conducted on reabstracted DAD in each age group to extract ICD-10 coding association rules at the three and four digits levels. The rule strength was assessed using support and confidence. The rules will be reviewed by a panel of 5 physicians and 2 coding specialists to assess their appropriateness from clinical and coding perspectives using a modified Delphi rating ResultsIn total, 975 rules at the three digits level and 822 rules at the four digits level were learned from the data. Half of the rules were in the age group of ≥65 and no rules were found in the age group of 5 to 19. The interquartile range of rule confidences were 0.112 to 0.425 in the three digits level and 0.073 to 0.222 in the four digits level. Two-thirds of rules had the diagnosis codes related to endocrine and metabolic disorders and diseases of circulatory, respiratory and genitourinary systems. The panel review will be conducted in early April and will have the final set of rules available before the conference. Conclusion/ImplicationsThis study developed a set of validated ICD-10 coding association rules and creates a useful tool to cost-effectively assess data quality in routinely collected administrative health data.
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 | 0.115 | 0.248 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.003 |
| Bibliometrics | 0.011 | 0.009 |
| Science and technology studies | 0.002 | 0.002 |
| Scholarly communication | 0.007 | 0.003 |
| Open science | 0.003 | 0.003 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.001 | 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".