Impact of Implementing an Evidence-Based Decision Support Tool for Reducing Inappropriate Medical Imaging Services
Bibliographic record
Abstract
Objective: To reduce the incidence of inappropriate diagnostic imaging, thereby resulting in a reduction of the inappropriate utilisation of scarce health care resources. Methods: An ‘on-line’ application called Diagnostic Imaging Pathways (DIP) was developed by the Division of Imaging Services at Royal Perth Hospital (RPH), Western Australia’s largest public hospital. DIP is an evidence-based education and decision support tool designed to assist clinicians to request the most appropriate examinations and in the best sequence to achieve a diagnosis. The application is accessed by employees of Western Australia’s public health system from the ‘desk-top’ by selecting a prominent icon, and it may also be accessed from the Internet (www.imagingpathways.health.wa.gov.au). Clinicians at RPH were regularly alerted to the desirability of complying to DIP recommendations in their diagnostic practice at presentations at the ‘Grand Round’, in items in the medical newsletter, and in the orientation and induction programme of junior doctors. Retrospective audits were then carried out in relation to referrals for medical imaging from the Emergency Department (ED) of RPH, the busiest in Australia, to establish compliance between diagnostic imaging referral practice and DIP recommendations. ‘Suspected Pulmonary Embolism’ (http://www.imagingpathways.health.wa.gov.au/includes/DIPMenu/pe/chart.html): patients referred over a 6 month period for a CT Pulmonary Angiogram (N = 94) or a Radionuclide Scan (N = 100). ‘Acute Ankle Sprain’ (http://www.imagingpathways.health.wa.gov.au/includes/DIPMenu/ankle/chart.html): patients presenting over a 4 month period with acute blunt ankle or mid-foot trauma (N = 160). ‘Suspected Renal Colic’ (http://www.imagingpathways.health.wa.gov.au/includes/DIPMenu/rencolic/chart.html): patients referred over a 3 month period for medical imaging with a provisional diagnosis of renal colic (N = 89). ‘Non Traumatic Acute Abdominal Pain’ (http://www.imagingpathways.health.wa.gov.au/includes/DIPMenu/axr/Summary.html): a random sample of patients referred over a 2 month period for an abdominal plain film for investigation of ‘acute abdomen’ (N = 215). In each audit, deviations between diagnostic practice and DIP recommendations were documented and the impact on ‘work-flow’ was assessed. Results: ‘Suspected Pulmonary Embolism’: 173 patients (89%) did not have a risk assessment using the Wells Score recorded in the clinical notes, casting doubt on the appropriateness of subsequent investigations. 32 (16%) did not have a positive D-Dimer result, which was needed as an indication for an examination. ‘Acute Ankle Sprain’: 112 (70%) were not assessed according to the Ottawa Ankle Rules (OAR) but received an x-ray examination. 10 (6%) who had a negative OAR received an x-ray inappropriately. ‘Suspected Renal Colic’: 47 (53%) had an initial inappropriate x-ray. 18 of these and a further 18 (total 40%) should have had further imaging during the investigation of their condition but did not. ‘Non Traumatic Acute Abdominal Pain’: 69 (32%) received an inappropriate x-ray of the abdomen. In 88 of the remaining 146, the eventual diagnosis or management was not affected by the x-ray result. Of the total cohort (N= 658), 285 (43%) received an imaging examination of doubtful appropriateness and 158 (24%) received examinations without indications. Focusing on plain film radiographs, 42 hours of staff time in the Division of Imaging Services was spent providing examinations which did not have appropriate indications. To this must be added unspecified extra time in ED waiting for an examination booking. Conclusions: The easy availability and active marketing of DIP within RPH has not eliminated inappropriate diagnostic imaging. Currently, interventions are underway to achieve greater compliance between diagnostic referral practice and DIP recommendations, via proof that pre-requisites for each imaging request have been met. Electronic requesting linked to DIP for the creation of an electronic decision support system is planned.
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 | 0.099 | 0.329 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.004 |
| Bibliometrics | 0.007 | 0.003 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.009 | 0.005 |
| Open science | 0.004 | 0.006 |
| Research integrity | 0.006 | 0.004 |
| Insufficient payload (model declined to judge) | 0.007 | 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".