Incorporation of expert knowledge in the statistical detection of diagnosis related group misclassification. (April 2020)
- Record Type:
- Journal Article
- Title:
- Incorporation of expert knowledge in the statistical detection of diagnosis related group misclassification. (April 2020)
- Main Title:
- Incorporation of expert knowledge in the statistical detection of diagnosis related group misclassification
- Authors:
- Suleiman, Mani
Demirhan, Haydar
Boyd, Leanne
Girosi, Federico
Aksakalli, Vural - Abstract:
- Graphical abstract: Highlights: Weakly informative Bayesian models are effective in detecting DRG misallocation. Prediction is improved by incorporating subjective opinion elicited from experts as modelling inputs. A hybrid prior model using elicited expert guesses is proposed. Expert guesses are only used if their accuracy is better than random. The hybrid prior is best in 14 out of 20 trials, outperforming benchmark models. Abstract: Background: In activity based funding systems, the misclassification of inpatient episode Diagnostic Related Groups (DRGs) can have significant impacts on the revenue of health care providers. Weakly informative Bayesian models can be used to estimate an episode's probability of DRG misclassification. Methods: This study proposes a new, Hybrid prior approach which utilises guesses that are elicited from a clinical coding auditor, switching to non-informative priors where this information is inadequate. This model's ability to detect DRG revision is compared to benchmark weakly informative Bayesian models and maximum likelihood estimates. Results: Based on repeated 5-fold cross-validation, classification performance was greatest for the Hybrid prior model, which achieved best classification accuracy in 14 out of 20 trials, significantly outperforming benchmark models. Conclusions: The incorporation of elicited expert guesses via a Hybrid prior produced a significant improvement in DRG error detection; hence, it has the ability to enhance theGraphical abstract: Highlights: Weakly informative Bayesian models are effective in detecting DRG misallocation. Prediction is improved by incorporating subjective opinion elicited from experts as modelling inputs. A hybrid prior model using elicited expert guesses is proposed. Expert guesses are only used if their accuracy is better than random. The hybrid prior is best in 14 out of 20 trials, outperforming benchmark models. Abstract: Background: In activity based funding systems, the misclassification of inpatient episode Diagnostic Related Groups (DRGs) can have significant impacts on the revenue of health care providers. Weakly informative Bayesian models can be used to estimate an episode's probability of DRG misclassification. Methods: This study proposes a new, Hybrid prior approach which utilises guesses that are elicited from a clinical coding auditor, switching to non-informative priors where this information is inadequate. This model's ability to detect DRG revision is compared to benchmark weakly informative Bayesian models and maximum likelihood estimates. Results: Based on repeated 5-fold cross-validation, classification performance was greatest for the Hybrid prior model, which achieved best classification accuracy in 14 out of 20 trials, significantly outperforming benchmark models. Conclusions: The incorporation of elicited expert guesses via a Hybrid prior produced a significant improvement in DRG error detection; hence, it has the ability to enhance the efficiency of clinical coding audits when put into practice at a health care provider. … (more)
- Is Part Of:
- International journal of medical informatics. Volume 136(2020)
- Journal:
- International journal of medical informatics
- Issue:
- Volume 136(2020)
- Issue Display:
- Volume 136, Issue 2020 (2020)
- Year:
- 2020
- Volume:
- 136
- Issue:
- 2020
- Issue Sort Value:
- 2020-0136-2020-0000
- Page Start:
- Page End:
- Publication Date:
- 2020-04
- Subjects:
- Bayesian analysis -- Clinical coding -- DRGs -- Health informatics -- Statistical modeling
Medical informatics -- Periodicals
Information science -- Periodicals
Computers -- Periodicals
Medical technology -- Periodicals
Medical Informatics -- Periodicals
Technology, Medical -- Periodicals
Computers
Information science
Medical informatics
Medical technology
Electronic journals
Periodicals
Electronic journals
610.285 - Journal URLs:
- http://www.sciencedirect.com/science/journal/13865056 ↗
http://www.clinicalkey.com/dura/browse/journalIssue/13865056 ↗
http://www.clinicalkey.com.au/dura/browse/journalIssue/13865056 ↗
http://www.elsevier.com/journals ↗ - DOI:
- 10.1016/j.ijmedinf.2020.104086 ↗
- Languages:
- English
- ISSNs:
- 1386-5056
- Deposit Type:
- Legaldeposit
- View Content:
- Available online (eLD content is only available in our Reading Rooms) ↗
- Physical Locations:
- British Library DSC - 4542.345250
British Library DSC - BLDSS-3PM
British Library HMNTS - ELD Digital store - Ingest File:
- 13430.xml