scholarly journals Failure of Administrative Data to Guide Asthma Care

Spectrum ◽  
2018 ◽  
Author(s):  
Joel Agarwal ◽  
Jennifer LaBranche ◽  
Jessica Cohen ◽  
Chris De Gara ◽  
Dilini Vethanayagam
Keyword(s):  
Administrative Data ◽  
Time Frame ◽  
Chronic Inflammatory Disease ◽  
Asthma Diagnosis ◽  
Learning Program ◽  
Diagnostic Codes ◽  
Canadian Health ◽  
Physician Learning ◽  
Alberta Health ◽  
Administrative Datasets

Rationale: Asthma is a chronic inflammatory disease of the airways that is very common (7.9% ofCanadians over the age of 12). Despite numerous clinical guidelines, education events and administrativedata reviews, there has been little change to the way asthma is managed in the Canadian health caresystem for nearly 30 years. We evaluated, through the Physician Learning Program (PLP) in Alberta,possible reasons why administrative datasets have not been able to provide meaningful information toadjust health policy. Methods: Provincial data was attained through Alberta Health Service and Alberta Health on pulmonaryfunction testing from 2005-2011 (through the PLP). The number of asthma diagnosis made during the sametime frame were then compared. Results: The preliminary results of the PLP found that spirometry was billed for roughly half as often asthe asthma diagnostic codes were utilized during the same time frame. However, the review also revealedinconsistencies in how administrative data are captured, making it difficult to determine whetherspirometry is being underutilized by physicians in making asthma diagnoses. Conclusions: Inconsistencies in how administrative data are captured in Alberta may be contributingto an incomplete picture of the rates of asthma diagnosis and physiological testing, and may explain, inpart, the limited influence of administrative datasets on guiding meaningful change within the healthcaresystem.

scholarly journals Simulated Physician Learning Program Improves Glucose Control in Adults With Diabetes

Diabetes Care ◽  
2010 ◽  
Vol 33 (8) ◽  
pp. 1727-1733 ◽  
Author(s):  
J. M. Sperl-Hillen ◽  
P. J. O'Connor ◽  
W. A. Rush ◽  
P. E. Johnson ◽  
T. Gilmer ◽  
...  
Keyword(s):  
Glucose Control ◽  
Learning Program ◽  
Physician Learning

scholarly journals Accuracy of Administrative Data for the Coding of Acute Stroke and TIAs

2016 ◽  
Vol 43 (6) ◽  
pp. 765-773 ◽  
Author(s):  
Ruth Hall ◽  
Luke Mondor ◽  
Joan Porter ◽  
Jiming Fang ◽  
Moira K. Kapral
Keyword(s):  
Ischemic Stroke ◽  
Acute Stroke ◽  
Administrative Data ◽  
Data Validation ◽  
Reference Standard ◽  
Stroke Registry ◽  
Diagnostic Codes ◽  
Icd 10 ◽  
Stroke Type ◽  
Ischemic Attack

AbstractObjective: Administrative data validation is essential for identifying biases and misclassification in research. The objective of this study was to determine the accuracy of diagnostic codes for acute stroke and transient ischemic attack (TIA) using the Ontario Stroke Registry (OSR) as the reference standard. Methods: We identified stroke and TIA events in inpatient and emergency department (ED) administrative data from eight regional stroke centres in Ontario, Canada, from April of 2006 through March of 2008 using ICD–10–CA codes for subarachnoid haemorrhage (I60, excluding I60.8), intracerebral haemorrhage (I61), ischemic (H34.1 and I63, excluding I63.6), unable to determine stroke (I64), and TIA (H34.0 and G45, excluding G45.4). We linked administrative data to the Ontario Stroke Registry and calculated sensitivity and positive predictive value (PPV). Results:: We identified 5,270 inpatient and 4,411 ED events from the administrative data. Inpatient administrative data had an overall sensitivity of 82.2% (95% confidence interval [CI95%]=81.0, 83.3) and a PPV of 68.8% (CI95%=67.5, 70.0) for the diagnosis of stroke, with notable differences observed by stroke type. Sensitivity for ischemic stroke increased from 66.5 to 79.6% with inclusion of I64. The sensitivity and PPV of ED administrative data for diagnosis of stroke were 56.8% (CI95%=54.8, 58.7) and 59.1% (CI95%=57.1, 61.1), respectively. For all stroke types, accuracy was greater in the inpatient data than in the ED data. Conclusion: The accuracy of stroke identification based on administrative data from stroke centres may be improved by including I64 in ischemic stroke type, and by considering only inpatient data.

scholarly journals Predicting postoperative surgical site infection with administrative data: a random forests algorithm

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Yelena Petrosyan ◽  
Kednapa Thavorn ◽  
Glenys Smith ◽  
Malcolm Maclure ◽  
Roanne Preston ◽  
...  
Keyword(s):  
Machine Learning ◽  
Administrative Data ◽  
Risk Score ◽  
Random Forests ◽  
Learning Algorithms ◽  
External Validation ◽  
Machine Learning Algorithms ◽  
Improvement Program ◽  
Health Administrative Data ◽  
Administrative Datasets

Abstract Background Since primary data collection can be time-consuming and expensive, surgical site infections (SSIs) could ideally be monitored using routinely collected administrative data. We derived and internally validated efficient algorithms to identify SSIs within 30 days after surgery with health administrative data, using Machine Learning algorithms. Methods All patients enrolled in the National Surgical Quality Improvement Program from the Ottawa Hospital were linked to administrative datasets in Ontario, Canada. Machine Learning approaches, including a Random Forests algorithm and the high-performance logistic regression, were used to derive parsimonious models to predict SSI status. Finally, a risk score methodology was used to transform the final models into the risk score system. The SSI risk models were validated in the validation datasets. Results Of 14,351 patients, 795 (5.5%) had an SSI. First, separate predictive models were built for three distinct administrative datasets. The final model, including hospitalization diagnostic, physician diagnostic and procedure codes, demonstrated excellent discrimination (C statistics, 0.91, 95% CI, 0.90–0.92) and calibration (Hosmer-Lemeshow χ2 statistics, 4.531, p = 0.402). Conclusion We demonstrated that health administrative data can be effectively used to identify SSIs. Machine learning algorithms have shown a high degree of accuracy in predicting postoperative SSIs and can integrate and utilize a large amount of administrative data. External validation of this model is required before it can be routinely used to identify SSIs.

Success Rates for Retinal Detachment Repair in Alberta: A Physician Learning Program Initiative

2018 ◽  
Vol 241 (3) ◽  
pp. 170-172
Author(s):  
Assaf Dotan ◽  
Dianne Johnson ◽  
Amin Kherani ◽  
Kahrram Jahangir ◽  
Matthew T.S. Tennant
Keyword(s):  
Retinal Detachment ◽  
Success Rates ◽  
Learning Program ◽  
Physician Learning

scholarly journals Predicting postoperative surgical site infection with administrative data: a Machine Learning algorithm

2020 ◽  
Author(s):  
Yelena Petrosyan ◽  
Kednapa Thavorn ◽  
Glenys Smith ◽  
Malcolm Maclure ◽  
Roanne Preston ◽  
...  
Keyword(s):  
Machine Learning ◽  
Administrative Data ◽  
Risk Score ◽  
Learning Algorithms ◽  
External Validation ◽  
Machine Learning Algorithms ◽  
Primary Data ◽  
Improvement Program ◽  
Health Administrative Data ◽  
Administrative Datasets

Abstract Background: Since primary data collection can be time-consuming and expensive, surgical site infections (SSIs) could ideally be monitored using routinely collected administrative data. We derived and internally validated efficient algorithms to identify SSIs within 30 days after surgery with health administrative data, using Machine Learning algorithms. All patients enrolled in the National Surgical Quality Improvement Program from the Ottawa Hospital were linked to administrative datasets in Ontario, Canada. Machine Learning approaches, including a Random Forests algorithm and the high-performance logistic regression, were used to derive parsimonious models to predict SSI status. Finally, a risk score methodology was used to transform the final models into the risk score system. The SSI risk models were validated in the validation datasets.Results: Of 14,351 patients, 795 (5.5%) had an SSI. First, separate predictive models were built for three distinct administrative datasets. The final model, including hospitalization diagnostic, physician diagnostic and procedure codes, demonstrated excellent discrimination (C statistics, 0.91, 95% CI, 0.90-0.92) and calibration (Hosmer-Lemeshow χ2 statistics, 4.531, p=0.402). Conclusion: We demonstrated that health administrative data can be effectively used to identify SSIs. Machine learning algorithms have shown a high degree of accuracy in predicting postoperative SSIs and can integrate and utilize a large amount of administrative data. External validation of this model is required before it can be routinely used to identify SSIs.

scholarly journals Measuring Dementia Incidence Within A Cohort Of 267,153 Older Australians Using Routinely Collected Linked Administrative Data

2020 ◽  
Vol 5 (5) ◽  
Author(s):  
Heidi J Welberry ◽  
Henry Brodaty ◽  
Benjumin Hsu ◽  
Sebastiano Barbieri ◽  
Louisa R Jorm
Keyword(s):  
Administrative Data ◽  
Observational Studies ◽  
Data Linkage ◽  
Aged Care ◽  
Incidence Rates ◽  
Residential Aged Care ◽  
Care Needs ◽  
Dementia Incidence ◽  
Linked Administrative Data ◽  
Administrative Datasets

IntroductionThere is no gold standard method for monitoring dementia incidence in Australia. Routinely collected linked administrative data are increasingly being used to monitor endpoints in observational studies and clinical trials and could benefit dementia research. Objectives and ApproachThis study examines dementia incidence within different Australian administrative datasets and how characteristics vary across datasets for groups detected as having dementia. This was an observational data linkage study based on a prospective cohort of 267,153 people in New South Wales, Australia from the 45 and Up Study. Participants completed a survey in 2006-2009 and dementia was identified using linked pharmaceutical claims (provided by Services Australia), hospitalisations, assessments of aged care eligibility, care needs at entry to residential aged care and death certificates. Data linkage was undertaken by the Centre for Health Record Linkage (CHeReL) and the Australian Institute of Health and Welfare. Age-specific and age-standardised incidence rates, incidence rate ratios and survival from first dementia diagnosis were calculated. ResultsAge-standardised dementia incidence was 16.9 cases per 1000 person years (PY) for people aged 65 years and over. Estimates for those aged 80-89 years were closest to published incidence rates (91% of rates for high-income countries). Relationships with dementia incidence were inconsistent across datasets for characteristics including sex, relative socio-economic disadvantage, support network size, marital status, functional limitations and diabetes. Median survival from first pharmaceutical claim for an anti-dementia medicine was 3.7 years compared to 3.0 years from first aged care eligibility assessment, 2.0 years from a dementia-related hospitalisation and 1.8 years from first residential aged care needs assessment. Conclusion / ImplicationsPeople identified with dementia in different administrative datasets have different characteristics, reflecting the factors that drive interaction with specific services. Bias may be introduced if single data sources are used to identify dementia as an outcome in observational studies.

scholarly journals A method for Linking Multiple De-identified Datasets

2018 ◽  
Vol 3 (2) ◽  
Author(s):  
Andrew Waugh ◽  
David Rowley ◽  
Auren Clarke
Keyword(s):  
Administrative Data ◽  
Bloom Filters ◽  
Population Estimates ◽  
Data Team ◽  
Scoring Algorithm ◽  
Run Time ◽  
The Difference ◽  
Link Strength ◽  
And Gender ◽  
Administrative Datasets

BackgroundNational Statistics Institutes have been exploring the value of using administrative data. The Administrative Data Team within the Scotland’s Census 2021 Programme are exploring bringing administrative datasets together to support the censusand produce alternative population estimates. ObjectivesWe are developing methods to link de-identified administrative datasets, drawing on existing methods. MethodsOur method uses hashed linking variables, derived from name, address, date of birth and gender. One linking variable is a names correction, produced by comparing names to each name in a reference set and scoring the difference. The scoring algorithm developed considers transpositions, deletions, insertions, substitutions and moves, and is sensitive to the particular letters involved. Linking variables are combined at run time to produce thousands of matchkeys, allowing more matches to be linked deterministically using hashed data. Overall link strength scores are calculated as a combination of: Penalties associated with the matchkey, based on the linking variables used, and Similarity on dates of birth, measured at run time using weighted Bloom Filters. We concatenate all the datasets and link the resulting dataset to itself. This allows simultaneous linking across all datasets and resolution of duplicate records within each dataset. This results in potentially complex patterns of links. By considering the records and links as a graph we allocaterecords to unique individuals through a vertex colouring algorithm on the complement of each component. The link strength is considered to prioritize allocation. FindingsClerical review on links made found that those with stronger scores were more likely to be considered a match. ConclusionsThis linking method is being used and tested further in linking admin datasets for population estimates. We also plan to use it for several linking tasks in the processing of Scotland’s Census 2021.

The UK Administrative Data Research Network: Its Genesis, Progress, and Future

2017 ◽  
Vol 675 (1) ◽  
pp. 184-201 ◽  
Author(s):  
Peter Elias
Keyword(s):  
Administrative Data ◽  
Historical Context ◽  
Research Network ◽  
Network Infrastructure ◽  
The Uk ◽  
Administrative Datasets

This article describes the ways in which UK administrative data are becoming more widely available for research. I outline the historical context of these developments, detail the network infrastructure that has now been put in place and discuss the continuing legal measures that are required to bring this network to fruition. I focus on the lessons that have been learned as work has progressed, drawing on this experience to elucidate some principles that may have relevance to similar attempts to promote better access to and linkage between administrative datasets in different cultural and legal settings.

Maximising the value of hospital administrative datasets

2010 ◽  
Vol 34 (2) ◽  
pp. 216 ◽  
Author(s):  
Shyamala G. Nadathur
Keyword(s):  
Administrative Data ◽  
Performance Monitoring ◽  
International Classification Of Diseases ◽  
Audit And Feedback ◽  
Management Tools ◽  
Classification Of Diseases ◽  
Diagnostic Related Groups ◽  
Data Collections ◽  
Hospital Administrative ◽  
Administrative Datasets

Mandatory and standardised administrative data collections are prevalent in the largely public-funded acute sector. In these systems the data collections are used for financial, performance monitoring and reporting purposes. This paper comments on the infrastructure and standards that have been established to support data collection activities, audit and feedback. The routine, local and research uses of these datasets are described using examples from Australian and international literature. The advantages of hospital administrative datasets and opportunities for improvement are discussed under the following headings: accessibility, standardisation, coverage, completeness, cost of obtaining clinical data, recorded Diagnostic Related Groups and International Classification of Diseases codes, linkage and connectivity. In an era of diminishing resources better utilisation of these datasets should be encouraged. Increased study and scrutiny will enhance transparency and help identify issues in the collections. As electronic information systems are increasingly embraced, administrative data collections need to be managed as valuable assets and powerful operational and patient management tools.

Sign in / Sign up

Export Citation Format

Share Document