Accuracy of Manitoba administrative health databases to identify patients with cirrhosis
| dc.contributor.author | Faisal, Nabiha | |
| dc.contributor.examiningcommittee | Singh, Harminder (Community Health Sciences) | |
| dc.contributor.examiningcommittee | Renner, Eberhard (Internal Medicine) | |
| dc.contributor.examiningcommittee | Singer, Alexander (Family Medicine) | |
| dc.contributor.supervisor | Mahar, Alyson | |
| dc.contributor.supervisor | Lix, Lisa | |
| dc.date.accessioned | 2023-09-07T12:05:55Z | |
| dc.date.available | 2023-09-07T12:05:55Z | |
| dc.date.issued | 2023-06-23 | |
| dc.date.submitted | 2023-08-12T21:39:08Z | en_US |
| dc.date.submitted | 2023-09-04T16:34:48Z | en_US |
| dc.date.submitted | 2023-09-06T13:59:05Z | en_US |
| dc.degree.discipline | Community Health Sciences | en_US |
| dc.degree.level | Master of Science (M.Sc.) | |
| dc.description.abstract | Introduction: Cirrhosis is associated with a substantial clinical and economic burden on healthcare systems worldwide. Currently, health system planning for cirrhosis is challenged by a lack of provincial prevalence and incidence information. Administrative health data may be one way to complete surveillance and monitoring of cirrhosis burden in Manitoba; however, an algorithm to identify cases of cirrhosis does not exist. Therefore, the objectives of this thesis, were to develop and validate administrative data algorithms for cirrhosis and describe the epidemiology of cirrhosis in Manitoba overtime and investigate changes in its incidence and prevalence across various age groups and between males and females. Methods: A retrospective study was conducted using linked provincial administrative health data to primary care electronic medical records (EMR) to develop and validate algorithms for identifying cases of cirrhosis. A validated case definition from EMR data was the reference standard. Two optimal algorithms were used to estimate the burden and epidemiology of cirrhosis overtime. Annual incidence rates were estimated using a generalized linear model and generalized estimating equations with a negative binomial distribution adjusting for age and sex. Results: The estimates of sensitivity, specificity, and positive predictive value (PPV) of the two optimal algorithms when compared to a reference standard of validated primary care case definition ranged 42.8%-67.9%, 95.8%-97.5%, 18.4%-19.3% respectively. Application of these algorithms to provincial data demonstrated an increase in age- and sex-adjusted incidence and prevalence rates of cirrhosis between 2010-2019. The incidence of cirrhosis showed an average annual increase of 4-6%, with the most significant increases ranged 6-8% per year (p < 0.0001) in young adults aged 18-44 years. Regardless of the algorithm employed, the incidence rates among females were consistently higher than those among males (p <0.0001). Furthermore, the rate of change in cirrhosis incidence was significantly higher among young adults (p = 0.001) and females (p=.03). Conclusion: Cirrhosis patients can be identified from administrative health data with modest accuracy when a validated case definition for primary care EMRs was the reference standard. The incidence and prevalence of cirrhosis substantially increased overtime, signaling an alarming shift in the disease burden towards younger individuals and females. | |
| dc.description.note | October 2023 | |
| dc.identifier.uri | http://hdl.handle.net/1993/37596 | |
| dc.language.iso | eng | |
| dc.rights | open access | en_US |
| dc.subject | Epidemiology | |
| dc.subject | Incidence | |
| dc.subject | Prevalence | |
| dc.subject | Electronic Medical Records | |
| dc.subject | Case Definitions | |
| dc.subject | Algorithm | |
| dc.subject | Primary Care | |
| dc.subject | Reference Standard | |
| dc.title | Accuracy of Manitoba administrative health databases to identify patients with cirrhosis | |
| dc.type | master thesis | en_US |
| local.subject.manitoba | yes |