Related Experiment Videos
Linking mortality data with hospital morbidity data--a pilot study
1Public Health Branch, Health & Community Services, Melbourne, Victoria, Australia.
Abstract:
Record linkage is becoming an increasingly important technique in moni toring health outcomes. In this study we wought to examine the feasibility of linking mortality data held by the Registry of Birth, Deaths and Marriages with the Patient Master Index (PMI) at St. Vincent's Hospital, Melbourne, in order to document deaths soon after hospital discharge. Linkage was carried out using demographic details. Registry data required extensive editing and formatting prior to being matched with the PMI. 86.2% of cases identified by the computer algorithm were actual cases. Mortality data are an important source of information on health outcome but a process is needed to make them more usable. We have described one such method and suggested where the process could have been streamlined. The use of unique identifiers would have facilitated the matching process, curtailing several processing steps.
Insights
Linking hospital patient data with mortality records is feasible for tracking post-discharge deaths. This health outcomes monitoring method achieved 86.2% accuracy, highlighting the value of accessible mortality data.
Area of Science:
- Health Informatics
- Public Health Surveillance
- Data Linkage
Background:
- Effective health outcomes monitoring relies on accurate mortality data.
- Linking disparate health datasets is crucial for comprehensive patient tracking.
- Challenges exist in integrating data from vital statistics registries and hospital systems.
Purpose of the Study:
- To assess the feasibility of linking mortality data from the Registry of Birth, Deaths and Marriages with a hospital's Patient Master Index (PMI).
- To document deaths occurring shortly after hospital discharge using linked data.
- To evaluate the effectiveness of demographic data for record linkage in a healthcare setting.
Main Methods:
- Utilized demographic details for record linkage between the Registry of Birth, Deaths and Marriages and St. Vincent's Hospital's PMI.
- Performed extensive editing and formatting on Registry data before matching.
- Employed a computer algorithm to identify potential matches.
Main Results:
- Achieved an 86.2% accuracy rate for cases identified by the computer algorithm.
- Demonstrated the potential for linking mortality data with hospital records for outcome monitoring.
- Identified areas for process improvement in data linkage.
Conclusions:
- Record linkage between vital statistics and hospital data is a viable method for monitoring health outcomes.
- Improvements in data standardization and the use of unique identifiers could streamline the linkage process.
- Accessible and usable mortality data are essential for robust public health surveillance.