Laura McDonald, Immunization and Respiratory Infections Branch/Public Health, Public Health Agency of Canada/Alberta Health and Wellness, 24th Floor 10024 Jasper Ave, Edmonton, AB, Canada
Learning Objectives for this Presentation:
By the end of this presentation participants will be able to:
Understand the complexity of using standards developed for surveys on registry based data
Background:
The Alberta Immunization Registry was developed in 2002 and continues to increase in functionality. Once fully populated, this system will be a rich source of information for determining vaccine uptake and coverage rates. To date, survey data have been the standard for coverage rate calculation, however these data have known shortcomings (such as parental recall and ambiguous denominator populations) making the interpretation of the information challenging. With a fully populated registry, coverage rate calculation can be automated and more precise. Also, the opportunities for novel measures of vaccine delivery success and population health are extensive.
Objectives:
The object of this study was to examine the current coverage standards and determine if refinement was needed to apply them to registry data.
Methods:
A working group was established within the province to look at the current standards and decide the ideal future measures. Iterations of SAS code were developed and tested based on the group's decisions. Rates produced by the new program were compared against those reported using the traditional method to validate the results.
Results:
There was consensus that it was ideal to break down coverage measures into two categories: those that determine population immunity, and those that measure programmatic success. Some of the main benefits to using registry data include: 1) the ability to match denominators to numerators, 2) flexibility in selecting coverage period and age of interest and 3) the ability to report information at higher geographic resolutions than typically available from surveys.
Conclusions:
Exact immunization coverage rates can be measured using registry data, however standards taking into account the granularity of these data have yet to be published. This study is a first attempt at developing these standards.