Calculating vaccination coverage estimates from the Australian Childhood Immunisation Register

The cohort method for calculating vaccination coverage was applied to ACIR data to measure vaccination coverage. This method has been routinely used in Australia for the reporting of vaccination coverage, and details have been previously published.116 For the purpose of this report, 12-month birth cohorts have been used. Each cohort includes children born between 1 January and 31 December for each respective 12-month period, 2007, 2008, 2009 and 2010. Vaccination coverage has previously been reported for children at 12, 24 and 72 months of age.13 Since 2008, ‘fully vaccinated’ coverage estimates have been reported at 60 months of age rather than 72 months.157 The assessment date for determining vaccination coverage was at least 3 months following each milestone age for the cohort, to allow for delayed reporting to the ACIR.

For the purpose of calculating vaccination coverage estimates, only children who were registered with Medicare were included in the analysis, to minimise the potential for duplicate records in the ACIR. As per previous coverage reports, the ‘third-dose assumption’, by which record of receipt of a later dose of a vaccine with a multiple dose schedule implies receipt of earlier doses of the schedule, was used.116 Evaluation of this assumption has concluded that it is appropriate for the reporting of population vaccination coverage rates.158,159

Vaccination coverage for Aboriginal and Torres Strait Islander children was calculated by dividing the numerator (children registered with Medicare as Aboriginal and Torres Strait Islander and reported as vaccinated) by the denominator (total number of children registered with Medicare as Aboriginal and Torres Strait Islander) within each birth cohort. Children for whom Aboriginal and Torres Strait Islander status was missing or unknown on their ACIR record have been categorised as other for the purpose of this report. Vaccination coverage of these children was calculated in a similar manner. The definition of a child being ‘fully vaccinated’ is based on the National Immunisation Program (NIP) schedule as per The Australian Immunisation Handbook, 9th edition.9 Additional vaccines will be included in this calculation from July 2013, and results for these extended definitions are also presented (Table A.1.1).

Table A.1.1: Inclusion criteria for 'fully vaccinated' definitions, by milestone
'Fully vaccinated' category Vaccine
DTPa Polio Hib HepB MMR MenC 7vPVC Rotavirus Varicella

* Applicable from 1996 to June 2013

† Applicable from July 2013

12 months – old*
y
y
y
y
12 months – new
y
y
y
y
y
12 months – new inc. rotavirus
y
y
y
y
y
y
24 months – old*
y
y
y
y
y
24 months – new
y
y
y
y
y
y
y
60 months
y
y
y

Top of page

Vaccination coverage expressed as children ‘fully vaccinated’ was calculated for Aboriginal and Torres Strait Islander and other children to allow for comparison. The standard definition of ‘fully vaccinated’ at 12 months of age is documented receipt on the ACIR of a third dose of a diphtheria-, tetanus- and acellular pertussis-containing vaccine; a third dose of polio vaccine; a second or third dose of a PRP-OMP-containing vaccine or a third dose of any other Hib vaccine; and a second or third dose of a Comvax hepatitis B vaccine or a third dose of any other hepatitis B vaccine.

The extended definition for ‘fully vaccinated’ at 12 months of age is as per the standard definition, but with the addition of a third dose of 7vPCV (as per the definition applicable from July 2013), and also with the addition of a second or third dose of rotavirus vaccine.

The standard definition for ‘fully vaccinated’ at 24 months of age is documented receipt on the ACIR of a third dose of a diphtheria-, tetanus- and acellular pertussis-containing vaccine; a third dose of polio vaccine; a third or fourth dose of a PRP-OMP-containing vaccine or a fourth dose of any other Hib vaccine; a third or fourth dose of Comvax hepatitis B vaccine or a fourth dose of any other hepatitis B vaccine; and a first dose of a measles-, mumps- and rubella-containing vaccine.

The extended definition for ‘fully vaccinated’ at 24 months of age is as per the standard definition, but with the addition of a first dose of varicella vaccine and a first dose of meningococcal C vaccine, as per the definition applicable from July 2013.

The definition used for ‘fully vaccinated’ at 60 months of age is documented receipt on ACIR of a fourth or fifth dose of a diphtheria-, tetanus- and acellular pertussis-containing vaccine; a fourth dose of polio vaccine; and a second dose of a measles-, mumps- and rubella-containing vaccine.

The differing number of doses in the inclusion criteria for various vaccines (such as Hib and hepatitis B vaccines) corresponds to the recommended schedule for the specific formulation of vaccine that a child received.

Vaccination coverage was also calculated for selected individual vaccines provided under the NIP, including conjugate and polysaccharide pneumococcal vaccines and hepatitis A, rotavirus and MMR vaccines.

Top of page

Calculating vaccination timeliness from the Australian Childhood Immunisation Register

In addition to vaccination coverage, we report timeliness of vaccination. As per other national reports,157 age-appropriate vaccination was defined as receipt of a scheduled vaccine dose within 30 days of the schedule point (age) recommended in The Australian Immunisation Handbook.160

For the purpose of reporting, vaccine doses given too early or not reported were excluded from the analysis. The proportion of children who received the dose within 30 days of the schedule point were defined as age-appropriately immunised. Remaining children were categorised by the length of time between the vaccine schedule point and when the vaccine dose was administered (i.e. vaccine received 1–2 months after schedule point, 3–6 months after schedule point, or >6 months after schedule point).

In response to high rates of pertussis infection in Australia since 2008,161 the Australian Technical Advisory Group on Immunisation, in February 2011, endorsed the recommendation to bring forward the first dose of DTP vaccine for infants from 8 weeks (2 months) of age to 6 weeks of age. To assess the implementation of this intervention, timeliness of the first dose of DTP was calculated, comparing the proportion vaccinated at 42–60 days of age with the proportion vaccinated at 61–92 days of age.

Geographic variations of vaccination coverage

Vaccination coverage estimates for being ‘fully vaccinated’ at 12 months of age by Australian Statistical Division (SD) were used as an indicator for geographic variations of vaccination coverage among Aboriginal and Torres Strait Islander children. ACIR child records were allocated to SDs by the postcode of residence recorded in the ACIR. Coverage estimates were calculated for each SD using the methods described previously; results were then mapped.

Vaccination coverage data and timeliness data were analysed by remoteness status based on the Accessibility/Remoteness Index of Australia (ARIA), developed by the Australian Government Department of Health.162 The ARIA system is based on the road distance of a location to the nearest service centre, with locations categorised into five groups. For the purpose of this report, the two groups with the most restricted access to services (remote and very remote) are classified as ‘remote’; the other three categories (highly accessible, accessible and moderately accessible) are classified as ‘accessible’. Vaccine timeliness of the second dose of DTPa (DTP2) and the first dose of MMR (MMR1) was assessed for both remote and non-remote categories.

Top of page

Vaccine refusal

The parents of Medicare registered children may lodge a conscientious objection to immunisation if they do not wish their children to be immunised. The percentage of conscientious objectors was measured by dividing the total number of objectors by the total number of Medicare registered children. The proportion of conscientious objectors was calculated from children born in the cohort from January 2004 to December 2008, to allow a large enough sample for statistical accuracy and sufficient time for parents to have lodged a conscientious objection form with Medicare. The proportions of conscientious objectors among Aboriginal and Torres Strait Islander and other children were compared.

Data quality and notes on interpreting coverage data

General under-reporting in the Australian Childhood Immunisation Register

It should be noted that there is a potential for general under-reporting of children and vaccine doses to the ACIR. Firstly, vaccination coverage estimates do not account for children not registered with Medicare; however, the impact of this is minimal. It is reported that by 12 months of age 99% of Australian children have been registered with Medicare.115 Secondly, the ACIR is dependent on reporting of vaccination doses by service providers. Providers report to the ACIR via the Medicare Australia website or through submission of a paper form which is then centrally updated to the ACIR.160 A national population-based survey was conducted in 2001 to measure the proportion of under-reporting to the ACIR and to calculate corrected vaccination coverage estimates. The study found that at 12 months of age the ACIR underestimated vaccination coverage by 2.7% (95% CI 2.4–3.0) and at 24 months of age by 6.5% (95% CI 6.1–6.9).163 However, since the time of this study, incentives have been introduced for both parents and providers to encourage reporting of vaccination to the ACIR (detailed elsewhere)160 and rates of electronic reporting have increased markedly. The impact of these incentives on reporting to the ACIR by service providers and the subsequent effect on vaccination coverage estimates has not been evaluated.

Indigenous status identification in the Australian Childhood Immunisation Register

In addition to general under-reporting, completion and accuracy of Indigenous status in the ACIR has, in the past, been suboptimal. Despite past data quality issues, significant improvements to data quality have been made since the inception of the ACIR. The first national surveillance report on vaccine preventable diseases and vaccination coverage in Aboriginal and Torres Strait Islander Australians, which covered the period 1999 to 2002, analysed coverage data from New South Wales, the Northern Territory, South Australia, Victoria and Western Australia only, as Indigenous status reporting was inadequate in other states.132 Indigenous status identification greatly improved between 2003 and 2004 and in the second of these reports, for the period 2003 to 2006, data quality had improved enough to allow all states and jurisdictions to be included in the analysis.13

The improvement in data quality is supported by several studies. Vlack et al. demonstrated that Aboriginal and Torres Strait Islander coverage estimates obtained from the ACIR were comparable to estimates provided by a survey of Aboriginal and Torres Strait Islander children in Queensland.164 Rank and Menzies demonstrated that Aboriginal and Torres Strait Islander status identification in the ACIR increased from 42% of the national cohort of Aboriginal and Torres Strait Islander children aged 12–14 months in 2002 to 95% in 2005.135