Healthy Communities: HPV immunisation rates in 2014–15 presents information about the percentages of adolescent girls and boys who were fully immunised against HPV according to the recommended schedule.3
De-identified unit record data were extracted from the HPV Register on 4 August 2016 to estimate HPV immunisation rates.
The indicator used in the report was calculated using the following method.
The number of adolescents aged 15 at 30 June 2015 who had received three doses of the HPV vaccine according to the schedule by 30 June 2015.
Australian Bureau of Statistics Estimated Resident Population (ERP) aged 13 as at 30 June 2013.
Note: The ERP at age 13 as at 30 June 2013 reflects the population at the time the vaccinations were likely administered.
(Numerator ÷ denominator) × 100, reported at one decimal place
Sex and geography: state and territory, Primary Health Network (PHN) area and Statistical Area Level 4 (SA4).
Calculation of variability in the population
Due to known issues with variability in population estimates by sex and single year of age for some areas of geography, the variability in the population for the same cohort of adolescents between two time points was calculated to identify areas with a substantial population increase or decrease over the period.
Variability in the population was calculated as the absolute per cent change between the number of adolescents aged 13 (vaccination year) and the number aged 15 (reporting year) two years later, using the ERP summarised to geography (either PHN area or SA4) for each financial year reported.
Results for areas with a substantial population change have either been suppressed (NP) or marked as ‘interpret with caution’ (#).
Method for measuring ‘significant change over time’ in HPV immunisation rates for girls
In the report, the following method was used to determine which areas had a significant increase or decrease over time in the percentage of girls fully immunised against HPV. This method accounts for differences in the size of populations in local areas, allowing fairer comparisons across them.
A two-proportion z-test was used to determine which areas had a statistically significant increase or decrease in the percentage of fully immunised girls from 2012–13 to 2013–14 and 2013–14 to 2014–15. The explanation below uses the change from 2013–14 to 2014–15 as an example.
For each local area, a pooled sample proportion (p) was used to compute the standard error of the sampling distribution:
Where p1 is the proportion of girls fully immunised in 2013–14, p2 is the proportion of girls fully immunised in 2014–15, n1 is the number of girls on the National HPV Register for the area in 2013–14, and n2 is the number of girls in 2014–15.
To calculate p, take n1 over n1 plus n2, all multiplied by p1. Then add the result of the following calculation: n2 over n1 plus n2, multiplied by p2.
The standard error (SE) of the sampling distribution difference between two proportions was then calculated as follows:
Where p is the pooled sample proportion, n1 is the number of girls in 2013–14, and n2 is the number of girls in 2014–15.
To calculate SE, take the square root of the addition of two results, 1 minus p multiplied by p divided by n1, and repeat for n2 for the second results to add to the first.
The test statistic (z) was then defined by the following equation:
Where p1 is the proportion of girls fully immunised in 2013–14, p2 is the proportion of girls fully immunised in 2014–15, and the SE is the standard error of the sampling distribution. This z-statistic has approximately a standard normal distribution when there is no statistically significant change in HPV immunisation rates.
To calculate z, take p1 from p2 and divide the result by SE.
Statistically significant changes were determined using the z critical value of 95% (0.05), as follows:
|z >-1.0 and z <1.0:||
Considered to be no change in HPV immunisation rates
|z <-1.0 and z>-1.96 or z > 1.0 and z <1.96:||
Considered to be no change in HPV immunisation rates, deviation which was considered to be an increase or decrease that was not statistically significant
|z <-1.96 or z>1.96:||
Considered to be statistically significant increase or decrease in HPV immunisation rates.
3. DoHA (Department of Health and Ageing) 2009. Chief Medical Officer Guidance on revaccination where HPV vaccine doses have been given at less than recommended minimum intervals January 2009. Canberra: DoHA. Viewed 20 February 2017, http://www.immunise.health.gov.au/internet/immunise/publishing.nsf/Content/cmo-full-advice-hpv-cnt/$File/CMO-full-advice-hpv.pdfExternal link, opens in a new window..