Statistics and Indicators

The Kessler 6 (K6) scale was developed to discriminate cases of serious mental illness from non-cases (https://www.hcp.med.harvard.edu/ncs/k6_scales.php). It uses a five-level response scale about how often someone reports feeling nervous, hopeless, restless or fidgety, that everything was an effort, so sad that nothing could cheer them up and worthless in the past four weeks.

Between 2020 and 2021, the proportion of respondents to the ACT General Health Survey aged 18 years and over who have probable serious mental illness based on their K6 score has remained stable (2020: 5.1%; 2021: 5.5%). In 2021, there was no significant difference by sex (males: 5.3%; females: 5.6%).

Probable serious mental illness is based on a score of 19-30. This corresponds with the score categorisation used by the Australian Bureau of Statistics. For more information, visit: https://www.abs.gov.au/ausstats/abs@.nsf/Lookup/4817.0.55.001Chapter92007-08.

If a respondent was missing one value, the missing value was replaced with the mean of the five non-missing values. If a respondent was missing more than one value, they were excluded from analysis.

For the purpose of reporting the ACT General Health Survey data on HealthStats, if the 95% confidence intervals of the estimates do not overlap, they are considered to be significantly different.

Note: The indicator shows self-reported data collected through Computer Assisted Telephone Interviewing (CATI). Estimates were weighted to adjust for differences in the probability of selection among respondents and were benchmarked to the estimated residential population using the latest available Australian Bureau of Statistics population estimates.

Persons includes male, female, other and refused sex respondents and may not always add to the sum of male and female.

The following estimates have a relative standard error between 25% and 50% and should be used with caution:

- 2011/12, 2013/14, 2015/16, 2018, 2020 males
- 2013/14 females.

Statistically significant differences are difficult to detect for smaller jurisdictions such as the Australian Capital Territory. Sometimes, even large apparent differences may not be statistically significant. This is particularly the case in breakdowns of small populations because the small sample size means that there is not enough power to identify even large differences as statistically significant.