The health of people in Australia's prisons 2022

All data contained in the 2022 NPHDC (National Prisoner Health Data Collection) report are based on information obtained from people aged 18 and over from participating prisons from all states and territories, except Victoria –which did not participate in the 2022 NPHDC.

Terminology

Accelerated ageing and age profiles

Various definitions of an ‘older prisoner’ exist in academic and government research. The threshold for what constitutes an older person in prison varies between 45 to 65 years and above and no consensus currently exists in the evidence base.

Accelerated ageing is common among people in prison who may experience a life expectancy gap of 10 years or more compared with the general Australian population (Baidawi et al. 2011; Turner and Trotter 2010). This means that a person in prison in their 50s may have the physical appearance and health problems of someone at least 10 years older in the general community and, so, are functionally older than their chronological age (Codd 2018).

Accelerated ageing occurs to a greater extent among people in prison because they are more likely to live in poverty, achieve a lower standard of education, and experience housing instability and a lack of employment (Turner and Trotter 2010). As well, having risky lifestyles and behaviours, being unable to access health-care services and experiencing incarceration are associated with poorer physical and mental health and can contribute to age-related illnesses and conditions (AIHW 2020; Codd 2018).

The 2022 NPHDC report therefore uses the threshold of 45 years and above to define an older person in prison (compared with 65 years and above in the general community). This lower age threshold (of 45+ years) was chosen to allow the report to better capture data on the health of older people in prison, as the sample size of older people in prison, particularly older females, restricts some detailed analyses. The national supplementary data tables are disaggregated by 5 age groups (18–24, 25–34, 35–44, 45–54 and 55+) where appropriate, so individual breakdowns can be viewed.

Disability

Standardised Disability Flag

The NPHDC collects information about people in prison with self-reported disability or long‑term health conditions using a version of the AIHW Standardised Disability Flag. The flag derives from a standard set of questions that assess a person’s level of functioning and need for support in everyday activities. The set is based on the International Classification of Functioning, Disability and Health, and is broadly consistent with the short disability questions that the Australian Bureau of Statistics (ABS) uses in a number of its social surveys. The Standardised Disability Flag includes 3 components:

• activity and participation need for assistance cluster
• education participation restriction flag
• employment participation restriction flag.

The activity and participation need for assistance cluster collects data on individuals’ perception of whether a long-term health condition or disability restricts their everyday activities. This component involves assigning the degree of assistance and/or supervision required to perform 8 everyday activities. For each of the 8 activities, there is a 5-level response scale. Data are then used to derive an activity limitation flag, the extent of activity limitation and an optional extent of core activity limitation (see Core activity limitation).

The education participation restriction flag and the employment participation restriction flag are derived from 2 separate questions; these are, in turn, independent from the activity limitation extent and flag items.

The Standard Disability Flag relies on self-reported impairment and restriction and does not require respondents to have a medical diagnosis. The flag does not include a question on disability type (AIHW 2016).

Core activity limitation

Core activities are everyday activities in communication, mobility, and self-care. These are activities deemed to be essential to normal everyday life. The extent of an individual’s core activity limitation is derived from the activity and participation need for assistance cluster from the Standardised Disability Flag (see ‘Standardised Disability Flag’ above).

Four levels of core activity limitation are determined based on whether a person needs help and/or supervision with, has difficulty with, or uses aids, equipment or medications for any of the 3 core activities. A person’s overall level of core activity limitation is determined by their highest level of limitation in these activities. The 4 levels of limitation are:

• profound limitation (people with the greatest need for help – who always need help, or who are unable to do an activity)
• severe limitation (people who sometimes need help and/or have difficulty)
• moderate limitation (people who need no help but have difficulty)
• mild limitation (people who need no help and have no difficulty, but use aids, equipment, or medications).

In the NPHDC, a profound or severe core activity limitation or restriction is defined as those who sometimes, or always need help and/or supervision with core activities of daily living (mobility, self-care, or communication).

Kessler Psychological Distress Scale-10

Psychological distress data presented in this report and supplementary tables are scored using the Kessler Psychological Distress Scale-10 (K10). The K10 was developed as a short dimensional measure of non-specific psychological distress in the anxiety-depression spectrum (Kessler et al. 2002; Andrews and Slade 2001). The K10 involves 10 questions about emotional states experienced in the previous 4 weeks. For each question, there is a 5-level response scale. Scores for each item are summed, yielding a minimum possible score of 10 and a maximum possible score of 50. Individuals must answer all 10 questions to be given a valid K10 score.

Results are grouped into the following 5 levels of psychological distress scoring categories:

• low (indicating little, or no psychological distress) – indicated by a score of 10–15
• moderate – indicated by a score of 16–21
• high – indicated by a score of 22–29
• very high – indicated by a score of 30–50.

Slightly different scoring for the K10 is sometimes used in other surveys and research (for example, low 10–19, moderate 20–24, high 25–29, very high 30–50), so caution should be used when comparing results. The K10 scoring used in this report is the same scoring used in ABS surveys to enable comparisons between the prison and general Australian populations (ABS 2012).

AUDIT-C

The data on alcohol consumption and alcohol-related harm presented in this report and supplementary tables were determined using questions on alcohol consumption from the World Health Organization’s Alcohol Use Disorder Identification Test (AUDIT) screening instrument (Babor et al. 2001). The AUDIT was developed as a simple method of screening for people with hazardous and harmful patterns of alcohol consumption.

The consumption component of this instrument (AUDIT-C) contains 3 consumption questions from the AUDIT, with each question scoring 0–4 points. Scores for the 3 questions are summed, with a minimum possible score of 0 (reflecting no alcohol use) and a maximum possible score of 12. Individuals must answer all 3 questions – or answer ‘0’ for question 1 only (reflecting no alcohol use) – to be given a valid AUDIT-C score. 

Results are grouped into risk categories of low, moderate, high, and very high depending on an individual’s sex. Generally, the higher the score, the more likely it is that a person’s drinking is affecting his or her safety. For both males and females, a score of 6 of more indicates a high risk of alcohol‑related harm.

It should be noted that the questions included in the 2022 NPHDC have additional wording added at the start of each AUDIT-C question relating to a time frame of ‘over the past 12 months’, so caution should be used when comparing other AUDIT-C results with those for the NPHDC.

Illicit drug use

The NPHDC collects self-reported data on the use of drugs for non-medical purposes in people entering and leaving prison. The term ‘illicit drugs’ in this report includes the following:

• illegal drugs (such as cocaine, heroin, and amphetamine type stimulants)
• pharmaceutical drugs (such as opioid-based pain relief medications, benzodiazepines and steroids) when used for non-medical purposes
• other substances, legal or illegal, used inappropriately, such as inhalants from petrol, paint or glue.

Presentation of data

Most data in this report have been rounded for readability. The following rounding rules have been applied:

• Numbers over 100,000 and under a million are rounded to the nearest multiple of 1,000.
• Numbers between 1,000 and 100,000 are rounded to the nearest multiple of 100.
• Numbers between 500 and 999 are rounded to the nearest multiple of 10.
• Numbers between 100 and 499 are rounded to the nearest multiple of 5.
• Numbers between 10 and 99 are rounded to the nearest whole number.
• Numbers under 10 are rounded to one decimal place.

Proportions presented in the report are shown as percentages, rounded to 1 decimal place when less than 10% and to whole numbers when over 10%.

While the NPHDC report focuses on national level data, supplementary data tables for national, state and territory data are available – see Data. State and territory data tables are available for indicators only, however, due to the low response rates in some states and territories, data should be interpreted with caution. Readers should satisfy themselves if the data are fit for purpose if they intend to use it where response rates are low.

No data have been age standardised.

Mean, median, and age range

Two measures of central tendency are sometimes used for reporting data in the NPHDC:

• Average (mean) – the average, or mean, is calculated by summing all the values for a particular data item, and dividing by the total number of observations. In the NPHDC, averages are used for some indicators that report age at which an event occurred, or as a summary statistic for survey sample data sets.
• Median – the median is the middle value of a set of observations, when arranged in ascending order. As a result, the median is not affected greatly by small or large outlier numbers. In the NPHDC, the median is often used in conjunction with the mean as a summary statistic for survey sample data sets, or where data are not normally distributed, or include extreme values that would distort the mean – for example, the median time that those sentenced could expect to serve.

Other measures of location sometimes used for reporting in the NPHDC reports and supplementary data tables include the minimum and maximum. In the NPHDC, this is often used to present the upper and lower bounds of ages (in years) for a particular indicator or data item. All means, medians and age ranges in the report and supplementary data tables have been indicated.

Standard deviation

One measure of spread – standard deviation – is sometimes used in the supplementary tables for reporting data in the NPHDC. The standard deviation measures how dispersed a set of observations are in relation to the mean.

In the NPHDC, the standard deviation is reported in conjunction with the mean as a survey sample data set summary statistic and/or for indicators reporting average age at which an event occurred.

All standard deviations reported in the supplementary data tables have been indicated.

Population rates

Population rates allow different groups to be compared while taking into account differences in population sizes.

In the NPHDC, rates are used for some indicators which report the number of events occurring in the population over a specified period of time. Rates derived from NPHDC data are expressed as the number per 100, or 1,000, people in or received into custody in 2021. All rates reported have been indicated.

Denominators for population rates for people in prison aged 18 and over are sourced from the ABS, unless otherwise stated.

Crude mortality rates

Data on deaths following release from prison came from the Department of Social Services, and are expressed as crude mortality rates (crude death rates). In the NPHDC, crude death rates are calculated both:

• as a proportion of the number of deaths occurring per 1,000 people released in the preceding 28 days (days 0–28 post-release) or 365 days (0–365 days post-release) (incidence proportion)
• as a proportion of per 1,000 person years (incidence rate). 

Readers should keep in mind that incidence proportions are not adjusted for the amount of time the group spent at risk of the event’s (death) occurring. Hence, the incidence proportion for a group calculated using the number of deaths in the first 28 days post-release will always be lower than, or equal to, the proportion for the same group calculated using the number of deaths in the first 365 days post-release.

Incidence rates, however, take into account the amount of time a group spent at risk of an outcome (death) and so allow for comparison, with the effect of differing times at risk removed. Incidence rates are presented in this report to assist in comparing crude death rates within 28 days and 365 days post-release.

Person-time crude death rates were calculated by dividing the number of deaths observed during the period (either 28 or 365 days post release) by the amount of time those released spent at risk of dying within the period.

Data cautions

Significance testing

Significance testing for the NPHDC was not undertaken due to the sample design and method of data collection. The collection was designed to be a census, capturing data on the entire population of interest at a given point in time. However, to date, this has not been achieved, as not all people in prison (especially entrants and dischargees) could be involved in data collection for various reasons. Of those who could be approached, some do not provide consent to participate.

The sample is therefore not probabilistic sampling, but rather nonprobability sampling using a convenience sample. In future NPHDC collections, the AIHW will aim to incorporate deeper levels of analysis where feasible.

Comparability of data in the 2022 NPHDC

Comparability of data between each NPHDC

There have been changes between each NPHDC collection – with indicators being added or deleted, and some changes made to definitions and data collection methods. Structural changes to questions and response options in the survey forms have also been made between collections in the NPHDC – meaning that certain aspects of the data may not be comparable between collections.

As the data in the NPHDC are based on a convenience sample, the coherence of the data across collection periods is difficult to assess, due to sampling and non-sampling biases present in each collection period.

Time-series analyses of data have been avoided in the 2022 NPHDC, except for certain data items from coherent external collections or that have good coherence across NPHDC collections.

For these reasons, caution should be used in comparing data for different years of the collection.

Comparability of NPHDC and general community data

This report compares NPHDC data with data for the general Australian community (where available) to provide additional context. These comparisons are made by sex, Indigenous identity, or age group where possible.

It should be noted, however, that the data are not directly comparable due to the different survey sampling or data collection techniques used, and to the substantial differences in the demographic profile of people in prison compared with those in the general community. For example, the age structure of the general prison population is lower than that of the general community, and the prison population has a higher proportion of males and First Nations people than the general community. No significance testing has been undertaken, so caution is advised when interpreting results.

Comparisons with the general community are at a national level, and include all states and territories, in line with the national focus of the NPHDC report. However, Victoria did not participate in the 2022 NPHDC, and data for that state are therefore not captured in 2022 data. Victorian data are, however, included in NPHDC indicators sourced from external data sources, including:

• the Productivity Commission (for prisoner employment)
• the Australian Institute of Criminology (for deaths in custody)
• the Department of Social Services (for deaths following custodial release).

Caution should therefore be used in interpreting comparisons between the NPHDC sample and the general community, and should be used as a guide only.