Key findings from the Drug Use Careers of Female Offenders study

Foreword | The Australian Institute of Criminology is undertaking research on the drug use careers of adult males, females and juveniles incarcerated in Australian prisons. The objective of the Drug use careers of offenders (DUCO) female study is to contribute to the empirical evidence about the interaction between drug use and criminal offending among incarcerated women. The results of the DUCO male study were released in 2003 and the results of the DUCO juvenile study are expected in 2005. The female study comprised 470 women who were incarcerated in prisons in six jurisdictions in Australia in 2003. The majority of these women offenders reported persistent offending and extensive drug use histories. Results have identified important differences in the patterns of drug use of women as compared to men. Risk factors for drug use have also been identified, including early exposure to drug and alcohol problems by family members, incarceration as a juvenile, mental health problems, and sexual and physical abuse. Understanding patterns in offending and drug use, and the connection between the two, may assist in the development of interventions and crime reduction strategies for women offenders.

Toni MakkaiActing Director

Illegal drug use has been identified as an important factor in criminal offending (White & Gorman 2000; Chaiken & Chaiken 1990). However, until very recently, studies of drug users and the relationship between drug use and crime have focused almost exclusively on male offenders (Willis & Rushforth 2003). Women who had drug or alcohol problems or who broke the law were considered somehow worse than men with the same problem. These women were seen as 'doubly deviant', as women who have not only violated legal conventions but who have also violated social norms about appropriate feminine behaviour (Broom & Stevens 1991; Lloyd 1995).

A growing body of research has examined female offenders and female drug users separately, in comparison to males, and in relation to their gendered social roles. A number of gender differences have been found that highlight the importance of examining women's drug use separately from men's. For example, addicted women are more likely to suffer from depression and other mental health problems, more likely to combine drugs and alcohol, and to begin and sustain an addiction through association with an addicted male associate (Pohl & Boyd 1992).

Many studies have noted the high prevalence of drug use among incarcerated women both in Australia and overseas (Hockings et al. 2002; Denton 1994; Kevin 1994; Ekstrand 1999). However, there is considerable debate as to whether the link between drugs and crime is causal. There are essentially three main explanatory models for the relationship between drugs and crime (White & Gorman 2000: 170):

  1. drug use leads to crime;
  2. crime leads to drug use; and
  3. drug use and crime are not causally related but are the result of a third factor.

Under the first scenario, drug use leads to crime due to the pharmacological properties of drugs, to acquire money to pay for drugs, or to the violence associated with the drug trade (Goldstein 1985). According to the second explanation, those who engage in crime are exposed to social situations in which alcohol and drugs are readily available and their use is reinforced. According to the third model, drug use and crime occur simultaneously due to a common cause, such as childhood abuse, early school failure, family characteristics, or neighbourhood disorganisation (White & Gorman 2000). Rather than being two separate populations, drug users and offenders coexist in some social groups and the motivating factors for drug use and crime are the same - excitement or risk-taking (Chaiken & Chaiken 1990; Denton 2001). The DUCO female study provides empirical data through which to explore the relationships between drug use and offending among women offenders.

About the DUCO female sample

The 470 women interviewed for this study were incarcerated in prisons in South Australia, Western Australia, Victoria, Queensland, Tasmania and Northern Territory. These offenders included 128 Indigenous women and 340 non-Indigenous women. Offenders were younger on average than the general adult population, had lower education levels, and were disproportionately single. They were a relatively impoverished group with 30 per cent living in public housing and five per cent living on the street prior to entering prison. Indigenous offenders were younger and had lower levels of education than non-Indigenous offenders. Sixty-nine per cent had been living in public housing or on the street prior to entering prison as compared to 22 per cent of non-Indigenous women.

Face-to-face interviews were conducted with offenders using a structured questionnaire. Due to the fact that prisons for women in Australia hold a relatively small number of inmates at any one time, a census of the sentenced prison population in each jurisdiction was attempted (this was not always possible for a variety of reasons, such as difficulty gaining access to prisoners and labour disputes). Women who were a danger to themselves or others and those who were non-English-speaking were also excluded. A total of 561 offenders were approached for an interview and 471 agreed to participate, giving a total response rate of 84 per cent. Only one failed to complete the interview. This compares to 996 female prisoners in the six jurisdictions in the prison census conducted by the Australian Bureau of Statistics on 30 June 2003. The DUCO study and the prison census are not directly comparable since both represent a snapshot at a single point in time, which can be expected to fluctuate.

Table 1 : Demographic characteristics of offenders
  n per cent
Source: Australian Institute of Criminology, DUCO female survey, 2003 [computer file]
Under 20 years 13 3
20-24 86 18
25-29 99 21
30-34 77 16
35-39 82 17
40 and older 113 24
Total 470 100
Mean age 33.1  
Primary school 109 23
Year 10 107 23
Apprenticeship 12 3
Year 12 31 7
TAFE/technical college 164 35
University 45 10
Total 468 100
Age left school (mean) 15.6  
Housing prior to prison
Rented/owned house/apartment 267 58
Public housing 137 30
Street 22 5
Someone else's home 14 3
Other 19 4
Total 459 100
Have children 336 71

History of offending

The DUCO study has several advantages over previous research on women offenders and their drug use as it:

  • covers a wide range of offence types in offenders' criminal history;
  • covers a wide range of drug types in offenders' drug use history;
  • covers most jurisdictions in Australia; and
  • is a relatively large sample of offenders.

While the ability to measure history of offending is a major advantage of the DUCO study, lengthy criminal histories can complicate the classification of offenders for analytical purposes. In this analysis, offenders are classified according to:

  • the main offence for which they are serving their current sentence of incarceration organised according to ABS counting rules; and
  • past offending.

'Regular' offending is used as the primary basis for classification of offenders in this analysis and was self-defined by offenders. Information about criminal careers provides a more accurate and complete picture of offending than relying simply on the primary offence that resulted in the current incarceration, since a majority of offences do not come to the attention of police. In fact, 70 per cent of women who committed burglaries, 44 per cent of those involved in fraud or forgery, and over 90 per cent of those involved in drug buying and selling said the police never or rarely found out about these crimes. However, a substantial percentage of these offenders had been convicted of a crime and incarcerated previously: 43 per cent had been in adult prison previously and 11 per cent had been in prison at least five times before.

The complexity of their offending patterns and the limits imposed by a relatively small sample means that very few offenders could be classified according to a single crime type. The principal focus of this analysis will be on the following categories of offenders:

  • regular property offenders (excluding women who had any regular involvement in violent offending or sex work);
  • regular violent offenders (whether or not they were also involved in any other offence types);
  • regular sex workers (whether or not they were also involved in any other offence types); and
  • regular drug offenders (whether or not they were also involved in any other offence types).

A majority of women were regular property and drug offenders:

  • three-quarters considered themselves to be 'regular' offenders;
  • half were regular property offenders (burglary, stealing without break-in, trading in stolen goods, vandalism and fraud);
  • 62 per cent were regular drug sellers or buyers;
  • 13 per cent were regular sex workers; and
  • 10 per cent were regular violent offenders (assault, armed and unarmed robbery).

The most common types of regular offending by individual offence type were buying drugs, stealing without break-in, selling drugs, trading in stolen goods, fraud and burglary (Figure 1). Assault had been committed by half of all offenders, but rarely escalated to regular offending. Escalation from 'ever' to 'regular' offending is an indicator of progression to more frequent offending. Escalation was most common for illegal drug buyers (91 per cent of those who ever bought drugs became regular drug buyers), drug sellers (72 per cent) and women working in the sex trade (74 per cent). However, just 18 per cent of offenders had ever engaged in sex work and 13 per cent became regular sex workers.

Figure 1 : Lifetime prevalence of offending and regular offending (per cent)

Source: Australian Institute of Criminology, DUCO female survey, 2003 [computer file]

Drug use

The majority of offenders reported using illegal drugs. Eighty per cent had experimented with drugs, 62 per cent were regular drug users in the six months prior to their arrest, and 39 per cent of all offenders were using more than one type (Figure 2). This is many times higher than the 14 per cent of women in the general population who reported recent drug use (AIHW 2002). In terms of drug type by offenders:

  • 40 per cent used cannabis;
  • 37 per cent used amphetamines illegally;
  • 27 per cent used heroin;
  • 15 per cent used benzodiazepines illegally; and
  • 6 per cent used cocaine.

Figure 2 : Lifetime prevalence and current regular use of illegal drugs (per cent)

1 Includes hallucinogens, ecstasy, street methadone and morphine

Source: Australian Institute of Criminology, DUCO female survey, 2003 [computer file]

Escalation from experimentation to regular drug use was reported by 78 per cent of offenders who ever tried drugs and 60 per cent of multiple drug users.

Temporal order of drug use and offending

Whether drug use 'leads' to crime or the reverse has been the subject of debate. Much of the research focusing on male offenders has found that criminal activity tends to precede illegal drug use, but that offending, in particular property crime, escalates as drug use increases (Makkai & Payne 2003). Somewhat different results were found for women in this study. Figure 3 illustrates the drug/crime career by type of offender. Regular fraud offenders were separated out from other property offenders due to their later start into drug use and offending:

  • initiation into illegal drugs preceded any offending for regular fraud offenders, other types of property offenders, and sex workers;
  • violent offenders began their crime/drug careers with first and regular property offending, followed by drug use prior to any violent offence;
  • fraud offenders began using illegal drugs and became regular drug users many years prior to committing their first fraud offence, suggesting a strong causal link between drug use and offending; and
  • women involved in the sex trade became regular drug users many years prior to regular property offending, and prior to first and regular sex work (age 19.5 and 20.4).

Figure 3 : Mean age progression of onset and regular offending and drug use

Note: Drug use includes cannabis, heroin, cocaine, hallucinogens, ecstasy, street methadone, and illegal use of morphine, amphetamines and benzodiazepines

Estimates include regular offenders only (property n=39, violent n=48, sex workers n=62, fraud n=91)

Source: Australian Institute of Criminology, DUCO female survey, 2003 [computer file]

For all groups of offenders, with the exception of fraud offenders, there is a brief period, between about 13 and 16, when they are initiated into drug use, become regular users, and commit their first offence. For fraud offenders, drug use happens at a later age and offending several years later than other types of offenders. Property offenders commit their first property crime after trying drugs but before regular drug use. This differs from male property offenders who tend to commit their first offence prior to using drugs (Makkai & Payne 2003: 44). Female violent offenders are more similar to males whereby initiation into offending preceded first use of illegal drugs by almost two years. However, unlike males, female violent offenders became regular drug users more than two years prior to their first violent offence. Males began violent offending at the same time as regular drug use (Makkai & Payne 2003: 61).

In the DUCO male study, 17 per cent of drug users began using drugs prior to committing any crime, 29 per cent began drugs and crime in the same year, and 54 per cent began offending prior to drug use (Makkai & Payne 2003). Among women the pattern is different:

  • 35 per cent had used illegal drugs prior to the first offence;
  • 31 per cent began using illegal drugs and began offending within the same year; and
  • 34 per cent began offending prior to any drug use.

This suggests that, for a substantial portion of female offenders, illegal drug use plays a role in shaping onset into a criminal career. Fraud offenders, sex workers and non-regular offenders are the groups most likely to have begun drug use prior to committing their first offence (Figure 4). Women involved in the sex trade, for example, tended to begin sex work after becoming regular users of amphetamines and heroin. Violent offenders were most likely to have begun drug use after the onset of offending and most closely resemble male offending and drug use patterns.

Figure 4 : Mean age progression for illegal drug use and crime

Source: Australian Institute of Criminology, DUCO female survey, 2003 [computer file]

'Non-regular' offenders differed from 'regular' offenders in important ways: they were more likely to be currently incarcerated for murder, assault or drug offences but had lower rates of illicit drug use. Just 15 per cent were regular drug users at the time of their arrest compared to 78 per cent of regular offenders.

Links between drugs and crime

The DUCO female study provides additional evidence of a connection between drug use and criminal offending, including:

  • 42 per cent of offenders were under the influence of drugs at the time of the offence for which they were incarcerated;
  • half of property offenders, and 87 per cent of those with a preference for heroin, attributed their crimes to the need to obtain money to buy drugs;
  • 44 per cent of women whose main current offences were burglary or theft and 29 per cent of those who were incarcerated for robbery were under the influence of heroin at the time of the offence;
  • amphetamines were used at the time of the offence by 35 per cent of women incarcerated for robbery or driving offences, and 30 per cent of those whose main offence was burglary or theft;
  • property offenders were actively involved in the drug market in the form of buying or selling drugs; and
  • 55 per cent of offenders were drug-dependent.

This is clear evidence of a link between drug use and crime. However, is there a portion of all crimes that can be said to be caused by crime? A measure of causation was derived in this study from a combination of addiction or intoxication at the time of the current offence, for offenders who stated that the reason they committed the offence was related to drugs or alcohol. In total, 41 per cent of offenders causally attributed their offending to their drug and alcohol abuse: 31 per cent to illegal drugs, nine per cent to alcohol and one per cent to both. Men in the DUCO male study were more likely to attribute their offending to both illegal drugs and alcohol (12 per cent), less likely to attribute their offending to drugs alone (18 per cent), and equally likely to say alcohol alone was the cause of their offending at nine per cent (Makkai & Payne 2003).

Risk factors for drug use and offending

A number of factors can influence an offender's drug use and criminal career. Current research identifies common risk factors for both drug use and criminal offending among women, which include sexual and physical abuse, mental health conditions, a history of family problems, and criminal behaviour as a juvenile (McClellan, Farabee & Crouch 1997; Browne, Miller & Maguin 1999). Women in the DUCO study report that:

  • 87 per cent were victims of sexual, physical or emotional abuse in either childhood (63 per cent) or adulthood (78 per cent); the majority were victims of multiple forms of abuse;
  • 60 per cent had mental health problems while growing up, and 62 per cent of these women said these problems significantly interfered with their lives;
  • 44 per cent grew up in families with alcohol problems, and 26 per cent in families with drug problems;
  • 50 per cent were often in trouble at school; and
  • 17 per cent spent time in juvenile detention.
Table 2 : Risk factors for drug dependency and regular offending (per cent)
  Drug dependent Property offender Violent offender Sex work
Yes No Yes No Yes No Yes No
* statistically significant, chi square, p
Source: Australian Institute of Criminology, DUCO female survey, 2003 [computer file]
Child abuse 71 54* 68 61 69 63 85 60*
Adult abuse 84 70* 76 48 73 78 90 76*
Mental health problems 68 51* 61 60 75 58* 76 58*
Juvenile detention 23 10* 22 15* 40 14* 20 17
Family had drug problems 38 11* 35 19* 33 25 53 42
Family had alcohol problems 50 36* 48 40 58 42* 34 24
(n) (257) (213) (160) (310) (48) (422) (62) (408)

These risk factors are correlated with drug dependency and offending (Table 2):

  • rates of childhood and adult abuse were significantly higher among drug-dependent women and those involved in the sex trade;
  • physical abuse in childhood was a predictor of violent offending;
  • mental health problems were correlated with drug dependency, violent offending and involvement in the sex trade;
  • early onset offending, as indicated by having spent time in juvenile detention, was related to drug dependency and property and violent offending as adults;
  • drug-dependent women and property offenders were significantly more likely to have grown up in families with drug problems; and
  • drug-dependent women and violent offenders were more likely to have grown up in families with alcohol problems.

Indigenous offenders

Indigenous women were incarcerated for somewhat different offences as compared to non-Indigenous women: they were more likely to be serving their current prison sentence for a violent offence (57 per cent compared to 21 per cent of non-Indigenous women), and less likely to be serving a sentence for property (21 per cent compared to 35 per cent) or drug offences (two per cent compared to 18 per cent). Difference are also shown in the drug use of these two groups of women offenders: non-Indigenous women were more likely than Indigenous women to be regular users of drugs other than cannabis, and more likely to be multiple drug users. Indigenous women, on the other hand, had higher levels of alcohol and cannabis use.

The higher prevalence of alcohol use among Indigenous women results in a higher level of alcohol dependency and attributions that their criminal offending was related to alcohol use. For example:

  • 60 per cent of Indigenous women were under the influence of alcohol at the time of the offence compared to 16 per cent of non-Indigenous women;
  • 54 per cent of Indigenous offenders were dependent on alcohol (or a combination of alcohol and drugs) compared to 17 per cent of non-Indigenous offenders; and
  • Indigenous offenders attributed their criminal offending almost equally to alcohol (24 per cent) or drugs (21 per cent) while non-Indigenous women attributed their offending primarily to illegal drugs (33 per cent) (Table 3).
Table 3 : Model attributions for intoxication and addiction by Indigenous status (per cent)
  Indigenous Non-Indigenous Total
- fewer than 5 cases
* difference between Indigenous and non-Indigenous offenders are statistically significant, chi square, p
Source: Australian Institute of Criminology, DUCO female survey, 2003 [computer file]
No attribution 52 62* 59
Illegal drugs 21 33* 31
Alcohol 24 2* 9
Both - 1 1
Total 100 100 100

Policy implications

Risk factors for drug dependency and persistent offending among these incarcerated offenders include victimisation, mental health problems, drug and alcohol abuse among family members, and delinquency. These results have important implications for the development of intervention and treatment programs for women and the prevention of crime and drug use. In order to be effective, interventions and treatment must take into account the following:

  • whole-of-government approaches and inter-agency cooperation to ensure that the range of factors that can lead to drug use and offending are addressed;
  • the prevention of drug dependency through rapid intervention with drug users and effective drug treatment programs;
  • early interventions with families and children of high-risk women in order to help avert intergenerational drug abuse and offending; and
  • programs targeted specifically to women and their personal histories and drug use patterns.

The results of this study suggest that treatment for women offenders needs to consider a wide range of issues, and must include agencies that normally lie outside the criminal justice system, such as drug and alcohol services, mental health services, child abuse and domestic violence treatment, and family services. Interventions based on empirical evidence of patterns in women's offending and drug use histories could help divert novice drug users from becoming dependent and help avoid harms associated with drug dependency, including criminal offending.


The DUCO project is funded by the Australian Government Attorney-General's Department under the National Illicit Drug Strategy. It represents a major partnership between the Attorney-General's Department, the Australian Institute of Criminology and the corrections agencies in the participating jurisdictions.


  • Australian Institute of Health and Welfare (AIHW) 2002. 2001 National drug strategy household survey: first results. Canberra: Australian Institute of Health and Welfare
  • Broom D & Stevens A 1991. Doubly deviant: women using alcohol and other drugs. International journal on drug policy vol 2: 25-27
  • Browne A, Miller B & Maguin E 1999. Prevalence and severity of lifetime physical and sexual victimization among incarcerated women. International journal of law and psychiatry vol 22: 301-322
  • Chaiken J & Chaiken M 1990. Drugs and predatory crime. In M Tonry & JQ Wilson (eds) Drugs and crime. Chicago: University of Chicago Press, pp 241-320
  • Denton B 2001. Dealing: women in the drug economy. Sydney: University of New South Wales Press
  • Denton B 1994. Prison, drugs and women: voices from below. Report for the National campaign against drug abuse: research into drug abuse grants
  • Ekstrand L 1999. Women in prison: issues and challenges confronting US correctional systems. Washington DC: United States General Accounting Office
  • Goldstein P 1985. The drugs/violence nexus: a tripartite conceptual framework. Journal of drug issues vol 15: 493-506
  • Hockings B, Young M, Falconer A & O'Rourke P 2002. Queensland women prisoner's health survey. Brisbane: Queensland Department of Corrective Services
  • Kevin M 1994. Women in prison with drug-related problems, part I: background characteristics. Research publication no 32. Sydney: New South Wales Department of Corrective Services
  • Lloyd A 1995. Doubly deviant, doubly damned: society's treatment of violent women. London: Penguin
  • Makkai T & Payne J 2003. Drugs and crime: a study of incarcerated male offenders. Research and public policy series no 52. Canberra: Australian Institute of Criminology
  • McClellan D, Farabee D & Crouch B 1997. Early victimization, drug use and criminality. Criminal justice and behavior vol 24: 455-476
  • Polh J & Boyd C 1992. Female addiction: a concept analysis. In T Mieczkowski (ed) Drugs, crime and social policy: research, issues and concerns. Boston: Allyn & Bacon, pp 138-152
  • White H & Gorman DM 2000. Dynamics of the drug crime relationship. In G Lafree (ed.) Criminal justice 2000: volume 1: the nature of crime: continuity and change. Washington DC: National Institute of Justice
  • Willis K & Rushforth C 2003. The female criminal: an overview of women's drug use and offending behaviourTrends and issues in crime and criminal justice no 264. Canberra: Australian Institute of Criminology.

About the authors

Holly Johnson is Visiting Fellow at the Australian Institute of Criminology