- Email: [email protected]
Substance use, dependence and treatment seeking in the United States and Australia: A cross-national comparison
Drug and Alcohol Dependence 81 (2006) 149–155
Substance use, dependence and treatment seeking in the United States and Australia: A cross-national comparison Maree Teesson a,∗ , Andrew Baillie b , Michael Lynskey c , Barry Manor a , Louisa Degenhardt a a
National Drug and Alcohol Research Centre, University of New South Wales, Sydney, NSW 2052, Australia b Department of Psychology, Macquarie University, Sydney, Australia c Washington University in St. Louis, Mo, USA. Received 29 November 2004; received in revised form 10 June 2005; accepted 21 June 2005
Abstract Aims: To compare the prevalence of alcohol and drug use, dependence and treatment seeking in the United States of America and Australia. Design: Two cross-sectional national surveys assessing substance use and DSM-IV substance dependence in the USA and Australia. Setting and participants: Age-matched cohorts (18–54 years old) were selected from nationally representative Australian (National Survey of Mental Health and Well-being, 1997, n = 7570) and American (National Comorbidity Survey, 1992, n = 7423) household surveys. Measurements: Both studies utilised a structured interview based on the Composite International Diagnostic Interview (CIDI). Findings: The 12-months prevalence of alcohol use was substantially higher in Australia (77.2%) than in the U.S. (46.3%) and the rates of alcohol dependence were also higher in Australia, although rates of alcohol dependence conditional on use were similar (6.8 and 6.5%, respectively). In contrast, although rates of use of drugs were similar in the two countries, rates of drug dependence and the probability of dependence conditional on use were higher in Australia than in the U.S. Importantly, the absence of significant interactions between correlates of alcohol and drug use disorders and country indicated that the influence of these factors was consistent across the two countries. Conclusions: Despite relatively similar cultural influences in Australia and the U.S. interesting cross-national differences emerged in the use of alcohol and drug dependence among those who used drugs and treatment seeking among people diagnosed with dependence. The cross-national generalizability of the associations between common correlates and rates of alcohol and drug use and dependence indicates that similar process of vulnerability to dependence may be operating in the two countries. Future research could usefully exploit these cross-national differences to help elucidate the cultural and structural factors influencing drug use, the development of dependence and treatment seeking. © 2005 Published by Elsevier Ireland Ltd. Keywords: Alcohol and drug use; Dependence; Treatment-seeking; Cross-national comparison; Epidemiology
1. Introduction Substance abuse and dependence leads to a significant burden on many societies. Cross-national comparisons are important in understanding this burden. The introduction of the Diagnostic and Statistical Manual of Mental Disorders, 3rd Edition (American Psychiatric Association, 1980) in 1980 revolutionised both psychiatry and psychiatric
∗
Corresponding author. Tel.: +61 2 9385 0333; fax: +61 2 9385 0222. E-mail address: [email protected] (M. Teesson).
0376-8716/$ – see front matter © 2005 Published by Elsevier Ireland Ltd. doi:10.1016/j.drugalcdep.2005.06.007
epidemiology and made international comparisons a possibility. Since then there have been a number of large-scale epidemiological studies of the prevalence and correlates of psychiatric conditions including drug and alcohol dependence. The first such major study was the ECA, conducted in five states, and involving structured interviews with 20,000 individuals, across the United States (Robins and Regier, 1991). There have now been a number of replications of the ECA, conducted both in the U.S. (Kessler et al., 1994) and other countries (e.g. Ross, 1995). These studies have confirmed the general pattern of results evident from the ECA: alcohol and drug dependence are common within the
150
M. Teesson et al. / Drug and Alcohol Dependence 81 (2006) 149–155
general population and only a minority of those meeting diagnostic criteria seek treatment for their condition. One of the strengths of these investigations is the application of rigorous methodologies, which allows the potential for comparisons across studies and countries. Even though there have been an ever growing number of epidemiological surveys of substance use disorders, to date there have been relatively few cross-national comparisons of the prevalence and correlates of alcohol and drug use disorders. The few well-conducted comparisons, which have been undertaken consistently find evidence that drug dependence is more prevalent in the U.S. than other countries (Vega et al., 2002; Maxwell, 2003; Furr-Holden and Anthony, 2003). In contrast, a number of publications (Chen et al., in press; Anthony et al., 1994; Swift et al., 2001a,b; Degenhardt et al., 2001) indicate that cannabis dependence is more prevalent in Australia and New Zealand (Poulton et al., 1997) compared to American populations (Anthony, in press). Interpretation of these crude comparisons is compromised by differences between the samples (for example, birth cohort follow-up in New Zealand compared to cross-sectional sampling in the USA and differing ages of the samples). The present paper conducts new analyses of existing data to derive more directly comparative estimates of drug use disorders in Australia and the USA. The value of such cross-national comparisons has long been recognised (Kessler et al., 1997) and includes: assessment of the impact of existing health care systems and social and cultural factors on both the prevalence and service response to mental health and substance dependence. For example, the availability of different drugs or restrictions on access to alcohol across countries may affect prevalence of use and dependence. Similarly, health service utilisation by people with alcohol or drug use disorders may be influenced by the structure of health care services. The current study reports a cross-national comparison of the prevalence, correlates and treatment seeking for substance dependence in Australia and the United States of America. Comparison between these countries are of value as, although culturally the two countries appear quite similar (they are both relatively young countries, English is the first language of the majority of people in both countries and they both enjoy high standards of living and access to health care, there are also a number of important differences between the two countries: 1. Rates of alcohol use appear to vary between the two countries with alcohol use being more accepted and widespread in Australia than in the United States (Maxwell, 2003). 2. Legal and policy approaches towards the reduction of drug use and the avoidance of drug related harm vary between the two countries: compared with a more prohibitionist stance in the United States, Australian alcohol and drug health service policy has more of a focus on harm reduction. Specific differences between the two countries include (but are not limited to) a higher legal drinking age in many of the United States (21 in USA versus 18 in Aus-
tralia) and the more widespread availability of needle and syringe programs in Australia than in the United States (Drucker et al., 1998). 3. Substantial differences in the organisation and funding of health care services with Australia having a universal health care system funded through taxation while the United States system relies more heavily on private funding. The aim of this paper is to present the 12 months DSM-IV prevalence of alcohol and drug use disorders in the Australian general population, and to compare these findings with the United States data from the National Comorbidity Survey (NCS).
2. Method This study presents a new analysis of data from both the NCS (Kessler et al., 1994) and the NSMHWB (Andrews et al., 1999; Australian Bureau of Statistics, 1998) on the prevalence, correlates and treatment seeking associated with DSM-IV substance use and substance dependence in the United States of America and Australia. Detailed descriptions of the U.S. and Australian surveys are available elsewhere (Anthony et al., 1994; Teesson et al., 2000; Hall et al., 1998; Teesson et al., 2002). These studies have a number of features that make them ideal for conducting a cross-national comparison. They were both based on personal interviews with a large and representative sample of the household population. The NCS interviewed a sample of 8098 people aged 15–54 from throughout the contiguous United States and the NSMHWB interviewed a total of 10,641 people aged 18 years or older from across Australia. Both of these studies achieved similar – and high – response rates (Australia 78.1%; USA 82.4%) and they both employed sophisticated methods of weighting to adjust for the complex sampling frames used. Similar data collection instruments were used (CIDI 1.1 in the NCS and CIDI 2.1 in the NSMHWB), which made it possible to derive standardised criteria for DSM disorders of anxiety, affective and alcohol and drug abuse and dependence. Both studies collected similar information on a range of socio-demographic features and on patterns of treatment seeking among respondents. There were some important differences in the two studies. These differences are outlined below, along with the adjustments conducted to allow comparisons between the NCS and NSMHWB. 1. The age range of respondents in the NSMHWB was 18 years and over, while the NCS included persons aged 15–54 years. Data analysis was restricted in both surveys to those in the age range 18–54. There were 7423 in the NCS and 7571 in the NSMHWB in this age range. 2. The NCS used the CIDI 1.1, which provided DSM-IIIR diagnoses, the NSMHWB used the CIDI 2.1 to provide DSM-IV diagnoses. In the current study, the original
M. Teesson et al. / Drug and Alcohol Dependence 81 (2006) 149–155
DSM-III-R diagnosis scripting for the NCS was modified to provide DSM-IV diagnoses of drug and alcohol dependence, after consultation with the NCS data analysis manager. The principle differences between DSM versions in the diagnosis of substance dependence were a reduction from nine criteria in DSM-III-R to seven in DSM-IV. Underlying these changes were the combination of two items probing withdrawal in DSM-III-R, and the transfer of criterion four (failure to fulfil major role obligations) from dependence to abuse. The original DSM-III-R diagnosis scripting was modified by removing criterion four from the past year diagnosis section.1 The approximation for DSM-IV diagnosis was validated by comparison of DSM-III-R and DSM-IV prevalence of substance dependence. As expected, dependence rates were slightly lower using DSM-IV diagnostic criteria (relative to the original DSM-III-R rates in the NCS), reflecting a more specific distinction between substance dependence and abuse. 3. The NSMHWB assessed diagnoses in the past 12 months, whereas the NCS inquired about both lifetime and 12months prevalence. Comparisons were thus restricted to 12-months diagnoses. 4. The categories of substances for which information was acquired differed in each survey. Cocaine use was not evaluated in the NSMHWB as only those drugs with an expected prevalence of greater than 1% were included and the preceding National Household Survey (Commonwealth Department of Health and Family Services, 1996) indicated prevalence of cocaine use in Australia was less than 1%. Analyses were restricted to categories of drug use that were common across the two surveys, and that are therefore reported in the current analyses. These were • Cannabis (marijuana and hashish). • Stimulants (3,4-methylenedioxy-methamphetamine (MDMA), amphetamines and other stimulants obtainable by medical prescription). • Sedatives (including barbiturates, tranquillisers and other sedatives). • Opioids, including heroin and opium, and opiate analgesics which can be obtained on medical prescription. • Drug use disorder data is presented on subjects who had used drugs more than five times in the past 12 months. 2.1. Analysis Prevalence estimates for substance use and dependence in both surveys were derived using a complex estimation procedure to account for the stratified multistage survey design. Part I weightings were used in the NCS for all analyses except those involving country of birth and employment status variables The Strata and ECU weightings were used in 1 The combination of criteria 8 and 9 in this section was not possible, since Criterion 9 was not available.
151
the SUDAAN jack-knife analysis to adjust for the complex survey design. The person-weightings and replicate weights in the NSMHWB were divided by the estimated Australian adult population size to adjust the proportional sample size for direct comparability with the NCS data. Odds ratios (based upon weighted estimates) were used to examine the strength of bivariate associations between substance use/dependence and other variables. All covariates for modelling the probability of alcohol or drug dependence were chosen because they have been shown previously to be associated with alcohol or drug use dependence. These include the social demographic variables of age, gender, marital status, employment, country of birth, education and urban location of living.
3. Results 3.1. Prevalence of substance use in the past 12 months Table 1 shows that recent alcohol use is higher in Australian adults, compared with Americans. Within the Australian sample, 77.2% of respondents aged 18–54 reported consumption of at least 12 drinks of alcohol in the year preceding the NSMHWB (1997), compared with 46.3% of Americans in the 12 months preceding the NCS (1990/1991). The prevalence of use of the drugs assessed was considerably lower than alcohol use, but similar, in both countries. An estimated 10.8% of Australians and 11.9% of Americans used at least one drug from the other classes (cannabis, stimulants, sedatives and opioids) more than five times in the past year. The most commonly used drug, both in Australia and the USA, was cannabis, with 9.8% of Australians and 8.1% of Americans reporting use on at least five occasions in the past year. The prevalence of use of stimulants, sedatives and opioids was around 0.3–1.3% for Australians and 2.4–2.5% for Americans. 3.2. Twelve-months prevalence of substance dependence In Australia, 5.3% of the sample met DSM-IV criteria for alcohol dependence, and 2.7% met criteria for drug dependence within the last 12 months. In the USA, these figures were 3.1 and 1.1%, respectively. Aside from alcohol, cannabis accounted for the highest levels of drug dependence. The estimated past-year prevalence of cannabis dependence in Australia was 2.0%, while in the USA it was 0.5%. Of the remaining classes of drugs, sedative dependence was next in rank in Australia, while in the USA this position was occupied by opioid dependence (see Table 1). The conditional prevalence of dependence (dependence among users only) on alcohol was similar in Australia (6.8%) and the USA (6.5%). However, for drugs a different picture emerged, with Australians in general having higher conditional prevalence’s than their American counterparts.
M. Teesson et al. / Drug and Alcohol Dependence 81 (2006) 149–155
152
Table 1 Twelve-months prevalence (%, S.E.) of substance use and DSM-IV dependence in the United States of America and Australia Use in past 12 months
Alcohol Any drug Cannabis Stimulants Sedatives Opioids
DSM-IV dependence
DSM-IV dependence among users
USA
Australia
USA
Australia
USA
Australia
46.3 (1.1) 11.9 (0.7) 8.1 (0.5) 2.5 (0.3) 2.4 (0.3) 2.5 (0.3)
77.2 (1.1) 10.8 (0.5) 9.8 (0.5) 1.3 (0.2) 1.1 (0.1) 0.3 (0.1)
3.1 (0.4) 1.1 (0.2) 0.5 (0.1) 0.2 (0.1) 0.2 (0.1) 0.3 (0.1)
5.3 (0.5) 2.7 (0.3) 2.0 (0.2) 0.3 (0.1) 0.5 (0.1) 0.3 (0.1)
6.5 (0.8) 9.0 (1.7) 6.5 (1.5) 7.8 (2.7) 8.4 (3.8) 9.1 (2.9)
6.8 (0.6) 22.7 (2.6) 20.9 (2.2) 22.4 (5.0) 25.9 (4.9) 43.1 (15.9)
The overall prevalence of alcohol and drug dependence was higher among males than among females. Specifically, 7.5% of males and 3.0% of females met DSM-IV criteria for alcohol dependence in Australia, and 3.7% of males and 1.6% of females met the criteria for drug dependence. In the USA, the prevalence of alcohol dependence was 5.1% for males and 1.2% for females, while for drugs it was 1.2% for males and 1.0% for females. With the exception of sedatives, all substances were more often used by males than by females, in both countries. 3.3. Correlates of substance dependence in Australia and the USA Associations between substance dependence and a range of socio-demographic characteristics were explored. Two multiple logistic regression analyses were undertaken with alcohol and drug dependence as dependent variables and nation (USA versus Australia) and other demographic variables as independent variables. Table 2 shows the results of these analyses for alcohol and drug use separately. Interactions between country and other correlates were tested but are not reported as they were non-significant and did not add to the explanation of the models.
The final model for alcohol dependence (−2 log likelihood 4819; χ = 384.2, p < 0.000) is presented in Table 2. The odds ratios presented are adjusted for the effect of all other variables in the model. Younger respondents were more likely than those over 44 years to meet criteria for alcohol dependence. Males were three times more likely than females to be alcohol dependent (OR = 3.0). Those who were never married were more likely to have alcohol dependence (OR = 1.7). Persons with less than 12 years education were more likely (OR = 1.4) than those with 12 years or more to be dependent on alcohol. Australians were 1.5 times more likely to be alcohol dependent than Americans. The final model for drug dependence (−2 log likelihood 2515; χ = 282.3, p < 0.000) is presented in Table 2. Younger respondents were also more likely than those over 44 years to meet criteria for drug dependence. Males were twice as likely as females to be drug dependent (OR = 2.2). Those who were never married were more likely to have a drug dependence (OR = = 1.8). Persons with less than 12 years education were more likely (OR = 1.7) than those with 12 years or more to be dependent on alcohol. Those who were unemployed were twice as likely to be dependent on drugs (2.2). Australians were nearly three times as likely to meet criteria for drug dependence than Americans. 3.4. Treatment seeking
Table 2 Adjusted odds ratios for alcohol and drug dependence, United States vs. Australia, full model Variable
Age 18–24 25–34 35–44 45–54 Male Never Married Urban Less education (<12 years) Unemployed Nationa
Alcohol dependence
Drug dependence
Adjusted odds ratio
95% CI
Adjusted odds ratio
95% CI
2.9* 1.9* 1.7* 1.0
2.1–4.0 1.5–2.6 1.3–2.3 –
5.2* 3.7* 2.4 1.0
3.0–9.1 2.2–6.2 1.4–4.1 –
3.0* 1.7* 1.0 1.4*
2.5–3.6 1.4–2.1 0.8–1.2 1.2–1.7
2.2* 1.8* 1.3 1.7*
1.7–2.8 1.3–2.4 1.0–1.8 1.3–2.2
0.8 1.5*
0.7–1.0 1.3–1.8
2.2* 2.9*
1.7–2.9 2.2–3.8
a Coded as Australia = 1 and USA = 0 odds ratios adjusted for all other covariates in the model: significance set at *p < 0.001.
Table 3 summarises logistic regression analyses in which treatment seeking was the dependent variable, and other socio demographic variables (including nation) were entered as the independent variables. Interaction effects were tested but did not add to the explanatory power of the model and are therefore not reported. The results of the analysis for those with alcohol and drug dependence revealed Australians were no more likely to seek professional help than Americans. Males were half as likely to seek help for alcohol dependence than females (OR = 0.46 95% CI 0.31–0.69). Males were even less likely to seek help for drug dependence than females (OR 0.33 95% CI 0.19–0.59) and 18–24 years old were less likely to seek help than those aged over 44 (OR 0.15 95% CI 0.07–0.32). Australians more often consulted family physicians or general practitioners, while Americans were almost equally likely to consult a general practitioner, a psychiatrist or a psychologist.
M. Teesson et al. / Drug and Alcohol Dependence 81 (2006) 149–155
153
Table 3 Twelve months treatment seeking (%, S.E.) among those with DSM-IV alcohol or drug dependence in the United States of America and Australia by gender Males Australia
Females USA
Australia
USA
Alcohol dependence General practitioner Psychiatrist Psychologist Other professional Hospital/drug unit admission Any treatment
6.8 (0.8) 1.7 (0.3) 1.4 (0.3) 3.8 (0.5) 0.3 (0.1) 9.2 (1.1)
2.0 (0.6) 1.7 (0.3) 1.3 (0.3) 3.1 (0.5) 0.5 (0.2) 7.1 (0.9)
11.8 (1.3) 2.3 (0.4) 2.4 (0.4) 7.8 (0.5) 0.3 (0.1) 16.2 (1.2)
4.4 (1.0) 2.8 (0.9) 2.7 (0.5) 6.4 (0.9) 0.6 (0.3) 12.9 (1.5)
Drug dependence General practitioner Psychiatrist Psychologist Other professional Hospital/drug unit admission Any treatment
12.0 (2.6) 4.3 (1.1) 2.6 (0.6) 9.2 (2.9) 0.5 (0.3) 17.4 (3.6)
3.6 (1.5) 3.2 (0.9) 4.3 (1.0) 7.6 (1.5) 1.8 (0.8) 15.2 (2.0)
24.5 (3.4) 6.0 (1.6) 7.6 (1.9) 16.3 (2.4) 1.0 (0.9) 33.4 (3.9)
7.4 (2.1) 5.6 (2.0) 5.7 (1.6) 10.8 (2.0) 2.5 (1.3) 21.4 (2.9)
4. Discussion The major finding from this comparison, with a few exceptions, is the consistency of rates and patterns of 12-months dependence on alcohol and drug dependence in the U.S. and Australia. Both surveys reported a higher prevalence of substance dependence in males than females, and the prevalence peaked in young adults and declined linearly with age. Importantly, the lack of any significant interaction between the common correlates of alcohol and other drug dependence and country indicates that the effects of correlates such as age, gender, and unemployment were similar in the two countries. The striking difference was the higher conditional prevalence for drug dependence in Australia compared to the USA, yet consistent rates of conditional prevalence for alcohol dependence. The differences in dependence rates between Australia and the USA remained when other demographic variables were accounted for in statistical models. The limitations of this cross-national comparison should be acknowledged. Firstly, the NSMHWB was not able to correct for non-response. However, the non-responders in the NCS had a higher rate of psychopathology and therefore the omission of non-responders in the Australian sample should bias against the finding of higher rates of conditional drug prevalence in Australia. Secondly, there is a 5-years gap between the Australian and U.S. samples (Australia 1997, USA 1992). Chen et al. (in press) have shown that age of initiation of cannabis use has reduced during the mid nineties in the U.S., and Compton et al. (2004) found some evidence suggestive of an increased occurrence of cannabis dependence in years 2000–2001 as compared to corresponding estimates 10 years ago. These changes may also have occurred in Australia, but are unlikely to fully explain the much higher rates of conditional drug prevalence in Australia compared to the USA. In most other respects, the findings of the NSMHWB were similar to those of the NCS. The NCS used the CIDI Version 1 to ascertain DSM-III-R diagnoses,
although data was collected so as to allow comparisons with DSM-IV. The NSMHWB used a slightly different version of the CIDI to ascertain DSM-IV diagnoses. Even so the differences in conditional drug prevalence were so large that these minor methodological differences cannot be expected to fully explain them. Despite the similar prevalence of alcohol and drug dependence in the U.S. and Australia, treatment-seeking patterns, although not rates of treatment seeking, differed. While specialist care was equally sought in the USA and Australia, individuals with substance dependence in Australia were more likely to seek care from general practitioners. Factors that may influence treatment seeking include health care infrastructure, degree of debility, socio-economic status and individual attitudes toward mental health. We were not able to measure such structural determinants in such a cross sectional survey, although they should form part of future work in the area. The analyses identified that several demographic correlates were related to the likelihood of having alcohol or drug dependence in both countries. Those who were married were less likely than those never married to have alcohol or drug dependence. The unemployed were more likely to have drug dependence than those in the labour force. Interestingly, unemployment was not correlated with alcohol dependence, which again may reflect its widespread use. Similarly, there were observed differences in the pattern of substance dependence according to level of education attained. There was no difference in the likelihood of having alcohol or drug dependence according to a person’s location of residence. This may reflect an increasing level of exposure to drugs in regional areas such that the risk of dependence is similar in urban and rural areas. Importantly, the absence of significant interactions between the above factors and country indicated that the influence of these factors was consistent across the two countries. The cross-national generalizability of the associations
154
M. Teesson et al. / Drug and Alcohol Dependence 81 (2006) 149–155
between these factors and rates of alcohol or drug use and dependence indicates that similar process of vulnerability to dependence may be operating in the two countries. While our results are consistent with previous findings regarding the correlates of alcohol and other drug dependence in Australia and the U.S. to the best of our knowledge this is the first time that potential differences in the magnitude of these associations have been formally tested. Finally, nation (Australia versus the USA) was associated with the likelihood of developing alcohol or drug dependence. Compared to those in the U.S., Australians are more likely to have alcohol or drug dependence. This difference remained after controlling for other demographic variables. The generally lower rate of conditional prevalence for alcohol use disorders indicates that non-dependent alcohol consumption among adult drinkers is more common than non-dependent illicit drug use. This is probably explained by the large number of older, moderate drinkers and the relative absence of older, moderate users of illicit drugs (Swift et al., 2001a,b). This was consistently found in both the Australian and U.S. samples. The similarity of conditional prevalence (dependence rate among alcohol users only) in the two countries suggests that the higher prevalence in Australia is simply a reflection of more widespread alcohol use in Australia. More Australians drink alcohol so more are dependent. The differences in conditional prevalence between the USA and Australia for illicit drug dependence are not so easily explained. The finding is consistent across drug classes and higher rates of drug dependence remain when other demographic variables are considered in statistical models. A limitation of this study is the lack of comparable data in the surveys on frequency and quantity of use. There are likely to have been differences between the two countries in consumption patterns, especially the consumption of the most widely used illicit drug in cannabis. One crucial issue is the delta-9-tetrahydrocannabinol (THC) content of cannabis products. The THC content of Australian cannabis products has not been systematically tested and the USA is the only country that has regularly collected data on the THC content of cannabis plants over the past several decades. The data shows only a small increase in THC content (Hall and Swift, 2000). Changes in the patterns of cannabis use may also increase the amount of THC being consumed. These changes include more regular use of more potent cannabis products such as the flowering tops of the stems, the greater use of water pipes and a possible increase in cannabis potency. Whether these changes occurred in Australia more so than the USA is unknown. Compton et al. (2004) suggested that increases in the THC concentration might account for apparent increases in prevalence of cannabis dependence in recent years. However, Chen et al. (in press) did not observe an increase in the risk of developing cannabis dependence problems soon after onset over the same time period. If the THC concentration has increased it may be that users compensate for increased
THC concentration (e.g., smoking fractions of a ‘joint’ as opposed to a complete ‘joint’) (Chen et al., in press). Thus, the relationship between use, patterns of use and dependence are not always clear.
5. Conclusions The drug dependence conditional prevalence rates differ considerably between the USA and Australia, yet the alcohol dependence conditional prevalence rates are very similar. The differences are so marked that the small differences in methodology are unlikely to explain the findings. The findings raise doubts as to the generalizability of research findings regarding risk factors cross-nationally and for the potential for targeted interventions based on identification of risk factors to be applicable cross-nationally.
Acknowledgements The National Survey of Mental Health and Well-being was a major piece of applied social research that involved the cooperation of a large number of individuals and organisations that we wish to thank. The design, development, and conduct of the survey were funded by the Mental Health Branch of the Commonwealth Department of Health and Family Services. The development and testing of the computerised survey instrument was undertaken by Gavin Andrews, Lorna Peters, and other staff at the Clinical Research Unit for Anxiety Disorders, and the WHO Collaborating Centre in Mental Health at St. Vincent’s Hospital. The design of the survey was overseen and approved by a technical advisory committee comprising: Professor A. Scott Henderson, Chair; Professor Gavin Andrews, Professor Wayne Hall, Professor Helen Herman, Professor Assen Jablensky and Professor Bob Kosky. The Australian Bureau of Statistics conducted the fieldwork and implementation of the Survey, the enumeration, compilation and initial analyses of the data. We thank Tony Cheshire, Gary Sutton and Marelle Lawson for their assistance. We thank Wendy Swift for statistical advice. We would also like to thank the many individuals who gave their time to participate in this survey. The NCS was funded by the National Institute of Mental Health (Grants R01 MH/DA46376 and R01 MH49098), the National Institute of Drug Abuse (through a supplement to R01 MH/DA46376) and the W.T. Grant Foundation (Grant 90135190). The NCS data was obtained from; Kessler, Ronald C. NATIONAL COMORBIDITY SURVEY, 1990–1992 [Computer file]. Conducted by University of Michigan, Survey Research Center. ICPSR ed. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [producer and distributor], 2000. We would like to acknowledge the assis-
M. Teesson et al. / Drug and Alcohol Dependence 81 (2006) 149–155
tance of the data manager in obtaining and using this information.
References Andrews, G., Hall, W., Teesson, M., Henderson, S., 1999. The Mental Health of Australians (2). Commonwealth Department of Health and Family Services, Canbera Australia. Anthony, J.C., Warner, L., Kessler, R., 1994. Comparative epidemiology of dependence on tobacco, alcohol, controlled substances, and inhalants: basic findings from the National Comorbidity Survey. Exp. Clin. Psychopharmacol. 2 (3), 244–268. Anthony, J.C., in press. The epidemiology of cannabis dependence. In: Roffman, R., Stephens, R. (Eds.). Cannabis dependence: Its Nature, Consequences and Treatment. Cambridge University Press, Cambridge. Australian Bureau of Statistics, 1998. National Survey of Mental Health and Wellbeing of Adults: Users’ Guide. Australian Bureau of Statistics, Canberra. Chen, C-Y., O’Brien, M.S., Anthony, J.S., 1999. Who becomes cannabis dependent soon after onset of use? Drug Alcohol Depend. Commonwealth Department of Health and Family Services, 1996. National Drug Strategy Household Survey, Survey report, 1995. Commonwealth Department of Health and Family Services, Canberra. Compton, W., et al., 2004. Prevalence of Marijuana use disorders in the United States: 1991–1992 and 2001 and 2002. JAMA 291, 2114. Degenhardt, L., Hall, W., Lynskey, M., 2001. Alcohol, cannabis and tobacco use among Australians: a comparison of their associations with other drug use and use disorders, affective and anxiety disorders, and psychosis. Addiction 96, 1603–1614. Drucker, E., Lurie, P., Wodak, A., Alcabes, P., 1998. Measuring harm reduction: the effects of needle and syringe exchange programs and methadone maintenance on the ecology of HIV. Aids 12 (Suppl. A), S217–S230. Furr-Holden, C.D.M., Anthony, J.C., 2003. Epidemiologic difference in drug dependence: a U.S.–U.K. cross-national comparison. Social Psychiatr. Psychiatr. Epidemiol. 38, 165–172. Hall, W., Teesson, M., Lynskey, M., Degenhardt, L., 1998. The 12-months prevalence of substance use and ICD-10 substance use disorders in Australian adults: findings from the National Survey of Mental Health and Well-being. Addiction 94 (10), 1541–1550.
155
Hall, W., Swift, W., 2000. The THC content of cannabis in Australia: evidence and implications. Aust. N. Z. J. Public Health 24, 503– 508. Kessler, R.C., Frank, R.G., Edlund, M., Katz, S.J., Lin, E., Leaf, P., 1997. Differences in the use of psychiatric outpatient services between the United States and Ontario [see comments]. N. Engl. J. Med. 336 (8), 551–557. Kessler, R.C., McGonagle, K.A., Zhao, S., Nelson, C.B., Hughes, M., Eshleman, S., Wittchen, H.U., Kendler, K.S., 1994. Lifetime and 12months prevalence of DSM-III-R psychiatric disorders in the United States. Results from the National Comorbidity Survey. Arch. Gen. Psychiatr. 51 (1), 8–19. Maxwell, J.C., 2003. Update: comparison of drug use in Australia and the United States as seen in the 2001 National Household Surveys. Drug Alcohol Rev. 22 (3), 347–357. Poulton, R.G., Brooke, M., Moffitt, T.E., Stanton, W.R., Silva, P.A., 1997. Prevalence and correlates of cannabis use and dependence in young New Zealanders. N. Z. Med. J. 110, 68–70. Robins, L.N., Regier, D.A. (Eds.), 1991. The Epidemiological Catchment Area Study. The Free Press, New York. Ross, H., 1995. DSM-III-R alcohol abuse and dependence and psychiatric comorbidity in Ontario: results from the Mental Health Supplement to the Ontario Health Survey. Drug Alcohol Depend. 39, 111–128. Swift, W., Hall, W., Teesson, M., 2001a. Cannabis use and dependence among Australian adults: results from the National Survey of Mental Health and Well-being. Addiction 96, 737–748. Swift, W., Hall, W., Teesson, M., 2001b. Characteristics of DSM-IV and ICD-10 cannabis dependence among Australian adults: results from the National Survey of Mental Health and Well-being. Drug Alcohol Depend. 63, 147–153. Teesson, M., Hall, W., Lynskey, M., Degenhardt, L., 2000. Alcohol and drug use disorders in Australia: implications of the National Survey of Mental Health and Well-being. Aust. N. Z. J. Psychiatr. 34, 206–213. Teesson, M., Lynskey, M., Manor, B., Baillie, A., 2002. The structure of DSM-IV cannabis use disorders in the community. Drug Alcohol Depend. 68, 255–262. Vega, W.A., Aguilar-Gaxiola, S., Andrade, L., Bijl, R., Borges, G., Caraveo-Anduaga, J.J., DeWit, D.J., Heeringa, S.G., Kessler, R.C., Kolody, B., Merikangas, K.R., Molnar, B.E., Walters, E.E., Warner, L.A., Wittchen, H.U., 2002. Prevalence and age of onset for drug use in seven international sites: results from the international consortium of psychiatric epidemiology. Drug Alcohol Depend. 68, 285–297.
Substance use, dependence and treatment seeking in the United States and Australia: A cross-national comparison Maree Teesson a,∗ , Andrew Baillie b , Michael Lynskey c , Barry Manor a , Louisa Degenhardt a a
National Drug and Alcohol Research Centre, University of New South Wales, Sydney, NSW 2052, Australia b Department of Psychology, Macquarie University, Sydney, Australia c Washington University in St. Louis, Mo, USA. Received 29 November 2004; received in revised form 10 June 2005; accepted 21 June 2005
Abstract Aims: To compare the prevalence of alcohol and drug use, dependence and treatment seeking in the United States of America and Australia. Design: Two cross-sectional national surveys assessing substance use and DSM-IV substance dependence in the USA and Australia. Setting and participants: Age-matched cohorts (18–54 years old) were selected from nationally representative Australian (National Survey of Mental Health and Well-being, 1997, n = 7570) and American (National Comorbidity Survey, 1992, n = 7423) household surveys. Measurements: Both studies utilised a structured interview based on the Composite International Diagnostic Interview (CIDI). Findings: The 12-months prevalence of alcohol use was substantially higher in Australia (77.2%) than in the U.S. (46.3%) and the rates of alcohol dependence were also higher in Australia, although rates of alcohol dependence conditional on use were similar (6.8 and 6.5%, respectively). In contrast, although rates of use of drugs were similar in the two countries, rates of drug dependence and the probability of dependence conditional on use were higher in Australia than in the U.S. Importantly, the absence of significant interactions between correlates of alcohol and drug use disorders and country indicated that the influence of these factors was consistent across the two countries. Conclusions: Despite relatively similar cultural influences in Australia and the U.S. interesting cross-national differences emerged in the use of alcohol and drug dependence among those who used drugs and treatment seeking among people diagnosed with dependence. The cross-national generalizability of the associations between common correlates and rates of alcohol and drug use and dependence indicates that similar process of vulnerability to dependence may be operating in the two countries. Future research could usefully exploit these cross-national differences to help elucidate the cultural and structural factors influencing drug use, the development of dependence and treatment seeking. © 2005 Published by Elsevier Ireland Ltd. Keywords: Alcohol and drug use; Dependence; Treatment-seeking; Cross-national comparison; Epidemiology
1. Introduction Substance abuse and dependence leads to a significant burden on many societies. Cross-national comparisons are important in understanding this burden. The introduction of the Diagnostic and Statistical Manual of Mental Disorders, 3rd Edition (American Psychiatric Association, 1980) in 1980 revolutionised both psychiatry and psychiatric
∗
Corresponding author. Tel.: +61 2 9385 0333; fax: +61 2 9385 0222. E-mail address: [email protected] (M. Teesson).
0376-8716/$ – see front matter © 2005 Published by Elsevier Ireland Ltd. doi:10.1016/j.drugalcdep.2005.06.007
epidemiology and made international comparisons a possibility. Since then there have been a number of large-scale epidemiological studies of the prevalence and correlates of psychiatric conditions including drug and alcohol dependence. The first such major study was the ECA, conducted in five states, and involving structured interviews with 20,000 individuals, across the United States (Robins and Regier, 1991). There have now been a number of replications of the ECA, conducted both in the U.S. (Kessler et al., 1994) and other countries (e.g. Ross, 1995). These studies have confirmed the general pattern of results evident from the ECA: alcohol and drug dependence are common within the
150
M. Teesson et al. / Drug and Alcohol Dependence 81 (2006) 149–155
general population and only a minority of those meeting diagnostic criteria seek treatment for their condition. One of the strengths of these investigations is the application of rigorous methodologies, which allows the potential for comparisons across studies and countries. Even though there have been an ever growing number of epidemiological surveys of substance use disorders, to date there have been relatively few cross-national comparisons of the prevalence and correlates of alcohol and drug use disorders. The few well-conducted comparisons, which have been undertaken consistently find evidence that drug dependence is more prevalent in the U.S. than other countries (Vega et al., 2002; Maxwell, 2003; Furr-Holden and Anthony, 2003). In contrast, a number of publications (Chen et al., in press; Anthony et al., 1994; Swift et al., 2001a,b; Degenhardt et al., 2001) indicate that cannabis dependence is more prevalent in Australia and New Zealand (Poulton et al., 1997) compared to American populations (Anthony, in press). Interpretation of these crude comparisons is compromised by differences between the samples (for example, birth cohort follow-up in New Zealand compared to cross-sectional sampling in the USA and differing ages of the samples). The present paper conducts new analyses of existing data to derive more directly comparative estimates of drug use disorders in Australia and the USA. The value of such cross-national comparisons has long been recognised (Kessler et al., 1997) and includes: assessment of the impact of existing health care systems and social and cultural factors on both the prevalence and service response to mental health and substance dependence. For example, the availability of different drugs or restrictions on access to alcohol across countries may affect prevalence of use and dependence. Similarly, health service utilisation by people with alcohol or drug use disorders may be influenced by the structure of health care services. The current study reports a cross-national comparison of the prevalence, correlates and treatment seeking for substance dependence in Australia and the United States of America. Comparison between these countries are of value as, although culturally the two countries appear quite similar (they are both relatively young countries, English is the first language of the majority of people in both countries and they both enjoy high standards of living and access to health care, there are also a number of important differences between the two countries: 1. Rates of alcohol use appear to vary between the two countries with alcohol use being more accepted and widespread in Australia than in the United States (Maxwell, 2003). 2. Legal and policy approaches towards the reduction of drug use and the avoidance of drug related harm vary between the two countries: compared with a more prohibitionist stance in the United States, Australian alcohol and drug health service policy has more of a focus on harm reduction. Specific differences between the two countries include (but are not limited to) a higher legal drinking age in many of the United States (21 in USA versus 18 in Aus-
tralia) and the more widespread availability of needle and syringe programs in Australia than in the United States (Drucker et al., 1998). 3. Substantial differences in the organisation and funding of health care services with Australia having a universal health care system funded through taxation while the United States system relies more heavily on private funding. The aim of this paper is to present the 12 months DSM-IV prevalence of alcohol and drug use disorders in the Australian general population, and to compare these findings with the United States data from the National Comorbidity Survey (NCS).
2. Method This study presents a new analysis of data from both the NCS (Kessler et al., 1994) and the NSMHWB (Andrews et al., 1999; Australian Bureau of Statistics, 1998) on the prevalence, correlates and treatment seeking associated with DSM-IV substance use and substance dependence in the United States of America and Australia. Detailed descriptions of the U.S. and Australian surveys are available elsewhere (Anthony et al., 1994; Teesson et al., 2000; Hall et al., 1998; Teesson et al., 2002). These studies have a number of features that make them ideal for conducting a cross-national comparison. They were both based on personal interviews with a large and representative sample of the household population. The NCS interviewed a sample of 8098 people aged 15–54 from throughout the contiguous United States and the NSMHWB interviewed a total of 10,641 people aged 18 years or older from across Australia. Both of these studies achieved similar – and high – response rates (Australia 78.1%; USA 82.4%) and they both employed sophisticated methods of weighting to adjust for the complex sampling frames used. Similar data collection instruments were used (CIDI 1.1 in the NCS and CIDI 2.1 in the NSMHWB), which made it possible to derive standardised criteria for DSM disorders of anxiety, affective and alcohol and drug abuse and dependence. Both studies collected similar information on a range of socio-demographic features and on patterns of treatment seeking among respondents. There were some important differences in the two studies. These differences are outlined below, along with the adjustments conducted to allow comparisons between the NCS and NSMHWB. 1. The age range of respondents in the NSMHWB was 18 years and over, while the NCS included persons aged 15–54 years. Data analysis was restricted in both surveys to those in the age range 18–54. There were 7423 in the NCS and 7571 in the NSMHWB in this age range. 2. The NCS used the CIDI 1.1, which provided DSM-IIIR diagnoses, the NSMHWB used the CIDI 2.1 to provide DSM-IV diagnoses. In the current study, the original
M. Teesson et al. / Drug and Alcohol Dependence 81 (2006) 149–155
DSM-III-R diagnosis scripting for the NCS was modified to provide DSM-IV diagnoses of drug and alcohol dependence, after consultation with the NCS data analysis manager. The principle differences between DSM versions in the diagnosis of substance dependence were a reduction from nine criteria in DSM-III-R to seven in DSM-IV. Underlying these changes were the combination of two items probing withdrawal in DSM-III-R, and the transfer of criterion four (failure to fulfil major role obligations) from dependence to abuse. The original DSM-III-R diagnosis scripting was modified by removing criterion four from the past year diagnosis section.1 The approximation for DSM-IV diagnosis was validated by comparison of DSM-III-R and DSM-IV prevalence of substance dependence. As expected, dependence rates were slightly lower using DSM-IV diagnostic criteria (relative to the original DSM-III-R rates in the NCS), reflecting a more specific distinction between substance dependence and abuse. 3. The NSMHWB assessed diagnoses in the past 12 months, whereas the NCS inquired about both lifetime and 12months prevalence. Comparisons were thus restricted to 12-months diagnoses. 4. The categories of substances for which information was acquired differed in each survey. Cocaine use was not evaluated in the NSMHWB as only those drugs with an expected prevalence of greater than 1% were included and the preceding National Household Survey (Commonwealth Department of Health and Family Services, 1996) indicated prevalence of cocaine use in Australia was less than 1%. Analyses were restricted to categories of drug use that were common across the two surveys, and that are therefore reported in the current analyses. These were • Cannabis (marijuana and hashish). • Stimulants (3,4-methylenedioxy-methamphetamine (MDMA), amphetamines and other stimulants obtainable by medical prescription). • Sedatives (including barbiturates, tranquillisers and other sedatives). • Opioids, including heroin and opium, and opiate analgesics which can be obtained on medical prescription. • Drug use disorder data is presented on subjects who had used drugs more than five times in the past 12 months. 2.1. Analysis Prevalence estimates for substance use and dependence in both surveys were derived using a complex estimation procedure to account for the stratified multistage survey design. Part I weightings were used in the NCS for all analyses except those involving country of birth and employment status variables The Strata and ECU weightings were used in 1 The combination of criteria 8 and 9 in this section was not possible, since Criterion 9 was not available.
151
the SUDAAN jack-knife analysis to adjust for the complex survey design. The person-weightings and replicate weights in the NSMHWB were divided by the estimated Australian adult population size to adjust the proportional sample size for direct comparability with the NCS data. Odds ratios (based upon weighted estimates) were used to examine the strength of bivariate associations between substance use/dependence and other variables. All covariates for modelling the probability of alcohol or drug dependence were chosen because they have been shown previously to be associated with alcohol or drug use dependence. These include the social demographic variables of age, gender, marital status, employment, country of birth, education and urban location of living.
3. Results 3.1. Prevalence of substance use in the past 12 months Table 1 shows that recent alcohol use is higher in Australian adults, compared with Americans. Within the Australian sample, 77.2% of respondents aged 18–54 reported consumption of at least 12 drinks of alcohol in the year preceding the NSMHWB (1997), compared with 46.3% of Americans in the 12 months preceding the NCS (1990/1991). The prevalence of use of the drugs assessed was considerably lower than alcohol use, but similar, in both countries. An estimated 10.8% of Australians and 11.9% of Americans used at least one drug from the other classes (cannabis, stimulants, sedatives and opioids) more than five times in the past year. The most commonly used drug, both in Australia and the USA, was cannabis, with 9.8% of Australians and 8.1% of Americans reporting use on at least five occasions in the past year. The prevalence of use of stimulants, sedatives and opioids was around 0.3–1.3% for Australians and 2.4–2.5% for Americans. 3.2. Twelve-months prevalence of substance dependence In Australia, 5.3% of the sample met DSM-IV criteria for alcohol dependence, and 2.7% met criteria for drug dependence within the last 12 months. In the USA, these figures were 3.1 and 1.1%, respectively. Aside from alcohol, cannabis accounted for the highest levels of drug dependence. The estimated past-year prevalence of cannabis dependence in Australia was 2.0%, while in the USA it was 0.5%. Of the remaining classes of drugs, sedative dependence was next in rank in Australia, while in the USA this position was occupied by opioid dependence (see Table 1). The conditional prevalence of dependence (dependence among users only) on alcohol was similar in Australia (6.8%) and the USA (6.5%). However, for drugs a different picture emerged, with Australians in general having higher conditional prevalence’s than their American counterparts.
M. Teesson et al. / Drug and Alcohol Dependence 81 (2006) 149–155
152
Table 1 Twelve-months prevalence (%, S.E.) of substance use and DSM-IV dependence in the United States of America and Australia Use in past 12 months
Alcohol Any drug Cannabis Stimulants Sedatives Opioids
DSM-IV dependence
DSM-IV dependence among users
USA
Australia
USA
Australia
USA
Australia
46.3 (1.1) 11.9 (0.7) 8.1 (0.5) 2.5 (0.3) 2.4 (0.3) 2.5 (0.3)
77.2 (1.1) 10.8 (0.5) 9.8 (0.5) 1.3 (0.2) 1.1 (0.1) 0.3 (0.1)
3.1 (0.4) 1.1 (0.2) 0.5 (0.1) 0.2 (0.1) 0.2 (0.1) 0.3 (0.1)
5.3 (0.5) 2.7 (0.3) 2.0 (0.2) 0.3 (0.1) 0.5 (0.1) 0.3 (0.1)
6.5 (0.8) 9.0 (1.7) 6.5 (1.5) 7.8 (2.7) 8.4 (3.8) 9.1 (2.9)
6.8 (0.6) 22.7 (2.6) 20.9 (2.2) 22.4 (5.0) 25.9 (4.9) 43.1 (15.9)
The overall prevalence of alcohol and drug dependence was higher among males than among females. Specifically, 7.5% of males and 3.0% of females met DSM-IV criteria for alcohol dependence in Australia, and 3.7% of males and 1.6% of females met the criteria for drug dependence. In the USA, the prevalence of alcohol dependence was 5.1% for males and 1.2% for females, while for drugs it was 1.2% for males and 1.0% for females. With the exception of sedatives, all substances were more often used by males than by females, in both countries. 3.3. Correlates of substance dependence in Australia and the USA Associations between substance dependence and a range of socio-demographic characteristics were explored. Two multiple logistic regression analyses were undertaken with alcohol and drug dependence as dependent variables and nation (USA versus Australia) and other demographic variables as independent variables. Table 2 shows the results of these analyses for alcohol and drug use separately. Interactions between country and other correlates were tested but are not reported as they were non-significant and did not add to the explanation of the models.
The final model for alcohol dependence (−2 log likelihood 4819; χ = 384.2, p < 0.000) is presented in Table 2. The odds ratios presented are adjusted for the effect of all other variables in the model. Younger respondents were more likely than those over 44 years to meet criteria for alcohol dependence. Males were three times more likely than females to be alcohol dependent (OR = 3.0). Those who were never married were more likely to have alcohol dependence (OR = 1.7). Persons with less than 12 years education were more likely (OR = 1.4) than those with 12 years or more to be dependent on alcohol. Australians were 1.5 times more likely to be alcohol dependent than Americans. The final model for drug dependence (−2 log likelihood 2515; χ = 282.3, p < 0.000) is presented in Table 2. Younger respondents were also more likely than those over 44 years to meet criteria for drug dependence. Males were twice as likely as females to be drug dependent (OR = 2.2). Those who were never married were more likely to have a drug dependence (OR = = 1.8). Persons with less than 12 years education were more likely (OR = 1.7) than those with 12 years or more to be dependent on alcohol. Those who were unemployed were twice as likely to be dependent on drugs (2.2). Australians were nearly three times as likely to meet criteria for drug dependence than Americans. 3.4. Treatment seeking
Table 2 Adjusted odds ratios for alcohol and drug dependence, United States vs. Australia, full model Variable
Age 18–24 25–34 35–44 45–54 Male Never Married Urban Less education (<12 years) Unemployed Nationa
Alcohol dependence
Drug dependence
Adjusted odds ratio
95% CI
Adjusted odds ratio
95% CI
2.9* 1.9* 1.7* 1.0
2.1–4.0 1.5–2.6 1.3–2.3 –
5.2* 3.7* 2.4 1.0
3.0–9.1 2.2–6.2 1.4–4.1 –
3.0* 1.7* 1.0 1.4*
2.5–3.6 1.4–2.1 0.8–1.2 1.2–1.7
2.2* 1.8* 1.3 1.7*
1.7–2.8 1.3–2.4 1.0–1.8 1.3–2.2
0.8 1.5*
0.7–1.0 1.3–1.8
2.2* 2.9*
1.7–2.9 2.2–3.8
a Coded as Australia = 1 and USA = 0 odds ratios adjusted for all other covariates in the model: significance set at *p < 0.001.
Table 3 summarises logistic regression analyses in which treatment seeking was the dependent variable, and other socio demographic variables (including nation) were entered as the independent variables. Interaction effects were tested but did not add to the explanatory power of the model and are therefore not reported. The results of the analysis for those with alcohol and drug dependence revealed Australians were no more likely to seek professional help than Americans. Males were half as likely to seek help for alcohol dependence than females (OR = 0.46 95% CI 0.31–0.69). Males were even less likely to seek help for drug dependence than females (OR 0.33 95% CI 0.19–0.59) and 18–24 years old were less likely to seek help than those aged over 44 (OR 0.15 95% CI 0.07–0.32). Australians more often consulted family physicians or general practitioners, while Americans were almost equally likely to consult a general practitioner, a psychiatrist or a psychologist.
M. Teesson et al. / Drug and Alcohol Dependence 81 (2006) 149–155
153
Table 3 Twelve months treatment seeking (%, S.E.) among those with DSM-IV alcohol or drug dependence in the United States of America and Australia by gender Males Australia
Females USA
Australia
USA
Alcohol dependence General practitioner Psychiatrist Psychologist Other professional Hospital/drug unit admission Any treatment
6.8 (0.8) 1.7 (0.3) 1.4 (0.3) 3.8 (0.5) 0.3 (0.1) 9.2 (1.1)
2.0 (0.6) 1.7 (0.3) 1.3 (0.3) 3.1 (0.5) 0.5 (0.2) 7.1 (0.9)
11.8 (1.3) 2.3 (0.4) 2.4 (0.4) 7.8 (0.5) 0.3 (0.1) 16.2 (1.2)
4.4 (1.0) 2.8 (0.9) 2.7 (0.5) 6.4 (0.9) 0.6 (0.3) 12.9 (1.5)
Drug dependence General practitioner Psychiatrist Psychologist Other professional Hospital/drug unit admission Any treatment
12.0 (2.6) 4.3 (1.1) 2.6 (0.6) 9.2 (2.9) 0.5 (0.3) 17.4 (3.6)
3.6 (1.5) 3.2 (0.9) 4.3 (1.0) 7.6 (1.5) 1.8 (0.8) 15.2 (2.0)
24.5 (3.4) 6.0 (1.6) 7.6 (1.9) 16.3 (2.4) 1.0 (0.9) 33.4 (3.9)
7.4 (2.1) 5.6 (2.0) 5.7 (1.6) 10.8 (2.0) 2.5 (1.3) 21.4 (2.9)
4. Discussion The major finding from this comparison, with a few exceptions, is the consistency of rates and patterns of 12-months dependence on alcohol and drug dependence in the U.S. and Australia. Both surveys reported a higher prevalence of substance dependence in males than females, and the prevalence peaked in young adults and declined linearly with age. Importantly, the lack of any significant interaction between the common correlates of alcohol and other drug dependence and country indicates that the effects of correlates such as age, gender, and unemployment were similar in the two countries. The striking difference was the higher conditional prevalence for drug dependence in Australia compared to the USA, yet consistent rates of conditional prevalence for alcohol dependence. The differences in dependence rates between Australia and the USA remained when other demographic variables were accounted for in statistical models. The limitations of this cross-national comparison should be acknowledged. Firstly, the NSMHWB was not able to correct for non-response. However, the non-responders in the NCS had a higher rate of psychopathology and therefore the omission of non-responders in the Australian sample should bias against the finding of higher rates of conditional drug prevalence in Australia. Secondly, there is a 5-years gap between the Australian and U.S. samples (Australia 1997, USA 1992). Chen et al. (in press) have shown that age of initiation of cannabis use has reduced during the mid nineties in the U.S., and Compton et al. (2004) found some evidence suggestive of an increased occurrence of cannabis dependence in years 2000–2001 as compared to corresponding estimates 10 years ago. These changes may also have occurred in Australia, but are unlikely to fully explain the much higher rates of conditional drug prevalence in Australia compared to the USA. In most other respects, the findings of the NSMHWB were similar to those of the NCS. The NCS used the CIDI Version 1 to ascertain DSM-III-R diagnoses,
although data was collected so as to allow comparisons with DSM-IV. The NSMHWB used a slightly different version of the CIDI to ascertain DSM-IV diagnoses. Even so the differences in conditional drug prevalence were so large that these minor methodological differences cannot be expected to fully explain them. Despite the similar prevalence of alcohol and drug dependence in the U.S. and Australia, treatment-seeking patterns, although not rates of treatment seeking, differed. While specialist care was equally sought in the USA and Australia, individuals with substance dependence in Australia were more likely to seek care from general practitioners. Factors that may influence treatment seeking include health care infrastructure, degree of debility, socio-economic status and individual attitudes toward mental health. We were not able to measure such structural determinants in such a cross sectional survey, although they should form part of future work in the area. The analyses identified that several demographic correlates were related to the likelihood of having alcohol or drug dependence in both countries. Those who were married were less likely than those never married to have alcohol or drug dependence. The unemployed were more likely to have drug dependence than those in the labour force. Interestingly, unemployment was not correlated with alcohol dependence, which again may reflect its widespread use. Similarly, there were observed differences in the pattern of substance dependence according to level of education attained. There was no difference in the likelihood of having alcohol or drug dependence according to a person’s location of residence. This may reflect an increasing level of exposure to drugs in regional areas such that the risk of dependence is similar in urban and rural areas. Importantly, the absence of significant interactions between the above factors and country indicated that the influence of these factors was consistent across the two countries. The cross-national generalizability of the associations
154
M. Teesson et al. / Drug and Alcohol Dependence 81 (2006) 149–155
between these factors and rates of alcohol or drug use and dependence indicates that similar process of vulnerability to dependence may be operating in the two countries. While our results are consistent with previous findings regarding the correlates of alcohol and other drug dependence in Australia and the U.S. to the best of our knowledge this is the first time that potential differences in the magnitude of these associations have been formally tested. Finally, nation (Australia versus the USA) was associated with the likelihood of developing alcohol or drug dependence. Compared to those in the U.S., Australians are more likely to have alcohol or drug dependence. This difference remained after controlling for other demographic variables. The generally lower rate of conditional prevalence for alcohol use disorders indicates that non-dependent alcohol consumption among adult drinkers is more common than non-dependent illicit drug use. This is probably explained by the large number of older, moderate drinkers and the relative absence of older, moderate users of illicit drugs (Swift et al., 2001a,b). This was consistently found in both the Australian and U.S. samples. The similarity of conditional prevalence (dependence rate among alcohol users only) in the two countries suggests that the higher prevalence in Australia is simply a reflection of more widespread alcohol use in Australia. More Australians drink alcohol so more are dependent. The differences in conditional prevalence between the USA and Australia for illicit drug dependence are not so easily explained. The finding is consistent across drug classes and higher rates of drug dependence remain when other demographic variables are considered in statistical models. A limitation of this study is the lack of comparable data in the surveys on frequency and quantity of use. There are likely to have been differences between the two countries in consumption patterns, especially the consumption of the most widely used illicit drug in cannabis. One crucial issue is the delta-9-tetrahydrocannabinol (THC) content of cannabis products. The THC content of Australian cannabis products has not been systematically tested and the USA is the only country that has regularly collected data on the THC content of cannabis plants over the past several decades. The data shows only a small increase in THC content (Hall and Swift, 2000). Changes in the patterns of cannabis use may also increase the amount of THC being consumed. These changes include more regular use of more potent cannabis products such as the flowering tops of the stems, the greater use of water pipes and a possible increase in cannabis potency. Whether these changes occurred in Australia more so than the USA is unknown. Compton et al. (2004) suggested that increases in the THC concentration might account for apparent increases in prevalence of cannabis dependence in recent years. However, Chen et al. (in press) did not observe an increase in the risk of developing cannabis dependence problems soon after onset over the same time period. If the THC concentration has increased it may be that users compensate for increased
THC concentration (e.g., smoking fractions of a ‘joint’ as opposed to a complete ‘joint’) (Chen et al., in press). Thus, the relationship between use, patterns of use and dependence are not always clear.
5. Conclusions The drug dependence conditional prevalence rates differ considerably between the USA and Australia, yet the alcohol dependence conditional prevalence rates are very similar. The differences are so marked that the small differences in methodology are unlikely to explain the findings. The findings raise doubts as to the generalizability of research findings regarding risk factors cross-nationally and for the potential for targeted interventions based on identification of risk factors to be applicable cross-nationally.
Acknowledgements The National Survey of Mental Health and Well-being was a major piece of applied social research that involved the cooperation of a large number of individuals and organisations that we wish to thank. The design, development, and conduct of the survey were funded by the Mental Health Branch of the Commonwealth Department of Health and Family Services. The development and testing of the computerised survey instrument was undertaken by Gavin Andrews, Lorna Peters, and other staff at the Clinical Research Unit for Anxiety Disorders, and the WHO Collaborating Centre in Mental Health at St. Vincent’s Hospital. The design of the survey was overseen and approved by a technical advisory committee comprising: Professor A. Scott Henderson, Chair; Professor Gavin Andrews, Professor Wayne Hall, Professor Helen Herman, Professor Assen Jablensky and Professor Bob Kosky. The Australian Bureau of Statistics conducted the fieldwork and implementation of the Survey, the enumeration, compilation and initial analyses of the data. We thank Tony Cheshire, Gary Sutton and Marelle Lawson for their assistance. We thank Wendy Swift for statistical advice. We would also like to thank the many individuals who gave their time to participate in this survey. The NCS was funded by the National Institute of Mental Health (Grants R01 MH/DA46376 and R01 MH49098), the National Institute of Drug Abuse (through a supplement to R01 MH/DA46376) and the W.T. Grant Foundation (Grant 90135190). The NCS data was obtained from; Kessler, Ronald C. NATIONAL COMORBIDITY SURVEY, 1990–1992 [Computer file]. Conducted by University of Michigan, Survey Research Center. ICPSR ed. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [producer and distributor], 2000. We would like to acknowledge the assis-
M. Teesson et al. / Drug and Alcohol Dependence 81 (2006) 149–155
tance of the data manager in obtaining and using this information.
References Andrews, G., Hall, W., Teesson, M., Henderson, S., 1999. The Mental Health of Australians (2). Commonwealth Department of Health and Family Services, Canbera Australia. Anthony, J.C., Warner, L., Kessler, R., 1994. Comparative epidemiology of dependence on tobacco, alcohol, controlled substances, and inhalants: basic findings from the National Comorbidity Survey. Exp. Clin. Psychopharmacol. 2 (3), 244–268. Anthony, J.C., in press. The epidemiology of cannabis dependence. In: Roffman, R., Stephens, R. (Eds.). Cannabis dependence: Its Nature, Consequences and Treatment. Cambridge University Press, Cambridge. Australian Bureau of Statistics, 1998. National Survey of Mental Health and Wellbeing of Adults: Users’ Guide. Australian Bureau of Statistics, Canberra. Chen, C-Y., O’Brien, M.S., Anthony, J.S., 1999. Who becomes cannabis dependent soon after onset of use? Drug Alcohol Depend. Commonwealth Department of Health and Family Services, 1996. National Drug Strategy Household Survey, Survey report, 1995. Commonwealth Department of Health and Family Services, Canberra. Compton, W., et al., 2004. Prevalence of Marijuana use disorders in the United States: 1991–1992 and 2001 and 2002. JAMA 291, 2114. Degenhardt, L., Hall, W., Lynskey, M., 2001. Alcohol, cannabis and tobacco use among Australians: a comparison of their associations with other drug use and use disorders, affective and anxiety disorders, and psychosis. Addiction 96, 1603–1614. Drucker, E., Lurie, P., Wodak, A., Alcabes, P., 1998. Measuring harm reduction: the effects of needle and syringe exchange programs and methadone maintenance on the ecology of HIV. Aids 12 (Suppl. A), S217–S230. Furr-Holden, C.D.M., Anthony, J.C., 2003. Epidemiologic difference in drug dependence: a U.S.–U.K. cross-national comparison. Social Psychiatr. Psychiatr. Epidemiol. 38, 165–172. Hall, W., Teesson, M., Lynskey, M., Degenhardt, L., 1998. The 12-months prevalence of substance use and ICD-10 substance use disorders in Australian adults: findings from the National Survey of Mental Health and Well-being. Addiction 94 (10), 1541–1550.
155
Hall, W., Swift, W., 2000. The THC content of cannabis in Australia: evidence and implications. Aust. N. Z. J. Public Health 24, 503– 508. Kessler, R.C., Frank, R.G., Edlund, M., Katz, S.J., Lin, E., Leaf, P., 1997. Differences in the use of psychiatric outpatient services between the United States and Ontario [see comments]. N. Engl. J. Med. 336 (8), 551–557. Kessler, R.C., McGonagle, K.A., Zhao, S., Nelson, C.B., Hughes, M., Eshleman, S., Wittchen, H.U., Kendler, K.S., 1994. Lifetime and 12months prevalence of DSM-III-R psychiatric disorders in the United States. Results from the National Comorbidity Survey. Arch. Gen. Psychiatr. 51 (1), 8–19. Maxwell, J.C., 2003. Update: comparison of drug use in Australia and the United States as seen in the 2001 National Household Surveys. Drug Alcohol Rev. 22 (3), 347–357. Poulton, R.G., Brooke, M., Moffitt, T.E., Stanton, W.R., Silva, P.A., 1997. Prevalence and correlates of cannabis use and dependence in young New Zealanders. N. Z. Med. J. 110, 68–70. Robins, L.N., Regier, D.A. (Eds.), 1991. The Epidemiological Catchment Area Study. The Free Press, New York. Ross, H., 1995. DSM-III-R alcohol abuse and dependence and psychiatric comorbidity in Ontario: results from the Mental Health Supplement to the Ontario Health Survey. Drug Alcohol Depend. 39, 111–128. Swift, W., Hall, W., Teesson, M., 2001a. Cannabis use and dependence among Australian adults: results from the National Survey of Mental Health and Well-being. Addiction 96, 737–748. Swift, W., Hall, W., Teesson, M., 2001b. Characteristics of DSM-IV and ICD-10 cannabis dependence among Australian adults: results from the National Survey of Mental Health and Well-being. Drug Alcohol Depend. 63, 147–153. Teesson, M., Hall, W., Lynskey, M., Degenhardt, L., 2000. Alcohol and drug use disorders in Australia: implications of the National Survey of Mental Health and Well-being. Aust. N. Z. J. Psychiatr. 34, 206–213. Teesson, M., Lynskey, M., Manor, B., Baillie, A., 2002. The structure of DSM-IV cannabis use disorders in the community. Drug Alcohol Depend. 68, 255–262. Vega, W.A., Aguilar-Gaxiola, S., Andrade, L., Bijl, R., Borges, G., Caraveo-Anduaga, J.J., DeWit, D.J., Heeringa, S.G., Kessler, R.C., Kolody, B., Merikangas, K.R., Molnar, B.E., Walters, E.E., Warner, L.A., Wittchen, H.U., 2002. Prevalence and age of onset for drug use in seven international sites: results from the international consortium of psychiatric epidemiology. Drug Alcohol Depend. 68, 285–297.