Stimulant use disorders: Characteristics and ... - SAGE Journals

1 downloads 0 Views 689KB Size Report
Abstract. Objective:To describe the correlates of stimulant use disorders (abuse, dependence) in an Australian population sample, to compare the characteristics ...
461057 2012

ANP461210.1177/0004867412461057ANZJP ArticlesSara et al.

Research

Stimulant use disorders: Characteristics and comorbidity in an Australian population sample

Australian & New Zealand Journal of Psychiatry 46(12) 1173­–1181 DOI: 10.1177/0004867412461057 © The Royal Australian and New Zealand College of Psychiatrists 2012 Reprints and permission: sagepub.co.uk/journalsPermissions.nav anp.sagepub.com

Grant Sara1,2,3, Philip Burgess3, Meredith Harris3, Gin S Malhi2,4, Harvey Whiteford5 and Wayne Hall6

Abstract Objective: To describe the correlates of stimulant use disorders (abuse, dependence) in an Australian population sample, to compare the characteristics of stimulant users with and without stimulant use disorders and to describe the patterns of service use and help-seeking in people with stimulant use disorders. Methods: Data were drawn from the 2007 National Survey of Mental Health and Wellbeing, which sampled 8841 residents of private dwellings in Australia in 2007. Lifetime DSM-IV substance use and mental disorder diagnoses were obtained from interviews conducted by lay interviewers, using the Composite International Diagnostic Interview (CIDI). Socio-demographic, socio-economic and clinical correlates of stimulant use disorders were identified using binary logistic regression models. Stimulant users with and without stimulant use disorders were compared to non-stimulant users via multinomial logistic regression models. Results: Compared to Australians without stimulant use disorder, people with stimulant use disorders were younger, more likely to be male, of non-heterosexual orientation and born in Australia, but were not more socially disadvantaged. Lifetime comorbidity rates were high: 79% of persons with stimulant use disorders had a lifetime alcohol use disorder, 73% a lifetime cannabis use disorder, and more than one third a lifetime mood or anxiety disorder. Stimulant use disorders were associated with a family history of substance use, affective disorders and psychosis. One in five people with lifetime stimulant use disorders had been imprisoned, homeless or hospitalised for substance or mental health problems, and 13% reported at least one symptom of psychosis. Nearly half had sought help for substance or mental health problems, primarily from General Practitioners (GPs), psychologists or psychiatrists. Conclusions: Stimulant use disorders in a representative population sample are associated with significant comorbidity and harm. Many persons with stimulant use disorders had sought care and found this helpful. There is scope for screening and intervention in this group. Keywords Stimulant disorders, stimulant dependence, stimulant abuse, harmful use, amphetamine, methamphetamine, cocaine, prevalence, associations, service use

1InforMH,

Mental Health and Drug and Alcohol Office, NSW Health, NSW, Australia 2Discipline of Psychiatry, Sydney Medical School, University of Sydney, Sydney, Australia 3School of Population Health, University of Queensland, Brisbane, Australia 4CADE Clinic, Royal North Shore Hospital, Sydney, Australia 5Queensland Centre for Mental Health Research, University of Queensland, Brisbane, Australia

6Centre

for Clinical Research, University of Queensland, Brisbane, Australia Corresponding author: Grant Sara, Macquarie Hospital, PO Box 169, North Ryde, NSW 1670, Australia. Email: [email protected]

Australian & New Zealand Journal of Psychiatry, 46(12)

1174

Introduction Stimulants may cause serious physical and mental health problems (Darke et al., 2008), more often in regular or dependent users than in those who use less frequently (McKetin et al., 2006a; Degenhardt et al., 2008). In Australia the 12 month prevalence of stimulant use disorders (abuse or dependence) is estimated to be around 0.6% (McKetin et al., 2005; Sara et al., 2011), a rate comparable to the prevalence of dependent heroin use at the height of Australia’s heroin ‘epidemic’ in the late 1990s (Hall et al., 2000). Understanding the characteristics and comorbidities of individuals with stimulant use disorders may assist in planning health system responses for this group. The first aim of this study is to describe people with stimulant use disorders in a representative Australian population sample. Most descriptions of stimulant use disorders are based on treated or convenience samples such as injecting drug users or high risk populations. These studies find that stimulant use disorders are associated with very significant comorbidity. Dependent stimulant users are often dependent on other drugs, and have high rates of depression, anxiety, psychotic symptoms, suicidal ideation and suicide attempts (Glasner-Edwards et al., 2009; Kalechstein et al., 2000; Stafford and Burns, 2010; Wallace et al., 2009; Zweben et al., 2004; McKetin et al., 2010; McKetin et al., 2006b). Individuals with stimulant use disorders are significantly disabled or marginalised, with fewer years of education and significantly increased rates of unemployment, homelessness, criminal activity and imprisonment compared with those without stimulant use disorders (Copeland and Sorensen, 2001; Degenhardt et al., 2008; Korte et al., 2011; McKetin et al., 2006a; Stafford and Burns, 2010; Zweben et al., 2004; Farrell et al., 1998). However, treatment and convenience samples are likely to include a more severely dependent or disabled group, who may differ systematically from persons with stimulant use disorders in the general population. Population surveys of drug use can complement studies based on treatment or convenience samples. Most population surveys report only on stimulant use, and do not apply diagnostic criteria for stimulant abuse or dependence. Surveys from the US, UK, Canada, Australia and New Zealand (Adlaf et al., 2005; Durell et al., 2008; Substance Use and Mental Health Services Administration, 2007, 2010; Wilkins and Sweetsur, 2008) find that stimulant use is more common in men, peaks in the 20s, and is usually preceded by the use of other drugs, particularly cannabis. Australia’s National Drug Strategy Household Survey (NDSHS) found that stimulant use is usually infrequent; only 36% of those who had used any stimulant in the preceding year had done so monthly or more often. Few population surveys have examined regular stimulant users or those diagnosed with stimulant use disorders. The US National Epidemiological Survey of Alcohol and Related Conditions (Conway et al., 2006) found high rates of lifetime depressive disorders (51%) and anxiety Australian & New Zealand Journal of Psychiatry, 46(12)

ANZJP Articles disorders (39%) in individuals with amphetamine abuse or dependence. The second aim of this study is to compare stimulant users who do not have stimulant use disorders with those who do. We have previously reported that nearly half (46%) of people who have used stimulants on more than five occasions met criteria for a lifetime stimulant use disorder (Sara et al., 2011). For an individual, the vulnerability to develop abuse or dependence is likely to reflect a mix of genetic, psychological and environmental factors (Kendler et al., 2003; Yucel et al., 2007) along with drug-related factors (McKetin et al., 2006a). To our knowledge, no study has compared the personal, family and social correlates of persons with stimulant use disorders against those of stimulant users who do not have features of abuse or dependence. The third aim of this study is to examine help-seeking in people with stimulant use disorders. Australians with substance use disorders seek help less often than people with other mental disorders, and rates of help seeking are lowest in younger adults and males (Burgess et al., 2009), who are the group most likely to have stimulant use disorders. Understanding the extent of unmet need for treatment in this group may assist in identifying opportunities for prevention or intervention. This study addresses these three aims by examining stimulant use disorders in the 2007 Australian National Survey of Mental Health and Wellbeing (NSMHWB) (Australian Bureau of Statistics, 2008), which provided information about a representative sample of the Australian population. It collected diagnostic information on substance use and high prevalence mental disorders diagnosed using ICD-10 and DSM-IV criteria, along with information about personal and household characteristics and family history.

Setting and participants The 2007 NSMHWB was conducted by the Australian Bureau of Statistics (ABS) in late 2007. Respondents aged 16 to 85 years were identified from private dwellings selected by the ABS using a stratified, multi-stage area sample. One individual was selected from the pool of eligible adults in each dwelling, using a randomising algorithm implemented by a computer-assisted interview schedule, and invited to participate in an interview. Younger (16-24 years) and older people (65-85 years) were over-sampled to improve the reliability of estimates for these groups. The sampling process yielded 8,841 fully-responding households, a response rate of 60%.

Methods Measurements Stimulant use variables.  The 2007 NSMHWB reports the use of stimulant drugs in five categories: (i) Amphetamine/ Speed, (ii) Methamphetamine/Base/Ice, (iii) Ecstasy, (iv)

Sara et al. Cocaine and (v) Any stimulant (any of i – iv). For each of these drug types respondents were asked whether they had ever used, and if so their age at first use and whether they had used more than five times in their lifetime. Persons who reported using any stimulant more than five times in their lifetime were asked whether they had done so in the last twelve months. Diagnostic variables.  The 2007 NSMHWB used a modified version of the Composite International Diagnostic Interview (CIDI) to provide lifetime and twelve month diagnoses of stimulant abuse and stimulant dependence. The derived variable ‘any stimulant use disorder’ was positive if an individual met criteria for stimulant abuse or for stimulant dependence. NSMHWB did not report diagnoses of abuse or dependence for individual stimulants (e.g. methamphetamine, ecstasy, cocaine). DSM-IV criteria were used in this study.

Data analysis Data from the 2007 NSMHWB Basic Confidentialised Unit Record File (CURF), April 2009 version (Australian Bureau of Statistics, 2009) were analysed using PASW Statistics Version 18 (IBM Corporation, 2009) and STATA version 11 (StataCorp, 2009). Data were weighted using factors within the CURF which adjust for the differential probability of survey selection and for the age and sex distribution of the Australian population. For all analyses standard errors and 95% confidence intervals (CIs) were calculated using jackknife repeated replication to take account of the complex survey design. Because of the low prevalence of stimulant use disorders in the past 12 months, analyses were conducted on correlates of lifetime stimulant use disorders only. Analysis was conducted in two stages. Firstly, binary logistic regression was used to compare people with and without lifetime stimulant use disorders. Secondly, people without lifetime stimulant use disorders were divided into two groups; stimulant users and non-stimulant users. Multinomial logistic regression was then used to examine whether stimulant users with a lifetime disorder differed from stimulant users without disorder, when compared with people who had never used stimulants. For the first analysis, personal correlates and comorbidities of stimulant use disorders were examined by comparing persons with a lifetime history of DSM-IV stimulant use disorder against those with no stimulant use disorder. Odds ratios and CIs were calculated using binary logistic regression analyses conducted separately for a range of potential predictors, which included demographic variables and lifetime mental health and substance (alcohol, cannabis) disorders. Lifetime harms were not included in the multivariate analysis, as these are consequences of rather than risk factors for stimulant use disorders. Where necessary, variables with

1175 low frequency values and high standard errors were dichotomised or categories were combined. Multiple logistic regression was then performed using those variables with significant associations (p