- Email: [email protected]
Delay to first treatment contact for alcohol use disorder
Drug and Alcohol Dependence 147 (2015) 116–121
Contents lists available at ScienceDirect
Drug and Alcohol Dependence journal homepage: www.elsevier.com/locate/drugalcdep
Delay to first treatment contact for alcohol use disorder Cath Chapman a,∗ , Tim Slade a , Caroline Hunt b , Maree Teesson a a NHMRC Centre of Research Excellence in Mental Health and Substance Use (CREMS), National Drug and Alcohol Research Centre (NDARC), University of New South Wales, Sydney, NSW 2052, Australia b School of Psychology, University of Sydney, NSW 2006, Australia
a r t i c l e
i n f o
Article history: Received 11 August 2014 Received in revised form 25 November 2014 Accepted 28 November 2014 Available online 10 December 2014 Keywords: Alcohol abuse Alcohol dependence Treatment delay Service utilization
a b s t r a c t Background: This study explored the patterns and correlates of time to first treatment contact among people with alcohol use disorder (AUD) in Australia. Specifically it examined the relationship between sex, birth cohort, onset of AUD symptoms, severity, comorbidity, symptom type and time to first treatment contact (treatment delay) among those with alcohol abuse and dependence in a large population sample. Methods: Data came from the 2007 Australian National Survey of Mental Health and Wellbeing (N = 8841). A modified version of the World Health Organization’s Composite International Diagnostic Interview was used to determine the presence and age of onset of DSM-IV AUD and other mental disorders and the age at which respondents first sought treatment for alcohol or other drug-related problems. Results: Median time to first treatment contact for an AUD was 18 years (14 years dependence, 23 years abuse). Projected lifetime treatment rates were 78.1% for alcohol dependence and 27.5% for abuse. Those with earlier onset and from older cohorts reported longer delay and were less likely to ever seek treatment compared to those with later onset or from more recent cohorts. Those with comorbid anxiety but not mood disorder, or who reported alcohol-related role disruption or recurrent interpersonal problems were more likely to ever seek treatment and reported shorter delay compared to those who did not report these symptoms. Conclusions: Treatment delay for alcohol use disorder in Australia is substantial. Those with earlier onset and those with comorbid mood disorder should be a target for earlier treatment. © 2014 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Alcohol use disorders (AUD) are common. Prevalence estimates in the United States, the United Kingdom and Australia indicate between one in four and one in six adults in the population will meet criteria for DSM-IV alcohol abuse or dependence over their lifetime (Bunting et al., 2012; Hasin et al., 2007; Teesson et al., 2010). AUD is associated with substantial negative social and health consequences and poses significant public health concern (Whiteford et al., 2013). However, despite the high prevalence and negative impact of AUD, and the existence of effective interventions (Dawson et al., 2012; Jonas et al., 2012), most people with an AUD do not seek treatment (Edlund et al., 2012; Hasin et al., 2007). Moreover those who do, typically delay seeking treatment for many years following the onset of symptoms. Large community studies
∗ Corresponding author at: National Drug and Alcohol Research Centre, University of New South Wales, 22-32 King Street, Randwick Campus, Sydney, NSW 2052, Australia. Tel.: +61 2 93850333; fax: +61 2 9385 0222. E-mail address: [email protected] (C. Chapman). http://dx.doi.org/10.1016/j.drugalcdep.2014.11.029 0376-8716/© 2014 Elsevier Ireland Ltd. All rights reserved.
have reported median delays to first treatment contact of between 6 and 18 years after the onset of problems associated with alcohol (Bruffaerts et al., 2007; Bunting et al., 2012; Kessler et al., 2001; Keyes et al., 2010b; Wang et al., 2005, 2007b). Even if it is argued that a proportion of people with AUD will recover naturally over the course of their lives, these long delays are thought to represent considerable unmet need for care (ten Have et al., 2013; Witkiewitz et al., 2014). Identification of the factors associated with delay to seek treatment is key to understanding how to reduce these delays and lessen this unmet need. Several large community studies have examined the factors associated with delay to seek treatment for AUD over lifetime. Generally these studies have reported longer delays and lowered odds of ever seeking treatment among those with earlier onset of symptoms (Bruffaerts et al., 2007; Hingson et al., 2006; Kessler et al., 2001, 1998; Olfson et al., 1998; ten Have et al., 2013; Wang et al., 2005) and shorter delays and higher odds of ever seeking treatment among more recent cohorts (Bruffaerts et al., 2007; Kessler et al., 2001, 1998; Olfson et al., 1998; Wang et al., 2005) – although three studies have reported no effect of one or both of these variables (Borges et al., 2007; Keyes et al., 2010b; Lee et al., 2007).
C. Chapman et al. / Drug and Alcohol Dependence 147 (2015) 116–121
With the exception of one study (Alvanzo et al., 2014), previous community studies have not reported sex differences in delay to seek treatment. No other predictors of treatment delay have been consistently examined or reported. These studies are all based on large representative community surveys and are therefore a robust source of information. However, commonly these studies focused on one or two variables or examined only sociodemographic factors, thereby missing the impact of potentially important factors such as comorbidity, symptom type and severity. Both comorbidity and severity have been shown to be associated with a greater likelihood of seeking treatment among those with mental and substance use disorders (Cohen et al., 2007; Edlund et al., 2012; Ilgen et al., 2011; Wang et al., 2007a), and previous studies on treatment seeking for AUD have found particular symptoms or consequences of alcohol use to be associated increased odds of treatment seeking (Dawson et al., 2012; Naughton et al., 2013; Saunders et al., 2006). It is reasonable to expect that that these factors may also be associated with shorter treatment delays among those with an AUD. The present study sought to address this gap by simultaneously examining the relationship between sex, birth cohort, age of onset, comorbidity, severity and symptom type and delay to first treatment contact for AUD in a large representative community sample in Australia. Australia is a country with high rates of alcohol use and dependence (Teesson et al., 2010). This is the first time data on treatment delay among those with AUD in Australia has been reported. 2. Methods 2.1. Sample The 2007 Australian National Survey of Mental Health and Wellbeing (NSMHWB) is a nationally representative population survey with a sample size of 8841 (Slade et al., 2009). Respondents were selected at random from a stratified, multistage area probability sample of persons aged 16–85 years living in private dwellings and data were weighted according to the inverse probability of being selected. Interviews were conducted in respondent’s households using a computerassisted personal interview (CAPI). The survey received a response rate of 60%, which is commensurate with other major national surveys in mental health and substance use (Kerr et al., 2013; Kessler, 2008). 2.2. Measurement of AUD Experience of DSM-IV alcohol abuse, dependence and other mental disorders was assessed using a modified version of the World Health Organization’s Composite International Diagnostic Interview (WMH-CIDI; Kessler and Ustun, 2004), a highly structured interview with questions designed to operationalise the diagnostic criteria for each mental disorder. Respondents who reported they had consumed at least 12 alcoholic drinks in any 1 year over lifetime were asked if they had ever drunk alcohol on 3 or more days per week and/or have usually consumed 3+ drinks on the days they were drinking. AUD was assessed among respondents who answered ‘yes’ to this question (n = 5520; 64% of sample). A series of 18 questions operationalised the four alcohol abuse (major role disruption; hazardous use; recurrent legal consequences; and recurrent social or interpersonal problems) and seven alcohol dependence (tolerance; withdrawal; larger amounts/longer period of drinking; difficulty cutting down; significant time obtaining alcohol; important activities given up; and continued use despite problems) criteria. The sample for the present study comprised all respondents who met criteria for lifetime DSM-IV alcohol abuse or dependence (n = 1847). 2.3. Measurement of treatment delay Respondents who completed the alcohol module of the survey were asked about the age at which they first experienced symptoms of alcohol abuse (1+ symptoms in any year) or dependence (3+ symptoms in any year) and this was defined as the age of onset of AUD. Respondents were also asked if they had ever talked to a medical doctor or other professional (psychologist, social worker, counsellor, herbalist, acupuncturist or other healing professional) about their use of alcohol or drugs and if so, how old they were the first time they did so. Treatment contact for alcoholrelated problems was not differentiated from treatment contact for drug-related problems. Treatment delay was defined as the number of years between onset of AUD and first treatment contact.
117
2.4. Covariates Age at onset of AUD symptoms was coded in four categories of early onset (<19 years), early average (19–20 years), late average (21–29 years), and late onset (>29 years) based on the distribution of the age of onset (Wang et al., 2005). Birth cohort was defined in 10 year bands based on age at interview. Severity was examined in terms of whether respondents met criteria for alcohol abuse only or for alcohol dependence (with or without abuse) and in terms of the total number of abuse and/or dependence symptoms endorsed (Hingson et al., 2006). Comorbidty with DSM-IV mood (depression, dysthymia, bipolar disorder), anxiety (panic disorder, agoraphobia, social phobia, generalised anxiety disorder, posttraumatic stress disorder) and other drug use disorders was examined in two ways; (i) whether individuals met criteria for at least one disorder in each of these three categories, and (ii) whether the onset of the comorbid disorder pre-dated the onset of AUD symptoms. In these models, people who reported onset of a comorbid disorder after first treatment contact for AUD were excluded (Keyes et al., 2010b). Each of the four abuse and seven dependence symptoms were included as dichotomous variables to examine whether endorsement of specific symptoms was related to treatment delay (Kessler et al., 2001).
2.5. Statistical analysis Survey procedures in Stata Release 12 (StataCorp, 2012) were used for all analyses and standard errors obtained through the delete-a-group jack-knife variance technique to account for complex sampling procedures. Projected lifetime probability of treatment contact, proportion who made contact within a year and median duration of delay were obtained using Kaplan–Meier survival estimates among all respondents with an AUD, and separately for those with abuse (n = 1500) and with dependence with or without abuse (n = 347). This method of analysis allows for data to be modelled over time to include all respondents including those who had not sought treatment at the time of interview, often referred to as ‘censored cases’ (Hosmer, 1999). Survival time was defined as the number of years from onset of AUD to age at first treatment contact, or to age at interview, whichever came first. The relationship between various covariates and survival time was examined in logistic regression models using discrete-time survival analysis where sex, birth cohort, age at onset of AUD, number of symptoms, type of symptoms and presence of comorbid mood, anxiety and other drug use disorders were treated as covariates predicting treatment delay. Models were run separately for respondents with alcohol abuse and dependence. Individual contributions of covariates to survival time were assessed through Wald F-statistics and associated p-values and estimates of likelihood of treatment seeking in any year over lifetime are expressed as odds ratios (OR) with 95% confidence intervals, and interpreted in a similar way to hazard ratios. The effect of time in both models was approximated by including a linear term. Onset age of regular drinking (defined as 12+ drinks in any 1 year period) was included in final models to control for potential confounding on onset of AUD. Preliminary modelling ensured that models met the assumption of proportionality of hazards and where this assumption was not met models included interaction terms between covariates and time (Grambsch and Therneau, 1994).
3. Results 3.1. Lifetime probability of treatment contact and treatment delay Fig. 1 displays Kaplan–Meier failure curves for the cumulative lifetime probability of treatment contact after onset of AUD stratified by alcohol abuse and dependence. Table 1 displays the expected lifetime treatment rates, proportion who made contact within 1 year of onset and median duration of delay among those who eventually made treatment contact. Just over one third of people with an AUD were estimated to eventually make treatment contact with a median treatment delay of 18 years among those who did. Rates of expected lifetime treatment contact were lower and estimated duration of delay longer for those with alcohol abuse compared to those with alcohol dependence (Wald F = 12.0, p = 0.001). After adjusting for the effects of sex, birth cohort, onset of AUD, number of symptoms and the presence of comorbid disorders, those with alcohol dependence were 2.4 times more likely than those with alcohol abuse to make treatment contact (OR = 2.4, 95% CI 1.4–3.9). Median treatment delays among those with alcohol abuse and dependence were 23 years and 14 years, respectively.
118
C. Chapman et al. / Drug and Alcohol Dependence 147 (2015) 116–121
Table 1 Proportional treatment contact within 1 year of onset, projected lifetime probability of treatment contact and median duration of delay among people with alcohol use disorders in the Australian population. Lifetime treatment for AUD
Alcohol abuse (N = 1500) Alcohol dependence (N = 347) Alcohol use disorder (N = 1847) a
Contact w/in 1 year %
Projected lifetime probability of contact %
Median duration of delaya Years
1.3 8.9 2.6
27.5 78.1 34.6
23 14 18
Median delay in years among those who eventually made treatment contact.
obligations were 2.3 times more likely to make treatment contact than those who did not (Wald F = 5.58, p = 0.022; 9 vs 29 years).
Fig. 1. Cumulative Lifetime probability of treatment contact for persons with AUD in the Australian population, stratified by abuse and dependence.
3.2. Predictors of treatment delay Table 2 presents the factors associated with treatment delay expressed as adjusted ORs for probability of treatment contact in any year among those with alcohol abuse and those with alcohol dependence. 3.2.1. Sex and birth cohort. Sex was not associated with treatment delay in either model. Individuals from more recent birth cohorts were significantly more likely to make treatment contact and to experience shorter treatment delay compared to individuals born before 1958 in both models (abuse: Wald F = 3.68, p = 0.017; 7–10 vs 29 years; dependence: Wald F = 5.23, p = 0.003; 4–10 vs 21 years). 3.2.2. Onset of AUD symptoms. Those with early onset of AUD symptoms were significantly less likely to make treatment contact and to experience longer treatment delay, than those with late onset, among those with alcohol abuse (Wald F = 2.81, p = 0.047; 7–19 vs 6 years) and dependence (Wald F = 5.99, p = 0.001; 13–16 vs 5 years). 3.2.3. Number and type of AUD symptoms. Among those with alcohol abuse the total number of abuse symptoms endorsed was not associated with treatment delay (Wald F = 0.86, p = 0.470). Similarly, among those with alcohol dependence the total number of abuse (Wald F = 0.85, p = 0.472) or dependence symptoms (Wald F = 1.53, p = 0.207) endorsed was not associated with treatment delay. Among those with alcohol abuse, individuals who endorsed recurrent social or interpersonal problems were 1.8 times more likely to make treatment contact and reported shorter treatment delay than those who reported never experiencing this symptom (Wald F = 4.8, p = 0.032; 21 vs 23 years). Among people with alcohol dependence those who endorsed failure to fulfil important role
3.2.4. Comorbidity. The presence of comorbid mood disorder was not associated with treatment delay among those with alcohol abuse or dependence (Wald F = 3.30, p = 0.075; Wald F = 0.57, p = 0.452, respectively). Conversely, comorbid anxiety disorder was associated with shorter treatment delay among those with abuse (Wald F = 6.74, p = 0.012) and dependence (Wald F = 4.65, p = 0.035). When the temporal relationship between anxiety disorder onset and AUD onset was considered, anxiety disorders that predated AUD symptom onset were significantly associated with shorter treatment delay and greater likelihood of treatment contact among those with abuse (OR = 2.1; 95% CI: 1.1–4.1; Wald F = 4.00, p = 0.024; 14 vs 19 years) and dependence (OR = 2.6; 95% CI: 1.2–5.7; p = 0.015; 11 vs 19 years) compared to those without a comorbid anxiety disorder. The presence of a comorbid drug use disorder was significantly associated with shorter treatment delay among those with alcohol abuse (Wald F = 23.36, p < 0.001) but not dependence (Wald F = 0.25, p = 0.619). Among those with alcohol abuse, both pre-existing drug use disorders and those with onset after AUD were significantly associated with shorter treatment delay and greater likelihood of treatment contact (OR = 4.4; 95% CI: 2.1–9.0; OR = 5.1; 95% CI: 2.0–12.9, Wald F = 11.06, p < 0.001; 8–19 vs 27 years). 4. Discussion The present study estimated that one in three people with an AUD in Australia will make treatment contact over lifetime and those who do will delay seeking treatment for a median of 18 years after onset of AUD symptoms. Expected rates of 1 year and lifetime treatment contact are higher among those with alcohol dependence than those with alcohol abuse and shorter median delay. Individuals from more recent cohorts, with later onset, or with comorbid anxiety disorder are more likely to make treatment contact and to do so sooner after onset. Among those with alcohol abuse, individuals with comorbid drug use disorder and those who experience role disruption are also more likely to make treatment contact and to experience shorter treatment delay. Among people with alcohol dependence, those who experience recurrent social or interpersonal problems are more likely to make treatment contact than those who do not. Sex, number of AUD symptoms and comorbid mood disorder were not associated with treatment delay. Previous estimates of treatment delay for AUD vary substantially, even among studies with comparable methodology. Within this context, the median treatment delays for AUD in Australia are longer than estimates from other countries. For alcohol abuse, the probability of ever seeking treatment in Australia of 27.5% is substantially lower than estimates from the United Kingdom: 74.6% (Bunting et al., 2012) and the United States: 52.7% (Wang et al., 2005) and the median treatment delay substantially longer (23 years vs 8 and 16 years). This difference is smaller for alcohol dependence where lifetime probability of treatment seeking is 78.1% in
C. Chapman et al. / Drug and Alcohol Dependence 147 (2015) 116–121
119
Table 2 Factors associated with delay to first treatment contact among people with alcohol use disorders in the Australian population. Likelihood of treatment contact Alcohol abuse Odds ratio
Alcohol dependence 95% CI
p-value
Odds ratio
95% CI
p-value
0.763
0.7
0.4–1.6
5.6* 3.8* 3.1* 1.0
2.2–14.0 1.7–8.2 1.3–7.1 –
0.2* 0.2* 0.4 1.0
0.1–0.7 0.1–0.5 0.1–1.4 –
0.470 – 0.123 0.498 0.411
1.0 2.0 2.2 2.0
– 0.6–6.2 0.8–5.7 0.8–5.3
0.472 – 0.240 0.110 0.156
– – – – –
1.0 1.7 1.6 2.3 2.8*
– 0.6–5.0 0.7–3.8 0.9–6.1 1.2–8.3
0.207 – 0.355 0.231 0.092 0.019
0.203 0.914 0.897 0.032
2.3* 0.3 0.5 1.3
1.1–4.7 0.1–1.1 0.2–1.4 0.6–2.9
0.022 0.067 0.169 0.551
– – – – – – –
1.3 1.4 – 1.3 1.0 1.3 1.9
0.7–2.4 0.7–3.0 – 0.6–2.7 0.4–2.5 0.7–2.2 0.9–4.3
0.487 0.349 – 0.514 0.983 0.410 0.103
0.075 0.012 0.000
1.2 2.1* 0.8
0.7–2.0 1.1–4.1 0.4–1.7
0.452 0.035 0.619
Sex 0.9
0.5–1.8
Birth cohort 1978–1991 1968–1977 1958–1967 < 1958
5.5* 2.5* 1.1 1.0
1.4–21.6 1.1–6.0 0.4–2.9 –
Onset of AUD symptoms Early Early average Late average Late
0.4* 0.5 0.3* 1.0
0.1–0.9 0.2–1.2 0.1–0.7 –
Number of abuse symptoms One Two Three Four
1.0 1.8 1.3 1.9
– 0.9–3.7 0.6–3.0 0.4–9.2
Number of dependence symptoms Three Four Five Six Seven
– – – – –
Type of abuse symptomsa Failure to fulfil role Drinking in hazardous situations Recurrent legal problems Recurrent social/interpersonal problems
1.5 0.9 1.0 1.8*
Type of dependence symptomsa Tolerance Withdrawal Larger amounts/longer periodsb Unable to cut-down Time spent obtaining alcohol Important activities given up Persistent physical/psychological problem
– – – – – – –
Comorbid mental disorderc Any mood disorder Any anxiety disorder Any drug use disorder
1.8 2.0* 4.7*
Female
a b c *
– – – – – 0.8–3.0 0.3–2.7 0.5–2.1 1.1–3.1 – – – – – – – 0.9–3.3 1.2–3.3 2.5–8.8
0.017 0.016 0.038 0.773 0.047 0.024 0.108 0.007
0.444 0.003 0.000 0.001 0.010 0.001 0.010 0.002 0.144
Type of abuse and dependence symptoms were examined in models controlling for the number of symptoms other than the target symptom endorsed. This symptom was not examined as it was endorsed by almost all of those with alcohol dependence. Comorbid mental disorders were also examined in terms of whether their onset was before or after the onset of AUD symptoms (see Section 3.2.4). Significant compared to reference group.
Australia, 88.8% in the United Kingdom, and 69.8% in the United States and median treatment delays 14, 10 and 8 years, respectively. In all three countries rates of expected lifetime treatment are higher and median treatment delays longer than those presented in a comparable study in The Netherlands: 6.5%, 36.9%, 1 and 4 years, respectively (ten Have et al., 2013). Studies of the natural history and course of AUD suggest many people, particularly those at the lower end of the severity spectrum, will remit over lifetime without treatment (Cunningham and McCambridge, 2012). One could therefore argue that expected treatment rates presented here may be appropriate, especially given that in this study, as with previous studies, alcohol dependence is associated with shorter delay and higher likelihood of treatment contact than alcohol abuse without dependence. Similarly, the consistent finding that more recent cohorts are more likely to seek treatment may positively reflect increased availability of services or changing attitudes towards treatment (ten Have et al., 2013). However, median delays of 14 and 23 years among those
who eventually make treatment contact are long, and the implications of almost two decades of untreated AUD and related harms substantial, particularly in a country with some of the highest AUD prevalence rates in the world (Teesson et al., 2010). It is likely that factors contributing to these long delays are attitudinal, financial and structural (Keyes et al., 2010a; Mojtabai et al., 2011). However, given evidence for the effectiveness of brief interventions, particularly early in the course of AUD (Babor et al., 2007; Jonas et al., 2012) it is imperative to better understand and address these barriers to care for Australians with AUD. Within the context of these broad implications there are several specific findings that warrant discussion. Consistent with previous reports, earlier onset of AUD was associated with lowered treatment rates and longer delay (Bruffaerts et al., 2007; Hingson et al., 2006; ten Have et al., 2013; Wang et al., 2005). This may reflect ‘normalisation’ of risky alcohol use among young people (Davies et al., 2013), the need for parental involvement in treatment among adolescents (ten Have et al., 2013) or reluctance of clinicians to ask
120
C. Chapman et al. / Drug and Alcohol Dependence 147 (2015) 116–121
young people about alcohol use (Hingson et al., 2013). Given the substantial impact of alcohol related harms among young people (Keyes et al., 2012), and amid increasing evidence for the effectiveness of prevention in delaying onset of drinking and reducing these harms, the findings highlight the continued need to evaluate and fund prevention of AUD among adolescents and young adults (Teesson et al., 2012). Two novel findings from the study deserve particular attention. These are the findings related to symptom type and treatment delay, and comorbidity and treatment delay. Recurrent social or interpersonal problems (among those with abuse) and role disruption (among those with dependence) were both associated with shorter treatment delay and greater likelihood of treatment over lifetime. Previous studies on the treatment seeking for AUD have found social consequences of drinking to be strong predictors of perceived need for, and lifetime seeking of treatment, among both clinical and community samples (Cohen et al., 2007; Dawson et al., 2012; Edlund et al., 2009; Hedden and Gfroerer, 2011; Saunders et al., 2006). To the best of our knowledge, this is the first large community study to suggest that these same factors are associated with seeking treatment sooner after alcohol related problems begin. These consequences are often emphasised in the content of effective prevention (Champion et al., 2013) and screening and brief intervention programmes (Heather, 2012). This study lends further support to the importance of these factors as determinants of helpseeking and therefore as targets for education about alcohol use and its consequences. Whilst previous population studies have indicated that comorbidity is associated with higher rates of lifetime treatment seeking among those with mental and substance use disorders (Cohen et al., 2007; Wang et al., 2007a) the role of comorbidity as a predictor of treatment delay for AUD has not commonly been examined. In this study comorbid drug use disorder, with onset either before or after AUD symptoms was associated with shorter treatment delay among those with alcohol abuse. It is possible that social and/or legal problems associated with use of illicit substances increases the likelihood of more timely treatment for AUD (Grella et al., 2009) or that greater perceived need among those with comorbid drug use disorders results in shorter treatment delay (Hedden and Gfroerer, 2011). It is less clear why comorbid drug use disorder was not associated with treatment delay among those with alcohol dependence. It could be that among those at the more severe end of the AUD spectrum the impact of comorbid drug use disorders is small, particularly given models also controlled for the number and type of symptoms as well as comorbid mental disorders. Comorbid anxiety but not mood disorder was associated with shorter treatment delay among those with alcohol abuse and dependence. Mood and anxiety disorders are generally associated with higher lifetime treatment rates (Bunting et al., 2012; ten Have et al., 2013; Wang et al., 2007a) and greater perceived need for treatment (Edlund et al., 2009; Mojtabai et al., 2011) than AUD and the findings with respect to comorbid anxiety disorder may well reflect this. However, the fact that comorbid mood disorder was not associated with shorter treatment delay or higher probability of treatment is perhaps surprising. One possibility is that people with comorbid AUD and mood disorder are seeking treatment for their mood disorder as opposed to their AUD (Edlund et al., 2012). However, Olfson et al. (2012) using data from the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC) in the United States, conducted a similar study of treatment delay among people with major depression. Using comparable analyses they determined that comorbid AUD was not associated with shorter delay to first treatment. Ilgen et al. (2011) similarly reported another study from the NESARC examining the longitudinal predictors of treatment contact at Wave 2, 3 years after Wave 1, and determined that comorbid anxiety, but not mood disorder was
associated with subsequent treatment for AUD. Mood disorders and AUD commonly co-occur (Hasin et al., 2007), and this comorbidity is associated with significant impairment and poorer prognosis (Burns et al., 2005). Taken together, the findings of this and other studies clearly argue that those with comorbid AUD and mood disorder should be a priority for treatment earlier, rather than later in the course of illness. The findings of the study need to be considered within the context of some limitations. Firstly, the survey did not differentiate between treatment contact for alcohol and drug-related problems. Whilst the majority of the sample (76%) met criteria for an AUD but not another drug use disorder, we cannot determine whether the remainder of the sample was seeking treatment for alcohol or drug-related problems. Secondly, the survey was cross-sectional and therefore relied on retrospective recall of illness onset and treatment seeking. Despite the fact that the survey used similar methodology to that used in the World Mental Health Surveys to improve accuracy of recall (Kessler et al., 2007), the estimates are subject to biases including the general limitations of memory and the cognitive schemas people might have about the actual versus typical age of illness onset (Simon and VonKorff, 1995). The limited dating of events also makes it difficult to model sociodemographic and other factors as truly time-varying covariates. The relationship between specific symptom endorsement and treatment delay for AUD for example, would need to be modelled prospectively to truly capture the relationship between experience of symptoms and treatment seeking over time. Thirdly, increased mortality from years of chronic heavy drinking cannot be assessed in a crosssectional survey, so it is not possible to comment on treatment delays for those most severely affected by alcohol-related physical conditions. However, given the low rates of treatment seeking among the general population of people with AUD and the long treatment delays, prospective studies would need to follow people up for very long periods of time to capture their first treatment episode. Within this context, cross-sectional population studies such as this one make an important contribution. To the best of our knowledge, this is the first study to simultaneously examine the relationship between symptom onset, birth cohort, number and type of symptoms, comorbidity and treatment delay for AUD in a large representative population sample. While lifetime probability of treatment contact was found to be relatively high, median treatment delays were long. Even if some individuals recover without formal intervention, the implications of almost two decades of untreated AUD and related harms are likely to be substantial. The study highlights the need to better understand barriers to AUD treatment and points to the continued need to invest in prevention as well as screening and brief intervention early in the course of AUD. Specifically, those with earlier onset of symptoms and those with comorbid mood disorder should be a target for earlier treatment. Role of funding source The 2007 National Survey of Mental Health and Wellbeing (NSMHWB) was funded by the Australian Government and conducted by the Australian Bureau of Statistics. The Centre of Research Excellence in Mental Health and Substance Use is funded by the National Health and Medical Research Council. The National Drug and Alcohol Research Centre is supported by funding from the Australian Government under the Substance Misuse Prevention and Service Improvements Grants Fund. Contributors Dr Chapman conceptualised the research question in consultation with the other authors and drafted the initial manuscript.
C. Chapman et al. / Drug and Alcohol Dependence 147 (2015) 116–121
Dr Chapman and A/Prof Slade conducted the analysis. All authors took an active role in interpreting the results and revising the manuscript. All authors have seen and approved the final version of the manuscript. Conflict of interest No conflict declared. Acknowledgements The authors would like to thank the NSMHWB reference group for their input in the survey’s design. The authors would also like to thank all those who participated in the survey. References Alvanzo, A.A., Storr, C.L., Mojtabai, R., Green, K.M., Pacek, L.R., La Flair, L.N., Cullen, B.A., Crum, R.M., 2014. Gender and race/ethnicity differences for initiation of alcohol-related service use among persons with alcohol dependence. Drug Alcohol Depend. 140, 48–55. Babor, T.F., McRee, B.G., Kassebaum, P.A., Grimaldi, P.L., Ahmed, K., Bray, J., 2007. Screening, brief intervention, and referral to treatment (SBIRT): toward a public health approach to the management of substance abuse. Subst. Abuse 28, 7–30. Borges, G., Wang, P.S., Medina-Mora, M.E., Lara, C., Chiu, W.T., 2007. Delay of first treatment of mental and substance use disorders in Mexico. Am. J. Public Health 97, 1638–1643. Bruffaerts, R., Bonnewyn, A., Demyttenaere, K., 2007. Delays in seeking treatment for mental disorders in the Belgian general population. Soc. Psychiatry Psychiatr. Epidemiol. 42, 937–944. Bunting, B.P., Murphy, S.D., O’Neill, S.M., Ferry, F.R., 2012. Lifetime prevalence of mental health disorders and delay in treatment following initial onset: evidence from the Northern Ireland Study of Health and Stress. Psychol. Med. 42, 1727–1739. Burns, L., Teesson, M., O’Neill, K., 2005. The impact of comorbid anxiety and depression on alcohol treatment outcomes. Addiction 100, 787–796. Champion, K.E., Newton, N.C., Barrett, E.L., Teesson, M., 2013. A systematic review of school-based alcohol and other drug prevention programs facilitated by computers or the internet. Drug Alcohol Rev. 32, 115–123. Cohen, E., Feinn, R., Arias, A., Kranzler, H.R., 2007. Alcohol treatment utilization: findings from the National Epidemiologic Survey on Alcohol and Related Conditions. Drug Alcohol Depend. 86, 214–221. Cunningham, J.A., McCambridge, J., 2012. Is alcohol dependence best viewed as a chronic relapsing disorder? Addiction 107, 6–12. Davies, E.L., Martin, J., Foxcroft, D.R., 2013. Young people talking about alcohol: focus groups exploring constructs in the prototype willingness model. Drugs Educ. Prev. Policy 20, 269–277. Dawson, D.A., Goldstein, R.B., Grant, B.F., 2012. Factors associated with first utilization of different types of care for alcohol problems. J. Stud. Alcohol Drugs 73, 647–656. Edlund, M.J., Booth, B.M., Feldman, Z.L., 2009. Perceived need for treatment for alcohol use disorders: results from two national surveys. Psychiatr. Serv. 60, 1618–1628. Edlund, M.J., Booth, B.M., Han, X., 2012. Who seeks care where? Utilization of mental health and substance use disorder treatment in two national samples of individuals with alcohol use disorders. J. Stud. Alcohol Drugs 73, 635–646. Grambsch, P.M., Therneau, T.M., 1994. Proportional hazards tests and diagnostics based on weighted residuals. Biometrika 81, 515–526. Grella, C.E., Karno, M.P., Warda, U.S., Moore, A.A., Niv, N., 2009. Perceptions of need and help received for substance dependence in a national probability survey. Psychiatr. Serv. 60, 1068–1074. Hasin, D.S., Stinson, F.S., Ogburn, E., Grant, B.F., 2007. Prevalence, correlates, disability, and comorbidity of DSM-IV alcohol abuse and dependence in the United States: results from the National Epidemiologic Survey on Alcohol and Related Conditions. Arch. Gen. Psychiatry 64, 830–842. Heather, N., 2012. Can screening and brief intervention lead to population-level reductions in alcohol-related harm? Addict. Sci. Clin. Pract. 7, 15. Hedden, S.L., Gfroerer, J.C., 2011. Correlates of perceiving a need for treatment among adults with substance use disorder: results from a national survey. Addict. Behav. 36, 1213–1222. Hingson, R.W., Heeren, T., Winter, M.R., 2006. Age of alcohol-dependence onset: associations with severity of dependence and seeking treatment. Pediatrics 118, e755–e763. Hingson, R.W., Zha, W., Iannotti, R.J., Simons-Morton, B., 2013. Physician advice to adolescents about drinking and other health behaviors. Pediatrics 131, 249–257. Hosmer, D.W., Lemeshow, S., 1999. Applied Survival Analysis. John Wiley & Sons, Inc., New York. Ilgen, M.A., Price, A.M., Burnett-Zeigler, I., Perron, B., Islam, K., Bohnert, A.S., Zivin, K., 2011. Longitudinal predictors of addictions treatment utilization in treatmentnaive adults with alcohol use disorders. Drug Alcohol Depend. 113, 215–221. Jonas, D.E., Garbutt, J.C., Amick, H.R., Brown, J.M., Brownley, K.A., Council, C.L., Viera, A.J., Wilkins, T.M., Schwartz, C.J., Richmond, E.M., Yeatts, J., Evans, T.S., Wood, S.D., Harris, R.P., 2012. Behavioral counseling after screening for alcohol misuse
121
in primary care: a systematic review and meta-analysis for the U.S. Preventive Services Task Force. Ann. Intern. Med. 157, 645–654. Kerr, W.C., Greenfield, T.K., Ye, Y., Bond, J., Rehm, J., 2013. Are the 1976–1985 birth cohorts heavier drinkers? Age-period-cohort analyses of the national alcohol surveys 1979–2010. Addiction 108, 1038–1048. Kessler, R.C., Aguilar-Gaxiola, S., Berglund, P.A., Caraveo-Anduaga, J.J., DeWit, D.J., Greenfield, S.F., Kolody, B., Olfson, M., Vega, W.A., 2001. Patterns and predictors of treatment seeking after onset of a substance use disorder. Arch. Gen. Psychiatry 58, 1065–1071. Kessler, R.C., Amminger, G.P., Aguilar-Gaxiola, S., Alonso, J., Lee, S., Ustun, T.B., 2007. Age of onset of mental disorders: a review of recent literature. Curr. Opin. Psychiatry 20, 359–364. Kessler, R.C., Olfson, M., Berglund, P.A., 1998. Patterns and predictors of treatment contact after first onset of psychiatric disorders. Am. J. Psychiatry 155, 62–69. Kessler, R.C., Ustun, T.B., 2004. The World Mental Health (WMH) survey initiative version of the World Health Organization (WHO) Composite International Diagnostic Interview (CIDI). Int. J. Methods Psychiatr. Res. 13, 93–121. Kessler, R.C., Ustun, T.B., 2008. The WHO World Mental Health Surveys: Global Perspectives on the Epidemiology of Mental Disorders. Cambridge University Press, Cambridge. Keyes, K.M., Hatzenbuehler, M.L., McLaughlin, K.A., Link, B., Olfson, M., Grant, B.F., Hasin, D., 2010a. Stigma and treatment for alcohol disorders in the United States. Am. J. Epidemiol. 172, 1364–1372. Keyes, K.M., Martins, S.S., Blanco, C., Hasin, D.S., 2010b. Telescoping and gender differences in alcohol dependence: new evidence from two national surveys. Am. J. Psychiatry 167, 969–976. Keyes, K.M., Schulenberg, J.E., O’Malley, P.M., Johnston, L.D., Bachman, J.G., Li, G., Hasin, D., 2012. Birth cohort effects on adolescent alcohol use: the influence of social norms from 1976 to 2007. Arch. Gen. Psychiatry 69, 1304–1313. Lee, S., Fung, S.C., Tsang, A., Zhang, M.Y., Huang, Y.Q., He, Y.L., Liu, Z.R., Shen, Y.C., Kessler, R.C., 2007. Delay in initial treatment contact after first onset of mental disorders in metropolitan China. Acta Psychiatr. Scand. 116, 10–16. Mojtabai, R., Olfson, M., Sampson, N.A., Jin, R., Druss, B., Wang, P.S., Wells, K.B., Pincus, H.A., Kessler, R.C., 2011. Barriers to mental health treatment: results from the National Comorbidity Survey Replication. Psychol. Med. 41, 1751–1761. Naughton, F., Alexandrou, E., Dryden, S., Bath, J., Giles, M., 2013. Understanding treatment delay among problem drinkers: what inhibits and facilitates help-seeking? Drug Educ. Prev. Policy 20, 297–303. Olfson, M., Kessler, R.C., Berglund, P.A., Lin, E., 1998. Psychiatric disorder onset and first treatment contact in the United States and Ontario. Am. J. Psychiatry 155, 1415–1422. Olfson, M., Liu, S.M., Grant, B.F., Blanco, C., 2012. Influence of comorbid mental disorders on time to seeking treatment for major depressive disorder. Med Care 50, 227–232. Saunders, S.M., Zygowicz, K.M., D’Angelo, B.R., 2006. Person-related and treatmentrelated barriers to alcohol treatment. J. Subst. Abuse Treat. 30, 261–270. Simon, G.E., VonKorff, M., 1995. Recall of psychiatric history in cross-sectional surveys: implications for epidemiologic research. Epidemiol. Rev. 17, 221–227. Slade, T., Johnston, A., Oakley Browne, M.A., Andrews, G., Whiteford, H., 2009. 2007 National Survey of Mental Health and Wellbeing: methods and key findings. Aust. N. Z. J. Psychiatry 43, 594–605. StataCorp, 2012. Stata Statistical Software: Release 12. StataCorp LP, College Station, TX. Teesson, M., Hall, W., Slade, T., Mills, K., Grove, R., Mewton, L., Baillie, A., Haber, P., 2010. Prevalence and correlates of DSM-IV alcohol abuse and dependence in Australia: findings of the 2007 National Survey of Mental Health and Wellbeing. Addiction 105, 2085–2094. Teesson, M., Newton, N.C., Barrett, E.L., 2012. Australian school-based prevention programs for alcohol and other drugs: a systematic review. Drug Alcohol Rev. 31, 731–736. ten Have, M., de Graaf, R., van Dorsselaer, S., Beekman, A., 2013. Lifetime treatment contact and delay in treatment seeking after first onset of a mental disorder. Psychiatr. Serv. 64, 981–989. Wang, P.S., Aguilar-Gaxiola, S., Alonso, J., Angermeyer, M.C., Borges, G., Bromet, E.J., Bruffaerts, R., de Girolamo, G., de Graaf, R., Gureje, O., Haro, J.M., Karam, E.G., Kessler, R.C., Kovess, V., Lane, M.C., Lee, S., Levinson, D., Ono, Y., Petukhova, M., Posada-Villa, J., Seedat, S., Wells, J.E., 2007a. Use of mental health services for anxiety, mood, and substance disorders in 17 countries in the WHO world mental health surveys. Lancet 370, 841–850. Wang, P.S., Angermeyer, M., Borges, G., Bruffaerts, R., Tat Chiu, W., Girolamo, D.E.G., Fayyad, O., Gureje, J., Haro, J.M., Huang, Y., Kessler, R.C., Kovess, V., Levinson, D., Nakane, Y., Oakley Brown, M.A., Ormel, J.H., Posada-Villa, J., Aguilar-Gaxiola, S., Alonso, J., Lee, S., Heeringa, S., Pennell, B.E., Chatterji, S., Ustun, T.B., 2007b. Delay and failure in treatment seeking after first onset of mental disorders in the World Health Organization’s World Mental Health Survey Initiative. World Psychiatry 6, 177–185. Wang, P.S., Berglund, P., Olfson, M., Pincus, H.A., Wells, K.B., Kessler, R.C., 2005. Failure and delay in initial treatment contact after first onset of mental disorders in the National Comorbidity Survey Replication. Arch. Gen. Psychiatry 62, 603–613. Whiteford, H.A., Degenhardt, L., Rehm, J., Baxter, A.J., Ferrari, A.J., Erskine, H.E., Charlson, F.J., Norman, R.E., Flaxman, A.D., Johns, N., Burstein, R., Murray, C.J., Vos, T., 2013. Global burden of disease attributable to mental and substance use disorders: findings from the Global Burden of Disease Study 2010. Lancet 382, 1575–1586. Witkiewitz, K., Dearing, R.L., Maisto, S.A., 2014. Alcohol use trajectories among nontreatment-seeking heavy drinkers. J. Stud. Alcohol Drugs 75, 415–422.
Contents lists available at ScienceDirect
Drug and Alcohol Dependence journal homepage: www.elsevier.com/locate/drugalcdep
Delay to first treatment contact for alcohol use disorder Cath Chapman a,∗ , Tim Slade a , Caroline Hunt b , Maree Teesson a a NHMRC Centre of Research Excellence in Mental Health and Substance Use (CREMS), National Drug and Alcohol Research Centre (NDARC), University of New South Wales, Sydney, NSW 2052, Australia b School of Psychology, University of Sydney, NSW 2006, Australia
a r t i c l e
i n f o
Article history: Received 11 August 2014 Received in revised form 25 November 2014 Accepted 28 November 2014 Available online 10 December 2014 Keywords: Alcohol abuse Alcohol dependence Treatment delay Service utilization
a b s t r a c t Background: This study explored the patterns and correlates of time to first treatment contact among people with alcohol use disorder (AUD) in Australia. Specifically it examined the relationship between sex, birth cohort, onset of AUD symptoms, severity, comorbidity, symptom type and time to first treatment contact (treatment delay) among those with alcohol abuse and dependence in a large population sample. Methods: Data came from the 2007 Australian National Survey of Mental Health and Wellbeing (N = 8841). A modified version of the World Health Organization’s Composite International Diagnostic Interview was used to determine the presence and age of onset of DSM-IV AUD and other mental disorders and the age at which respondents first sought treatment for alcohol or other drug-related problems. Results: Median time to first treatment contact for an AUD was 18 years (14 years dependence, 23 years abuse). Projected lifetime treatment rates were 78.1% for alcohol dependence and 27.5% for abuse. Those with earlier onset and from older cohorts reported longer delay and were less likely to ever seek treatment compared to those with later onset or from more recent cohorts. Those with comorbid anxiety but not mood disorder, or who reported alcohol-related role disruption or recurrent interpersonal problems were more likely to ever seek treatment and reported shorter delay compared to those who did not report these symptoms. Conclusions: Treatment delay for alcohol use disorder in Australia is substantial. Those with earlier onset and those with comorbid mood disorder should be a target for earlier treatment. © 2014 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Alcohol use disorders (AUD) are common. Prevalence estimates in the United States, the United Kingdom and Australia indicate between one in four and one in six adults in the population will meet criteria for DSM-IV alcohol abuse or dependence over their lifetime (Bunting et al., 2012; Hasin et al., 2007; Teesson et al., 2010). AUD is associated with substantial negative social and health consequences and poses significant public health concern (Whiteford et al., 2013). However, despite the high prevalence and negative impact of AUD, and the existence of effective interventions (Dawson et al., 2012; Jonas et al., 2012), most people with an AUD do not seek treatment (Edlund et al., 2012; Hasin et al., 2007). Moreover those who do, typically delay seeking treatment for many years following the onset of symptoms. Large community studies
∗ Corresponding author at: National Drug and Alcohol Research Centre, University of New South Wales, 22-32 King Street, Randwick Campus, Sydney, NSW 2052, Australia. Tel.: +61 2 93850333; fax: +61 2 9385 0222. E-mail address: [email protected] (C. Chapman). http://dx.doi.org/10.1016/j.drugalcdep.2014.11.029 0376-8716/© 2014 Elsevier Ireland Ltd. All rights reserved.
have reported median delays to first treatment contact of between 6 and 18 years after the onset of problems associated with alcohol (Bruffaerts et al., 2007; Bunting et al., 2012; Kessler et al., 2001; Keyes et al., 2010b; Wang et al., 2005, 2007b). Even if it is argued that a proportion of people with AUD will recover naturally over the course of their lives, these long delays are thought to represent considerable unmet need for care (ten Have et al., 2013; Witkiewitz et al., 2014). Identification of the factors associated with delay to seek treatment is key to understanding how to reduce these delays and lessen this unmet need. Several large community studies have examined the factors associated with delay to seek treatment for AUD over lifetime. Generally these studies have reported longer delays and lowered odds of ever seeking treatment among those with earlier onset of symptoms (Bruffaerts et al., 2007; Hingson et al., 2006; Kessler et al., 2001, 1998; Olfson et al., 1998; ten Have et al., 2013; Wang et al., 2005) and shorter delays and higher odds of ever seeking treatment among more recent cohorts (Bruffaerts et al., 2007; Kessler et al., 2001, 1998; Olfson et al., 1998; Wang et al., 2005) – although three studies have reported no effect of one or both of these variables (Borges et al., 2007; Keyes et al., 2010b; Lee et al., 2007).
C. Chapman et al. / Drug and Alcohol Dependence 147 (2015) 116–121
With the exception of one study (Alvanzo et al., 2014), previous community studies have not reported sex differences in delay to seek treatment. No other predictors of treatment delay have been consistently examined or reported. These studies are all based on large representative community surveys and are therefore a robust source of information. However, commonly these studies focused on one or two variables or examined only sociodemographic factors, thereby missing the impact of potentially important factors such as comorbidity, symptom type and severity. Both comorbidity and severity have been shown to be associated with a greater likelihood of seeking treatment among those with mental and substance use disorders (Cohen et al., 2007; Edlund et al., 2012; Ilgen et al., 2011; Wang et al., 2007a), and previous studies on treatment seeking for AUD have found particular symptoms or consequences of alcohol use to be associated increased odds of treatment seeking (Dawson et al., 2012; Naughton et al., 2013; Saunders et al., 2006). It is reasonable to expect that that these factors may also be associated with shorter treatment delays among those with an AUD. The present study sought to address this gap by simultaneously examining the relationship between sex, birth cohort, age of onset, comorbidity, severity and symptom type and delay to first treatment contact for AUD in a large representative community sample in Australia. Australia is a country with high rates of alcohol use and dependence (Teesson et al., 2010). This is the first time data on treatment delay among those with AUD in Australia has been reported. 2. Methods 2.1. Sample The 2007 Australian National Survey of Mental Health and Wellbeing (NSMHWB) is a nationally representative population survey with a sample size of 8841 (Slade et al., 2009). Respondents were selected at random from a stratified, multistage area probability sample of persons aged 16–85 years living in private dwellings and data were weighted according to the inverse probability of being selected. Interviews were conducted in respondent’s households using a computerassisted personal interview (CAPI). The survey received a response rate of 60%, which is commensurate with other major national surveys in mental health and substance use (Kerr et al., 2013; Kessler, 2008). 2.2. Measurement of AUD Experience of DSM-IV alcohol abuse, dependence and other mental disorders was assessed using a modified version of the World Health Organization’s Composite International Diagnostic Interview (WMH-CIDI; Kessler and Ustun, 2004), a highly structured interview with questions designed to operationalise the diagnostic criteria for each mental disorder. Respondents who reported they had consumed at least 12 alcoholic drinks in any 1 year over lifetime were asked if they had ever drunk alcohol on 3 or more days per week and/or have usually consumed 3+ drinks on the days they were drinking. AUD was assessed among respondents who answered ‘yes’ to this question (n = 5520; 64% of sample). A series of 18 questions operationalised the four alcohol abuse (major role disruption; hazardous use; recurrent legal consequences; and recurrent social or interpersonal problems) and seven alcohol dependence (tolerance; withdrawal; larger amounts/longer period of drinking; difficulty cutting down; significant time obtaining alcohol; important activities given up; and continued use despite problems) criteria. The sample for the present study comprised all respondents who met criteria for lifetime DSM-IV alcohol abuse or dependence (n = 1847). 2.3. Measurement of treatment delay Respondents who completed the alcohol module of the survey were asked about the age at which they first experienced symptoms of alcohol abuse (1+ symptoms in any year) or dependence (3+ symptoms in any year) and this was defined as the age of onset of AUD. Respondents were also asked if they had ever talked to a medical doctor or other professional (psychologist, social worker, counsellor, herbalist, acupuncturist or other healing professional) about their use of alcohol or drugs and if so, how old they were the first time they did so. Treatment contact for alcoholrelated problems was not differentiated from treatment contact for drug-related problems. Treatment delay was defined as the number of years between onset of AUD and first treatment contact.
117
2.4. Covariates Age at onset of AUD symptoms was coded in four categories of early onset (<19 years), early average (19–20 years), late average (21–29 years), and late onset (>29 years) based on the distribution of the age of onset (Wang et al., 2005). Birth cohort was defined in 10 year bands based on age at interview. Severity was examined in terms of whether respondents met criteria for alcohol abuse only or for alcohol dependence (with or without abuse) and in terms of the total number of abuse and/or dependence symptoms endorsed (Hingson et al., 2006). Comorbidty with DSM-IV mood (depression, dysthymia, bipolar disorder), anxiety (panic disorder, agoraphobia, social phobia, generalised anxiety disorder, posttraumatic stress disorder) and other drug use disorders was examined in two ways; (i) whether individuals met criteria for at least one disorder in each of these three categories, and (ii) whether the onset of the comorbid disorder pre-dated the onset of AUD symptoms. In these models, people who reported onset of a comorbid disorder after first treatment contact for AUD were excluded (Keyes et al., 2010b). Each of the four abuse and seven dependence symptoms were included as dichotomous variables to examine whether endorsement of specific symptoms was related to treatment delay (Kessler et al., 2001).
2.5. Statistical analysis Survey procedures in Stata Release 12 (StataCorp, 2012) were used for all analyses and standard errors obtained through the delete-a-group jack-knife variance technique to account for complex sampling procedures. Projected lifetime probability of treatment contact, proportion who made contact within a year and median duration of delay were obtained using Kaplan–Meier survival estimates among all respondents with an AUD, and separately for those with abuse (n = 1500) and with dependence with or without abuse (n = 347). This method of analysis allows for data to be modelled over time to include all respondents including those who had not sought treatment at the time of interview, often referred to as ‘censored cases’ (Hosmer, 1999). Survival time was defined as the number of years from onset of AUD to age at first treatment contact, or to age at interview, whichever came first. The relationship between various covariates and survival time was examined in logistic regression models using discrete-time survival analysis where sex, birth cohort, age at onset of AUD, number of symptoms, type of symptoms and presence of comorbid mood, anxiety and other drug use disorders were treated as covariates predicting treatment delay. Models were run separately for respondents with alcohol abuse and dependence. Individual contributions of covariates to survival time were assessed through Wald F-statistics and associated p-values and estimates of likelihood of treatment seeking in any year over lifetime are expressed as odds ratios (OR) with 95% confidence intervals, and interpreted in a similar way to hazard ratios. The effect of time in both models was approximated by including a linear term. Onset age of regular drinking (defined as 12+ drinks in any 1 year period) was included in final models to control for potential confounding on onset of AUD. Preliminary modelling ensured that models met the assumption of proportionality of hazards and where this assumption was not met models included interaction terms between covariates and time (Grambsch and Therneau, 1994).
3. Results 3.1. Lifetime probability of treatment contact and treatment delay Fig. 1 displays Kaplan–Meier failure curves for the cumulative lifetime probability of treatment contact after onset of AUD stratified by alcohol abuse and dependence. Table 1 displays the expected lifetime treatment rates, proportion who made contact within 1 year of onset and median duration of delay among those who eventually made treatment contact. Just over one third of people with an AUD were estimated to eventually make treatment contact with a median treatment delay of 18 years among those who did. Rates of expected lifetime treatment contact were lower and estimated duration of delay longer for those with alcohol abuse compared to those with alcohol dependence (Wald F = 12.0, p = 0.001). After adjusting for the effects of sex, birth cohort, onset of AUD, number of symptoms and the presence of comorbid disorders, those with alcohol dependence were 2.4 times more likely than those with alcohol abuse to make treatment contact (OR = 2.4, 95% CI 1.4–3.9). Median treatment delays among those with alcohol abuse and dependence were 23 years and 14 years, respectively.
118
C. Chapman et al. / Drug and Alcohol Dependence 147 (2015) 116–121
Table 1 Proportional treatment contact within 1 year of onset, projected lifetime probability of treatment contact and median duration of delay among people with alcohol use disorders in the Australian population. Lifetime treatment for AUD
Alcohol abuse (N = 1500) Alcohol dependence (N = 347) Alcohol use disorder (N = 1847) a
Contact w/in 1 year %
Projected lifetime probability of contact %
Median duration of delaya Years
1.3 8.9 2.6
27.5 78.1 34.6
23 14 18
Median delay in years among those who eventually made treatment contact.
obligations were 2.3 times more likely to make treatment contact than those who did not (Wald F = 5.58, p = 0.022; 9 vs 29 years).
Fig. 1. Cumulative Lifetime probability of treatment contact for persons with AUD in the Australian population, stratified by abuse and dependence.
3.2. Predictors of treatment delay Table 2 presents the factors associated with treatment delay expressed as adjusted ORs for probability of treatment contact in any year among those with alcohol abuse and those with alcohol dependence. 3.2.1. Sex and birth cohort. Sex was not associated with treatment delay in either model. Individuals from more recent birth cohorts were significantly more likely to make treatment contact and to experience shorter treatment delay compared to individuals born before 1958 in both models (abuse: Wald F = 3.68, p = 0.017; 7–10 vs 29 years; dependence: Wald F = 5.23, p = 0.003; 4–10 vs 21 years). 3.2.2. Onset of AUD symptoms. Those with early onset of AUD symptoms were significantly less likely to make treatment contact and to experience longer treatment delay, than those with late onset, among those with alcohol abuse (Wald F = 2.81, p = 0.047; 7–19 vs 6 years) and dependence (Wald F = 5.99, p = 0.001; 13–16 vs 5 years). 3.2.3. Number and type of AUD symptoms. Among those with alcohol abuse the total number of abuse symptoms endorsed was not associated with treatment delay (Wald F = 0.86, p = 0.470). Similarly, among those with alcohol dependence the total number of abuse (Wald F = 0.85, p = 0.472) or dependence symptoms (Wald F = 1.53, p = 0.207) endorsed was not associated with treatment delay. Among those with alcohol abuse, individuals who endorsed recurrent social or interpersonal problems were 1.8 times more likely to make treatment contact and reported shorter treatment delay than those who reported never experiencing this symptom (Wald F = 4.8, p = 0.032; 21 vs 23 years). Among people with alcohol dependence those who endorsed failure to fulfil important role
3.2.4. Comorbidity. The presence of comorbid mood disorder was not associated with treatment delay among those with alcohol abuse or dependence (Wald F = 3.30, p = 0.075; Wald F = 0.57, p = 0.452, respectively). Conversely, comorbid anxiety disorder was associated with shorter treatment delay among those with abuse (Wald F = 6.74, p = 0.012) and dependence (Wald F = 4.65, p = 0.035). When the temporal relationship between anxiety disorder onset and AUD onset was considered, anxiety disorders that predated AUD symptom onset were significantly associated with shorter treatment delay and greater likelihood of treatment contact among those with abuse (OR = 2.1; 95% CI: 1.1–4.1; Wald F = 4.00, p = 0.024; 14 vs 19 years) and dependence (OR = 2.6; 95% CI: 1.2–5.7; p = 0.015; 11 vs 19 years) compared to those without a comorbid anxiety disorder. The presence of a comorbid drug use disorder was significantly associated with shorter treatment delay among those with alcohol abuse (Wald F = 23.36, p < 0.001) but not dependence (Wald F = 0.25, p = 0.619). Among those with alcohol abuse, both pre-existing drug use disorders and those with onset after AUD were significantly associated with shorter treatment delay and greater likelihood of treatment contact (OR = 4.4; 95% CI: 2.1–9.0; OR = 5.1; 95% CI: 2.0–12.9, Wald F = 11.06, p < 0.001; 8–19 vs 27 years). 4. Discussion The present study estimated that one in three people with an AUD in Australia will make treatment contact over lifetime and those who do will delay seeking treatment for a median of 18 years after onset of AUD symptoms. Expected rates of 1 year and lifetime treatment contact are higher among those with alcohol dependence than those with alcohol abuse and shorter median delay. Individuals from more recent cohorts, with later onset, or with comorbid anxiety disorder are more likely to make treatment contact and to do so sooner after onset. Among those with alcohol abuse, individuals with comorbid drug use disorder and those who experience role disruption are also more likely to make treatment contact and to experience shorter treatment delay. Among people with alcohol dependence, those who experience recurrent social or interpersonal problems are more likely to make treatment contact than those who do not. Sex, number of AUD symptoms and comorbid mood disorder were not associated with treatment delay. Previous estimates of treatment delay for AUD vary substantially, even among studies with comparable methodology. Within this context, the median treatment delays for AUD in Australia are longer than estimates from other countries. For alcohol abuse, the probability of ever seeking treatment in Australia of 27.5% is substantially lower than estimates from the United Kingdom: 74.6% (Bunting et al., 2012) and the United States: 52.7% (Wang et al., 2005) and the median treatment delay substantially longer (23 years vs 8 and 16 years). This difference is smaller for alcohol dependence where lifetime probability of treatment seeking is 78.1% in
C. Chapman et al. / Drug and Alcohol Dependence 147 (2015) 116–121
119
Table 2 Factors associated with delay to first treatment contact among people with alcohol use disorders in the Australian population. Likelihood of treatment contact Alcohol abuse Odds ratio
Alcohol dependence 95% CI
p-value
Odds ratio
95% CI
p-value
0.763
0.7
0.4–1.6
5.6* 3.8* 3.1* 1.0
2.2–14.0 1.7–8.2 1.3–7.1 –
0.2* 0.2* 0.4 1.0
0.1–0.7 0.1–0.5 0.1–1.4 –
0.470 – 0.123 0.498 0.411
1.0 2.0 2.2 2.0
– 0.6–6.2 0.8–5.7 0.8–5.3
0.472 – 0.240 0.110 0.156
– – – – –
1.0 1.7 1.6 2.3 2.8*
– 0.6–5.0 0.7–3.8 0.9–6.1 1.2–8.3
0.207 – 0.355 0.231 0.092 0.019
0.203 0.914 0.897 0.032
2.3* 0.3 0.5 1.3
1.1–4.7 0.1–1.1 0.2–1.4 0.6–2.9
0.022 0.067 0.169 0.551
– – – – – – –
1.3 1.4 – 1.3 1.0 1.3 1.9
0.7–2.4 0.7–3.0 – 0.6–2.7 0.4–2.5 0.7–2.2 0.9–4.3
0.487 0.349 – 0.514 0.983 0.410 0.103
0.075 0.012 0.000
1.2 2.1* 0.8
0.7–2.0 1.1–4.1 0.4–1.7
0.452 0.035 0.619
Sex 0.9
0.5–1.8
Birth cohort 1978–1991 1968–1977 1958–1967 < 1958
5.5* 2.5* 1.1 1.0
1.4–21.6 1.1–6.0 0.4–2.9 –
Onset of AUD symptoms Early Early average Late average Late
0.4* 0.5 0.3* 1.0
0.1–0.9 0.2–1.2 0.1–0.7 –
Number of abuse symptoms One Two Three Four
1.0 1.8 1.3 1.9
– 0.9–3.7 0.6–3.0 0.4–9.2
Number of dependence symptoms Three Four Five Six Seven
– – – – –
Type of abuse symptomsa Failure to fulfil role Drinking in hazardous situations Recurrent legal problems Recurrent social/interpersonal problems
1.5 0.9 1.0 1.8*
Type of dependence symptomsa Tolerance Withdrawal Larger amounts/longer periodsb Unable to cut-down Time spent obtaining alcohol Important activities given up Persistent physical/psychological problem
– – – – – – –
Comorbid mental disorderc Any mood disorder Any anxiety disorder Any drug use disorder
1.8 2.0* 4.7*
Female
a b c *
– – – – – 0.8–3.0 0.3–2.7 0.5–2.1 1.1–3.1 – – – – – – – 0.9–3.3 1.2–3.3 2.5–8.8
0.017 0.016 0.038 0.773 0.047 0.024 0.108 0.007
0.444 0.003 0.000 0.001 0.010 0.001 0.010 0.002 0.144
Type of abuse and dependence symptoms were examined in models controlling for the number of symptoms other than the target symptom endorsed. This symptom was not examined as it was endorsed by almost all of those with alcohol dependence. Comorbid mental disorders were also examined in terms of whether their onset was before or after the onset of AUD symptoms (see Section 3.2.4). Significant compared to reference group.
Australia, 88.8% in the United Kingdom, and 69.8% in the United States and median treatment delays 14, 10 and 8 years, respectively. In all three countries rates of expected lifetime treatment are higher and median treatment delays longer than those presented in a comparable study in The Netherlands: 6.5%, 36.9%, 1 and 4 years, respectively (ten Have et al., 2013). Studies of the natural history and course of AUD suggest many people, particularly those at the lower end of the severity spectrum, will remit over lifetime without treatment (Cunningham and McCambridge, 2012). One could therefore argue that expected treatment rates presented here may be appropriate, especially given that in this study, as with previous studies, alcohol dependence is associated with shorter delay and higher likelihood of treatment contact than alcohol abuse without dependence. Similarly, the consistent finding that more recent cohorts are more likely to seek treatment may positively reflect increased availability of services or changing attitudes towards treatment (ten Have et al., 2013). However, median delays of 14 and 23 years among those
who eventually make treatment contact are long, and the implications of almost two decades of untreated AUD and related harms substantial, particularly in a country with some of the highest AUD prevalence rates in the world (Teesson et al., 2010). It is likely that factors contributing to these long delays are attitudinal, financial and structural (Keyes et al., 2010a; Mojtabai et al., 2011). However, given evidence for the effectiveness of brief interventions, particularly early in the course of AUD (Babor et al., 2007; Jonas et al., 2012) it is imperative to better understand and address these barriers to care for Australians with AUD. Within the context of these broad implications there are several specific findings that warrant discussion. Consistent with previous reports, earlier onset of AUD was associated with lowered treatment rates and longer delay (Bruffaerts et al., 2007; Hingson et al., 2006; ten Have et al., 2013; Wang et al., 2005). This may reflect ‘normalisation’ of risky alcohol use among young people (Davies et al., 2013), the need for parental involvement in treatment among adolescents (ten Have et al., 2013) or reluctance of clinicians to ask
120
C. Chapman et al. / Drug and Alcohol Dependence 147 (2015) 116–121
young people about alcohol use (Hingson et al., 2013). Given the substantial impact of alcohol related harms among young people (Keyes et al., 2012), and amid increasing evidence for the effectiveness of prevention in delaying onset of drinking and reducing these harms, the findings highlight the continued need to evaluate and fund prevention of AUD among adolescents and young adults (Teesson et al., 2012). Two novel findings from the study deserve particular attention. These are the findings related to symptom type and treatment delay, and comorbidity and treatment delay. Recurrent social or interpersonal problems (among those with abuse) and role disruption (among those with dependence) were both associated with shorter treatment delay and greater likelihood of treatment over lifetime. Previous studies on the treatment seeking for AUD have found social consequences of drinking to be strong predictors of perceived need for, and lifetime seeking of treatment, among both clinical and community samples (Cohen et al., 2007; Dawson et al., 2012; Edlund et al., 2009; Hedden and Gfroerer, 2011; Saunders et al., 2006). To the best of our knowledge, this is the first large community study to suggest that these same factors are associated with seeking treatment sooner after alcohol related problems begin. These consequences are often emphasised in the content of effective prevention (Champion et al., 2013) and screening and brief intervention programmes (Heather, 2012). This study lends further support to the importance of these factors as determinants of helpseeking and therefore as targets for education about alcohol use and its consequences. Whilst previous population studies have indicated that comorbidity is associated with higher rates of lifetime treatment seeking among those with mental and substance use disorders (Cohen et al., 2007; Wang et al., 2007a) the role of comorbidity as a predictor of treatment delay for AUD has not commonly been examined. In this study comorbid drug use disorder, with onset either before or after AUD symptoms was associated with shorter treatment delay among those with alcohol abuse. It is possible that social and/or legal problems associated with use of illicit substances increases the likelihood of more timely treatment for AUD (Grella et al., 2009) or that greater perceived need among those with comorbid drug use disorders results in shorter treatment delay (Hedden and Gfroerer, 2011). It is less clear why comorbid drug use disorder was not associated with treatment delay among those with alcohol dependence. It could be that among those at the more severe end of the AUD spectrum the impact of comorbid drug use disorders is small, particularly given models also controlled for the number and type of symptoms as well as comorbid mental disorders. Comorbid anxiety but not mood disorder was associated with shorter treatment delay among those with alcohol abuse and dependence. Mood and anxiety disorders are generally associated with higher lifetime treatment rates (Bunting et al., 2012; ten Have et al., 2013; Wang et al., 2007a) and greater perceived need for treatment (Edlund et al., 2009; Mojtabai et al., 2011) than AUD and the findings with respect to comorbid anxiety disorder may well reflect this. However, the fact that comorbid mood disorder was not associated with shorter treatment delay or higher probability of treatment is perhaps surprising. One possibility is that people with comorbid AUD and mood disorder are seeking treatment for their mood disorder as opposed to their AUD (Edlund et al., 2012). However, Olfson et al. (2012) using data from the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC) in the United States, conducted a similar study of treatment delay among people with major depression. Using comparable analyses they determined that comorbid AUD was not associated with shorter delay to first treatment. Ilgen et al. (2011) similarly reported another study from the NESARC examining the longitudinal predictors of treatment contact at Wave 2, 3 years after Wave 1, and determined that comorbid anxiety, but not mood disorder was
associated with subsequent treatment for AUD. Mood disorders and AUD commonly co-occur (Hasin et al., 2007), and this comorbidity is associated with significant impairment and poorer prognosis (Burns et al., 2005). Taken together, the findings of this and other studies clearly argue that those with comorbid AUD and mood disorder should be a priority for treatment earlier, rather than later in the course of illness. The findings of the study need to be considered within the context of some limitations. Firstly, the survey did not differentiate between treatment contact for alcohol and drug-related problems. Whilst the majority of the sample (76%) met criteria for an AUD but not another drug use disorder, we cannot determine whether the remainder of the sample was seeking treatment for alcohol or drug-related problems. Secondly, the survey was cross-sectional and therefore relied on retrospective recall of illness onset and treatment seeking. Despite the fact that the survey used similar methodology to that used in the World Mental Health Surveys to improve accuracy of recall (Kessler et al., 2007), the estimates are subject to biases including the general limitations of memory and the cognitive schemas people might have about the actual versus typical age of illness onset (Simon and VonKorff, 1995). The limited dating of events also makes it difficult to model sociodemographic and other factors as truly time-varying covariates. The relationship between specific symptom endorsement and treatment delay for AUD for example, would need to be modelled prospectively to truly capture the relationship between experience of symptoms and treatment seeking over time. Thirdly, increased mortality from years of chronic heavy drinking cannot be assessed in a crosssectional survey, so it is not possible to comment on treatment delays for those most severely affected by alcohol-related physical conditions. However, given the low rates of treatment seeking among the general population of people with AUD and the long treatment delays, prospective studies would need to follow people up for very long periods of time to capture their first treatment episode. Within this context, cross-sectional population studies such as this one make an important contribution. To the best of our knowledge, this is the first study to simultaneously examine the relationship between symptom onset, birth cohort, number and type of symptoms, comorbidity and treatment delay for AUD in a large representative population sample. While lifetime probability of treatment contact was found to be relatively high, median treatment delays were long. Even if some individuals recover without formal intervention, the implications of almost two decades of untreated AUD and related harms are likely to be substantial. The study highlights the need to better understand barriers to AUD treatment and points to the continued need to invest in prevention as well as screening and brief intervention early in the course of AUD. Specifically, those with earlier onset of symptoms and those with comorbid mood disorder should be a target for earlier treatment. Role of funding source The 2007 National Survey of Mental Health and Wellbeing (NSMHWB) was funded by the Australian Government and conducted by the Australian Bureau of Statistics. The Centre of Research Excellence in Mental Health and Substance Use is funded by the National Health and Medical Research Council. The National Drug and Alcohol Research Centre is supported by funding from the Australian Government under the Substance Misuse Prevention and Service Improvements Grants Fund. Contributors Dr Chapman conceptualised the research question in consultation with the other authors and drafted the initial manuscript.
C. Chapman et al. / Drug and Alcohol Dependence 147 (2015) 116–121
Dr Chapman and A/Prof Slade conducted the analysis. All authors took an active role in interpreting the results and revising the manuscript. All authors have seen and approved the final version of the manuscript. Conflict of interest No conflict declared. Acknowledgements The authors would like to thank the NSMHWB reference group for their input in the survey’s design. The authors would also like to thank all those who participated in the survey. References Alvanzo, A.A., Storr, C.L., Mojtabai, R., Green, K.M., Pacek, L.R., La Flair, L.N., Cullen, B.A., Crum, R.M., 2014. Gender and race/ethnicity differences for initiation of alcohol-related service use among persons with alcohol dependence. Drug Alcohol Depend. 140, 48–55. Babor, T.F., McRee, B.G., Kassebaum, P.A., Grimaldi, P.L., Ahmed, K., Bray, J., 2007. Screening, brief intervention, and referral to treatment (SBIRT): toward a public health approach to the management of substance abuse. Subst. Abuse 28, 7–30. Borges, G., Wang, P.S., Medina-Mora, M.E., Lara, C., Chiu, W.T., 2007. Delay of first treatment of mental and substance use disorders in Mexico. Am. J. Public Health 97, 1638–1643. Bruffaerts, R., Bonnewyn, A., Demyttenaere, K., 2007. Delays in seeking treatment for mental disorders in the Belgian general population. Soc. Psychiatry Psychiatr. Epidemiol. 42, 937–944. Bunting, B.P., Murphy, S.D., O’Neill, S.M., Ferry, F.R., 2012. Lifetime prevalence of mental health disorders and delay in treatment following initial onset: evidence from the Northern Ireland Study of Health and Stress. Psychol. Med. 42, 1727–1739. Burns, L., Teesson, M., O’Neill, K., 2005. The impact of comorbid anxiety and depression on alcohol treatment outcomes. Addiction 100, 787–796. Champion, K.E., Newton, N.C., Barrett, E.L., Teesson, M., 2013. A systematic review of school-based alcohol and other drug prevention programs facilitated by computers or the internet. Drug Alcohol Rev. 32, 115–123. Cohen, E., Feinn, R., Arias, A., Kranzler, H.R., 2007. Alcohol treatment utilization: findings from the National Epidemiologic Survey on Alcohol and Related Conditions. Drug Alcohol Depend. 86, 214–221. Cunningham, J.A., McCambridge, J., 2012. Is alcohol dependence best viewed as a chronic relapsing disorder? Addiction 107, 6–12. Davies, E.L., Martin, J., Foxcroft, D.R., 2013. Young people talking about alcohol: focus groups exploring constructs in the prototype willingness model. Drugs Educ. Prev. Policy 20, 269–277. Dawson, D.A., Goldstein, R.B., Grant, B.F., 2012. Factors associated with first utilization of different types of care for alcohol problems. J. Stud. Alcohol Drugs 73, 647–656. Edlund, M.J., Booth, B.M., Feldman, Z.L., 2009. Perceived need for treatment for alcohol use disorders: results from two national surveys. Psychiatr. Serv. 60, 1618–1628. Edlund, M.J., Booth, B.M., Han, X., 2012. Who seeks care where? Utilization of mental health and substance use disorder treatment in two national samples of individuals with alcohol use disorders. J. Stud. Alcohol Drugs 73, 635–646. Grambsch, P.M., Therneau, T.M., 1994. Proportional hazards tests and diagnostics based on weighted residuals. Biometrika 81, 515–526. Grella, C.E., Karno, M.P., Warda, U.S., Moore, A.A., Niv, N., 2009. Perceptions of need and help received for substance dependence in a national probability survey. Psychiatr. Serv. 60, 1068–1074. Hasin, D.S., Stinson, F.S., Ogburn, E., Grant, B.F., 2007. Prevalence, correlates, disability, and comorbidity of DSM-IV alcohol abuse and dependence in the United States: results from the National Epidemiologic Survey on Alcohol and Related Conditions. Arch. Gen. Psychiatry 64, 830–842. Heather, N., 2012. Can screening and brief intervention lead to population-level reductions in alcohol-related harm? Addict. Sci. Clin. Pract. 7, 15. Hedden, S.L., Gfroerer, J.C., 2011. Correlates of perceiving a need for treatment among adults with substance use disorder: results from a national survey. Addict. Behav. 36, 1213–1222. Hingson, R.W., Heeren, T., Winter, M.R., 2006. Age of alcohol-dependence onset: associations with severity of dependence and seeking treatment. Pediatrics 118, e755–e763. Hingson, R.W., Zha, W., Iannotti, R.J., Simons-Morton, B., 2013. Physician advice to adolescents about drinking and other health behaviors. Pediatrics 131, 249–257. Hosmer, D.W., Lemeshow, S., 1999. Applied Survival Analysis. John Wiley & Sons, Inc., New York. Ilgen, M.A., Price, A.M., Burnett-Zeigler, I., Perron, B., Islam, K., Bohnert, A.S., Zivin, K., 2011. Longitudinal predictors of addictions treatment utilization in treatmentnaive adults with alcohol use disorders. Drug Alcohol Depend. 113, 215–221. Jonas, D.E., Garbutt, J.C., Amick, H.R., Brown, J.M., Brownley, K.A., Council, C.L., Viera, A.J., Wilkins, T.M., Schwartz, C.J., Richmond, E.M., Yeatts, J., Evans, T.S., Wood, S.D., Harris, R.P., 2012. Behavioral counseling after screening for alcohol misuse
121
in primary care: a systematic review and meta-analysis for the U.S. Preventive Services Task Force. Ann. Intern. Med. 157, 645–654. Kerr, W.C., Greenfield, T.K., Ye, Y., Bond, J., Rehm, J., 2013. Are the 1976–1985 birth cohorts heavier drinkers? Age-period-cohort analyses of the national alcohol surveys 1979–2010. Addiction 108, 1038–1048. Kessler, R.C., Aguilar-Gaxiola, S., Berglund, P.A., Caraveo-Anduaga, J.J., DeWit, D.J., Greenfield, S.F., Kolody, B., Olfson, M., Vega, W.A., 2001. Patterns and predictors of treatment seeking after onset of a substance use disorder. Arch. Gen. Psychiatry 58, 1065–1071. Kessler, R.C., Amminger, G.P., Aguilar-Gaxiola, S., Alonso, J., Lee, S., Ustun, T.B., 2007. Age of onset of mental disorders: a review of recent literature. Curr. Opin. Psychiatry 20, 359–364. Kessler, R.C., Olfson, M., Berglund, P.A., 1998. Patterns and predictors of treatment contact after first onset of psychiatric disorders. Am. J. Psychiatry 155, 62–69. Kessler, R.C., Ustun, T.B., 2004. The World Mental Health (WMH) survey initiative version of the World Health Organization (WHO) Composite International Diagnostic Interview (CIDI). Int. J. Methods Psychiatr. Res. 13, 93–121. Kessler, R.C., Ustun, T.B., 2008. The WHO World Mental Health Surveys: Global Perspectives on the Epidemiology of Mental Disorders. Cambridge University Press, Cambridge. Keyes, K.M., Hatzenbuehler, M.L., McLaughlin, K.A., Link, B., Olfson, M., Grant, B.F., Hasin, D., 2010a. Stigma and treatment for alcohol disorders in the United States. Am. J. Epidemiol. 172, 1364–1372. Keyes, K.M., Martins, S.S., Blanco, C., Hasin, D.S., 2010b. Telescoping and gender differences in alcohol dependence: new evidence from two national surveys. Am. J. Psychiatry 167, 969–976. Keyes, K.M., Schulenberg, J.E., O’Malley, P.M., Johnston, L.D., Bachman, J.G., Li, G., Hasin, D., 2012. Birth cohort effects on adolescent alcohol use: the influence of social norms from 1976 to 2007. Arch. Gen. Psychiatry 69, 1304–1313. Lee, S., Fung, S.C., Tsang, A., Zhang, M.Y., Huang, Y.Q., He, Y.L., Liu, Z.R., Shen, Y.C., Kessler, R.C., 2007. Delay in initial treatment contact after first onset of mental disorders in metropolitan China. Acta Psychiatr. Scand. 116, 10–16. Mojtabai, R., Olfson, M., Sampson, N.A., Jin, R., Druss, B., Wang, P.S., Wells, K.B., Pincus, H.A., Kessler, R.C., 2011. Barriers to mental health treatment: results from the National Comorbidity Survey Replication. Psychol. Med. 41, 1751–1761. Naughton, F., Alexandrou, E., Dryden, S., Bath, J., Giles, M., 2013. Understanding treatment delay among problem drinkers: what inhibits and facilitates help-seeking? Drug Educ. Prev. Policy 20, 297–303. Olfson, M., Kessler, R.C., Berglund, P.A., Lin, E., 1998. Psychiatric disorder onset and first treatment contact in the United States and Ontario. Am. J. Psychiatry 155, 1415–1422. Olfson, M., Liu, S.M., Grant, B.F., Blanco, C., 2012. Influence of comorbid mental disorders on time to seeking treatment for major depressive disorder. Med Care 50, 227–232. Saunders, S.M., Zygowicz, K.M., D’Angelo, B.R., 2006. Person-related and treatmentrelated barriers to alcohol treatment. J. Subst. Abuse Treat. 30, 261–270. Simon, G.E., VonKorff, M., 1995. Recall of psychiatric history in cross-sectional surveys: implications for epidemiologic research. Epidemiol. Rev. 17, 221–227. Slade, T., Johnston, A., Oakley Browne, M.A., Andrews, G., Whiteford, H., 2009. 2007 National Survey of Mental Health and Wellbeing: methods and key findings. Aust. N. Z. J. Psychiatry 43, 594–605. StataCorp, 2012. Stata Statistical Software: Release 12. StataCorp LP, College Station, TX. Teesson, M., Hall, W., Slade, T., Mills, K., Grove, R., Mewton, L., Baillie, A., Haber, P., 2010. Prevalence and correlates of DSM-IV alcohol abuse and dependence in Australia: findings of the 2007 National Survey of Mental Health and Wellbeing. Addiction 105, 2085–2094. Teesson, M., Newton, N.C., Barrett, E.L., 2012. Australian school-based prevention programs for alcohol and other drugs: a systematic review. Drug Alcohol Rev. 31, 731–736. ten Have, M., de Graaf, R., van Dorsselaer, S., Beekman, A., 2013. Lifetime treatment contact and delay in treatment seeking after first onset of a mental disorder. Psychiatr. Serv. 64, 981–989. Wang, P.S., Aguilar-Gaxiola, S., Alonso, J., Angermeyer, M.C., Borges, G., Bromet, E.J., Bruffaerts, R., de Girolamo, G., de Graaf, R., Gureje, O., Haro, J.M., Karam, E.G., Kessler, R.C., Kovess, V., Lane, M.C., Lee, S., Levinson, D., Ono, Y., Petukhova, M., Posada-Villa, J., Seedat, S., Wells, J.E., 2007a. Use of mental health services for anxiety, mood, and substance disorders in 17 countries in the WHO world mental health surveys. Lancet 370, 841–850. Wang, P.S., Angermeyer, M., Borges, G., Bruffaerts, R., Tat Chiu, W., Girolamo, D.E.G., Fayyad, O., Gureje, J., Haro, J.M., Huang, Y., Kessler, R.C., Kovess, V., Levinson, D., Nakane, Y., Oakley Brown, M.A., Ormel, J.H., Posada-Villa, J., Aguilar-Gaxiola, S., Alonso, J., Lee, S., Heeringa, S., Pennell, B.E., Chatterji, S., Ustun, T.B., 2007b. Delay and failure in treatment seeking after first onset of mental disorders in the World Health Organization’s World Mental Health Survey Initiative. World Psychiatry 6, 177–185. Wang, P.S., Berglund, P., Olfson, M., Pincus, H.A., Wells, K.B., Kessler, R.C., 2005. Failure and delay in initial treatment contact after first onset of mental disorders in the National Comorbidity Survey Replication. Arch. Gen. Psychiatry 62, 603–613. Whiteford, H.A., Degenhardt, L., Rehm, J., Baxter, A.J., Ferrari, A.J., Erskine, H.E., Charlson, F.J., Norman, R.E., Flaxman, A.D., Johns, N., Burstein, R., Murray, C.J., Vos, T., 2013. Global burden of disease attributable to mental and substance use disorders: findings from the Global Burden of Disease Study 2010. Lancet 382, 1575–1586. Witkiewitz, K., Dearing, R.L., Maisto, S.A., 2014. Alcohol use trajectories among nontreatment-seeking heavy drinkers. J. Stud. Alcohol Drugs 75, 415–422.