RESEARCH REPORT doi:10.1111/j.1360-0443.2010.03194.x Probability and predictors of remission from life-time nicotine, alcohol, cannabis or cocaine dependence: results from the National Epidemiologic Survey on Alcohol and Related Conditionsadd_3194 657..669 Catalina Lopez-Quintero1, Deborah S. Hasin1,2, José Pérez de los Cobos3, Abigail Pines, Shuai Wang1, Bridget F. Grant4 & Carlos Blanco1 New York State Psychiatric Institute, Department of Psychiatry, College of Physicians and Surgeons, Columbia University, New York, NY, USA,1 Department of Epidemiology, Mailman School of Public Health, Columbia University, NewYork, NY,USA,2 Addictive Behaviors Unit of Psychiatry Department, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain3 and Laboratory of Epidemiology and Biometry, Division of Intramural Clinical and Biological Research, National Institute on Alcohol Abuse and Alcoholism, National Institutes of Health, Bethesda, MD, USA4 ABSTRACT Aim To estimate the general and racial/ethnic specific cumulative probability of remission from nicotine alcohol cannabis or cocaine dependence, and to identify predictors of remission across substances. Design Data were col- lected from structured diagnostic interviews using the Alcohol Use Disorder and Associated Disabilities Interview Schedule—DSM-IV version. Setting The 2001–2002 National Epidemiological Survey of Alcohol and Related Con- ditions (NESARC) surveyed a nationally representative sample from US adults (n = 43 093) selected in a three-stage sampling design. Participants The subsamples of individuals with life-time DSM-IV diagnosis of dependence on nicotine (n = 6937), alcohol (n = 4781), cannabis (n = 530) and cocaine (n = 408). Measurements Cumulative prob- ability estimates of dependence remission for the general population and across racial/ethnic groups. Hazard ratios for remission from dependence. Findings Life-time cumulative probability estimates of dependence remission were 83.7% for nicotine, 90.6% for alcohol, 97.2% for cannabis and 99.2% for cocaine. Half of the cases of nicotine, alcohol, cannabis and cocaine dependence remitted approximately 26, 14, 6 and 5 years after dependence onset, respectively. Males, Blacks and individuals with diagnosis of personality disorders and history of substance use comor- bidity exhibited lower hazards of remission for at least two substances. Conclusions A significant proportion of individuals with dependence on nicotine, alcohol, cannabis or cocaine achieve remission at some point in their life-time, although the probability and time to remission varies by substance and racial/ethnic group. Several predictors of remission are shared by at least two substances, suggesting that the processes of remission overlap. The lower rates of remission of individuals with comorbid personality or substance use disorders highlight the need for providing coordinated psychiatric and substance abuse interventions. Keywords Alcohol, cannabis, cocaine, dependence, nicotine, remission. Correspondence to: Carlos Blanco, New York State Psychiatric Institute, Department of Psychiatry, College of Physicians and Surgeons of Columbia University, 1051 Riverside Drive, unit 69, New York, NY 10032, USA. E-mail: [email protected] Submitted 29 March 2010; initial review completed 21 May 2010; final version accepted 31 August 2010 INTRODUCTION wide (approximately 91 million deaths) were attributed to substance use disorders (SUDs) [1]. However, a signifi- Substance dependence is associated with increased risk of cant proportion of individuals with these disorders physical and mental illness, disability, lost work produc- achieve remission at some point in their lives [2–8]. tivity, financial problems, perpetrating and suffering vio- Therefore, recognizing the patterns and predictors of lence, accidents and death [1]. In 2004, 0.5% of disability remission from substance dependence is essential for adjusted life years (DALY) and 0.2% of all deaths world- increasing our understanding of its natural history © 2010 The Authors, Addiction © 2010 Society for the Study of Addiction Addiction, 106, 657–669 658 Catalina Lopez-Quintero et al. and developing timely prevention and treatment dardized under the direction of the National Institute on strategies. Alcohol Abuse and Alcoholism Alcohol (NIAAA). All Results from clinical and population-based studies procedures, including informed consent, received full have shown higher remission rates for dependence on ethical review and approval from the US Census Bureau cannabis [7–9] and alcohol [4,5,10], followed by cocaine and the US Office of Management and Budget. [2,6,11] and nicotine [3,12]. These studies have also identified several factors associated with remission, MEASURES including being female, older, White, married, with higher educational attainment, and a later onset of sub- Socio-demographics stance use [5,7,10,13–15]. Despite this extensive body of Self-reported race/ethnicity was recoded into five groups: research, important questions remain. To date, compari- Whites, Blacks, Hispanics, Native Hawaiians or other sons of remission rates across all racial/ethnic groups in Pacific Islanders (NH/PI) and American Indians or the United States are not available, due to limited sample Alaskan Natives (AI/AN). Other socio-demographic vari- size for minority groups in most data sets [3,5]. Scarce ables included gender, age, nativity, level of education, information exists regarding the commonality of remis- individual income and marital status. sion predictors across substances. Few studies have examined the role of psychiatric comorbidity on the prob- Substance dependence and remission ability of remission, and even fewer have assessed time- All diagnoses were made according to the DSM-IV criteria varying predictors of remission [5,14,16,17]. using the NIAAA Alcohol Use Disorder and Associated The goal of this study was to advance knowledge in Disabilities Interview Schedule—DSM-IV version these areas. Specifically, we sought to: (i) estimate the (AUDADIS) [19], which allows measurement of sub- time from onset of nicotine, alcohol, cannabis or cocaine stance use and mental disorders in large-scale surveys. dependence until remission of all dependence criteria Detailed information on survey methods is described else- (herein remission) for each substance; (ii) examine the where [18]. Dependence diagnoses require the existence cumulative probability of remission from dependence of three or more of the seven DSM-IV diagnostic criteria across racial/ethnic groups; and, (iii) identify predictors within a 12-month period [20]. Age of dependence onset of remission across substances. for each substance was defined as the age at which the respondent first met dependence diagnostic criteria for that particular substance. METHODS The outcome variables remission and age at remission Sample were determined by asking individuals with a life-time diagnosis of dependence: ‘about how old were you when The 2001–02 National Epidemiological Survey of you finally stopped having any of these experiences Alcohol and Related Conditions (NESARC) surveyed a (dependence criteria) with (name of drug)? By finally representative sample of the US population [18]. The stopped, I mean they never started happening again.’ target population was the civilian non-institutionalized Time to remission for each specific substance was defined population 18 years and older residing in households and as the time interval between age of dependence onset and group quarters (e.g. college quarters, group homes, age of remission. boarding houses and non-transient hotels). The survey Early substance use onset (defined as before age 14) included residents of the continental United States, Dis- and family history of SUDs (any alcohol or drug use dis- trict of Columbia, Alaska and Hawaii. Face-to-face order among first-degree relatives) were also included as computer-assisted interviews were conducted among a substance use-related covariates. The good-to-excellent multi-stage cluster sample of 43 093 respondents. The ( 0.54–0.91) test–retest reliability and validity of overall survey response rate was 81%. Blacks, Hispanics k= AUDADIS-IV SUD diagnoses is well documented in clini- and adults aged 18–24 years were oversampled, with cal and general population samples [21–24]. data adjusted for oversampling, household- and person- level non-response [18]. This report examined data of the Psychiatric disorders subsample of respondents with life-time history of dependence on nicotine (n = 6937), alcohol (n = 4781), Mood disorders included DSM-IV primary major depres- cannabis (n = 530) or cocaine (n = 408). sive disorder (MDD), dysthymia and bipolar disorders. Interviews were conducted by professional interview- Anxiety disorders included DSM-IV primary panic disor- ers from the US Census Bureau. On average, the inter- der (with and without agoraphobia), social anxiety disor- viewers had 5 years of experience working on census and der, specific phobias and generalized anxiety disorder. other health-related national surveys. Training was stan- ConsistentwithDSM-IV,‘primary’AUDADIS-IVdiagnoses © 2010 The Authors, Addiction © 2010 Society for the Study of Addiction Addiction, 106, 657–669 Predictors of life-time dependence remission 659 excluded disorders that are substance-induced or due to RESULTS general medical conditions. Diagnoses of MDD ruled out Characteristics of the study population bereavement [20]. AUDADIS-IV diagnostic methods and advantages over other survey interviews are described