Course of Drug Use Disorders
For adults in general community settings, only sparse data exist regarding the long-term course and outcome of illicit drug use. In addition, definitions of remission in these studies are often imprecise and inconsistent. Nevertheless, results suggest that remission from drug dependence may occur somewhat frequently but that rates vary across different drugs.23
In community samples, a study of cannabis users (34.2%) in a sample of German 14- to 24-year-olds found that 8.1% met diagnostic criteria for cannabis abuse, and 4.3% met criteria for dependence when first assessed.24 At a 4-year follow-up, only 22.8% and 25.2% of those initially diagnosed with abuse and dependence, respectively, had been abstinent from cannabis in the past year. At a later 10-year follow-up, the percentage abstinent from cannabis was lower for those initially diagnosed with abuse (29.6%) than for those initially diagnosed with dependence (56.4%), similar to Vaillant’s findings regarding AUDs.1125 Not surprisingly, some individuals not meeting abuse or dependence criteria at baseline did so at later points, thus highlighting that studies that begin with and follow only an initial sample of individuals with SUD diagnoses can only provide an incomplete picture of the onset and course of SUDs.
More frequent remissions (65%) were found by Gilder et al.26 among 159 Southwest California Indian participants, aged 18 to 70 years, who had a lifetime diagnosis of cannabis dependence. Remission was defined liberally as not meeting dependence criteria for 6 months or more. Remission was more likely among those who had more cannabis-induced symptoms and a shorter period of cannabis dependence.
To describe trajectories of using cocaine, opioids, and amphetamines among adults in community settings, a cohort of 4,301 men and women was followed for 18 years, from young adulthood to middle age.27 Four trajectories were identified: nonusers (86%), early occasional users (8%), persistent occasional users (4%), and early frequent/later occasional users (3%), the latter of whom had a higher all-cause mortality rate. Of adult users, more than one-third continued using illicit drugs into middle age.
In a review of prospective studies (a mix of community and treated samples) with follow-ups ranging from 3 to 33 years, remission was more likely among individuals using amphetamines (45%) than those using opioids (22%) or cocaine (14%). According to Calabria and colleagues,23 about 25% of persons dependent on amphetamines, 20% of those dependent on cocaine, 15% of those dependent on heroin, and 10% of those dependent on cannabis remitted from active drug dependence in a given year.
In a representative sample of individuals in the United States with a lifetime diagnosis of alcohol, cannabis, or cocaine dependence, lifetime cumulative probability estimates of remission were 91% for alcohol, 97% for cannabis, and 99% for cocaine dependence.28 Thus, the vast majority of individuals with lifetime diagnoses of dependence on these substances remitted at some point in their lives, although men were less likely than women to remit from dependence on all the substances, and dependence on one substance tended to decrease the probability of remission from dependence on another substance. One-half of the individuals with alcohol, cannabis, and cocaine dependence remitted approximately 14, 6, and 5 years, respectively, after onset of dependence. That is, remission from cannabis or cocaine dependence occurred faster than remission from alcohol dependence. Indeed, more than two-thirds of remissions from cannabis and cocaine dependence occurred within the first decade after onset of dependence, whereas only one-fifth of remissions from alcohol dependence occurred within that period. The differences in rates of remission across substances may be explained partly by the variation in the speed with which physical, psychological, and socially adverse consequences manifest after the onset of dependence; the legal status of alcohol; and the greater social acceptability of cannabis use relative to cocaine use.28
With respect to long-term outcome, Finney et al.6 presented data from long-term follow-ups (at least 8 years) in 16 studies of patients treated for drug use disorders, including 11 studies of treatment for opioid dependence, three for various illicit drugs, and two for cocaine dependence. The percentage of followed patients in remission ranged from 30% to 100% across the 15 studies for which remission rates could be calculated; annualized remission varied from 2.0%29 to 7.4% in a study by Byrne.30 However, in the Byrne study, the rate may have been inflated by classifying patients on methadone as “in remission.” The 4% annualized remission rate for these studies was slightly lower than the 4.9% rate noted earlier for persons treated for AUDs.
An additional analysis of data from male heroin users29 followed over a 33-year period after treatment identified three groups with distinctive profiles: (1) stable high-level users (59%) who maintained consistent high levels of heroin use since initiation, (2) decelerated users (32%) who started at a high level but decreased use over time, and (3) early quitters (9%) who quickly dropped to no use within 10 years of initial use.31 Early quitters had initiated heroin use at older ages and had the lowest mortality rate. Among those alive at the 33-year follow-up, stably high-level users had the lowest rate of employment and highest rates of legal and mental health problems.
Another study focused on 581 abstainers (85%) and 103 (15%) non-problem users at 1 year following drug and alcohol treatment, who then were studied 5, 7, 9, and 11 years post-baseline.32 In all, 48% of those initially remitted at 1 year exhibited stable remission over the subsequent 10 years. However, initial non-problem users had twice the odds of relapsing between years 2 and 11 relative to initial abstainers, thus suggesting that non-problem use is a less optimal initial outcome of SUD treatment than is abstinence.