The Role of Stressful Events in Drug Use Disorder Relapse

There is strong evidence of a link between stressful life events, and substance use disorder relapse. What is less clear is the directionality of this relationship; i.e., whether stressful life events drive substance use disorder relapse, or substance use disorder relapse generates stressful life events. This study sought to better understand whether stressful life events influence someone’s ability to initiate and sustain recovery.


Researchers are constantly seeking to understand factors that may influence relapse or an individuals’ likelihood of successfully recovering from substance use disorder.

One problem with much existing research in this area is that it is cross-sectional in design, meaning that data are collected at a single point in time. While cross-sectional studies can suggest a relationship between individual factors such as an individual’s stress levels, and substance use and substance use related problems, they cannot establish causality, in part, because it’s hard to know which precedes which. In other words, it cannot be determined whether the individual factor, such as stress, is driving the substance use and substance use problems, or vice-versa, or perhaps both are “causal”, having reciprocal effects.

Establishing causality provides valuable information that can inform numerous aspects of care and support for individuals with substance use disorder, from the clinic to the community.


This study by McCabe and colleagues analyzed survey data collected from waves 1 and 2 of the 2001–2002 and 2004–2005 National Epidemiologic Survey on Alcohol and Related Conditions (NESARC-I and NESARC-II), which surveyed the United States, non-institutionalized adult population, aged 18 years or older.


At wave 1 (baseline for the study reviewed here), the NESARC achieved an overall response rate of 81%, with 43,093 respondents completing face-to-face personal interviews. Three years after wave 1 was completed, an attempt was made to re-interview all 43,093 wave 1 respondents via face-to-face personal interviews. A response rate of 87% was achieved at wave 2 for a total sample of 34,653, resulting in an overall response rate of 70%.

Using DSM-IV criteria, each participant was assessed for drug use disorder in the past year at each assessment time-point (i.e., NESARC wave 1, and wave 2). It should be noted that DSM-IV drug dependence would be consistent with moderate or severe drug use disorder in DSM-5, the latest version of the DSM.

The study focused on the sub-sample of 921 persons with drug dependence at wave 1, and did not include individuals dependent solely on alcohol. Past-year remission was based on DSM-IV definitions and broken into the following four sub-categories:


  1. past-year abstinence: no drug use in the past 12 months
  2. past-year asymptomatic drug use: used at least one drug at least once but did not experience any DSM-IV drug dependence criteria in the past 12 months
  3. past-year partial remission: past-year recurrence of DSM-IV drug dependence criteria but did not meet full criteria for DSM-IV drug dependence in the past 12 months
  4. past-year still drug dependent: continued to meet criteria for DSM-IV drug dependence for at least one drug class in the past 12 months.


Stressful life events were assessed by asking respondents if they had experienced each of the following stressful life events:

“In the last 12 months…


  1. Did any of your family members or close friends die?
  2. Did any of your family members or close friends have a serious illness or injury?
  3. Did you move or has anyone new come to live with you?
  4. Were you fired or laid off from a job?
  5. Were you unemployed and looking for a job for more than a month?
  6. Have you had trouble with your boss or a coworker?
  7. Did you change jobs, job responsibilities or work hours?
  8. Did you get separated or divorced or break off a steady relationship?
  9. Have you had serious problems with a neighbor, friend or relative?
  10. Have you experienced a major financial crisis, declared bankruptcy or more than once been unable to pay your bills on time?
  11. Did you or a family member have trouble with the police, get arrested or get sent to jail?
  12. Were you or a family member the victim of any type of crime?


The 12 items were summed to create an index of the number of stressful life events (range 0–12) experienced during the last 12 months (see Dawson, Grant, Stinson, et al., 2005).

In their statistical analysis, study authors assessed whether sex, race, age, education level, income, marital status, stressful life events, having prior-to-past-year alcohol use disorder, drug treatment utilization, and having a prior-to-past-year drug use disorder influenced remission status within each assessment time-point (i.e., within NESARC wave 1, and within wave 2) over and above these individual characteristics. In addition, the authors tested whether stressful life events predict remission status approximately 3 years later on, after statistically adjusting for these same individual factors.


Overall, higher levels of stressful life events at study baseline were associated with a lower likelihood of participants being in partial or full remission from drug use disorder at follow-up assessment three years later. The findings of this study offer evidence that more stressful life events are barriers to remission from drug use disorder both over the short term (i.e., during the same time period as the stressful events) and the longer term (i.e., years after experiencing the stressful events).

Specifically, compared to someone with zero stressful life events, someone with one stressful life event is 20% more likely to report problematic drug use (i.e., one or more drug use disorder symptoms) than abstinence 3 years later, someone with two stressful life events is 44% more likely to report problematic drug use, and someone with three stressful life events is 72% more likely to report problematic drug use. This unique role of stressful life events on chances of remission exists over and above individual demographic characteristics, history of alcohol use disorder, and whether or not the individual sought substance use disorder treatment.

In addition to this finding, having a high school education or less, relative to having more than a high school education was associated with greater likelihood of meeting criteria for drug dependence at 3-year follow-up assessment. Also, being a young adult at baseline made it less likely that such participants were in full DSM-IV drug dependence remission at follow-up, and more likely that they were in partial remission, meaning they were still experiencing drug use disorder related problems, but not with the same degree of severity. Other individual factors such as sex, race, and marital status at baseline were not related to whether participants met criteria for drug dependence versus being abstinent at 3-year follow-up.

Notably, abstinence at 3-year follow-up was more prevalent among participants formerly meeting criteria for drug dependence who were abstaining from drugs at the time of baseline assessment, relative to those who reported drug use but no drug use disorder symptoms, partial remission, or met criteria for drug use disorder at baseline. Also of note, abstinence was the most stable drug use status over time, with 85.3% of participants in this category at study baseline still reporting abstinence at follow-up.


Though to many it may seem obvious that stressful life events increase individuals’ risk for a return to problematic substance use (because it is widely understood that many people with substance use disorder use alcohol and other drugs to cope), previous research had not established a clear causal relationship. It was equally likely that substance use disorder relapse was causing stressful life events to arise. After all, relationship and work/school related problems are diagnostic criteria for substance use disorder.

In keeping with stress and coping theories of substance use, this study supports the notion that stressful life events may, in fact, lead to greater likelihood of relapse, and makes the case for emphasizing effective stress management/coping skills training in treatments for substance use disorder, as well as finding ways to reduce stress among individuals seeking substance use disorder recovery, including providing training, education, and employment opportunities, particularly to those with criminal records. It is important to note that findings from this study do not mean that substance use disorder relapse cannot lead to stressful life events. Both are likely true.

Notably, this study also found that the 23.4% of individuals with baseline drug dependence continued to meet criteria for drug dependence 3 years later, which is lower than a prior study that found the 3-year persistence rate of drug dependence was 30.9% (Fenton et al., 2012). Also of note, McCabe and colleagues’ observed that 85.3% of participants who were abstinent from drugs at baseline (wave 1) were still abstinent at 3-year follow-up. This is consistent with a similar study that examined alcohol dependence recovery using waves 1 and 2 of the NESARC data.

  1. While it was shown that greater levels of stressful life events at the beginning of the study predicted lower likelihood of drug dependence remission at 3-year follow-up, the authors did not mention whether they assessed/controlled for the effects of stressful life events at 3-year follow-up in their statistical models that measured the longer-term influence of stressful life events from 3 years prior. If they did not, it means we can be less certain that greater stress is driving higher future likelihood of drug use disorder relapse, because stress at 3-year follow-up could be influencing these findings.
  2. The NESARC did not include institutionalized adults who typically have higher rates of substance use disorder. This study may have therefore underestimated substance use disorder rates. Thus, the study’s findings may not be applicable to those institutionalized.
  3. This study considered remission from several drug classes combined; stress could play a different role in remission from different drugs. For example, stress may affect individuals with opiate use disorder in different ways to those with cocaine use disorder because of different ways these drugs affect stress systems in the body such as the hypothalamic-pituitary-adrenal-axis.
  4. Despite a large national sample, the sizes of some subgroups in this study’s analyses were relatively small, thus reducing confidence that these results generalize to the U.S. population.
  5. Findings in this study are based on DSM-IV criteria for drug dependence; were the new DSM-5 criteria for substance use disorder to be employed, the findings may differ. For a summary of DSM-5 drug use disorder prevalence in wave 3 of the NESARC see the following review of drug use disorder prevalence.


Future studies will ideally test for factors that may be moderating or influencing the relationship between stressful life events and substance use disorder relapse. For instance, it is possible individuals’ coping skills, degree of general or recovery-specific social support, or socioeconomic status, influence their vulnerability to substance use disorder relapse following stressful life events.


  • For individuals & families seeking recovery: This study suggests you would do well to minimize exposure to life stressors (in as much as that is possible), and learn effective coping skills for managing stress. The results also suggest individuals in recovery from drug use disorder would benefit from accessing resources in the community that may help reduce stress, such as peer-support.
  • For scientists: This study suggests stressful life events are a significant predictor of drug use disorder remission status. Though as much has been assumed based on previous cross-sectional research, this study is important as it adds support to stress and coping theories of substance relapse.
  • For policy makers: This study suggests stressful life events are associated with subsequent drug use disorder remission status. Legal, education, and employment policies that increase stress among individuals with histories of drug use disorder should be reconsidered. This study also highlights the importance of providing individuals with accessible recovery support services that can address this risk factor.
  • For treatment professionals and treatment systems: This study suggests stressful life events are a significant future predictor of subsequent drug use disorder remission status. The results suggest arming patients with coping skills to more effectively manage stressors may reduce substance use and substance use disorder relapse risk.


McCabe, S. E., Cranford, J. A., & Boyd, C. J. (2016). Stressful events and other predictors of remission from drug dependence in the United States: Longitudinal results from a national survey. Journal of Substance Abuse Treatment, 71, 41-47.