摘要：Objectives. We examined the relationships between substance abuse treatment, abstinence, and mortality in a sample of individuals entering treatment. We also estimated overall mortality rates and the extent to which they varied according to demographic, clinical severity, and treatment variables. Methods. We used data from a 9-year longitudinal study of 1326 adults entering substance abuse treatment on the west side of Chicago, of whom 131 died (11.0 per 1000 person-years). Baseline predictors, initial and long-term treatment response, and substance use patterns were used to predict mortality rates and time to mortality. Results. Older age, health problems, and substance use were associated with an increased risk of mortality, and higher percentages of time abstinent and longer durations of continuous abstinence were associated with a reduced risk of mortality. Treatment readmission in the first 6 months after baseline was related to an increased likelihood of abstinence, whereas readmission after 6 months was related to a decreased likelihood of abstinence, suggesting that treatment timing is significant. Conclusions. Our findings suggest the need to shift the addiction treatment field from an acute care model to a chronic disease management paradigm and the need for more aggressive screening, intervention, and addiction management over time. The strong association of illicit drug use with other prominent causes of mortality (e.g., tobacco and alcohol use, accidents, and high-risk sexual behaviors), as well as its contribution to significant chronic health problems (e.g., cancer, heart disease, liver disease, and respiratory illness), renders it one of the most fundamental mortality risks. 1 , 2 The statistics are staggeringly clear: individuals with a drug dependence diagnosis die on average 22.5 years earlier than those without such a diagnosis. 3 Furthermore, mortality in the subgroup of individuals with substance use disorders who enter substance abuse treatment is 3 or more times higher than that in the US population as a whole after adjustment for age. 4 – 15 How should US public health officials address the insidious mortality risk associated with illicit drug use, which currently threatens the approximately 6.9 million people (2.8% of the US population) who meet the diagnostic criteria for substance use disorders? 2 , 16 Before a coherent and effective public health strategy to address the mortality risk associated with drug use can be formulated, the research community must achieve a greater understanding of the specific mechanisms that drive the association between addiction treatment and reduced mortality. Although the relationship between addiction treatment, abstinence, and mortality is complex, we do know that the variables that may affect this relationship include the point in their addiction career at which people enter treatment and the amount of treatment they receive. A large number of studies as well as expert reviews of the literature in the past 2 decades have consistently concluded that participation in substance abuse treatment increases the likelihood of short-term abstinence. 17 – 25 Particularly, more intense initial treatment, higher cumulative treatment dosage, and early reintervention have been associated with sustained abstinence over multiple years. 26 – 29 A smaller body of evidence also indicates that those who enter treatment sooner and stay in longer are less at risk for mortality. 26 , 30 In addition, the US Preventive Services Task Force, which is charged with evaluating the benefits of strategies that address the leading causes of mortality, found “good evidence” that drug use is related to mortality and that various treatments designed to reduce illicit drug use in the short term are effective. 31 Nevertheless, it concluded that insufficient evidence exists to link treatment to longer term improvements in morbidity (prevalence and duration of abstinence) and, consequently, reduced mortality. A rigorous examination of the relationships between illicit drug use, treatment, and long-term outcomes, including mortality, requires consideration of the variables that affect these relationships not only in the short term but also in the long term. One such group of variables consists of an individual's characteristics upon entering treatment: age, gender, living arrangements, employment, criminal justice history, substance use history, and preexisting health conditions. 6 , 7 , 32 – 34 In addition to this first set of variables, results from a range of studies indicate the need to consider various aspects of treatment participation. Because substance use disorders are best conceptualized as chronic conditions, a number of addiction scientists have argued that research in this area adopt a life course, developmental perspective to assess the role of treatment in substance use and related outcomes. 35 – 39 From this perspective, mortality risk among substance users may be related to their initial response to treatment and their cumulative duration of abstinence. Duration of sustained abstinence may also play a role in mortality risk. Finally, previous studies have shown that longer periods of abstinence (% of days whether consecutive or not) and longer durations of sustained abstinence (consecutive days only) yield benefits in a wide array of physical, psychological, and social functioning domains, including social network improvements, increased vocational involvement, and better mental health. 27 , 28 , 40 – 43 Despite the knowledge that such changes represent an increase in recovery capital that may be related to a wide range of health outcomes with an impact on mortality risk, it remains unclear whether characteristics of an individual's abstinence (e.g., timing or duration), the proximal goal of treatment, mediate that association. Capitalizing on a 9-year longitudinal investigation of individuals presenting to community-based substance abuse treatment facilities, we addressed 2 research questions in this study: (1) What are the overall mortality rates in this sample, and how do they vary according to demographic, clinical severity, and treatment variables? and (2) How and to what extent do treatment and abstinence (i.e., the proximal outcome of treatment) mediate these relationships?
