Several findings can be observed across the literature at this point, including some surprises. Even for clinicians who do not specialize in PTSD/SUD treatment, it is worth understanding the current state of the field. As it is said, every clinician has PTSD/SUD patients in their practice, whether they know it or not.

For details on the findings, see Najavits and Hien (2013), which is a comprehensive literature review on all outcome studies for PTSD/SUD. That review was written for clinicians as well as researchers and provides extensive descriptions of each study’s methodology and results. Recent studies that emerged after that review are cited in Najavits (2013b); Hien et al. in press. Other reviews are available but are not comprehensive (e.g., Torchalla et al. 2012). Also more research is needed, given the methodology limitations of some studies (Najavits 2013b).

PTSD/SUD studies consistently show positive outcomes. In the 38 outcome studies conducted thus far, the pattern of results has consistently been positive. Improvements were found in PTSD, SUD, and other domains such as self-compassion, cognitions, coping skills, psychopathology, and functioning. Treatment satisfaction was strong in studies that addressed it. Early concerns that addressing PTSD in the context of SUD would worsen patients’ state have not been borne out. But it is important to remember that all studies used new models specifically designed for PTSD/SUD or made major changes to classic PTSD therapies to make them tolerable and feasible for SUD samples.

All studies using a PTSD exposure (past-focused) approach combined it with a SUD coping skills (present-focused) approach—but none outperformed models that were present-focused alone. A major current discussion in the field is the relative merit of present- versus past-focused approaches to PTSD treatment. Broadly speaking, models that focus on exposure-based or other emotionally intense exploration of trauma memories are termed here past-focused. In contrast, present-focused models focus on coping skills and psychoeducation but do not explore trauma memories in detail (Najavits 2013a). Note that trauma-focused is often used to refer to exposure-based models. However, all present-focused PTSD models directly focus on trauma. The difference is how they approach it. Exposure-based models focus primarily on the past by exploring the intense trauma narrative and memories. Present-focused PTSD models explicitly omit detailed exploration of the past and instead offer psychoeducation and coping skills to help patients work on PTSD in the present (e.g., learn to identify and manage trauma symptoms; improve functioning; increase safety in their current actions, thinking, and behavior; and promote overall stabilization). Moreover, the term non-trauma-focused treatment for present-focused PTSD models is problematic; it is comparable to referring to women as “non-men” or children as “non-adults.” Thus the terms present- and past-focused are used here.

The majority of PTSD/SUD studies thus far use present- rather than past-focused approaches. This is convergent with the widely endorsed stage-based approach to PTSD treatment in which present-focused stabilization occurs before moving into past-focused exposure (Cloitre et al. 2011; Herman 1992). This framework also helps explain why most of the PTSD literature has excluded SUD patients.

In recent years, there has been the healthy development of trying to evaluate whether past-focused approaches may be safely used with PTSD/SUD populations. Importantly, every study using a past-focused PTSD approach combined it with a present-focused SUD model. Concurrent Treatment of PTSD and Substance Use Disorders Using Prolonged Exposure (COPE; Mills et al. 2012) combines PTSD exposure therapy (Foa and Rothbaum 1998) with two CBT SUD models (Baker et al. 2003; Carroll 1998). The PE study (Foa et al. 2013) combined PE (Foa et al. 2007) with a motivational interviewing SUD model (Volpicelli et al. 2001). The Integrated Treatment Study (Sannibale et al. 2013) combined PTSD therapies (exposure and PTSD cognitive restructuring) with SUD treatment manuals from Project MATCH (Kadden et al. 1995) and Project COMBINE (Miller 2004). A study by van Dam et al. (2013) combined Structured Writing Therapy for PTSD (SWT; van Emmerik et al. 2008) with SUD group CBT (Emmelkamp and Vedel 2006). Creating Change uses a gentle approach to explore the past in relation to both PTSD and SUD, including preparation for the work, readiness evaluation, strong safety monitoring, and theme-based session topics (Najavits 2013a). In sum, no investigator has used any past-focused PTSD approach as-is with a SUD population.

Moreover, it is notable that all studies that included a past-focused component were delivered in individual modality rather than group and were almost always restricted to less complex samples than present-focused studies, in keeping with the PTSD-alone literature. “Less complex” means that patients were typically excluded if they had drug use disorders (rather than alcohol only), current domestic violence, homelessness, suicidality, violence, cognitive impairment, serious mental illness, and/or criminal justice involvement. In contrast, present-focused models were primarily group modality and accepted a much broader range of patients (Najavits and Hien 2013). (See below for more on this point.)

Many people believe that past-focused models are more powerful than present-focused models, perhaps because they are experienced as more emotionally intense. Yet all four RCTs that included past-focused PTSD treatment found null results (no difference) on either PTSD or SUD at the end of treatment compared to a control conditions that was present-focused only (Mills et al. 2012; Foa et al. 2013; Sannibale et al. 2013; van Dam et al. 2013) See Najavits 2013b for a summary. End of treatment is emphasized as that is the most rigorous time point for evaluating the impact of a model relative to a control. Both past- and present-focused models worked, but past-focused was not superior to present-focused, even on PTSD where it would be expected to if the “emotional intensity” hypothesis held. One explanation for the null results is that combining past-focused methods with present-focused diluted the past-focused work (Foa et al. 2013). Another explanation is that past-focused models may be too intense for patients who are struggling with SUD, which is consistent with the dropout problem Hoge et al. (2014), in various past-focused studies (e.g., (Foa et al. 2013; Mills et al. 2012; Brady et al. 2001)). See also the recent meta-analysis by Gerger et al. (2013), which found that the PTSD treatment models they reviewed, which were predominantly past-focused, worked best with simpler rather than more complex patients, when compared to nonspecific therapies such as supportive therapy and relaxation training. A study sample was identified as complex if 80 % met at least two of four clinical criteria: (a) duration of symptoms lasting more than 6 months; (b) presence of multiple problems (e.g., comorbid mental disorders, being in an ongoing violent relationship; being a refugee); (c) presence of a complex psychological traumatization, that is, childhood, multiple, or intentional trauma; and (d) the presence of a formal PTSD diagnosis per the DSM.

Overall, with PTSD/SUD patients, greater emotional intensity in sessions does not equal better outcomes. Both present- and past-focused models may be helpful to patients, based on readiness of the patient and clinician, training, setting, and other contextual factors. Such findings are consistent with psychotherapy research broadly, which shows that manual-based models perform equally well, including those developed for PTSD and those developed for SUD (Imel et al. 2008; Benish et al. 2007; Powers et al. 2010). The bottom line is that clinicians have a lot of choice in which models to use.

The most evidence-based model at this point is Seeking Safety (SS). SS has been very widely implemented in treatment programs for PTSD/SUD (as well as for either alone and for subthreshold patients). It has been the subject of the majority of PTSD/SUD studies, including 13 pilots, 3 controlled studies, and 7 RCTs (Najavits and Hien, 2013; Hien et al. in press). It is also the model with the most number of studies by independent investigators, which are less subject to positive bias (Chambless and Hollon 1998). SS has had consistently positive outcomes and is the only model thus far to outperform a control on both PTSD and SUD (Najavits and Hien 2013). However, some partial-dose SS studies were more mixed. Partial-dose studies used just 24–48 % of the model, including the largest study of SS, the National Institute on Drug Abuse Clinical Trials Network. SS is currently the only model for PTSD/SUD listed as having strong research support by professional entities, such as the International Society for Traumatic Stress Studies and Divisions 12 and 50 of the American Psychological Association.

Most studies addressed complex PTSD/SUD populations. It is heartening that the majority of PTSD/SUD studies addressed a broad range of patients: those with substance dependence rather than just substance abuse, those with drug disorders rather than just alcohol, and often including those with issues such as homelessness, domestic violence, suicidality, violence, serious and persistent mental illness, criminal justice involvement, unemployment, multiple prior treatment episodes, and low education. Inclusions and exclusions varied by study, but generally there were low to moderate exclusions in contrast to the relatively high exclusions in the PTSD-alone literature. Among PTSD/SUD studies, those with past-focused models had the most exclusions, in keeping with the PTSD-alone literature from which they derived. Exceptions, however, were Mills et al. (2012), Najavits and Johnson (2014), and Najavits et al. (2005), all of which had a broader range of patients.

Most studies used lower-cost formats for delivery of treatment models. The PTSD/SUD literature primarily uses group rather than individual therapy, open rather than closed groups, frontline clinicians who were native to the setting rather than brought in from the outside, and clinicians who were less highly trained (e.g., without advanced degrees). Such features are common in SUD treatment settings, which is where most of the studies were conducted. Here too, past-focused models differed overall, being conducted in individual modality and generally by highly trained clinicians brought in from the outside.

It appears easier to change PTSD than SUD. In the literature thus far, when there were differences between conditions, they were more often on PTSD or other mental health variables and less often on SUD. This may indicate that in PTSD/SUD patients, PTSD and mental health issues may be easier to treat than SUD. That remains a question for future research but does fit clinicians’ perceptions (Back et al. 2009). This pattern also fits the current view of PTSD as amenable to time-limited treatment, whereas SUD (severe SUD in particular) is conceptualized as a chronic relapsing disorder needing ongoing care (Arria and McLellan 2012).