Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Sep 14.
Published in final edited form as: Behav Ther. 2020 Apr 2;52(1):162–169. doi: 10.1016/j.beth.2020.03.008

Estimated Intelligence Moderates Cognitive Processing Therapy Outcome for Posttraumatic Stress Symptoms

Brian P Marx 1, Johanna Thompson-Hollands 1, Daniel J Lee 1, Patricia A Resick 2, Denise M Sloan 3
PMCID: PMC8439555  NIHMSID: NIHMS1738566  PMID: 33483114

Abstract

Although patient intelligence may be an important determinant of the degree to which individuals may comprehend, comply with, and ultimately benefit from trauma-focused treatment, no prior studies have examined the impact of patient intelligence on benefit from psychotherapies for PTSD. We investigated the degree to which educational achievement, often used as a proxy for intelligence, and estimated full scale intelligence quotient (FSIQ) scores themselves moderated treatment outcomes for two effective psychotherapies for PTSD: Cognitive Processing Therapy (CPT) and Written Exposure Therapy (WET). Participants, 126 treatment-seeking adults with PTSD (52% male; mean age = 43.9, SD = 14.6), were equally randomized to CPT and WET; PTSD symptom severity was measured at baseline and 6-, 12-, 24-, 36-, and 60-weeks following the first treatment session. Multilevel models revealed that participants with higher FSIQ scores experienced significantly greater PTSD symptom reduction through the 24-week assessment in CPT but not WET; this effect did not persist through the 60-week assessment. Educational achievement did not moderate symptom change through either 24- or 60-weeks. Individuals with higher FSIQ who are treated with CPT may experience greater symptom improvement in the early stages of recovery.

Keywords: PTSD, trauma, cognitive-behavioral treatment, moderators

Evidence-based, trauma-focused psychotherapies for posttraumatic stress disorder (PTSD) typically employ techniques that require higher levels of cognitive functioning (e.g., to verbally describe traumatic experiences in detail, to appraise and correct dysfunctional thinking patterns, and to make sense of experiences). As such, intelligence may be an important determinant of the degree to which patients may be able to comprehend, comply with, and ultimately benefit from trauma-focused treatment. This assumption is consistent with a well-established understanding that individuals with higher intelligence usually receive greater benefits from psychotherapy in general (e.g., Smith & Glass, 1977).

Yet, to date, no prior studies have examined the impact of intelligence on benefit from psychotherapies for PTSD. Researchers have explored the relation between education level (a possible proxy for intelligence) and PTSD treatment outcome and found inconsistent results. For example, in a small randomized controlled trial (N=28), Ehlers and colleagues (2005) compared cognitive therapy for PTSD (CT-PTSD) and a 3-month waitlist condition (WL) and found that, relative to WL, CT led to large reductions in PTSD symptoms, disability, depression, and anxiety. They also found that lower educational attainment was associated with better outcomes. The authors suggested that these results may represent a chance finding, given its inconsistency with a priori expectations. In a subsequent, much larger study, Ehlers et al. (2013) examined the degree to which CT-PTSD could be effectively implemented into a United Kingdom National Health Service Outpatient Clinic. Results of this investigation showed that education level was not associated with treatment outcomes. Finally, Cusack et al. (2019) examined moderators of treatment outcomes associated with group therapy for PTSD among veterans. These researchers used archival data from male veterans (N = 450) and employed multilevel modeling to examine change in self-efficacy (SE) over the course of treatment as well as moderation of change in SE as a function of race and educational attainment. Results indicated there was significant improvement in SE, with significantly different improvement based on education. Specifically, higher levels of education were associated with greater increases in SE after treatment. Although SE was significantly correlated with PTSD symptom severity, researchers did not examine the degree to which education level was associated with changes in PTSD symptoms.

Although intelligence and education level are significantly correlated, intelligence is a better measure of cognitive ability than educational level; an individual can achieve a relatively low education level (e.g., not complete high school) but have a high intelligence. Indeed, social and economic factors can influence educational achievement independent of intelligence (e.g., Israel, Beaulieu, & Hartless, 2001; Marks, Cresswell, & Ainley, 2006; Stewart, 2008; Travis & Kohli, 1995).

In this study, we examined the extent to which both education level and estimated intelligence quotient (IQ) moderated the relation between treatment condition and outcome using data from a study comparing two evidence-based, trauma-focused psychotherapies (Sloan, Marx, Lee, & Resick, 2018), Written Exposure Therapy (WET; Sloan & Marx, 2019) and Cognitive Processing Therapy (CPT; Resick, Monson, & Chard, 2016). The primary study found that WET was noninferior to CPT in terms of change in PTSD symptom severity (Sloan et al., 2018). Both WET and CPT are recommended treatments for PTSD (Department of Veteran Affairs and Department of Defense, 2017), but they have different emphases; CPT largely focuses on cognitive “stuck points,” whereas WET focuses on processing trauma memories in a more adaptive and functional manner. Consistent with results from prior studies of the influence of intelligence on psychotherapy outcomes (e.g., Smith & Glass, 1977), we expected that patients with higher estimated IQ scores would have significantly greater reductions in PTSD symptom severity treatment outcome than those with lower estimated IQ scores. Given that IQ and education are highly correlated and educational level has been found to moderate PTSD psychotherapy outcome (e.g., Cusack et al., 2019), we also expected to find that higher educational level would predict better outcome for participants randomized to both treatment conditions.

Method

PARTICIPANTS

Participants were 126 treatment-seeking adults (52% male) with PTSD. The mean age of the sample was 43.9 (SD = 14.6), and participants were largely White (54.8%) or African American (34.1%) and non-Hispanic (90.5%). To be eligible, individuals were required to meet DSM-5 criteria for PTSD (American Psychiatric Association, 2013) and if taking psychotropic medication, to be on a stable dose for at least 4 weeks prior to study entry. We excluded potential participants if they were currently at high risk for suicide, were actively psychotic or manic, had severe cognitive impairment, currently met criteria for a substance dependence diagnosis, or were currently engaged in other PTSD-focused psychotherapy. As stated elsewhere (Sloan, Marx, & Resick, 2016), cognitive impairment was evaluated with the Mini Mental State Exam (Folstein & McHugh, 1975); however, no participants were excluded from the study due to cognitive impairment (Sloan et al., 2018).

We included individuals regardless of trauma type, provided that their identified traumatic stressor met PTSD Criterion A. All participants provided written informed consent. Upon determining eligibility, participants were randomized 1:1 to either WET or CPT, resulting in a final sample of 126 individuals. The study was approved by local institutional review boards. Detailed study procedures can be found in “Brief Treatment for PTSD: A Noninferiority Trial” (Sloan et al., 2016).

MEASURES

We conducted assessments at baseline and 6, 12, 24, 36, and 60 weeks after the first treatment session. The 24-week assessment best represents the posttreatment assessment time point for CPT participants (100% of CPT participants had concluded their treatment at the time of their 24-week assessment), whereas the 6-week assessment best represents the posttreatment assessment time point for the WET participants (89% of WET participants had concluded their treatment at the time of their 6-week assessment; see, Sloan et al., 2018). The 36- and 60-week assessments constituted the follow-up period. Independent evaluators (IEs) who were unaware of participants’ treatment condition assignment conducted diagnostic interviews. Each diagnostic interview was digitally recorded, and 20% of each IE’s interviews were randomly selected for reliability analyses.

We used the Clinician Administered PTSD Scale for DSM-5 (CAPS-5; Weathers et al., 2013; Weathers et al., 2018) to assess PTSD diagnostic status and symptom severity. The CAPS-5 is a structured diagnostic interview; clinicians rate each PTSD symptom on a 5-point scale ranging from absent to extreme, considering both symptom frequency and intensity. Items are summed to create a total score reflecting overall PTSD symptom severity; higher scores indicate greater severity. Interrater reliability of the CAPS in this study was very good (κ = .85).

The Wechsler Test of Adult Reading (WTAR; Wechsler, 2001) is a brief premorbid estimate of intelligence. This measure consists of 50 words of increasing pronunciation difficulty that participants read aloud. The number of correctly pronounced words is used to calculate various estimates of IQ (e.g., Full Scale IQ, Verbal IQ). These estimates of IQ were co-normed along with the Wechsler Adult Intelligence Test, third edition (WAIS-III), and the Wechsler Memory Scale, fourth edition, among nationally representative samples in both the United States and the United Kingdom (Wechsler, 2001; Wechsler, 2009). Norms are stratified by age and educational attainment.

When considering respondents ages 18 or older (consistent with the minimum age for participation in this study), internal consistency of the WTAR scores were between .90–.97 for the American normative sample, and the standard error of measurement was between 2.6–4.4. Test-retest stability of the WTAR was .90–.94 (Wechsler, 2001). The WTAR is highly correlated with Verbal IQ (VIQ; rs = .72–.80) and Full-Scale IQ (FSIQ, rs = .68–.80), as measured by the Wechsler Adult Intelligence Scales-III (WAIS-III; Wechsler, 2001). In this study, we only used participants’ WTAR-estimated FSIQ scores. IEs administered the WTAR during the baseline assessment. An advantage of using the WTAR to estimate IQ is that this measure is a good estimate of premorbid IQ level, including estimating premorbid IQ after traumatic brain injury (Steward et al., 2017), which is not uncommon among individuals with PTSD. Average estimated FSIQ among the full sample was 103.03 (SD = 11.38, minimum = 71, maximum = 119), in the WET condition was 101.37 (SD = 11.95, minimum = 71, maximum = 119), and in the CPT condition was 104.70 (SD = 10.61, minimum = 83, maximum = 119).

TREATMENTS

WET is a five-session treatment during which patients write about a specific traumatic event for 30 minutes each session. During the writing sessions, individuals provide detailed descriptions of their traumatic experiences, focusing on the thoughts and emotions felt during or immediately after exposure, and make meaning of the experience within the broader context of their lives. There is no homework assigned between the treatment sessions. The range of time to complete the treatment protocol was 3 to 6 weeks (average completion time = 4.17 weeks, SD = 1.24).

CPT is a 12-session treatment during which patients learn to challenge dysfunctional cognitions about their traumatic event, as well as challenge distorted thoughts about themselves, others, and the world. CPT does include homework that patients are assigned to complete between sessions; specifically, patients complete a trauma impact statement, a detailed account of their trauma, and various worksheets focused on challenging cognitions. Participants were asked to attend CPT sessions twice per week; however, the range of time to complete CPT was 6 to 20 weeks (average completion time = 12.70 weeks, SD = 3.44). The CPT protocol used in this study is now referred to as Cognitive Processing Therapy Plus Written Accounts (CPT+A; Resick et al., 2016).

Therapists were doctoral-level psychologists; each completed a 2-day CPT workshop and 2-hour WET workshop prior to treating participants in the study. All therapists administered both WET and CPT. Therapists were supervised throughout the study by the treatment developers, who also had access to recorded treatment sessions. Treatment adherence and competence was evaluated by independent CPT and WET clinicians who rated 15% of the sessions that were randomly selected.

DATA ANALYSIS

We used multilevel modeling (MLM; Raudenbush & Bryk, 2002) in R (R Core Team, 2017) to examine the effect of the hypothesized moderators on changes in PTSD symptoms over time. MLM permits the exploration of within-person and between-person change simultaneously, to more appropriately capture contextual effects.

We conducted analyses using all data points through 24 weeks, and separately through 60 weeks, for all participants who were randomized (i.e., intent to treat). Two-level models (within-person level 1, between-person level 2) were run, with CAPS-5 total score as the dependent variable. We created models of change in PTSD symptom severity over time including fixed effects of the intercept, linear time, and quadratic time (because quadratic time had been shown to improve fit in our previous analyses; see Thompson-Hollands et al., 2018), as well as random effects for the intercept and linear slope. To test the potential effect of our moderating variables of interest, we added fixed effects for each moderator (in separate models), a Moderator × Linear Time interaction, and a Moderator × Linear Time × Treatment Condition interaction. The following variables were centered: linear and quadratic time (entered as number of weeks, centered at baseline) and FSIQ (centered at 100). Educational achievement was a categorical dichotomized variable (0 = less than a 4-year college degree, 1 = bachelor’s degree or higher). Missing data were minimal, with only 10.6% of all planned CAPS-5 assessments missed and none of the moderator variables missed. MLM is well-suited to handling missing data, producing robust inferences when the proportion of missing data is small (Raudenbush & Bryk, 2002).

POWER ANALYSIS

Using the Mplus statistical software (Muthén & Muthén, 1998–2009), we estimated power to detect significant moderators of treatment outcome (i.e., change over time or Moderator × Time interactions) using the Monte Carlo method (Muthén & Muthén, 2002). This method involves specifying parameter estimates for a given model and generating 10,000 datasets to determine the percentage of time the parameter estimates were statistically significant (at the p < .05 level). We used parameter estimates from Resick et al. (2008). Two of the three conditions of Resick et al.’s dismantling trial were very similar to the two conditions of this study (i.e., CPT full treatment protocol and the written trauma account only). We conducted growth curve models using data from the CPT and written accounts conditions from Resick et al. (2008) and examined dissociation as a moderator (Resick et al., 2012) to produce parameter estimates for a viable example of a potential moderator. This provided us with a precise estimate of power for all pertinent parameters. The power analysis estimated the power to detect a significant effect that included treatment Condition × Moderator × Time interaction was .92 and .99 to detect a medium and large effect size, respectively.

Results

Descriptive statistics for included measures, partitioned by treatment condition, are presented in Table 1; conditions did not differ in age, sex, race, baseline PTSD symptom severity, or any examined moderator variables. For the models examining FSIQ through 24 weeks, while there was no effect of FSIQ × Time (B=<0.01, SE=0.01, t=0.49, p= .626), we found a significant moderating effect of FSIQ × Time × Treatment (B=−0.02, SE=0.01, t=−2.55, p=.012; see Table 2). Specifically, individuals who received CPT and had higher FSIQ scores experienced significantly greater symptom reduction from baseline to 24 weeks post first session. For illustrative purpose, Figure 1 displays the different average symptom trajectories for CPT and WET participants with varying FSIQ scores; although it should be noted that moderator analyses were conducted using a continuous value for FSIQ scores, the categorical depiction is provided to visualize the FSIQ findings. As illustrated in Figure 1, participants with varying degrees of FSIQ appear to benefit from CPT, but those with the highest estimated FSIQ appear to benefit the most. Post-hoc within-condition analyses through 24 weeks confirmed that while the moderation effect of FSIQ × Time was significant in CPT (B=−0.02, SE=0.01, t=−2.70, p=.010), there was no corresponding effect on symptom change among participants who received WET (B=<0.01, SE=<0.01, t=0.56, p=.580). In our whole-sample analyses through 60 weeks, there was no moderating effect of either FSIQ × Time (B=<0.01, SE=<0.01, t = 0.85, p = .397) or FSIQ × Time × Treatment (B =<0.01, SE =<0.01, t = 1.16, p = .249). Thus, FSIQ did not moderate PTSD outcome for either WET or CPT at any assessment period examined. For the model testing the effect of educational achievement through 24-weeks, we found no moderating effect for either Education × Time (B=−0.10, SE=0.15, t=−0.68, p=.497) or for Education × Time × Treatment (B=−0.19, SE=0.21, t =−0.91, p = .363). The same was true when examining the results through the 60-week time-point; neither Education × Time (B=−0.06, SE=0.06, t=−1.04, p=.299) nor Education × Time × Treatment (B=0.07, SE=0.08, t=0.80, p=.424) were significant.

Table 1.

Descriptive Statistics of Included Measures by Condition

Measure/Demographic WET CPT t, df
Mean SD Mean SD
Age 44.89 14.81 42.83 14.40 0.79, 124
WTAR Estimated FSIQ 101.41 12.26 104.79 10.90 −1.64, 124
WTAR Estimated Verbal IQ 101.37 11.95 104.70 10.61 −1.66, 124
Symptom duration (months) 201.40 193.54 187.35 166.88 0.43, 122
# comorbid diagnoses 1.32 1.40 1.00 1.12 1.40, 124
BDI-II 21.11 10.49 22.87 12.80 −0.84, 123
Baseline CAPS-5 total score 36.13 8.89 37.10 10.07 −0.57, 124
n % n % X2,df
Male 33 52.38 33 52.38 0.00, 1
White 36 57.14 33 52.38 0.29, 1
MDD diagnosis 18 28.57 19 30.16 0.06, 1
Open in a new tab

Note. BDI-II = Beck Depression Inventory (2nd edition); CPT = Cognitive Processing Therapy; FSIQ = Full-Scale IQ from the Wechsler Test of Adult Reading; MDD = major depressive disorder; WET = Written Exposure Therapy; WTAR = Wechsler Test of Adult Reading;

*

p<.05.

Table 2.

Within-Condition Moderating Variables Through the Posttreatment and Follow-up Periods

Through 24 weeks Through 60 weeks
WET CPT WET CPT
Time by: B SE B SE B SE B SE
 Sex 0.05 0.12 0.19 0.17 0.08 0.05 0.06 0.06
 Age 0.01 0.00 0.01 0.01 0.00 0.00 0.00 0.00
 College graduate −0.10 0.13 −0.29 0.16 −0.06 0.05 0.01 0.07
 FSIQ 0.00 0.01 −0.02 0.01 ** 0.00 0.00 −0.00 0.00
 Symptom duration 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
 Number of comorbid diagnoses −0.05 0.04 −0.04 0.07 −0.01 0.02 −0.01 0.03
 BDI-II 0.00 0.01 −0.01 0.01 −0.00 0.00 −0.00 0.00
 MDD diagnosis −0.01 0.14 −0.15 0.19 −0.04 0.06 0.03 0.07
Open in a new tab

Note. N = 63 for each treatment. WET = Written Exposure Therapy; CPT = Cognitive Processing Therapy; FSIQ = Full-Scale IQ from the Wechsler Test of Adult Reading; BDI-II = Beck Depression Inventory (2nd edition); MDD = major depressive disorder.

*

p<.05,

**

p<.01

FIGURE 1.

FIGURE 1

Open in a new tab

Moderation of slope of change in PTSD symptoms by estimated full-scale IQ in Cognitive Processing Therapy and Written Exposure Therapy. Please note that multilevel modeling was used to examine the effect of estimated IQ on PTSD outcome.

Given between-condition differences in treatment dropout observed in this trial (6.4% in WET, 39.7% in CPT), follow-up analyses were conducted to examine if the observed effect of IQ on symptom course in CPT could be attributable to IQ predicting dropout in CPT. However, IQ did not predict dropout among participants who received CPT (b = −0.03, SE = 0.03, p = .19).

Discussion

We found that individuals randomized to receive CPT who had higher estimated FSIQ, based on WTAR scores, experienced significantly greater reduction in PTSD symptom severity through the 24-week assessment time point relative to those with lower estimated FSIQ; there was no significant moderating effect of estimated FSIQ on PTSD symptom severity at the 60-week assessment time point. In contrast, we found no effect for estimated FSIQ through 24 or 60 weeks among those who received WET. These findings suggest that CPT may be particularly beneficial for individuals with higher IQ, in that they may be more likely to experience significantly greater early symptom improvement. Prior research suggests that such early symptom improvement may be an important predictor of treatment response to CPT (Holmes et al., 2019), although our findings do not suggest any association between estimated FSIQ and long-term outcomes. Individuals with higher IQ may be better able to engage in key features of CPT such as challenging dysfunctional cognitions about the trauma and distorted thoughts about themselves, others, and the world. Higher IQs are associated with better working memory and recall, information processing speed, attentional control, and the ability to inhibit behavior (Best, Miller, & Naglieri, 2011; Clark, Prior, & Kinsella, 2002). These cognitive abilities may, in turn, be associated with an increased ability to shift cognitive sets and perspectives (skill sets critical in CPT). However, studies specifically designed to measure these processes are needed to establish that link. Crucially, the findings do not suggest that CPT should only be used with individuals with higher IQs, as those with lower estimated IQs also showed significant improvement. Rather, the findings suggest that the early treatment gains associated with CPT might be larger for those individuals with higher IQ. Thus, clinicians should consider assessing patients’ estimated IQ prior to treatment using the WTAR or another brief test of cognitive ability. If a patient has a particularly high IQ they should consider selecting CPT.

Another important aspect of understanding the potential role of IQ in treatment outcome among those who receive CPT is that prior studies have shown that cognitive development and subsequent functioning may be affected negatively by adverse and traumatic experiences, especially in childhood (Enlow et al., 2012; Kuo et al., 2011). In such cases, trauma exposure may be the cause of both PTSD and impaired cognitive functioning. Research demonstrating that individuals with lower IQs are at significantly greater risk for developing PTSD in the wake of trauma exposure (Macklin et al., 1998; McNally & Shin, 1995) may even suggest that IQ mediates the association between trauma exposure and PTSD. Our findings, taken together with those of past studies, suggest that additional research is needed to more fully understand the degree to which individuals’ cognitive resources can affect recovery from the sequelae to trauma exposure.

The finding that estimated IQ did not moderate outcome for participants assigned to WET suggests that the essential elements of WET (writing in detail about and making sense of one’s traumatic experiences) may be less affected by lower IQ and associated impairments in executive functioning processes, perhaps because they rely less on abstract thinking skills. Because WET primarily involves patients writing about their traumatic experiences, it may be assumed that individuals who are less educated or have lower IQs have more limited verbal ability and would therefore reap less benefit from a treatment that relies on writing. However, prior research has not supported that assumption (Sloan & Marx, 2019) and the findings from this study do not support that assumption. To protect against any concerns or biases that patients may have about their ability to complete the writing assignments in appropriate fashion, the treatment protocol emphasizes that the quality of the writing (e.g., grammar, spelling, sophistication of the writing) has no bearing on treatment outcome. Rather, the most important aspect of the writing is to confront one’s trauma memory, which is achieved by following the instructions for the written narrative.

Although we did find a moderating effect for estimated FSIQ among participants who received CPT in the more immediate aftermath of treatment, we did not observe any moderating effects for education level at any time point. These results support our contention that education level is an inadequate proxy for intelligence and should not be used as a moderator variable in PTSD treatment outcome studies because it is influenced by other social and cultural factors.

This study has several strengths, including the large sample size, patient heterogeneity, and the inclusion of two trauma-focused treatments that differ in the emphasis on cognitive restructuring and imaginal exposure. An important limitation of this study is that, although the WTAR has strong psychometric properties and its scores are significantly correlated with scores from more comprehensive intelligence tests (Wechsler, 2001), they are only an estimate of FSIQ. Administering a more comprehensive IQ test (e.g., Wechsler Adult Intelligence Scales) would provide a more precise determination of IQ than the WTAR. However, such an assessment in the context of an RCT, or everyday clinical practice, is impractical, time consuming, and overly burdensome to both clinicians and participants/patients. Time efficient, less burdensome measures that provide a reliable and valid estimate of premorbid FSIQ, such as the WTAR, are reasonable alternatives for clinicians and their clients. That being said, it will be important for continued examination of IQ as a moderator of PTSD treatment outcome to both replicate the findings of the current study but also examine whether IQ moderates PTSD symptom changes associated with different types of PTSD treatments. There is a relatively large number of evidence-based treatments for PTSD that vary in treatment components (e.g., focus on exposure therapy versus focus on cognitive therapy). Another limitation of this study is that we did not examine whether type of traumatic event influenced moderator findings because of the small cell sizes for some trauma types reported by our study participants, but this would be an important avenue for future investigations.

We have made significant advances in identifying effective PTSD treatments over the past 30 years. The challenge for the field now is to understand which treatments are best for which patients under what set of circumstances (Paul, 1967). Gaining a better understanding of patient characteristics that predict success in specific PTSD treatments will further advance the effectiveness and decrease dropout among patients who receive evidence-based interventions.

Acknowledgments

This work was supported by grant R01 MH095737 from the National Institute of Mental Health. Dr. Thompson-Hollands was supported by the U.S. Department of Veterans Affairs (Clinical Sciences Research and Development Service) under Career Development Award # IK2 CX001589. Dr. Lee was supported by National Institute of Mental Health award #5T32MH019836-16.

Footnotes

Conflict of Interest Statement

Drs. Sloan and Marx receive royalties for Written Exposure Therapy treatment protocol and Dr. Resick receives royalties for Cognitive Processing Therapy treatment manual. These two treatments were examined in the study described in this manuscript.

Contributor Information

Patricia A. Resick, Duke University Medical Center

Denise M. Sloan, National Center for PTSD at VA Boston Healthcare System, Boston University School of Medicine

References

  1. American Psychiatric Association (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Washington, DC: American Psychiatric Association. [Google Scholar]
  2. Best JR, Miller PH, & Naglieri JA (2011). Relations between executive function and academic achievement from ages 5 to 17 in a large, representative national sample. Learning and Individual Differences, 21(4), 327–336. 10.1016/j.lindif.2011.01.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Clark C, Prior M, & Kinsella G (2002). The relationship between executive function abilities, adaptive behaviour, and academic achievement in children with externalizing behaviour problems. Journal of Child Psychology and Psychiatry, 43(6), 785–796. 10.1111/1469-7610.00084 [DOI] [PubMed] [Google Scholar]
  4. Cusack SE, Coleman JA, Rappaport LM, & Sheerin C (2019). Moderation of improvement in self-efficacy following group psychotherapy for PTSD. Psychological Services, 16(4), 657–663. 10.1037/ser0000260 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Department of Veteran Affairs and Department of Defense (2017). Clinical practice guideline for the management of posttraumatic stress disorder and acute stress disorder. https://www.healthquality.va.gov/guidelines/mh/ptsd
  6. Ehlers A, Clark DM, Hackmann A, McManus F, & Fennell M (2005). Cognitive therapy for post-traumatic stress disorder: Development and evaluation. Behaviour Research and Therapy, 43, 413–431. 10.1016/j.brat.2004.03.006 [DOI] [PubMed] [Google Scholar]
  7. Ehlers A, Grey N, Wild J, Stott R, Liness S, Deale A, … Manley J (2013). Implementation of cognitive therapy for PTSD in routine clinical care: Effectiveness and moderators of outcome in a consecutive sample. Behaviour Research and Therapy, 51(11), 742–752. 10.1016/brat.2013.08.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Enlow MB, Egeland B, Blood EA, Wright RO, & Wright RJ (2012). Interpersonal trauma exposure and cognitive development in children to age 8 years: A longitudinal study. Journal of Epidemiological Community Health, 66(11), 1005–1011. 10.1136/jech-2011-200727 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Folstein M, & McHugh P (1975). Mini mental state a practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research, 12, 189–198. 10.1016/0022-3956(75)90026-6 [DOI] [PubMed] [Google Scholar]
  10. Holmes SC, Johnson CM, Suvak MK, Sijercic I, Monson CM, & Stirman SW (2019). Examining patterns of dose response for clients who do and do not complete cognitive processing therapy. Journal of Anxiety Disorders, 68, 102–120. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Israel GD, Beaulieu LJ, & Hartless G (2001). The influence of family and community social capital on educational achievement. Rural Sociology, 66, 43–68. [Google Scholar]
  12. Kuo JR, Goldin PR, Werner K, Heimberg RG, & Gross JJ (2011). Childhood trauma and current psychological functioning in adults with social anxiety disorder. Journal of Anxiety Disorders, 25, 467–473. 10.1016/j.janxdis.2010.11.011 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Macklin ML, Metzger LJ, Litz BT, McNally RJ, Lasko NB, Orr SP, & Pitman RK (1998). Lower precombat intelligence is a risk factor for posttraumatic stress disorder. Journal of Consulting and Clinical Psychology, 66, 323–326. 10.1037/0022-006X.66.2.323 [DOI] [PubMed] [Google Scholar]
  14. Marks GN, Cresswell J, & Ainley J (2006). Explaining socioeconomic inequalities in student achievement: The role of home and school factors. Educational Research and Evaluation, 12, 105–128. 10.1080/13803610600587040 [DOI] [Google Scholar]
  15. McNally RJ, & Shin LM (1995). Association of intelligence with severity of posttraumatic stress disorder symptoms in Vietnam combat veterans. American Journal of Psychiatry, 152, 936–938. 10.1176/aip.152.6.936 [DOI] [PubMed] [Google Scholar]
  16. Muthén LK, & Muthén BO (1998–2009). Mplus User’s Guide (5th ed.). Muthén & Muthén [Google Scholar]
  17. Muthén LK, & Muthén BO (2002). How to use a Monte Carlo study to decide on sample size and determine power. Structural Equation Modeling, 4, 599–620. [Google Scholar]
  18. Paul GL (1967). Strategy of outcome research in psychotherapy. Journal of Consulting Psychology, 31(2), 109–118. 10.1037/h0024436 [DOI] [PubMed] [Google Scholar]
  19. R Core Team (2017). R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. URL https://www.R-project.org [Google Scholar]
  20. Raudenbush SW, & Bryk A (2002). Hierarchical linear models: Applications and data analysis methods. Vol. 1. (pp.): Sage. [Google Scholar]
  21. Resick PA, Galovski TE, Uhlmansiek MO, Scher CD, Clum GA, & Young-Xu Y (2008). A randomized clinical trial to dismantle components of cognitive processing therapy for posttraumatic stress disorder in female victims of interpersonal violence. Journal of Consulting and Clinical Psychology, 76, 243–258. 10.1037/0022-006X.76.2.243 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Resick PA, Monson CM, & Chard KM (2016). Cognitive Processing Therapy for PTSD: A comprehensive manual. Guilford Press. [Google Scholar]
  23. Resick PA, Suvak MK, Johnides BD, Mitchell KS, & Iverson KM (2012). The impact of dissociation on PTSD treatment with Cognitive Processing Therapy. Depression and Anxiety. 10.1002/da.21938 [DOI] [PubMed] [Google Scholar]
  24. Sloan DM, & Marx BP (2019). Written Exposure Therapy for PTSD: A brief treatment approach for mental health professionals. American Psychological Association. [Google Scholar]
  25. Sloan DM, Marx BP, Lee DJ, & Resick PA (2018). A brief exposure-based treatment versus cognitive processing therapy for posttraumatic stress disorder: A randomized noninferiority clinical trial. JAMA Psychiatry, 75(3), 233–239. 10.1001/jamapsychiatry.2017.4249 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Sloan DM, Marx BP, & Resick PA (2016). Brief treatment for PTSD: A noninferiority trial. Contemporary Clinical Trials, 48, 76–82. 10.1016/j.cct.2016.04.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Smith ML, & Glass GV (1977). Meta-analysis of psychotherapy outcome studies. American Psychologist, 32, 752–760. 10.1037/0003-066X.32.9.752 [DOI] [PubMed] [Google Scholar]
  28. Steward KA, Novack TA, Kennedy R, Crowe M, Marson DC, & Triebel KL (2017). The Wechsler Test of Adult Reading as a measure of premorbid intelligence following traumatic brain injury. Archives of Clinical Neuropsychology, 32, 98–103. 10.1093/arclin/acw081 [DOI] [PubMed] [Google Scholar]
  29. Stewart EB (2008). School structural characteristics, student effort, peer associations, and parental involvement: The influence of school- and individual-level factors on academic achievement. Education and Urban Society, 40, 179–204. 10.1177/0013124507304167 [DOI] [Google Scholar]
  30. Thompson-Hollands J, Marx BP, Lee DJ, Resick PA, & Sloan DM (2018). Long-term treatment gains of a brief exposure-based treatment for PTSD. Depression and Anxiety, 35, 985–991. 10.1002/da.22825 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Travis R, & Kohli V (1995). The birth order factor: Ordinal position, social strata, and educational achievement. Journal of Social Psychology, 135, 499–507. 10.1080/00224545.1995.9712218 [DOI] [PubMed] [Google Scholar]
  32. Weathers FW, Blake DD, Schnurr PP, Kaloupek DG, Marx BP, & Keane TM (2013). The Clinician-Administered PTSD Scale for DSM-5 (CAPS-5). Interview available from the National Center for PTSD; at www.ptsd.va.gov [Google Scholar]
  33. Weathers FW, Bovin MJ, Lee DJ, Sloan DM, Schnurr PP, Kaloupek DG, … Marx BP (2018). The Clinician-Administered PTSD Scale for DSM-5 (CAPS-5): Development and initial psychometric evaluation in military veterans. Psychological Assessment, 30, 383–395. 10.1037/pas0000486 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Wechsler D (2001). The Wechsler test of adult reading (WTAR). The Psychological Corporation. [Google Scholar]
  35. Wechsler D (2009). Wechsler memory scale (4th ed.). The Psychological Corporation [Google Scholar]

RESOURCES