Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Feb 1.
Published in final edited form as: Psychol Trauma. 2022 Jun 6;16(2):331–339. doi: 10.1037/tra0001249

The Effect of Time Since Index Trauma on Trauma-Related Beliefs

Travis A Cole 1, Lillian Reuman 1,2,3, Daniel J Lee 1,2, Chelsea Shotwell Tabke 1, Brian P Marx 1,2, Denise M Sloan 1,2
PMCID: PMC9896149  NIHMSID: NIHMS1866363  PMID: 35666937

Abstract

Objective:

Greater duration of negative trauma-related beliefs may inhibit improvements in these cognitions during posttraumatic stress disorder (PTSD) treatment. The goal of the current study was to examine the impact of time since trauma on change in negative trauma-related beliefs during PTSD treatment.

Method:

A sample of 126 adults diagnosed with PTSD were randomized to Cognitive Processing Therapy (CPT) or Written Exposure Therapy (WET) and completed the Posttraumatic Cognitions Inventory (PTCI) at pretreatment and 6, 12, 24, 36, and 60 weeks following the first treatment session.

Results:

Consistent with past findings, negative cognitions related to the self, others, and self-blame decreased during and after trauma-focused therapy. Greater time since the trauma was associated with less improvement in negative trauma-related beliefs about the world and less early improvement in negative trauma-related beliefs about the self and self-blame. The effect of time since trauma on reduction in negative trauma-related beliefs about the world was stronger among participants randomized to CPT.

Conclusions:

These results indicate that trauma-focused therapies become less potent in reducing trauma-related beliefs as time progresses since trauma exposure, particularly in CPT, for beliefs about the world. Results indicate PTSD treatment should be provided in closer proximity to trauma exposure before negative trauma-related beliefs become inflexible.

Keywords: posttraumatic stress disorder, cognitions, cognitive processing therapy, written exposure therapy, posttraumatic cognitions inventory

The Effect of Time Since Trauma Event on Trauma-Related Beliefs

Theoretical models and empirical findings suggest that negative trauma-related beliefs, or distorted interpretations of one’s safety, agency, trust in others, and blame following a traumatic event, play an important role in the development and maintenance of posttraumatic stress disorder (PTSD; e.g., Brown et al., 2019; Foa & Kozak, 1986; Foa & Rothbaum, 1998). These beliefs relate to several domains, including the self (e.g., “I am a weak person”), the world (e.g., “The world is completely dangerous”), and blame (e.g., “The event happened because of the way I acted”). Negative trauma-related cognitions are thought to arise from and contribute to distress as well as prompt avoidance of negative trauma-related beliefs and stimuli (Ehlers & Clark, 2000; Foa & Rothbaum, 1998).

Although research has been sparse, evidence indicates that when left untreated, negative trauma-related beliefs tend to worsen over time (Dunmore et al., 2001; O’Donnell et al., 2007). In a study of 253 survivors of physical injury, O’Donnell and colleagues (2007) found that negative posttraumatic cognitions not only intensified over time, but also directly and indirectly predicted PTSD symptom severity 12 months later. Results from a longer term (17 year), prospective study revealed that negative trauma-related cognitions worsened over time among Israeli combat veterans with PTSD (Dekel et al., 2013). The investigators also found that negative trauma-related cognitions were positively associated with the trajectory of PTSD.

Some interventions for PTSD, such as Cognitive Processing Therapy (CPT; Resick et al., 2016), specifically target challenging and changing negative trauma-related beliefs. Other PTSD treatments, such as Written Exposure Therapy (WET; Sloan & Marx, 2019), also target negative trauma-related beliefs but in indirect ways. WET is a five session, exposure-based PTSD treatment in which individuals write trauma narratives as a form of exposure to their traumatic memory. Through these written narratives, corrective information about the trauma memory becomes more accessible allowing for reductions in the strength of negative trauma-related beliefs throughout treatment. Findings indicate that WET is efficacious (Sloan et al., 2013, 2018) and effective (LoSavio et al., 2021) in the treatment of PTSD. Trauma-focused PTSD interventions, both those with a cognitive focus and those with an exposure focus, result in significant reductions in negative trauma-related beliefs (Holliday et al., 2014; Iverson et al., 2015; Kumpula et al., 2017; Lee et al., 2021; Scher et al., 2017; Sobel et al., 2009). These reductions are consistent regardless of trauma type, veteran status, or gender (Holliday et al., 2018). Reductions in negative trauma-related beliefs may even function as a mediator of treatment outcome (e.g., Holliday et al., 2018; Zalta et al., 2014), although further research in this area is needed (Lee et al., 2021).

Despite the evidence that negative trauma-related beliefs generally decrease over the course of PTSD treatment, there has been sparse investigation of whether time since the trauma event affects the degree to which negative trauma beliefs change during and after treatment. Like behavior malleability, which relies heavily on the quantity and frequency of a reinforcement (Kassin, 2003), repetitive thought is a crucial part of belief development (Watkins, 2008). Unlike most other beliefs, those related to a traumatic event have an approximated origin date, for which “time since trauma” can serve as a proxy. Increased time since the development of negative trauma-related beliefs represent a larger window for appraisals—and functionally related behaviors—to be repeated and confirmed, providing more opportunities to engage in a negative feedback loop with other co-occurring PTSD symptoms at greater quantities. For example, an individual may believe the world is dangerous and avoid going outside, which also reinforces the belief. Some evidence suggests that time since the traumatic event has no or little effect on the outcome of trauma-focused treatments (Barawi et al., 2020). As summarized by Barawi et al. (2020), results from two studies revealed a nonsignificant association between time since index trauma and treatment outcome in cognitive therapy (Ehlers et al., 2003, 2005). Similar results were also found in studies examining CPT and Eye Movement Desensitization and Reprocessing (Karatzias et al., 2007; Roberge et al., 2019). However, Lewis et al. (2017) found that participants with a more recent trauma exhibited a slightly improved treatment outcome after completing an Internet-based guided self-help treatment. No studies to date have examined the impact of time since the traumatic event on negative trauma-related belief change within trauma-focused treatment.

In this study, we examined the degree to which time since trauma event was related to changes in negative trauma-related beliefs during and after treatment for individuals who received either CPT or WET. Although we expected that significant reductions in negative trauma-related beliefs would be observed for both trauma-focused treatments, we expected that longer time since trauma event would be associated with less improvement in all three domains of negative trauma-related beliefs (i.e., negative beliefs about the self, negative beliefs about the world, and self-blame) following both trauma-focused treatments.

Method

The current study represents a secondary analysis of a larger study in which the goal was to examine if WET was noninferior to CPT, despite the more time intensive nature of CPT. The study found that WET was noninferior to CPT and WET was associated with a significantly lower dropout rate (Sloan et al., 2018).

Participants

One hundred and twenty-six adults (47.62% female; n = 60) participated in the study, which was conducted in a large city in the northeastern section of the United States. The participants were diverse in terms of racial background, with 69 participants (54.76%) identifying as White, 43 (34.13%) identifying as Black, seven (5.56%) identifying as “other,” four (3.17%) identifying as American Indian/Alaskan Native, two (1.59%) identifying as Asian, and one (.79%) identifying as Pacific Islander/Native Hawaiian. Participants’ M age was 43.86 years (SD = 14.6).

As reported by Sloan et al. (2018), eligibility criteria required a PTSD diagnosis as determined by the Clinician-Administered PTSD Scale for DSM–5 (CAPS-5; Weathers et al., 2018) and a stable dose of psychotropic medication for at least 4 weeks (if taking any). Exclusion criteria included high risk of suicide, current psychosis or mania, or current psychotherapy for PTSD. The index event for which the participant met PTSD diagnosis was varied for the sample, including adult nonsexual assault (n = 24; 19%), childhood sexual assault (n = 20; 15.9%), adult sexual assault (n = 19; 15.1%), and combat-related events (n = 16; 12.7%). The average time since the index trauma event was 219.56 months (SD = 190.41 months; range: 2 to 684 months). Table S1 in the online supplemental materials presents a full summary of participant demographics and information regarding participants’ reported index trauma.

Measures

The Traumatic Life Events Questionnaire (TLEQ)

The TLEQ (Kubany et al., 2000) is a self-report questionnaire that assesses exposure to 21 potentially traumatic events. The number of times an event has been experienced is scored on a Likert scale ranging from 0 (never) to 6 (more than five times). In the case of exposure to multiple traumatic events, individuals determine which event they perceive to be the most distressing; this event is referred to as their index trauma. Participants also provide additional information about the index trauma, including their age at the time of the event. Time since index trauma (calculated by subtracting age at index trauma from age at baseline assessment) was used as a proxy for the length of time participants have held negative trauma-related beliefs. All participants completed the TLEQ at the baseline assessment.

Posttraumatic Cognitions Inventory (PTCI)

The PTCI (Foa et al., 1999) is a 33-item self-report measure on which participants rate their agreement with statements assessing their posttraumatic maladaptive beliefs (e.g., “I cannot rely on myself”) on a Likert scale ranging from 1 (totally disagree) to 7 (totally agree). The PTCI consists of three subscales: Negative Cognitions About Self (NCS), Negative Cognitions About the World (NCW), and Self-Blame (SB). Higher scores indicate greater endorsement of maladaptive cognitions. Previous research has established strong psychometric properties for the PTCI (Beck et al., 2004; Foa et al., 1999). Participants completed the PTCI at baseline and 6, 12, 24, 36, and 60 weeks following the first treatment session. In the current sample, internal consistency was good to excellent for each of the three subscales across all assessment time points (NCS, αs = .95–.96; NCW, αs = .87–.95; SB, αs = .82–.87).

Treatment

Written Exposure Therapy (WET; Sloan & Marx, 2019) is a five session, weekly treatment in which patients are instructed to write about a specific traumatic event for 30 minutes during each of five sessions. Participants were instructed to focus on event details, thoughts and feelings that occurred during the event, and the impact of the event. Following each writing session, therapists conduct a 10-minute check-in with each participant about their experiences on writing about the traumatic event. Participants are not assigned any out-of-session assignments in WET. CPT (Resick et al., 2016) consists of 12, one-hour sessions. CPT includes psychoeducation about PTSD followed by the introduction of skills to identify and challenge trauma-related beliefs. Later sessions explicitly address themes of safety, trust, power/control, esteem, and intimacy in relation to PTSD. The original protocol includes a written trauma account (after Sessions 3 and 4) that included sensory details, thoughts, and feelings about a specific traumatic event. These accounts are completed as homework assignments to be conducted outside of session and then brought into the subsequent session. Participants had unlimited time to complete two trauma accounts.

Treatment for both conditions was delivered in an individual, weekly format by 10 doctoral-level therapists who had extensive PTSD treatment experience. All therapists administered both CPT and WET. Sessions were recorded, and independent evaluators rated treatment adherence and competence for 20% of randomly selected sessions (Sloan et al., 2018).

Procedure

All study procedures were approved by the local Institutional Review Board. Recruitment consisted of referrals from community mental health clinics and advertisements. A total of 126 participants met eligibility criteria following the baseline assessment and were randomized to receive either WET (n = 63) or CPT (n = 63).

After completing the baseline assessment, which included the TLEQ and PTCI, all participants completed the PTCI during assessments at 6, 12, 24, 36, and 60 weeks following the first treatment session. Given the substantial difference in treatment length between the two treatment conditions (i.e., 5 vs 12 weeks), assessments were structured across time with the intention of capturing posttreatment (6 weeks and 12 weeks) for the two treatment conditions and several follow up assessments (24, 36, and 60 weeks). All participants randomized to the WET condition completed treatment by the 6-week assessment. Participants randomized to the CPT condition completed treatment between 6 and 20 weeks.

Data Analytic Strategy

All analyses were conducted using Mplus Version 8 (Muthén & Muthén, 1998–2017) with the intent-to-treat sample that was assessed at baseline and 6, 12, 24, 36, and 60 weeks following the first treatment session. We investigated change in trauma-related beliefs by examining means and standard deviations of PTCI scores over time, effect sizes of change over time, and growth curve models of change over time.

Effect Size Calculations

To measure the magnitude of within- and between-condition change (between baseline and a given timepoint) in PTCI subscale scores, we calculated Cohen’s d effect sizes (Cohen, 1988). To quantify magnitude of change on a standardized metric, we used Cohen’s d with a correction for repeated measures designs (Morris & DeShon, 2002). Effect sizes were assessed according to Cohen’s (1988) recommendations: small (d = .20), medium (d = .50), and large (d = .80).

Growth Curve Modeling

Five growth curve models for each PTCI subscale were investigated. First, we examined linear and quadratic models in which time was coded as weeks since baseline (e.g., 0, 6, 12). Next, we examined slope factor models. In this approach, the loadings of the first and last observation to the slope factor were fixed to 0 and 1, respectively, and the remaining loadings were freely estimated. Next, we examined logarithmic models. Finally, we examined piecewise growth curve models. This approach allowed for distinct periods of change. For these models, the first slope measured change in weeks 0–24, and the second slope measured change in weeks 24–60. We examined piecewise growth curve models in which time was matched to weeks and another model utilizing a slope factor. When we observed negative variance, we fixed variance for that parameter to zero and reran the model. We examined the effect of time since index trauma on change in PTCI score change in the best fitting model for each PTCI subscale.

We estimated all parameters using maximum likelihood estimation. We examined model fit using χ2, Bentler Comparative Fit Index (CFI), Tucker-Lewis Index (TLI), Root Mean Square Error of Approximation (RMSEA), and Standardized Root Mean Square Residual (SRMR). We interpreted fit statistics collectively using established criteria for close fit: χ2 p values > .05, CFI and TLI ≥ .95, RMSEA close to .06, and SRMR ≤ .08 (Bentler, 1990; Brown, 2006; Browne & Cudeck, 1992; Hu & Bentler, 1999; Kline, 2011). We compared model fit across configurations using review of fit indices. Given that many of the models were not nested, direct comparison was not possible using a chi-square difference test. Instead, we compared models as they fit the data relative to recommended ranges for CFI, TLI, RMSEA, and SRMR.

We reviewed the covariance coverage matrix prior to running the planned analyses to identify the amount of missing data. A small portion of the data was missing; the covariance coverage matrices ranged from .730–.984. We handled missing data using full information maximum likelihood estimation (FIML).

Results

Effects of Treatment on Trauma-Related Beliefs

Participants randomized to CPT and WET did not significantly differ in any demographic characteristic (see Sloan et al., 2018). Means, standard deviations, and effect size values by treatment condition, time point, and PTCI subscale are presented in Table 1. Small to moderate magnitude decreases in self-reported negative trauma-related beliefs among participants randomized to WET were observed, with the smallest observed changes in self-blame. For participants randomized to CPT, moderate to large magnitude decreases in self-reported negative trauma-related beliefs were observed, again with the smallest observed changes in self-blame.

Table 1.

FIML Estimated Means, Standard Deviations, and Effect Sizes for PTCI Subscales

Subscales Full sample WET CPT Between condition
d
M SD d M SD d M SD d
NCS
 Baseline 73.32 30.47 71.33 26.42 75.24 33.92
 6-week 69.83 30.59 −0.14 67.15 27.68 −0.18 72.03 33.21 −0.17 0.17
 12-week 63.82 29.72 −0.31 66.70 30.21 −0.17 60.57 28.38 −0.71 0.21
 24-week 56.81 28.10 −0.50 62.62 25.91 −0.34 49.98 28.23 −1.06 0.46
 36-week 56.39 31.18 −0.52 62.35 30.04 −0.34 50.90 32.11 −1.00 0.37
 60-week 53.59 30.28 −0.57 55.56 28.28 −0.55 50.22 31.82 −0.96 0.17
NCW
 Baseline 35.46 9.51 35.62 9.58 35.29 9.50
 6-week 33.95 9.35 −0.19 34.21 9.80 −0.18 33.57 8.99 −0.19 0.07
 12-week 31.93 11.24 −0.33 34.78 10.72 −0.09 28.82 11.02 −0.54 0.55
 24-week 29.96 11.72 −0.50 32.57 10.93 −0.33 27.12 12.02 −0.67 0.50
 36-week 29.20 12.80 −0.52 32.06 12.39 −0.33 26.17 12.77 −0.71 1.09
 60-week 28.72 12.17 −0.56 31.56 10.93 −0.43 25.55 13.14 −0.70 0.50
SB
 Baseline 18.68 9.31 17.04 8.14 20.29 10.09
 6-week 17.58 9.34 −0.13 15.76 8.26 −0.18 19.42 10.02 −0.09 0.41
 12-week 16.37 8.96 −0.24 16.81 8.43 −0.03 15.84 9.43 −0.42 0.11
 24-week 13.94 7.89 −0.48 14.81 7.62 −0.27 12.87 8.05 −0.69 0.24
 36-week 15.08 8.95 −0.34 15.58 8.12 −0.15 14.70 9.87 −0.48 0.09
 60-week 14.07 8.53 −0.42 13.58 8.11 −0.36 14.12 9.02 −0.50 0.06
Open in a new tab
Note.

FIML = Full Information Maximum Likelihood; PTCI = Posttraumatic Cognitions Inventory; WET = Written Exposure Therapy; CPT = Cognitive Processing Therapy; M = mean; SD = standard deviation; d = Cohen’s d; NCS = Negative Cognitions–Self; NCW = Negative Cognitions–World; SB = Self-Blame.

Growth curve model fit statistics are presented in Table 2. Quadratic models failed to converge for all three subscales. We constrained variance for the second slope in all models as the parameters were so small that estimating them caused convergence problems. Among the growth curve models for the NCS subscale, only the linear model provided poor fit to the data across all fit statistics; the others provided generally adequate fit. Of the remaining models, CFI and TLI values were within recommended ranges for all models. The SRMR values for the logarithmic and piecewise slope factor models fell within the recommended range. Collectively, the piecewise model generally provided the best fit to the data. This model indicated scores began at a mean level of 74.076, decreased from baseline to 24 weeks, and then decreased at a more gradual rate from 24 to 60 weeks. Parameter estimates for the best fitting growth curve model of each subscale are presented in Table 3.

Table 2.

Growth Curve Model Fit Statistics

Model X2 DF CFI TLI RMSEA [90% CI] SRMR
Negative Cognitions–Self
 Linear Model 120.234* 21 .798 .855 .194 [.161, .228] .136
 Slope Factor Model 33.260* 19 .971 .977 .077 [.029, .120] .082
 Logarithmic Model 24.357 12 .975 .968 .090 [.036, .142] .050
 Piecewise Model 29.160 * 18 .977 .981 .070 [.009, .115] .085
 Piecewise Slope Model 26.816* 15 .976 .976 .079 [.024, .127] .075
Negative Cognitions–World
 Linear Model 83.420* 21 .862 .902 .154 [.120, .189] .190
 Slope Factor Model 14.08 10 1.000 1.000 <.001 [<.001, .066] .046
 Logarithmic Model 17.255* 12 .988 .986 .059 [<.001, .116] .095
 Piecewise Model 13.045 17 1.000 1.000 <.001 [<.001, .060] .063
 Piecewise Slope Model 12.330 15 1.000 1.000 <.001 [<.001, .070] .049
Self-Blame
 Linear Model 135.567* 21 .744 .817 .208 [.175, .242] .158
 Slope Factor Model 44.171* 10 .939 .946 .113 [.072, .154] .099
 Logarithmic Model 25.936 12 .969 .961 .096 [.044, .147] .057
 Piecewise Model 25.479 13 .972 .968 .087 [.034, .137] .080
 Piecewise Slope Model 40.591 15 .943 .943 .116 [.074, .160] .097
Open in a new tab
Note.

Models in bold font represent the selected models; CFI = Bentler Comparative Fit Index; TLI = Tucker Lewis Index; RMSEA = Root Mean Square Error of Approximation; SRMR = Standardized Root Mean Square Residual.

*

p < .05.

Table 3.

Growth Curve Model Parameter Estimates

Model NCS NCW SB
Unstandardized
 Means
  Intercept 74.076* (2.683) 35.756* (0.815) 18.907* (0.823)
  Slope 1 −5.338* (1.023) −6.419* (1.088) −1.460* (0.301)
  Slope 2 −2.005* (0.921) — −0.102 (0.250)
 Variances
  Intercept 754.336* (114.824) 58.344* (10.295) 61.747* (11.066)
  Slope 1 80.811* (80.811) 84.457* (18.663) 5.678* (1.537)
  Slope 2 0+ — 0+
 Covariances
  Int. – Slp. 1 −116.484* (35.070) −7.665 (10.222) −8.843* (3.407)
  Int. – Slp. 2 2.656 (28.630) — −1.814* (2.348)
  Slp. 1 – Slp. 2 −4.343 (9.943) — −0.102 (0.250)
Standardized
 Means
  Slope 1 −.594* −.698* −.613*
  Slope 2 0+ — —
 Covariances
  Int. – Slp. 1 −.472* −.109 −.472*
  Int. – Slp. 2 N/A — N/A
  Slp. 1 – Slp. 2 N/A — N/A
Open in a new tab
Note.

Standard errors of the coefficients are presented in parentheses; Int. = intercept; slp = slope; + = parameter fixed to zero; NCS = Negative Cognitions–Self; NCW = Negative Cognitions–World; SB = Self-Blame.

*

p < .05.

As with the NCS subscale, only the linear model provided poor fit to the data across all fit statistics. The CFI, TLI, and RMSEA values for all other models fell within the recommended range. The SRMR values for all except the logarithmic model fell within the recommended range. Given limited distinction between models in terms of fit, we elected to use the slope factor model because it is more parsimonious than the piecewise models. This model indicated that NCW scores began at a mean level of 35.756 and decreased from baseline to 60 weeks. Given that the loadings fluctuate across time points in a slope factor model, interpreting slope is somewhat challenging; loadings indicated the period of greatest change was from baseline to 36 weeks.

Among the SB subscale, only the linear model provided poor fit to the data across all fit statistics; the others provided generally adequate fit. Of the remaining models, the CFI and TLI values fell within the recommended range. The RMSEA values for all of the models did not fall within the recommended range, with the piecewise being the closest to .06. The SRMR values for the logarithmic and piecewise models were within the recommended range. Although none of these models provided excellent fit to the data, we elected to use the piecewise model as it provided the best fit to the data and could be characterized as providing generally adequate fit. This model indicated scores began at a mean level of 18.907, decreased from baseline to 24 weeks, then did not change significantly from 24 to 60 weeks.

Effect of Time Since Index Trauma on Change in Trauma-Related Beliefs

Table 4 presents results of growth curve models examining the effect of time since index trauma on trauma-related belief change. We present fit statistics for the models examining time since trauma and treatment condition by time since trauma interaction in Table S2 in the online supplemental materials; all models provided adequate fit or better. We did not estimate the effects of time since index trauma on the second slope for NCS and SB since we fixed the variance for these parameters to zero. For the NCS subscale, in Model 1, greater time since trauma was associated with less reduction in trauma-related beliefs from baseline to 24 weeks. In Model 2, treatment condition, but not time since trauma or the time since trauma by treatment condition interaction, was predictive of change in trauma-related beliefs from baseline to 24 weeks. For the NCW subscale, in Model 1, greater time since trauma was associated with less reduction in trauma-related beliefs from baseline to 60 weeks. In Model 2, both treatment condition and the treatment condition by time since trauma interaction were significant. For the SB subscale, in Model 1, greater time since trauma was associated with less reduction in trauma-related beliefs from baseline to 24 weeks. In Model 2, treatment condition, but not time since trauma or the time since trauma by treatment condition interaction, was predictive of change in trauma-related beliefs from baseline to 24 weeks.

Table 4.

Effects of Time Since Index Trauma on Change in Trauma-Related Beliefs

Model/Predictor Intercept Slope 1
Est. SE β p Est. SE β p
NCS
 Model 1
  TST −0.014 0.020 −.07 .503 0.018 0.008 .24 .025
 Model 2
  TST 0.078 0.061 .38 .201 −0.023 0.024 −.30 .329
  Condition 22.485 11.499 .29 .051 −16.001 4.582 −.55 <.001
  TST × Cond. −0.062 0.040 −.48 .112 0.028 0.016 .57 .082
NCW
 Model 1
  TST −0.004 0.004 −.09 .399 0.014 0.005 .28 0.013
 Model 2
  TST 0.012 0.013 .29 .375 −0.018 0.016 −.36 .261
  Condition 1.855 2.456 .12 .450 −10.440 3.064 −.56 .001
  TST × Cond. −0.010 0.009 −.41 .221 0.021 0.010 .68 .045
SB
 Model 1
  TST −0.007 0.004 −.17 .113 0.003 0.002 .27 .028
 Model 2
  TST 0.009 0.013 .21 .508 −0.005 0.004 −.41 .249
  Condition 6.224* 2.455 .39 .011 −3.282 0.855 −.69 <.001
  TST × Cond. −0.010 0.009 −.38 .224 0.003 0.003 .70 .060
Open in a new tab
Note.

β = standardized parameter estimate; Est. = unstandardized parameter estimates; NCS = Negative Cognitions–Self; NCW = Negative Cognitions–World; SB = Self-Blame; SE = standard error; TST = time since trauma; TST × Cond. = time since trauma by treatment condition interaction.

*

p < .05.

We present figures of observed interaction effects in Figure 1 in which participants are portioned by treatment condition and one standard deviation above or below average in time since index trauma. Within CPT, the difference between individuals with more recent versus more distant index traumas in changes in NCW appeared quite pronounced. Although visually this appeared consistent across subscales, these interaction effects were not significant for NCS or SB subscales.

Figure 1. Trajectories of Negative Trauma-Related Beliefs (By Condition × Time Since Trauma) Over Time.

Figure 1

Open in a new tab

Note. PTCI = Posttraumatic Cognitions Inventory; WET = Written Exposure Therapy; CPT = Cognitive Processing Therapy; NCS = Negative Cognitions–Self; NCW = Negative Cognitions–World; SB = Self-Blame.

Discussion

In this study, consistent with previous research, both trauma-focused treatments facilitated significant reductions in trauma-related beliefs. Given that CPT directly targets trauma-related beliefs, it is not surprising that observed decreases were greater in the CPT condition compared with WET, an exposure-based treatment. We observed greater change in negative trauma-related beliefs during the periods roughly corresponding to treatment for CPT (i.e., pretreatment through 24 weeks after the first treatment session) relative to follow-up (from 24 to 60 weeks after the first treatment session). Of note, we only observed continued significant decreases in negative trauma-related beliefs related to the self between 24 and 60 weeks. Change in self-blame during the 24–60 week period was not significant, indicating possible maintenance of treatment gains for these trauma-related beliefs.

The impact of time since index trauma on the effect of interventions on negative trauma-related belief change varied by treatment condition and belief domain (i.e., PTCI subscale). Time since index trauma had a greater impact of the effect of treatment on trauma-related belief improvement among participants randomized to CPT. This finding suggests that a larger window of time between index trauma and treatment may negatively impact the malleability of negative trauma-related beliefs about the world during CPT. Significant effects of time since index trauma on all belief domains among the full sample indicate treatment should be provided in closer proximity to trauma exposure before negative trauma-related beliefs become inflexible and, in the case of CPT, particularly for beliefs about the world.

Two other observed patterns of change warrant discussion. First, significant change in all belief domains was limited to the 24-week period following baseline. Unlike PTSD symptoms that often continue to decline following treatment completion in RCTs (Kline et al., 2018), which are often marked by high treatment fidelity, change in trauma-related beliefs appeared to plateau. Previous longitudinal research investigating negative-trauma related belief change has indicated a similar trajectory following the completion of treatment (Scher et al., 2017).

Second, changes in self-blame were the smallest in magnitude in both treatment conditions. Our findings are in line with other studies, that have found that the PTCI self-blame performed differently from the other PTCI subscales (e.g., Beck et al., 2004; Startup et al., 2007). Thoughts related to self-blame may be more persistent, more ingrained, and/or slower to change over time. Several factors may have contributed to the patterns observed in the PTCI self-blame scale over time. For some individuals, there could be a benefit to retaining thoughts related to self-blame, and self-blame may not be seen by all as a “negative” cognition. For example, blaming oneself for the cause of an event may allow the individual to maintain some semblance of control or belief in the “controllability” of the event, which may feel empowering. From a psychometric standpoint, the self-blame subscale includes fewer items in comparison to the longer PTCI subscales, and the subscale—and its associated smaller range of subscale scores—may affect the ability to detect change.

Although changes in negative trauma-related beliefs were expected for CPT, it is encouraging that these changes were also observed in the brief, exposure-based treatment of WET. Additionally, time since index trauma did not significantly affect changes in negative trauma-related beliefs among individuals receiving WET. The fact that the negative trauma-related beliefs did not change as much in WET as for CPT (despite overall symptom improvement; Sloan et al., 2018) suggests that substantial belief change may not be necessary for overall PTSD symptom change to occur. Our results suggest repeated exposure to the traumatic memory through writing may allow individuals to sufficiently examine and challenge inaccurate trauma-related beliefs (e.g., “I could have done something to have stopped the event from taking place.”)

There are several study limitations that should be considered. Broadly, participant self-report of trauma-related beliefs is challenging by definition, as individuals with PTSD may have limited insight into distorted thinking patterns. Thus, the use of a self-report measure to assess negative trauma-related beliefs may be limited. Second, recent findings suggest the PTCI factor structure may not be as stable as initially assumed (e.g., Whiteman et al., 2022); this study relied on the assumption that the PTCI measures three distinct belief domains. Future research is needed to confirm this assumption or find a more robust solution. Despite these potential limitations, prior research in the area has also used the PTCI to investigate trauma-related beliefs, including changes during treatment. Third, we used “time since index trauma” as a proxy for the “origin” of negative trauma-related beliefs. We acknowledge, however, that the self-report nature may not be an accurate estimate of time since the onset of trauma-related beliefs. Further, individuals who have experienced multiple traumatic events may have developed negative trauma-related beliefs earlier or later than the identified index event. Lastly, change in negative trauma-related beliefs was characterized by significant heterogeneity, but our sample was not adequately powered to examine clinical characteristics that may provide additional nuance to our findings. Factors such as trauma type and the number of Criterion A events may impact the degree that to which time affects entrenchment and change in negative trauma-related beliefs.

Future studies should consider novel and nuanced measures of belief onset and belief change to examine possible granular changes in cognition. For example, researchers might consider exploring the timing of the first traumatic event to measure possible onset of beliefs instead of the timing of the index event, which may not have been the first traumatic event experienced by a participant. Behavioral measures or ecological momentary assessment may provide alternatives to retrospective self-report measures. Assessment with multiple measures, including self-report instruments besides the PTCI that measure cognitive change, is warranted. Finally, given the quest within precision medicine to determine which treatments work best for whom, future research should consider which factors are likely to contribute to the greatest treatment gains. Relative to this goal and the current study, future research may consider whether time since index trauma is one characteristic that may guide intervention selection.

Supplementary Material

supplemental1

Clinical Impact Statement.

This study examined the role of time since index trauma on the effects of trauma-focused therapies on negative trauma-related beliefs. Time since trauma impacted the effect of treatment on negative trauma-related beliefs in cognitive processing therapy (CPT). Results from this study indicate that providing treatment earlier following trauma exposure may optimize the impact of CPT on negative trauma-related beliefs, particularly negative cognitions about to the world.

Acknowledgments

This research was supported by grants from the National Institute of Mental Health (R0MH095737). Lillian Reuman was supported by the National Institute of Mental Health Award T32MH019836.

Footnotes

The data reported in this article have been previously published and were collected as part of a larger data collection. Findings from the data collection have been reported in separate manuscripts. Sloan et al. (2018) focuses on PTSD symptom reduction and Tx dropout rate; while Thompson-Hollands et al. (2021) focuses on longitudinal changes in Peritraumatic disassociation over time following PTSD treatment. Alpert et al. (2020) investigates predictors of dropout in cognitive processing therapy for PTSD; whereas Lee et al. (2021) examines if extinction or changes negative trauma-related beliefs occur either prior to or concurrently with changes in posttraumatic stress symptoms among for participants who received WET or CPT. Marx et al. (2021) examines the degree to which estimated intelligence moderates Cognitive Processing Therapy Outcome. Finally, Thompson-Hollands et al. (2018) investigates longitudinal treatment gains from a brief exposure-based treatment and the effect of treatment on depressive symptoms.

References

  1. Alpert E, Hayes AM, Barnes JB, & Sloan DM (2020). Predictors of dropout in cognitive processing therapy for PTSD: An examination of trauma narrative content. Behavior Therapy, 51(5), 774–788. 10.1016/j.beth.2019.11.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Barawi KS, Lewis C, Simon N, & Bisson JI (2020). A systematic review of factors associated with outcome of psychological treatments for post-traumatic stress disorder. European Journal of Psychotraumatology, 11(1), 1774240. 10.1080/20008198.2020.1774240 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Beck JG, Coffey SF, Palyo SA, Gudmundsdottir B, Miller LM, & Colder CR (2004). Psychometric properties of the Posttraumatic Cognitions Inventory (PTCI): A replication with motor vehicle accident survivors. Psychological Assessment, 16(3), 289–298. 10.1037/1040-3590.16.3.289 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bentler PM (1990). Comparative fit indexes in structural models. Psychological Bulletin, 107(2), 238–246. 10.1037/0033-2909.107.2.238 [DOI] [PubMed] [Google Scholar]
  5. Brown TA (2006). Confirmatory factor analysis for applied research. Guilford Press. [Google Scholar]
  6. Brown LA, Belli GM, Asnaani A, & Foa EB (2019). A review of the role of negative cognitions about oneself, others, and the world in the treatment of PTSD. Cognitive Therapy and Research, 43(1), 143–173. 10.1007/s10608-018-9938-1 [DOI] [Google Scholar]
  7. Browne MW, & Cudeck R (1992). Alternative ways of assessing model fit. Sociological Methods & Research, 21(2), 230–258. 10.1177/0049124192021002005 [DOI] [Google Scholar]
  8. Cohen J (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Erlbaum. [Google Scholar]
  9. Dekel S, Peleg T, & Solomon Z (2013). The relationship of PTSD to negative cognitions: A 17-year longitudinal study. Psychiatry, 76(3), 241–255. 10.1521/psyc.2013.76.3.241 [DOI] [PubMed] [Google Scholar]
  10. Dunmore E, Clark DM, & Ehlers A (2001). A prospective investigation of the role of cognitive factors in persistent posttraumatic stress disorder (PTSD) after physical or sexual assault. Behaviour Research and Therapy, 39(9), 1063–1084. 10.1016/S0005-7967(00)00088-7 [DOI] [PubMed] [Google Scholar]
  11. Ehlers A, & Clark DM (2000). A cognitive model of posttraumatic stress disorder. Behaviour Research and Therapy, 38(4), 319–345. 10.1016/S0005-7967(99)00123-0 [DOI] [PubMed] [Google Scholar]
  12. 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(4), 413–431. 10.1016/j.brat.2004.03.006 [DOI] [PubMed] [Google Scholar]
  13. Ehlers A, Clark DM, Hackmann A, McManus F, Fennell M, Herbert C, & Mayou R (2003). A randomized controlled trial of cognitive therapy, a self-help booklet, and repeated assessments as early interventions for posttraumatic stress disorder. Archives of General Psychiatry, 60(10), 1024–1032. 10.1001/archpsyc.60.10.1024 [DOI] [PubMed] [Google Scholar]
  14. Foa EB, & Kozak MJ (1986). Emotional processing of fear: Exposure to corrective information. Psychological Bulletin, 99(1), 20–35. 10.1037/0033-2909.99.1.20 [DOI] [PubMed] [Google Scholar]
  15. Foa EB, & Rothbaum BO (1998). Treating the trauma of rape: Cognitive-behavioral therapy for PTSD. Guilford Press. [Google Scholar]
  16. Foa EB, Ehlers A, Clark DM, Tolin DF, & Orsillo SM (1999). The posttraumatic cognitions inventory (PTCI): Development and validation. Psychological Assessment, 11(3), 303–314. 10.1037/1040-3590.11.3.303 [DOI] [Google Scholar]
  17. Holliday R, Holder N, & Surís A (2018). A single-arm meta-analysis of cognitive processing therapy in addressing trauma-related negative cognitions. Journal of Aggression, Maltreatment & Trauma, 27(10), 1145–1153. 10.1080/10926771.2018.1429511 [DOI] [Google Scholar]
  18. Holliday R, Link-Malcolm J, Morris EE, & Surís A (2014). Effects of cognitive processing therapy on PTSD-related negative cognitions in veterans with military sexual trauma. Military Medicine, 179(10), 1077–1082. 10.7205/MILMED-D-13-00309 [DOI] [PubMed] [Google Scholar]
  19. Hu L, & Bentler PM (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6(1), 1–55. 10.1080/10705519909540118 [DOI] [Google Scholar]
  20. Iverson KM, King MW, Cunningham KC, & Resick PA (2015). Rape survivors’ trauma-related beliefs before and after cognitive processing therapy: Associations with PTSD and depression symptoms. Behaviour Research and Therapy, 66, 49–55. 10.1016/j.brat.2015.01.002 [DOI] [PubMed] [Google Scholar]
  21. Karatzias A, Power K, McGoldrick T, Brown K, Buchanan R, Sharp D, & Swanson V (2007). Predicting treatment outcome on three measures for post-traumatic stress disorder. European Archives of Psychiatry and Clinical Neuroscience, 257(1), 40–46. 10.1007/s00406-006-0682-2 [DOI] [PubMed] [Google Scholar]
  22. Kassin S (2003). Essentials of psychology. Prentice Hall. [Google Scholar]
  23. Kline AC, Cooper AA, Rytwinksi NK, & Feeny NC (2018). Long-term efficacy of psychotherapy for posttraumatic stress disorder: A meta-analysis of randomized controlled trials. Clinical Psychology Review, 59, 30–40. 10.1016/j.cpr.2017.10.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Kline RB (2011). Principles and practice of structural equation modeling (3rd ed.). Guilford Press. [Google Scholar]
  25. Kubany ES, Haynes SN, Leisen MB, Owens JA, Kaplan AS, Watson SB, & Burns K (2000). Development and preliminary validation of a brief broad-spectrum measure of trauma exposure: The Traumatic Life Events Questionnaire. Psychological Assessment, 12(2), 210–224. 10.1037/1040-3590.12.2.210 [DOI] [PubMed] [Google Scholar]
  26. Kumpula MJ, Pentel KZ, Foa EB, LeBlanc NJ, Bui E, McSweeney LB, Knowles K, Bosley H, Simon NM, & Rauch SAM (2017). Temporal sequencing of change in posttraumatic cognitions and PTSD symptom reduction during prolonged exposure therapy. Behavior Therapy, 48(2), 156–165. 10.1016/j.beth.2016.02.008 [DOI] [PubMed] [Google Scholar]
  27. Lee DJ, Marx BP, Thompson-Hollands J, Gallagher MW, Resick PA, & Sloan DM (2021). The temporal sequence of change in PTSD symptoms and hypothesized mediators in Cognitive Processing Therapy and Written Exposure Therapy for PTSD. Behaviour Research and Therapy, 144, 103918. 10.1016/j.brat.2021.103918 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Lewis CE, Farewell D, Groves V, Kitchiner NJ, Roberts NP, Vick T, & Bisson JI (2017). Internet-based guided self-help for post-traumatic stress disorder (PTSD): Randomized controlled trial. Depression and Anxiety, 34(6), 555–565. 10.1002/da.22645 [DOI] [PubMed] [Google Scholar]
  29. LoSavio ST, Worley CB, Aajmain ST, Rosen CS, Wiltsey Stirman S, & Sloan DM (2021). Effectiveness of written exposure therapy for posttraumatic stress disorder in the Department of Veterans Affairs Healthcare System. Psychological Trauma: Theory, Research, Practice and Policy. Advance online publication. 10.1037/tra0001148 [DOI] [PubMed] [Google Scholar]
  30. Marx BP, Thompson-Hollands J, Lee DJ, Resick PA, & Sloan DM (2021). Estimated intelligence moderates Cognitive Processing Therapy outcome for posttraumatic stress symptoms. Behavior Therapy, 52(1), 162–169. 10.1016/j.beth.2020.03.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Morris SB, & DeShon RP (2002). Combining effect size estimates in meta-analysis with repeated measures and independent-groups designs. Psychological Methods, 7(1), 105–125. 10.1037/1082-989X.7.1.105 [DOI] [PubMed] [Google Scholar]
  32. Muthén LK, & Muthén BO (1998–2017). Mplus Version 7 user’s guide. [Google Scholar]
  33. O’Donnell ML, Elliott P, Wolfgang BJ, & Creamer M (2007). Post-traumatic appraisals in the development and persistence of posttraumatic stress symptoms. Journal of Traumatic Stress, 20(2), 173–182. 10.1002/jts.20198 [DOI] [PubMed] [Google Scholar]
  34. Resick PA, Monson CM, & Chard KM (2016). Cognitive processing therapy for PTSD: A comprehensive manual. Guilford Press. [Google Scholar]
  35. Roberge EM, Weinstein HR, & Bryan CJ (2019). Predicting response to cognitive processing therapy: Does trauma history matter? Psychological Trauma: Theory, Research, Practice and Policy, Advance online publication. 10.1037/tra0000530 [DOI] [PubMed] [Google Scholar]
  36. Scher CD, Suvak MK, & Resick PA (2017). Trauma cognitions are related to symptoms up to 10 years after cognitive behavioral treatment for posttraumatic stress disorder. Psychological Trauma: Theory, Research, Practice and Policy, 9(6), 750–757. 10.1037/tra0000258 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Sloan DM, & Marx BP (2019). Written exposure therapy for PTSD: A brief treatment approach for mental health professionals. American Psychological Association. [Google Scholar]
  38. Sloan DM, Lee DJ, Litwack SD, Sawyer AT, & Marx BP (2013). Written exposure therapy for veterans diagnosed with PTSD: A pilot study. Journal of Traumatic Stress, 26(6), 776–779. 10.1002/jts.21858 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Sloan DM, Marx BP, Lee DJ, & Resick PA (2018). A brief exposure-based treatment vs 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]
  40. Sobel AA, Resick PA, & Rabalais AE (2009). The effect of cognitive processing therapy on cognitions: Impact statement coding. Journal of Traumatic Stress, 22(3), 205–211. 10.1002/jts.20408 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Startup M, Makgekgenene L, & Webster R (2007). The role of self-blame for trauma as assessed by the Posttraumatic Cognitions Inventory (PTCI): A self-protective cognition? Behaviour Research and Therapy, 45(2), 395–403. 10.1016/j.brat.2006.02.003 [DOI] [PubMed] [Google Scholar]
  42. 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(10), 985–991. 10.1002/da.22825 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Thompson-Hollands J, Marx BP, Lee DJ, & Sloan DM (2021). Longitudinal change in self-reported peritraumatic dissociation during and after a course of PTSD treatment: Contributions of symptom severity and time. Psychological Trauma: Theory, Research, Practice and Policy, 13(6), 665–672. 10.1037/tra0000951 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Watkins ER (2008). Constructive and unconstructive repetitive thought. Psychological Bulletin, 134(2), 163–206. 10.1037/0033-2909.134.2.163 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Weathers FW, Bovin MJ, Lee DJ, Sloan DM, Schnurr PP, Kaloupek DG, Keane TM, & Marx BP (2018). The Clinician-Administered PTSD Scale for DSM-5 (CAPS-5): Development and initial psychometric evaluation in military veterans. Psychological Assessment, 30(3), 383–395. 10.1037/pas0000486 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Whiteman SE, Kramer LB, Silverstein MW, Witte TK, & Weathers FW (2022). Evaluating the factor structure of the Posttraumatic Cognitions Inventory. Assessment, 29(2), 128–135. 10.1177/1073191120954915 [DOI] [PubMed] [Google Scholar]
  47. Zalta AK, Gillihan SJ, Fisher AJ, Mintz J, McLean CP, Yehuda R, & Foa EB (2014). Change in negative cognitions associated with PTSD predicts symptom reduction in prolonged exposure. Journal of Consulting and Clinical Psychology, 82(1), 171–175. 10.1037/a0034735 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

supplemental1
NIHMS1866363-supplement-supplemental1.docx (22.7KB, docx)

RESOURCES