Authors: Anna C. Barbano, Danielle M. Piggott, Ian M. Raugh, Jessica R. Ellem, Minden B. Sexton, RaeAnn E. Anderson
Categories: Article, tonic immobility, freezing response, rape, posttraumatic stress disorder, dissociation
Source: Journal of traumatic stress
Doi: 10.1002/jts.23182
Authors: Anna C. Barbano, Danielle M. Piggott, Ian M. Raugh, Jessica R. Ellem, Minden B. Sexton, RaeAnn E. Anderson
Posttraumatic stress disorder (PTSD) is a common psychopathological outcome of sexual assault. Peritraumatic tonic immobility (TI; involuntary “freezing”) during a sexual assault has been linked with worsened PTSD symptoms, yet little research has examined these effects longitudinally. The present study aimed to examine the effects of TI severity on PTSD symptom, negative posttraumatic cognition (NPC), and dissociative symptom severity over the course of one year among 112 female survivors of sexual assault aged 16–30. Multilevel linear regressions evaluated the effects of time since sexual assault, TI severity, and their interaction on PTSD symptom, NPC (total, blame, self, world), and dissociation severity. Results indicated that TI severity was not associated with PTSD symptom severity (β = 0.25, p = 0.116) but was associated with increased NPC (β = 0.22, p = 0.018) and dissociative symptom severity (β = 0.65, p = 0.037). These relationships remained consistent over time. Given the salience of NPCs, individuals seeking care for TI-associated sexual trauma may benefit from interventions with demonstrated ability to reduce NPCs (e.g., Prolonged Exposure, Cognitive Processing Therapy). Further evaluating these treatments’ utility in reducing dissociation may be of particular benefit to those with histories of TI reactions during trauma. Finally, coercive rape tactics were associated with increased PTSD symptoms (β = 0.48, p = 0.002), whereas forcible (β = −0.04, p = 0.805) and drug- or alcohol-facilitated rape tactics (β = 0.35, p = 0.285) were not, indicating the importance of assessing verbal coercion in research and clinical work with individuals who have experienced sexual assault.
Sexual violence is endemic in the United States, and women are at particularly high risk for exposure to sexual assault (Smith et al., 2018; Tjaden & Thoennes, 2006). The negative impact of sexual assault on mental health outcomes is well-established (Campbell et al., 2009), including increased risk for developing posttraumatic stress disorder (PTSD; Dworkin et al., 2017; Dworkin, 2020).
Understanding factors that contribute to the development and maintenance of PTSD symptoms following sexual assault is crucial for prevention and intervention efforts. Numerous pre-existing, peritraumatic, and posttraumatic factors influence the development and course of PTSD symptoms following sexual assault. Peritraumatic risk factors include those occurring during or shortly after the traumatic event, such as rape tactics or assault characteristics. Among sexual assault survivors, PTSD severity appears to vary based on assaultive tactics used, including forcible rape (FR; e.g., use of harm or a weapon), drug- or alcohol-facilitated rape (DAFR; i.e., victim is incapacitated due to effects of drugs or alcohol), and coercive rape (CR; e.g., use of manipulation, lies, threats, or pressure; Zinzow, Resnick, McCauley et al., 2010; Zinzow, Resnick, Amstadter et al., 2010). FR, with or without incapacitation due to substances, has been associated with worse PTSD symptoms relative to other rape tactics (McConnell et al., 2020; Zinzow, Resnick, McCauley et al., 2010). However, some studies show verbal anger or criticism expressed by an aggressor (consistent with CR type) is associated with higher PTSD symptoms regardless of whether force was used (Kern et al., 2024). CR is also associated with lower self-esteem relative to FR (Zweig et al., 1997).
Another peritraumatic risk factor that has gained increasing research attention is the peritraumatic occurrence of tonic immobility (TI), which can occur when a danger overwhelms an individual’s ability to effectively fight or escape (Gallup & Rager, 1996). Although theories of threat response vary in their conceptualization of TI etiology (for a review, see Beutler et al., 2022), most agreed upon definitions of TI include elements of both physical immobility (e.g., paralysis, numbness, inability to vocalize) and mental numbing without loss of consciousness (e.g., dissociation, mind going blank; TeBockhorst et al., 2014; de la Torre Laso, 2023). TI is an involuntary response to danger and is differentiated from a voluntary “freezing” response, which may be done purposefully to avoid or minimize physical harm (Gallup, 1977). Nonetheless, TI – which can appear as “playing dead” – is considered evolutionarily advantageous for survival during attack, as predators are less likely to attack prey that is already perceived as dead (Bovin & Marx, 2011). TI is a particularly common response to sexual assault compared to other trauma types (Kalaf et al., 2017). To date, there is little research examining the prevalence of TI experiences in relation to rape tactics used. Although TI can occur during sexual assault regardless of rape tactics used by an aggressor, TI may be more common during DAFR given victims’ already impaired ability to move or react due to intoxication.
Extant research suggests that TI is associated with worsened mental health outcomes post-assault (e.g., Coimbra et al., 2023; Lima et al., 2010; Möller et al., 2017; Rizvi et al., 2008). Although TI may be adaptive in preventing further damage or risk, many adult sexual assault survivors experience increased shame and self-blame due to societal and/or self-imposed expectations that they should have been able to defend themselves against the attack (Galliano et al., 1993; TeBockhorst et al., 2015). Distorted posttraumatic cognitions such as those related to self-blame and one’s own ability to respond to perceived danger may be particularly toxic in PTSD development and maintenance (LoSavio et al., 2017) as well as powerful intervention targets (Brown et al., 2019; Schumm et al., 2015). Moreover, TI may also be a risk factor for posttraumatic dissociation. Conceptually, the occurrence of TI, which includes dissociative-type experiences (Abrams et al., 2009), may result in disturbances in the formation of concrete, accessible trauma memories (e.g., autobiographical details) versus more contextually-based, involuntary memories (e.g., intrusions), consistent with the Dual Representation Theory of PTSD (Brewin et al., 2007). Alterations in consciousness, such as those often seen during TI responses, are associated with dissociative symptoms or experiences; yet, the effect of peritraumatic TI on persistent posttraumatic dissociation is not yet known.
Overall, there is scant research investigating these experiences longitudinally, particularly with the use of clinical interview data in addition to self-report (Coimbra et al., 2023). However, a few existing studies have demonstrated associations between TI and PTSD symptoms over time. For example, de Souza Junior and colleagues (2024) found that among health care professionals exposed to Covid-19-related traumatic events, TI severity was positively associated with PTSD symptom severity and probable PTSD diagnosis at 6- and 12-months post-baseline. Similarly, a study by Bøgelund Dokkedahl and Lahav (2023) found that among Israeli civilians exposed to rocket shelling, TI and peritraumatic dissociation severity fully mediated the relationship between perceived threat and PTSD re-experiencing, avoidance, and negative alteration in mood and cognition symptoms at 1- to 2-months post-baseline.
The present study examined the associations of TI severity at baseline interview with PTSD symptoms, posttraumatic cognitions, and dissociative symptoms assessed each at baseline, 6-months, and 12-months among young adult women exposed to sexual assault within the past seven years, extending upon previous research through the use of interview data, consideration of rape tactics, and examination of broad trauma-related psychopathology (i.e., posttraumatic cognitions and dissociation). Given the literature suggesting deleterious effects of TI on posttraumatic psychopathology, it was hypothesized higher TI scores would be associated with higher PTSD symptoms, higher posttraumatic cognitions (particularly those related to blame and the self), and higher dissociative symptoms at 6- and 12-months post-baseline interview. These hypotheses were based on documented cross-sectional associations of TI with dissociation and with increased experiences with shame and self-blame. Given their relevance to TI and later PTSD symptoms, rape tactics were explored as covariates.
Participants were recruited as part of a larger, longitudinal study of sexual violence experiences via social media advertisements, flyers, and through the psychology department subject pool of a large Midwestern university. Eligibility criteria included female sex, age 16–35, and experience of rape within the last seven years. Rape was defined by behavioral descriptions measured via three questions adapted from the Sexual Experiences Survey (SES; Koss et al., 2007). All participants endorsed having experienced FR or DAFR at least once in the past seven years on this measure. Other questions from the screener included those pertaining to age, gender, and timing of the experience. One hundred and twelve women between ages 16 and 35 who completed all necessary components for the present study were included in analyses; complete demographic characteristics are presented in Table 1.
There were no significant differences on outcome variables (i.e., PTSD symptoms, maladaptive posttraumatic cognitions, or dissociative symptoms), other clinical variables (e.g., depressive symptoms), or demographic variables (ps > .05), with the exception of age, between study completers and non-completers. Study completers were older at the time of the study and at the time of the index sexual assault (see below).
This study’s research protocol was approved by the Institutional Review Board of the local university at which it was conducted. The study was advertised with multiple titles to recruit women with varying experiences of sexual assault regardless of how they perceived or labeled the sexual assault (e.g., “a miscommunication” versus “rape”). Study advertisements led to a screening survey.
Eligible participants met with a graduate student clinician who confirmed eligibility, obtained informed consent, gathered a brief description of their sexual violence experience, and administered self-report questionnaires related to their reactions to the experience, general health, and treatment-seeking behaviors, and conducted a diagnostic assessment using the Diagnostic Interview for Anxiety, Mood, and OCD and Related Neuropsychiatric Disorders (DIAMOND; Tolin et al., 2018). The longitudinal elements of the study were structured so that participants completed initial questionnaires with a graduate student clinician, but all follow-up components were delivered via self-report surveys emailed to participants every six months following their initial interview date. The present study includes data from initial baseline interviews and the first and second follow-up surveys, representing 12 months of follow-up. Follow-up surveys included several questionnaires from the initial interview to track symptom presentation changes. Of the included measures, measures of posttraumatic stress symptoms were given only to participants meeting PTSD criteria at baseline (n = 30; due to a software error) and to all participants at follow-up assessments. TI and assault characteristics were measured at the initial baseline interview only, whereas changes in dissociative experiences, PTSD symptoms, and posttraumatic cognitions were measured across each follow-up.
The Diagnostic Interview for Anxiety, Mood, and OCD and Related Neuropsychiatric Disorders (DIAMOND) is a semi-structured, clinician-administered diagnostic interview for assessing psychopathology (Tolin et al., 2018). Interrater reliability for the DIAMOND ranges from very good to excellent across disorders, and test-retest reliability ranges from good to excellent (Tolin et al., 2018). On the PTSD module of the DIAMOND, interviewers directed participants to identify the most upsetting specific sexual assault experience to keep in mind while completing the inventory (i.e., the index event).
The Assault Characteristics Questionnaire (ACQ; Littleton et al., 2009) is a 21-item self-report inventory assessing the characteristics and circumstances of a specified sexual assault experience, such as the tactics used by the aggressor (e.g., “verbally threatening your relationship,” “holding you down,” “choking or beating you”), self-defense behaviors, and relationship to the aggressor. Interviewers once again directed participants to the index event while completing the inventory using the following “If you have had more than one such experience, please complete the following questions regarding what you would consider to be your worst or most memorable experience with unwanted sex.” These data were used to identify violence type (i.e., coercive rape [CR], forcible rape [FR], and drug- or alcohol-facilitated rape [DAFR]), which was used as a control variable throughout analyses. Of note, violence types were not mutually exclusive; thus, participants may have endorsed more than one type of rape tactic experienced during the index assault.
Tonic immobility was measured by the Tonic Immobility Scale (TIS; Forsyth et al., 2000). The 10-item self-report measure prompts participants to recall a sexual assault experience and rate the degree to which they experienced specific sensations or feelings. Items were rated on a 7-point Likert scale ranging from having not experienced it at all to having experienced it to a high degree. Items included sensations and experiences including felt cold, felt numb or no pain, and felt detached from yourself. Three items are reverse scored to minimize biased responses. Higher scores indicate greater endorsement of immobilized responses. TIS scores have demonstrated adequate test-retest reliability in samples of women sexual assault survivors (Fuse et al., 2006); α = 0.85 in this study. This study used items 1–2, 4–6, and 8–9 to calculate TI severity, consistent with the two-factor model proposed by Fusé and colleagues (2007). Participants in this study were instructed to anchor answers to “your personal experience during an episode of unwanted sex, like those on the questionnaire you just completed.”
The PTSD Checklist for DSM-5 (PCL-5; Weathers et al., 2013) was used to investigate change in PTSD symptoms over time. The PCL-5 is a 20-item self-report measure that assesses the corresponding 20 PTSD symptoms as outlined in the DSM-5. Respondents indicate the frequency with which they have experienced that symptom over the last month. Items are rated on a 5-point Likert scale ranging from not at all to extremely. Ratings are summed for a total score, with higher scores indicating greater symptom endorsement and severity. Total scores for this measure have demonstrated adequate internal consistency and test-retest reliability in previous samples (Blevins et al., 2015); α = 0.94 in this study. At all time points, participants were asked to anchor their responses to the index assault, unless a new, more upsetting assault had occurred during the follow-up period. Only 14 participants at the 6-month follow-up and 26 participants at the 12-month follow up reported instances of revictimization (not necessarily more distressing than the index assault), which was considered an insufficient frequency for separate analyses.
The Dissociative Experiences Measure, Oxford (DEMO; Cernis et al., 2018) assessed dissociative symptoms. The DEMO includes 30 items and asks participants to rate the frequency of specified dissociative experiences over the last month on a 5-point scale anchored by not at all and most of the time. The scale produces five subscales including unreality (e.g., I feel like I am in a parallel world), numb and disconnected (e.g., I feel like I’m ‘just existing’), memory blanks (e.g., I find myself somewhere and do not remember how I got there), zoned out (e.g., I stare aimlessly without thinking about anything), and vivid internal world (e.g., Unwanted images from my past come into my head). Scores are totaled, and higher values are associated with higher severity of dissociative experiences. DEMO scores have demonstrated adequate reliability and convergent validity with other measures of dissociation (Cernis et al., 2018); α = 0.96 in this study.
Changes in thinking following sexual assault experiences were assessed by the Posttraumatic Cognitions Inventory (PCTI; Foa et al., 1999). The 33-item self-report measure describes various thought patterns to which participants rank their level of personal agreement on a 7-point Likert scale anchored by totally disagree and totally agree. In addition to a total scale, the measure produces three self-blame (e.g., The event happened because of the way I acted), negative cognitions about the self (e.g., I am a weak person), and negative cognitions about the world (e.g., People can’t be trusted). Higher scores represent greater endorsement of negative trauma-related beliefs. The measure correlates with PTSD symptom severity and total scores have shown adequate internal consistency and reliability (Foa et al., 1999, Sexton et al., 2018); α = 0.96 in this study.
Hypotheses were evaluated in RStudio using a series of multilevel linear regressions evaluating the fixed effects of time (T1, T2, T3), TIS scores, and the interaction thereof on PCL-5, PTCI, and DEMO scores; observations were nested within participant (random slopes). Each model included time since the assault and assault type (FR, CR, and DAFR) as covariates. Restricted estimate maximum likelihood was used to account for missing data and Benjamini-Hochberg correction for multiple comparisons within each model.
Secondary analyses examined the impact of the same predictors on PTSD symptom clusters. Sensitivity analyses controlling for age at the time of the study and number of prior sexual victimization experiences were conducted and did not significantly improve model fit (see Supplementary Material).
Demographic variables of gender, age, sexual orientation, race, age of assault, and years since assault were not significantly associated with TI severity (i.e., TIS), PTSD symptom severity (i.e., PCL-5), posttraumatic cognitions (i.e., PTCI), or dissociative symptom severity (i.e., DEMO; all ps > .05). Because fewer than 10 participants with complete PCL-5 data and DEMO data were Latina, ethnicity was not included in analyses. See Table 2 for observations of key outcomes at each time; notably, only 31 and 34 participants had complete PCL-5 and DEMO data (respectively) at Time 1.
Full results of regression analyses are presented in Table 3. The model of PTSD symptoms accounted for 60% of the variance (R^2^), 20% of which was due to the fixed effects (time, tonic immobility, time since assault, assault characteristics, and interactions). Contrary to hypotheses regarding TI and PTSD symptoms, greater TIS scores were not significantly associated with greater PCL-5 scores (β = 0.25, p = .116). However, CR was significantly associated with greater posttraumatic stress symptoms (β = 0.48, p = .002). Results by PCL-5 subscale are presented in Table 4.
Regarding the relationship between TI and maladaptive posttraumatic cognitions, hypotheses were again supported in a model that accounted for 62% of the variance with 10% due to fixed effects. Greater tonic immobility scores were significantly associated with higher maladaptive posttraumatic cognitions (β = 0.22, p = .018). This association was consistent across time, indicated by nonsignificant interactions with timepoint. DAFR was associated with higher overall maladaptive posttraumatic cognitions (β = 0.77, p = .029). Results of regression analyses related to PTCI subscales are presented in Table 4.
Regarding the relationship between TI and dissociative experiences, hypotheses were supported in a model that accounted for 77% of the variance with 7% from fixed effects. Greater TI scores were significantly associated with greater dissociative experiences (β = 0.28, p = .037). This association was consistent across time, indicated by nonsignificant interactions with timepoint.
Secondary analyses by PCL-5 subscale showed TI severity significantly predicted re-experiencing symptoms (Table 4). Sensitivity analyses did not significantly differ from primary analyses; full details are in Supplementary Materials.
The majority of results did not survive correction for multiple comparisons; see Tables 3–4 for adjusted values and which effects survived correction.
The present study examined the associations of TI severity with PTSD symptoms, posttraumatic cognitions, and dissociative symptoms concurrently and at 6- and 12-months post-baseline interview among young women exposed to sexual assault within the past seven years. We hypothesized that higher TI severity would be associated with higher PTSD symptom severity, increased dissociative experiences, and increased maladaptive posttraumatic cognitions at the time of the baseline interview, as well as at 6- and 12-months post-baseline. The study also examined the influence of rape tactics in these associations due to their associations with TI and posttraumatic psychological sequalae.
In this study, higher TI severity was not significantly associated with PTSD symptom severity at any time point, contrary to prior findings and to hypotheses. There are several possible explanations for these findings. First, given that most individuals in this sample had experienced a sexual assault at least one year ago, it is possible that in the chronic PTSD phase, symptoms may be relatively stable, regardless of TI severity. Indeed, PTSD symptom severity scores broadly decreased among participants after the initial interview (likely influenced by potentially distressing elements of a trauma-focused study interview; Table 3) but remained overall high and maintained a consistent relationship with TI severity (see Figure 1). Over time, differences in coping style (e.g., approach- vs. avoidance-oriented) may attenuate PTSD symptomatology more so than TI severity. On the other hand, higher TI severity was associated with higher re-experiencing symptoms of PTSD, consistent with the Dual Representation Theory of trauma memories (Brewin et al., 2007), though these differences did not drive higher symptom severity overall.
Consistent with previous research, TI severity was associated with increased maladaptive posttraumatic cognitions, including those specifically related to the world (e.g., “People are not what they seem,” “The world is dangerous”), that were maintained in the long-term. Interestingly and contrary to previous research, TI severity was not associated with increased self-blame, perhaps suggesting TI severity may be associated more with perceiving the sexual assault as completely uncontrollable. Because maladaptive posttraumatic cognitions are powerful mechanisms in the maintenance of PTSD (LoSavio et al., 2017) and the relationship between TI and maladaptive posttraumatic cognitions appears stable over time, results have important implications for therapeutic work with sexual assault survivors who have experienced TI. It is likely particularly important to target maladaptive posttraumatic beliefs about the world in cognitive restructuring and emotional processing elements of trauma-focused treatments, as TI appears to be associated with prolonged negative changes in perceived sense of safety and trust in survivors of sexual assault. These treatments appear to alleviate PTSD symptoms via changes in maladaptive posttraumatic cognitions (Kangaslampi & Peltonen, 2019; Zalta et al., 2015), and they are suitable for individuals with elevated dissociative symptoms (Hagenaars et al., 2010; van Minnen et al., 2015; Zoet et al., 2018). Future studies may wish to explore the efficacy of these treatments in addressing maladaptive cognitions specifically related to TI experiences.
Few studies have investigated the relationship between TI and dissociative symptoms longitudinally. Consistent with hypotheses, higher TI severity was associated with increased dissociative symptoms at later time points. The relationship between TI scores and dissociative symptoms did not significantly change at the 6-and 12-month follow-ups, further demonstrating the lasting effects of TI on dissociative symptoms over time. Findings diverge from previous literature showing that peritraumatic dissociation minimally predicts posttraumatic psychopathology compared to persistent posttraumatic dissociation (Briere et al., 2005; Werner & Griffin, 2012). In this regard, findings from this study may suggest that peritraumatic TI is associated with posttraumatic dissociation beyond the potential contributions of TI-associated peritraumatic dissociation. According to a review by Carlson and colleagues (2012), posttraumatic dissociative symptoms typically increase immediately after a traumatic event and subsequently decline thereafter; for others, these symptoms persist without treatment. The latter trajectory appears to have been the typical response among the present sample, for whom the average time since assault at baseline was a few years. Clinically, dissociative symptoms can respond well to trauma-focused treatment (Resick et al., 2012; Zoet et al., 2018); future research could investigate if dissociative symptoms follow the typical pattern suggested by Carlson et al. (2012) for those who experience TI at an acute phase of recovery.
Regarding rape tactics, it is interesting that although all participants had experienced a FR or DAFR to be included in the study, almost half the sample indicated that their index assault included CR tactics (without force). CR tactics were associated with increased PTSD symptoms at each time point, whereas FR and DAFR tactics were not. CR may be associated with higher PTSD symptoms primarily via changes in arousal, behavior, and emotion, as changes in cognition (measured via the PTCI) were associated with DAFR and FR only. These results diverge from those of previous studies indicating that FR and DAFR tactics were significantly associated with increased PTSD symptoms (Zinzow, Resnick, McCauley et al., 2010; Zinzow, Resnick, Amstadter et al., 2010). However, these studies have typically not considered the impact of CR specifically. CR is considerably understudied compared to FR and DAFR, potentially due to narrow legal definitions of rape (Hannan et al., 2020), wherein the use of verbal pressure or manipulation to compel an unwilling person to have sex is not illegal in most states. Many previous studies of CR tactics have also only focused on threats of violence rather than nonviolent threats, such as threatening suicide or infidelity or threatening to release sensitive photographs if the victim does not comply (e.g., McConnell et al., 2017; Snipes et al., 2017). Studying consequences of rape in this way may significantly underestimate the psychological burden of sexual violence (Broach & Petretic, 2006), particularly given the high prevalence of verbal coercion (Smith et al., 2018). Findings from this study indicate CR should not be ignored in research or clinical work as it relates to the development of PTSD symptoms.
Finally, counter to prior research (Margarita et al., 2013; Sigurvinsdottir & Ullman, 2016), there were no significant associations between demographic variables (e.g., sexual orientation, racial/ethnic background) and psychological outcomes. However, this is unsurprising given that the sample was relatively homogenous in demographic characteristics, except sexual orientation.
Results of this study are interpreted with consideration of several limitations. First, participants in this study were all women between the ages of 16–35 and primarily of White racial background, limiting the generalizability of these findings to individuals of other genders and other age and racial/ethnic groups. As previously noted, inclusion criteria related to the time since the sexual assault (i.e., past seven years) also limit the ability to draw conclusions about the effects of TI over the developmental course of PTSD, as most participants within this sample meeting criteria for probable PTSD were already following a chronic, unremitting trajectory. In other words, it remains unclear how TI may affect PTSD symptoms, dissociative experiences, and negative posttraumatic cognitions during more acute phases of PTSD development (i.e., within three months), where intervention is most desirable. In a similar vein, when participants report on events from the past, as in the present study, responses are subject to inaccuracies and recall bias, which may unduly influence results (Coughlin, 1990). Given that TI itself is associated with mental numbing, memory inaccuracies are of particular concern. Though clinically memory accuracy is somewhat less important in achieving therapeutic results (trauma-focused therapies can address events for which memory is partially lost, such as due to head injury or loss of consciousness), prospective research is needed to further clarify the relationship between TI and posttraumatic psychopathology phenomenologically.
Similarly, though the longitudinal nature of the present study provides important information on the stability of posttraumatic symptomatology over time following TI, other variables not captured in this study – such as social support, coping strategies, or treatment engagement – may also influence the course of these experiences. Understanding factors that might reduce the severity of posttraumatic psychopathology following TI would be of clinical utility, and future research may examine these variables prospectively, particularly during the more acute post-trauma period, to clarify the nature of these changes.
A notable strength of the study procedure was the use of clinical interviews during the baseline assessments to more reliably assess psychopathology within the sample. However, longitudinal data relied upon self-report measures, which are limited by participants’ understanding of constructs presented and adherence to the instructions to respond to self-report measures keeping in mind the index assault. The extent to which measures of psychopathology at follow-up would be validated by clinician-administered measures is unknown. PCL-5 scores in particular should be interpreted as measures of participant distress rather than concretely established diagnoses.
In addition , a substantial number of participants did not have PCL-5 or DEMO data at the initial timepoint, as only those meeting diagnostic criteria for PTSD received those measures at baseline due to a software error. Utilizing this small subsample for PCL-5 analyses at the baseline timepoint may skew results. Moreover, this study was underpowered to detect some of the small effects. We interpret the corrected p values cautiously, as corrections further reduce power. As such, more highly powered studies are required to replicate preliminary results presented here.
Finally, although this study fills an important gap in the literature regarding the psychological effects of TI post-trauma over time, research on TI in humans is still relatively new, and its adoption as a phenomenon within public consciousness and policy is even newer. For example, public health messaging regarding consent has only recently shifted from “no means no” to encouraging affirmative consent (“yes means yes”) – that is, individuals engaging in sexual activity should not assume that the absence of verbal refusal implies consent. As these constructs continue to be studied, the conceptualization and measurement of TI is likely to be refined. Despite limitations and small effects, given the salience of the TI response in sexual assault, this preliminary study contributes to our understanding of the lasting deleterious psychological effects of TI and suggests pathways to mitigate them, including increased education regarding affirmative consent and the importance of addressing trauma-related beliefs about the world and posttraumatic dissociation in trauma-focused therapies.