The effect of autonomic nervous system reactivity to a stressor on
intrusion development after watching a trauma film
University of Amsterdam bachelor thesis
Kim Cramer
Student number: 11279826 Daily supervisor: Lotte Hilberdink External assessor: Mirjam van Zuiden
Word count: 5672 Abstract word count: 195
Abstract
Intrusions are a hallmark symptom of post-traumatic stress disorder (PTSD) that bring great
discomfort and inconvenience to the patients experiencing them. Previous research has shown that acute stress and the autonomic nervous system (ANS) reactivity this stress evokes might play a key role in intrusion development, but the mechanisms underlying the development of intrusions and their characteristics still remain largely unclear. The current study helped gather information about intrusion development by investigating the relationship between acute systolic and diastolic blood pressure (sBP and dBP), heart rate (HR) and pre-ejection period (PEP) reactivity and intrusion frequency, associated vividness and distress. The current study achieved this by combining the socially-evaluated cold pressor test and trauma film paradigms to simulate physiological and mental stress before and during simulated trauma. In the following week, participants recorded their
experienced intrusions and their associated vividness and distress. Linear mixed models and multiple regression analyses revealed that acute sBP and dBP reactivity after trauma may serve as a predictor for the vividness of developed intrusions. These findings add to the existing knowledge about intrusion development, offering information that can help to understand the processes of ANS reactivity and intrusion development in PTSD.
1. Introduction
Posttraumatic stress disorder (PTSD) is an anxiety disorder that approximately 10% of individuals develops after exposure to a traumatic event such as physical harm or sexual assault (De Vries & Olff, 2009). Symptoms of the disorder include re-experiencing of the traumatic event through intrusive memories, avoidance of trauma-related stimuli, hyperarousal and negative alterations in mood and cognition (DSM-V, American Psychiatric Association, 2013). PTSD is often comorbid with other mental disorders, as 78% of male patients get diagnosed with an additional disorder (Spijker et al., 2018). Patients are also at high risk for developing non-psychiatric comorbidities such as cardiovascular diseases, endocrine and immune disorders (Seng et al., 2005). Symptoms of PTSD could lead to changes in behavior and social isolation (DSM-V, American Psychiatric Association, 2013). The disorder can also substantially affect the emotional wellbeing of family and acquaintances of the person diagnosed with PTSD (Galovski et al., 2004) and the high economic costs of PTSD treatment are partially financed by society (Creamer et al. 2011). Since PTSD has major impact on an emotional, social and economic level, proper treatment is needed. Knowledge on the development of PTSD and
its pathology is progressing, but current pharmacological and therapeutic treatment options are ineffective in treatment-resistant patients and suboptimal for other, largely chronic PTSD patients (Friedman et al., 2007). Next to this, most current therapies are focused on treating the disorder after it has already been established, rather than attempting to target mechanisms underlying the
development of PTSD as effective preventive strategy. With research advancing more interventional methods are being developed. However, most interventions focus follow a psychological approach, rather than taking the biological responses before and during trauma into account (Karatzias et al., 2019, Roberts et al., 2019).
One of the hallmark symptoms of PTSD is intrusive memories, also called intrusions. Intrusions are recurrent, involuntary and unwanted memories through which an individual re-experiences a traumatic event. Intrusions are distressing, evoking anxiety or unease when the intrusion appears. Intrusions are also vivid, meaning that the memory is clear and powerful (Hackmann et al., 2004). Intrusions take on various forms, such as nightmares or imagery, but can also appear as mere thoughts or a mix of both. It is important to focus on intrusions when conducting research with the intent of improving PTSD treatment, as intrusions evoke many negative emotions and directly influence other symptoms of PTSD. For example, avoidance of stimuli triggering intrusions may result in social isolation, drastically decreasing PTSD patients’ quality of life (McFarlane, 1992).
The most widely accepted cognitive model for PTSD developed by Ehlers and Clark (2000) suggests that intrusions are the result of altered emotional memory consolidation and contextualization during encoding of the traumatic memory (Ehlers and Clark, 2000; Brewin, 2015). Memory contextualization is part of memory consolidation in which an event and the context in which it has happened are linked together. In PTSD, it is thought that memory consolidation does not happen properly during and after a traumatic event. Disturbed memory contextualization leads to emotional events being incorrectly linked to certain contexts. This incorrect linking facilitates retrieval of the trauma memory in situations not directly related to the traumatic event and could lead to intrusions (Ehlers et al., 2004, Brewin, 2015). Memory consolidation has been found to be vulnerable to a variety of factors, such as stress (Corbett et al., 2017). Acute stress affects memory encoding by enhancing
consolidation for emotional memories. This ultimately leads to both healthy individuals and PTSD patients remembering emotional stimuli better than neutral stimuli (Kuhlmann and Wolf, 2006). Research by Stawski et al. (2009) has shown that while episodic memory is not affected by acute stress, participants exposed to a stressor did exhibit increases in intrusions compared to controls. Keeping in mind the disturbed memory contextualization, facilitated memory retrieval and enhanced consolidation for emotional memories, it is clear that these processes may play a role in PTSD pathology and intrusion development.
Stress can influence memory, but the autonomic nervous system (ANS) also responds to stress. The stress reactions that start when a healthy individual is exposed to a stressor can narrow arteries, leading to an increase in blood pressure (BP) (Eliot, 1992). Stress can also decrease the time between heart contractions, leading to a decreased pre-ejection period (PEP) (Berntson et al., 1994) and heart rate (HR) (Vrijkotte et al., 2000). These physiological changes could be helpful when an individual experiences a dangerous situation, enabling the individual to respond adequately to possible threats. However, high ANS reactivity to stressors is also known to be a vulnerability factor for various
diseases (Schlotz, 2013), possibly including PTSD. Results from a study by Minassian et al (2015) and Morris et al. (2016) suggest that reduced heart rate pre-deployment may serve as a predictor for PTSD development during and after deployment in soldiers, and a meta-analysis by Morris et al. (2006) even brings forward heart rate variability within 24 hours post-trauma as a predictor as a biomarker for PTSD. Additionally, research by Buckley et al. (2004) has shown an increase in BP reactivity in PTSD patients compared to healthy controls. However, the relations between ANS reactivity and PTSD development, before the disorder manifests itself, are less clear. High stress reactivity of the autonomic nervous system is suspected to be an endophenotype for several mental disorders, such as psychosis and depression (Myin-Germeys, 2007; van Winkel et al., 2015), so more research into the impact of acute stress on disorder development is needed. While investigating acute stress reactivity during and after trauma is impossible in most cases, the influence of acute stress reactivity during and after simulated trauma on intrusion development can possibly be investigated in experimental settings.
In an experimental setting, changes to sympathetic nervous system functioning such as an increased blood pressure and heart rate and decreased PEP have been found to be easily evocable by using the socially evaluated cold-pressor test (henceforth referred to as SECPT). This stress test has been found to reliably evoke biological stress in individuals undergoing the test, increasing autonomic nervous system responses as well as salivary cortisol (Sänger et al., 2014, Schwabe et al., 2008). This makes the SECPT an appropriate way to simulate pre-trauma physiological stress in experimental settings. Many factors possibly influencing the development of PTSD and intrusions have been identified, but the exact relationships between ANS reactivity and intrusion development and the mechanisms underlying intrusion development remain partially unclear. Taking into account the current knowledge gaps, as well as the available information on the effects of stress on memory and autonomic nervous system functioning on PTSD pathology, leads to the current research question: what is the effect of autonomic nervous system reactivity on the development of intrusions? It is hypothesized that high autonomic nervous system reactivity, measured as a high difference between resting and stress heart rate, systolic and diastolic blood pressure and PEP, increases the total number
and associated vividness and distress of intrusions individuals experience. The current study
investigated this hypothesis by admitting participants to the SECPT or a control condition. Participants watched a trauma film to induce intrusions. Intrusions and their associated distress and vividness were reported by participants in the following week through a smartphone app.
2. Materials and methods
2.1 Participants
The current study investigated 68 healthy males between 18 and 40 years old. Only male participants were included in this study as studies by Zuiden et al. (2013) and Pineles et al. (2017) have found that PTSD and intrusion development, pathology and acute stress reactivity vary between genders. Thus, including measurements from both male and female participants could negatively influence the validity of the study’s results.
Participants were included in the study if they met the inclusion criteria, which were as follows: having a Caucasian ethnic background, a BMI between 18.5 and 30, fluency in Dutch language and possessing a smartphone to download the application for intrusion reporting. Participants were excluded from participation if they suffered from a current major medical or mental disorder, high levels of anxiety, depressive or PTSD related symptoms. Participants did not use any medication affecting the ANS and HPA-axis functioning and did not smoke on a regular basis.
2.2
Recruitment procedures
Information about the study was distributed at appropriate locations, such as university campuses and relevant social media groups. Participants who signed up for the study through e-mail or lab.uva.nl received additional information about the study and were asked to provide their phone number for the telephonic screening.
2.3 Experimental procedure
The study consisted of 1 telephonic and 3 in-person assessments. Eligibility of participants was assessed through a telephonic and in-person screening (T0, T1), followed by the experimental paradigm (T2) and a follow-up session (T3). Between the T2 and T3, intrusion reports made by participants were repeatedly checked by researchers during the following reporting week (T2.1-2.7). After completion of the T3, participants who completed the study were awarded 40 euros or 4 UvA course credits. If participants partially completed the study, they were rewarded accordingly. T0: Telephonic screening
During the telephonic screening eligibility of the participant was assessed according to the in- and exclusion criteria of the study. If participants met the inclusion criteria during the telephonic
screening, they were sent additional information through e-mail and the appointment for the second assessment (T1) was planned.
T1: In-person screening
Upon the start of the T1 assessment participants signed an informed consent form. Exact weight was determined using a regular scale, participants’ BMI was calculated and participants were asked to fill out several questionnaires to measure presence of depression, anxiety and stress symptomatology. Depressive symptoms experienced in the past week were assessed using DASS-21 (Depression Anxiety Stress Scale, de Beurs et al., 2001). Participants could score a total of 21 points on each subscale and were included only if they scored <12 points on the depression subscale and <5 points on the anxiety subscale. LEC-5 (Life Event Checklist, Gray et al., 2004) was administered to assess traumatic events participants had experienced throughout their life. If the researcher deemed the participant to be at risk for adverse events due to prior traumatic experiences that were possibly similar to the content of the trauma film, the participant was excluded from participation. Finally, PTSD checklist for DSM-5 (PCL-5), used to screen for current PTSD, was administered to assess the DSM-5 symptoms of PTSD an individual has experienced in the past week. Participants who met the threshold for at least two out of four clusters of the PCL-5 were excluded from participating as this score may indicate current PTSD. All questionnaires used are appropriate ways to measure disorder symptomatology, as validated by Henry & Crawford (2005) and Blevins et al. (2015).
If all inclusion criteria were met, the experimental (T2) and follow-up (T3) assessments were planned. Between the T1 and T2, participants were asked to collect saliva for two days to measure participants’ cortisol awakening response. Participants took a dexamethasone tablet on the evening before the second saliva collection day to facilitate assessing hypothalamic-pituitary-adrenal axis functioning. T2: Experimental assessment
In order to minimize individual differences in bodily functioning or possible effects of substance use on stress reactivity, participants were asked to not take alcohol, drugs or medication 24 hours prior to the appointment, could not work out or have lunch the day of the appointment and could not smoke, drink anything other than water or brush their teeth in the last 2 hours before the assessment. Experimental assessments started between 13:00 and 15:30 to control for diurnal cortisol cycles, were performed in the UvA Lab L1.26-27 at the Roeterseiland campus and had a duration of approximately 95 minutes.
Stress measurements
During the stess measurements blood pressure, ECG and mood were measured using a blood pressure meter, the VU-AMS (VU Ambulatory Monitoring system) and a VAS (visual analog scale) questionnaire. PEP and heart rate data were collected using the VU-AMS, whereas blood pressure data was collected using a blood pressure meter. Six stress measurements were carried out during the assessment to assess the levels of physical and emotional stress participants experienced throughout the experimental procedure. One single stress measurement had a duration of 150 seconds and prior to the start of the assessment, participants practiced the stress measurement once.
The experimental assessment started with 15 minutes of rest as the participant read magazines while staying seated. Afterwards, the first stress measurement was conducted which served as a baseline measurement (stress and rest measurement 1). Participants then underwent either the stressful SECPT or the non-stressful control procedure. Immediately after the water test, participants underwent the second stress measurement. Participants then watched the 15-minute trauma film clip, and underwent the third stress measurement. After watching the trauma film, participants again rested for 10 minutes by reading magazines while seated and underwent the fourth stress
measurement (second resting measurement). After the fourth stress measurement participants filled out three memory questionnaires regarding the movie fragment, of which the results will not be evaluated in the current thesis. Participants then underwent the fifth stress measurement. Afterwards, participants were asked to download the smartphone application “Intrude AMC” by Mecosud b.v. through which participants were able to report intrusions and their characteristics in the following week. After successful installation, participants received information about the app and learned how to make reports before starting the sixth and final stress measurement. Participants could contact the researchers if any problems arose during the following week.
Figure 1: Schematic overview of the experimental assessment (T2), with the duration and the measurements conducted during each activity.
Water test
Prior to the experimental assessment, participants were randomly admitted to the experimental condition where they underwent the SECPT, or the control group. After the first stress measurement, participants underwent the water test. Participants admitted to the experimental condition received verbal and written instructions from an unknown, neutral researcher and signed an additional informed consent form. Participants were instructed to submerge their hand and wrist in cold water with a temperature ranging from 1 to 3 degrees Celsius) as long as they could. Participants were not informed about the duration of the test, which was 180 seconds. Participants were instructed to look into a recording video camera used for facial analyses. Both the signed informed consent form and camera recordings were performed to increase the level of psychological stress experienced by the participants and were not used for further analyses. The SECPT has been shown by Schwabe et al. (2008) to be a reliable way to induce autonomic nervous system responses. Participants admitted to the control condition were greeted by a friendly unknown researcher prior to the water test, received instructions verbally and no video recordings were made. The water used for the control condition had a temperature between 37 and 39 degrees Celsius and did not evoke any biological or emotional stress (Schwabe et al., 2008).
Trauma film paradigm
After the second stress measurement, participants were asked to view a movie clip with an aversive nature participants were informed about prior to watching. The fragment used was a 15-minute clip from the movie ‘Irreversible’ (2002) and contained footage of a woman being physically, sexually and
verbally assaulted by a man. Participants viewed the movie on a computer screen and audio was provided through headphones. Previous research has shown that exposing people to this particular movie clip is a reliable way to induce distress, disgust and sympathetic nervous system activation, as well as intrusions in the days following exposure to the movie (Weidmann et al., 2009). After watching the movie, participants were asked to fill out additional VAS questionnaires about their feelings related to the movie.
T2.1-2.7: Intrusion reports
The Intrude app was used as a diary to report intrusions during the 7 days following the experimental assessment. Participants were instructed to report their intrusions as they happened though the “NOW” option, but could also report intrusions experienced earlier through the “EARLIER” option if immediate reporting was impossible. Participants were asked when and how the memory occurred (as thoughts, images or both), to rate the associated vividness and distress and to give a short description of the intrusion. These questions were asked to allow researchers to verify the validity of the reported intrusions, as intrusions require to be distressing and vivid in order to be considered valid.
Participants received daily reminder notifications at 10:00 AM and PM . Participants were asked to respond to the notifications by making intrusion reports with 0 associated distress and vividness. This way, researchers could verify that participants were still actively participating and responses to the notifications were not counted as actual intrusions. If participants did not respond to the notifications timely, they were contacted through e-mail or telephone.
T3: Follow-up assessment
The follow up assessment (T3) took place exactly a week after the experimental assessment. Participants were asked to fill in a questionnaire regarding their intrusion reporting and the content of their intrusions. Participants filled out the PCL-5 during the follow-up assessment to check for possible remaining PTSD symptoms due to the trauma film. If scores on the PCL-5 or participant’s remarks gave rise to concerns about the participant’s wellbeing, they were offered psychological help from AMC psychologists or their general practitioner. Ultimately, participants received a debriefing by the researcher.
2.4 Statistical analysis
Statistical analyses were carried out using Rstudio (version 3.6.1, 2019-07-05) and R packages nlme (version 3.1-140) and emmeans (version 1.4.3.01). For all analyses, a significance level of α=0.05 was used. 2 participants were excluded due to adverse events and 3 participants cancelled participation, resulting in a final sample size of n=63 participants that completed the experimental paradigm. Demographic variables were analyzed using Shapiro-Wilk and Levene’s test to assess violation of assumptions for normality and equal variances. If assumptions were met, Student’s t-test was used to assess differences between conditions. If assumptions were violated, a Mann-Whitney U test was used instead. If demographic variables were found to differ significantly between conditions, the variable was added to the further analyses as a covariate. Inspected variables included age, BMI and total DASS-21 and PCL scores.
An explorative linear mixed model (LMM) was performed to define the stress measurement where the effect of ANS stress reactivity was the highest, to use in the regression analyses. Assumptions were tested, however research by Jacqmin-Gadda et al. (2007) has shown that linear mixed effects models are robust against assumption violations. Stress measurement was added as a fixed effect, while condition was added as a random effect. An interaction term of time*condition was added to the model. Estimated marginal means were compared as a post-hoc test to identify significant differences between stress measurements using pairwise comparisons. Values were corrected using the Bonferroni correction and were considered significant if p<0.05.
In order to investigate the relationship between ANS reactivity and intrusions, linear regression analyses were performed. ANS reactivity was defined by subtracting the HR, BP and PEP values of the first stress measurement, a baseline measurement, from the HR, BP and PEP values of the
measurement found with the exploratory LMM for every participant. This difference in ANS activity between a resting state and stressful state indicates reactivity of ANS measures (Sharpley, 1994). The total number of intrusions per participant was determined by counting the number of valid
intrusions. The mean associated vividness and distress was determined by adding up the reported values for vividness and distress of each intrusion, then dividing the value by the number of
intrusions for each participant. For linear regression analyses, sBP, dBP, HR and PEP reactivity values, condition and ANS reactivity*condition effects were included as predictors for total number of intrusions and average intrusion vividness and distress.
Results
3.1 Demographic characteristics
A total of n=68 healthy male participants were allocated to the warm or cold water intervention conditions. 5 participants were excluded from further analyses due to cancellation of participation (n=3) or discontinuation during the intervention (n=2), ultimately resulting in a total sample of n=63 (experimental condition n=29, control condition n=34). Participants of the experimental condition had a mean age of 22.30 (SD=4.83) and BMI of 22.92 (SD=2.58) and participants of the control condition had a mean age of 22.31 (SD=3.88) and BMI of 22.61 (SD=2.17).
No significant intergroup differences were found for age, BMI, DASS and PCL scores (table 1).
Mean (SD) Experimental condition Control condition P-value, W-value
Age 22.30 (4.83) 22.31 (3.88) 0.947, 532.5
BMI 22.92 (2.58) 22.61 (2.17) 0.477, 458
DASS total score 6.90 (6.12) 5.37 (3.09) 0.164, 633
PCL total score 2.77 (3.24) 3.86 (3.97) 0.456, 583.5
Table 1: Participant characteristics, showing no significant intergroup differences as measured during the in-person screening (T1), analyzed using the Mann-Whitney U test
3.2 Explorative LMM
In order to determine the stress measurement and accompanying ANS data that was most suited to include in the linear regressions, an explorative linear mixed model was performed. No covariates were added to the linear mixed models. Additionally, no differences were found between groups for baseline autonomic nervous system activity measurements. Stress measurement was included as fixed effects, whereas condition was included as random effect. 12 more participants were excluded from the linear mixed model analyses using data collected with the VUAMS (PEP and HR
measurements) due to measuring equipment failure, resulting in a final number of n=51 participants (n=25 for the experimental condition, n=26 for the control condition) included in PEP and HR
analyses.
Residuals for sBP, dBP, PEP and HR were found to be significantly not normally distributed for at least one of two conditions. After performing the exploratory LMM, no significant time*group interaction effects were found for sBP, dBP, PEP and HR. Additionally, no significant group effects were found for
sBP, dBP, PEP and HR. However, multiple significant time effects were found for sBP, dBP and HR. P-values were adjusted for the multiple comparisons using the Bonferroni correction.
For diastolic blood pressure, several significant time effects were found (figure 2). A significantly higher dBP was found for the second stress measurement compared to the first stress measurement (T(63)=-3.124, p=0.029), for the third stress measurement compared to the first stress measurement (T(63)=-5.009, p=<0.0001), for the fifth stress measurement compared to the first stress
measurement (T(63)=-3.174, p=0.025) and for the sixth stress measurement compared to the first stress measurement (T(63)=-3.310, p=0.016)
Diastolic blood pressure throughout stress measurements
Figure 2: Results for the diastolic blood pressure LMM displayed in a graph, showing significant time effects and standard deviations. * = p<0.05, **=p<0.001, n=62
For systolic blood pressure, several significant time effects were found (figure 3). A significantly higher sBP was found for the second stress second stress measurement compared to the first stress
measurement (T(63)=-3.151, p=0.027), for the second stress measurement compared to the fourth stress measurement (T(63)=5.477, p<0.0001), for the second stress measurement compared to the fifth stress measurement (T(63)=5.013, p<0.001), for the second stress measurement compared to the sixth stress measurements (T(63)=4.270, p=<0.001), for the third stress measurement compared to the fourth stress measurements (T(63)=5.191, p=<0.0001), for the third stress measurement
compared to the fifth stress measurements (T(63)=4.729, p<0.001) and for the third stress measurement compared to the sixth stress measurement (T(63)=3.990, p<0.001).
Systolic blood pressure throughout stress measurements
Figure 3: Results for the systolic blood pressure LMM displayed in a graph, showing significant time effects and standard deviations. * = p<0.05, **=p<0.001, n=62
For heart rate, several significant time effects were found (figure 4). Due to measuring equipment failure, 11 participants were excluded from the analyses, resulting in n=51 participants included in the analysis. A significantly higher HR was found for the first stress measurement compared to the sixth stress measurement (T(52)=4.386, p=0.0002), for the third stress measurement compared to the second stress measurement (T(52)=-3.928, p=0.002), for the third stress measurement compared to the fourth stress measurement (T(52)=3.263, p=0.019), for the third stress measurement compared to the fifth stress measurement (T(52)=3.820, p=0.003), for the third stress measurement compared to the sixth stress measurement (T(52)=6.476, p<0.0001) and the fourth stress measurement compared to the sixth stress measurement (T(52)=3.213, p=0.022).
Heart rate throughout stress measurements
Figure 4: Results for the heart rate LMM displayed in a graph, showing significant time effects and standard deviations. * = p<0.05, **=p<0.001, n=51
For PEP, no significant time effects were found. Due to measuring equipment failure, 11 participants were excluded from the analyses, resulting in n=51 participants included in the analysis.
In sum, these results suggest that the third stress measurement showed the largest effects on stress reactivity. Therefore, the linear regressions were performed for the systolic and diastolic BP, HR and PEP data of the third stress measurements in the linear regressions after subtracting baseline values for each participant, to investigate the relationship between ANS reactivity and the development of intrusions.
3.3 Linear regressions
For all separate linear regression analyses, participants with incomplete data and extreme values were removed from the datasets. 16 participants did not report any intrusions at all in the week following exposure to the trauma film, and were thus excluded from the regression analyses
regarding intrusion vividness and associated distress, as data was unavailable for these variables. This resulted in a total of n=56 participants for the blood pressure and intrusion frequency analyses and n=42 for the blood pressure and intrusion characteristic analyses. For HR and PEP, this results in a total of n=49 participants for the intrusion frequency analyses and n=36 for the intrusion
characteristic analyses.
3.3.1 Relation between blood pressure and intrusion frequency, vividness and distress
Systolic blood pressure
For intrusion frequency, a regression equation of F(1, 57) = 0.66, p = 0.580 with an R^2 of 0.033 was found. This shows that the model used is not suitable to predict intrusion frequency (table 2 and 3). For intrusion vividness, a regression equation of F(1,43) = 2.968, p = 0.042 with an R^2 of 0.172 was found. This shows that the model used is suited to predict intrusion vividness and implies a possible relationship (table 2 and 3, figure 5)
For intrusion distress, a regression equation of F(1, 43) = 1.771, p = 0.167 with an R^2 of 0.11 was found. This shows that the model used is not suitable to intrusion distress (table 2 and 3)
→ Main/interaction effects ↓Dependent variables
sBP reactivity Condition sBP reactivity * Condition
Intrusion frequency T(1, 57) = -1.030 p = 0.307 T(1, 57) = 0.311 p = 0.757 T(1, 57) = 1.277 p = 0.207 Intrusion vividness T(1, 43) = 1.498 p = 0.142 T(1, 43) = -0.361 p = 0.720 T(1, 43) = -0.597 p = 0.554 Intrusion distress T(1, 43) = 1.834 p = 0.072 T(1, 43) = -0.468 p = 0.642 T(1, 43) = -1.799 p = 0.079
Table 2: Degrees of freedom and T- and p-values for main and interaction effects for intrusion frequency, vividness and distress ~ sBP reactivity + Condition + sBP*Condition multiple regression analyses
Intrusion frequency 0.034 0.66 0.580
Intrusion vividness 0.172 2.968 0.040 (*)
Intrusion distress 0.11 1.771 0.167
Table 3: Multiple R^2, F-statistics and P-values for intrusion frequency, vividness and distress ~ sBP reactivity + Condition + sBP*Condition regression analyses
The effect of systolic blood pressure variability on mean intrusion vividness
Figure 5: Graph for the linear regression investigating the relationship between systolic blood pressure and intrusion vividness
For intrusion frequency, a regression equation of F(1, 57) = 0.787 , p = 0.506 with an R^2 of 0.039 was found for the model. This shows that the model used is not suitable to predict intrusion frequency (table 4 and 5)
For intrusion vividness, a regression equation of F(1,43) = 1.751, p = 0.032 with an R^2 of 0.109 was found for the model. This shows that the model used is suitable to predict intrusion vividness and implies a possible relationship (table 4 and 5, figure 6)
For intrusion distress, a regression equation of F(1, 43) = 1.529, p = 0.221 with an R^2 of 0.096 was found. This shows that the model used is not suitable to predict intrusion frequency (table 4 and 5)
→ Main/interaction effects ↓Dependent variables
dBP reactivity Condition dBP reactivity * Condition
Intrusion frequency T(1, 57) = 0.144 P = 0.886 T(1, 57) = -0.076 P = 0.939 T(1, 57) = 0.240 P = 0.811 Intrusion vividness T(1, 43) = 1.556 P = 0.127 T(1, 43) = 0.114 P = 0.910 T(1, 43) = -0.974 P = 0.336 Intrusion distress T(1, 43) = 1.653 P = 0.106 T(1, 43) = 0.235 P = 0.815 T(1, 43) = -1.547 P = 0.129
Table 4: Degrees of freedom and T- and p-values for main and interaction effects for intrusion frequency, vividness and distress ~ dBP reactivity + Condition + dBP*Condition multiple regression analyses
Multiple R2 F-statistic P-value
Intrusion frequency 0.040 0.787 0.506
Intrusion vividness 0.109 1.751 0.032 (*)
Intrusion distress 0.096 1.529 0.221
Table 5: Multiple R^2, F-statistics and P-values for intrusion frequency, vividness and distress ~ dBP reactivity + Condition + dBP*Condition reactivity regression analyses
Figure 6: Graph for the linear regression investigating the relationship between diastolic blood pressure and intrusion vividness
For intrusion frequency, a regression equation of F(1, 50) = 1.583 , p = 0.205 with an R^2 of 0.087 was found. This shows that the model used is not suitable to predict intrusion frequency (table 6 and 7). For intrusion vividness, a regression equation of F(1,37) = 2.353, p = 0.088 with an R^2 of 0.160 was found. This shows that the model used is not suitable to predict intrusion vividness (table 6 and 7). For intrusion distress, a regression equation of F(1, 37) = 0.515, p = 0.675 with an R^2 of 0.040 was found. This shows that the model used is not suitable to predict intrusion distress (table 6 and 7).
→ Main/interaction effects ↓Dependent variables
HR reactivity Condition HR reactivity * Condition
Intrusion frequency T(1, 50) = 0.808 P = 0.429 T(1, 50) = -0.224 P = 0.808 T(1, 50) = -0.233 P = 0.816 Intrusion vividness T(1, 37) = -1.832 P = 0.075 T(1, 37) = 1.472 P = 0.150 T(1, 37) = 1.347 P = 0.186 Intrusion distress T(1, 37) = -0.320 P = 0.751 T(1, 37) = 0.694 P = 0.492 T(1, 37) = -0.031 P = 0.976
Table 6: Degrees of freedom and T- and p-values for main and interaction effects for intrusion frequency, vividness and distress ~ HR reactivity + Condition + HR*Condition multiple regression analyses
Multiple R2 F-statistic P-value
Intrusion frequency 0.087 1.583 0.205
Intrusion vividness 0.160 2.353 0.088
Intrusion distress 0.040 0.515 0.675
Table 7: Multiple R^2, F-statistics and P-values for intrusion frequency, vividness and distress ~ HR reactivity + Condition + HR*Condition regression analyses
3.3.3 Pre-ejection period linear regressions
For intrusion frequency, a regression equation of F(1, 50) = 1.583 , p = 0.205 with an R^2 of 0.087 was found. This shows that the model used is not suitable to predict intrusion frequency (table 8 and 9).
For intrusion vividness, a regression equation of F(1,37) = 2.353, p = 0.088 with an R^2 of 0.160 was found. This shows that the model used is not suitable to predict intrusion vividness (table 8 and 9). For intrusion distress, a regression equation of F(1, 37) = 0.515, p = 0.675 with an R^2 of 0.040 was found. This shows that the model used is not suitable to predict intrusion distress (table 8 and 9).
→ Main/interaction effects ↓Dependent variables
PEP reactivity Condition PEP reactivity * Condition
Intrusion frequency T(1, 49) = -0.898 P = 0.374 T(1, 49) = -0.390 P = 0.706 T(1, 49) = 1.103 P = 0.275 Intrusion vividness T(1, 37) = -1.183 P = 0.245 T(1, 37) = 1.447 P = 0.156 T (1,37) = 0.707 P = 0.484 Intrusion distress T(1, 37) = -1.763 P = 0.086 T(1,37) = 0.210 P = 0.834 T(1, 37) = 1.807 P = 0.079
Table 8: Degrees of freedom and T- and p-values for main and interaction effects for intrusion frequency, vividness and distress ~ PEP reactivity + Condition + PEP*Condition multiple regression analyses
Multiple R2 F-statistic P-value
Intrusion frequency 0.400 0.787 0.506
Intrusion vividness 0.126 1.733 0.178
Intrusion distress 0.096 1.529 0.221
Table 9: Multiple R^2, F-statistics and P-values for the complete intrusion frequency, vividness and distress ~ PEP reactivity + Condition + PEP*Condition regression analyses model
Discussion
More knowledge on the development of intrusions is important for improving PTSD prevention and treatment. However, the trajectory leading up to PTSD psychopathology and factors that could influence this development remain partially unclear, and more research is needed. The current study helped gather more information on intrusion development by investigating the relationship between autonomic nervous system reactivity and the development of intrusions in the week after watching a
trauma film. In the current study, the relationship between the addition of a psychosocial stressor to the trauma film paradigm and the development of intrusions was measured. This is a relatively novel approach, which was expected to lead to a more robust stress response and a more realistic model for the development of intrusions compared to previous studies. Results of the current study have shown possible relationships between ANS reactivity, measured as the differences in sBP and dBP between rest and post-trauma film measurements, and mean intrusion vividness.
P-values found for the multiple regression models were significant, however individual p-values for the predictors were not. This can indicate a problem with multicollinearity between predictor variables, or might imply that there are other factors that have influenced the obtained sBP and dBP reactivity values that were not explicitly added into the model. However, as the overall model was significant, the models investigating sBP or dBP reactivity, condition and the interaction between them are suitable to predict intrusion vividness and imply a possible relationship. No significant relationships between sBP, dBP, PEP and HR reactivity and the total number of intrusions and mean associated distress were found, and no significant relationships between PEP and HR and associated intrusion vividness were found.
Regarding our main research question about effect of autonomic nervous system reactivity on the development of intrusions, the results suggest that there is a positive relationship between systolic and diastolic blood pressure and the vividness of intrusions developed after watching a trauma film. High sBP and dBP reactivity during and after times of distress have been found to be increased in PTSD patients after the disorder has already been developed (Buckley et al., 2004), which is in line with the findings of the current study.
However, no influence of heart rate and PEP variability on the development of intrusions was found, and none of the ANS reactivity variables were able to evoke a measurable effect on the number of intrusions and the distress of the intrusions participants experienced in the current study. This is partially in line with our expectations. The combination of psychological and physiological stress evoked in the current study’s experimental condition was expected to increase ANS reactivity, which was in turn expected to significantly increase intrusion frequency, vividness and distress. However, intrusion frequency and distress did not seem to be affected by ANS reactivity, and HR and PEP reactivity was also not found to influence intrusion development. An explanation for this could be the amount of time that has passed between the traumatic event and HRV measurements, which could possibly influence the ability to detect a relationship between PEP and HR reactivity and intrusion development. For example, cortisol levels in healthy individuals do not peak during or shortly after exposure to acute stress, but can take up to 20 minutes to reach their peak (Quaedflieg et al., 2015).
Additionally, Buckley et al. (2004) did not find a relationship between HR reactivity and PTSD symptoms in patients, which is in line with the current findings.
Limitations
Mind should be paid to limitations that arose during the study. Due to failure of electronic measuring equipment, less HR and PEP reactivity output was gathered than was initially desired. This decrease in sample size for the HR and PEP reactivity variables negatively influenced the power of the analyses, and thus the current study may have failed to pick up small effects of HR and PEP reactivity on intrusion development that would have been significant when using a larger sample, and were found for the BP analyses.
It could be that exposure to media that an individual could consider traumatizing does not lead to the same psychopathological course of events as exposure to real-life trauma. During the current study, sixteen participants reported they did not experience any intrusions. Whereas normally, 15% of participants do not report any intrusions after watching the trauma film (James et al., 2016), this percentage rose to approximately 25% for the current study. The trauma film paradigm was validated by Holmes et al. in 2004, but might not be valid nowadays. Individuals undergo a much greater exposure to verbal, sexual and violent graphic content through social media exposure now compared to eleven years ago, possibly leading to emotional and physiological desensitization in regards to explicit content (De Choudhury et al., 2014; Mrug et al., 2015) and thus less intrusions. Another factor that may have contributed to the lack of power for the analyses, due to some participants not reporting any intrusions, is the decision to include only male participants in the current study. PTSD research in women is important as women have a 1.5-2 times as high chance of developing PTSD throughout their life, a heightened number that may be associated with female sex hormones. Female estrogen levels have been found to be a predictor of intrusion quantity (Cheung, 2013), implying a difference in pathological processes of PTSD development between genders. Stress reactions differ between genders and may impact emotional memory consolidation and recall (Cheung, 2013), and thus relationships between ANS reactivity and the development of intrusions may be present in females but not in males.
More research into the development of PTSD symptomatology is needed, and the current thesis gathered additional information on intrusion development and factors possibly influencing this development. Gathering more information on intrusion development during or shortly after trauma is important to uncover the mechanisms underlying intrusion development, which could then possibly facilitate the development of new preventive and therapeutic strategies for PTSD. Despite the
limitations that arose during the study, the outcomes offer information that can help to understand the processes of ANS reactivity and intrusion development in PTSD. However, more research is needed. For example, the paradigm could be repeated in females, possibly leading to different results due to hormonal differences. Focusing on the impact of ANS reactivity on different PTSD symptoms, such as avoidance, could also offer a greater understanding of the development of other PTSD symptoms, possibly contributing to novel preventive and treatment targets. Additionally,
Furthermore, replicating the current study with more statistical power could show results that the current study was unable to find. Research into PTSD and intrusion development remains important, and more research is needed to improve treatment and preventive strategies for patients suffering from this life-changing disorder.
References
de Vries, G. J., & Olff, M. (2009). The lifetime prevalence of traumatic events and posttraumatic stress disorder in the Netherlands. Journal of Traumatic Stress: Official Publication of The International Society for Traumatic Stress Studies, 22(4), 259-267.
Arlington, VA: American Psychiatric Association (2013). American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders: Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition.
Spijker, E. E., Jones, K., Duijff, J. W., Smith, A., & Christey, G. R. (2018). Psychiatric comorbidities in adult survivors of major trauma: findings from the Midland Trauma Registry. Journal of primary health care, 10(4), 292-302.
Seng, J. S., Graham-Bermann, S. A., Clark, M. K., McCarthy, A. M., & Ronis, D. L. (2005). Posttraumatic stress disorder and physical comorbidity among female children and adolescents: results from service-use data. Pediatrics, 116(6), e767-e776.
Galovski, T., & Lyons, J. A. (2004). Psychological sequelae of combat violence: A review of the impact of PTSD on the veteran's family and possible interventions. Aggression and violent behavior, 9(5), 477-501.
Creamer, M., Wade, D., Fletcher, S., & Forbes, D. (2011). PTSD among military personnel. International Review of Psychiatry, 23(2), 160-165.
Friedman, M. J., Marmar, C. R., Baker, D. G., Sikes, C. R., & Farfel, G. M. (2007). Randomized, double-blind comparison of sertraline and placebo for posttraumatic stress disorder in a Department of Veterans Affairs setting. Journal of Clinical Psychiatry, 68(5), 711-720.
Karatzias, T., Murphy, P., Cloitre, M., Bisson, J., Roberts, N., Shevlin, M., ... & Mason-Roberts, S. (2019). Psychological interventions for ICD-11 complex PTSD symptoms: systematic review and meta-analysis. Psychological medicine, 1-15.
Roberts, N. P., Kitchiner, N. J., Kenardy, J., & Bisson, J. I. (2009). Multiple session early psychological interventions for the prevention of post traumatic stress disorder. Cochrane Database of Systematic ‐ Reviews, (3).
Hackmann, A., Ehlers, A., Speckens, A., & Clark, D. M. (2004). Characteristics and content of intrusive memories in PTSD and their changes with treatment. Journal of Traumatic Stress: Official Publication of The International Society for Traumatic Stress Studies, 17(3), 231-240.
McFarlane, A. C. (1992). Avoidance and intrusion in posttraumatic stress disorder. Journal of Nervous and Mental Disease.
Ehlers, A., Hackmann, A., & Michael, T. (2004). Intrusive re experiencing in post traumatic stress ‐ ‐ disorder: Phenomenology, theory, and therapy. Memory, 12(4), 403-415.
Brewin, C. R. (2015). Re-experiencing traumatic events in PTSD: New avenues in research on intrusive memories and flashbacks. European Journal of Psychotraumatology, 6(1), 27180.
Kuhlmann, S., & Wolf, O. T. (2006). Arousal and cortisol interact in modulating memory consolidation in healthy young men. Behavioral neuroscience, 120(1), 217.
Stawski, R. S., Sliwinski, M. J., & Smyth, J. M. (2009). The effects of an acute psychosocial stressor on episodic memory. European Journal of Cognitive Psychology, 21(6), 897-918.
Eliot, R. S. (1988). Stress and the heart.
Berntson, G. G., Cacioppo, J. T., Binkley, P. F., Uchino, B. N., Quigley, K. S., & Fieldstone, A. (1994). Autonomic cardiac control. III. Psychological stress and cardiac response in autonomic space as revealed by pharmacological blockades. Psychophysiology, 31(6), 599-608.
Vrijkotte, T. G., Van Doornen, L. J., & De Geus, E. J. (2000). Effects of work stress on ambulatory blood pressure, heart rate, and heart rate variability. Hypertension, 35(4), 880-886.
Schlotz, W. (2013). Stress reactivity. In Encyclopedia of behavioral medicine (pp. 1891-1894). Springer. Minassian, A., Maihofer, A. X., Baker, D. G., Nievergelt, C. M., Geyer, M. A., & Risbrough, V. B. (2015). Association of predeployment heart rate variability with risk of postdeployment posttraumatic stress disorder in active-duty marines. JAMA psychiatry, 72(10), 979-986.
Morris, M. C., Hellman, N., Abelson, J. L., & Rao, U. (2016). Cortisol, heart rate, and blood pressure as early markers of PTSD risk: A systematic review and meta-analysis. Clinical psychology review, 49, 79-91.
Morris Jr, J. A., Norris, P. R., Ozdas, A., Waitman, L. R., Harrell Jr, F. E., Williams, A. E., ... & Jenkins, J. M. (2006). Reduced heart rate variability: an indicator of cardiac uncoupling and diminished
physiologic reserve in 1,425 trauma patients. Journal of Trauma and Acute Care Surgery, 60(6), 1165-1174.
Buckley, T. C., Holohan, D., Greif, J. L., Bedard, M., & Suvak, M. (2004). Twenty-four-hour ambulatory assessment of heart rate and blood pressure in chronic PTSD and non-PTSD veterans. Journal of traumatic stress, 17(2), 163-171.
Myin-Germeys, I., & van Os, J. (2007). Stress-reactivity in psychosis: evidence for an affective pathway to psychosis. Clinical psychology review, 27(4), 409-424.
Van Winkel, M., Nicolson, N. A., Wichers, M., Viechtbauer, W., Myin-Germeys, I., & Peeters, F. (2015). Daily life stress reactivity in remitted versus non-remitted depressed individuals. European Psychiatry, 30(4), 441-447.
Sänger, J., Bechtold, L., Schoofs, D., Blaszkewicz, M., & Wascher, E. (2014). The influence of acute stress on attention mechanisms and its electrophysiological correlates. Frontiers in behavioral neuroscience, 8, 353.
Schwabe, L., Haddad, L., & Schachinger, H. (2008). HPA axis activation by a socially evaluated cold-pressor test. Psychoneuroendocrinology, 33(6), 890-895.
van Zuiden, M., Kavelaars, A., Geuze, E., Olff, M., & Heijnen, C. J. (2013). Predicting PTSD: pre-existing vulnerabilities in glucocorticoid-signaling and implications for preventive interventions. Brain,
Behavior, and Immunity, 30, 12-21.
Pineles, S. L., Hall, K. A. A., & Rasmusson, A. M. (2017). Gender and PTSD: different pathways to a similar phenotype. Current opinion in psychology, 14, 44-48.
de Beurs, E., Van Dyck, R., Marquenie, L. A., Lange, A., & Blonk, R. W. (2001). De DASS: een vragenlijst voor het meten van depressie, angst en stress. Gedragstherapie, 34(1), 35-54.
Gray, M. J., Litz, B. T., Hsu, J. L., & Lombardo, T. W. (2004). Psychometric properties of the life events checklist. Assessment, 11(4), 330-341.
Henry, J. D., & Crawford, J. R. (2005). The short form version of the Depression Anxiety Stress Scales ‐ (DASS 21): Construct validity and normative data in a large non clinical sample. British journal of ‐ ‐ clinical psychology, 44(2), 227-239.
Blevins, C. A., Weathers, F. W., Davis, M. T., Witte, T. K., & Domino, J. L. (2015). The posttraumatic stress disorder checklist for DSM 5 (PCL 5): Development and initial psychometric evaluation. Journal ‐ ‐ of traumatic stress, 28(6), 489-498.
Weidmann, A., Conradi, A., Gröger, K., Fehm, L., & Fydrich, T. (2009). Using stressful films to analyze risk factors for PTSD in analogue experimental studies–which film works best?. Anxiety, Stress, & Coping, 22(5), 549-569.
Jacqmin-Gadda, H., Sibillot, S., Proust, C., Molina, J. M., & Thiébaut, R. (2007). Robustness of the linear mixed model to misspecified error distribution. Computational Statistics & Data Analysis, 51(10), 5142-5154.
Sharpley, C. F. (1994). Differences in pulse rate and heart rate and effects on the calculation of heart rate reactivity during periods of mental stress. Journal of Behavioral Medicine, 17(1), 99-109.
Quaedflieg, C. W. E. M., Van De Ven, V., Meyer, T., Siep, N., Merckelbach, H. L. G. J., & Smeets, T. (2015). Temporal dynamics of stress-induced alternations of intrinsic amygdala connectivity and neuroendocrine levels. PloS one, 10(5).
De Choudhury, M., Monroy-Hernandez, A., & Mark, G. (2014, April). " Narco" emotions: affect and desensitization in social media during the mexican drug war. In Proceedings of the SIGCHI conference on Human Factors in Computing Systems (pp. 3563-3572).
Mrug, S., Madan, A., Cook, E. W., & Wright, R. A. (2015). Emotional and physiological desensitization to real-life and movie violence. Journal of youth and adolescence, 44(5), 1092-1108.
Cheung, J., Chervonsky, L., Felmingham, K. L., & Bryant, R. A. (2013). The role of estrogen in intrusive memories. Neurobiology of learning and memory, 106, 87-94.
