All patients referred to the psychiatric outpatient clinic 15 at Södra Älvsborgs Sjukhus between August and December 2021 were assessed with a test version of the International Trauma Interview 3.2 (ITI) (17) for diagnosis of PTSD and C-PTSD. Inclusion criteria were: C-PTSD according to ICD-11, PTSD with at least one comorbid disorder requiring psychiatric care or PTSD with previous unsuccessful trauma-focused treatment. Exclusion criteria were: insufficient knowledge of the Swedish language to complete questionnaires, ongoing substance abuse, use of benzodiazepines, acute suicidality, or dissociative identity disorder (DID). No patients were excluded on these criteria during the inclusion phase. Of 29 eligible patients, 17 declined to participate in the 8-day ITP. 12 patients accepted the ITP. The final sample was 11 participants, as one participant dropped out on day 4 due to current stress and increased risk factors in the home environment during treatment. This participant did not attend posttreatment evaluation and was not included in the data analysis.

All procedures were approved by the Ethics Committee in Lund on 2021-06-15. Informed consent was obtained before inclusion in the study. All patients who met the inclusion criteria obtained information about the study verbally and in writing. Accepting participation was not necessary to receive treatment.

Screening for eligibility was the first step. Data collection started if the participant met the inclusion criteria, accepted to participate in the study and signed the informed consent (see Table 1). Between the last day of the treatment program and 3-month follow-up, no further interventions towards PTSD were initiated. Two participants did not complete measures of symptoms for multiple sessions (4 and 5, respectively) as they ended treatment prematurely due to reduced symptoms. The two participants completed posttreatment evaluation. Self-reported data were collected using REDCap (Research Electronic Data Capture) (18, 19), an electronic data capture tool hosted at Gothia Forum REDCap (RRID: SCR_003445). Participants entered symptom ratings via a QR code or web link. Participants who were uncomfortable using electronic devices were offered paper versions of self-report measures.

The primary outcome measure was the Post Traumatic Checklist for DSM-5 (PCL-5) (20), a self-report questionnaire for PTSD symptoms. PCL-5 has shown good internal reliability (0.96) and test–retest reliability (21), with good sensitivity to clinical change (22). The International Trauma Questionnaire (ITQ) (23) is a questionnaire in two parts developed to measure PTSD symptoms and Disturbance in Self Organization (DSO) symptoms (associated with C-PTSD). ITQ has shown high internal reliability (0.89 to 0.94) (24). Life Events Checklist for DSM-5 (LEC-5) (25) was used to identify traumatic events. LEC-5 has shown varying test–retest reliability between items (kappa = 0.36–0.80) (26).

To assess diagnostic criteria for PTSD and C-PTSD, a test version of the International Trauma Interview 3.2 (ITI) (17) was used. ITI has satisfying interrater reliability (α = 0.76) (27). Beck’s scale for suicide ideation (SSI) (28) in Swedish translation was used to screen high levels of suicidal intentions at pretreatment and to assess change in suicidal risk during the treatment. SSI has shown good internal reliability (0.89) (29). The Swedish translation of Dissociative Experience Scale II (DES) (30) was used to screen dissociative symptoms. DES II has good internal reliability (0.87) (30). Structured Clinical Interview for DSM-IV (R) Dissociative Disorders (SCID-D) (31) is a diagnostic interview for dissociative syndromes and was used to identify diagnosis of DID if scores on DES II indicated pathological dissociation. SCID-D has shown good interrater reliability (kappa = 0.72) and test–retest reliability (kappa = 0.88) (32).

The 8-day ITP included PE, EMDR, physical activity and psychoeducation, in accordance with the treatment program developed at PSYTREC and described by Van Woudenberg et al. (11) During each day of treatment, the participant had 90 min of PE (6) in the morning and 90 min of EMDR (5) in the afternoon. Both treatments were slightly modified to adapt to the intensive format. In vivo exposure between sessions was removed from PE and stabilization phase prior to trauma processing in EMDR was removed. Prior to and after each PE and EMDR session, participants had 30–60 min of physical activity guided by a physiotherapist. The purpose of the physical activity was to help participants to increase body awareness, regulate physiological tension and counteract rumination and passive re-experiencing of trauma. The physical activities included high-pulse exercise and relaxation, e.g., boxing, medicine-ball exercise, tai-chi, mindful forest walk. In the evening, psychoeducation about PTSD and coping strategies was given in a group format. The treatment schedule is summarized in Table 2. Participants stayed at the hospital hotel at Södra Älvsborgs Sjukhus during the treatment, where all meals were served. Participants had the possibility to use telephone coaching outside sessions during the treatment. Therapist rotation was used for both PE and EMDR, with the assumption that less focus on the relationship between patient and therapist would facilitate increased focus on processing traumatic memories (33). Typically, participants met 8 different therapists during the treatment. The therapists were all trained in PE or EMDR, or both methods. Information on treatment progress was reported and discussed between therapists at a 30-min meeting every morning.

All statistical analyzes were performed with IBM SPSS for Windows (version 28). The authors were blinded to the identity of the participants during data management procedures. Data were screened for data-entry errors and outliers, and assumptions associated with planned statistical analysis were checked via explore function in SPSS. Means, standard deviation and frequency for demographic and clinical variables in the sample were calculated. The total missing data was 6.5%. No systematic pattern could be found in the missing data. Imputation was used for missing data, the value for the previous measure was repeated. To determine whether the intervention brought a change in PTSD symptoms (measured by PCL-5) and DSO symptoms (measured by ITQ-part 2) from pre-to posttreatment and 3-month follow-up, a repeated measures analysis of variance (ANOVA) was used, two-tailed using a 95% level of significance. Two effect size measures were calculated: partial eta squared and Cohens’d for paired samples. To determine to which degree the intervention led to a clinically meaningful change at posttreatment and 3-month follow-up on primary outcome measure (PCL-5), two definitions were used. National Center for PTSD (NCPTSD) (34) recommend 10 points as minimum change. Marx et al. (22) recommend a PCL-5 value of <28 points at posttreatment to measure of clinically meaningful treatment effect. The proportion of participants reaching thresholds at pre-, posttreatment and 3-month follow-up was calculated. The total value of SSI items 4 and 5 during the interview was used to determine whether suicidal intentions changed during the intervention. Friedman’s test was used to compare median values pretreatment, midtreatment, posttreatment and 3-month follow-up.

Of 11 participants, 83% were female and 17% male. The mean age was 26,2 (sd 7,73). 72% were living with their spouse or primary family, 28% in a single household. 36% were employed, and 18% had a temporary employment. Mean year of education was 11,45 (sd 1,82). 27% of the participants met the criteria for C-PTSD diagnosis. Sexual violence was the most common traumatic event (92%), followed by physical violence (82%). For both sexual and physical violence, the traumatic event was experienced before age 13 in 45% of the cases. 45% had a history of previously unsuccessful trauma-focused treatment. Nine participants completed the whole treatment. Two participants ended treatment earlier than expected due to reduced PTSD symptoms and fully processed trauma memories, one participant on day 2 and one participant on day 3. Both participants completed posttreatment evaluation and were included in the analysis.

The mean and standard deviation for PTSD symptoms (PCL-5) and DSO symptoms (ITQ-part 2) are presented in Table 3. The ANOVA of PTSD symptoms indicated a significant change over time (Wilks Lambda = 0.18, F (df1 = 2, df2 = 9) = 19.96, p < 0.001, partial eta squared = 0.82). The ANOVA of DSO symptoms (Table 3) indicated a significant change over time (Wilks Lambda = 0.44, F (df1 = 2, df2 = 9) = 5,75, p < 0.025, partial eta squared = 0.56). Effect sizes comparing pretreatment to posttreatment and 3-month follow-up respectively, were large for both PTSD and DSO symptoms. 91% of participants who completed treatment reached the threshold for clinically meaningful change in PTSD symptoms at posttreatment and 100% at 3-month follow-up. Using alternative definitions of clinically meaningful change in PTSD symptoms recommended by Marx et al. (22), 72% reached the threshold at posttreatment and 3-month follow-up. At posttreatment, 91% of participants completing treatment no longer meet the criteria for PTSD according to ICD-11. No participants met the criteria for C-PTSD. At 3-month follow-up, none of the participants met the criteria for PTSD.

In the daily PCL-5 scores (Figure 1), no symptom exacerbation could be observed at group level. At group level, no part of the treatment process could be associated with worsening or breach of a positive trend in PCL-5 scores throughout the treatment.

One of the participants (9%) who completed treatment (n = 11) had an increased score (+10 points) on PCL-5 at posttreatment compared to pretreatment. The same participant, however, had a lowered score on PCL-5 at 3-month follow-up and no longer met the criteria for PTSD. Regarding change in suicidal intentions during treatment, Friedmans test (0.05 significance level) indicated no statistically significant difference between any of the measuring points (pretreatment, day 4, posttreatment, 3-month follow-up) (x² (df = 3, n = 11) = 6,000, p = 0.112). One participant used the coaching telephone on two occasions outside office hours to manage anxiety, this participant completed treatment.

Both due to design and small sample size, this feasibility study has several limitations regarding which conclusions can be drawn. The small sample reduces the possibility of generalizing the results to patients seeking healthcare due to severe PTSD. No control group was used, which make conclusions regarding the actual effect of the treatment impossible. Apart from studies with larger sample sizes with control group, there is also a need to examine the effect of the 8-day ITP in patients diagnosed with C-PTSD in Swedish public healthcare, as only three patients diagnosed with C-PTSD were included in this study.

The results indicate that an 8-day ITP could be a useful treatment for severe PTSD in a Swedish public healthcare setting. A positive change in symptom levels was observed and no adverse events or increase in suicidal ideation was detecded. Based on these results, we conclude that support for the 8-day ITP as a safe form of treatment has grown even further. Therefore we believe this form of treatment should be further explored as an option to traditionally paced treatments in public healthcare, with larger sample size and further exploration of how the presence of C-PTSD interacts with treatment effects. Although the short time in which substantial symptom relief can be achieved is appealing in itself, there is also a need for further exploration of potential advantages or disadvantages of the 8-day ITP by comparing treatment effects directly to traditionally paced PTSD treatments.