This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Formative Research, is properly cited. The complete bibliographic information, a link to the original publication on https://formative.jmir.org, as well as this copyright and license information must be included.
The COVID-19 pandemic has accelerated the worldwide need for simple remotely delivered (digital) scalable interventions that can also be used preventatively to protect the mental health of health care staff exposed to psychologically traumatic events during their COVID-19–related work. We have developed a brief behavioral intervention that aims to reduce the number of intrusive memories of traumatic events but has only been delivered face-to-face so far. After digitalizing the intervention materials, the intervention was delivered digitally to target users (health care staff) for the first time. The adaption for staff’s working context in a hospital setting used a co-design approach.
The aims of this mixed method exploratory pilot study with health care staff who experienced working in the pandemic were to pilot the intervention that we have digitalized (for remote delivery and with remote support) and adapted for this target population (health care staff working clinically during a pandemic) to explore its ability to reduce the number of intrusive memories of traumatic events and improve related symptoms (eg, posttraumatic stress) and participant’s perception of their functioning, and to explore the feasibility and acceptability of both the digitalized intervention and digitalized data collection.
We worked closely with target users with lived experience of working clinically during the COVID-19 pandemic in a hospital context (registered nurses who experienced intrusive memories from traumatic events at work; N=3). We used a mixed method design and exploratory quantitative and qualitative analysis.
After completing the digitalized intervention once with remote researcher support (approximately 25 minutes) and a brief follow-up check-in, participants learned to use the intervention independently. All 3 participants reported zero intrusive memories during week 5 (primary outcome: 100% digital data capture). Prior to study inclusion, two or more intrusions in the week were reported preintervention (assessed retrospectively). There was a general pattern of symptom reduction and improvement in perceived functioning (eg, concentration) at follow-up. The digitalized intervention and data collection were perceived as feasible and rated as acceptable (eg, all 3 participants would recommend it to a colleague). Participants were positive toward the digital intervention as a useful tool that could readily be incorporated into work life and repeated in the face of ongoing or repeated trauma exposure.
The intervention when delivered remotely and adapted for this population during the pandemic was well received by participants. Since it could be tailored around work and daily life and used preventatively, the intervention may hold promise for health care staff pending future evaluations of efficacy. Limitations include the small sample size, lack of daily intrusion frequency data in the week before the intervention, and lack of a control condition. Following this co-design process in adapting and improving intervention delivery and evaluation, the next step is to investigate the efficacy of the digitalized intervention in a randomized controlled trial.
The mental health of health care staff exposed to stressful and traumatic events during their work in the COVID-19 pandemic is a number one research priority internationally [
Here, we focus on a brief intervention to target one focal symptom that can arise after exposure to traumatic events—intrusive memories [
Exposure to psychologically traumatic events presents a problem for health care staff working in the pandemic and will continue to be a problem once the pandemic is over. Before the pandemic, experiencing different traumatic events (either as a direct threat to themselves or a witnessed threat to patients) could lead to mental health difficulties such as PTSD in health care staff [
Interventions to support health care staff need to be suitable for delivery in their work context. This includes being suitable for ongoing exposure to stressful events in the line of work (ie, repeatable); preventative use to keep staff well and working (ie, prevent the buildup of symptoms); scalability, simplicity, and brevity; and remote delivery to reduce the risk of virus transmission during a pandemic. The development of new interventions requires adaptation and detailed feedback from people with lived experience [
Good evidence-based treatments for PTSD exist (eg, trauma-focused cognitive behavioral therapy and eye movement desensitization and reprocessing [
The procedure takes approximately half an hour and is delivered according to a clear protocol and administered typically in one or two guided sessions (thereafter self-administered, if needed). Delivery currently requires guided support by someone trained in the intervention but does not always require a fully qualified mental health professional. The intervention consists of several components including a brief memory reminder to moments within the trauma (hotspots), training in mental rotation, and engaging in a visuospatial cognitive task (the computer game Tetris) for a specific time while actively using mental rotation (ie, planning ahead and visualizing in the mind’s eye how to rotate and move upcoming Tetris blocks to fit them into a horizontal line). The cognitive task aspect can be delivered on the participant’s own smartphone. This intervention does not require a detailed discussion of the traumatic event.
The rationale underlying the intervention includes the following. Engaging in the visuospatial task is hypothesized to compete for limited working memory resources [
The intervention can be used on the same day the traumatic event occurred (day 1 protocol) [
Early studies have shown that the intervention may prevent intrusive memory occurrence in patients soon after trauma [
To date the intervention has been delivered in person. So that it can be delivered remotely, we have first taken steps to digitalize the instructions for this brief intervention [
As there is currently no gold standard for a systematic stepwise approach to use for developing a successful intervention, it has been recommended to collect and apply relevant steps suggested from different methodologies, to use them with a “flexible” approach [
The aims of this mixed methods exploratory pilot study with health care staff with experience of working in the pandemic were to:
Pilot the intervention procedures that we have digitalized (for remote delivery and with remote support) and adapted for this target population (health care staff working during a pandemic) to explore its ability to reduce the number of intrusive memories of trauma (primary outcome: week 5 diary post intervention) and improve related symptoms (eg, posttraumatic stress symptoms) and participant’s perception of their functioning (eg, concentration)
Explore the feasibility and acceptability of both the digitalized intervention procedures and the digitalized data collection (eg, primary outcome measure)
Participants (N=3, all female) were Swedish registered nurses, all with a university education and in full-time employment, who had worked clinically in the ICU or ambulance service during the COVID-19 pandemic in the spring of 2020 and were still currently working. They were all around 50 years of age and had >30 years of experience in their work. They had specialized in anesthesia and intensive care or ambulance care. To protect participant anonymity, most demographic characteristics have been omitted.
The inclusion criteria were being 18 years or older; doing clinical work during the COVID-19 pandemic in hospital care facilities (eg, ICU, intermediate care, or ward), experiencing at least one traumatic event in relation to their clinical work as health care staff during the COVID-19 pandemic that met Diagnostic and Statistical Manual of Mental Disorders (Fifth Edition; DSM-5) [
Details of the traumatic events causing intrusive memories were assessed with a bespoke item (“Select which of the following category(s) best fits for the traumatic event(s) you have experienced during COVID-19 within the last three months and experience intrusive memories from”) and then 11 categories (including a category for
Characteristics of the worked-related traumatic events that participants had intrusive memories from and prior experiences of traumatic events reported per participant.
Variable | Participanta | Participant | Participant |
Time(s) of trauma | Between 1 and 3 months ago | Between 1 and 3 months; within the last 24 hours | Between 1 and 3 months ago; ongoing exposure |
Traumatic event causing intrusive memories | Traumatic or tragic patient death; situation in which patient care did not work as planned | Traumatic or tragic patient death | Situation in which patient care did not work as planned; confronted with distressed family members of patients; |
Retrospectively rated number of intrusions in the week before study participationb | 2 | 3 | 3 |
Prior experiences of traumatic event types (LEC-5c,d) | 6 | 12 | 10 |
aParticipant numbers are omitted to preserve anonymity.
bAssessed via a single item retrospective rating.
cLEC-5: Life Events Checklist for Diagnostic and Statistical Manual of Mental Disorders (5th Edition).
dA list of 17 traumatic event types. Here, we report the number of event types endorsed as “happened to me,” “witnessed it,” or “part of my job” [
The number of intrusive memories of the traumatic events was assessed with a digital adaptation of the pen-and-paper diary used in our previous work [
The number of intrusive memories of the traumatic events was also assessed in week 1 (starting after having received the intervention on day 1) in another identical 7-day digital diary.
The intrusion questionnaire (eg, [
The intrusion and avoidance subscales of the Impact of Event Scale-Revised (IES-R) [
Participants’ current symptoms of PTSD were assessed via the PTSD Checklist for DSM-5 (PCL-5) short version [
Two self-rated items assessed participants’ level of distress and vividness associated with the intrusions (11-point scales from 0
One item was used to assess how many intrusive memories the participant had experienced in the week prior to entering the study, from 2 to more than 7. If more than 7, a free-text response field to specify the number of intrusive memories was presented. This was followed by three self-rated items measuring the level of distress, vividness, or concentration disruption associated with the intrusions (11-point scales from 0
Participants rated their perceived level of concentration disruption associated with intrusions with a bespoke item adapted from Holmes et al [
A bespoke item was used to assess their perceived impact on daily functioning associated with the intrusions (“During the previous month how much did your intrusive memories of the traumatic event affect your functioning (social, occupational, or other important areas e.g. relationships with other people, work, parenting, schoolwork, housework, volunteer work etc.)?”) [
Before the intervention, participants completed the Credibility/Expectancy Questionnaire [
Subjective units of distress (SUD) were collected three times during the intervention process to measure participants’ level of distress (11-point scales from 0
Participants perceived coping during the COVID-19 pandemic was assessed via two free-text response field questions (eg, are there any specific factors you think have made it more difficult or easier for you to handle the COVID-19 situation and its consequences?).
Adverse events [
A feedback questionnaire consisting of nine bespoke items assessed participant’s experience of study participation. Items included (eg, “How acceptable was it to do the task?”; 11-point scale from 0
Participants were all recruited from the professional network of the research team (author AR). Data collection occurred from July 8 to August 28, 2020. Participants received study information materials via email and postally, and provided their written and informed consent prior to study procedures. They were contacted by the researcher (author MK) to set up a time for a remote digital meeting using Zoom [
The intervention was administered fully remotely and digitally. Remote researcher support during the procedure and delivery of the intervention was provided by a clinical psychologist with extensive experience in delivering the intervention to clinical participants (MK; see description of training in the Training to Deliver the Intervention section). During the remote meeting, participants received links to each step of the study (baseline measures; intervention package), and researchers (authors MK and KD) observed and took notes on feedback and questions that arose. The baseline and intervention session took approximately 1 hour in total (see
Flow of study procedures. Note that, in contrast to the planned follow-up randomized controlled trial, for these pilot participants, participation ended after the week 5 diary was complete.
The intervention procedure commenced after the completion of baseline measures (see
Next, a video and written instructions followed on how to play Tetris with mental rotation and how to access the game (via Tetris.com) and adjust the necessary settings (
After the intervention, participants were encouraged to keep using the intervention as needed and in a way that would fit with their work demands. They received instructions for completing an online daily diary registration on the number of intrusive memories over week 1 (see the Secondary Outcome Measures section). Incoming data was monitored approximately daily for the number of intrusive memories, and whenever intrusive memories were reported, participants had the option for
At the 1-week and 1-month follow-ups, participants completed self-report questionnaires via the electronic platform. After the 1-month follow-up, they commenced with the primary outcome—daily registrations on the number of intrusive memories during week 5. Participants received an end-of-study letter that included a graph depicting their change in the number of intrusions over time. They were invited to a phone call with a researcher (MK) for additional feedback on study procedures to help refine these prior to starting the main randomized controlled trial (RCT).
The researcher delivering the intervention (MK) had received prior detailed training in delivering the intervention and how to obtain the primary outcome. This included theoretical and procedural knowledge; observing and being observed by a supervisor (author EAH) in role plays; receiving in-vivo and group supervision; and in this study, receiving regular ongoing supervision (from EAH) as necessary.
We conducted an exploratory analysis of quantitative results using a descriptive approach. Analyses and graphs were performed in Excel Professional Plus Version 16.0.5065.1000 (Microsoft Corporation).
A flexible approach building on content analysis as outlined by Bengtsson [
Participant feedback was systematically evaluated to determine whether changes should be implemented, following the suggested person-based approach created by Yardley et al [
This study was approved by the Swedish Ethical Review Board before the start of the study: 2020-03085. All participants provided their written and informed consent in accordance with the guidelines of the Declaration of Helsinki; signed consent forms were returned digitally and on paper. These pilot cases were done as a precursor to the planned RCT registered under ClinicalTrials.gov NCT04460014 (July 7, 2020).
Two participants experienced one recurring visual intrusive memory of a work-related traumatic event, and 1 experienced three separate recurring intrusive memories (ie, different visual scenes). Examples of these intrusive memories’ content included an image of patient’s faces, image of patient’s relatives, or image of a dead person (see
All 3 participants completed all study procedures. In the past, we had used paper diaries, and here, the daily digital data capture of the number of intrusive memories via an electronic link (SMS text message or email) worked well (ie, the digital diary was completed by all 3 participants, and the feedback was positive). Our 3 participants answered 100% of the intrusive memory diary links, which were sent out four times a day for 7 days in a row, twice (primary outcome: week 5; secondary outcome: week 1 post intervention). The digital data capture for other questionnaires (secondary outcomes) was also successful (ie, questionnaires were completed by participants), with only two ratings missing for 1 participant (intrusion questionnaire: unwanted memory frequency and concentration disruption on day 2).
In contrast to baseline and week 1 postintervention levels (see the Secondary Outcomes: Number of Intrusive Memories [Week 1] section), all 3 participants reported 0 intrusive memories throughout the week 5 diary (
Graphs for visual inspection of the number of intrusive memories of trauma (sum for the week) during week 1 (secondary outcome) and week 5 (primary outcome) following the initial intervention session for each participant. After learning how to do the intervention (guided by the researcher) and practicing it independently during week 1, all 3 participants reported zero intrusions by week 5.
Bar graphs showing examples of intrusion-related symptoms (posttraumatic stress) and participant’s perception of their functioning in terms of concentration disruption (all secondary and other outcomes). All 3 participants reported reductions in intrusion (a) and avoidance (b) symptoms on the Impact of Event Scale-Revised from baseline to week 1 and the 1-month follow-up. Participant’s total scores on the Posttraumatic Stress Disorder Checklist for DSM-5 (c) decreased from baseline to week 1 and the 1-month follow-up. All 3 participants reported that their perceived concentration disruption associated with having an intrusive memory (d) decreased from baseline to week 1 and the 1-month follow-up. DSM-5: Diagnostic and Statistical Manual of Mental Disorders (5th Edition).
In week 1, participants reported 9 (P1), 7 (P2), and 4 (P3) intrusive memories in total. Self-reported diary accuracy ratings were high for P1 and P3 (10/10), and medium for P2 (4/10).
We discuss lessons learned during the first week on a case-by-case basis. For all 3 participants, the intervention was delivered once with researcher support (approximately 25 minutes), and they subsequently used the intervention on their smartphone on their own. Participants received encouragement about intervention use from the researchers via 1 to 4 SMS text messages or emails (all participants) and one phone call (1 participant).
P1 reported 1 intrusive memory in the first half of the week and then a spike in intrusive memories on day 5 (7 intrusive memories). In communication with the researchers, she described that on this day she experienced an event at work similar to the one represented in her intrusive memory, which triggered these intrusions. She felt too tired to use the intervention straight after that work shift but successfully used it for that intrusion the day after.
P2 completed the game play part of the intervention on her computer on day 1. She reported 4 intrusive memories on the same day and 3 more on day 2. On day 2, she was instructed over the phone to repeat the intervention and encouraged by researchers to use her smartphone instead and to bring the image to mind before the game play. From day 3 to day 7 she reported no more intrusive memories.
P3 reported 2 intrusive memories on day 1 and 1 each on day 2 and day 3. From day 4 to day 7 she was intrusion-free, thus showing a general decrease of intrusion frequency over the week. She kept using the intervention during week 1 and wrote that she was “incredibly impressed by how such a simple thing could make such a huge difference.”
From baseline to week 1, the frequency of intrusive or unwanted memories of the traumatic event during the previous week slightly decreased from
From baseline to 1 month, participants’ scores on the intrusion subscale decreased (from a range of 8-15 to a range of 0-2); similarly on the avoidance subscale, all scores decreased to 0 after 1 month (
From baseline to week 1, total scores on the PCL-5 decreased for all 3 participants and remained low at 1 month (range 0-1;
From baseline to week 1, vividness and distress associated with the intrusive memories decreased. P1 showed a strong decrease of vividness (7 to 3) and distress (7 to 2) from baseline to week 1, P2 showed a slight decrease from baseline to week 1 (vividness 10 to 9, distress 5 to 4), and P3 showed a strong decrease of vividness (10 to 2) and a slight decrease in distress (1 to 0) from baseline to week 1. At 1 month, none of our 3 participants reported intrusions anymore, thus related vividness or distress was not rated.
Here we present outcomes regarding functioning and intrusive memories as used in our previous work [
At baseline, concentration disruption associated with having an intrusive memory was medium to high (5 and 7 out of 10) and reported as lasting for 1 to 5 minutes per intrusive memory. At 1 month, concentration disruption had decreased to 0 or 1 out of 10 (
From baseline to 1 month post intervention, 2 participants reported that the impact of intrusive memories on functioning decreased (from 5 [P1] or 2 [P3] out of 10 at baseline to 0 at 1 month), and it remained low for P2 (1 out of 10 at baseline and 1 month).
Credibility ratings taken after a brief description of the intervention prior to engaging in it were low to midrange for all 3 participants (38 out of 50 by P1, 35 by P2, and 21 by P3), and this was also reflected in qualitative feedback (eg, P3 wrote at the 1-month follow-up, “...[I] didn’t believe at all that this would work!”).
During the intervention session with the researcher, participants completed distress ratings (SUD) before and after describing their intrusive memories, and after playing Tetris. For 2 participants, distress increased (P1 from 2 to 6; P3 from 0 to 1) but did not reach ceiling after describing and choosing an intrusion to target with the intervention, indicating successful emotional memory activation (note, P2’s ratings decreased from 2 to 0). Critically, after game play, distress decreased or remained at zero for all 3 participants (P1: 1; P2: 0; P3: 0).
In the end of the intervention session, we asked participants if any specific situations or factors made it difficult to cope with the COVID-19 situation at work and which made it easier to cope. They noted that, for example, support from the employer and help and support from colleagues made it easier to cope, and a lack of knowledge about COVID-19, lack of competence, difficulties in accessing and using personal protective equipment, and not being able to give person-centered care made it more difficult.
There were no reported adverse events within the platform at week 1 and the 1-month follow-up.
The digitalized intervention (approximately 25 minutes) was delivered with the option for researcher support (via video call). There was no dropout in our 3 participants. The session with the researcher and subsequent use of the intervention was flexibly scheduled into participant’s daily work and life. After this initial contact, limited researcher support was needed and none of the 3 participants requested additional support on how to use the intervention. Only 1 participant received a phone call on day 2; for the others, only brief encouragement to use the intervention via SMS text message or email was used.
All primary and secondary outcome data was successfully collected digitally or remotely. Digital procedures (eg, SMS text message) for reminding participants to fill in the daily registration on intrusions were used if needed.
In a feedback questionnaire completed at 1 month, the intervention task was rated by participants 1 and 3 to be easy, and low in terms of how upsetting it was (see
Responses on feedback questionnaire about participation from each participant at 1 month.
Feedback questionnaire about participation | P1 | P2 | P3 |
“If a friend or colleague has gone through a similar event, how likely is it that you would recommend this task?”a | 10 | 10 | 10 |
“How easy did you find it to do the task?”a,b | 9 | 1 | 10 |
“How upsetting did you find it to do the task?”b | 1 | 0 | 0 |
“How much did you appreciate having something to do?”b | 9 | 1 | 2 |
“How acceptable did you find the task?”b | 9 | 10 | 10 |
“Received other treatment due to the traumatic event”c | No | No | No |
“Did task on their own (i.e., after the researcher-led session)”c | Yes | Yes | Yes |
“How often?”c | 3-4 | 8 | Daily for the first 2 weeks, after that every now and then |
“Have you talked to others (e.g. friends, colleagues) about the task?”d | Yes | Yes | Yes |
aRated on an 11-point scale from 0 (not at all) to 10 (extremely/very much).
bIn this study,
cFree-text response field.
dYes/no response.
In the open-ended feedback collected within the platform, the 3 participants expressed overall positive feedback about their outcomes after the intervention, that their experience with using the platform was overall good, and that they found the instructions and videos to be clear and helpful. P3 wrote as part of the open-ended feedback question given at the 1-month follow-up that the intervention was “excellent” in her case. P1 wrote in the platform both at the 1-week and 1-month follow-ups that she had not experienced any more intrusive memories since she had used the intervention on her own in the beginning of week 1. She noted that even when choosing to talk about the event of the memory with her colleagues she no longer experienced intrusive memories. Additionally, when contacted for a follow-up via phone call, participants reported that seeing their results gave them a sense of self-achievement (P1) and that they have learned a tool that they can use in future situations (P3).
In terms of impact on work functioning, P1 reported in the follow-up phone call that when she had intrusive memories it affected her empathy and her interactions with patients and relatives: “I did not dare to let them in, I was afraid something similar would happen and I would get more intrusive memories.”
We systematically analyzed participant feedback for changes or improvement, and three overarching themes emerged:
Getting the data right: This refers to minor changes in data collection (eg, clarifying the wording to some questions, removing items that were poorly understood, and adding a baseline diary; ensure quality of data).
Doing the digital intervention right: This refers to any changes in instructions, adding more information to further emphasize the important aspects, and for tech or intervention procedure (ensure intervention fidelity).
Feeling that study participation is alright: This refers to tailoring study procedures (eg, adding examples of staff categories in questions or examples) and adapting data collection as necessary (ensure participants feel included, can relate, and that participant burden is minimized).
Based on participant feedback, 18 changes were made (see
Some changes that were suggested were not implemented, such as altering the original or translated wording of an official or pre-established questionnaire.
Further, for the daily registration of intrusive memories, 1 participant said that it was easy to press the incorrect option indicating how many intrusions she had (eg, P1 accidently pressed “1” when she meant to press “0” for the week 5 diary). Overall, even though the digital intrusive memory registration was said to be a feasible and acceptable data collection method, there is a need for researchers to closely monitor and check in to verify the accuracy, as a participant can easily select the wrong option without being able to edit their response. Furthermore, we decided to take out the question about “diary accuracy,” as low responses given by some of the participants were due to confusion on what this really meant (ie, how precisely did they count their work-related intrusions or how well or how swiftly they entered their data into the platform after having received the link).
Examples of changes made to study materials and procedure following participant feedback [
Theme | Participant feedback as expressed by one or more participants | Changes made to the study based on feedback |
Doing the digital intervention right | It is unclear what the next step is after filling out the baseline questionnaires, and after completing the assigned task. “ |
Added flowcharts that show the participant journey at the beginning and end of each module link that is sent out. It includes a green arrow with the text “You are here” to demonstrate where they are in the study and the next step. |
Getting the data right | It is difficult to remember exactly how many intrusions one had in the last week (for the retrospective rating at baseline). (P1) | We kept this question as part of inclusion criteria and ask for a general estimate (ie, “have you had at least 2 intrusive memories in the last week”), but also added a baseline (week -1) daily electronic registration/diary of intrusive memories that participants are asked to complete during the week prior to filling in the baseline questionnaires and completing the intervention/control session. |
Feeling that participation is alright | It looks like there are going to be a lot of questions in the SmartTrial list of questionnaires. [You] could lump some of the single-item, or shorter questionnaires together. (P3) | We did this with four questionnaires, where we combined two of them together twice. We also removed one of the work-related questionnaires. |
In this study with health care staff (3 registered nurses) who experienced intrusive memories in the context of the COVID-19 pandemic, we piloted a simple and brief intervention (here, approximately 25 minutes) with digitalized study procedures (for remote delivery and with remote support) that was adapted for this population (health care staff working during a pandemic). We explored whether the intervention reduced the number of intrusive memories of trauma (primary outcome) and improved related symptoms and participant’s perception of their functioning. All 3 participants reported a reduction in their intrusive memories (to zero) at week 5 post intervention, a greater reduction than expected; though, this has to be interpreted with caution given the small sample size. They also reported a reduction in other related symptoms—posttraumatic stress symptoms on the IES-R intrusion and avoidance subscales and the PCL-5. Of interest, all 3 participants perceived that their functioning improved (eg, they reported less disruption of their concentration). For example, P1 described that having intrusive memories affected her empathy and her interactions with patients and their relatives. This highlights that a reduction of intrusive memories might be beneficial for health care staff and for their patients and care.
We also explored the feasibility and acceptability of both the digitalized intervention and digitalized data collection. Participants perceived the intervention as feasible and rated it as acceptable (ie, they rated acceptability as 9, 10, and 10 out of 10). There was no dropouts (ie, all 3 participants completed the intervention), indicating favorable acceptability according to National Institute of Health and Care Excellence guidelines [
Of particular interest is the notion that participants readily engaged in repeated self-use of the intervention to reduce remaining intrusive memories, those not targeted in the session with the researcher, or when their work triggered intrusive memories. This indicates high levels of user engagement with the intervention [
To support health care staff during the pandemic and beyond, and to address broader critical challenges of reaching people affected by trauma at scale [
We note the number of intrusions reported at baseline was 2 to 3 per week, which is indicative of 8 to 12 per month. This may sound like a low number, but to the contrary, such levels can potentially cause significant distress and functional impairment; the gold standard assessment tool for PTSD (Clinical Administered PTSD Scale for DSM-5) states that an intrusion frequency of
In terms of perception of functioning, we note that participant’s self-rated concentration improved as their intrusions reduced. This is important as participants reported at baseline that their concentration was disrupted for approximately 1 to 5 minutes each time they experienced an intrusion. Why might having an intrusion, which disrupts concentration, at times lead to a problem of functioning at work? First, having even a brief but sudden and unplanned lapse in concentration after an intrusion can have the potential to interfere with the type of work-related duties that require focused attention, such as monitoring patients who are critically ill on a respirator. In our previous work, we have seen indication that intrusive memories can have a significant impact on people’s perception of their ability to concentrate [
To take steps that might help prevent burnout among health care staff is of high importance due to the fact that even before COVID-19 there was a global shortage of nurses [
We noted several limitations of the current mixed methods exploratory pilot study. First and most important, since our study included a small sample (N=3), any changes in quantitative data reported by our 3 participants have to be interpreted with great caution.
Second, as a precursor to a planned RCT, we did not preregister this pilot study separately but did preregister the RCT before commencing this pilot (ClinicalTrials.gov NCT04460014, July 7, 2020), which reflects the outcomes reported here.
Third, we piloted only our intervention procedures in this study and can therefore only interpret reported intrusive memory data (primary outcome) compared to baseline assessment, not compared to a placebo control group. We will use an attention placebo control task, which we have piloted previously [
Fourth, we assessed the number of intrusive memories in the week prior to the intervention with a single time point retrospective rating. However, participants reported that it was difficult to remember how many intrusive memories they had during the previous week, posing a risk of the numbers we obtained through retrospective ratings being an under- or overestimation of the actual number of intrusions at baseline and making it difficult to compare numbers obtained through a retrospective rating at baseline with numbers obtained through a daily diary during week 1 and week 5. Following this limitation in this pilot study and because participants were positive toward completing an additional diary to monitor their intrusive memories at baseline, we will add a baseline diary in the planned RCT.
Strengths of this study include our use of different methodologies as part of stepwise intervention development [
Based on the target user feedback and lessons learned from this pilot study, we will make the following adjustments to the main RCT. In addition to adding a diary assessing the number of intrusive memories during the week before the intervention, we will also make changes to the diary itself. We will remove the accuracy rating at the end of the diary we had used in this pilot based on our previous pen-and-paper diary because it was unclear for our participants how to rate the item in relation to the digital diary. By collecting the diary data electronically (four times a day), we automatically obtain data on how compliant participants completed it (ie, how often they recorded their intrusive memories) in a more objective way compared to a subjective rating scale. Thus, we will use this data to report diary accuracy in the main RCT. Furthermore, being at work may trigger intrusive memories in this population of health care staff whose work also represents the context in which the traumatic events occurred. Therefore, we will add two ratings assessing the number of days at work and the number of night shifts in the previous week at the end of each diary. These ratings may help the researcher understand patterns of intrusions reported by participants (eg, high frequency of intrusions at night but none during the day could be because a person works night shifts).
We will also adapt the item assessing perceived functional impairment at baseline and follow-ups. Because participants found it difficult to answer our question on perceived functional impairment (since it included many different areas of functioning) and because our questionnaire measures of work-related stress did not seem to capture impairment at work due to intrusive memories very well, we will split this item into perceived work-related functional impairment (rating and an option for a free-text response on how work-related functioning was impaired because of having intrusive memories) and perceived functional impairment in other areas (eg, relationships with other people, social life, schoolwork, housework, or volunteer work). When asked in a phone call after week 5, participants were positive to this change. They also provided examples of perceived work-related functional impairment (see the Clinical Implications section).
We will also remove questionnaires on work functioning that did not reveal any effects in this pilot study to reduce participant burden. For example, a questionnaire assessing intention to leave the profession or workplace will be removed completely, and a questionnaire assessing burnout symptoms will only be assessed at baseline and the 6-month follow-up in the main RCT, since the effects of having intrusive memories on burnout might only unfold after longer time periods. Additionally, we will slightly change the wording of the intrusion questionnaire [
There was also some feedback from our participants we could not implement because of the limitations of the electronic data collection platform we are using in this study. For instance, we could not embed instruction videos directly in the electronic platform. Instead, participants had to click on a link directing them to a separate site on which we host the videos, which impaired study flow for participants. We also had to use a freely available version of
As illustrated with participants P1 and P3, the recurrence of one specific intrusive memory (eg, recurrent images of a specific patient’s face) can directly impact related emotion and behavior (eg, having intrusions was reported as having an impact on the capacity to provide empathic patient-centered care or experiencing intrusions was related to negative anticipations about going to work). The perceived work-related impact from having intrusions described by P1 (see the Open-ended Feedback From Participants section) reflects a typical avoidance circle commonly seen in PTSD where similar stimuli as the trauma stimuli (here patients) triggered subtle avoidance behavior for P1. For P3, intrusive memories made her feel reluctant about going to work and worried that something similar would happen. P3 noted that prior to the intervention it felt as if her mental “backpack was totally full” and that experiencing intrusive memories warned her that she might experience more events at work. She was worried that encountering similar events would cause additional intrusions, that is, intrusive imagery in only
In this study, the intervention was delivered by a researcher who had been trained and had extensive experience on delivering the intervention with clinical participants (MK) and who received additional supervision as needed (from EAH). As training and supervision are currently deemed crucial to deliver the intervention adequately, we note that next steps in addition to the RCT include creating more standardized training material that can be delivered remotely (eg, in the form of an online training course) to researchers and procedures for role play, observation and feedback, and
Further, in this study, participants were recruited from within the professional network of the researchers. A potential barrier for recruiting a large sample for an RCT could be the heavy burden placed on health care staff during the pandemic, which might prevent them from engaging in a research study. Thus, recruitment strategies for an RCT need careful consideration and optimization.
In this mixed method exploratory pilot study, we piloted a brief intervention that we have digitalized (for remote delivery and with remote support) and adapted for this target population (health care staff working during a pandemic). We explored whether the intervention reduced the number of intrusive memories of trauma (primary outcome) a month after the intervention session and improved related symptoms and participant’s perception of their functioning. Our participants (3 registered nurses who had experienced intrusive memories in the context of their work during the COVID-19 pandemic) reported zero intrusive memories at week 5 post intervention, a greater reduction than expected, which has to be interpreted with caution given the small sample size. Participants also reported a reduction in other related clinical symptoms (eg, posttraumatic stress symptoms) and perceived that their functioning improved (eg, they reported less concentration disruption).
We also explored the feasibility and acceptability of both the digitalized intervention and the digitalized data collection. After an initial session with the researcher, all 3 participants continued to use the intervention independently and perceived it as feasible and acceptable. The primary outcome measure (a daily digital intrusion diary) and secondary outcome measures (eg, digital questionnaires) were successfully completed. Should the intervention prove effective in future clinical trials, it could hold promise as a helpful tool, which because of its brevity and digital format, could be incorporated into health care staff’s work life and be repeated as needed in the face of numerous intrusive memories or ongoing trauma exposure. One next step will be to investigate the efficacy of the intervention in reducing the number of intrusive memories in an RCT with health care staff with exposure to traumatic events during the COVID-19 pandemic.
Supplementary materials.
Diagnostic and Statistical Manual of Mental Disorders (Fifth Edition)
intensive care unit
Impact of Event Scale-Revised
must have, should have, could have, won’t have
Posttraumatic Stress Disorder Checklist for Diagnostic and Statistical Manual of Mental Disorders (Fifth Edition)
posttraumatic stress disorder
randomized controlled trial
subjective units of distress
We thank Evelina Kontio for support with the formatting of tables; Katy Metcalf for her animation work; Beau Gamble for a discussion around guiding principles; Lalitha Iyadurai for helpful input in the preparation stages of, for example, adapting study materials for health care staff; and Michelle Moulds for input in moving from this pilot study to the RCT.
This project was mainly supported by grants to EAH from the Swedish Research Council (2017-00957 and 2020-00873), the Oak Foundation (OKAY-18-442), AFA Insurance (200342), and the Lupina Foundation. LS was supported by a Swiss National Science Foundation grant (P2BEP1_184378) and a Thunberg Fellowship by the Swedish Collegium for Advanced Study. MK was supported by a FO Medical Psychology grant from Karolinska Institutet and Karolinska University Hospital. KEG was supported by a grant from FO Akut, Karolinska University Hospital. AR was supported by a grant from AFA Insurance (200311). Funders were not involved in the study design, collection, analysis and interpretation of data; writing of the paper; or decision to submit for publication.
LS contributed to study preparation; was responsible for data monitoring and analysis, tables, and graphs; and wrote the first draft of the manuscript with MK.
MK contributed to study preparation; was responsible for intervention delivery, data collection, and qualitative data analysis; and wrote the first draft of the manuscript with LS.
KD provided assistance during data collection and contributed to qualitative data analysis and write up under supervision.
ACF, VL, OD, KEG, and AR all provided nursing expertise during study preparation and throughout recruitment, data collection, and write up.
EAH conceived the study, including study design, interpretation of data, and write up, and supervised the study.
EAH reports serving on the board of trustees of the charity MQ: Transforming Mental Health but receives no remuneration for this role. EAH receives royalties from books and occasional fees for workshops and invited addresses; she receives occasional consultancy fees from the Swedish agency for health technology assessment and assessment of social services. The other authors have no conflicts of interest to declare.