A Quality Improvement Project to Support Post-Intensive Care Unit Patients with COVID-19: Structured Telephone Support

Background: More than 50% of intensive care unit (ICU) survivors suffer from long-lasting physical, psychosocial, and cognitive health impairments, also called “post-intensive care syndrome” (PICS). Intensive care admission during the COVID-19 pandemic was especially uncertain and stressful, both for patients and for their family. An additional risk of developing symptoms of PICS was feared in the absence of structural aftercare for the patient and family shortly after discharge from the hospital. The purpose of this quality improvement study was to identify PICS symptoms and to support post-intensive care patients and families in the transition from the hospital to the home. Therefore, we offered post-ICU patients and families structured telephone support (STS). Methods: This was a quality improvement study during the 2019 COVID-19 pandemic. A project team developed and implemented a tool to structure telephone calls to identify and order symptoms according to the PICS framework and to give individual support based on this information. We supported post-ICU patients diagnosed with COVID-19 pneumonia and their family caregivers within four weeks after hospital discharge. The reported findings were both quantitative and qualitative. Results: Forty-six post-ICU patients received structured telephone support and reported symptoms in at least one of the three domains of the PICS framework. More than half of the patients experienced a loss of strength or condition and fatigue. Cognitive and psychological impairments were reported less frequently. Family caregivers reported fewer impairments concerning fatigue and sleeping problems and expressed a need for a continuity of care. Based on the obtained information, the ICU nurse practitioners were able to check if individual care plans were optimal and clear and, if indicated, initiated disciplines to optimize further follow-up. Conclusions: The implementation of the STS tool gave insight in the impairments of post-ICU patients. Surprisingly, family caregivers expressed fewer impairments. Giving support early after hospital discharge in a structured way may contribute to providing guidance in the individual care plans and treatment of the early symptoms of PICS (-F).

The primary aim of this quality improvement project was to provide support to postintensive care patients and families; therefore, we developed and implemented an STS tool for post-intensive care patients and families in the transition from the hospital to the home. The STS intervention was undertaken by ICU nurse practitioners and underpinned by the PICS framework [9]. It comprised several short questions focusing on the different post-ICU problems patients can experience with PICS (i.e., physical, mental, and cognitive consequences). Second, to provide a deeper understanding of the long-term impairments of post-ICU patients and their families, we report the findings of the STS in the PICS framework [9].

Materials and Methods
In this quality improvement study, post-ICU patients treated for COVID-19 pneumonia received an STS within four weeks post-discharge to home. We used both quantitative and qualitative methods integrated in the Deming cycle, as reported by the criteria of the revised Standards for Quality Improvement Reporting Excellence 2.0 framework [37].

Study Design and Setting
This was a quality improvement initiative during the COVID-19 pandemic in a Dutch tertiary hospital. This hospital has a 34-bed, mixed medical/(neuro) surgical ICU with a 24/7 intensive care service delivered by ICU nurses and intensivists [38]. During the outbreak of COVID-19, the ICU capacity in this hospital was upgraded to 40 beds to guarantee care for both COVID-19 patients and regular ICU patients.

Population and Sampling
We offered STS to all post-ICU patients diagnosed with COVID-19 pneumonia, as proven by a reverse transcription-polymerase chain reaction (RT-PCR) test, who were discharged home in the period of 7 March 2020 to 15 May 2021. If family members were involved, they were also asked to participate.

Plan
At the start of the COVID-19 pandemic, the project team, including three ICU nurse practitioners (SH, MO, MP), an ICU physician, and nurse researchers, rapidly developed an STS tool-the so-called "STS Post-ICU COVID-19 tool". The purpose of developing and implementing the tool was to structure a telephone support intervention to screen the symptoms experienced by post-ICU patients, divided into the three domains of the PICS framework. By asking about all the domains of PICS in a structured way, we tried to identify where the most symptoms and thus the greatest needs are in their individual treatment plan. During the phone interview, all symptoms were asked about in a structured way and noted using the STS Post-ICU COVID-19 tool. Then, the ICU nurse practitioner discussed with the patient and/or family whether there was sufficient support at home or whether additional support was required. Based on a rapid search of the literature from Medline/PubMed, focusing on systematic reviews, we identified evidence of the effectiveness of the use of STS intervention on patient outcomes [26,27]. Consequently, we combined the well-known PICS framework [9,39] and local knowledge of our team to define the content of the STS intervention for our population. Figure A1, Appendix A visualizes the final developed STS Post-ICU COVID-19 tool, which is a structured digital pocket card that focuses on the three domains of the PICS framework (physical, cognitive, and psychological function) [9]. In addition to this, we added one question about ICU experiences and two questions (5a and 5b, Figure A1, Appendix A) to assess the mental health and symptoms of the family caregiver.

Do
After development, we implemented the STS Post-ICU COVID-19 tool into daily practice. To support the use of the tool, we developed smart phrases (i.e., blocks of text that can automatically be pasted into the patient's Electronic Health Record (EHR)).
Three ICU nurse practitioners (SH, MO, and MP) conducted the intervention within four weeks after the patients' hospital discharge to home. Due to the rapid implementation strategy due to COVID-19, the ICU nurse practitioners were not trained in the telephone support. Therefore, the ICU nurse practitioners discussed a strategy for the telephone call beforehand, and the tool provided sufficient structure for the telephone support to have a uniform procedure. All post-intensive care patients who were discharged from hospital to home were offered an STS within four weeks after hospital discharge.
Each STS call started with a short review of the patient's experiences in the ICU. All questions, per the PICS domain, were asked using the structured format of the STS intervention. The occurrence of PICS symptoms per domain was assessed by patients using a four-point scale ranging from "no burden" to "very much". In addition to this score, patients were asked to provide additional information to concretize the symptoms and report-initiated interventions or advice. This information was also reported in the EHR. Lastly, if family members were involved, they were asked about their psychosocial symptoms. If not, then patients were asked to assess the symptoms of their family members or next of kin. Each call lasted 20-60 min.

Check
We described the gathered symptoms that patients and family caregivers reported during the STS. Therefore, we collected demographical data including age, gender, body mass index (BMI), and pre-existing comorbidities. Furthermore, we used the Acute Physiology and Chronic Health Evaluation IV (APACHE IV) to estimate the risk of short-term mortality and the length of the ICU stay [40]. The Sequential Organ Failure Assessment [41] was used to describe the severity of organ failure during ICU admission. We recorded the number of mechanical ventilation days, the length of stay (LOS) in the ICU, and the LOS in the hospital. All clinical data were retrospectively collected from the EHR. The data from the STS Post-ICU COVID-19 tool were used to describe identified domains of PICS in numbers and percentages. As appropriate, quantitative and continuous data were expressed as means with the standard deviation (SD) or as medians with interquartile ranges (IQRs). All data were analyzed using a statistical software package (SPSS Inc., version 24, Chicago, IL, USA). Additionally, we collected the free text of the reported consultations from the EHR and processed it anonymously into a text database for qualitative analysis. First, we checked if there was additional information reported about PICS symptoms according to the PICS framework (Table A1, Appendix B). Second, we collected data on reported patients' needs, initiated interventions, and nursing problems. To support the data collection, we constructed an easy-to-use table that structured the data per patient. The data collection tables were pilot tested by three researchers (SH, AE, and JB) through independent data collection. Discrepancies in interpretation were resolved through discussion.
After pilot testing, two independent researchers read all of the notes from the structured telephone reports line by line (JB and SH). Both researchers (JB and SH) scrutinized the data by coding and extracting quotes [42]. In addition, a final category frame was made to structure all the relevant data into the three predetermined categories. The thematic matrix is attached in Supplementary Table S1. Both of the researchers categorized the data separately and discussed differences until a consensus was reached.

Results
In total, 49 post-ICU patients received an audio/video call in which the STS tool was used. The patient flow diagram shows the flow of patients ( Figure 1); in three cases, data were missing because no information was reported. The first post-ICU patient who participated was admitted to the ICU on March 12, 2020, and the last patient was admitted on March 2, 2021. The median time between the ICU discharge and the interview was 37 days (IQR 24-54). The demographics and clinical characteristics of the included patients are presented in Table 1.
matrix is attached in Supplementary Table S1. Both of the researchers categorized t separately and discussed differences until a consensus was reached.

Results
In total, 49 post-ICU patients received an audio/video call in which the STS to used. The patient flow diagram shows the flow of patients ( Figure 1); in three case were missing because no information was reported. The first post-ICU patient w ticipated was admitted to the ICU on March 12, 2020, and the last patient was ad on March 2, 2021. The median time between the ICU discharge and the interview days (IQR 24-54). The demographics and clinical characteristics of the included p are presented in Table 1.    Table 2 details the scored symptoms of PICS with the Post-ICU COVID-19 tool. We described each element of PICS and combined the quantitative data from the structured tool and the qualitative data extracted from the text of the consultation reports of the ICU nurse practitioners as quotes (R#). After the structured questioning with the tool, there was also room to discuss other symptoms or to question symptoms in more depth. This additional information was reported as loose text. From this text, we organized the additional symptoms and describe them in Table A1 (Appendix B). Table A2 (Appendix B) shows an inventory of the professionals involved in the individual care plans of the post-ICU patients included.

Physical Symptoms
All the patients experienced physical symptoms (n = 46). Over 60% of the patients experienced a loss of strength or condition. Fifty-four percent of patients reported the loss of muscle strength as "quite a lot", and 59% reported the loss of condition as "quite a lot". Dyspnea was scored as "very much" by 7% (n = 3) of the patients and as moderate by 26% (n = 12 patients). Fatigue was reported by 89% of patients (n = 39), and 22% (n = 10) of the patients reported this as intensely present.
The most frequent additional physical symptoms reported by the patients were dyspnea, poor condition, pain, and limited mobility. In particular, pain while breathing and fatigue were often mentioned as a limitation of daily function: "In his own words, he sometimes 'gasps for breath'".(R#7) Patients experienced additional physical symptoms such as edema of the legs and sleeping problems. Sleeping problems were often mentioned, sometimes due to dyspnea, nocturnal urination, or psychological symptoms. If sleep problems were mentioned, a classification was lacking.
"I still have a lot of thoughts; I sleep an average of 2 to 3 h a night".(R#15) In addition, patients experienced weight loss and ICU-acquired weakness.
"Fine motor skills can still be improved, tingling in toes, colder hands are described".(R#46)

Cognitive Symptoms
Cognitive symptoms were less frequently reported; the inability to plan (13%, n = 6), the inability to multitask (15%; n = 7), and overstimulation (13%; n = 6) were reported as "quite a lot". In addition, patients' symptoms regarding cognitive dysfunction were reported, especially memory loss (9% reported this as "very much") and concentration problems. Memory loss causes limitations in daily functioning and sometimes in the context of work.
"Patient cannot carry out his work as a lawyer because of fatigue and memory loss".(R#3)

Psychological Symptoms
All patients experienced psychological symptoms; most of the symptoms were signs of Post-Traumatic Stress Disorder (PTSD) and were reported as "quite a lot" in 17% of cases (n = 8). The most mentioned symptoms of anxiety were excessive worry and irritability. Patients expressed their concerns about COVID-19 and feared a recurrence of illness.
"Patient expresses difficulties to deal with visitors because of fearing a recurrence of COVID-19".(R#14) During the telephone consultations, emotions were often still present. For some patients, the conversation was still too tiring or had too much of an impact. Some patients explained a state of avoidance around memories of the ICU and described flashbacks, including nightmares.
"Patient becomes emotional several times-nightmares, thoughts of the delirium he experienced. He mentioned, he was tied up, wanted to take out the ventilator. He understands that this was necessary, but now he has terrible thoughts about it that continue to haunt him".(R#15)

Caregivers' Symptoms
In nine cases, caregivers participated in the video or telephone call. In 91.3% (n = 42) of cases, caregivers reported experiencing "not very many" symptoms. In four cases, the caregivers did report psychological symptoms. Most symptoms concerned fatigue and sleeping problems: "The caregiver is more tired lately, sleeps worse, since last days" (R#5). When caregivers participated in the phone calls, they could explain which symptoms were present; in other cases, the patients themselves described the symptoms of their relative or caregiver. Fatigue, anxiety, flashbacks, and nightmares were experienced as symptoms. In one case, relational problems were reported: "Patient experiences stress from memories of the ICU admission, and this manifested itself in problems in communication with his family members (informal caregivers)".(R#26)

Patients' Needs
Based on the reported symptoms, the nurse practitioners detected patients' needs based on their reported experience and symptoms. Patients especially mentioned the need for a continuity of care. This manifested itself primarily in the specific need for clear communication with their (primary) healthcare practitioners and clarity about follow-up at home. In addition, family caregivers expressed their need to be seen and heard in their role as informal caregivers. As a result, the ICU nurse specialist scheduled a follow-up meeting to listen to their experiences again as an aftercare measure. When healthcare practitioners were not involved, the ICU nurse specialist actively referred patients or caregivers to healthcare practitioners. In several cases, the ICU nurse specialist contacted the general practitioner or a medical specialist to improve the care plan-for instance, when symptoms had worsened or new symptoms had appeared. For some patients, the ICU nurse specialist could provide clarity about their admission into the ICU and provide information to meet their needs. Patients and family members always received an informal invitation to seek contact with the ICU nurse specialist in case of new problems. In some cases, the ICU nurse practitioners initiated extra follow-up calls for evaluating initiated advice and monitoring symptoms. Table A2, Appendix B shows an overview of the inventory of professionals already involved in the individual care plans of the post-ICU patients included.

Discussion
This study describes a quality improvement during the COVID-19 pandemic using the "STS Post-ICU COVID-19 tool". We offered structured telephone follow-up care to post-ICU patients who were discharged from the hospital to the home within four weeks after hospital discharge. Forty-six post-ICU patients experienced physical symptoms, whereas cognitive and psychological symptoms were reported less. Almost two-thirds of the participating patients mentioned a loss of strength and condition. Family caregivers expressed the need for a continuity of care and experienced fewer symptoms.
Recent studies focused on physical function, reported fatigue, and muscle weakness as the most common symptoms after ICU admission with COVID-19 pneumonia [43]. We compared our study with a recent Dutch multicenter study that included 246 patients who were alive one year following ICU treatment for COVID-19 [44]. In this study, 74.3% reported physical symptoms, 26.2% reported mental symptoms, and 16.2% reported cognitive symptoms [44]. These findings are in line with our own. It is notable that our intervention was a self-report and was conducted shortly after hospital discharge (within 4 weeks). Another Dutch study including post-ICU patients without COVID-19 reported similar rates in terms of physical and cognitive symptoms. The rates of psychological symptoms were higher in this study [44]. Martillo et al. found a high prevalence (91%) of PICS elements in 45 COVID-19 post-ICU patients in a similar study with telephone follow-up after 1 month post-discharge. In this cohort, 87% reported physical impairments [32]. A study from Italy found reduced functional capacity in post-ICU patients with COVID-19 at 2 months post-hospital discharge [45]. The cardiopulmonary performance was presumably better in this study than in cohorts with other forms of ARDS. Carenzo et al. also highlighted the mildly reduced overall quality of life and a high proportion of PTSD symptoms at 6 months [45]. Halpin et al. (2020) implemented a similar rapid structured telephone tool, ordered by the International Classification of Functioning, Disability, and Health instead of the PICS classification we used [46]. Several studies reported physical symptoms such as fatigue, breathlessness, and psychological distress in COVID-19 survivors (n = 201) 14 and 20 days after hospital discharge [46,47]. In our study, we performed a follow-up at a median of 37 days (IQR , and our findings are in line with these aforementioned studies [46,47] A possible explanation for the high rate of physical symptoms is the higher proportion of ICU patients who received mechanical ventilation. A recent review showed that the number of ICU patients requiring mechanical ventilation during ICU admission increased from 20-40% in the period before COVID-19 to 63-87.3% during the COVID-19 pandemic [48]. Furthermore, the usual standards of care, such as the ABCDEF bundle, were not always practically feasible, which ideally would contribute to preventing PICS-F [49][50][51]. The symptoms of the family caregivers in our study were unexpectedly low. In nine cases, caregivers reported their concerns, and in the other cases, patients were asked to assess the symptoms of their caregiver; these assessments may have differed from the perceptions of the caregivers themselves. Nevertheless, the risk of suicide and self-harm has been associated with ICU admission when patients have pre-existent psychological conditions such as anxiety, depression, or PTSD [52]. This new insight supports the need for an individual assessment to determine the risk of developing symptoms of PICS and to initiate early interventions. In our study, cognitive function scored low ("not"/"not often"), which is not in line with the current literature. The incidence of cognitive impairment after one year post-ICU was determined to be 43% [53]. In our study, patients were interviewed shortly after hospital discharge and were asked to self-assess their cognitive function. Early after discharge, the focus of most patients was on physical rehabilitation. Cognitive impairments may reveal themselves later; several studies have reported cognitive symptoms after a follow-up period of 2-156 months [53][54][55]. The self-assessment of cognitive function may be insufficient. Moreover, cognitive problems such as difficulty concentrating and multitasking often come to light when work and social obligations are resumed. However, the early onset of symptoms and the need for extra help could be detected and could be helpful in reducing long-term PICS complaints.
The incidence of psychological symptoms was relatively low overall, except for the number of patients who reported PTSD symptoms (n = 8, 17%). A meta-analysis found that one-fifth of ICU survivors reported symptoms of PTSD after one year in the period before COVID-19. These numbers suggest that the psychological impact is high among survivors who have had "good" outcomes because they were discharged to home. We expect the numbers to be higher at a 6-or 12-month follow-up, as the impact on daily life becomes more apparent. However, recent studies report otherwise [44]. The caregivers in this study are underrepresented because they only participated in the consultations in a few cases. The few who reported their experience did describe psychological symptoms. Experts caution that the COVID-19 pandemic may have a greater impact on PICS-F numbers as family caregivers experience more awareness of the limitations of this crisis [56]. For family caregivers, a proactive follow-up call may have a positive impact in allowing them to share their experience and burden in order to access help early.
Although several studies have described the burden on ICU patients with COVID-19 and their families after hospital discharge, there is less evidence of initiated interventions. Before the COVID-19 pandemic, there was little evidence of interventions that could prevent long-term symptoms such as those caused by PICS [12].
This study describes a quality improvement in a single-center context and with a small sample size. Another limitation to mention is the fact that we only included patients discharged from the hospital to the home, which may not be representative of the whole population of ICU patients in this period. Almost one-third of COVID-19 patients admitted to Dutch ICUs died in the hospital [57]. The majority of the ICU patients admitted during this study period were transferred to a rehabilitation center or nursing home and received a structured rehabilitation care plan. For patients discharged to home, a structured care plan was lacking. Secondly, not all eligible patients received interventions due to limited staff capacity during the COVID-19 pandemic, and not all calls from patients were answered. In addition, a substantial group of ICU patients were transferred to another ICU because of hospital bed capacity problems during the pandemic. This may have led to selection bias. In addition, we know that having a loved one admitted to the ICU can have a major impact on family caregivers, especially during the COVID-19 pandemic.
Nevertheless, family caregivers are underrepresented in this study, but they may have suffered more because the limited access to healthcare affected this group during the COVID-19 pandemic. This study substantiates that a structured, early inventory of PICS may potentially contribute to the proactive deployment of quality improvements. However, owing to a lack of time due to COVID-19, we did not perform an appropriate validation of the protocol. This tool may be beneficial for post-ICU patients and families in general in a non-pandemic setting. Further research is needed to validate and evaluate the effectiveness of structured tools to help professionals identify patients' needs. In addition, cognitive function was measured by self-assessment; this may have resulted in reporting bias. Lastly, the effect of the coordinating role of nurse practitioners remains underrepresented in this study. In further research, we recommend focusing on the process of individual care planning during complex transitions after ICU admission. Before COVID-19, Dutch ICU aftercare was not structured or structural [8]. Whether aftercare should be structured or more individually based is a topic of debate. Therefore, an early inventory may provide guidance for individual care plans during the transition period from the hospital to the home.

Conclusions
To conclude, this quality improvement study shows that post-ICU patients and their family caregivers experienced symptoms of the PICS framework, identified at an early stage of post-discharge with the STS tool. Post-ICU patients diagnosed with COVID-19 pneumonia all reported symptoms in at least one of the three elements of the PICS framework. The most common symptom was physical burden in all patients, and symptoms of PTSD were also notable. We recommend individually planned early-onset rehabilitation and for more structured attention to be paid to family caregivers. Further research and follow-up are crucial, as COVID-19 is a new illness and post-discharge symptoms and long-term follow up are yet to be researched.

Institutional Review Board Statement:
The study was conducted in accordance with the Medical Research Ethics Committee Brabant, The Netherlands, and approved by the ETZ research board under the number NW2020-76, and the local hospital research protocol was provided with the research number L1079.2020.

Informed Consent Statement: Not applicable.
Data Availability Statement: The data informing the findings of this study are available from the corresponding author upon reasonable request.

Acknowledgments:
The authors would like to thank all the participants and professionals involved in providing care for this study.  Table A1. Additional symptoms ordered by Post Intensive Care Syndrome (PICS).

Physical Elements
Dyspnoea "In his own words, he sometimes 'gasps for breath'" (r. 7) Poor condition "Showering and climbing stairs still consume a lot of energy" (r. 29) Poor mobility "Walking short distances in the house are still quite difficult" (r. 48) Sleeping problems "Still has a lot of thoughts, sleeps an average of 2 to 3 hours a night" (r. 15) Oedema in legs "Patient suffers from swollen feet, especially during the day" (r. 36) Weight loss "Patient lost weight, 18 kg in total. Now, gained another 2.5 kg" (r. 7) ICU-acquired weakness "Fine motor skills can still be improved, tingling in toes, colder hands are described" (r. 47)

Cognitive elements
Memory loss "Patient cannot carry out his work as a lawyer because of fatigue and memory loss" (r. 3) Attention/concentration problems "Slightly slower thinking ability, suited to the situation" (r. 26)

Psychological element
Anxiety, excessive worry "Patient expresses difficulties to deal with visitors because of fearing a recurrence of COVID-19" (r. 14) Anxiety, irritability "Patient states that sometimes he is a bit more/quicker irritated" (r. 33) Avoidance "Patient explains he experiences avoidance around the memories" (r. 37)

Flashbacks, nightmares
"Patient becomes emotional several times -nightmares, thoughts of the delirium he experienced. He was tied up and wanted to take out the ventilator. He understands that this was necessary, but now he has terrible thoughts about it that continue to haunt him" (r. 15)

Caregivers' burden
Fatigue "Caregiver is a bit more tired" (r. 47)