Symptoms of psychological distress among bereaved relatives of deceased patients in the intensive care unit during the first COVID-19 wave.

Dear Editor,

During the coronavirus disease 2019 (COVID-19) pandemic, policies of family-centered care in the intensive care unit (ICU) were impacted by visiting restrictions and undesirable physical distance to the patient [1]. This might have increased psychological distress, including complicated grief (CG) among bereaved relatives [2]. CG is characterized by intense symptoms of grief, separation distress, intrusive thoughts, and feelings of emptiness or meaninglessness lasting for more than 6 months post-loss [3]. A French qualitative study reported that relatives’ experiences of ICU care and mourning were disrupted during the pandemic [4]. We aimed to compare symptoms of CG and psychological distress such as depression, anxiety and posttraumatic stress in bereaved relatives of deceased ICU patients before and during the COVID-19 pandemic. To the best of our knowledge, this is the first study exploring a hypothesized raise in symptoms of CG in bereaved relatives as a result of the COVID-19 pandemic.

This cross-sectional survey study is part of the BRIC-study [5]. We stratified the post-intervention group by date of patient’s deaths: September 2019–February 2020 (non-COVID-group; n = 114, patients dying before the pandemic) and March–May 2020 (COVID-group; n = 43, patients dying during the 1st wave). All first and second contact persons were approached for study participation (n = 269). After providing informed consent, relatives were invited to complete a questionnaire measuring CG, anxiety, depression, and posttraumatic stress 8 weeks after bereavement.

In total, 121 (45%) respondents completed the questionnaire (non-COVID-group n = 86, COVID-group n = 35). Most respondents were partners (52%), female (71%), and had a mean age of 52 years (range 20–83). We found no difference between groups regarding age, relationship to the patient and religion. Symptoms of CG were above the cutoff in 1.3% in the non-COVID-group and 48.6% in the COVID-group. All outcome variables showed a normal distribution, with both skewness and kurtosis between 1 and − 1. Therefore, using a Student’s T-test was allowed. Table 1 shows levels of psychological distress indicating significant worse results for the COVID-group compared to the non-COVID-group.

Table 1

Levels of psychological distress in the non-COVID-group and the COVID-group

	Mean (± SD)		Standardizera	Point estimate	95% CIb
	Non-COVID	COVID			Lower	Upper
CG*	37.4 ± 8.8	60.9 ± 9.4	8.95	− 2.61	− 3.14	− 2.09
Anxiety**	6.9 ± 2.3	9.2 ± 1.9	2.16	− 1.08	− 1.55	− 0.6
Depression**	9.4 ± 2.4	11.8 ± 1.6	2.16	− 1.12	− 1.61	− 0.62
PTS***	31.6 ± 10.5	56.2 ± 11.2	10.68	− 2.3	− 2.83	− 1.78

SD standard deviation to the mean, CI confidence interval

aStandardizer as the denominator used in estimating the effect sizes

bIndependent samples effect sizes with Cohen’s d using a Student’s T-test analysis

*CG = symptoms of complicated grief measured after 8 weeks with traumatic grief inventory-self report version, 19 items, scores range from 17 minimum to 85 maximum

**Anxiety and depression symptoms measured with Hospital Anxiety and Depression Scale, 7 items in each subscale, scores range from 0 minimum to 21 maximum

***PTS = symptoms of posttraumatic stress measured with Impact of Events Scale, 22 items, scores range from 0 minimum to 88 maximum

Our study has limitations. First, we measured CG earlier than 6 months post-death, potentially leading to overestimation. However, as the measurements were conducted at similar moments in the non-COVID-group and COVID-group, the comparison is meaningful. Second, during the pandemic, visiting hours in our ICU were reduced from almost all day (11.00 AM–21.00 PM) to maximum 1 h in person for two relatives per patient. Although all relatives were physically present during the dying process, visiting restrictions could have impacted relatives’ psychological distress. Watchful waiting and coping mechanisms are common strategies to recover. However, if this period was overlapping with start of the pandemic, which was the case for 27% of bereaved relatives in the non-COVID-group, the process could be disrupted by traumatic circumstances and less social support. Unmeasured confounders, such as lack of social support, confrontations with COVID-19 due to ongoing restrictions, and media reports, could have contributed to psychological distress in both groups differently.

Our findings emphasize the need for unrestricted visiting policies and social support for bereaved relatives, who are at high risk of developing psychological distress.

Trial registration

Netherlands Trial Register NL 7875, Registered on 27/07/2019, http://www.trialregister.nl.