Impact of COVID-19 infection experience on mental health status of intensive care unit patients’ family members: a real-world study

Summary

Purpose

Family members of patients hospitalized in intensive care unit (ICU) are susceptible to adverse psychological outcomes. However, there is a paucity of studies specifically examining the mental health symptoms in ICU patients’ family members with a prior history of coronavirus disease 2019 (COVID-19) infection.

Aim

This study aimed to investigate mental health status and its influencing factors of ICU patients’ family members with previous COVID-19 infection experience in China.

Design

Nationwide, cross-sectional cohort of consecutive participants of family members of ICU patients from 10 provinces randomly selected in mainland China conducted between October 2022 and May 2023.

Methods

The basic information scale, Self-rating depression scale, Self-rating Anxiety Scale, Impact of Event Scale-Revised, Pittsburgh sleep quality index, Perceived Stress Scale, Connor-Davidson resilience scale, Simplified Coping Style Questionnaire were employed to explore mental health status among participants.

Results

A total of 463 participants, comprising 156 individuals in Covid-19 family member cohort (infection group) and 307 individuals in control family member cohort (control group), met inclusion criteria. The infection group exhibited significantly higher incidence of composite mental health symptoms compared to control group (P = 0.017). Multivariable logistic regression analysis revealed that being female, engaging in physical/mental labor, residing in rural areas, and having children were identified as risk factors for the development of depression, anxiety, and post-traumatic stress disorder symptoms, while medical history of surgery was protective factor. A predictive model demonstrated a favorable discriminative ability and excellent calibration.

Conclusion

COVID-19 infection experience regarded as new traumatic stressors worsen mental health status of ICU patients’ family members.

Introduction

The sudden onset of critical illness, coupled with inadequate understanding of the intensive care unit (ICU) environment and uncertainty prognosis of condition, present complex psychosomatic challenges not only for patients but also for their family members.¹ Research indicates that after the ICU stay, 6–69% of family members endure post-intensive care syndrome-family (PICS-F).¹ And not just after discharge, a considerable proportion of family members during ICU hospitalization also encounter a range of psychological challenges, commonly known as family intensive care unit syndrome (FICUS). FICUS is associated with substantial morbidity and functional impairment among family members of patients admitted to ICU wards.^2,3 Regrettably, despite healthcare professionals acknowledging the importance of addressing the physical and mental well-being of family members during and after ICU care, there is currently a dearth of tangible interventions in this regard.⁴

In early 2020, the swift transmission of coronavirus disease 2019 (COVID-19) exerted substantial effects on social, economic, and global healthcare systems. By 17 May 2023, the cumulative number of reported cases surpassed 766 million, resulting in a global death toll of over 6.93 million.⁵ As the global COVID-19 pandemic nears its conclusion through enhanced medical conditions and precise scientific control measures, the enduring psychological and social repercussions it has engendered remain prevalent. Empirical evidence indicates that COVID-19, as a novel traumatic stressor, poses a risk for post-traumatic stress disorder (PTSD), anxiety, and depression.⁶ Post-acute COVID-19 infection individuals may experience a protracted duration of persistent symptoms referred to as ‘long COVID’.⁷ Extensive research has consistently demonstrated that individuals who have recovered from the acute infection exhibit enduring psychosomatic symptoms, including depression, anxiety, PTSD, obsessive-compulsive disorder, and insomnia.^7–9 The enduring symptoms following COVID-19 infection exert a detrimental influence on patients’ cognitive functioning and mental well-being, resulting in a decline in their overall quality of life and imposing substantial burdens on society and economy.^7,10 To date, the effects of the COVID-19 infection experience on the mental health status of family members of ICU patients remain uncertain.

Given the enduring impact of COVID-19 on the psychosomatic symptoms, the situation of family members may be more intricate. However, to the best of our knowledge, no studies have examined the influence of COVID-19 infection experiences on the mental health status of family members of ICU patients, particularly among vulnerable populations. Therefore, the aim of this nationwide study was to investigate the mental health status and its influencing factors of ICU patients’ family members with a prior history of COVID-19 infection during post-pandemic era.

Materials and methods

Study design, setting and participants

This study was conducted between 1 October 2022 and 1 May 2023. Estimating that 46% of family members of ICU patients would experience psychosomatic symptoms based on previous studies,⁹ a required sample size of 399 was calculated with a type I error probability of 0.05. A random sampling method was employed to select participants from 10 provincial-level administrative regions in mainland China for questionnaire surveys. Ultimately, a total of 463 valid questionnaires from 6 provinces, 2 autonomous regions, and 2 municipalities were collected and utilized (Supplementary Figure S1).

This study was granted approval by the Ethics Committee of Tianjin Medical University General Hospital (No. IRB2023-wz-068). The inclusion criteria encompassed ICU patients admitted between October 2022 and May 2023, with no predetermined specifications regarding admission dates or disease types, but with adherence to ICU admission criteria (Supplementary Figure S2 for flow chart). We define the ‘Covid-19 infection experience’: family members (participants in our research) contacted through the electronic medical record system having at least one documented history of Covid-19 infection prior to the patient’s ICU admission.

Exclusion criteria consisted of age less than 18 years and ICU hospital stays less than 4 days. Following patient selection, the closest family members were contacted through the electronic medical record system, and the survey’s purpose was explained.

The term ‘closest’ family member refers to the immediate family members or primary caregivers who have a close relationship with the patient and are involved in their care and support. This typically includes spouses, parents, children, or siblings. To ensure accurate recollection of experiences and emotions, data collection from relatives using items of questionnaire were conducted within two weeks of entering the ICU ward. The interviews took place in a private room, and informed consent was obtained from all patient’s family members.

This nationwide cross-sectional study adheres strictly to the STROBE checklist guidelines¹¹ (Supplementary Table S1). Our aim is to advocate for increased psychosocial support for family members of ICU patients. and all reasonable requests that promote this cause can be shared the original data.

Outcomes and covariates

Demographic characteristics encompassed age, gender, relationship to the patient, education level, occupation, income level, residential address (urban or rural), reproductive history, insurance status, medical history, history of COVID-19 infection, vaccination, participation in epidemic prevention work, and experience of Quarantine.

The Self-rating depression scale (SDS) Self-rating Anxiety Scale (SAS), Impact of Event Scale-Revised (IES-R), Pittsburgh sleep quality index (PSQI), Perceived Stress Scale (PSS), Connor-Davidson resilience scale (CDRISC), Simplified Coping Style Questionnaire (SCSQ) were employed to explore mental health status among participants (Supplementary Tables S2 and S3), with anxiety, depression, and PTSD as the primary outcome measures. The specific scale score calculation and reliability test are shown in Supplementary Table S4.

Survey method and quality control

The survey questionnaires underwent validation by experts from the Department of Clinical Psychology, Tianjin Anding Hospital, and the Department of Psychology, Zhejiang Normal University. Subsequently, the validated questionnaires were made available to participants in either paper or online format, based on their preferences. Standardized instructions were employed to mitigate bias arising from varied interpretations. Surveyors received ethical and technical training to ensure the strict confidentiality and proper handling of personal data. The data collection process adhered strictly to the principles outlined in the Helsinki Declaration and its subsequent amendments, under the supervision of the Ethics Committee of Tianjin Medical University General Hospital. To identify and exclude false or invalid questionnaires, deceptive detection questions were incorporated. Two experts (R.J. and Ch.L.) independently reviewed the questionnaires, addressing logical errors, significant data omissions, and resolving any ambiguities through discussion.

Statistical analysis

Continuous variables were described using mean with standard deviation (SD) or median with interquartile range (IQR) depending on distribution. Categorical variables were expressed as percentages. Based on their prior COVID-19 infection experience, the participants were categorized into Covid-19 family member cohort (infection group) versus control family member cohort (control group). The t-test or Mann–Whitney U test was employed to compare continuous variables, while Pearson χ² test and Fisher’s exact test were used to compare categorical variables between two groups.

Based on cutoff values (Supplementary Table S4), the results of scales were transformed into categorical variables. The artificial variable, termed the composite mental health symptoms (CMHSs), was defined to reflect the presence or absence of anxiety, depression, and PTSD. If at least one of these three symptoms was present, the CMHSs were classified as positive. The presence of mental health symptoms was examined through logistic regression analysis, and the results were presented as crude odds ratios (ORs) in univariable regression and adjusted odds ratios (aORs) in multivariable regression analyses, along with corresponding 95% confidence intervals (CIs). The subgroup analysis of basic information was used to explore the impact of each variable on the CMHSs between the two groups.

Multivariable logistic regression was used to analyze the relationship between potential risk factors in demographic characteristics and the occurrence of CMHSs. Only risk factors with a P-values <0.20 in univariable logistic regression analysis were included in the multivariable logistic regression (backward stepwise regression LR). The ORs, 95% CIs, and P-values were reported for both the univariable and multivariable regression analyses. Based on the results, a predictive model for risk factors of CMHSs was established. Receiver operating characteristic (ROC) curves and the area under the curve (AUC) were used to evaluate the discriminative ability, Calibration plots and the Hosmer & Lemeshow test were used to assess the calibration.

Questionnaires with severe missing data (less than half of the questions answered) were excluded, and missing data in the scale results were imputed with mean values. The significance level (α) for all statistical results in this study was set at a two-tailed P-values <0.05. Data analysis was performed using IBM SPSS version 26 (Chicago, IL, USA) and R version 4.2.2.

Results

Baseline characteristics

The average age of these participants was (39.20 ± 15.68) years, with 192 (41.5%) females. The baseline demographic characteristics of both groups are presented in Table 1.

Distribution of scale scores

Table 2 presents the results indicating that the experimental group exhibited significantly higher scores on the SDS, SAS, IES-R, and PSQI scales compared to the control group (SDS: 51.96 ± 3.743 vs. 48.97 ± 6.756, P < 0.001; SAS: 49.51 ± 4.725 vs. 45.25 ± 9.544, P < 0.001; IES-R: 22[21–25] vs. 21[17–24]; PSQI: 5.78 ± 1.426 vs. 4.72 ± 1.803, P < 0.001). However, no significant differences were observed between the two groups in the scores of the PSS, CD-RISC, and SCSQ scales. After adjusting for potential confounding factors, the multivariable logistic regression analysis revealed that the experimental group had a significantly higher occurrence rate of CMHSs, depression, and poor sleep quality (Table 3). Although the univariate logistic analysis showed significantly higher incidence of anxiety and PTSD compared to the control group, there were no significant differences after adjusting (P > 0.05). Subsequent post hoc exploratory subgroup analysis illustrated the influence of demographic characteristics on CMHSs between the two groups (Figure 1).

Risk factors associated with CMHSs

Multivariable logistic analysis revealed that female gender, residing in rural areas, and having children were significantly associated with an increased risk of experiencing CMHSs (Table 4). Furthermore, compared to individuals not engaged in any occupation, those involved in mental labor or physical labor occupations exhibited a higher risk of experiencing CMHSs. Conversely, a history of surgery was identified as a protective factor, as participants with a history of surgery had a lower risk of experiencing such symptoms.

Model for predicting CMHSs

Based on previous analysis, a predictive model was developed to identify risk factors for CMHSs. The model included five predictive factors: gender, occupation, residential area, having children, and history of surgery. The parameter estimates and test results are presented in Supplementary Table S5. The logistic regression equation is as follows: Logit(p)=−9.428 + 2.397 × Female sex+3.611 × Physical occupations+4.435 × Intellectual occupations+2.577 × Living in rural areas+2.980 × Having children−2.243 × Operation history. ROC curves of the predictive model and the five variables included were plotted. And the AUC and its 95% CI of the model and each variable were calculated: 0.918 (0.875–0.961), 0.579 (0.489–0.669), 0.635 (0.521–0.749), 0.662 (0.565–0.759), 0.693 (0.592–0.793), 0.624 (0.624–0.727) respectively (Supplementary Figure S3). The calibration plot demonstrated a close correspondence between the predicted probabilities and the actual probabilities, and the Hosmer–Lemeshow goodness-of-fit test showed χ²=5.279, P = 0.727 > 0.05 (Supplementary Figure S4).

Discussion

In this nationwide cross-sectional study, the Covid-19 family member cohort showed a significantly higher prevalence of CMHSs compared to the control family member cohort (82.7% vs. 66.1%).These findings suggest that there may be an association between COVID-19 infection experience, considered as potential new traumatic stressors, and the exacerbation of mental health symptoms in these populations.

Previous study found that, although the severity of symptoms and health impairments diminish over time, up to 18% of the infected individuals are still affected by COVID-19 sequelae two years after the infection.¹² Our Covid-19 family member cohort observed a higher prevalence rate of 82.7% compared to the control family member cohort, the observed differences in prevalence rates could be influenced by COVID-19 traumatic sequelae. Additionally, we found a significantly higher prevalence of poor sleep quality in Covid-19 family member cohort (84.6%), indicating that previous COVID-19 infection experience also leads to decreased sleep quality among ICU family members. As a novel source of traumatic stress, COVID-19 exerts enduring effects not only on the individuals who have been infected but also on their close relatives. Among family members of patients admitted to ICU with acute respiratory distress syndrome (ARDS), COVID-19 infection, in comparison to other etiologies of ARDS, was found to be significantly associated with a heightened risk of PTSD symptoms at 90 days following discharge from the ICU.¹³ The considerable prevalence and severity of mental health symptoms during the ICU stay and PTSD symptoms after discharge can primarily be attributed to the suspension of ICU visits.^14,15

Considering the impact of FICUS, which is frequently observed in ICU family members, the detrimental effects of COVID-19 infection on mental health symptoms of ICU family members may represent another blow in addition to FICUS. This ‘double-hit’ or ‘cumulative effect’ explains why the incidence of mental health symptoms in Covid-19 family member cohort is relatively high.

Studies have reported that perceived stress, psychological resilience, and coping strategies can influence the occurrence of mental health symptoms.^16–21 However, in our study, there were no significant differences in perceived stress, psychological resilience, and coping strategies between the two groups, suggesting that COVID-19 infection may primarily affect participants through neurophysiological mechanisms rather than by altering other factors. It has been found that many patients experience long-term symptoms of depression, anxiety, PTSD, and insomnia following acute COVID-19 infection recovery.^7,22–24 This may be related to the ability of the coronavirus to infect the central nervous system through hematogenous or retrograde neuronal invasion pathways, leading to chronic neuroinflammation.^25,26 However, it is not yet clear whether those symptoms are caused by the viral infection itself or by the host immune response.²⁷

In the absence of considering the impact of COVID-19, risk factors for the occurrence of psychological problems among ICU family members include being female, younger age, types of kinship, and lower socioeconomic status.²¹ In our study, for ICU family members previously infected with COVID-19, multivariable logistic regression analysis showed that being female, engaging in physical/mental labor, living in rural areas, and having children were risk factors, while a medical history of surgery was a protective factor against the occurrence of these symptoms.

Regarding the variable of having children, we agree that it may not be a typical risk factor for FICUS. Caring for children while simultaneously coping with the stress and uncertainty of a family member’s critical illness in the ICU could contribute to the heightened psychological symptoms observed. Similarly, the association between rural residency and psychological symptoms among family members is worth exploring further. Living in rural areas might entail additional challenges, such as limited access to healthcare resources, reduced social support networks, and longer distances to the ICU, which could contribute to increased psychological burden. The protective factor refers to past medical history, that is, the patient’s health status and history of illness before medical treatment or admission. Family members with a previous history of surgery may have already experienced the ICU environment and procedures, thereby acquiring a certain level of familiarity and adaptability. This previous exposure could potentially alleviate the psychological stress experienced by their family members. Furthermore, these family members may have received some preoperative education in the past, which could contribute to reducing their anxiety and uncertainty.

This suggests that the impact of COVID-19 infection on mental health symptoms of different populations is variable, and healthcare professionals should pay more attention and provide support to ICU family members who have risk factors. It is worth noting that although many studies have reported the impact of quarantine experiences during the pandemic on individuals’ psychological health,^28,29 in our study, quarantine experience was not identified as a risk factor. This may be because, in the post-pandemic era, as epidemic prevention and control measures gradually become normalized and the public’s understanding and familiarity with the transmission capability of the COVID-19 and preventive measures increase, the perceived stress and stress responses related to quarantine among the Chinese population have decreased compared to the early stages of the pandemic and have become relatively stable.³⁰ According to multivariable logistic regression analysis, a predictive model was developed with five factors: gender, occupation, residential area, childbirth status, and surgical history. ROC analysis was used to assess the discriminative ability of the predictive model,³¹ while calibration plot analysis and the Hosmer–Lemeshow test were employed to evaluate the calibration.³² The individual AUCs for the five predictive factors were all below 0.70, suggesting a correlation rather than a causal relationship between these risk/protective factors and psychological health symptoms. However, the combined predictive model, incorporating all five risk/protective factors, yielded an AUC of 0.918, indicating a strong discriminative ability of the model. Both the calibration plot and Hosmer–Lemeshow test results showed no significant differences between the predicted probabilities and actual probabilities, indicating good calibration of the model.

Medical workers should not only be satisfied with reducing the mortality rate of ICU patients, but should also prioritize the patients’ good post-recovery and high quality of life as their pursuit. This is in line with the requirements of the modern medical model (biopsychosocial approach). Given the complex etiology and clinical manifestations of FICUS, compounded by the long-term effects of ‘long COVID’, it undoubtedly exacerbates the physical and mental burden on the patients’ families. Currently, research on PICUS in many countries is still in the early stages, requiring further investigation to propose effective and feasible intervention measures. Commonly employed intervention measures at present include patient and family-centered care (PFCC) and ICU diaries.^4,33–35 PFCC, in particular, is highly necessary during major public health crises.³⁶ Establishing outpatient clinics can also effectively address FICUS, but the outpatient system for FICUS in China has not yet been established and needs to be gradually improved and popularized. It is urgently needed to actively engage in interdisciplinary research to benefit patients.

Limitations

However, our study also has several limitations. Firstly, this is a cross-sectional study, and the results only reflect the short-term impact. Follow-up studies are needed to determine the long-term effects. Moreover, the associations between mental health symptoms and risk/protective factors do not necessarily imply causation. Additionally, the occurrence of mental health symptoms is based on self-report measures rather than clinical diagnosis, which may introduce certain biases.

Conclusions

Prior COVID-19 infection in ICU patients’ families is significantly associated with increased rates of depression, anxiety, PTSD, and other psychological symptoms, which may represent a second impact following FICUS. In the post-pandemic era, healthcare practitioners should be particularly attentive to the heightened prevalence of psychological health concerns, especially among vulnerable populations with risk factors. Subsequent studies should concentrate on devising efficacious interventions to mitigate the occurrence of psychological symptoms among ICU families affected by COVID-19 and promote their well-being throughout the patient care journey.

Supplementary material

Supplementary material is available at QJMED online.

Acknowledgements

We appreciate the participation of all the family members. In addition, we thank all the ICU staff involved in the hospital for their valuable efforts in this study. MD Liu would like to give special thanks to PhD, Jing Hu, Tongji University, MD, Renmin Xue, Capital University of Medical, MD, Renzhi Li, Rheinisch-Westfaelische Technische Hochschule Aachen, MD, Xudong Zhang, Fudan University, MD, Pengyu Chen, Peking Union Medical College Hospital, MD, Wentai Zhang, Peking University, MD, Hui Jiang, Zhejiang University, MD, Chao Li, Tianjin Medical University General Hospital, MD, Qiuyu Zhang, Tianjin Medical University General Hospital, MSc, Yuzhi Li, Zhejiang Normal University, they was not compensated for their contribution.

Author contributions

Tao Liu (Conceptualization [lead], Data curation [lead], Formal analysis [lead], Investigation [lead], Methodology [lead], Project administration [lead], Writing—original draft [lead], Writing—review & editing [lead]), Zhihao Zhao (Conceptualization [equal], Data curation [equal], Formal analysis [equal], Funding acquisition [equal], Investigation [equal], Methodology [equal], Project administration [equal], Software [equal], Validation [equal], Visualization [equal], Writing—original draft [equal], Writing—review & editing [equal]), Chenrui Wu (Conceptualization [equal], Data curation [equal], Investigation [equal], Methodology [equal]), Chenghao Lu (Data curation [equal], Investigation [equal], Methodology [equal], Project administration [equal]), Mingqi Liu (Conceptualization [equal], Data curation [equal], Investigation [equal], Methodology [equal]), Xu An (Data curation [equal], Funding acquisition [equal], Methodology [equal], Validation [equal], Writing—original draft [equal]), Zhuang Sha (Investigation [equal]), Xin Wang (Investigation [equal], Methodology [equal]), Zhenyu Luo (Investigation [equal]), Lvjian Chen (Investigation [equal]), Chenglong Liu (Investigation [equal], Methodology [equal]), Peiyu Cao (Investigation [equal], Methodology [equal]), Dewei Zhang (Resources [equal], Software [equal], Supervision [equal], Validation [equal], Visualization [equal], Writing—original draft [equal]), and Rongcai Jiang (Project administration [equal], Software [equal], Supervision [equal], Validation [equal], Visualization [equal], Writing—original draft [equal], Writing—review & editing [equal])

Funding

This work was supported by grants from the National Natural Science Foundation of China (via Grant No. 82071390 to R Jiang, Grant No. 82101434 to J Huang, Grant No. 82001323 to C Gao).

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Ethics statement

This research was approved by the Ethics Committee of Tianjin Medical University General Hospital (No. IRB2023-wz-068).

Conflict of interest

None declared.

References

Author notes