Reply to Shi et al.: ICU visitation policies: no harm found, or harm missed?

From the Authors: We thank Shi and colleagues for their interest in our study1 and for their comments. We would like to address several points they raise. First, we acknowledge the apparent contrast between our findings in families and factors associated with psychological symptoms among intensive care unit (ICU) clinicians during the COVID-19 pandemic.2 We believe these results are not contradictory but rather complementary. Clinicians’ regret about restricted visitation policies and their experiences during end-of-life care reflect professional challenges and ethical dilemmas, which may understandably affect their own psychological well-being. Our study, in contrast, specifically assessed family members 3 months after ICU discharge. The different populations, outcomes, and methodologies limit direct comparison but together provide a broader perspective. Second, regarding potential sources of bias, illness severity scores or sedation/consciousness levels are relevant to patient outcomes but are not established predictors of long-term psychological symptoms in relatives. In contrast, variables such as invasive mechanical ventilation, relationship to the patient, patient death, and gender of the relative are consistently associated with family psychological outcomes,3 and these were collected in our study. Patient-to-nurse and patient-to-intensivist ratios remained stable and consistent with French guidelines, further reducing confounding from variations in care resources. The quasi-experimental design, based on 3 successive national policies, was the most pragmatic and ethically sound approach during the pandemic. It would not have been feasible in 2020 to randomize ICUs within the same country to divergent visitation policies. Nonresponse remains a limitation, but it is unlikely to have substantially altered our findings because our participants were representative of the overall study population in terms of major characteristics such as mortality, COVID-19 status, age, gender, and the relationship with the patient. Third, regarding videoconferencing, we agree that not all remote visitation programs are equivalent. Our study did not find long-term benefits in psychological symptoms, whereas a recent Korean randomized trial reported short-term improvements with a structured protocol.4 Although sociocultural differences among countries might explain this discrepancy, this highlights the need for further work to evaluate the design, timing, and long-term impact of remote communication strategies. Finally, we concur that, in retrospect and based on several years of experience with the COVID-19 pandemic, the beneficial effect of visitation restrictions in preventing COVID-19 infection may have been limited. However, during the first wave of the COVID-19 pandemic, when uncertainties were high, such measures were widely endorsed by health authorities and experts. Our results suggest that these restrictions, although difficult, were not a major determinant of long-term psychological symptoms in families. Instead, nonmodifiable factors had a greater impact, and the resilience of relatives in this crisis deserves recognition. These findings provide robust data that can inform balanced decisions on visitation policies in future global health crises. A.L., B.P., and J.-C.R. conceptualized and wrote the original draft. M.C., N.E., and C.D. reviewed and edited the draft. All authors approved the final version of the manuscript. Supplementary material is available at American Journal of Respiratory and Critical Care Medicine online. We declare no conflicts of interest. Please see the ICMJE disclosure forms, which have been provided as supplementary material.