As care for critically ill patients has advanced, so too has our recognition of the multidimensional and persistent morbidities that affect patients and their families following an intensive care unit (ICU) stay (1). Survivors of critical illness experience impairments across multiple domains of health—physical, cognitive, emotional, social, and financial—that can persist months to years after hospital discharge (2–4). Family members frequently serve as surrogate decision-makers for their loved ones in the ICU, and many become caregivers for their loved ones after hospital discharge. Increasingly, the critical care community has become aware of the impacts of these experiences on family members of ICU survivors and decedents (5–7), recognizing that ICU family members experience high rates of anxiety, depression, posttraumatic stress symptoms (PTSS), and prolonged or complicated grief, outcomes that have been very difficult to mitigate in intervention studies to date (1, 5).

In this issue of AnnalsATS, Wendlandt and colleagues (pp. 782–793) challenge the assumption that measures of PTSS are the most important family-centered outcomes for ICU family caregivers (8). In their qualitative study of 21 family caregivers, they explored how participants defined experiences of wellness and distress 6 months after a loved one’s ICU admission using semistructured interviews. In so doing, they expand our understanding of how to describe post-ICU discharge outcomes for family members and identify opportunities for more complete measurement and eventual intervention development. They identified four main outcome themes: 1) positive versus negative physical and psychological outcomes (which overlap well with existing measures of PTSS), 2) high versus limited capacity for self-care, 3) comfort versus struggle in the caregiving role, and 4) regaining a sense of normalcy versus ongoing life disruption. Additionally, participants described the experiences that served as sources of distress as well as moderators that influenced family members’ resiliency and adaptation to stress events. Importantly, the key stress events as well as the identified moderators often involved events or circumstances that extended beyond the ICU admission.

A key strength of this study is that, in its design and implementation, the investigators center the voices of family caregivers. By allowing participants to define ICU family caregiver outcomes of wellness and distress in their own words, we gain insight into outcome domains that have not traditionally been assessed in ICU research. Although there has been increasing interest in understanding resiliency and adaptation among ICU survivors, this study illustrates the relevance of these constructs in family caregivers (9, 10). The clear description of rigorous qualitative methods and application of the well-established chronic traumatic stress framework are additional strengths of the study.

One central limitation of this study, and of many prior intervention studies seeking to impact the long-term outcomes of ICU family members, is the inability to disentangle the differential exposure to and impacts of: 1) family members’ experiences within the ICU (e.g., surrogate decision-making, communication with the medical team, observing their loved one in distress or pain), 2) longitudinal caregiving for a critical illness survivor who is experiencing new or worsened disability outside of the hospital and who may have repeated hospital or ICU admissions, 3) bereavement after an in-ICU or post-ICU death, and 4) the everyday-life stressors that continue even when under the stress of a loved one’s critical illness. Wendlandt and colleagues importantly recognize that these exposures cannot be studied in isolation, and therefore chose a conceptual model that encompasses multiple stress events. In prior work, Wendlandt and colleagues demonstrated the association between patients’ recovery trajectories and surrogates’ reports of mental health among surrogates of patients with chronic critical illness (11). Their findings, in this study and the prior one, suggest that there may be important differences among family members whose loved ones survived versus died and who recovered versus had persistent caregiving needs.

Other studies have attempted to isolate populations with more uniform stressors, such as family surrogate decision-makers whose loved ones have died in the ICU (12, 13). Even among this more uniform population (i.e., without variability in patient recovery trajectory or in the stressor of longitudinal caregiving after the ICU stay), these studies have illustrated the complex and interwoven trajectories of prolonged grief, posttraumatic stress disorder, and depression for bereaved family surrogate decision-makers using quantitative metrics. However, they have lacked the rich qualitative data that might have provided insights about individual stressors and modifiers that may shape distinct trajectories for each bereaved surrogate decision-maker.

To leverage the strengths of quantitative and qualitative data, future research ought to capitalize on mixed-methods designs. Sufficient sample sizes and prospective data collection to stratify analyses across key exposures over time while ensuring thematic saturation within relevant subgroups would strengthen our ability to identify specific relationships and develop tailored interventions. To support further thinking and research, we have created a draft conceptual model that builds on Wendlandt and colleagues’ adaptation of the chronic traumatic stress framework and qualitative findings (8) (Figure 1).

Figure 1. A conceptual model depicting relationships between intensive care unit (ICU) stressors and outcomes of wellness and distress among family members of ICU patients. Family members of ICU patients are at risk of stress events related to the ICU admission, the aftermath of the ICU stay, and everyday life stressors. Different colors (blue, purple, red) represent hypothesized relationships between different exposures and outcomes. Future research is needed to disentangle these relationships between specific stress events and family member outcomes, thus informing targeted interventions.

Download Figure | Download Powerpoint

[More] [Minimize]

In this model, family members of ICU patients are at risk of exposure to stress events related to the ICU admission, the aftermath of the ICU stay, and everyday life. The intensity of each of these stressors may vary with time, and, as such, reported family outcomes are also likely to change in relation to such stressors, highlighting the importance of longitudinal data. Moderators are also likely to be present and influence outcomes and trajectories throughout. Future research is needed to disentangle the relationships among specific stress events, moderators, and family member outcomes, thus informing the interventions that can be targeted to the specific stressors and the study designs that can target the relevant outcomes. For example, investigations of the impact of bereavement after the death of a loved one during or after a critical illness requires extended longitudinal follow-up, as outcomes assessed at 6 months are inherently unable to distinguish between prolonged grief disorder versus a “normal” grief reaction after the loss of a loved one (14, 15).

Individual family members may also be exposed to different stressors, depending on their role in supporting or caring for their loved one. Thus, recognizing the presence of varied family structures and division of caregiving responsibilities may also be informative for future research. For example, the family member who serves as surrogate decision-maker in the ICU and is susceptible to that stressor may be a different family member than the one who assumes primary caregiver duties after ICU discharge, and, for many patients, these responsibilities may be shared across a family unit (16). Our conceptual model attempts to highlight that these differential exposures may represent opportunities for unique, tailored interventions, even within the same family.

In conclusion, Wendlandt and colleagues have provided an important contribution about family caregiver experiences after an ICU admission of a loved one. Their work highlights the need for ongoing study in this field to better elucidate the different impacts of different, but frequently concurrent, stressors on family members: communication and surrogate decision-making in the ICU, longitudinal caregiving for a critical illness survivor, and bereavement. Moving forward, the intentional design of mixed-methods studies will help advance this understanding, which will be crucial to informing potential interventions to improve care for patients and family members after critical illness.