Abstract
Objective:
While palliative care needs are assumed to improve during intensive care unit (ICU) care, few empiric data exist on need trajectories or their impact on long-term outcomes. We aimed to describe trajectories of palliative care needs during ICU care and to determine if changes in needs over 1 week was associated with similar changes in psychological distress symptoms at 3 months.
Design:
Prospective cohort study.
Setting:
6 adult medical and surgical ICUs.
Participants:
Patients receiving mechanical ventilation for ≥2 days and their family members.
Measurements and Main Results:
The primary outcome was the 13-item Needs at the End-of-Life Screening Tool (NEST; total score range 0-130) completed by family members at baseline, 3, and 7 days. The Patient Health Questionnaire-9 (PHQ-9), Generalized Anxiety Disorder-7 (GAD-7), and Post-Traumatic Stress Scale (PTSS) were completed at baseline and 3 months. General linear models were used to estimate differences in distress symptoms by change in need (NEST improvement ≥10 points or not). 159 family members participated (median age 54.0 years, [IQR 44.0, 63.0], 125 [78.6%] female, 54 [34.0%] African American). At 7 days 53 (33%) a serious level of overall need and 35 (22%) ranked ≥1 individual need at the highest severity level. NEST scores improved ≥10 points in only 47 (30%). Median NEST scores were 22 (IQR 12, 40) at baseline and 19 (IQR 9, 37) at 7 days (change −2.0 [IQR −11.0, 5.0; p=0.12). There were no differences in PHQ-9, GAD-7, or PTSS change scores by change in NEST score (all p>0.15).
Conclusions:
Serious palliative care needs were common and persistent among families during ICU care. Improvement in needs was not associated with less psychological distress at 3 months. Serious needs may be commonly underrecognized in current practice.
Keywords: palliative care, palliative care needs, clinical palliative care triggers, care delivery models, psychological distress
Introduction
Critically ill patients and their family members commonly experience serious stressors and symptoms during their time in the intensive care unit (ICU) that may also persist long after discharge. (1, 2) Palliative care provided by either ICU clinicians or palliative care specialists is therefore an important component of critical care because it aims to optimize quality of life by addressing physical, emotional, social, and spiritual needs. (3, 4)
Reliably delivering high-quality palliative care in ICU settings is challenging because there are no replicable systems able to both ascertain the presence, severity, or type of needs among patients and their family members and then sensibly align the appropriate level of support from ICU teams and palliative care specialists. (5, 6) During the past decade, health systems increasingly implemented protocols that triggered palliative care specialist consultation based on clinical characteristics associated with mortality and resource utilization that were assumed to be reasonable proxy measures of need. (7) However, we recently observed that clinical trigger status assessed within the first few days of the ICU course was not associated with the presence, severity, or type of palliative care needs reported directly by patients’ family members. (8)
While this observation would seem to raise concern about contemporary trends in the organization of inpatient palliative care delivery, it may be an incomplete characterization. Given the cross-sectional nature of this work, it remains unclear if needs change over time or if clinical triggers may in fact predict incident needs later in ICU care. Additionally, it is unclear if long term outcomes such as post-discharge psychological distress symptoms are associated with the presence of unmet needs or clinical palliative care triggers present during ICU care.
To address this knowledge gap, we conducted a prospective cohort study to characterize trajectories of needs during ICU care and to test associations between need trajectories and post-discharge symptoms of depression, anxiety, and post-traumatic stress disorder (PTSD). We hypothesized that needs would generally improve during ICU care and that compared to family members without improvement in needs, those whose needs improved would have less psychological distress 3 months later.
Materials and Methods
Study design, setting, and participants
This observational prospective cohort study with 3-month follow up was conducted in 6 adult medical and surgical ICUs in one academic hospital and one large community hospital in the Duke University Health System between January 2019 - January 2021 (STROBE checklist in Digital Supplemental Content). Written informed consent was obtained directly from each patient’s legal representative, generally a family member, per Duke University Institutional Review Board Protocol #00090202.
Study staff screened all consecutive patients in study ICUs who were ≥18 years of age and had received mechanical ventilation for ≥48 hours. This strategy was chosen to ensure that an ICU team-led family meeting could be conducted within an expert guideline-recommended timeframe before completion of the initial study survey. (9) Exclusions were attending physician expectation of a transition to fully comfort-focused care or death within 24 hours, patient decisional capacity, patient imprisonment, and absence of a family member available in person or by phone to participate in study activities. We enrolled one adult family member per patient self-described as the individual most involved in the patient’s care. Family members were excluded if they reported insufficient confidence in their English fluency to complete study surveys.
Data collection and outcomes
Surveys were self-completed by family members through unique secure weblinks emailed or texted from the study data system at the time of consent as well as 3 days, 7 days, and 3 months later. (10) Study staff recorded clinical data from the electronic medical record. ICU clinicians were not aware of survey results.
Family member-completed measures
The Needs at the End-of-Life Screening Tool (NEST) was the primary study outcome. (10–12) The 13-item NEST (score range 0 [no need] to 130 [highest need]) measures needs from all eight core domains of palliative care quality including structure and processes of care, physical and psychological symptoms, social support, spiritual and cultural aspects of care, end-of-life care, and ethical aspects of care. (13) Our previous work demonstrated the NEST’s adaptation to the ICU setting, evidence of content and criterion validity, and association with measures of psychological distress. (8, 10) We defined a clinically important decrease in palliative care needs as a reduction in NEST total score of ≥10 units between baseline and 7 days later, a value that is nearly identical to the median change value of 9 as well as a 25-30% reduction in baseline score observed in past ICU-based observation and intervention research. (8, 10)
Our secondary outcomes of interest were symptoms of depression, anxiety, and post-traumatic stress disorder (PTSD) measured with the Patient Health Questionnaire 9-item scale (PHQ-9; range 0 [lowest] to 27 [highest]), (14) the Generalized Anxiety Disorder 7-item scale (GAD-7; range 0 [lowest] to 21 [highest]), (15) and the Post-Traumatic Stress Syndrome inventory (PTSS; range 10 [lowest] to 70 [highest]), (16) respectively, at baseline and 3 months.
Other measures of family perceptions of ICU care included the receipt of goal-concordant care (concordance in actual vs. preferred treatment; baseline and day 3), (17) the Quality of Communication scale summary item (range 0 [worst] to 10 [best]; baseline and day 3), (18) the Interpersonal Processes of Care (IPC) instrument’s Patient-Centeredness of Decision Making, Eliciting Concerns, and Discrimination scales (scale score range 1 [worst] to 5 [best]; baseline and day 3); (19) family members’ relationship with the ICU attending physician (baseline and day 3); expected hospital survival for the patient (baseline and day 3); and degree of financial stress. (20)
Patient-level clinical variables
In addition to sociodemographic variables, clinical characteristics, (21) and descriptors of the ICU and hospital stay, staff recorded from the medical record the presence or absence of any of 9 clinical triggers for palliative care consultation present within the first 48 hours of ICU admission derived from the medical literature; (7, 22–24) details in Supplement Table 1.
Statistical analysis
We assessed the associations of sociodemographic and clinical characteristics with primary and secondary outcomes using a univariable logistic regression, with a dichotomous variable indicating clinically important change in the survey score as the outcome (baseline – 7 days for NEST and baseline – 3 months for psychological distress symptom scale secondary outcomes) and each characteristic as the single covariate. Odds ratios (ORs) and 95% confidence intervals (CIs) are reported for each comparison. We conducted one sample t-tests to determine if the continuous survey scores differed significantly between baseline and the subsequent time point of interest.
We used general linear models to examine the association between improvement in NEST scores and 3-month changes in psychological distress symptom scores (GAD-7, PHQ-9, and PTSS). Each model’s parameters included the improvement in NEST score (improved=NEST decrease ≥10 points from baseline to day 7, not improved=NEST did not decrease ≥10 points), time (0 days, 3 days, 3 months), and the interaction between NEST improvement and time. An unstructured covariance was used to account for the correlation of participants’ repeated survey measurements over time. For both NEST improvement status groups at each time point we calculated mean estimates, group differences, and differences in differences with 95% confidence intervals. Analyses were performed using SAS software version 9.4 (SAS Institute Cary, NC).
Results
Patient and family member characteristics
Among 1,179 consecutive patients screened, 360 potentially eligible patient-family member dyads were approached for enrollment, 262 (71.4 %) provided informed consent, and 159 dyads (61%; 318 participants) with complete data were included in analyses. Family members were predominantly middle-aged (median age 54.0 years, IQR 44.0, 63.0), female (n=125, 78.6%), patients’ spouse or partner (n=71, 44.6%), and attained at least some college education (n=78, 49.4%); Table 1. Patients were generally middle-aged (median age 57.0 years, interquartile range [IQR] 46.0, 68.0), male (n=86, 54.1%), White (n=88, 55.4%) or Black or African American race (n=55, 34.6%), and managed in a medical ICU (n=65, 40.9%); 59 (37.1%) met at least one palliative care trigger condition (Supplement Table 2). No dyad had received a palliative care consultation at the time of informed consent.
Table 1:
Family member characteristics by change in needs between baseline and 1 week later
| No decrease in needs n=112 (70%) |
Decrease in needs n=47 (30%) |
Total n=159 (100%) |
Odds ratio (95% CI) |
p | |
|---|---|---|---|---|---|
| Age, median (IQR) years | 53.5 (42.5, 61.0) | 56.0 (46.0, 65.0) | 54.0 (44.0, 63.0) | 1.01 (0.99, 1.04) | 0.254 |
| Gender, n (%) | |||||
| Female | 86 (76.8%) | 39 (83.0%) | 125 (78.6%) | 1.42 (0.59, 3.42) | 0.438 |
| Male | 25 (22.3%) | 8 (17.0%) | 33 (20.8%) | ref | |
| Transgender | 1 (0.9%) | 0 (0.0%) | 1 (0.6%) | -- | |
| Race, n (%) | |||||
| Black or African American | 36 (32.1%) | 18 (38.3%) | 54 (34.0%) | 1.23 (0.60, 2.55) | 0.598 |
| White | 64 (57.1%) | 26 (55.3%) | 90 (56.6%) | ref | |
| Other or more than one | 6 (5.4%) | 0 (0.0%) | 6 (3.7%) | -- | |
| American Indian or Alaska Native | 2 (1.8%) | 2 (4.3%) | 4 (2.5%) | -- | |
| Asian | 4 (3.6%) | 0 (0.0%) | 4 (2.5%) | -- | |
| Native Hawaiian or Pacific Islander | 0 (0.0%) | 1 (2.1%) | 1 (0.6%) | -- | |
| Hispanic or Latinx ethnicity, n (%) | 5 (4.5%) | 2 (4.3%) | 7 (4.4%) | 0.95 (0.18, 5.09) | 0.954 |
| Married or live with partner, n (%) | 78 (69.6%) | 32 (68.1%) | 110 (69.2%) | 1.08 (0.52, 2.24) | 0.846 |
| Relationship to patient, n (%) | 0.488 | ||||
| Spouse or partner | 52 (46.4%) | 19 (40.4%) | 71 (44.6%) | 1.28 (0.64, 2.55) | |
| Other | 60 (53.6%) | 28 (59.6%) | 88 (55.4%) | ref | |
| Employed or housemaker, n (%) | 82 (73.2%) | 25 (53.2%) | 107 (67.3%) | 2.41 (1.18, 4.89) | 0.015 |
| Education college degree or more, n (%) | 60 (54.0%) | 20 (42.6%) | 80 (50.6%) | 0.63 (0.32, 1.25) | 0.188 |
| Financial distress, n (%) | 54 (48.2%) | 21 (44.7%) | 75 (47.2%) | 0.87 (0.44, 1.72) | 0.684 |
| Expect hospital survival, n (%) | 0.020 | ||||
| Almost certainly | 69 (61.6%) | 26 (55.3%) | 95 (59.8%) | Ref | |
| Most likely | 42 (37.5%) | 16 (34.0%) | 58 (36.5%) | 1.01 (0.45, 2.22) | |
| Probably or almost definitely not | 1 (0.9%) | 5 (10.6%) | 6 (3.8%) | 12.90 (1.35, 636.36) |
Values shown as n (%), median (interquartile range), and odds ratio (95% confidence interval [CI]). Odds ratios and p values from univariable logistic regression.
Overall improvement in needs during ICU care
Fewer than a third (n=47, 29.6%) of family members had NEST total scores that improved ≥10 points between baseline and day 7 (Table 2). Among those whose total NEST scores improved, the individual NEST items that most frequently improved were patient discomfort (n=37, 78.7%; mean change −1.7 [SD 2.8]), financial distress (n=32, 68.0%; mean change −2.3 [SD 2.9]), uncertainty about what to expect (n=31, 66.0%; mean change −2.3 [SD 3.0]), spirituality (n=30, 63.8%; mean change −2.0 [SD 3.4]), decision making (n=30, 63.8%; −1.7 [SD 3.1]), and anxiety and stress (n=29, 61.7%; mean change −2.0 [SD 3.1]; Supplement Figure 1). Among the 112 (70.4%) whose NEST total score did not improve significantly, the needs most likely to worsen included patient discomfort (n=45, 40.2%; mean change 0.5 [SD 2.8]), lack of participation in decision making (n=41, 36.6%; 0.9 [SD 2.4]), and uncertainty in what to expect (n=36, 32.1%; mean change 0.2 [SD 2.9]).
Table 2:
Primary and secondary outcomes at each follow up
| Outcome | Baseline n=159 |
3 days n=159 |
7 days n=159 |
3 months n=159 |
Change | p |
|---|---|---|---|---|---|---|
| NEST | 0.117 | |||||
| Mean (SD) | 28.2 (21.4) | 26.2 (20.8) | 25.8 (22.2) | - | −2.4 (19.0) | |
| Median (IQR) | 22.0 (12.0, 40.0) | 20.0 (11.0, 37.0) | 19.0 (9.0, 37.0) | - | −2.0 (−11.0, 5.0) | |
| Range | 0.0-103.0 | 0.0-107.0 | 0.0-113.0 | - | −71.0 - 82.0 | |
| ≥10-point improvement, n (%)* | 47 (29.6%) | |||||
| PHQ-9 | 0.001 | |||||
| Mean (SD) | 7.1 (6.0) | 6.5 (5.5) | - | 5.7 (5.5) | −1.5 (5.7) | |
| Median (IQR) | 5.0 (3.0, 11.0) | 5.0 (2.0, 9.0) | - | 4.0 (2.0, 8.0) | −1.0 (−4.0, 2.0) | |
| Range | 0.0-25.0 | 0.0-27.0 | - | 0.0-24.0 | −19.0-16.0 | |
| Moderate depression symptoms (score ≥10) | 47 (29.6%) | 36 (22.6%) | - | 38 (23.9%) | −6.0% | |
| ≥3-point improvement, n (%)† | 98 (64.5%) | |||||
| GAD-7 | <0.001 | |||||
| Mean (SD) | 7.4 (5.8) | 6.8 (5.6) | - | 5.0 (4.9) | −2.4 (5.7) | |
| Median (IQR) | 6.0 (3.0, 11.0) | 6.0 (2.0, 10.0) | - | 4.0 (1.0, 7.5) | −2.0 (−6.0, 1.0) | |
| Range | 0.0-21.0 | 0.0-21.0 | - | 0.0-19.0 | −18.0-13.0 | |
| Moderate anxiety symptoms (score ≥10) | 51 (36.4%) | 42 (26.4%) | - | 35 (22.0%) | −14.4% | |
| ≥3-point improvement, n (%)† | 82 (54.3%) | |||||
| PTSS | 0.498 | |||||
| Mean (SD) | 21.7 (11.6) | - | 21.4 (12.3) | −0.6 (11.3) | ||
| Median (IQR) | 18.0 (13.0, 28.0) | - | 16.5 (12.0, 26.0) | −1.0 (−6.0, 5.0) | ||
| Range | 10.0-70.0 | - | 10.0-60.0 | −34.0-38.0 | ||
| Moderate PTSD symptoms (score ≥25) | 47 (29.6%) | - | 49 (30.8%) | 1.2% | ||
| ≥5-point improvement, n (%)† | 106 (69.7%) |
Between baseline and 7 days
Between baseline and 3 months
GAD-7: Generalized anxiety disorder 7-item scale, ICU: intensive care unit, IQR: interquartile range, NEST: Needs at the End-of-Life Screening Tool, PHQ-9: Patient health questionnaire 9-item depression scale, PTSD: Post-traumatic stress disorder, PTSS: Post-traumatic stress scale, SD: standard deviation.
Trajectories of overall need severity during ICU care
At baseline, 57 (36%) had a serious level of need (NEST total score ≥30) and a median NEST of 51.0 (36.5, 57.5), and by 7 days there were still 53 (33%) with serious need (median NEST 43.0 [37.0, 59.0]; Figure 2). Similarly, at baseline 48 (30%) had at least one NEST item ranked at the highest severity (item score=10), falling slightly to 32 (22%) at 7 days (Supplement Figure 2).
Figure 2: Serious individual palliative care needs reported during ICU care.
The percentage of family members with a NEST item score ≥5 is shown for each 13 of the NEST’s items across the time span of the study at baseline, day 3, and day 7 day. An asterisk denotes the two items that increased in prevalence between baseline and day 3.
Among all participants, median NEST total scores declined from 22.0 (IQR 12.0, 40.0; range 0-103) at baseline to 19.0 (IQR 9.0, 37.0; range 0-113) at one week, a change of −2.0 (IQR −11.0, 5.0; range −71 to 82; p=0.117) that was not statistically significant (Table 2, Supplement Figure 3).
Family members’ overall severity of needs based on total NEST scores was further categorized as high (NEST ≥30; n=57, 36%), moderate (NEST 10-29; n=74, 46%), low (NEST 1-9; n=25, 16%), or no needs (NEST 0; n=3, 2%). There were notable changes in need severity categories over time at the individual level, with a third of participants changing severity category during ICU care through 25 unique trajectories of need (Figure 1). NEST total score improvement over a week was greatest for those with high baseline needs (mean −9.6 [95% CI −15.9, −3.3]), though minimal for those with lower baseline needs (Supplement Figure 4).
Figure 1: Trajectories of family members’ severity of overall needs during ICU care.
Participants’ need severity category as defined by NEST scores (high: NEST ≥30, moderate NEST 10-29, low: NEST 1-9, none: NEST 0) are shown across the time span of the study. The height of boxes is proportional to the percentage in each category; the number (percent) of participants is shown at T1 (baseline), T2 (3 days), and T3 (1 week). Lines between boxes demonstrate the number of participants who either remain in the same need category or transition to another severity category; the width of the line is proportional to the number (also shown to the left of each severity grouping) entering a category at each time point. For example, of the 57 who had high needs at baseline, 41 had persistently high needs at 3 days and 35 had persistently high needs at 1 week. The number (%) worsened, improved, or unchanged in need severity category between baseline and day 7 is shown to the far right of the figure.
Individual needs and change across ICU care
Figure 3 shows that individual NEST items reflecting serious needs (item score ≥5) that were most frequently reported among family members during ICU care included financial stress (range 50%−57%), spirituality (range 44%−58%), uncertainty about what to expect (range 19-32%), perceived patient discomfort (25%−38%), and anxiety and stress (19%−23%). The serious needs that improved most frequently were uncertainty about what to expect (39%), questions not answered (25%), and spiritual needs (24%), while the need most likely to worsen was not being included in decision making (24%).
Figure 3: Serious total palliative care needs versus less serious needs during ICU care.
The number (%) of family members reporting serious needs (NEST total score ≥30) and needs that were less serious (NEST total score <30) at baseline, 3 days, and 7 days are shown. Median (interquartile range) NEST total scores are also shown for each group. At 7 days, the 53 still reporting serious needs had a median NEST total score of 43.0 (37.0, 59.0). The 106 with less serious needs had a median NEST total score of 12.5 (6.0, 19.0).
Clinical characteristics and outcomes by need category
Patient-family member dyads completed their initial surveys after a median of 4 (interquartile range 2, 5) days of ventilation. Patients were intubated generally within 24 hours of hospital admission (median 0 days [interquartile range 0, 1]). Median APACHE II scores were similar for those with no decrease versus a decrease in needs (21.0 [18.0, 25.5] vs. 23.0 [17.0, 28.0], p=0.40); Supplement Table 2. Days of ventilation, hospital duration, code status at discharge, and discharge disposition were similar by group (Supplement Table 3). The group with no decrease in needs had longer ICU duration in days (14.5 [8.0, 23.5]] vs. 11.0 [5.0, 17.0], p=0.048).
Long-term symptoms of psychological distress and association with needs
Moderate symptoms of depression (PHQ-9 score ≥10), anxiety (GAD-7 score ≥10), and PTSD (PTSS score ≥25) were reported at 3 months by 38 (23.9%), 35 (22.0%), and 49 (30.8%) family members, respectively (Table 2). Statistically significant improvement between baseline and 3 months was observed on the PHQ-9 (5.0 [3.0, 11.0] vs. 4.0 [2.0, 8.0], difference: −1.0 [−4.0, 2.0]; p=0.001) and GAD-7 (6.0 [3.0, 11.0] vs. 4.0 [1.0, 7.5], difference: −2.0 [−6.0, 1.0]; p <0.001), though not the PTSS (18.0 [13.0, 28.0] vs. 16.5 [12.0, 26.0], difference: −1.0 [−6.0, 5.0]; p=0.498); Table 2, Supplement Figure 3. Notable clinical and sociodemographic factors associated with changes of clinical importance in psychological distress symptoms scores are shown in Supplement Tables 4–7. However, model estimates of differences between baseline and 3 months in PHQ-9 scores were not significantly different between NEST improvement groups (NEST improved: −0.19 [−1.66,1.27] vs. NEST not improved: −1.39 [−2.34, −0.44]; difference in differences: 1.20 [−0.55, 2.94]; p=0.177); Table 3. Similarly, there was no significant difference in estimated mean GAD-7 scores between NEST groups at 3 months (−1.73 [−3.18, −0.28] vs. −2.25 [−3.19, −1.31]; difference in differences: 0.53 [−1.20, 2.25]; p=0.548) or PTSS scores (1.53 [−1.78, 4.83] vs. −1.37 [−3.52, 0.77]; difference in differences: 2.90 [−1.04, 6.84]; p=0.148).
Table 3:
Difference between baseline and 3 months in model-estimated means of psychological distress symptoms by improvement in needs during hospitalization
| Baseline | 3 months | Difference (3 months – baseline) | |
|---|---|---|---|
| PHQ-9 | |||
| Decrease in NEST ≥10 points | 6.24 (4.68, 7.80) | 6.05 (4.46, 7.64) | −0.19 (−1.66, 1.27) |
| No decrease in NEST ≥10 points | 6.93 (5.92, 7.94) | 5.54 (4.51, 6.57) | −1.39 (−2.34, −0.44) |
| Estimated difference | −0.69 (−2.55, 1.16) | 0.50 (−1.39, 2.40) | 1.20 (−0.55, 2.94) |
| p value for difference* | 0.462 | 0.600 | 0.177 |
| GAD-7 | |||
| Decrease in NEST ≥10 points | 6.86 (5.32, 8.41) | 5.14 (3.71, 6.56) | −1.73 (−3.18, −0.28) |
| No decrease in NEST ≥10 points | 7.22 (6.22, 8.22) | 4.97 (4.04, 5.89) | −2.25 (−3.19, −1.31) |
| Estimated difference | −0.36 (−2.20, 1.48) | 0.17 (−1.53, 1.87) | 0.53 (−1.20, 2.25) |
| p value for difference* | 0.703 | 0.844 | 0.548 |
| PTSS | |||
| Decrease in NEST ≥10 points | 20.38 (17.03, 23.74) | 21.91 (18.31, 25.51) | 1.53 (−1.78, 4.83) |
| No decrease in NEST ≥10 points | 22.31 (20.14, 24.49) | 20.94 (18.60, 23.27) | −1.37 (−3.52, 0.77) |
| Estimated difference | −1.93 (−5.93, 2.07) | 0.97 (−3.32, 5.26) | 2.90 (−1.04, 6.84) |
| p value for difference* | 0.342 | 0.655 | 0.148 |
p value is calculated from the general linear model. Values shown as model-estimated mean (95% CI).
GAD-7: Generalized Anxiety Disorder 7-item scale, NEST: Needs at the End-of-Life Screening Tool, PHQ-9: Patient Health Questionnaire 9-item depression scale, PTSS: Post-Traumatic Stress Scale.
Discussion
We observed that palliative care needs reported by ICU patients’ family members were common, complex, and persistent, with one-third still experiencing serious overall needs after a week of ICU care. These results are compelling because while palliative care needs are common, (8, 25) it has generally been assumed that they improve over time in ICU settings.
Relevance to clinical care and research
These findings highlight the complexity of developing a simple management solution for current gaps in care delivery. Despite greater awareness of palliative care principles and widely publicized recommendations by professional organizations, many family members had serious needs that remained unchanged or even worsened during ICU care. Furthermore, trajectories of need at the individual level were complex and dynamic, with frequent shifts in the severity and predominant types of needs. Yet few baseline sociodemographic and clinical factors were associated with either the presence of serious needs or change in these needs. There were no group differences in days of ventilation, hospital duration, code status at discharge, and discharge disposition. While the group with no decrease in needs had longer a ICU duration, the median length of stay exceeded the target day for the final in-hospital interview. A humbling takeaway is that contemporary ICU care may be suboptimally addressing basic needs that are important to patients and family members—needs that could also persist far beyond the ICU. It is unclear if the persistence of needs is related to clinician ‘blind spots,’ clinician attention focused on issues they incorrectly assume to be most problematic, or other factors. (26)
These results underscore the importance of developing better strategies to proactively identify and address unmet needs in clinical settings. They also provide greater clarity about which individual needs are most pervasive and static. Serious financial distress was reported throughout by most family members, closely followed by spiritual needs and not knowing what to expect. Among those whose needs did not improve, worsening of perceived patient discomfort was reported by 40%, with nearly a third reporting a worsened understanding of medical issues, distrust in medical information, failing to get questions answered, and lacking social support. Among the minority whose needs improved substantially, perception of patient discomfort and stress associated with financial strain—a construct that is often believed to be immutable—decreased notably over time. These results suggest that needs are potentially malleable targets for future interventions.
Clinical palliative care trigger status was not associated with either change in need or psychological distress symptoms, similar to our recent finding that baseline needs did not differ in severity between those with or without a trigger. (8) This adds further concern about using clinical triggers as a proxy for serious need in specialist palliative care protocols.
Are inpatient palliative care measures valuable?
Measures of core components of high-quality care including needs, quality of communication, and patient-centeredness of care were not associated with long-term psychological distress symptoms. In the greater context of two decades of ICU-based palliative care clinical trials often focused on post-discharge distress symptoms, many negative, this finding raises a fundamental methodological and philosophical question: is an inpatient outcome important if it does not demonstrate a strong relationship with person-centered outcomes measured many weeks or months later? We believe so.
First, addressing needs in the moment is a core aim of palliative care and these issues are of great importance to patients and family members. (27) Second, needs are clearly more proximal to an ICU-based palliative care intervention’s effect than are long-term psychological distress symptoms. In fact, these symptoms themselves often demonstrate weak or no association with many standard ICU outcomes that seem intuitively important and may be explained by the numerous stressors one may experience across inpatient and post-discharge periods. (20, 28–33) Third, individual NEST items including lack of social support and financial distress did demonstrate significant correlations with psychological distress symptoms similar to that observed in others’ past work, perhaps reflecting stressors that may have existed to some extent before hospitalization. (31, 34, 35) Fourth, needs are personal, specific, and dynamic, and therefore potentially valuable measures by which to inform person-centered care. In fact, the NEST instrument captured a wide range of need severity and type, demonstrated no ceiling or floor effect, was responsive to change in total score as well as severity category, and correlated strongly with quality of communication and patient-centeredness. As such, this metric could potentially play a useful role in future trials of ICU-based interventions. Last, knowledge of need severity and type could be used to inform future models of ICU-based palliative care delivery focused on efficiently aligning the skills of the ideal provider (e.g., nurse, chaplain, social worker, physician) with the needs of greatest importance. (8, 36)
Limitations
This study has limitations. Given its conduct in two hospitals in a similar geographic area, its findings may not be broadly generalizable. However, the cohort included a high proportion of African American participants (34%) and was drawn from medical and surgical ICUs in academic and community hospital settings. Also, while there is no gold standard for palliative care need assessment, the NEST’s items assess all eight core domains of high-quality palliative care. Additionally, although there is no accepted scoring threshold for defining serious need, our past work informed the cutoffs used in the current study. (10) A final limitation is the lack of non-English-speaking participants, whose palliative care needs may differ somewhat from those of English-speaking participants due to cultural variations in family structure, approaches to decision making, and palliative and end-of-life care needs.
Conclusions
Palliative care needs were common and persistent overall in a critical care setting, with one-third of family members reporting serious needs after a week of ICU care. Trajectories of need type and severity were also dynamic and complex at the individual family member level, likely reflecting a consistently evolving environment. Improvement in needs was not associated with improvement in symptoms of depression, anxiety, and PTSD at 3 months. Because needs may have been underrecognized based on the small longitudinal changes observed, knowledge of their type and severity could help to improve future models of palliative care delivery.
Supplementary Material
Key Points.
Question:
Because little is known about unmet palliative care needs in ICU settings, we aimed to determine if changes in needs during ICU care was associated with 3-month psychological distress symptoms.
Findings:
In this prospective cohort study of ICU patients and their family members, we found that unmet palliative care needs were very common and changed little in severity through the course of ICU care. However, those whose needs decreased in severity during ICU care did not experience fewer post-discharge depression, anxiety, or PTSD symptoms.
Meaning:
Serious palliative care needs are common and likely underrecognized in ICU settings, though not clearly associated with psychological distress symptoms months later.
Acknowledgements
The authors would like to thank the patients, family members, and ICU teams involved as well as Andrew Corcoran for his invaluable technological support.
Sources of funding and support:
U54 MD012530 (Johnson, Cox, Docherty), R01 AG058915 (Cox, Docherty), R21 NR016743 (Cox, Docherty)
Role of the funders/sponsors:
The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of data; preparation, review, or approval of the manuscript; or decision to submit the manuscript for publication.
Footnotes
Conflict of interest disclosure:
No author has potential conflicts of interest, including financial interests, activities, relationships, and affiliations
Meeting presentation:
Portions of this work addressing baseline NEST scores were presented on May 19, 2021 at the American Thoracic Society International Conference. Portions of this work addressing NEST scores and long-term outcomes were presented on May 17, 2022 at the American Thoracic Society International Conference.
References
- 1.Cameron JI, Chu LM, Matte A, et al. : One-year outcomes in caregivers of critically ill patients. N Engl J Med 2016; 374(19):1831–1841 [DOI] [PubMed] [Google Scholar]
- 2.Schofield-Robinson OJ, Lewis SR, Smith AF, et al. : Follow-up services for improving long-term outcomes in intensive care unit survivors. Cochrane Database Syst Rev 2018; 11:CD012701. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Truog RD, Campbell ML, Curtis JR, et al. : Recommendations for end-of-life care in the intensive care unit: a consensus statement by the American College of Critical Care Medicine. Crit Care Med 2008; 36(3):953–963 [DOI] [PubMed] [Google Scholar]
- 4.Ernecoff NC, Check D, Bannon M, et al. : Comparing specialty and primary palliative care interventions: analysis of a systematic review. J Palliat Med 2020; 23(3):389–396 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Kruser JM, Aaby DA, Stevenson DG, et al. : Assessment of variability in end-of-life care delivery in intensive care units in the United States. JAMA Netw Open 2019; 2(12):e1917344. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Schoenherr LA, Bischoff KE, Marks AK, et al. : Trends in hospital-based specialty palliative care in the United States Ffom 2013 to 2017. JAMA Netw Open 2019; 2(12):e1917043. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Kistler EA, Stevens E, Scott E, et al. : Triggered palliative care consults: a systematic review of interventions for hospitalized and emergency department patients. J Pain Symptom Manage 2020; 60(2):460–475 [DOI] [PubMed] [Google Scholar]
- 8.Cox CE, Ashana DC, Haines KL, et al. : Assessment of clinical palliative care trigger status vs actual needs among critically ill patients and their family members. JAMA Netw Open 2022; 5(1):e2144093. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Davidson JE, Aslakson RA, Long AC, et al. : Guidelines for family-centered care in the neonatal, pediatric, and adult ICU. Crit Care Med 2017; 45(1):103–128 [DOI] [PubMed] [Google Scholar]
- 10.Cox CE, Jones DM, Reagan W, et al. : Palliative Care Planner: a pilot study to evaluate acceptability and usability of an electronic health records system-integrated, needs-targeted app platform. Annals of the American Thoracic Society 2018; 15(1):59–68 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Emanuel LL, Alpert HR, Emanuel EE: Concise screening questions for clinical assessments of terminal care: the Needs Near the End-of-Life Care Screening Tool. J Palliat Med 2001; 4(4):465–474 [DOI] [PubMed] [Google Scholar]
- 12.Scandrett KG, Reitschuler-Cross EB, Nelson L, et al. : Feasibility and effectiveness of the NEST13+ as a screening tool for advanced illness care needs. J Palliat Med 2010; 13(2):161–169 [DOI] [PubMed] [Google Scholar]
- 13.National Consensus Project for Quality Palliative Care. Clinical practice guidelines for quality palliative care. 2018. Available at: https://www.nationalcoalitionhpc.org/wp-content/uploads/2018/10/NCHPC-NCPGuidelines_4thED_web_FINAL.pdf
- 14.Kroenke K, Spitzer RL, Williams JB: The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med 2001; 16(9):606–613 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Spitzer RL, Kroenke K, Williams JB, et al. : A brief measure for assessing generalized anxiety disorder: the GAD-7. Arch Intern Med 2006; 166(10):1092–1097 [DOI] [PubMed] [Google Scholar]
- 16.Stoll C, Kapfhammer HP, Rothenhausler HB, et al. : Sensitivity and specificity of a screening test to document traumatic experiences and to diagnose post-traumatic stress disorder in ARDS patients after intensive care treatment. Intensive Care Med 1999; 25(7):697–704 [DOI] [PubMed] [Google Scholar]
- 17.Teno JM, Fisher ES, Hamel MB, et al. : Medical care inconsistent with patients’ treatment goals: association with 1-year Medicare resource use and survival. J Am Geriatr Soc 2002; 50(3):496–500 [DOI] [PubMed] [Google Scholar]
- 18.Engelberg R, Downey L, Curtis JR: Psychometric characteristics of a quality of communication questionnaire assessing communication about end-of-life care. J Palliat Med 2006; 9(5):1086–1098 [DOI] [PubMed] [Google Scholar]
- 19.Stewart AL, Napoles-Springer AM, Gregorich SE, et al. : Interpersonal Processes of Care survey: patient-reported measures for diverse groups. Health services research 2007; 42(3 Pt 1):1235–1256 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Cox CE, Hough CL, Carson SS, et al. : Effects of a telephone- and web-based coping skills training program compared with an education program for survivors of critical illness and their family members. a randomized clinical trial. Am J Respir Crit Care Med 2018; 197(1):66–78 [DOI] [PubMed] [Google Scholar]
- 21.Knaus WA, Draper EA, Wagner DP, et al. : APACHE II: a severity of disease classification system. Crit Care Med 1985; 13(10):818–829 [PubMed] [Google Scholar]
- 22.Aslakson R, Cheng J, Vollenweider D, et al. : Evidence-based palliative care in the intensive care unit: a systematic review of interventions. J Palliat Med 2014; 17(2):219–235 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Nelson JE, Curtis JR, Mulkerin C, et al. : Choosing and using screening criteria for palliative care consultation in the ICU: a report from the Improving Palliative Care in the ICU Advisory Board. Crit Care Med 2013; 41(10):2318–2327 [DOI] [PubMed] [Google Scholar]
- 24.Hua MS, Li G, Blinderman CD, et al. : Estimates of the need for palliative care consultation across united states intensive care units using a trigger-based model. Am J Respir Crit Care Med 2014; 189(4):428–436 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Creutzfeldt CJ, Engelberg RA, Healey L, et al. : Palliative care needs in the neuro-ICU. Crit Care Med 2015; 43(8):1677–1684 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Schindler AW, Schindler N, Enz F, et al. : ICU personnel have inaccurate perceptions of their patients’ experiences. Acta Anaesthesiol Scand 2013; 57(8):1032–1040 [DOI] [PubMed] [Google Scholar]
- 27.World Health Organization. WHO defintion of palliative care. 2022. Available at: https://www.who.int/health-topics/palliative-care
- 28.Jackson JC, Pandharipande PP, Girard TD, et al. : Depression, post-traumatic stress disorder, and functional disability in survivors of critical illness in the BRAIN-ICU study: a longitudinal cohort study. Lancet Respir Med 2014; 2(5):369–379 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Davydow DS, Gifford JM, Desai SV, et al. : Posttraumatic stress disorder in general intensive care unit survivors: a systematic review. Gen Hosp Psychiatry 2008; 30(5):421–434 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Mikkelsen ME, Christie JD, Lanken PN, et al. : The adult respiratory distress syndrome cognitive outcomes study: long-term neuropsychological function in survivors of acute lung injury. Am J Respir Crit Care Med 2012; 185(12):1307–1315 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Huff NG, Nadig N, Ford DW, et al. : Therapeutic alliance between the caregivers of critical illness survivors and intensive care unit clinicians. Annals of the American Thoracic Society 2015; 12(11):1646–1653 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Parker AM, Sricharoenchai T, Raparla S, et al. : Posttraumatic stress disorder in critical illness survivors: a meta-analysis. Crit Care Med 2015; 43(5):1121–1129 [DOI] [PubMed] [Google Scholar]
- 33.Bienvenu OJ, Colantuoni E, Mendez-Tellez PA, et al. : Depressive symptoms and impaired physical function after acute lung injury: a 2-year longitudinal study. Am J Respir Crit Care Med 2012; 185(5):517–524 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Nadig N, Huff NG, Cox CE, et al. : Coping as a multifaceted construct: associations with psychological outcomes among family members of mechanical ventilation survivors. Crit Care Med 2016; 44(9):1710–1717 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Khandelwal N, Hough CL, Downey L, et al. : Prevalence, risk factors, and outcomes of financial stress in survivors of critical illness. Crit Care Med 2018; 46(6):e530–e539 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Cox CE, Curtis JR: Using technology to create a more humanistic approach to integrating palliative care into the intensive care unit. Am J Respir Crit Care Med 2016; 193(3):242–250 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.