Utilization of Palliative Care for Patients with COVID-19 and Acute Kidney Injury during a COVID-19 Surge

Abstract

Visual Abstract

Background and objectives 

AKI is a common complication of coronavirus disease 2019 (COVID-19) and is associated with high mortality. Palliative care, a specialty that supports patients with serious illness, is valuable for these patients but is historically underutilized in AKI. The objectives of this paper are to describe the use of palliative care in patients with AKI and COVID-19 and their subsequent health care utilization.

Design, setting, participants, & measurements 

We conducted a retrospective analysis of New York University Langone Health electronic health data of COVID-19 hospitalizations between March 2, 2020 and August 25, 2020. Regression models were used to examine characteristics associated with receiving a palliative care consult.

Results 

Among patients with COVID-19 (n=4276; 40%), those with AKI (n=1310; 31%) were more likely than those without AKI (n=2966; 69%) to receive palliative care (AKI without KRT: adjusted odds ratio, 1.81; 95% confidence interval, 1.40 to 2.33; P<0.001; AKI with KRT: adjusted odds ratio, 2.45; 95% confidence interval, 1.52 to 3.97; P<0.001), even after controlling for markers of critical illness (admission to intensive care units, mechanical ventilation, or modified sequential organ failure assessment score); however, consults came significantly later (10 days from admission versus 5 days; P<0.001). Similarly, 66% of patients initiated on KRT received palliative care versus 37% (P<0.001) of those with AKI not receiving KRT, and timing was also later (12 days from admission versus 9 days; P=0.002). Despite greater use of palliative care, patients with AKI had a significantly longer length of stay, more intensive care unit admissions, and more use of mechanical ventilation. Those with AKI did have a higher frequency of discharges to inpatient hospice (6% versus 3%) and change in code status (34% versus 7%) than those without AKI.

Conclusions 

Palliative care was utilized more frequently for patients with AKI and COVID-19 than historically reported in AKI. Despite high mortality, consultation occurred late in the hospital course and was not associated with reduced initiation of life-sustaining interventions.

Podcast 

This article contains a podcast at https://www.asn-online.org/media/podcast/CJASN/2022_02_24_CJN11030821.mp3

Introduction

In the spring of 2020, New York City was the epicenter of the coronavirus disease 2019 (COVID-19) pandemic, with a peak of over 150,000 cases by early April (¹). Although initial attention to COVID-19 was as a severe respiratory illness, it is now known that AKI is common, reported in 20%–56% of hospitalized COVID-19 cases (^2–6). A recent meta-analysis found that AKI occurred in 46% of patients with COVID-19 admitted to intensive care units (ICUs), with 19% requiring KRT (⁵). Similar to pre–COVID-19 data (⁷), patients with COVID-19 and AKI often have severe illness marked by mechanical ventilation and a high risk for in-hospital death (⁶).

The significant morbidity and mortality associated with AKI in the setting of COVID-19 creates complex medical decisions concerning the initiation or withdrawal of life-sustaining interventions, particularly KRT. The emotional effect of these acute life-threatening situations was exacerbated during the COVID-19 surge by restrictive visitation policies. These clinical scenarios are ones in which palliative care, a specialty that focuses on the relief of suffering and provides expert communication (⁸), becomes a valuable and essential service. For critically ill patients with COVID-19 and AKI, initiation of life-sustaining interventions that may lead to a medicalized end-of-life experience elevates the importance of conversations that establish an individual’s goals of treatment.

Historically, there has been low utilization of palliative care in patients with AKI requiring KRT (⁹), but to our knowledge, there are no published analyses of the use of palliative care for hospitalized patients with both AKI and COVID-19. The objectives of this paper are (1) to describe the provision of palliative care consultation and subsequent trends in health care utilization in patients with COVID-19 and AKI and (2) to examine the associations of patient characteristics and health care utilization with palliative care consultation.

Materials and Methods

Study Design and Setting

We analyzed electronic health record (EHR; Epic Systems, Verona, WI) data at New York University Langone Health (NYULH) between the dates of March 2, 2020 and January 6, 2021. Only completed hospitalizations were used; therefore, patients were followed up until hospital discharge or death. The range of follow-up was 1–279 days. NYULH consists of three acute care hospitals located in Manhattan, Brooklyn, and Long Island.

Study Cohort

We identified all hospital admissions for individuals ≥18 years old who tested positive for COVID-19 by real-time RT-PCR assay or nasopharyngeal or oropharyngeal swab specimens 7 days pre- or postadmission between March 2, 2020 and August 25, 2020 with discharge or death between March 13, 2020 and January 6, 2021. For patients with multiple admissions, we focused on the admission with the most severe illness (defined hierarchically by death, KRT initiation, use of mechanical ventilation, or longest length of stay). We identified AKI using the Acute Kidney Injury Network (AKIN) creatinine criteria. Urine output was not included due to inconsistencies in documentation. Use of KRT was extracted directly from the EHR as a procedure order. Outpatient creatinine values, when available within 6 months prior to admission, were utilized to define baseline kidney function. eGFR on admission was determined using baseline creatinine values and calculated with the Chronic Kidney Disease Epidemiology Collaboration equation (¹⁰). When no outpatient value was available, we used the admission creatinine.

Covariates

Demographic data, health care utilization (length of stay and days in the ICU), maximum modified sequential organ failure assessment (mSOFA) score, and procedures (KRT, initiation of mechanical ventilation, and cardiopulmonary resuscitation [CPR]/rapid response, with rapid response defined as a clinically unstable patient seen by a specialized team including an intensivist) were identified through dedicated fields in EHR flow sheets. We identified code status through an EHR report and smoking status through recorded social history in the EHR. We used International Classification of Diseases (ICD)-10 codes to assign the following clinical characteristics: hypertension, hyperlipidemia, coronary heart disease, congestive heart failure, peripheral vascular disease, obesity, diabetes, pulmonary disease (defined as chronic obstructive pulmonary disease, interstitial lung disease, or pulmonary fibrosis), cancer, and cirrhosis. Discharge disposition was grouped as died, home, skilled nursing facility, inpatient hospice rehabilitation facility, against medical advice, and other facility.

Main Outcomes

We assessed four primary outcomes: (1) utilization of palliative care, defined by completion of a palliative care note by a palliative care provider (this is usually done within 24 hours of placing a consult order); (2) health care utilization (ICU admissions, length of hospital or ICU stay, and use of life-sustaining interventions, defined as initiation of KRT or mechanical ventilation); (3) characteristics associated with palliative care use; and (4) patterns of discharge to inpatient hospice versus death in the hospital. In our institution, actively dying patients with a prognosis of hours to days can be admitted to inpatient hospice with the patient’s or family’s consent and without transfer to a different facility or location. This enables access to hemodynamically unstable patients, including those in the ICU. Given that it is readily available in our institution and it is considered a “gold standard” for end-of-life experience, we chose this as an outcome.

Statistical Analyses

Baseline variables are reported using means ±SD for continuous variables and n (percentage) for categorical variables, stratified by presence of AKI. Descriptive comparisons are performed using a two-sample unpaired t test for continuous variables or a chi-squared test for categorical variables.

We used multivariable logistic regression to examine the odds of receiving a palliative care consult. The main covariate of interest is AKI or AKI with KRT compared with those without AKI. Covariates include age, sex, race, ethnicity, presence of cirrhosis, cancer, body mass index, need for mechanical ventilation, ICU admission, baseline eGFR, maximum mSOFA score, CPR/rapid response, and length of stay. These covariates were chosen as they are hypothesized to be important predictors of receiving palliative care services. We conducted multivariable imputation by chained equations to fill in missing values (race [6%], ethnicity [6%], body mass index [3%], eGFR [0.02%], and mSOFA [19%]). We used R version 3.6.2 for all analyses. P<0.05 was considered significant.

This study was approved by the New York University Grossman Institutional Review Board as an expedited review with a waiver of informed consent.

Results

Baseline Characteristics

Of the 10,793 patients admitted to NYULH during the study time frame, 6517 (60%) were excluded for the following reasons: absence of positive test for COVID-19 within a week of admission (n=6197; 95%), diagnosis of kidney failure (n=239; 4%), or <18 years of age (n=81; 1%). Our final cohort included 4276 patients (40%); 1310 (31%) developed AKI, 201 (5%) of whom required KRT (Figure 1).

Figure 1.: Consolidated Standards of Reporting Trials (CONSORT) diagram of the analytical cohort. COVID-19, coronavirus disease 2019.

Those with AKI (n=1310) were older than those without AKI (n=2966), had lower baseline eGFR, were more often men, had higher mSOFA scores, and had more comorbid illnesses with the exclusion of obesity. Nearly half of the population was White; however, self-defined race was similar in both groups. A higher proportion of patients with AKI identified as non-Hispanic. Admission do not resusitate (DNR) code status was similar in those with and without AKI (Table 1).

Table 1. - Demographics of coronavirus disease 2019–positive inpatients with or without AKI and those with AKI who did and did not receive KRT

Demographic and Clinical Characteristics	No AKI, n=2966	AKI, n=1310	AKI, No KRT, n=1109	AKI + KRT, n=201
Age, mean (SD)	61 (18)	68 (15)	69 (15)	63 (12)
Men, n (%)	1589 (54)	864 (66)	695 (63)	169 (84)
Race, n (%)
White	1335 (48)	624 (50)	532 (51)	92 (48)
Black	435 (16)	189 (15)	161 (15)	28 (15)
Pacific Islander/Native American	22 (0.8)	4 (0.3)	4 (0.4)	0 (0)
Asian	212 (8)	98 (8)	80 (8)	18 (9)
Other/multiracial	785 (28)	329 (26)	276 (26)	53 (28)
Missing	177	66	56	10
Ethnicity, n (%)
Hispanic	847 (30)	320 (26)	264 (26)	56 (30)
Non-Hispanic	1943 (70)	897 (74)	769 (74)	128 (70)
Missing	176	93	76	17
Comorbidities, n (%)
Hypertension	1523 (51)	930 (71)	804 (73)	126 (63)
Hyperlipidemia	1210 (41)	663 (51)	582 (53)	81 (40)
Heart failure	220 (7)	227 (17)	214 (19)	13 (7)
Coronary artery disease	365 (12)	285 (22)	253 (23)	32 (16)
Peripheral vascular disease	145 (5)	112 (9)	106 (10)	6 (3)
Obesity	471 (16)	210 (16)	182 (16)	28 (14)
Diabetes	941 (32)	549 (42)	474 (43)	75 (37)
CKD	145 (5)	216 (17)	188 (17)	28 (14)
Pulmonary disease	665 (22)	350 (27)	313 (28)	37 (18)
Cancer	344 (12)	209 (16)	186 (17)	23 (11)
Cirrhosis	43 (1)	30 (2)	27 (2)	3 (2)
Body mass index, mean (SD)	30 (7)	30 (8)	30 (7)	32 (8)
Missing	89	26	25	1
Smoking, n (%)
Never	1760 (66)	677 (58)	562 (57)	115 (67)
Active	193 (7)	79 (7)	67 (7)	12 (7)
Former	602 (22)	325 (28)	294 (30)	31 (18)
Not asked	132 (5)	78 (7)	63 (6)	15 (9)
Missing	279	151	123	28
Admission code status, n (%)
CPR	2744 (93)	1199 (92)	1000 (90)	199 (99)
Do not resuscitate	209 (7)	109 (8)	107 (10)	2 (1)
mSOFA maximum value, n (%)
0–7	2267 (99)	651 (56)	641 (65)	10 (5)
8–11	25 (1)	334 (28)	271 (28)	63 (34)
>11	0 (0)	181 (16)	68 (7)	113 (61)
Missing	674	144	129	15
Baseline eGFR, n (%)
<15	33 (1)	82 (6)	61 (5)	21 (10)
15 to <30	131 (4)	205 (16)	186 (17)	19 (10)
30 to <60	559 (19)	444 (34)	389 (35)	55 (27)
≥60	2242 (76)	579 (44)	473 (43)	106 (53)
Missing	1	0	0	0

Means (SD) for continuous variables and counts (percentage) for categorical variables are shown to compare demographic and clinical characteristics. CPR, cardiopulmonary resuscitation; mSOFA, modified sequential organ failure assessment.

Among those with AKI, those who received KRT were younger, had a higher proportion of patients with an eGFR <15 ml/min per 1.73 m², were more likely to be men, had higher mSOFA scores, and had higher body mass index. Those who received KRT were equally or less likely to have comorbid illnesses. This population also had a lower proportion of individuals who were DNR at admission (Table 1).

Utilization and Timing of Palliative Care and Discharge Outcomes

Compared with those without AKI, individuals with AKI received more palliative care consults (42% versus 7%; P<0.001), but they occurred significantly later (10 days from admission versus 5 days; P<0.001). Sixty-six percent of patients with AKI initiated on KRT received palliative care compared with 37% (P<0.001) of those with AKI not started on KRT. Palliative care consults also occurred later for those started on KRT compared with those who were not (12 days from admission versus 9 days; P=0.002) (Table 2).

Table 2. - Health care utilization among coronavirus disease 2019–positive inpatients with versus without AKI and among those with AKI who did versus did not receive KRT

Characteristic	No AKI, n=2966	AKI, n=1310	P Value	AKI, No KRT, n=1109	AKI + KRT, n=201	P Value
Palliative care consult, n (%)	218 (7)	543 (42)	<0.001	410 (37)	133 (66)	<0.001
Days from admission to palliative care consult, d, mean (SD, range)	5 (7, 0–58)	10 (10, 0–67)	<0.001	9 (10, 0–67)	12 (11, 0–67)	0.002
Length of stay, d, mean (SD, range)	7 (8, 1–86)	20 (24, 1–279)	<0.001	18 (21, 1–248)	32 (33, 3–279)	<0.001
ICU admission, n (%)	234 (8)	749 (57)	<0.001	560 (51)	189 (94)	<0.001
Duration of ICU stay, d, mean (SD, range)	8 (11, 0–63)	19 (25, 0–271)	<0.001	17 (23, 0–245)	26 (30, 1–270)	<0.001
Mechanical ventilation, n (%)	117 (4)	708 (54)	<0.001	520 (47)	188 (94)	<0.001
Duration of mechanical ventilation, d, mean (SD, range)	9 (13, 0–61)	18 (23, 0–241)	<0.001	16 (23, 0–241)	23 (23, 1–135)	<0.001
CPR or rapid response, n (%)	183 (6)	541 (41)	<0.001	425 (38)	116 (58)	<0.001
Change in code status, n (%)	214 (7)	450 (34)	<0.001	349 (32)	101 (50)	<0.001
Discharge disposition, n (%)			<0.001			<0.001
Died	149 (5)	608 (46)		460 (42)	148 (74)
Home	2273 (77)	318 (24)		311 (28)	7 (4)
Skilled nursing facility	343 (12)	188 (14)		164 (15)	24 (12)
Inpatient hospice	80 (3)	77 (6)		71 (6)	6 (3)
Rehabilitation facility	56 (2)	93 (7)		81 (7)	12 (6)
Against medical advice	22 (0.7)	3 (0.2)		3 (0.3)	0 (0)
Other facility	43 (1)	23 (2)		19 (2)	4 (2)

ICU, intensive care unit; CPR, cardiopulmonary resuscitation.

Discharge outcomes differed markedly by acuity of illness. For example, compared with those without AKI, higher proportions of those with AKI died during the hospitalization (46% versus 5%) or were discharged to inpatient hospice (6% versus 3%), whereas lower proportions of those with AKI were discharged home (24% versus 77%). Similar trends were seen when comparing those started on KRT versus those who were not, with the exception of inpatient hospice use (Table 2).

In analyses restricted to patients admitted to the ICU (n=983), palliative care consultation occurred at 12, 11, and 10 days in those with AKI started on KRT, AKI not started on KRT, and without AKI, respectively (P=0.36). Figure 2 shows the timing of palliative care consultation in relation to major hospital events for patients admitted to the ICU. In patients with AKI, average palliative care consultation occurred after AKI and after undergoing other life-sustaining procedures, such as intubation, CPR, and KRT.

Figure 2.: Timing of major hospital events for patients in the intensive care unit (ICU) by categories of kidney injury (AKI with KRT, AKI without KRT, and no AKI). Major hospital events were defined as ICU admission, AKI, intubation, undergoing cardiopulmonary resuscitation (CPR), initiation of KRT, code status change, death in the hospital, discharge to inpatient hospice, or alive discharge. For each category, the numbers shown are the number of patients. The day listed is the mean number of days from admission for each event.

An analysis of all patients who died or were discharged to inpatient hospice showed that those with AKI who were started on KRT had a longer time between palliative care consult and death compared with those who had AKI not started on KRT or those without AKI (16 days before death versus 7 days versus 5 days, respectively; P<0.001). In this same group, those with AKI requiring KRT again received palliative later than those with AKI without KRT or those without AKI (11 days versus 7 days versus 5 days; P<0.001).

Health Care Utilization, Life-Sustaining Interventions, and Change in Code Status

Patients with AKI had significantly higher health care utilization than those without, characterized by longer lengths of stay, more ICU admissions of longer duration, and greater use of mechanical ventilation. They also had more changes in code status (34% versus 7%; P<0.001) than those with a lower acuity of illness (Table 2).

Most of the patients who received palliative care had significantly higher health care utilization (Supplemental Figure 1). Patients with AKI who received KRT had similar frequency of death (73% with palliative care and 75% without) and hospice use (3% for both groups), regardless of receiving palliative care (Figure 3).

Figure 3.: Disposition ( i.e. , death in the hospital or discharge to inpatient hospice) comparison between patients who received a palliative care consult versus those who did not for different AKI groups. Percentage was calculated on the basis of the total study population. Supplemental Tables 1 and 2 have more descriptions of the data used. P values were estimated by the chi-squared test to compare disposition between palliative care groups (yes or no). **P<0.001.

Factors Associated with Palliative Care Consultation

Compared with all other conditions, patients with AKI started on KRT had the strongest association with palliative care consultation even after adjusting for markers of serious acute illness, such as ICU admission, mechanical ventilation, mSOFA score, and having undergone CPR (Table 3). AKI with KRT (adjusted odds ratio, 2.45; 95% confidence interval, 1.52 to 3.97; P<0.001) or without KRT (adjusted odds ratio, 1.81; 95% confidence interval, 1.40 to 2.33; P<0.001) was associated with a higher likelihood of receiving a palliative care consult compared with no AKI. Other markers of health care utilization, such as ICU admission, CPR, or rapid response, and length of stay were also associated with a higher likelihood of receiving a palliative care consultation. We saw similar trends when analysis was restricted to patients in the ICU (n=785), where AKI both with or without KRT had a strong association with receiving a palliative care consult, even after adjustment for mechanical ventilation, CPR, or length of stay (Supplemental Table 3).

Table 3. - Associations of AKI and initiation of KRT with receipt of a palliative care consult

Covariates	Odds of Receiving a Palliative Care Consult, n=4276 a
	Unadjusted Odds Ratio (95% Confidence Interval)	Adjusted Odds Ratio (95% Confidence Interval)	P Value
AKI
No AKI	1	1	Reference
AKI, no KRT	7.39 (6.16 to 8.90)	1.81 (1.40 to 2.33)	<0.001
AKI with KRT	24.65 (17.92 to 34.22)	2.45 (1.52 to 3.97)	<0.001
Age, 10 yr	1.47 (1.40 to 1.55)	1.71 (1.56 to 1.88)	<0.001
Sex
Women	1	1	Reference
Men	1.09 (0.93 to 1.28)	0.66 (0.53 to 0.81)	<0.001
Race
White	1	1	Reference
Black	0.70 (0.55 to 0.89)	0.97 (0.72 to 1.30)	0.82
Asian	0.91 (0.67 to 1.22)	0.87 (0.59 to 1.27)	0.48
Other/multiracial	0.68 (0.56 to 0.82)	1.13 (0.83 to 1.54)	0.44
Pacific Islander/Native American	0.45 (0.11 to 1.29)	2.24 (0.47 to 7.92)	0.25
Ethnicity
Hispanic	1	1	Reference
Non-Hispanic	1.67 (1.39 to 2.02)	1.36 (0.99 to 1.87)	0.06
Cirrhosis	1.20 (0.65 to 2.07)	1.12 (0.53 to 2.22)	0.76
Cancer	2.21 (1.80 to 2.70)	2.03 (1.58 to 2.60)	<0.001
Body mass index, 5 kg/m2	0.92 (0.87 to 0.97)	1.01 (0.94 to 1.09)	0.76
eGFR, 15 ml/min per 1.73 m2	0.76 (0.73 to 0.79)	1.02 (0.96 to 1.09)	0.79
mSOFA maximum value	1.32 (1.30 to 1.35)	1.17 (1.12 to 1.23)	<0.001
ICU admission	8.03 (6.77 to 9.54)	1.55 (1.10 to 2.16)	0.010
Mechanical ventilation	9.77 (8.19 to 11.67)	1.24 (0.83 to 1.86)	0.30
CPR/rapid response	6.37 (5.33 to 7.61)	1.50 (1.18 to 1.90)	0.001
Length of stay, d	1.05 (1.05 to 1.06)	1.02 (1.02 to 1.03)	<0.001

The adjusted odds ratios and P values were calculated using AKI group and adjusting for age, sex, race, ethnicity, body mass index, eGFR, mSOFA score, comorbidities (cirrhosis and cancer), and health care utilization (ICU admission, mechanical ventilation, CPR/rapid response, and length of stay). mSOFA, modified sequential organ failure assessment; ICU, intensive care unit; CPR, cardiopulmonary resuscitation.

^aMultivariable imputation by chained equations was used to fill in missing data for race, ethnicity, body mass index, eGFR, and mSOFA.

Discussion

To our knowledge, this is the first study to describe the utilization of palliative care in hospitalized patients with COVID-19 and AKI. We found that <50% of those with AKI received a palliative care consult. However, compared with those without AKI, more patients with AKI received a palliative care consult, were discharged to inpatient hospice, and changed their code status, although often in the context of high health care utilization. Furthermore, palliative care consultations occurred on average 5 days later in the hospital course for those with AKI. Similar trends were seen for those with AKI who received KRT compared with those with AKI who were not started on KRT.

The provision of palliative care to hospitalized patients with AKI has received little attention in the literature. This comes despite published evidence of the benefits of palliative care for seriously ill patient as measured by higher hospice use, completion of advance directives, documentation of end-of-life conversations, fewer ICU admissions, and lower use of life-sustaining interventions (^11–16). In 2017, Chong et al. (⁹) published an analysis of over 3 million hospitalized patients with AKI identified by ICD-9 codes comparing utilization of palliative care in those with AKI receiving KRT with those with AKI without KRT or without AKI. In their entire cohort, only 2% of patients received a palliative consult: 8% of those with AKI receiving KRT and 5% of those with AKI without KRT. In a smaller study of 230 patients receiving continuous KRT (CKRT) in an institution with a dedicated palliative care unit, 21% of patients received palliative care (¹¹). Our results suggest that practice patterns deviated from pre–COVID-19 practice, with much higher utilization of palliative care. Nevertheless, a majority of patients with AKI in our analysis still did not receive this service. Notably, those who did receive this consultation did so markedly later in the hospital course than those without AKI. Despite higher utilization of palliative care, we did not identify clear benefit of palliative care’s effect on health care utilization. In fact, resource utilization was markedly higher, and we found similar rates of utilization of inpatient hospice for those started on KRT, regardless of palliative care’s involvement. This is similar to the findings of Okon et al. (¹¹), in which patients on CKRT who received palliative care had longer hospital lengths of stay, longer ICU stays, and more days of CKRT.

Explanations for the practice patterns we identified are challenging given the constant evolution of knowledge of COVID-19 and the unique environment of visitation restrictions for families. The greater use of palliative care in a sicker population may simply be a reflection of triage of palliative care to higher-acuity patients. Nevertheless, >50% of the AKI population still did not receive a consult. Another explanation could be lower access to palliative care given the strain on our hospital system or use of this service to triage resources in a time of crisis. Although this is plausible, our institution’s palliative care team did reorganize early in the pandemic to meet the increased volume and never had to limit use of resources, making triage unnecessary (¹⁷).

In our findings, we observed that the later use of palliative care in higher-acuity patients often resulted in use of the service only after the patient had received life-sustaining interventions, such as KRT, mechanical ventilation, or CPR. The later consultation of palliative care in patients with AKI compared with those without may reflect a perceived better prognosis for this population and, consequently, more use of life-sustaining interventions. Another explanation may be that palliative care was used for conflict resolution (when the team and family differed on the appropriate direction of care). Clinically, the time needed for conflict resolution could explain the large amount of time between palliative care and death for patients who died. A third explanation is that AKI was not recognized as the marker for poor prognosis, and therefore, palliative care was deemed unnecessary earlier in the hospitalization. These perceptions of palliative care as an add-on service rather than a proactive consultation that can empathically guide patients and families through difficult discussions about serious illness, mitigate caregiver distress, and alleviate physical and emotional suffering for patients suggest a missed opportunity for patients and families to benefit from added support. Additionally, earlier consultation also may have lowered the use of nonbeneficial invasive medical interventions.

Our study has several limitations. First, our data were extracted from a database that was dependent upon clinical and administrative data. There may have been misclassification, despite extensive data cleaning completed prior to analysis, and the use of the AKIN criteria to identify AKI likely increased accuracy. Second, although we can hypothesize the reasons for the observations we report, the nuances of clinical practice cannot be truly captured through secondary data analysis. Third, the larger implications of these findings are unknown, as practice patterns and systems were modified during COVID-19 to meet the needs of a potentially unique increase in volume, and our results should be generalized cautiously to other settings. Finally, we analyzed data from a single health care system with a mostly White, non-Hispanic population from an urban area, limiting its generalizability.

Despite these limitations, our study demonstrates important findings with considerations that extend beyond the COVID-19 pandemic. Currently, the benefits of palliative care for patients with AKI are a knowledge gap in nephrology. The high mortality of those with AKI and the greater use of life-sustaining interventions that often do not prevent death suggest that AKI can serve as a trigger for proactive and early involvement of palliative care. Further multicentered studies of the timing of palliative care in critically ill patients may lead to opportunities for system-based interventions that identify patients at high risk of mortality or suffering who may benefit from early palliative care consultation.

Disclosures

D.M. Charytan reports consultancy agreements with Allena Pharmaceuticals (Data and Safety Monitoring Board), Amgen, AstraZeneca, CSL Behring, Eli Lilly/Boehringer Ingelheim, Fresenius, Gilead, GSK, Janssen (steering committee), Medtronic, Merck, Novo Nordisk, and PLC Medical (clinical events committee); research funding from Amgen, Bioporto (clinical trial support), Gilead, Medtronic (clinical trial support), and Novo Norodisk; serving as an associate editor of CJASN; and expert witness fees related to proton pump inhibitors. J. Chodosh reports consultancy agreements with the Gerontological Association of America (GSA) and the West Health Foundation, honoraria from GSA and West Health, and serving as a scientific advisor or member of the Aging New York Fund. J.S. Scherer reports serving as one-time scientific advisor for CARA Therapeutics and with Vifor Pharmaceuticals and honoraria from Cara Therapeutics, Vifor Pharma, and UpToDate as a peer reviewer. All remaining authors have nothing to disclose.

Funding

J.S. Scherer is supported by National Institute of Diabetes and Digestive and Kidney Diseases grant K23DK-125840 and National Kidney Foundation Young Investigator Award, Satellite Dialysis Clinical Investigator Grant. J. Zhong is supported by National Institute on Aging grants 1R011AG065330-01, 5R01AG 054467-04, and 3R01AG54467-04S1.