Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Association of Intensive Care Unit Patient Load and Demand With Mortality Rates in US Department of Veterans Affairs Hospitals During the COVID-19 Pandemic

Association of Intensive Care Unit Patient Load and Demand With Mortality Rates in US Department... Key Points Question Is greater coronavirus disease IMPORTANCE Although strain on hospital capacity has been associated with increased mortality in 2019 (COVID-19) intensive care unit nonpandemic settings, studies are needed to examine the association between coronavirus disease (ICU) strain associated with increased 2019 (COVID-19) critical care capacity and mortality. COVID-19 mortality? Findings In this cohort study of 8516 OBJECTIVE To examine whether COVID-19 mortality was associated with COVID-19 intensive care patients with COVID-19 admitted to 88 unit (ICU) strain. US Veterans Affairs hospitals, strains on critical care capacity were associated DESIGN, SETTING, AND PARTICIPANTS This cohort study was conducted among veterans with with increased COVID-19 mortality. COVID-19, as confirmed by polymerase chain reaction or antigen testing in the laboratory from March Among patients with COVID-19, those through August 2020, cared for at any Department of Veterans Affairs (VA) hospital with 10 or more treated in the ICU during periods of peak patients with COVID-19 in the ICU. The follow-up period was through November 2020. Data were COVID-19 ICU demand had a nearly analyzed from March to November 2020. 2-fold increased risk of mortality compared with those treated during EXPOSURES Receiving treatment for COVID-19 in the ICU during a period of increased COVID-19 ICU periods of low demand. load, with load defined as mean number of patients with COVID-19 in the ICU during the patient’s hospital stay divided by the number of ICU beds at that facility, or increased COVID-19 ICU demand, Meaning These findings suggest that with demand defined as mean number of patients with COVID-19 in the ICU during the patient’s stay public health officials and hospital divided by the maximum number of patients with COVID-19 in the ICU. administrators should consider interventions that reduce COVID-19 ICU MAIN OUTCOMES AND MEASURES All-cause mortality was recorded through 30 days after demand to improve survival among discharge from the hospital. patients with COVID-19 in the ICU. RESULTS Among 8516 patients with COVID-19 admitted to 88 VA hospitals, 8014 (94.1%) were men Invited Commentary and mean (SD) age was 67.9 (14.2) years. Mortality varied over time, with 218 of 954 patients (22.9%) dying in March, 399 of 1594 patients (25.0%) dying in April, 143 of 920 patients (15.5%) dying in May, Multimedia 179 of 1314 patients (13.6%) dying in June, 297 of 2373 patients (12.5%) dying in July, and 174 of 1361 Supplemental content (12.8%) patients dying in August (P < .001). Patients with COVID-19 who were treated in the ICU during periods of increased COVID-19 ICU demand had increased risk of mortality compared with Author affiliations and article information are listed at the end of this article. patients treated during periods of low COVID-19 ICU demand (ie, demand of25%); the adjusted hazard ratio for all-cause mortality was 0.99 (95% CI, 0.81-1.22; P = .93) for patients treated when COVID-19 ICU demand was more than 25% to 50%, 1.19 (95% CI, 0.95-1.48; P = .13) when COVID-19 ICU demand was more than 50% to 75%, and 1.94 (95% CI, 1.46-2.59; P < .001) when COVID-19 ICU demand was more than 75% to 100%. No association between COVID-19 ICU demand and mortality was observed for patients with COVID-19 not in the ICU. The association between COVID-19 ICU load and mortality was not consistent over time (ie, early vs late in the pandemic). CONCLUSIONS AND RELEVANCE This cohort study found that although facilities augmented ICU capacity during the pandemic, strains on critical care capacity were associated with increased (continued) Open Access. This is an open access article distributed under the terms of the CC-BY License. JAMA Network Open. 2021;4(1):e2034266. doi:10.1001/jamanetworkopen.2020.34266 (Reprinted) January 19, 2021 1/11 JAMA Network Open | Public Health Association of Intensive Care Unit Load and Demand With COVID-19 Mortality in VA Hospitals Abstract (continued) COVID-19 ICU mortality. Tracking COVID-19 ICU demand may be useful to hospital administrators and health officials as they coordinate COVID-19 admissions across hospitals to optimize outcomes for patients with this illness. JAMA Network Open. 2021;4(1):e2034266. doi:10.1001/jamanetworkopen.2020.34266 Introduction Health policy interventions (eg, social distancing) implemented to avoid overloading health care systems have been associated with reduced coronavirus disease 2019 (COVID-19) hospitalization 1,2 rates. Hospital capacity strain resulting from increased patient volume or disease severity is associated with increased mortality in nonpandemic settings. The association between COVID-19 critical care strain and mortality has not been examined. The objective of this study was to examine the associations between 2 measures of intensive care unit (ICU) strain and mortality among patients with COVID-19 who were admitted to a Department of Veterans Affairs (VA) facility. Methods This cohort study was approved by the Indiana University School of Medicine institutional review board and the VA research and development committee at Richard L. Roudebush VA Medical Center. This was an observational study without any direct patient contact and considered to be of minimal risk; therefore, a waiver of informed consent was obtained. This study adheres to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline. This observational cohort study included patients with polymerase chain reaction or antigen test results positive for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) from March 1 to August 31, 2020. We included patients admitted to a VA hospital that had 10 or more patients with COVID-19 in the ICU during the study period. We excluded patients whose SARS-CoV-2 test date was more than 2 days after admission. Follow-up extended through November 1, 2020. Data from the VA Corporate Data Warehouse (CDW), including inpatient and outpatient data files from the 2 years before the patient’s COVID-19 diagnosis through the day of admission, were used to identify comorbid conditions, self-reported race/ethnicity, health care use, procedures 4,5 received, vital signs, and laboratory data. Death dates were obtained from the CDW, VA Vital Status File, and patient’s electronic health record. The modified Acute Physiology, Age, Chronic Health Evaluation (APACHE) III score was calculated at the time of hospital admission as a measure of illness severity. Number of ICU beds was obtained from the VA Bed Management Solution initiative. The number of ICU beds per facility described prepandemic ICU capacity, not any augmented capacity that may have been implemented during the COVID-19 pandemic. Facility complexity described the level of services provided at a VA facility, categorized as 1a, 1b, 1c, 2, or 3, with level 1a being the most complex and level 3 being the least complex. Facility complexity included ICU level, operative complexity level, patient clinical classification, teaching status characteristics, amount of research funding, complex clinical programs provided (eg, invasive catheterization laboratory, neurosurgery, and transplant), rurality, care provided in the community, and mental health programs provided. The primary outcome was all-cause mortality. Patients were followed up for 30 days after discharge. Measures of COVID-19 ICU Strain This study included 2 measures of COVID-19 critical care strain: ICU load and ICU demand; Figure 1 displays how these measures vary over time within an example facility. The COVID-19 ICU load described how the caseload of patients with COVID-19 in the ICU compared with the typical ICU bed JAMA Network Open. 2021;4(1):e2034266. doi:10.1001/jamanetworkopen.2020.34266 (Reprinted) January 19, 2021 2/11 JAMA Network Open | Public Health Association of Intensive Care Unit Load and Demand With COVID-19 Mortality in VA Hospitals count at each facility. The COVID-19 ICU load was calculated at the patient level as the mean number of patients with COVID-19 in the ICU during the patient’s hospital stay divided by the number of ICU beds at that facility. The number of ICU beds at each facility was a fixed number. The COVID-19 ICU load included only the number of patients with COVID-19 in the ICU, excluding patients with other critical illnesses. The COVID-19 ICU load ranged from 0% to greater than 100%; it exceeded 100% if the hospital increased critical care bed capacity during the pandemic (eg, by converting a sleep laboratory into an ICU) and if those beds were occupied by patients with COVID-19. The population of patients with COVID-19 in the ICU varied over time, with peak prevalence rates occurring early (eg, March) at some hospitals and later (eg, July) at other hospitals. The COVID-19 ICU demand described the caseload of patients with COVID-19 in the ICU when a patient was treated compared with peak COVID-19 ICU caseload. It was calculated at the patient level as the mean number of patients with COVID-19 in the ICU during the patient’s stay divided by the maximum number of patients with COVID-19 in the ICU at that facility during the study period. The COVID-19 ICU demand ranged from 0% to 100%. For example, if a hospital had 60 ICU beds before the pandemic and the mean number of patients with COVID-19 in the ICU during a patient’s stay was 20, then the COVID-19 ICU load would be 20 divided by 60, or 33%. If at that same facility the peak surge included 20 patients with COVID-19 in the ICU and a patient was treated during the period when the mean of number patients with COVID-19 in the ICU was 20, then the COVID-19 ICU demand would be 20 divided by 20, or 100%. Statistical Analysis We described differences over time in baseline characteristics and mortality among inpatients with COVID-19 using χ and Wilcoxon rank sum tests. We used Cox proportional hazard models to analyze the time in days from admission to death, either in the hospital or within 30 days after discharge, among patients who were admitted to the hospital (overall, in the general ward, and in the ICU). Patients who were still in the hospital or out of the hospital and alive at 30 days after discharge were treated as censored observations. We included a random effect for the facility to account for the correlation of mortality among patients in the same hospital. Analyses were performed using SAS Enterprise Guide statistical software version 7.11 (SAS Institute). P values were 2-sided, and statistical significance was set at P < .05. Data were analyzed from March to November 2020. Figure 1. Coronavirus Disease 2019 (COVID-19) Intensive Care Unit (ICU) Load and Demand at an Example Facility Patients with COVID-19 in the ICU ICU beds Maximum volume of patients with COVID-19 in the ICU Black solid vertical line with arrows indicates the COVID-19 ICU load exceeding 100% of ICU capacity numbers at 1 time point used to calculate COVID-19 ICU load, defined as the mean number of patients with COVID-19 in the ICU during a patient’s hospital stay divided by the number of ICU beds. Black dotted vertical line indicates the numbers at 1 time point used to calculate COVID-19 demand, defined as the mean number of patients with COVID-19 in the ICU during a patient’s hospital stay divided by the maximum number of patients with COVID-19 in the ICU during the study period. The results suggest that the risk of mortality would be highest if a COVID-19 patient’s stay March April May June July August was during the peak of ICU demand and if ICU caseload Time, mo approached or exceeded ICU bed capacity. JAMA Network Open. 2021;4(1):e2034266. doi:10.1001/jamanetworkopen.2020.34266 (Reprinted) January 19, 2021 3/11 Patients with COVID-19 in the ICU, No. JAMA Network Open | Public Health Association of Intensive Care Unit Load and Demand With COVID-19 Mortality in VA Hospitals Results The cohort included 8516 patients with COVID-19 admitted to 88 VA hospitals (Figure 2); 8014 (94.1%) were men, and mean (SD) age was 67.9 (14.2) years. The observed mortality was 218 of 954 patients (22.9%) in March. It increased to 399 of 1594 patients (25.0%) in April, then decreased to 143 of 920 patients (15.5%) in May, leveling off at 179 of 1314 patients (13.6%) in June, 297 of 2373 patients (12.5%) in July, and 174 of 1361 patients (12.8%) in August (P < .001) (Table 1). The proportion of patients ages 75 years or older fluctuated over time; 223 patients (23.4%) were in this age group in March, increasing to 534 patients (33.5%) in April and 311 patients (33.8%) in May, decreasing to 333 patients (25.3%) in June and 637 patients (26.8%) in July, and then increasing again to 435 patients (32.0%) in August (P < .001). The proportion of patients receiving care on general wards increased after the earliest months of the pandemic (March: 527 patients [55.2%]; April: 965 patients [60.5%]; May: 546 patients [59.3%]; June: 829 patients [63.1%]; July: 1603 patients [67.6%]; and August: 903 patients [66.3%]; P < .001). The COVID-19 ICU load and demand changed over time. The proportion of patients with COVID-19 treated during periods of low COVID-19 ICU load (ie,25%) increased over time, with 487 patients (51.0%) in March, 952 patients (59.7%) in April, 785 patients (85.3%) in May, 1170 patients (89.0%) in June, 1923 patients (81.0%) in July, and 1250 patients (91.8%) in August (P < .001). Similarly, the proportion of patients with COVID-19 treated during periods of peak COVID-19 ICU load (ie, >100%) decreased from 60 patients (6.3%) in March to 18 patients (1.1%) in April, to 0 patients in May through August. The proportion of patients with COVID-19 treated during periods of high COVID-19 ICU demand (ie, >75%) decreased during the first 3 months of the pandemic, with 233 patients (24.4%) in March, 322 patients (20.2%) in April, and 44 patients (4.8%) in May; increased to 142 patients (10.8%) in June and 413 patients (17.4%) in July; then decreased to 79 patients (5.8%) in August (P < .001) (Table 1). Patients with COVID-19 in the ICU treated during high COVID-19 ICU strain had increased risk of mortality. Table 2 provides the unadjusted results, and Table 3 displays the adjusted hazard ratios (HRs). Compared with patients with COVID-19 in the ICU treated during periods of low COVID-19 ICU load (ie,25%), the adjusted HR for all-cause mortality was 1.10 (95% CI, 0.88-1.37) for patients treated during periods when COVID-19 ICU load was greater than 25% to 50%, 1.15 (95% CI, 0.81-1.64) when COVID-19 ICU load was greater than 50% to 75%, 1.67 (95% CI, 1.08-2.60) when COVID-19 ICU load was greater than 75% to 100%, and 2.35 (95% CI, 1.25-4.39) when COVID-19 ICU Figure 2. Patient Flow Diagram 48 013 US veterans with suspected COVID-19 347 Excluded because of negative antibody tests 47 666 With COVID-19 38 030 Excluded because they received outpatient care only 9636 Admitted at any VA hospital 659 Excluded because admitted at VA hospitals with <10 ICU patients 8977 Admitted at VA hospital with ≥10 ICU patients 461 Excluded with COVID-19 diagnosis >2 days after admission 8516 Included in the analytic sample The figure displays the relative proportion of veterans with coronavirus disease 2019 (COVID-19) who were cared for in the general ward and intensive care 5275 Admitted to general wards (63.1%) 3143 Admitted to ICU (36.9%) unit (ICU). JAMA Network Open. 2021;4(1):e2034266. doi:10.1001/jamanetworkopen.2020.34266 (Reprinted) January 19, 2021 4/11 JAMA Network Open | Public Health Association of Intensive Care Unit Load and Demand With COVID-19 Mortality in VA Hospitals Table 1. Baseline Characteristics and Mortality Over Time Patients with COVID-19 by test date, No. (%) Characteristic, No. (%) March (n = 954) April (n = 1594) May (n = 920) June (n = 1314) July (n = 2373) August (n = 1361) P value Age, y <65 412 (43.2) 499 (31.3) 275 (29.9) 564 (42.9) 952 (40.1) 472 (34.7) 65-74 319 (33.4) 561 (35.2) 334 (36.3) 417 (31.7) 784 (33.0) 454 (33.4) <.001 75-84 147 (15.4) 306 (19.2) 173 (18.8) 207 (15.8) 410 (17.3) 269 (19.8) ≥85 76 (8.0) 228 (14.3) 138 (15.0) 126 (9.6) 227 (9.6) 166 (12.2) Men 907 (95.1) 1518 (95.2) 872 (94.8) 1216 (92.5) 2228 (93.9) 1273 (93.5) .03 Race/ethnicity African American 591 (61.9) 822 (51.6) 402 (43.7) 504 (38.4) 887 (37.4) 458 (33.7) Other or unknown 54 (5.7) 88 (5.5) 62 (6.7) 115 (8.8) 199 (8.4) 116 (8.5) <.001 White 309 (32.4) 684 (42.9) 456 (49.6) 695 (52.9) 1287 (54.2) 787 (57.8) Primary care visit in prior 2 y 912 (95.6) 1479 (92.8) 834 (90.7) 1254 (95.4) 2243 (94.5) 1256 (92.3) <.001 COVID-19 ICU load ≤25% 487 (51.0) 952 (59.7) 785 (85.3) 1170 (89.0) 1923 (81.0) 1250 (91.8) >25% to 50% 141 (14.8) 305 (19.1) 77 (8.4) 126 (9.6) 443 (18.7) 106 (7.8) >50% to 75% 178 (18.7) 181 (11.4) 53 (5.8) 17 (1.3) 7 (0.3) 5 (0.4) <.001 >75% to 100% 88 (9.2) 138 (8.7) 5 (0.5) 1 (0.1) 0 0 >100% 60 (6.3) 18 (1.1) 0 0 0 0 COVID-19 ICU demand ≤25% 241 (25.3) 336 (21.1) 381 (41.4) 472 (35.9) 516 (21.7) 542 (39.8) >25% to 50% 219 (23.0) 466 (29.2) 307 (33.4) 434 (33.0) 772 (32.5) 498 (36.6) <.001 >50% to 75% 261 (27.4) 470 (29.5) 188 (20.4) 266 (20.2) 672 (28.3) 242 (17.8) >75% 233 (24.4) 322 (20.2) 44 (4.8) 142 (10.8) 413 (17.4) 79 (5.8) Hypertension 727 (76.2) 1272 (79.8) 706 (76.7) 970 (73.8) 1774 (74.8) 1023 (75.2) .002 Atrial fibrillation 127 (13.3) 302 (18.9) 174 (18.9) 191 (14.5) 362 (15.3) 231 (17.0) <.001 Smoking history Never 336 (35.2) 449 (28.2) 248 (27.0) 480 (36.5) 780 (32.9) 442 (32.5) Current 89 (9.3) 213 (13.4) 128 (13.9) 164 (12.5) 346 (14.6) 204 (15.0) <.001 Former 420 (44.0) 661 (41.5) 364 (39.6) 546 (41.6) 1006 (42.4) 532 (39.1) Unknown 109 (11.4) 271 (17.0) 180 (19.6) 124 (9.4) 241 (10.2) 183 (13.4) BMI <25 (reference) 193 (20.2) 451 (28.3) 274 (29.8) 307 (23.4) 555 (23.4) 328 (24.1) 25-29 271 (28.4) 481 (30.2) 278 (30.2) 405 (30.8) 702 (29.6) 416 (30.6) 30-34 260 (27.3) 339 (21.3) 194 (21.1) 315 (24.0) 573 (24.1) 323 (23.7) <.001 ≥35 218 (22.9) 300 (18.8) 146 (15.9) 267 (20.3) 513 (21.6) 265 (19.5) Missing data 12 (1.3) 23 (1.4) 28 (3.0) 20 (1.5) 30 (1.3) 29 (2.1) Place of treatment General ward 527 (55.2) 965 (60.5) 546 (59.3) 829 (63.1) 1603 (67.6) 903 (66.3) ICU <.001 No ventilation 195 (20.4) 367 (23.0) 251 (27.3) 346 (26.3) 536 (22.6) 341 (25.1) Mechanical ventilation 232 (24.3) 262 (16.4) 123 (13.4) 139 (10.6) 234 (9.9) 117 (8.6) APACHE score, median (IQR) 12 (6-21) 13 (6-22) 12 (5-20) 10 (4-17) 10 (4-19) 10 (4-18) <.001 Charlson comorbidity index, 3 (1-6) 3 (1-6) 3 (1-6) 2 (1-5) 3 (1-5) 3 (1-5) <.001 median (IQR) Died 218 (22.9) 399 (25.0) 143 (15.5) 179 (13.6) 297 (12.5) 174 (12.8) <.001 Abbreviations: APACHE, Acute Physiology, Age, Chronic Health Evaluation; BMI, body Calculated as mean No. of patients with COVID-19 in the ICU during stay divided by mass index (calculated as weight in kilograms divided by height in meters squared); maximum No. of patients with COVID-19 in the ICU. COVID-19, coronavirus disease 2019; ICU, intensive care unit; IQR, interquartile range. d A physiologic measure in which increasing scores indicate worse health. This score Other includes Hispanic or Latino, Asian, Native Hawaiian or other Pacific Islander, includes laboratory data (eg, white blood cell count) as well as vital signs (eg, American Indian or Alaskan Native, or other race/ethnicity. oxygenation and blood pressure). b e Calculated as mean No. of patients with COVID-19 in the ICU during stay divided by No. A measure of medical comorbidity in which increasing values indicate a greater of ICU beds. comorbidity burden. JAMA Network Open. 2021;4(1):e2034266. doi:10.1001/jamanetworkopen.2020.34266 (Reprinted) January 19, 2021 5/11 JAMA Network Open | Public Health Association of Intensive Care Unit Load and Demand With COVID-19 Mortality in VA Hospitals load was 100% or more (P = .049). The association between COVID-19 ICU load and mortality among patients treated in the general ward was statistically significant but did not form a monotonic gradient; the adjusted HR for all-cause mortality was 1.30 (95% CI, 0.92-1.84) for patients treated during periods when COVID-19 ICU load was greater than 25% to 50%, 0.74 (95% CI, 0.42-1.32) when COVID-19 ICU load was greater than 50% to 75%, 1.90 (95% CI, 0.98-3.65) when COVID-19 ICU load was greater than 75% to 100%, and 1.14 (95% CI, 0.29-4.49) when COVID-19 ICU load was 100% or more (P = .04). Compared with patients with COVID-19 in the ICU treated during periods of low COVID-19 ICU demand (ie,25%), the adjusted HR for all-cause mortality was 0.99 (95% CI, 0.81-1.22; P =.93) for patients treated during periods when COVID-19 ICU demand was greater than 25% to 50%, 1.19 (95% CI, 0.95-1.48 P = .13) when COVID-19 ICU demand was greater than 50% to 75%, and 1.94 (95% CI, 1.46-2.59; P < .001) when COVID-19 ICU demand was greater than 75% to 100%. No statistically significant association between COVID-19 ICU demand and mortality was observed among patients with COVID-19 who were not in the ICU. The eTable in the Supplement provides the adjusted HRs for the 2 measures of COVID-19 ICU strain early in the pandemic (ie, March-May 2020) and later in the pandemic (ie, June-August 2020), and these data are consistent with overall study findings. Discussion In this cohort study of patients with COVID-19 in US VA hospitals, receiving treatment during peak COVID-19 ICU demand, with demand describing the caseload of patients with COVID-19 in the ICU when the patient was treated compared with peak COVID-19 ICU caseload, was consistently and independently associated with COVID-19 ICU mortality. In the extreme case, the adjusted hazard of death was 1.94 for patients with COVID-19 treated in the ICU during periods with greater than 75% to 100% of the peak COVID-19 ICU caseload. The finding that COVID-19 ICU demand was associated with increased mortality for patients with critical COVID-19 early in the pandemic (ie, March-May) and later in the pandemic (ie, June-August) supports the overall study results that suggested that strains on critical care capacity were associated with increased COVID-19 ICU mortality. Tracking COVID-19 ICU demand may be useful to hospital administrators and health officials as they seek to implement interventions to optimize outcomes for patients with COVID-19. COVID-19 ICU demand can be calculated only retrospectively (because the peak number of patients with COVID-19 in the ICU can be assessed only retrospectively). However, facilities can identify the peak surge caseload since the pandemic started, in March 2020, and prospectively monitor COVID-19 ICU demand. Facilities within a health care system or within a geographic region could collaborate to triage patients with critical COVID-19 to sites with greater ICU capacity to reduce strain on any 1 9,10 facility. Future research is urgently needed to investigate the mechanisms by which COVID-19 ICU Table 2. Unadjusted Mortality by COVID-19 ICU Strain Metrics COVID-19 mortality through 30 d postdischarge, No./total No. (%) ICU strain at patient level Overall General ward only ICU COVID-19 ICU load ≤25% 950/6567 (14.5) 313/4303 (7.3) 637/2264 (28.1) >25% to 50% 259/1198 (21.6) 75/662 (11.3) 184/536 (34.3) >50% to 75% 96/441 (21.8) 22/233 (9.4) 74/208 (35.6) >75% to 100% 79/232 (34.0) 28/132 (21.2) 51/100 (51.0) >100% 26/78 (33.3) 3/43 (7.0) 23/35 (65.7) Abbreviations: COVID-19, coronavirus disease 2019; b ICU, intensive care unit. COVID-19 ICU demand Calculated as No. patients with COVID-19 in the ICU ≤25% 281/2488 (11.3) 116/1815 (6.4) 165/673 (24.5) during stay divided by No. of ICU beds. >25% to 50% 429/2696 (15.9) 145/1683 (8.6) 284/1013 (28.0) Calculated as mean No. of patients with COVID-19 in >50% to 75% 443/2099 (21.1) 106/1107 (9.6) 337/992 (34.0) the ICU during stay divided by maximum No. of >75% 257/1233 (20.8) 74/768 (9.6) 183/465 (39.4) patients with COVID-19 in the ICU. JAMA Network Open. 2021;4(1):e2034266. doi:10.1001/jamanetworkopen.2020.34266 (Reprinted) January 19, 2021 6/11 JAMA Network Open | Public Health Association of Intensive Care Unit Load and Demand With COVID-19 Mortality in VA Hospitals Table 3. Proportional Hazard Results From Admission to 30 Days Postdischarge or Death Overall General ward only ICU Characteristic Adjusted HR (95% CI) P value Adjusted HR (95% CI) P value Adjusted HR (95% CI) P value Age, y <65 1 [Reference] 1 [Reference] 1 [Reference] 65-74 2.10 (1.75-2.52) 2.35 (1.55-3.57) 2.08 (1.70-2.55) <.001 <.001 <.001 75-84 3.04 (2.50-3.69) 4.79 (3.15-7.29) 2.73 (2.18-3.42) ≥85 6.85 (5.59-8.39) 11.15 (7.33-16.97) 5.39 (4.20-6.92) Women 0.86 (0.60-1.25) .44 0.68 (0.32-1.46) .32 0.97 (0.63-1.48) .88 Race/ethnicity African American 0.77 (0.68-0.87) 0.61 (0.48-0.77) 0.86 (0.74-1.00) Other or unknown 1.01 (0.81-1.26) 0.93 (0.64-1.36) 1.02 (0.78-1.34) <.001 <.001 .11 White 1 [Reference] 1 [Reference] 1 [Reference] Primary care within prior 2 y 0.86 (0.68-1.09) .20 0.78 (0.54-1.14) .20 0.93 (0.69-1.26) .63 COVID-19 ICU load ≤25% 1 [Reference] 1 [Reference] 1 [Reference] >25% to 50% 1.19 (0.99-1.43) 1.30 (0.92-1.84) 1.10 (0.88-1.37) >50% to 75% 1.03 (0.77-1.40) 0.74 (0.42-1.32) 1.15 (0.81-1.64) .01 .04 .049 >75% to 100% 1.63 (1.13-2.35) 1.90 (0.98-3.65) 1.67 (1.08-2.60) >100% 2.03 (1.16-3.56) 1.14 (0.29-4.49) 2.35 (1.25-4.39) COVID-19 ICU demand ≤25% 1 [Reference] 1 [Reference] 1 [Reference] >25% to 50% 1.11 (0.94-1.31) 1.41 (1.07-1.84) 0.99 (0.81-1.22) <.001 .09 <.001 >50% to 75% 1.25 (1.05-1.49) 1.30 (0.95-1.79) 1.19 (0.95-1.48) >75% 1.67 (1.33-2.11) 1.29 (0.85-1.97) 1.94 (1.46-2.59) Hypertension 0.94 (0.80-1.11) .46 0.92 (0.68-1.24) .55 0.94 (0.77-1.14) .51 Atrial fibrillation 1.00 (0.88-1.15) .95 1.01 (0.81-1.27) .93 1.00 (0.84-1.18) .96 History of smoking Never 1 [Reference] 1 [Reference] 1 [Reference] Current 0.70 (0.56-0.87) 0.73 (0.47-1.14) 0.74 (0.57-0.96) <.001 .001 <.001 Former 0.93 (0.81-1.06) 0.98 (0.77-1.26) 0.92 (0.78-1.08) Unknown 1.55 (1.30-1.84) 1.59 (1.19-2.12) 1.59 (1.27-1.98) BMI <25 1 [Reference] 1 [Reference] 1 [Reference] 25-29 0.83 (0.72-0.95) 0.75 (0.59-0.94) 0.88 (0.74-1.05) 30-34 0.80 (0.68-0.94) 0.72 (0.53-0.97) 0.84 (0.69-1.02) .001 .01 .12 ≥35 0.94 (0.79-1.11) 0.96 (0.68-1.36) 0.98 (0.79-1.20) Missing data 1.38 (0.97-1.98) 1.65 (0.92-2.93) 1.39 (0.87-2.22) Place of treatment General ward 1 [Reference] NA NA ICU <.001 NA <.001 No ventilation 1.93 (1.66-2.24) NA 1 [Reference] Mechanical ventilation 7.40 (6.47-8.48) NA 3.93 (3.39-4.55) APACHE score 1.02 (1.02-1.03) <.001 1.04 (1.03-1.05) <.001 1.02 (1.01-1.02) <.001 Charlson comorbidity score 1.05 (1.03-1.06) <.001 1.09 (1.06-1.12) <.001 1.02 (1.00-1.05) .03 Month of COVID diagnosis March 1.46 (1.15-1.84) 1.43 (0.92-2.22) 1.50 (1.13-2.00) April 1.47 (1.20-1.79) 1.48 (1.04-2.10) 1.52 (1.19-1.96) May 1 [Reference] 1 [Reference] 1 [Reference] <.001 .004 .009 June 1.15 (0.91-1.45) 1.11 (0.74-1.68) 1.22 (0.91-1.62) July 1.05 (0.85-1.30) 0.83 (0.57-1.20) 1.23 (0.94-1.61) August 1.23 (0.97-1.55) 0.84 (0.56-1.26) 1.50 (1.13-2.00) (continued) JAMA Network Open. 2021;4(1):e2034266. doi:10.1001/jamanetworkopen.2020.34266 (Reprinted) January 19, 2021 7/11 JAMA Network Open | Public Health Association of Intensive Care Unit Load and Demand With COVID-19 Mortality in VA Hospitals Table 3. Proportional Hazard Results From Admission to 30 Days Postdischarge or Death (continued) Overall General ward only ICU Characteristic Adjusted HR (95% CI) P value Adjusted HR (95% CI) P value Adjusted HR (95% CI) P value Facility complexity 1a 1.05 (0.67-1.66) 1.41 (0.54-3.68) 0.88 (0.52-1.50) 1b 0.94 (0.59-1.51) 1.29 (0.48-3.47) 0.82 (0.47-1.41) .08 .80 .08 1c 1.29 (0.81-2.07) 1.50 (0.56-4.04) 1.17 (0.68-2.02) 2 1 [Reference] 1 [Reference] 1 [Reference] Abbreviations: APACHE, Acute Physiology, Age, Chronic Health Evaluation; BMI, body A measure of medical comorbidity in which increasing values indicate a greater mass index (calculated as weight in kilograms divided by height in meters squared); comorbidity burden. COVID-19, coronavirus disease 2019; HR, hazard ratio; ICU, intensive care unit; NA, not f The complexity of the services provided at Veterans Affairs facilities, with level 1a being applicable. the most complex. The facility complexity model includes ICU level, operative The other race/ethnicity category includes Hispanic or Latino, Asian, Native Hawaiian complexity level, patient clinical classification, teaching status characteristics, amount or other Pacific Islander, American Indian or Alaskan Native, or other race/ethnicity. of research funding, complex clinical programs provided (eg, invasive catheterization b laboratory, neurosurgery, or transplant), rurality, care provided in the community, and Calculated as No. patients with COVID-19 in the ICU during stay divided by No. of mental health programs provided. ICU beds. Calculated as mean No. of patients with COVID-19 in the ICU during stay divided by maximum No. of patients with COVID-19 in the ICU. A physiologic measure in which increasing scores indicate worse health. This score includes laboratory data (eg, white blood cell count) as well as vital signs (eg, oxygenation and blood pressure). demand may be associated with increased mortality; it is imperative that we understand the degree to which patient characteristics (eg, disease severity) or facility issues (eg, staffing) contribute to the association between COVID-19 ICU strain and poor patient outcomes among patients with critical COVID-19. We did not have a formal measure of ICU capacity, because VA ICU bed availability is not fixed but instead depends on staffing availability; therefore, we calculated COVID-19 ICU load as the ratio of ICU COVID-19 occupancy to the maximum ICU bed number as a surrogate for COVID-19 ICU capacity. Although the association between COVID-19 ICU load and patient mortality was statistically significant, it was neither as consistent over time nor as robust as the association between COVID-19 ICU demand and mortality. We hypothesize that facilities increased their critical care capacity in response to the pandemic and that the degree of this augmentation varied across facilities. Therefore, the comparison with a fixed number of patient beds was likely a relatively poor measure of ICU capacity during the pandemic. Given that hospitals are charged with caring for patients with non–COVID-19 critical illness as well as patients with COVID-19, future studies should seek to examine whether measures of critical care strain that include all patients in the ICU (not just those with COVID-19) are associated with patient outcomes. Future studies should also evaluate whether ICU load provides an adequate measure of strain across the broad spectrum of VA and non-VA hospitals, which vary greatly in prepandemic ICU bed number and the potential to augment capacity during a 11 12 pandemic. Our overall study findings are supported by cohort studies from 2013 and 2018 11,12 demonstrating that as ICUs are strained, mortality increases. It may be the case that during periods of peak ICU caseload, patients who would be admitted to the ICU under more typical conditions are instead admitted to the ward. Our data did not allow us to examine this issue directly; however, we did examine outcomes associated with COVID-ICU strain separately among patients in the general ward and patients in the ICU. Although the association between COVID-19 ICU load and general ward mortality was statistically significant, it varied over time (ie, early vs later in the pandemic). Future research should examine how critical care strains may be associated with outcomes in the general ward for patients with COVID-19 and those without COVID-19. JAMA Network Open. 2021;4(1):e2034266. doi:10.1001/jamanetworkopen.2020.34266 (Reprinted) January 19, 2021 8/11 JAMA Network Open | Public Health Association of Intensive Care Unit Load and Demand With COVID-19 Mortality in VA Hospitals Limitations This study has several limitations. First, this study evaluated care of patients with COVID-19 at VA hospitals; future studies should examine the association between COVID-19 ICU burden and mortality in non-VA facilities. Second, this study focused on COVID-19 mortality; future studies should examine the potential associations of COVID-19 ICU load and demand with outcomes among patients without COVID-19. Third, the results of this study should not be interpreted as a statement on scarcity of critical care or mechanical ventilation; we have no data to suggest that patients needing critical care or mechanical ventilation did not receive this care. Fourth, although the risk adjustment models included demographic and clinical characteristics, they did not include social determinants of health (eg, income or education), which may contribute to COVID-19 mortality. Fifth, we did not examine changes in ICU staffing during the study period. Sixth, we do not have a measure of the degree to which facilities expanded ICU capacity during the pandemic. Seventh, patients with COVID-19 who were admitted to the ICU service could have physically been in diverse settings, including locations designated as the COVID-19 ICU, such as surgical ICUs; some patients with critical COVID-19 were cared for by the ICU team but were physically located in the emergency department. Eighth, related to the observed change in mortality over time, our results suggest that changes in patient characteristics and measures of COVID-19 ICU strain were associated with some of the variation in mortality over time; however, given the observational nature of these data, causality cannot be inferred. Other potential causes (eg, use of medications, such as remdesivir and dexamethasone; clinical practices, such as proning; and unmeasured changes in patient characteristics, such as susceptibility) may have contributed to changes in COVID-19 mortality. Conclusions In this cohort study of patients with COVID-19 in US VA hospitals, COVID-19 ICU demand—a measure of COVID-19 ICU caseload when a patient was treated compared with peak COVID-19 ICU caseload—was associated with mortality among patients with COVID-19 in the ICU. Public health officials and hospital administrators may seek to prevent high COVID-19 ICU demand to optimize outcomes for patients with COVID-19. ARTICLE INFORMATION Accepted for Publication: November 28, 2020. Published: January 19, 2021. doi:10.1001/jamanetworkopen.2020.34266 Open Access: This is an open access article distributed under the terms of the CC-BY License.©2021BravataDM et al. JAMA Network Open. Corresponding Author: Dawn M. Bravata, MD, Health Services Research and Development Center for Health Information and Communication, Richard L. Roudebush VA Medical Center, 1481 W 10th St, HSR&D Mail Code 11H, Indianapolis, IN 46202 (Dawn.Bravata2@va.gov). Author Affiliations: Precision Monitoring to Transform Care Quality Enhancement Research Initiative, Health Services Research and Development, Department of Veterans Affairs, Indianapolis, Indiana (Bravata, Perkins, L. J. Myers, Arling, Zhang, J. Myers, Austin, Sexson, Keyhani); Health Services Research and Development Center for Health Information and Communication, Richard L. Roudebush VA Medical Center, Department of Veterans Affairs, Indianapolis, Indiana (Bravata, L. J. Myers, J. Myers, Austin, Sexson, Dev); Medicine Service, Richard L. Roudebush VA Medical Center, Indianapolis, Indiana (Bravata, Reese, Dysangco, Agarwal, Dev); Department of Medicine, Indiana University School of Medicine, Indianapolis (Bravata, L. J. Myers, Reese, Dysangco, Agarwal); Department of Neurology, Indiana University School of Medicine, Indianapolis (Bravata); William M. Tierney Center for Health Services Research, Regenstrief Institute, Indianapolis, Indiana (Bravata, L. J. Myers, Dev); Department of Biostatistics, Indiana University School of Medicine, Indianapolis (Perkins); School of Nursing, Purdue University, West Lafayette, Indiana (Arling); Department of Biostatistics, College of Public Health, University of Nebraska Medical Center, Omaha (Zhang); Department of Pharmacy Practice, College of Pharmacy, Purdue University, West Lafayette, Indiana (Zillich); Northern California Institute for Research and Education, San Francisco (Leonard); San JAMA Network Open. 2021;4(1):e2034266. doi:10.1001/jamanetworkopen.2020.34266 (Reprinted) January 19, 2021 9/11 JAMA Network Open | Public Health Association of Intensive Care Unit Load and Demand With COVID-19 Mortality in VA Hospitals Francisco VA Medical Center, San Francisco, California (Keyhani); Division of General Internal Medicine, Department of Medicine, University of California, San Francisco (Keyhani). Author Contributions: Dr Bravata and Mr Perkins had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Concept and design: Bravata, Arling, Zhang, Zillich, Keyhani. Acquisition, analysis, or interpretation of data: Bravata, Perkins, L. Myers, Arling, Zhang, Zillich, Reese, Dysangco, Agarwal, J. Myers, Austin, Sexson, Leonard, Dev. Drafting of the manuscript: Bravata, Arling. Critical revision of the manuscript for important intellectual content: Bravata, Perkins, L. Myers, Zhang, Zillich, Reese, Dysangco, Agarwal, J. Myers, Austin, Sexson, Leonard, Dev, Keyhani. Statistical analysis: Bravata, Perkins, Arling, Zhang. Obtained funding: Bravata. Administrative, technical, or material support: L. Myers, Reese, J. Myers, Austin, Sexson, Keyhani. Supervision: Bravata, Zhang. Conflict of Interest Disclosures: Drs Bravata and Perkins reported receiving grants from the Department of Veterans Affairs during the conduct of the study. Dr Agarwal reported receiving personal fees and travel support from Bayer, Relypsa, Reata, Sanofi, Boehringer, and Merck; personal fees from Janssen, DiaMedica, Lexicon, Akebia, Eli Lilly, and Astra Zeneca; and travel support from Akebia outside the submitted work. No other disclosures were reported. Funding/Support: This work was supported by the US Department of Veterans Affairs Health Services Research and Development Service Precision Monitoring to Transform Care Quality Enhancement Research Initiative (QUE 15-280). Role of the Funder/Sponsor: The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. Disclaimer: The contents of this study do not represent the views of the US Department of Veterans Affairs or the US government. REFERENCES 1. Pan A, Liu L, Wang C, et al. Association of public health interventions with the epidemiology of the COVID-19 outbreak in Wuhan, China. JAMA. 2020;323(19):1915-1923. doi:10.1001/jama.2020.6130 2. Hartley DM, Perencevich EN. Public health interventions for COVID-19: emerging evidence and implications for an evolving public health crisis. JAMA. 2020;323(19):1908-1909. doi:10.1001/jama.2020.5910 3. Eriksson CO, Stoner RC, Eden KB, Newgard CD, Guise JM. The association between hospital capacity strain and inpatient outcomes in highly developed countries: a systematic review. J Gen Intern Med. 2017;32(6):686-696. doi:10.1007/s11606-016-3936-3 4. VA Health Services Research and Development. VA Informatics and Computing Infrastructure (VINCI). Accessed December 10, 2020. https://www.hsrd.research.va.gov/for_researchers/vinci/default.cfmhttps://vaww. vinci.med.va.gov/VinciCentral/Home/About 5. Borzecki AM, Wong AT, Hickey EC, Ash AS, Berlowitz DR. Can we use automated data to assess quality of hypertension care? Am J Manag Care. 2004;10(7 Pt 2):473-479. 6. Sohn MW, Arnold N, Maynard C, Hynes DM. Accuracy and completeness of mortality data in the Department of Veterans Affairs. Popul Health Metr. 2006;4:2. doi:10.1186/1478-7954-4-2 7. Knaus WA, Wagner DP, Draper EA, et al. The APACHE III prognostic system: risk prediction of hospital mortality for critically ill hospitalized adults. Chest. 1991;100(6):1619-1636. doi:10.1378/chest.100.6.1619 8. Kim JH, Hong SK, Kim Y, et al. Experience of augmenting critical care capacity in Daegu during COVID-19 incident in South Korea. Acute Crit Care. 2020;35(2):110-114. doi:10.4266/acc.2020.00275 9. Einav S, Hick JL, Hanfling D, et al; Task Force for Mass Critical Care. Surge capacity logistics: care of the critically ill and injured during pandemics and disasters: CHEST consensus statement. Chest. 2014;146(4)(suppl): e17S-e43S. doi:10.1378/chest.14-0734 10. Ma X, Vervoort D. Critical care capacity during the COVID-19 pandemic: global availability of intensive care beds. J Crit Care. 2020;58:96-97. doi:10.1016/j.jcrc.2020.04.012 11. Gabler NB, Ratcliffe SJ, Wagner J, et al. Mortality among patients admitted to strained intensive care units. Am J Respir Crit Care Med. 2013;188(7):800-806. doi:10.1164/rccm.201304-0622OC JAMA Network Open. 2021;4(1):e2034266. doi:10.1001/jamanetworkopen.2020.34266 (Reprinted) January 19, 2021 10/11 JAMA Network Open | Public Health Association of Intensive Care Unit Load and Demand With COVID-19 Mortality in VA Hospitals 12. Anesi GL, Liu VX, Gabler NB, et al. Associations of intensive care unit capacity strain with disposition and outcomes of patients with sepsis presenting to the emergency department. Ann Am Thorac Soc. 2018;15(11): 1328-1335. doi:10.1513/AnnalsATS.201804-241OC 13. Feinstein AR, Sosin DM, Wells CK. The Will Rogers phenomenon: stage migration and new diagnostic techniques as a source of misleading statistics for survival in cancer. N Engl J Med. 1985;312(25):1604-1608. doi: 10.1056/NEJM198506203122504 14. Yang X, Yu Y, Xu J, et al. Clinical course and outcomes of critically ill patients with SARS-CoV-2 pneumonia in Wuhan, China: a single-centered, retrospective, observational study. Lancet Respir Med. 2020;8(5):475-481. doi: 10.1016/S2213-2600(20)30079-5 15. Horby P, Lim WS, Emberson JR, et al; RECOVERY Collaborative Group. Dexamethasone in hospitalized patients with Covid-19 — preliminary report. N Engl J Med.2020. SUPPLEMENT. eTable. Proportional Hazard Results for COVID-19 ICU Strain: Time From Admission to 30 Days Postdischarge or Death by Temporal Period JAMA Network Open. 2021;4(1):e2034266. doi:10.1001/jamanetworkopen.2020.34266 (Reprinted) January 19, 2021 11/11 Supplemental Online Content Bravata DM, Perkins AJ, Myers LJ, et al. Association of intensive care unit patient load and demand with mortality rates in US Department of Veterans Affairs hospitals during the COVID-19 pandemic. JAMA Netw Open. 2021;4(1):e2034266. doi: 10.1001/jamanetworkopen.2020.34266 eTable. Proportional Hazard Results for COVID-19 ICU Strain: Time From Admission to 30 Days Postdischarge or Death by Temporal Period This supplemental material has been provided by the authors to give readers additional information about their work. © 2021 Bravata DM et al. JAMA Network Open. eTable. Proportional Hazard Results for COVID-19 ICU Strain: Time From Admission to 30 Days Postdischarge or Death by Temporal Period Baseline Characteristic Overall General Ward Only Intensive Care Unit (ICU) HR (95% CI) P-value HR (95% CI) P-value HR (95% CI) P-value March-May 2020* COVID-19 ICU-Load: Mean number COVID-19 ICU patients during stay/Number of 0.049 0.010 0.243 ICU Beds 1.00 1.00 1.00 COVID-19 ICU-Demand: Mean number COVID-19 ICU patients during <0.001 0.264 <0.001 stay/Maximum number of COVID patients 1.00 1.00 1.00 1.12 June-August 2020* COVID-19 ICU-Load: Mean number COVID-19 ICU patients during stay/Number of ICU Beds 1.00 1.00 1.00 -- -- -- -- -- -- COVID-19 ICU-Demand: Mean number COVID-19 ICU patients during 0.016 0.023 stay/Maximum number of COVID patients 1.00 1.00 1.00 *The temporal period refers to the date of the first COVID-19 laboratory test per patient. © 2021 Bravata DM et al. JAMA Network Open. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png JAMA Network Open American Medical Association

Association of Intensive Care Unit Patient Load and Demand With Mortality Rates in US Department of Veterans Affairs Hospitals During the COVID-19 Pandemic

Download PDF
13 pages

Loading next page...
 
/lp/american-medical-association/association-of-intensive-care-unit-patient-load-and-demand-with-RLffcvGxw5

References (40)

Publisher
American Medical Association
Copyright
Copyright 2021 Bravata DM et al. JAMA Network Open.
eISSN
2574-3805
DOI
10.1001/jamanetworkopen.2020.34266
Publisher site
See Article on Publisher Site

Abstract

Key Points Question Is greater coronavirus disease IMPORTANCE Although strain on hospital capacity has been associated with increased mortality in 2019 (COVID-19) intensive care unit nonpandemic settings, studies are needed to examine the association between coronavirus disease (ICU) strain associated with increased 2019 (COVID-19) critical care capacity and mortality. COVID-19 mortality? Findings In this cohort study of 8516 OBJECTIVE To examine whether COVID-19 mortality was associated with COVID-19 intensive care patients with COVID-19 admitted to 88 unit (ICU) strain. US Veterans Affairs hospitals, strains on critical care capacity were associated DESIGN, SETTING, AND PARTICIPANTS This cohort study was conducted among veterans with with increased COVID-19 mortality. COVID-19, as confirmed by polymerase chain reaction or antigen testing in the laboratory from March Among patients with COVID-19, those through August 2020, cared for at any Department of Veterans Affairs (VA) hospital with 10 or more treated in the ICU during periods of peak patients with COVID-19 in the ICU. The follow-up period was through November 2020. Data were COVID-19 ICU demand had a nearly analyzed from March to November 2020. 2-fold increased risk of mortality compared with those treated during EXPOSURES Receiving treatment for COVID-19 in the ICU during a period of increased COVID-19 ICU periods of low demand. load, with load defined as mean number of patients with COVID-19 in the ICU during the patient’s hospital stay divided by the number of ICU beds at that facility, or increased COVID-19 ICU demand, Meaning These findings suggest that with demand defined as mean number of patients with COVID-19 in the ICU during the patient’s stay public health officials and hospital divided by the maximum number of patients with COVID-19 in the ICU. administrators should consider interventions that reduce COVID-19 ICU MAIN OUTCOMES AND MEASURES All-cause mortality was recorded through 30 days after demand to improve survival among discharge from the hospital. patients with COVID-19 in the ICU. RESULTS Among 8516 patients with COVID-19 admitted to 88 VA hospitals, 8014 (94.1%) were men Invited Commentary and mean (SD) age was 67.9 (14.2) years. Mortality varied over time, with 218 of 954 patients (22.9%) dying in March, 399 of 1594 patients (25.0%) dying in April, 143 of 920 patients (15.5%) dying in May, Multimedia 179 of 1314 patients (13.6%) dying in June, 297 of 2373 patients (12.5%) dying in July, and 174 of 1361 Supplemental content (12.8%) patients dying in August (P < .001). Patients with COVID-19 who were treated in the ICU during periods of increased COVID-19 ICU demand had increased risk of mortality compared with Author affiliations and article information are listed at the end of this article. patients treated during periods of low COVID-19 ICU demand (ie, demand of25%); the adjusted hazard ratio for all-cause mortality was 0.99 (95% CI, 0.81-1.22; P = .93) for patients treated when COVID-19 ICU demand was more than 25% to 50%, 1.19 (95% CI, 0.95-1.48; P = .13) when COVID-19 ICU demand was more than 50% to 75%, and 1.94 (95% CI, 1.46-2.59; P < .001) when COVID-19 ICU demand was more than 75% to 100%. No association between COVID-19 ICU demand and mortality was observed for patients with COVID-19 not in the ICU. The association between COVID-19 ICU load and mortality was not consistent over time (ie, early vs late in the pandemic). CONCLUSIONS AND RELEVANCE This cohort study found that although facilities augmented ICU capacity during the pandemic, strains on critical care capacity were associated with increased (continued) Open Access. This is an open access article distributed under the terms of the CC-BY License. JAMA Network Open. 2021;4(1):e2034266. doi:10.1001/jamanetworkopen.2020.34266 (Reprinted) January 19, 2021 1/11 JAMA Network Open | Public Health Association of Intensive Care Unit Load and Demand With COVID-19 Mortality in VA Hospitals Abstract (continued) COVID-19 ICU mortality. Tracking COVID-19 ICU demand may be useful to hospital administrators and health officials as they coordinate COVID-19 admissions across hospitals to optimize outcomes for patients with this illness. JAMA Network Open. 2021;4(1):e2034266. doi:10.1001/jamanetworkopen.2020.34266 Introduction Health policy interventions (eg, social distancing) implemented to avoid overloading health care systems have been associated with reduced coronavirus disease 2019 (COVID-19) hospitalization 1,2 rates. Hospital capacity strain resulting from increased patient volume or disease severity is associated with increased mortality in nonpandemic settings. The association between COVID-19 critical care strain and mortality has not been examined. The objective of this study was to examine the associations between 2 measures of intensive care unit (ICU) strain and mortality among patients with COVID-19 who were admitted to a Department of Veterans Affairs (VA) facility. Methods This cohort study was approved by the Indiana University School of Medicine institutional review board and the VA research and development committee at Richard L. Roudebush VA Medical Center. This was an observational study without any direct patient contact and considered to be of minimal risk; therefore, a waiver of informed consent was obtained. This study adheres to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline. This observational cohort study included patients with polymerase chain reaction or antigen test results positive for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) from March 1 to August 31, 2020. We included patients admitted to a VA hospital that had 10 or more patients with COVID-19 in the ICU during the study period. We excluded patients whose SARS-CoV-2 test date was more than 2 days after admission. Follow-up extended through November 1, 2020. Data from the VA Corporate Data Warehouse (CDW), including inpatient and outpatient data files from the 2 years before the patient’s COVID-19 diagnosis through the day of admission, were used to identify comorbid conditions, self-reported race/ethnicity, health care use, procedures 4,5 received, vital signs, and laboratory data. Death dates were obtained from the CDW, VA Vital Status File, and patient’s electronic health record. The modified Acute Physiology, Age, Chronic Health Evaluation (APACHE) III score was calculated at the time of hospital admission as a measure of illness severity. Number of ICU beds was obtained from the VA Bed Management Solution initiative. The number of ICU beds per facility described prepandemic ICU capacity, not any augmented capacity that may have been implemented during the COVID-19 pandemic. Facility complexity described the level of services provided at a VA facility, categorized as 1a, 1b, 1c, 2, or 3, with level 1a being the most complex and level 3 being the least complex. Facility complexity included ICU level, operative complexity level, patient clinical classification, teaching status characteristics, amount of research funding, complex clinical programs provided (eg, invasive catheterization laboratory, neurosurgery, and transplant), rurality, care provided in the community, and mental health programs provided. The primary outcome was all-cause mortality. Patients were followed up for 30 days after discharge. Measures of COVID-19 ICU Strain This study included 2 measures of COVID-19 critical care strain: ICU load and ICU demand; Figure 1 displays how these measures vary over time within an example facility. The COVID-19 ICU load described how the caseload of patients with COVID-19 in the ICU compared with the typical ICU bed JAMA Network Open. 2021;4(1):e2034266. doi:10.1001/jamanetworkopen.2020.34266 (Reprinted) January 19, 2021 2/11 JAMA Network Open | Public Health Association of Intensive Care Unit Load and Demand With COVID-19 Mortality in VA Hospitals count at each facility. The COVID-19 ICU load was calculated at the patient level as the mean number of patients with COVID-19 in the ICU during the patient’s hospital stay divided by the number of ICU beds at that facility. The number of ICU beds at each facility was a fixed number. The COVID-19 ICU load included only the number of patients with COVID-19 in the ICU, excluding patients with other critical illnesses. The COVID-19 ICU load ranged from 0% to greater than 100%; it exceeded 100% if the hospital increased critical care bed capacity during the pandemic (eg, by converting a sleep laboratory into an ICU) and if those beds were occupied by patients with COVID-19. The population of patients with COVID-19 in the ICU varied over time, with peak prevalence rates occurring early (eg, March) at some hospitals and later (eg, July) at other hospitals. The COVID-19 ICU demand described the caseload of patients with COVID-19 in the ICU when a patient was treated compared with peak COVID-19 ICU caseload. It was calculated at the patient level as the mean number of patients with COVID-19 in the ICU during the patient’s stay divided by the maximum number of patients with COVID-19 in the ICU at that facility during the study period. The COVID-19 ICU demand ranged from 0% to 100%. For example, if a hospital had 60 ICU beds before the pandemic and the mean number of patients with COVID-19 in the ICU during a patient’s stay was 20, then the COVID-19 ICU load would be 20 divided by 60, or 33%. If at that same facility the peak surge included 20 patients with COVID-19 in the ICU and a patient was treated during the period when the mean of number patients with COVID-19 in the ICU was 20, then the COVID-19 ICU demand would be 20 divided by 20, or 100%. Statistical Analysis We described differences over time in baseline characteristics and mortality among inpatients with COVID-19 using χ and Wilcoxon rank sum tests. We used Cox proportional hazard models to analyze the time in days from admission to death, either in the hospital or within 30 days after discharge, among patients who were admitted to the hospital (overall, in the general ward, and in the ICU). Patients who were still in the hospital or out of the hospital and alive at 30 days after discharge were treated as censored observations. We included a random effect for the facility to account for the correlation of mortality among patients in the same hospital. Analyses were performed using SAS Enterprise Guide statistical software version 7.11 (SAS Institute). P values were 2-sided, and statistical significance was set at P < .05. Data were analyzed from March to November 2020. Figure 1. Coronavirus Disease 2019 (COVID-19) Intensive Care Unit (ICU) Load and Demand at an Example Facility Patients with COVID-19 in the ICU ICU beds Maximum volume of patients with COVID-19 in the ICU Black solid vertical line with arrows indicates the COVID-19 ICU load exceeding 100% of ICU capacity numbers at 1 time point used to calculate COVID-19 ICU load, defined as the mean number of patients with COVID-19 in the ICU during a patient’s hospital stay divided by the number of ICU beds. Black dotted vertical line indicates the numbers at 1 time point used to calculate COVID-19 demand, defined as the mean number of patients with COVID-19 in the ICU during a patient’s hospital stay divided by the maximum number of patients with COVID-19 in the ICU during the study period. The results suggest that the risk of mortality would be highest if a COVID-19 patient’s stay March April May June July August was during the peak of ICU demand and if ICU caseload Time, mo approached or exceeded ICU bed capacity. JAMA Network Open. 2021;4(1):e2034266. doi:10.1001/jamanetworkopen.2020.34266 (Reprinted) January 19, 2021 3/11 Patients with COVID-19 in the ICU, No. JAMA Network Open | Public Health Association of Intensive Care Unit Load and Demand With COVID-19 Mortality in VA Hospitals Results The cohort included 8516 patients with COVID-19 admitted to 88 VA hospitals (Figure 2); 8014 (94.1%) were men, and mean (SD) age was 67.9 (14.2) years. The observed mortality was 218 of 954 patients (22.9%) in March. It increased to 399 of 1594 patients (25.0%) in April, then decreased to 143 of 920 patients (15.5%) in May, leveling off at 179 of 1314 patients (13.6%) in June, 297 of 2373 patients (12.5%) in July, and 174 of 1361 patients (12.8%) in August (P < .001) (Table 1). The proportion of patients ages 75 years or older fluctuated over time; 223 patients (23.4%) were in this age group in March, increasing to 534 patients (33.5%) in April and 311 patients (33.8%) in May, decreasing to 333 patients (25.3%) in June and 637 patients (26.8%) in July, and then increasing again to 435 patients (32.0%) in August (P < .001). The proportion of patients receiving care on general wards increased after the earliest months of the pandemic (March: 527 patients [55.2%]; April: 965 patients [60.5%]; May: 546 patients [59.3%]; June: 829 patients [63.1%]; July: 1603 patients [67.6%]; and August: 903 patients [66.3%]; P < .001). The COVID-19 ICU load and demand changed over time. The proportion of patients with COVID-19 treated during periods of low COVID-19 ICU load (ie,25%) increased over time, with 487 patients (51.0%) in March, 952 patients (59.7%) in April, 785 patients (85.3%) in May, 1170 patients (89.0%) in June, 1923 patients (81.0%) in July, and 1250 patients (91.8%) in August (P < .001). Similarly, the proportion of patients with COVID-19 treated during periods of peak COVID-19 ICU load (ie, >100%) decreased from 60 patients (6.3%) in March to 18 patients (1.1%) in April, to 0 patients in May through August. The proportion of patients with COVID-19 treated during periods of high COVID-19 ICU demand (ie, >75%) decreased during the first 3 months of the pandemic, with 233 patients (24.4%) in March, 322 patients (20.2%) in April, and 44 patients (4.8%) in May; increased to 142 patients (10.8%) in June and 413 patients (17.4%) in July; then decreased to 79 patients (5.8%) in August (P < .001) (Table 1). Patients with COVID-19 in the ICU treated during high COVID-19 ICU strain had increased risk of mortality. Table 2 provides the unadjusted results, and Table 3 displays the adjusted hazard ratios (HRs). Compared with patients with COVID-19 in the ICU treated during periods of low COVID-19 ICU load (ie,25%), the adjusted HR for all-cause mortality was 1.10 (95% CI, 0.88-1.37) for patients treated during periods when COVID-19 ICU load was greater than 25% to 50%, 1.15 (95% CI, 0.81-1.64) when COVID-19 ICU load was greater than 50% to 75%, 1.67 (95% CI, 1.08-2.60) when COVID-19 ICU load was greater than 75% to 100%, and 2.35 (95% CI, 1.25-4.39) when COVID-19 ICU Figure 2. Patient Flow Diagram 48 013 US veterans with suspected COVID-19 347 Excluded because of negative antibody tests 47 666 With COVID-19 38 030 Excluded because they received outpatient care only 9636 Admitted at any VA hospital 659 Excluded because admitted at VA hospitals with <10 ICU patients 8977 Admitted at VA hospital with ≥10 ICU patients 461 Excluded with COVID-19 diagnosis >2 days after admission 8516 Included in the analytic sample The figure displays the relative proportion of veterans with coronavirus disease 2019 (COVID-19) who were cared for in the general ward and intensive care 5275 Admitted to general wards (63.1%) 3143 Admitted to ICU (36.9%) unit (ICU). JAMA Network Open. 2021;4(1):e2034266. doi:10.1001/jamanetworkopen.2020.34266 (Reprinted) January 19, 2021 4/11 JAMA Network Open | Public Health Association of Intensive Care Unit Load and Demand With COVID-19 Mortality in VA Hospitals Table 1. Baseline Characteristics and Mortality Over Time Patients with COVID-19 by test date, No. (%) Characteristic, No. (%) March (n = 954) April (n = 1594) May (n = 920) June (n = 1314) July (n = 2373) August (n = 1361) P value Age, y <65 412 (43.2) 499 (31.3) 275 (29.9) 564 (42.9) 952 (40.1) 472 (34.7) 65-74 319 (33.4) 561 (35.2) 334 (36.3) 417 (31.7) 784 (33.0) 454 (33.4) <.001 75-84 147 (15.4) 306 (19.2) 173 (18.8) 207 (15.8) 410 (17.3) 269 (19.8) ≥85 76 (8.0) 228 (14.3) 138 (15.0) 126 (9.6) 227 (9.6) 166 (12.2) Men 907 (95.1) 1518 (95.2) 872 (94.8) 1216 (92.5) 2228 (93.9) 1273 (93.5) .03 Race/ethnicity African American 591 (61.9) 822 (51.6) 402 (43.7) 504 (38.4) 887 (37.4) 458 (33.7) Other or unknown 54 (5.7) 88 (5.5) 62 (6.7) 115 (8.8) 199 (8.4) 116 (8.5) <.001 White 309 (32.4) 684 (42.9) 456 (49.6) 695 (52.9) 1287 (54.2) 787 (57.8) Primary care visit in prior 2 y 912 (95.6) 1479 (92.8) 834 (90.7) 1254 (95.4) 2243 (94.5) 1256 (92.3) <.001 COVID-19 ICU load ≤25% 487 (51.0) 952 (59.7) 785 (85.3) 1170 (89.0) 1923 (81.0) 1250 (91.8) >25% to 50% 141 (14.8) 305 (19.1) 77 (8.4) 126 (9.6) 443 (18.7) 106 (7.8) >50% to 75% 178 (18.7) 181 (11.4) 53 (5.8) 17 (1.3) 7 (0.3) 5 (0.4) <.001 >75% to 100% 88 (9.2) 138 (8.7) 5 (0.5) 1 (0.1) 0 0 >100% 60 (6.3) 18 (1.1) 0 0 0 0 COVID-19 ICU demand ≤25% 241 (25.3) 336 (21.1) 381 (41.4) 472 (35.9) 516 (21.7) 542 (39.8) >25% to 50% 219 (23.0) 466 (29.2) 307 (33.4) 434 (33.0) 772 (32.5) 498 (36.6) <.001 >50% to 75% 261 (27.4) 470 (29.5) 188 (20.4) 266 (20.2) 672 (28.3) 242 (17.8) >75% 233 (24.4) 322 (20.2) 44 (4.8) 142 (10.8) 413 (17.4) 79 (5.8) Hypertension 727 (76.2) 1272 (79.8) 706 (76.7) 970 (73.8) 1774 (74.8) 1023 (75.2) .002 Atrial fibrillation 127 (13.3) 302 (18.9) 174 (18.9) 191 (14.5) 362 (15.3) 231 (17.0) <.001 Smoking history Never 336 (35.2) 449 (28.2) 248 (27.0) 480 (36.5) 780 (32.9) 442 (32.5) Current 89 (9.3) 213 (13.4) 128 (13.9) 164 (12.5) 346 (14.6) 204 (15.0) <.001 Former 420 (44.0) 661 (41.5) 364 (39.6) 546 (41.6) 1006 (42.4) 532 (39.1) Unknown 109 (11.4) 271 (17.0) 180 (19.6) 124 (9.4) 241 (10.2) 183 (13.4) BMI <25 (reference) 193 (20.2) 451 (28.3) 274 (29.8) 307 (23.4) 555 (23.4) 328 (24.1) 25-29 271 (28.4) 481 (30.2) 278 (30.2) 405 (30.8) 702 (29.6) 416 (30.6) 30-34 260 (27.3) 339 (21.3) 194 (21.1) 315 (24.0) 573 (24.1) 323 (23.7) <.001 ≥35 218 (22.9) 300 (18.8) 146 (15.9) 267 (20.3) 513 (21.6) 265 (19.5) Missing data 12 (1.3) 23 (1.4) 28 (3.0) 20 (1.5) 30 (1.3) 29 (2.1) Place of treatment General ward 527 (55.2) 965 (60.5) 546 (59.3) 829 (63.1) 1603 (67.6) 903 (66.3) ICU <.001 No ventilation 195 (20.4) 367 (23.0) 251 (27.3) 346 (26.3) 536 (22.6) 341 (25.1) Mechanical ventilation 232 (24.3) 262 (16.4) 123 (13.4) 139 (10.6) 234 (9.9) 117 (8.6) APACHE score, median (IQR) 12 (6-21) 13 (6-22) 12 (5-20) 10 (4-17) 10 (4-19) 10 (4-18) <.001 Charlson comorbidity index, 3 (1-6) 3 (1-6) 3 (1-6) 2 (1-5) 3 (1-5) 3 (1-5) <.001 median (IQR) Died 218 (22.9) 399 (25.0) 143 (15.5) 179 (13.6) 297 (12.5) 174 (12.8) <.001 Abbreviations: APACHE, Acute Physiology, Age, Chronic Health Evaluation; BMI, body Calculated as mean No. of patients with COVID-19 in the ICU during stay divided by mass index (calculated as weight in kilograms divided by height in meters squared); maximum No. of patients with COVID-19 in the ICU. COVID-19, coronavirus disease 2019; ICU, intensive care unit; IQR, interquartile range. d A physiologic measure in which increasing scores indicate worse health. This score Other includes Hispanic or Latino, Asian, Native Hawaiian or other Pacific Islander, includes laboratory data (eg, white blood cell count) as well as vital signs (eg, American Indian or Alaskan Native, or other race/ethnicity. oxygenation and blood pressure). b e Calculated as mean No. of patients with COVID-19 in the ICU during stay divided by No. A measure of medical comorbidity in which increasing values indicate a greater of ICU beds. comorbidity burden. JAMA Network Open. 2021;4(1):e2034266. doi:10.1001/jamanetworkopen.2020.34266 (Reprinted) January 19, 2021 5/11 JAMA Network Open | Public Health Association of Intensive Care Unit Load and Demand With COVID-19 Mortality in VA Hospitals load was 100% or more (P = .049). The association between COVID-19 ICU load and mortality among patients treated in the general ward was statistically significant but did not form a monotonic gradient; the adjusted HR for all-cause mortality was 1.30 (95% CI, 0.92-1.84) for patients treated during periods when COVID-19 ICU load was greater than 25% to 50%, 0.74 (95% CI, 0.42-1.32) when COVID-19 ICU load was greater than 50% to 75%, 1.90 (95% CI, 0.98-3.65) when COVID-19 ICU load was greater than 75% to 100%, and 1.14 (95% CI, 0.29-4.49) when COVID-19 ICU load was 100% or more (P = .04). Compared with patients with COVID-19 in the ICU treated during periods of low COVID-19 ICU demand (ie,25%), the adjusted HR for all-cause mortality was 0.99 (95% CI, 0.81-1.22; P =.93) for patients treated during periods when COVID-19 ICU demand was greater than 25% to 50%, 1.19 (95% CI, 0.95-1.48 P = .13) when COVID-19 ICU demand was greater than 50% to 75%, and 1.94 (95% CI, 1.46-2.59; P < .001) when COVID-19 ICU demand was greater than 75% to 100%. No statistically significant association between COVID-19 ICU demand and mortality was observed among patients with COVID-19 who were not in the ICU. The eTable in the Supplement provides the adjusted HRs for the 2 measures of COVID-19 ICU strain early in the pandemic (ie, March-May 2020) and later in the pandemic (ie, June-August 2020), and these data are consistent with overall study findings. Discussion In this cohort study of patients with COVID-19 in US VA hospitals, receiving treatment during peak COVID-19 ICU demand, with demand describing the caseload of patients with COVID-19 in the ICU when the patient was treated compared with peak COVID-19 ICU caseload, was consistently and independently associated with COVID-19 ICU mortality. In the extreme case, the adjusted hazard of death was 1.94 for patients with COVID-19 treated in the ICU during periods with greater than 75% to 100% of the peak COVID-19 ICU caseload. The finding that COVID-19 ICU demand was associated with increased mortality for patients with critical COVID-19 early in the pandemic (ie, March-May) and later in the pandemic (ie, June-August) supports the overall study results that suggested that strains on critical care capacity were associated with increased COVID-19 ICU mortality. Tracking COVID-19 ICU demand may be useful to hospital administrators and health officials as they seek to implement interventions to optimize outcomes for patients with COVID-19. COVID-19 ICU demand can be calculated only retrospectively (because the peak number of patients with COVID-19 in the ICU can be assessed only retrospectively). However, facilities can identify the peak surge caseload since the pandemic started, in March 2020, and prospectively monitor COVID-19 ICU demand. Facilities within a health care system or within a geographic region could collaborate to triage patients with critical COVID-19 to sites with greater ICU capacity to reduce strain on any 1 9,10 facility. Future research is urgently needed to investigate the mechanisms by which COVID-19 ICU Table 2. Unadjusted Mortality by COVID-19 ICU Strain Metrics COVID-19 mortality through 30 d postdischarge, No./total No. (%) ICU strain at patient level Overall General ward only ICU COVID-19 ICU load ≤25% 950/6567 (14.5) 313/4303 (7.3) 637/2264 (28.1) >25% to 50% 259/1198 (21.6) 75/662 (11.3) 184/536 (34.3) >50% to 75% 96/441 (21.8) 22/233 (9.4) 74/208 (35.6) >75% to 100% 79/232 (34.0) 28/132 (21.2) 51/100 (51.0) >100% 26/78 (33.3) 3/43 (7.0) 23/35 (65.7) Abbreviations: COVID-19, coronavirus disease 2019; b ICU, intensive care unit. COVID-19 ICU demand Calculated as No. patients with COVID-19 in the ICU ≤25% 281/2488 (11.3) 116/1815 (6.4) 165/673 (24.5) during stay divided by No. of ICU beds. >25% to 50% 429/2696 (15.9) 145/1683 (8.6) 284/1013 (28.0) Calculated as mean No. of patients with COVID-19 in >50% to 75% 443/2099 (21.1) 106/1107 (9.6) 337/992 (34.0) the ICU during stay divided by maximum No. of >75% 257/1233 (20.8) 74/768 (9.6) 183/465 (39.4) patients with COVID-19 in the ICU. JAMA Network Open. 2021;4(1):e2034266. doi:10.1001/jamanetworkopen.2020.34266 (Reprinted) January 19, 2021 6/11 JAMA Network Open | Public Health Association of Intensive Care Unit Load and Demand With COVID-19 Mortality in VA Hospitals Table 3. Proportional Hazard Results From Admission to 30 Days Postdischarge or Death Overall General ward only ICU Characteristic Adjusted HR (95% CI) P value Adjusted HR (95% CI) P value Adjusted HR (95% CI) P value Age, y <65 1 [Reference] 1 [Reference] 1 [Reference] 65-74 2.10 (1.75-2.52) 2.35 (1.55-3.57) 2.08 (1.70-2.55) <.001 <.001 <.001 75-84 3.04 (2.50-3.69) 4.79 (3.15-7.29) 2.73 (2.18-3.42) ≥85 6.85 (5.59-8.39) 11.15 (7.33-16.97) 5.39 (4.20-6.92) Women 0.86 (0.60-1.25) .44 0.68 (0.32-1.46) .32 0.97 (0.63-1.48) .88 Race/ethnicity African American 0.77 (0.68-0.87) 0.61 (0.48-0.77) 0.86 (0.74-1.00) Other or unknown 1.01 (0.81-1.26) 0.93 (0.64-1.36) 1.02 (0.78-1.34) <.001 <.001 .11 White 1 [Reference] 1 [Reference] 1 [Reference] Primary care within prior 2 y 0.86 (0.68-1.09) .20 0.78 (0.54-1.14) .20 0.93 (0.69-1.26) .63 COVID-19 ICU load ≤25% 1 [Reference] 1 [Reference] 1 [Reference] >25% to 50% 1.19 (0.99-1.43) 1.30 (0.92-1.84) 1.10 (0.88-1.37) >50% to 75% 1.03 (0.77-1.40) 0.74 (0.42-1.32) 1.15 (0.81-1.64) .01 .04 .049 >75% to 100% 1.63 (1.13-2.35) 1.90 (0.98-3.65) 1.67 (1.08-2.60) >100% 2.03 (1.16-3.56) 1.14 (0.29-4.49) 2.35 (1.25-4.39) COVID-19 ICU demand ≤25% 1 [Reference] 1 [Reference] 1 [Reference] >25% to 50% 1.11 (0.94-1.31) 1.41 (1.07-1.84) 0.99 (0.81-1.22) <.001 .09 <.001 >50% to 75% 1.25 (1.05-1.49) 1.30 (0.95-1.79) 1.19 (0.95-1.48) >75% 1.67 (1.33-2.11) 1.29 (0.85-1.97) 1.94 (1.46-2.59) Hypertension 0.94 (0.80-1.11) .46 0.92 (0.68-1.24) .55 0.94 (0.77-1.14) .51 Atrial fibrillation 1.00 (0.88-1.15) .95 1.01 (0.81-1.27) .93 1.00 (0.84-1.18) .96 History of smoking Never 1 [Reference] 1 [Reference] 1 [Reference] Current 0.70 (0.56-0.87) 0.73 (0.47-1.14) 0.74 (0.57-0.96) <.001 .001 <.001 Former 0.93 (0.81-1.06) 0.98 (0.77-1.26) 0.92 (0.78-1.08) Unknown 1.55 (1.30-1.84) 1.59 (1.19-2.12) 1.59 (1.27-1.98) BMI <25 1 [Reference] 1 [Reference] 1 [Reference] 25-29 0.83 (0.72-0.95) 0.75 (0.59-0.94) 0.88 (0.74-1.05) 30-34 0.80 (0.68-0.94) 0.72 (0.53-0.97) 0.84 (0.69-1.02) .001 .01 .12 ≥35 0.94 (0.79-1.11) 0.96 (0.68-1.36) 0.98 (0.79-1.20) Missing data 1.38 (0.97-1.98) 1.65 (0.92-2.93) 1.39 (0.87-2.22) Place of treatment General ward 1 [Reference] NA NA ICU <.001 NA <.001 No ventilation 1.93 (1.66-2.24) NA 1 [Reference] Mechanical ventilation 7.40 (6.47-8.48) NA 3.93 (3.39-4.55) APACHE score 1.02 (1.02-1.03) <.001 1.04 (1.03-1.05) <.001 1.02 (1.01-1.02) <.001 Charlson comorbidity score 1.05 (1.03-1.06) <.001 1.09 (1.06-1.12) <.001 1.02 (1.00-1.05) .03 Month of COVID diagnosis March 1.46 (1.15-1.84) 1.43 (0.92-2.22) 1.50 (1.13-2.00) April 1.47 (1.20-1.79) 1.48 (1.04-2.10) 1.52 (1.19-1.96) May 1 [Reference] 1 [Reference] 1 [Reference] <.001 .004 .009 June 1.15 (0.91-1.45) 1.11 (0.74-1.68) 1.22 (0.91-1.62) July 1.05 (0.85-1.30) 0.83 (0.57-1.20) 1.23 (0.94-1.61) August 1.23 (0.97-1.55) 0.84 (0.56-1.26) 1.50 (1.13-2.00) (continued) JAMA Network Open. 2021;4(1):e2034266. doi:10.1001/jamanetworkopen.2020.34266 (Reprinted) January 19, 2021 7/11 JAMA Network Open | Public Health Association of Intensive Care Unit Load and Demand With COVID-19 Mortality in VA Hospitals Table 3. Proportional Hazard Results From Admission to 30 Days Postdischarge or Death (continued) Overall General ward only ICU Characteristic Adjusted HR (95% CI) P value Adjusted HR (95% CI) P value Adjusted HR (95% CI) P value Facility complexity 1a 1.05 (0.67-1.66) 1.41 (0.54-3.68) 0.88 (0.52-1.50) 1b 0.94 (0.59-1.51) 1.29 (0.48-3.47) 0.82 (0.47-1.41) .08 .80 .08 1c 1.29 (0.81-2.07) 1.50 (0.56-4.04) 1.17 (0.68-2.02) 2 1 [Reference] 1 [Reference] 1 [Reference] Abbreviations: APACHE, Acute Physiology, Age, Chronic Health Evaluation; BMI, body A measure of medical comorbidity in which increasing values indicate a greater mass index (calculated as weight in kilograms divided by height in meters squared); comorbidity burden. COVID-19, coronavirus disease 2019; HR, hazard ratio; ICU, intensive care unit; NA, not f The complexity of the services provided at Veterans Affairs facilities, with level 1a being applicable. the most complex. The facility complexity model includes ICU level, operative The other race/ethnicity category includes Hispanic or Latino, Asian, Native Hawaiian complexity level, patient clinical classification, teaching status characteristics, amount or other Pacific Islander, American Indian or Alaskan Native, or other race/ethnicity. of research funding, complex clinical programs provided (eg, invasive catheterization b laboratory, neurosurgery, or transplant), rurality, care provided in the community, and Calculated as No. patients with COVID-19 in the ICU during stay divided by No. of mental health programs provided. ICU beds. Calculated as mean No. of patients with COVID-19 in the ICU during stay divided by maximum No. of patients with COVID-19 in the ICU. A physiologic measure in which increasing scores indicate worse health. This score includes laboratory data (eg, white blood cell count) as well as vital signs (eg, oxygenation and blood pressure). demand may be associated with increased mortality; it is imperative that we understand the degree to which patient characteristics (eg, disease severity) or facility issues (eg, staffing) contribute to the association between COVID-19 ICU strain and poor patient outcomes among patients with critical COVID-19. We did not have a formal measure of ICU capacity, because VA ICU bed availability is not fixed but instead depends on staffing availability; therefore, we calculated COVID-19 ICU load as the ratio of ICU COVID-19 occupancy to the maximum ICU bed number as a surrogate for COVID-19 ICU capacity. Although the association between COVID-19 ICU load and patient mortality was statistically significant, it was neither as consistent over time nor as robust as the association between COVID-19 ICU demand and mortality. We hypothesize that facilities increased their critical care capacity in response to the pandemic and that the degree of this augmentation varied across facilities. Therefore, the comparison with a fixed number of patient beds was likely a relatively poor measure of ICU capacity during the pandemic. Given that hospitals are charged with caring for patients with non–COVID-19 critical illness as well as patients with COVID-19, future studies should seek to examine whether measures of critical care strain that include all patients in the ICU (not just those with COVID-19) are associated with patient outcomes. Future studies should also evaluate whether ICU load provides an adequate measure of strain across the broad spectrum of VA and non-VA hospitals, which vary greatly in prepandemic ICU bed number and the potential to augment capacity during a 11 12 pandemic. Our overall study findings are supported by cohort studies from 2013 and 2018 11,12 demonstrating that as ICUs are strained, mortality increases. It may be the case that during periods of peak ICU caseload, patients who would be admitted to the ICU under more typical conditions are instead admitted to the ward. Our data did not allow us to examine this issue directly; however, we did examine outcomes associated with COVID-ICU strain separately among patients in the general ward and patients in the ICU. Although the association between COVID-19 ICU load and general ward mortality was statistically significant, it varied over time (ie, early vs later in the pandemic). Future research should examine how critical care strains may be associated with outcomes in the general ward for patients with COVID-19 and those without COVID-19. JAMA Network Open. 2021;4(1):e2034266. doi:10.1001/jamanetworkopen.2020.34266 (Reprinted) January 19, 2021 8/11 JAMA Network Open | Public Health Association of Intensive Care Unit Load and Demand With COVID-19 Mortality in VA Hospitals Limitations This study has several limitations. First, this study evaluated care of patients with COVID-19 at VA hospitals; future studies should examine the association between COVID-19 ICU burden and mortality in non-VA facilities. Second, this study focused on COVID-19 mortality; future studies should examine the potential associations of COVID-19 ICU load and demand with outcomes among patients without COVID-19. Third, the results of this study should not be interpreted as a statement on scarcity of critical care or mechanical ventilation; we have no data to suggest that patients needing critical care or mechanical ventilation did not receive this care. Fourth, although the risk adjustment models included demographic and clinical characteristics, they did not include social determinants of health (eg, income or education), which may contribute to COVID-19 mortality. Fifth, we did not examine changes in ICU staffing during the study period. Sixth, we do not have a measure of the degree to which facilities expanded ICU capacity during the pandemic. Seventh, patients with COVID-19 who were admitted to the ICU service could have physically been in diverse settings, including locations designated as the COVID-19 ICU, such as surgical ICUs; some patients with critical COVID-19 were cared for by the ICU team but were physically located in the emergency department. Eighth, related to the observed change in mortality over time, our results suggest that changes in patient characteristics and measures of COVID-19 ICU strain were associated with some of the variation in mortality over time; however, given the observational nature of these data, causality cannot be inferred. Other potential causes (eg, use of medications, such as remdesivir and dexamethasone; clinical practices, such as proning; and unmeasured changes in patient characteristics, such as susceptibility) may have contributed to changes in COVID-19 mortality. Conclusions In this cohort study of patients with COVID-19 in US VA hospitals, COVID-19 ICU demand—a measure of COVID-19 ICU caseload when a patient was treated compared with peak COVID-19 ICU caseload—was associated with mortality among patients with COVID-19 in the ICU. Public health officials and hospital administrators may seek to prevent high COVID-19 ICU demand to optimize outcomes for patients with COVID-19. ARTICLE INFORMATION Accepted for Publication: November 28, 2020. Published: January 19, 2021. doi:10.1001/jamanetworkopen.2020.34266 Open Access: This is an open access article distributed under the terms of the CC-BY License.©2021BravataDM et al. JAMA Network Open. Corresponding Author: Dawn M. Bravata, MD, Health Services Research and Development Center for Health Information and Communication, Richard L. Roudebush VA Medical Center, 1481 W 10th St, HSR&D Mail Code 11H, Indianapolis, IN 46202 (Dawn.Bravata2@va.gov). Author Affiliations: Precision Monitoring to Transform Care Quality Enhancement Research Initiative, Health Services Research and Development, Department of Veterans Affairs, Indianapolis, Indiana (Bravata, Perkins, L. J. Myers, Arling, Zhang, J. Myers, Austin, Sexson, Keyhani); Health Services Research and Development Center for Health Information and Communication, Richard L. Roudebush VA Medical Center, Department of Veterans Affairs, Indianapolis, Indiana (Bravata, L. J. Myers, J. Myers, Austin, Sexson, Dev); Medicine Service, Richard L. Roudebush VA Medical Center, Indianapolis, Indiana (Bravata, Reese, Dysangco, Agarwal, Dev); Department of Medicine, Indiana University School of Medicine, Indianapolis (Bravata, L. J. Myers, Reese, Dysangco, Agarwal); Department of Neurology, Indiana University School of Medicine, Indianapolis (Bravata); William M. Tierney Center for Health Services Research, Regenstrief Institute, Indianapolis, Indiana (Bravata, L. J. Myers, Dev); Department of Biostatistics, Indiana University School of Medicine, Indianapolis (Perkins); School of Nursing, Purdue University, West Lafayette, Indiana (Arling); Department of Biostatistics, College of Public Health, University of Nebraska Medical Center, Omaha (Zhang); Department of Pharmacy Practice, College of Pharmacy, Purdue University, West Lafayette, Indiana (Zillich); Northern California Institute for Research and Education, San Francisco (Leonard); San JAMA Network Open. 2021;4(1):e2034266. doi:10.1001/jamanetworkopen.2020.34266 (Reprinted) January 19, 2021 9/11 JAMA Network Open | Public Health Association of Intensive Care Unit Load and Demand With COVID-19 Mortality in VA Hospitals Francisco VA Medical Center, San Francisco, California (Keyhani); Division of General Internal Medicine, Department of Medicine, University of California, San Francisco (Keyhani). Author Contributions: Dr Bravata and Mr Perkins had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Concept and design: Bravata, Arling, Zhang, Zillich, Keyhani. Acquisition, analysis, or interpretation of data: Bravata, Perkins, L. Myers, Arling, Zhang, Zillich, Reese, Dysangco, Agarwal, J. Myers, Austin, Sexson, Leonard, Dev. Drafting of the manuscript: Bravata, Arling. Critical revision of the manuscript for important intellectual content: Bravata, Perkins, L. Myers, Zhang, Zillich, Reese, Dysangco, Agarwal, J. Myers, Austin, Sexson, Leonard, Dev, Keyhani. Statistical analysis: Bravata, Perkins, Arling, Zhang. Obtained funding: Bravata. Administrative, technical, or material support: L. Myers, Reese, J. Myers, Austin, Sexson, Keyhani. Supervision: Bravata, Zhang. Conflict of Interest Disclosures: Drs Bravata and Perkins reported receiving grants from the Department of Veterans Affairs during the conduct of the study. Dr Agarwal reported receiving personal fees and travel support from Bayer, Relypsa, Reata, Sanofi, Boehringer, and Merck; personal fees from Janssen, DiaMedica, Lexicon, Akebia, Eli Lilly, and Astra Zeneca; and travel support from Akebia outside the submitted work. No other disclosures were reported. Funding/Support: This work was supported by the US Department of Veterans Affairs Health Services Research and Development Service Precision Monitoring to Transform Care Quality Enhancement Research Initiative (QUE 15-280). Role of the Funder/Sponsor: The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. Disclaimer: The contents of this study do not represent the views of the US Department of Veterans Affairs or the US government. REFERENCES 1. Pan A, Liu L, Wang C, et al. Association of public health interventions with the epidemiology of the COVID-19 outbreak in Wuhan, China. JAMA. 2020;323(19):1915-1923. doi:10.1001/jama.2020.6130 2. Hartley DM, Perencevich EN. Public health interventions for COVID-19: emerging evidence and implications for an evolving public health crisis. JAMA. 2020;323(19):1908-1909. doi:10.1001/jama.2020.5910 3. Eriksson CO, Stoner RC, Eden KB, Newgard CD, Guise JM. The association between hospital capacity strain and inpatient outcomes in highly developed countries: a systematic review. J Gen Intern Med. 2017;32(6):686-696. doi:10.1007/s11606-016-3936-3 4. VA Health Services Research and Development. VA Informatics and Computing Infrastructure (VINCI). Accessed December 10, 2020. https://www.hsrd.research.va.gov/for_researchers/vinci/default.cfmhttps://vaww. vinci.med.va.gov/VinciCentral/Home/About 5. Borzecki AM, Wong AT, Hickey EC, Ash AS, Berlowitz DR. Can we use automated data to assess quality of hypertension care? Am J Manag Care. 2004;10(7 Pt 2):473-479. 6. Sohn MW, Arnold N, Maynard C, Hynes DM. Accuracy and completeness of mortality data in the Department of Veterans Affairs. Popul Health Metr. 2006;4:2. doi:10.1186/1478-7954-4-2 7. Knaus WA, Wagner DP, Draper EA, et al. The APACHE III prognostic system: risk prediction of hospital mortality for critically ill hospitalized adults. Chest. 1991;100(6):1619-1636. doi:10.1378/chest.100.6.1619 8. Kim JH, Hong SK, Kim Y, et al. Experience of augmenting critical care capacity in Daegu during COVID-19 incident in South Korea. Acute Crit Care. 2020;35(2):110-114. doi:10.4266/acc.2020.00275 9. Einav S, Hick JL, Hanfling D, et al; Task Force for Mass Critical Care. Surge capacity logistics: care of the critically ill and injured during pandemics and disasters: CHEST consensus statement. Chest. 2014;146(4)(suppl): e17S-e43S. doi:10.1378/chest.14-0734 10. Ma X, Vervoort D. Critical care capacity during the COVID-19 pandemic: global availability of intensive care beds. J Crit Care. 2020;58:96-97. doi:10.1016/j.jcrc.2020.04.012 11. Gabler NB, Ratcliffe SJ, Wagner J, et al. Mortality among patients admitted to strained intensive care units. Am J Respir Crit Care Med. 2013;188(7):800-806. doi:10.1164/rccm.201304-0622OC JAMA Network Open. 2021;4(1):e2034266. doi:10.1001/jamanetworkopen.2020.34266 (Reprinted) January 19, 2021 10/11 JAMA Network Open | Public Health Association of Intensive Care Unit Load and Demand With COVID-19 Mortality in VA Hospitals 12. Anesi GL, Liu VX, Gabler NB, et al. Associations of intensive care unit capacity strain with disposition and outcomes of patients with sepsis presenting to the emergency department. Ann Am Thorac Soc. 2018;15(11): 1328-1335. doi:10.1513/AnnalsATS.201804-241OC 13. Feinstein AR, Sosin DM, Wells CK. The Will Rogers phenomenon: stage migration and new diagnostic techniques as a source of misleading statistics for survival in cancer. N Engl J Med. 1985;312(25):1604-1608. doi: 10.1056/NEJM198506203122504 14. Yang X, Yu Y, Xu J, et al. Clinical course and outcomes of critically ill patients with SARS-CoV-2 pneumonia in Wuhan, China: a single-centered, retrospective, observational study. Lancet Respir Med. 2020;8(5):475-481. doi: 10.1016/S2213-2600(20)30079-5 15. Horby P, Lim WS, Emberson JR, et al; RECOVERY Collaborative Group. Dexamethasone in hospitalized patients with Covid-19 — preliminary report. N Engl J Med.2020. SUPPLEMENT. eTable. Proportional Hazard Results for COVID-19 ICU Strain: Time From Admission to 30 Days Postdischarge or Death by Temporal Period JAMA Network Open. 2021;4(1):e2034266. doi:10.1001/jamanetworkopen.2020.34266 (Reprinted) January 19, 2021 11/11 Supplemental Online Content Bravata DM, Perkins AJ, Myers LJ, et al. Association of intensive care unit patient load and demand with mortality rates in US Department of Veterans Affairs hospitals during the COVID-19 pandemic. JAMA Netw Open. 2021;4(1):e2034266. doi: 10.1001/jamanetworkopen.2020.34266 eTable. Proportional Hazard Results for COVID-19 ICU Strain: Time From Admission to 30 Days Postdischarge or Death by Temporal Period This supplemental material has been provided by the authors to give readers additional information about their work. © 2021 Bravata DM et al. JAMA Network Open. eTable. Proportional Hazard Results for COVID-19 ICU Strain: Time From Admission to 30 Days Postdischarge or Death by Temporal Period Baseline Characteristic Overall General Ward Only Intensive Care Unit (ICU) HR (95% CI) P-value HR (95% CI) P-value HR (95% CI) P-value March-May 2020* COVID-19 ICU-Load: Mean number COVID-19 ICU patients during stay/Number of 0.049 0.010 0.243 ICU Beds 1.00 1.00 1.00 COVID-19 ICU-Demand: Mean number COVID-19 ICU patients during <0.001 0.264 <0.001 stay/Maximum number of COVID patients 1.00 1.00 1.00 1.12 June-August 2020* COVID-19 ICU-Load: Mean number COVID-19 ICU patients during stay/Number of ICU Beds 1.00 1.00 1.00 -- -- -- -- -- -- COVID-19 ICU-Demand: Mean number COVID-19 ICU patients during 0.016 0.023 stay/Maximum number of COVID patients 1.00 1.00 1.00 *The temporal period refers to the date of the first COVID-19 laboratory test per patient. © 2021 Bravata DM et al. JAMA Network Open.

Journal

JAMA Network Open – American Medical Association

Published: Jan 19, 2021

There are no references for this article.