https://orcid.org/0000-0002-8371-1507
Figures
Abstract
Background
The volume-outcome relationship in patients with severe Coronavirus disease 2019 (COVID-19) is unclear and is important for establishing a system for the medical care of severe COVID-19. This study aimed to evaluate the association between institutional case volume and outcomes in patients with ventilated COVID-19.
Methods
We analyzed patients with severe COVID-19 on ventilatory control aged > 17 years who were enrolled in the J-RECOVER study, which is a retrospective multicenter observational study conducted between January 2020 and September 2020 in Japan. Based on the ventilated COVID-19 case volume, the higher one-third of institutions were defined as high-volume centers, the middle one-third as middle-volume centers, and the lower one-third as low-volume centers. The primary outcome measure was in-hospital mortality during hospitalization due to COVID-19. Multivariate logistic regression analysis for in-hospital mortality and ventilated COVID-19 case volume was performed after adjusting for multiple propensity scores and in-hospital variables. To estimate the multiple propensity score, we fitted a multinomial logistic regression model, which fell into one of the three groups based on patient demographics and prehospital factors.
Results
We analyzed 561 patients who required ventilator management. In total, 159, 210, and 192 patients were admitted to low-volume (36 institutions, < 11 severe COVID-19 cases per institution during the study period), middle-volume (14 institutions, 11–25 severe cases per institution), and high-volume (5 institutions, > 25 severe cases per institution) centers, respectively. After adjustment for multiple propensity scores and in-hospital variables, admission to middle- and high-volume centers was not significantly associated with in-hospital death compared with admission to low-volume centers (adjusted odds ratio, 0.77 [95% confidence interval (CI): 0.46–1.29] and adjusted odds ratio, 0.76 [95% CI: 0.44–1.33], respectively).
Citation: Amagasa S, Uematsu S, Kubota M, Kashiura M, Yasuda H, Hayakawa M, et al. (2023) Relationship between institutional ventilated COVID-19 case volume and in-hospital death: A multicenter cohort study. PLoS ONE 18(6): e0287310. https://doi.org/10.1371/journal.pone.0287310
Editor: Chiara Lazzeri, Azienda Ospedaliero Universitaria Careggi, ITALY
Received: May 5, 2022; Accepted: June 3, 2023; Published: June 15, 2023
Copyright: © 2023 Amagasa et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The datasets analyzed during our study are not publicly available due to restrictions imposed by our Institutional Review Boards, which limited the use of the data to participating institutions only. However, study data can be obtained from the steering committee of J-RECOVER study upon a reasonable request. Contact e-mail: 2020covidstudy@gmail.com."
Funding: The authors received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Coronavirus disease 2019 (COVID-19) was first reported as an outbreak of pneumonia in Wuhan, a city in Hubei Province, China, at the end of 2019, and severe acute respiratory syndrome coronavirus 2 (SARS-CoV2) was confirmed to be the cause [1,2]. Subsequently, COVID-19 caused a global pandemic in March 2020 [3,4]. Since then, the pandemic has not abated, with more than 491 million people infected worldwide, and currently, more than 6.1 million deaths have been reported (as of April 4, 2022). COVID-19 causes many serious injuries, and 5%-20% of infected patients become severely ill and require treatment in intensive care units (ICU) [1,4]. Mortality rates for critically ill patients with COVID-19 are between 15% and 61% despite treatment in the ICU [5–8].
High-quality, specialized medical approaches are required for patients experiencing a critical illness or severe trauma and those requiring certain procedures, and volume-outcome relationships in these patients have been previously reported [9,10]. For various surgeries and conditions, including critical care, mechanical ventilation, pancreatic cancer, esophageal cancer, abdominal aortic aneurysm, and coronary artery bypass surgery, the larger the hospital case volume, the better the prognosis. Few studies have examined the volume-outcome relationship in critically ill patients with COVID-19, and the effect of case volume alone on outcomes in these cases remains unclear. In many of the regions that experienced the rapid expansion of COVID-19 earlier, the burden of disease was not uniform across regions due to restricted healthcare systems and limited resources [11]. In 2020, resources for intensive care treatment of severe COVID-19 were maintained in Japan [12]. A better understanding of the volume-outcome relationship in patients with severe COVID-19 in less healthcare-pressured areas could help create an efficient healthcare system and improve care and outcomes for patients with severe COVID-19.
To evaluate the volume-outcome relationship in patients with severe COVID-19 in areas where resources for critical care are maintained, we investigated the association between institutional case volume and outcome in ventilated critically ill patients with COVID-19.
Materials and methods
Study design and setting
The J-RECOVER study was a retrospective multicenter observational study conducted between January and September 2020 in Japan. The design, data collection methods, and protocols for the J-RECOVER study have been reported in detail previously [13]. This study was approved by the institutional ethics committee of each participating institution, with a waiver for informed consent to protect participant anonymity.
Patients
All patients with COVID-19 admitted to the participating institutions during the study period were included in the J-RECOVER study. The current study included patients with severe COVID-19 on ventilatory control aged >17 years who were enrolled in the J-RECOVER study.
Data collection
The existing clinical information of patients was obtained from diagnosis procedure combination (DPC) data and medical records. The DPC system is a comprehensive evaluation system for medical fees for acute inpatient care. Data were claimed to have been created and stored electronically at each facility under a comprehensive payment system based on DPC [13,14]. DPC data included sex, birth date, the main purpose of care during hospitalization, admission date, discharge date, patient transfer, route of admission, hospital referral or outpatient department admission, scheduled or emergency care admission, ambulance transport, discharge destination, and discharge outcome. Other necessary information that could not be obtained from the DPC data was obtained from the medical records of the institution and its personnel.
Measurements and definitions
The primary outcome measurement was in-hospital death during hospitalization with COVID-19.
Hospital volume was defined as the number of eligible patients treated at each hospital, and it was subcategorized into tertials, as per previous studies [15–17]. The cutoff value was defined to determine the number of patients as equally as possible. The higher one-third of institutions were defined as high-volume centers, the middle one-third as middle-volume centers, and the lower one-third as low-volume centers. The Charlson comorbidity index was used as a measure of underlying disease [18]. The Sequential Organ Failure Assessment (SOFA) score was used as the severity score [19].
Statistical analysis
We compared demographic factors, patient characteristics, hospital care, and outcomes among patients treated in high-, middle-, and low-volume centers. Trends in discrete variables among the three groups were analyzed using the Mantel–Haenszel trend test, and continuous variables were analyzed using the Kruskal–Wallis test.
Missing data were accounted for by undertaking multiple imputations to minimize bias and maximize the power of analyses [20]. Twenty multiple imputed datasets were generated, and the association between the severe COVID-19 case volume and in-hospital death in each dataset was evaluated using multivariate logistic regression analysis. The estimates were then combined.
We used multiple propensity score analysis in the multivariate analysis to adjust for confounding factors before admission that could be measured [21,22]. A propensity score represents the conditional probability of a particular exposure given a set of measured baseline covariates and is an established method for reducing confounding factors in observational studies. Although propensity score matching or stratification has commonly been used to assemble similar patient cohorts in terms of baseline covariates for 2-group comparisons, a multiple propensity score approach has been used for comparisons of three or more groups [22–24]. A multiple propensity score is defined as the conditional probability of being categorized into a particular group of three or more given a set of observed baseline covariates. To estimate the multiple propensity score for the current analysis, we fitted a multinomial logistic regression model, categorized into one of the three groups along with hospital characteristics, including teaching status, number of hospital beds, number of ICU beds, number of annual ICU admissions (2019), patient demographics (age, sex, and underlying disease), and prehospital factors (number of days from symptom onset to hospitalization, number of days from positive polymerase chain reaction test to hospitalization, transfer from another hospital, and severity of illness on admission). Next, we used multivariate logistic regression to analyze the association between institutional severe COVID-19 case volume and in-hospital mortality with adjustment for multiple propensity scores and in-hospital variables, including favipiravir, remdesivir, ICU admission, use of prone position, and use of extracorporeal membrane oxygenation (ECMO).
We also performed a sensitivity analysis using data from cases where all necessary values were recorded for robustness. This sensitivity analysis was performed with multivariate logistic regression analysis of institutional volume and in-hospital mortality adjusted for multiple propensity scores calculated from preadmission confounders and in-hospital variables as in the main analysis.
All statistical tests were two-sided, and a level of 0.05 was considered statistically significant. All data analyses were performed using SPSS statistical software package version 24 (IBM, New York, USA).
Results
Patients
In total, 4,700 patients with COVID-19 from 66 institutions were enrolled in the J-RECOVER study. Among these patients, 561 were adults with COVID-19 who required ventilator management (Fig 1). None of the patients met the exclusion criteria. A total of 159, 210, and 192 patients were admitted to low-volume (36 institutions, < 11 severe COVID-19 cases per institution during the study period), middle-volume (14 institutions, 11–25 severe cases per institution), and high-volume (5 institutions, >25 severe cases per institution) centers, respectively. There were no ventilated patients in 11 institutions.
*During the study period (9 months).
Institutional information and patient characteristics
There were no significant differences in the number of academic hospitals, hospital beds, ICU beds, or ICU admissions in 2019 among the three groups (Tables 1 and 2). The number of ECMO patients per institution in low-volume centers was significantly lower than that in the other two groups.
The age of the patients in the high-volume center was significantly higher than in the other two groups (< 0.001). Patients in high-volume centers had a higher probability of transfer from another hospital (p < 0.001), favipiravir administration (p = 0.031), ICU admission (p = 0.01), and a lower probability of ECMO (p < 0.001). Patients in low-volume centers had a lower probability of a prone position (p = 0.002).
The difference in outcomes among institutions
The length of stay in the hospital and ICU was significantly shorter as the case volume increased. In-hospital death was not significantly different among the three groups (Table 3).
Multivariate logistic regression analyses for in-hospital death adjusted by multiple propensity score and in-hospital variables
After adjustment for multiple propensity scores and in-hospital variables, admission to middle- and high-volume centers was not significantly associated with in-hospital mortality compared with admission to low-volume centers (adjusted odds ratio, 0.77 [95% confidence interval (CI): 0.46–1.29] and adjusted odds ratio, 0.76 [95% CI: 0.44–1.33], respectively) (Table 4).
A multiple propensity score was defined as the conditional probability of falling into a particular group of three groups of institutional case volume. We fitted a multinomial logistic regression model, which fell into one of the three groups as patient demographics and prehospital factors.
Sensitivity analysis of multivariate logistic regression analyses for in-hospital death adjusted by multiple propensity score and in-hospital variables using data of the cases with complete data
In the sensitivity analysis, admission to middle- and high-volume centers was also not significantly associated with in-hospital mortality compared with admission to low-volume centers (adjusted odds ratio, 0.76 [95% CI: 0.33–1.72] and adjusted odds ratio, 0.61 [95% CI: 0.29–1.30], respectively) (Table 5).
Discussion
The present study evaluated the volume-outcome relationship in critically ill patients with ventilated COVID-19. In Japan, we did not observe a significant association between institutional case volume and in-hospital mortality after adjusting for in-hospital confounding factors and multiple propensity scores (calculated based on patient characteristics and prehospital confounding factors). The findings from this study may help in the construction of a medical system for patients with severe COVID-19.
There are many observational studies on volume-outcome relationships in critical care, and the relationship remains a matter of debate [16,17,25]. One systematic review of the volume-outcome relationship in critical care reported that high volumes might improve survival outcomes, although there is no complete consistency in this relationship [10]. Regarding the volume of mechanical ventilation, several studies showed that high volume improved outcomes, while others found no such association. Two studies with data before 2003 [26,27], one with an instrumental variable analysis conducted between 2004 and 2006 [28], and a more recent study conducted with patients with acute respiratory distress syndrome [15] reported that a higher case volume of mechanical ventilation improved outcomes. No association was found between volume outcomes in ventilated patients in two studies using data from 2008 and 2009 [29,30]. In contrast, a recent large retrospective cohort study showed that patient outcomes were worse at hospitals with higher mechanical ventilation volumes [31]. These differences in results may be due to the influence of changes in mechanical ventilatory management over time, differences in health care systems, types of hospitals, statistical methods, or adjusting for confounding factors. Furthermore, unlike previous studies that examined the volume outcome of other specific diseases, procedures, or surgeries, previous studies on critical care had a heterogeneous population with a wide range of patient diseases and treatments. This heterogeneity may have greatly influenced the discrepancy in the findings of the previous studies.
In the present study, we observed that the institutional ventilated COVID-19 case volume was not associated with in-hospital mortality. This study is a relatively small number of patients and institutions for a study examining volume-outcome relationships. Issues such as low variability in the number of patients among participating institutions and the small number of institutions included in the high-volume centers may be related to the results. However, our study examined a single disease, making our study less heterogeneous than previous studies of volume outcomes in critical care. Moreover, multiple imputations to address missing data and multiple propensity score analysis was undertaken, allowing adjustment for many confounding factors, including hospital and patient characteristics, prehospital factors, and treatment. Age, which is an important factor with respect to mortality and is skewed by group, was also adjusted for. We did not observe a volume-outcome relationship after multiple propensity score analysis, adjusting for these factors. To the best of our knowledge, there are no publications on the volume-outcome relationship for COVID-19, and we believe this study is important for examining the same.
There are several possible reasons for the lack of a significant association between the institutional volume of ventilated COVID-19 cases and outcomes. In the present study, many participating hospitals were tertiary emergency centers, and both high- and low-volume centers were habitually proficient in ventilator management and could provide higher-level care, such as ECMO, if necessary. In other words, there may be equalization of care with respect to the participating institutions in this study. It is also possible that even high-volume centers may not be able to improve outcomes because COVID-19 is an emerging infection, and care has not yet been established. However, the mortality rate was lower than that in other countries, indicating a superior quality of care. In the future, it will be necessary to observe whether the development of COVID-19 treatment affects the volume outcome. The higher-volume facilities might not have enough resources due to the sudden COVID-19; pandemic. However, no shortage of resources for intensive care in Japan during the pandemic was observed. Our data suggest that institutional ventilated COVID-19 case volume was not associated with survival outcomes in an ICU with adequate resources.
The group with a higher institutional case volume had significantly shorter ICU stays and hospitalizations. It was thought that a large number of transfers to other departments and discharges due to transfers to other hospitals might be associated with shorter ICU and hospital stays. However, transfers to another department did not differ among the three groups, and discharges due to transfer to other hospitals tended to be higher in high-volume centers; nevertheless, this was not significant. Shorter ICU and hospital stays may also be related to practice. The high-volume center could treat more patients with severe COVID-19 due to shorter ICU and hospital stays, which is important in terms of the effective utilization of resources.
This study had several limitations. First, the J-RECOVER study was conducted only at certain hospitals and was not a population-based study; therefore, the generalizability of the results may be limited. Second, although understanding volume-outcome relationships requires sufficient hospital numbers to create stable comparisons, the number of eligible institutions and cases was relatively low. The lack of a large difference in the number of patients between high-volume and low-volume institutions and the number of only five high-volume institutions may be related to the result that the volume-outcome relationship was not found. Because this study targeted a single disease and was not highly heterogeneous, the results are considered meaningful; however, further large-scale studies are needed. Third, this study included patients in the early stages of the COVID-19 pandemic as an emerging infectious disease. The results may change in the future as effective treatment and care becomes more established. The volume-outcome relationship of COVID-19 as an emerging infectious disease needs to be examined over time, considering changes in treatment methods and protocols. Fourth, this is a retrospective study, and caution should be exercised in its interpretation because of missing values, unmeasured confounding, and other issues. Future prospective studies on the volume-outcome relationship of emerging infectious diseases are needed.
Conclusions
In the present study, institutional ventilated COVID-19 case volume was not associated with in-hospital death after adjusting for patient demographics and pre-hospital and in-hospital factors. The results of this study may aid in the construction of a medical system for patients with severe COVID-19. However, further investigation is needed to determine the association between institutional case volumes and outcomes in patients with severe COVID-19.
References
- 1. Guan WJ, Ni ZY, Hu Y, Ou CQ, He JX, Liu L, et al. Clinical Characteristics of Coronavirus Disease 2019 in China. N Engl J Med. 2020; 382: 1708–1720. pmid:32109013
- 2. Zhou F, Yu T, Du R, Fan G, Liu Y, Liu Z, et al. Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study. The Lancet. 2020; 395: 1054–1062. pmid:32171076
- 3. Grasselli G, Zangillo A, Zanella A, Antonelli M, Cabrini L, Castelli A, et al. Baseline Characteristics and Outcomes of 1591 Patients Infected With SARS-CoV-2 Admitted to ICUs of the Lombardy Region, Italy. JAMA. 2020; 323: 1574–1581. pmid:32250385
- 4. Richardson S, Hirsch JS, Narasimhan M, Crawford JM, McGinn T, Davidson KW, et al. Presenting Characteristics, Comorbidities, and Outcomes Among 5700 Patients Hospitalized With COVID-19 in the New York City Area. JAMA. 2020; 323: 2052–2059. pmid:32320003
- 5. Mitra AR, Fergusson NA, Lloyd-Smith E, Wormsbecker A, Foster D, Karpov A, et al. Baseline characteristics and outcomes of patients with COVID-19 admitted to intensive care units in Vancouver, Canada: a case series. CMAJ. 2020; 192: E694–701. pmid:32461326
- 6. Cummings MJ, Baldwin MR, Abrams D, Jacobson SD, Meyer BJ, Balough EM, et al. Epidemiology, clinical course, and outcomes of critically ill adults with COVID19 in New York city: A prospective cohort study. The Lancet. 2020; 395: 1763–1770.
- 7. Grasselli G, Greco M, Zanella A, Albano G, Antonelli M, Bellani G, et al. Risk factors associated with mortality among patients with COVID-19 in intensive care units in Lombardy, Italy. JAMA Intern Med. 2020; 180: 1–11. pmid:32667669
- 8. Yang X, Yu Y, Xu J, Shu H, Xia J, Liu H, 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: 475–481. pmid:32105632
- 9. Halm EA, Lee C, Chassin MR. Is volume related to outcome in health care? A systematic review and methodologic critique of the literature. Ann Intern Med. 2002; 137: 511–520. pmid:12230353
- 10. Nguyen YL, Wallace DJ, Yordanov Y, Trinquart L, Blomkvist J, Angus DC, et al. The volume-outcome relationship in critical care: a systematic review and meta-analysis. Chest. 2015; 148: 79–92. pmid:25927593
- 11. Lim ZJ, Subramaniam A, Ponnapa Reddy M, Blecher G, Kadam U, Afroz A, et al. Case Fatality Rates for Patients with COVID-19 Requiring Invasive Mechanical Ventilation. A Meta-analysis. Am J Respir Crit Care Med. 2021; 203: 54–66. pmid:33119402
- 12. Japan ECMOnet for COVID-19, Shime N. Save the ICU and save lives during the COVID-19 pandemic. J Intensive Care. 2020; 8: 40. pmid:32549989
- 13. Tagami T, Yamakawa K, Endo A, Hayakawa M, Ogura T, Hirayama A, et al. Japanese multicenter research of COVID-19 by assembling real-world data: A study protocol. Annals of Clinical Epidemiology (In press). https://doi.org/10.37737/ace.22012.
- 14. Yasunaga H. Real World Data in Japan: Chapter II The Diagnosis Procedure Combination Database. Annals of Clinical Epidemiology. 2019; 1: 76–79.
- 15. Ike JD, Kempker JA, Kramer MR, Martin GS. The Association Between Acute Respiratory Distress Syndrome Hospital Case Volume and Mortality in a U.S. Cohort, 2002–2011. Crit Care Med. 2018; 46: 764–773. pmid:29394181
- 16. Sasabuchi Y, Yasunaga H, Matsui H, Lefor AK, Horiguchi H, Fushimi K, et al. The Volume-Outcome Relationship in Critically Ill Patients in Relation to the ICU-to-Hospital Bed Ratio. Crit Care Med. 2015; 43: 1239–1245. pmid:25756414
- 17. Kluge GH, Brinkman S, van Berkel G, van der Hoeven J, Jacobs C, Snel YE, et al. The association between ICU level of care and mortality in the Netherlands. Intensive Care Med. 2015; 41: 304–311. pmid:25600188
- 18. Quan H, Sundararajan V, Halfon P, Fong A, Burnand B, Luthi JC, et al. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care. 2005; 43: 1130–1139. pmid:16224307
- 19. Vincent JL, Moreno R, Takala J, Willatts S, De Mendonça A, Bruining H, et al. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine. Intensive Care Med. 1996; 22: 707–710. pmid:8844239
- 20. Sterne JA, White IR, Carlin JB, Spratt M, Royston P, Kenward MG, et al. Multiple imputation for missing data in epidemiological and clinical research: potential and pitfalls. BMJ. 2009; 338: b2393. pmid:19564179
- 21. Feng P, Zhou XH, Zou QM, Fan MY, Li XS. Generalized propensity score for estimating the average treatment effect of multiple treatments. Stat Med. 2012; 31: 681–697. pmid:21351291
- 22. Spreeuwenberg MD, Bartak A, Croon MA, Hagenaars JA, Busschbach JJ, Andrea H, et al. The multiple propensity score as control for bias in the comparison of more than two treatment arm: an introduction from a case study in mental health. Med Care. 2010; 48: 166–174.
- 23. Tagami T, Matsui H, Fushimi K, Yasunaga H. Changes in Therapeutic Hypothermia and Coronary Intervention Provision and In-Hospital Mortality of Patients With Out-of-Hospital Cardiac Arrest: A Nationwide Database Study. Crit Care Med. 2016; 44: 488–495. pmid:26496447
- 24. Amagasa S, Kashiura M, Moriya T, Uematsu S, Shimizu N, Sakurai A, et al. Relationship between institutional case volume and one-month survival among cases of paediatric out-of-hospital cardiac arrest. Resuscitation. 2019; 137: 161–167. pmid:30802557
- 25. Lapichino G, Gattinoni L, Radrizzani D, Simini B, Bertolini G, Ferla L, et al. Volume of activity and occupancy rate in intensive care units. Association with mortality. Intensive Care Med. 2004; 30: 209–297. pmid:14685662
- 26. Kahn JM, Goss CH, Heagerty PJ, Kramer AA, O’Brien CR, Rubenfeld GD. Hospital volume and the outcomes of mechanical ventilation. N Engl J Med. 2006; 355: 41–50. pmid:16822995
- 27. Needham DM, Bronskill SE, Rothwell DM, Sibbald WJ, Pronovost PJ, Laupacis A, et al. Hospital volume and mortality for mechanical ventilation of medical and surgical patients: A population-based analysis using administrative data. Crit Care Med. 2006; 34: 2349–2354. pmid:16878036
- 28. Kahn JM, Ten Have TR, Iwashyna TJ. The relationship between hospital volume and mortality in mechanical ventilation: An instrumental variable analysis. Health Serv Res. 2009; 44: 862–879. pmid:19674428
- 29. Fernández R, Altaba S, Cabre L, Lacueva V, Santos A, Solsona JF, et al. Relationship between volume and survival in closed intensive care units is weak and apparent only in mechanically ventilated patients. Anesthesiology. 2013; 119: 871–879. pmid:23838712
- 30. Cooke CR, Kennedy EH, Wiitala WL, Almenoff PL, Sales AE, Iwashyna TJ. Despite variation in volume, Veterans Affairs hospitals show consistent outcomes among patients with non-postoperative mechanical ventilation. Crit Care Med. 2012; 40: 2569–2575. pmid:22732289
- 31. Mehta AB, Walkey AJ, Curran-Everett D, Matlock D, Douglas IS. Hospital Mechanical Ventilation Volume and Patient Outcomes: Too Much of a Good Thing? Crit Care Med. 2019; 47: 360–368. pmid:30585831