key: cord-1032965-7ey0ri09
authors: Elke, Berger; Juliane, Winkelmann; Helene, Eckhardt; Ulrike, Nimptsch; Dimitra, Panteli; Christoph, Reichebner; Tanja, Rombey; Reinhard, Busse
title: A country-level analysis comparing hospital capacity and utilisation during the first COVID-19 wave across Europe
date: 2021-12-01
journal: Health Policy
DOI: 10.1016/j.healthpol.2021.11.009
sha: 1420734b0c4ce71967b97060685e64a85fe5799d
doc_id: 1032965
cord_uid: 7ey0ri09

BACKGROUND: : The exponential increase in SARS-CoV-2 infections during the first wave of the pandemic created an extraordinary overload and demand on hospitals, especially intensive care units (ICUs), across Europe. European countries have implemented different measures to address the surge ICU capacity, but little is known about the extent. The aim of this paper is to compare the rates of hospitalised COVID-19 patients in acute and ICU care and the levels of national surge capacity for intensive care beds across 16 European countries and Lombardy region during the first wave of the pandemic (28 February to 31 July). METHODS: : For this country level analysis, we used data on SARS-CoV-2 cases, current and/or cumulative hospitalised COVID-19 patients and current and/or cumulative COVID-19 patients in ICU care. To analyse whether capacities were exceeded, we also retrieved information on the numbers of hospital beds, and on (surge) capacity of ICU beds during the first wave of the COVID-19 pandemic from the COVID-19 Health System Response Monitor (HSRM). Treatment days and mean length of hospital stay were calculated to assess hospital utilisation. RESULTS: : Hospital and ICU capacity varied widely across countries. Our results show that utilisation of acute care bed capacity by patients with COVID-19 did not exceed 38.3% in any studied country. However, the Netherlands, Sweden, and Lombardy would not have been able to treat all patients with COVID-19 requiring intensive care during the first wave without an ICU surge capacity. Indicators of hospital utilisation were not consistently related to the number of SARS-CoV-2 infections. The mean number of hospital days associated with one SARS-CoV-2 case ranged from 1∙3 (Norway) to 11∙8 (France). CONCLUSION: : In many countries, the increase in ICU capacity was important to accommodate the high demand for intensive care during the first COVID-19 wave.

The exponential increase in SARS-CoV-2 infections during the first wave of the pandemic created an extraordinary overload and demand on hospitals, especially intensive care units (ICUs), across Europe. European countries have implemented different measures to address the surge ICU capacity, but little is known about the extent. The aim of this paper is to compare the rates of hospitalised COVID-19 patients in acute and ICU care and the levels of national surge capacity for intensive care beds across 16 European countries and Lombardy region during the first wave of the pandemic (28 February to 31 July).

Methods For this country level analysis, we used data on SARS-CoV-2 cases, current and/or cumulative hospitalised COVID-19 patients and current and/or cumulative COVID-19 patients in ICU care. To analyse whether capacities were exceeded, we also retrieved information on the numbers of hospital beds, and on (surge) capacity of ICU beds during the first wave of the COVID-19 pandemic from the COVID-19 Health System Response Monitor (HSRM). Treatment days and mean length of hospital stay were calculated to assess hospital utilisation.

Hospital and ICU capacity varied widely across countries. Our results show that utilisation of acute care bed capacity by patients with COVID-19 did not exceed 38.3% in any studied country.

However, the Netherlands, Sweden, and Lombardy would not have been able to treat all patients with COVID-19 requiring intensive care during the first wave without an ICU surge capacity. Indicators of hospital utilisation were not consistently related to the number of SARS-CoV-2 infections. The mean number of hospital days associated with one SARS-CoV-2 case ranged from 1•3 (Norway) to 11•8 (France) .

Conclusion In many countries, the increase in ICU capacity was important to accommodate the high demand for intensive care during the first COVID-19 wave. patients to other hospitals across the country or even neighbouring countries to free up capacity [3] .

Throughout Europe, hospitals were required to postpone elective treatments to free up hospital beds and to add beds to ICUs equipped with ventilators while maintaining essential services such as urgent consultations, necessary treatments (e.g., chemotherapy, dialysis), maternal services, and rehabilitation [4] [5] .

During the second wave starting in autumn 2020, many countries again faced increasing hospital admission rates and substantial pressure on hospital wards.

While recognising that many health systems have at least partially been overburdened during the first wave, little is known about the extent of hospital surge capacity created in acute and intensive care units as a response to COVID-19 to accommodate the spike in the number of patients and the actual use of hospital capacities across European countries. However, information on capacities, both in terms of initial and reserve surge capacities, as well as resource utilisation by patients with COVID-19

is the key to inform pandemic preparedness and contingency planning within the hospital setting [6] .

At the time this article was written, existing studies only report on hospital surge capacities in single countries [7] or hospitals [8] ; to the best of our knowledge, no cross-country overview of additional surge capacity for patients with COVID-19 is available. Furthermore, evidence on the length of stay of patients with COVID-19 in acute and intensive care units is available only for a few European countries [9] [10] [11] .

This paper aims to analyse whether COVID-19 hospitalisations exceeded the national ICU surge capacity across 16 European countries and the Lombardy region during the first wave of the pandemic (28 February to 31 July). We complement this analysis with a comparison of the average length of stay and the cumulative number of days of hospitalisation for patients with COVID-19 across the countries.

The latter parameter is additionally compared to the rate of SARS-CoV-2 infections. The findings aim to support health care decision makers in refining contingency plans and improving hospital preparedness for future health emergencies.

This cross-country analysis is based on a systematic data collection encompassing data from 18 table 4 ). If updated data were available retrospectively, we adjusted the collected data in our database accordingly. The ECDC started to provide similar data on current hospital and ICU occupancy for COVID-19 in several European countries as of summer 2020 [15] .

We also retrieved information on the numbers of hospital beds prior to the pandemic from the OECD [16] and ICU beds prior to the pandemic from the OECD/European Union [17] . Information on the surge capacity of ICU beds during the first wave was obtained from the COVID-19 Health System

Response Monitor (HSRM) [18] , with some supplementary information available from national sources. The HSRM tool was established in March 2020 and designed in response to the COVID-19

outbreak to collect and disseminate up-to-date information on how countries, mainly those in the WHO European Region, are responding to the crisis with a primary focus on the responses of health systems.

For the analysis data on SARS-CoV-2 cases and currently and cumulative hospital and ICU admissions of patients with COVID-19 were used. Spain and Switzerland did not report on current numbers of patients in acute and intensive care units and were therefore excluded from the analysis. The

Lombardy region was included because it had the highest number of cases of COVID-19 in Europe and in the world during the early months of 2020 [19] . Single missing values for currently hospitalised patients were imputed linearly to facilitate a comparative analysis of countries with incomplete daily reporting (e.g., no data were reported on weekends).

For the data collected on hospitalised patients and patients in intensive care units, definitions and reported units differ between countries due to different data collection methods used (see supplemental tables 1-18, appendix pp 1-23). For most countries, information on whether hospitalised patients with COVID-19 represent confirmed cases only or confirmed and suspected cases combined is not available. When both confirmed and suspected cases were reported, we only retrieved data on confirmed cases. Furthermore, data collection varies in terms of the reason and trajectory of hospital admission, namely, whether patients were admitted only due to COVID-19 (and excluding other pathologies and transfers from other hospitals, i.e., in Belgium) or comprise all hospitalised patients who tested positive for COVID-19 (i.e., in Denmark and France). Moreover, some countries subsumed ICU patients in the total number of hospitalised patients, while others did not (e.g., the Netherlands).

We added the numbers of inpatients treated in normal wards and in ICUs to calculate the total number of currently hospitalised patients. The definition of ICU cases is unclear in some countries, e.g., regarding the inclusion of patients in surveillance beds. A detailed description of the data collected, including sources, variable definitions (i.e. type of tests), first or last date of reporting, and collection method for each country (and region), is provided in supplemental tables 1-18, appendix pp 1-23.

The analysis presented here focuses on a subset of 16 table 1 ). Thus, a three-fold variation in the acute care capacity and even a seven-fold variation in the intensive care bed capacity were observed between the two ends of the spectrum in these countries prior to the COVID-19 pandemic. However, it must be noted that the number of ICU beds in Ireland only include those in the public sector. Figure 1 further illustrates the association between acute care and ICU bed capacities in nine countries and the Lombardy region; countries with a high acute bed capacity also tend to have a high ICU bed

capacity. Denmark appears to be an exception, with relatively high ICU bed capacities compared to a lower acute hospital bed capacity. crisis. The number of ICU beds also increased markedly in Belgium, Germany, Greece and Italy.

According to Norway's contingency plan, the ICU capacity could more than double in an emergency situation. An increase in the number of beds likely occurred also in acute care units, but data were not consistently reported in the HSRM and country-specific sources. Figure 2A shows that the pre-pandemic capacity of acute care beds was not exceeded by COVID-19

hospitalisations in any of the 11 countries. In the Lombardy region, the percentage of acute care beds occupied by patients with COVID-19 peaked at 38.3%, followed by Italy at 21.1%, while in Austria, the maximum percentage was 2.3%.

In contrast, figure 2B shows that the intensive care capacity prior to the pandemic was exceeded by Figure 3A and 3B 3C Cumulative hospital days and ICU days of patients with COVID-19 (per 100,000) Figure 3B and Figure 

To the best of our knowledge, this study is the first cross-country overview of hospitalisations associated with COVID-19 and the creation of additional intensive care resources. Our results indicate that the pre-pandemic hospital capacities varied substantially between countries included in the analysis. Regardless of their starting point, countries have implemented several measures to increase the ICU capacity during the first wave of the pandemic and meet the spike in demand for hospital care, although to different extents. The highest increase in the number of ICU beds was achieved in Ireland, the Netherlands, and Sweden, where capacity at least doubled compared to pre-COVID levels. The

Lombardy region and Greece also increased their initial capacity of ICU beds per 100,000 by 86% and 80%, and Italy and Belgium increased their initial capacity by 63% and 38%, respectively. In contrast, Germany increased its ICU capacity by only 18%, which is likely to be linked to its high initial capacity of ICU beds. From the sample of countries included in our analysis, countries with low pre-pandemic ICU bed rates (Sweden, Greece, Ireland and the Netherlands) increased ICU capacity stronger than countries with high initial ICU capacities such as Belgium and Austria.

In response to the rapidly increasing numbers of patients with COVID-19, countries have implemented different strategies. The most common strategies were the postponement of elective surgery, the reconfiguration of hospital wards, the use of private hospitals or the setup of field hospitals, to rapidly create additional acute and ICU beds [4, 21] . In the first wave, countries seemed to have used these various strategies to increase hospital capacities to highest possible levels. These strategies were increasingly aligned to real need throughout the pandemic as knowledge on expected admissions, treatment and length of stay proliferated. Some countries that did not see a critical increase in COVID-19 cases and hospitalisations in the first wave, such as Denmark, Estonia, and Norway, had contingency plans in place to reserve surge capacity for extreme situations.

While countries faced different numbers of confirmed SARS-CoV-2 infections, the available capacities of acute care beds were not exceeded by patients with COVID-19 in most the included countries. Thus, in theory, these countries would have been able to manage more COVID-19 patients or non-COVID-19

patients requiring acute care, always provided that there are sufficient health professionals. However, based on our analysis, the Netherlands and Sweden would not have had sufficient capacity to treat all patients with COVID-19 requiring intensive care without the ICU surge capacity. In the Lombardy region, the ICU surge capacity was even exceeded.

Our calculations of hospital indicators showed that COVID-19-related hospital utilisation varied substantially across the countries included. We observed a large difference in the number of cumulative treatment days of patients with COVID-19 in acute and intensive care settings and were unable to reveal a consistent relation to the incidence of SARS-CoV-2. Thus, the SARS-CoV-2 incidence alone is not the driving force for the utilisation of acute and intensive care in the hospital, but many other factors, such as demographics and morbidity of patients infected with SARS-CoV-2 [22] testing strategies, treatment pathways, and service delivery patterns may play an important role and should be included in predictive models. For example, the true number of persons infected with SARS-CoV-2 might be underestimated, depending on the use of different testing strategies across countries [23] .

Consequently, the mean number of hospital days per SARS-CoV-2 case ranged from 1.3 (Norway) to 11.8 (France), and the number of ICU days per case ranged from 0.3 (Ireland) to 1.1 (Netherlands).

These figures may serve as basic landmarks for forecasting capacity requirements to meet the surge demand. In Belgium, for example, based on the data obtained from March to July 2020, 1,000 additional infections would cause an average need for 3,800 hospital bed days (mean days per case of 3.8) and 800 ICU bed days (mean days per case of 0.8).

The study has several limitations that should be acknowledged before discussing its usefulness, e.g., Similarly, our data do not allow us to identify whether the capacities of single hospitals were exceeded.

For example, Mateen et al. [24] reported that the hospital capacities of one-third of all hospitals in England were exceeded during the first COVID-19 wave. Fifth, the focus of our study is on hospital bed capacities without considering the capacities of health professionals and medical equipment such as PPE (personal protective equipment) and ventilators that are necessary to ensure that surge capacities for patients with COVID-19 remain operational. Any decision on planning capacity should take these aspects into account [6] . Sixth, disaggregation of data related to both intensive care capacity and utilisation, i.e., low, high or intermediate care, was not performed due to limited information on type of ICU beds in international data. Hence, differences in the notion of intensive care influencing the comparability of the data may exist. A consistent and harmonised definition of ICU beds across countries would enhance comparability of capacities. Currently, there is no such definition by international organisation such as the OECD. Furthermore, the informative value of our data in regard to the length of hospital stay of patients with COVID-19 is limited, as we were only able to approximate the mean length of stay instead of calculating the median, which was reported in the majority of existing studies [9, 10] . Finally, the comparability of data across countries is limited, e.g., due to the use of different data collection methods and definitions for variables, such as hospitalised patients with COVID-19, which in some countries also include unconfirmed cases or the numbers of SARS-CoV-2 cases (which were only confirmed by PCR tests in some countries, but also confirmed by antibody tests in other countries, see supplemental tables 1-18, appendix pp 1-23).

However, when comparing our data to the published literature, our findings appear to be plausible.

For example, the Norwegian ICU registry reported an average length of stay of 17 days in October 2020 [25] , which is similar to our data from 19 [28] and were therefore able to strictly monitor the daily situation during the pandemic, which is key for reacting in a timely manner. Other countries, such as Germany [29] , established ICU registries only after the pandemic had emerged.

Overall, European countries experienced hospital capacity utilisation differently. This result underlines the importance of data collection and monitoring for planning authorities. Generally, being overprepared in extreme situations might be preferable to risking overwhelmed capacities. This finding is substantiated by a recent study showing a higher mortality rate for patients with COVID-19 in an area without access to intensive care [30] . At the same time, in many countries hospital units were restructured with elective services being postponed and occupancy rates being low which had adverse effects on patient outcomes [31] .

Our study indicates that the SARS-CoV-2 incidence is not the only aspect during the first COVID-19

wave that contributed to the burden of hospital care for patients with COVID-19, but rather the utilisation of hospital resources, as indicated by cumulative hospital days and mean length of stay, is also important. Hospital resource utilisation also depends on demographics and morbidity of infected population groups, treatment pathways, and service delivery patterns, including reimbursement policies, public health strategies and the number of hospital beds and ICU beds within a health system.

The presented analysis on intensity and timing of COVID-19 related hospital admissions might contribute to preparedness (re-)planning for healthcare during emergency phases. In countries with low ICU capacities where pre-pandemic ICU capacities were exceeded, a potentially useful approach would be to consider strategies for reserving ICU beds for future health emergencies. Upcoming studies on hospital utilisation during the second wave of the pandemic might provide additional findings that will further contribute to preparedness activities, aiming to cope with future occurrences of unpredicted large-scale needs in acute health care.

EB, JW, HE, UN, CR, TR, DP and RB report grants from the Ministry of Education and Research (BMBF) and the federal state of Berlin during the conduct of the study. UN receives funding from the German Research Foundation. DP started working for the European Observatory on Health Systems and Policies on 1 October 2020, a partnership hosted by the World Health Organization. The submitted work is independent from this relationship. RB reports grants from the Federal Ministry of Research and Education, grants from the World Health Organization, grants and other funding from the Federal Ministry of Health, and grants from the Innovation Fund for projects outside the scope of the submitted work. 

Indications for healthcare surge capacity in European countries facing an exponential increase in coronavirus disease (COVID-19) cases

Spanish minister says older people found 'dead and abandoned': US scrambles to tackle coronavirus pandemic and Germany reports that new infections are levelling off

How do the worst-hit regions manage COVID-19 patients when they have no spare capacity left?: Health System Response Monitor

European countries' responses in ensuring sufficient physical infrastructure and workforce capacity during the first COVID-19 wave

Providing health services effectively during the first wave of COVID-19: A cross-country comparison on planning services, managing cases, and maintaining essential services

World Health Organization. Beyond Containment: Health systems responses to COVID-19 in the OECD

Adapting hospital capacity to meet changing demands during the COVID-19 pandemic

Hospital surge capacity in a tertiary emergency referralcentre during the COVID-19 outbreak in Italy

Country differences in hospitalisation, length of stay and admission to Intensive Care Units due to SARS-CoV-2 infection: a rapid review of available literature (preprint)

COVID-19 length of hospital stay: a systematic review and data synthesis

Case characteristics, resource use, and outcomes of 10 021 patients with COVID-19 admitted to 920 German hospitals: an observational study. The Lancet Respiratory Medicine

Coronavirus Pandemic (COVID-19): Statistics and Research

Department of Health Care Management (MiG)

Hospitalisation, cases and tests in 18 European countries

Data on hospital and ICU admission rates and current occupancy for COVID-19

Curative (acute) care beds

Health at a Glance: Europe 2020: State of Health in the EU Cycle

COVID-19 Health System Response Monitor: Main page

The early phase of the COVID-19 outbreak in Lombardy

Intensive care management of coronavirus disease 2019 (COVID-19): challenges and recommendations. The Lancet Respiratory Medicine

Risk factors for hospitalization among persons with COVID-19-Colorado

Evaluating Data Types: A Guide for Decision Makers using Data to Understand the Extent and Spread of COVID-19

A geotemporal survey of hospital bed saturation across England during the first wave of the COVID-19 Pandemic (preprint)

Norwegian COVID-19 patients: Physicians surprised by length of stay on ventilators

World Health Organization. Critical preparedness, readiness and response actions for COVID-19

Variations in the length of stay of intensive care unit nonsurvivors in three Scandinavian countries

National ICU registries

Access to intensive care in 14 European countries: a spatial analysis of intensive care need and capacity in the light of COVID-19

The COVID-19 pandemic: impact on surgical departments of non-university hospitals

Institutional support and physical resources were provided by the Technische Universität Berlin. Erin Webb was responsible for language editing and proofreading.

EB, JW and RB conceived the study and were responsible for its design. UN, HE, DP, CR and TR contributed ideas and feedback to the conceptualisation of the study. EB, JW, HE and UN contributed to data analysis and wrote the main draft of the manuscript. EB, JW, HE UN, CR, TR and DP contributed to the acquisition and curation of data. RB and DP contributed to the critical revision of the report and study supervision. All authors critically revised and approved the manuscript.Acute care bed capacity prior to COVID-19 [ n/a n/a n/a n/a n/a n/a 10.4 c 108% UK n/a n/a n/a n/a n/a n/a n/a n/a n/a Notes: a: Own calculation based on ICU beds prior to COVID-19 per 100,000; b: including 645 ICU beds with ventilators usually reserved for patients with elective surgery; c: own calculation based on ICU beds prior to COVID-19 in total; d: we were unable to determine whether these beds were created additionally or reserved for patients with COVID-19; e: calculation based on information from HSRM that ICU bed capacity increased by 86%; f: on April 15 th , the Regional Health Authorities provided detailed contingency plans on how to increase the ICU capacity to 1,200 beds, but underlined that such a capacity can only be sustained for a short period of time; g: includes high-care and low-care ICU beds (for adults and children) but excludes post-surgery recovery beds; h: includes high-care and low-care ICU beds (for adults and children); i: includes general level 3 ICU beds for adults and children; j: includes resuscitation beds (lits de réanimation adulte)(except severe burns) and intensive care beds (lits de soins intensifs)(except neonatology) but excludes surveillance beds for adults and children (lits de surveillance) and resuscitation beds for children (lits de réanimation enfants).