Measures and Impact of Caseload Surge During the COVID-19 Pandemic: A Systematic Review

Crit Care Med. 2024 Mar 22. doi: 10.1097/CCM.0000000000006263. Online ahead of print.

Abstract

Objectives: COVID-19 pandemic surges strained hospitals globally. We performed a systematic review to examine measures of pandemic caseload surge and its impact on mortality of hospitalized patients.

Data sources: PubMed, Embase, and Web of Science.

Study selection: English-language studies published between December 1, 2019, and November 22, 2023, which reported the association between pandemic "surge"-related measures and mortality in hospitalized patients.

Data extraction: Three authors independently screened studies, extracted data, and assessed individual study risk of bias. We assessed measures of surge qualitatively across included studies. Given multidomain heterogeneity, we semiquantitatively aggregated surge-mortality associations.

Data synthesis: Of 17,831 citations, we included 39 studies, 17 of which specifically described surge effects in ICU settings. The majority of studies were from high-income countries (n = 35 studies) and included patients with COVID-19 (n = 31). There were 37 different surge metrics which were mapped into four broad themes, incorporating caseloads either directly as unadjusted counts (n = 11), nested in occupancy (n = 14), including additional factors (e.g., resource needs, speed of occupancy; n = 10), or using indirect proxies (e.g., altered staffing ratios, alternative care settings; n = 4). Notwithstanding metric heterogeneity, 32 of 39 studies (82%) reported detrimental adjusted odds/hazard ratio for caseload surge-mortality outcomes, reporting point estimates of up to four-fold increased risk of mortality. This signal persisted among study subgroups categorized by publication year, patient types, clinical settings, and country income status.

Conclusions: Pandemic caseload surge was associated with lower survival across most studies regardless of jurisdiction, timing, and population. Markedly variable surge strain measures precluded meta-analysis and findings have uncertain generalizability to lower-middle-income countries (LMICs). These findings underscore the need for establishing a consensus surge metric that is sensitive to capturing harms in everyday fluctuations and future pandemics and is scalable to LMICs.