DISCUSSION

The reduced ED utilization during the early phase of the COVID-19 pandemic has been observed in other health care systems, too, with reduction rates varying from 30 to 63%.^{Reference Giamello, Abram, Bernardi and Lauria3,Reference Cohen, Wessling and Serina5} Likewise, ED census decreased during the initial stages of the 2003 SARS epidemic in Hong Kong.^{Reference Man, Yeung, Chung and Cameron6} Our study shows that ED visits also declined in a country with relatively low ED self-referral rates.^{Reference Thijssen, Kraaijvanger and Barten7}

The reduced ED volumes during the pandemic cannot be attributed to a single cause. Several papers have raised concerns about the impact of the fear of contracting COVID-19 in the hospital, and it has indeed been reported that patients have delayed seeking emergency care because of COVID-19 fears.^{Reference De Filippo, D’Ascenzo and Angelini1,Reference Mantica, Riccardi, Terrone and Gratarola2} This fear may be largely determined by the novelty of the disease and (social) media coverage.^{Reference Cohen, Wessling and Serina5} However, there probably is more than fear that keeps patients away. The (indirect) effects of lockdowns, social distancing, and improved personal hygiene have likely played a substantial role.

First, the lockdown has been associated with a vast reduction in workplace and traffic accidents.^{Reference Nunez, Sallent and Lakhani8} Similarly, due to the restrictions in the catering industry, we expect that nightlife-related ED attendances, such as injuries and intoxications, have declined, too. Some reports also argue that reduced access may be explained by reduced ED utilization for low-urgency complaints.^{Reference Mantica, Riccardi, Terrone and Gratarola2,Reference Giamello, Abram, Bernardi and Lauria3} However, this may be less evident in our EDs because of the strong gatekeeper role of primary care in the Netherlands. Self-referral rates in Dutch EDs are exceptionally low compared to other countries.^{Reference Thijssen, Kraaijvanger and Barten7} Second, hospitals canceled elective surgeries and chemotherapies, which may have caused fewer ED visits for iatrogenic complications. In addition, there may have been a tendency to postpone consultations with specialists.^{Reference Mantica, Riccardi, Terrone and Gratarola2} It is likely that this tendency affected ED utilization as well. Third, there are theories that a decrease in physical exertion by patients as well as improved air quality may be associated with reduced access for pulmonary and cardiovascular emergencies.^{Reference Giamello, Abram, Bernardi and Lauria3,Reference Metzger, Tolbert and Klein9} Finally, the lockdown and school closures were associated with a significant decrease in pediatric infectious diseases disseminated through airborne or fecal-oral transmission.^{Reference Angoulvant, Ouldali and Yang10} Improved hand hygiene in the community may have had similar effects.

The strengths of this study are its multicenter design and the discrimination between COVID- and non-COVID-related ED visits. Similar utilization patterns were observed in all 3 EDs, underscoring the relevance of this phenomenon. Furthermore, this is the first study to determine changes in ED utilization in a country where primary care is well developed and largely responsible for relatively low self-referral rates to EDs. Its limitations are the retrospective nature of the study and the lack of adjustment for multiple potentially confounding factors. Also, the discrimination between COVID- and non-COVID-related ED visits could be debated, because triage criteria continued to evolve. However, this was probably the case in EDs all around the world. Furthermore, our observational findings do not allow for firm conclusions about the causes of reduced access for emergency conditions. In order to quantify the impact of fear and other potential variables on ED utilization, prospective, population-based research is warranted during future waves of this and other pandemics. This should include a qualitative analysis of the risk perception of patients who visited the ED and those who did not. A better understanding of underlying mechanisms may help reduce preventable injury resulting from delayed emergency care.