Obstetrics and Gynecology Emergency Department Activity during Lockdown in a Teaching Hospital, Hub Center, for COVID-19

Background The lockdown related to the SARS-CoV-2 pandemic has imposed profound changes in the interaction of the population with hospitals and emergency departments. The main aim of this research was to evaluate the impact of lockdown on the activity of obstetrics and gynecology emergency department (OGED) in a teaching hospital, hub center, for COVID-19. Methods The study considers all visits to the OGED with their different triage color codes that represent the clinical severity of each case (from the most severe to the least one: red, yellow, green, white). Data were selected through the “PSNet” triage program and collected anonymously. We analyzed frequency distributions of the variables separately for each woman and calculated mean and standard deviations for continuous variables. We then analyzed the association between factors and outcomes for categorical variables (expressed as a number and percentage of the total) using the chi-square test (χ2). The level of significance was established with p < 0.05. Statistical analysis was performed using SPSS Statistics V20.0. Given the fact that the study has a retrospective observational nature and it is based on an anonymous routine database, approval by the Local Ethics Committee was not necessary. Results The relative decrease of patients presenting to OGED in 2020 was −50.96%. The percentage of nonpregnant women was significantly lower in 2020 compared to 2019 (p ≤ 0.0001; Δ = −79.46%). Regarding the obstetric group, we saw an important decrease of visits in 2020 compared to 2019 (p < 0.0001; Δ = −40%). The prevalence of yellow codes was significantly higher in 2020 (Δ = +29.72%), while that of white (Δ = −61.58%) and green (Δ = −52.22%) codes was significantly lower (p ≤ 0.0001). Comparing the diagnoses at discharge, we could highlight significant reductions in 2020 for more than one diagnosis: bleeding (p ≤ 0.0001; Δ = −70.42%), pain (p ≤ 0.0001; Δ = −81.22%), urinary diseases (p = 0.004; Δ = −75.64%), and gastrointestinal diseases (p ≤ 0.0001; Δ = −87.50%). Conclusions An evident change emerged in relation to the dynamics between the local obstetrical and gynecological population, and OGED resources. The COVID‐19 lockdown greatly reduced the rate of admission to OGED without time-related obstetric and gynecological complications. The reduction of admissions suggests a more appropriate use of the ED by patients that may inspire future policies for the implementation of emergency services.


Introduction
e main problem of an emergency department (ED) is overcrowding, increasing sanitary costs, decreasing patient satisfaction, decreasing morale of health workers, and even increasing morbidity and mortality according to the waiting time between the triage and the actual clinical examination [1][2][3]. Overcrowding and all the related problems are also open issues in the management of obstetrics and gynecology emergency department (OGED). Almost 1/3 of OGED visits are for nonurgent services [4,5].
e Italian government adopted several restrictive measures to implement social distancing and limit the spread of SARS-CoV-2 infection. On March 9 of 2020, a state of emergency was declared, commonly called "lockdown." All nonessential activities such as office work, commercial activities, sports, and recreational activities were banned, including the free travel of people in and outside the whole country. Hospitals greatly modified their routine   Obstetrics and Gynecology International 5 during and after lockdown by suspending all nonurgent outpatient consultations and planned surgery. All emergency and urgent consultations and procedures were guaranteed. Such strict measures were kept in place until May 4 of 2020. e main objective of the study was to evaluate the impact of lockdown on OGED activity in a teaching hospital, hub center, for COVID-19. is hospital is placed in the second most afflicted Italian region in terms of the number of cases. It is less than 20 km away from Lombardy, the region with most cases in the country.

Materials and Methods
is is a retrospective monocentric study, conducted in the Department of Obstetrics and Gynecology at the "Maggiore della Carità" University Hospital, in Novara.
e study compares all admissions to OGED during the lockdown period from March 9 to May 4 of 2020 with all the admissions of the same period in 2019. Our analysis starts from March 9 of 2020 because in that month Novara was one of the first Italian cities to be included in the red zone. e study considers all visits to OGED divided into triage color codes (red, yellow, green, and white).
is hospital is a secondary care center, and it is active 24 hours a day; all patients with gynecological and obstetric symptoms refer directly to the OGED without going through the general ED. A multidisciplinary team works in OGED, including gynecologists, midwives, anesthetists, neonatologists, social workers, and nurses. Our OGED collaborates with the radiology and laboratory departments, and it is active 24 hours a day. e triage is performed by trained midwives.
en, according to the codes, the patients are evaluated by the physician on duty. In the hospital protocol, the management of triage codes is divided as follows: the red code means "emergency," that is current impairment of a vital function of the woman or fetus at a gestational age ≥23 weeks; the yellow code means "urgency," that is the threat of impairment of a vital function of the woman or fetus at a gestational age ≥23 weeks to be evaluated within 15 minutes; green code means "non-urgency," that is service which can be postponed to be evaluated within 3 hours; white code means types of services similar to an outpatient service (Table1). Once the final diagnosis is made, the patient is dismissed from the OGED with a discharge color, often not the same color that was assigned at triage. In this     work, we considered the discharge color code, not the admission one, because this is the one that is saved by the triage program. Data were extracted from "PSNet" (HiTech SPA, Software Engineering, Via di Campigliano 51, 50012 Florence, Italy) and collected anonymously. We calculated for each patient the minutes spent from admission to final decision (less than 30 minutes; from 30 to 60 minutes; from 60 to 90 minutes; from 90 to 120 minutes; more than 120 minutes). We analyzed all age groups (less than 35, from 35 to 50, from 50 to 65, and over 65) and all nationalities (Italian and others). e study included all admissions, by considering all the pregnant women as obstetric patients (from 0 to 12 weeks, from 12 to 27 weeks, and from 27 weeks to term, first pregnancy or multiparous women) and also including puerperal patients in the gynecological patients. All patients who accessed OGED were included in the study results.
ere were no exclusions.  (i) Bleeding that includes spotting and abnormal uterine bleeding.
We also collected the diagnosis at discharge, selected by the doctor at the end of the access. e discharge diagnosis for obstetrical and gynecological patients was as follows:  and before going to the examination room, or being examined but leaving before a final diagnosis was produced).
Considering the change in the number of patients accessing OGED during lockdown, our additional aim was to evaluate whether any time-related obstetric and gynecological complications occurred during that period. e complications could have developed before or during lockdown, and they could have become clinically evident after this period. For this reason, we decided to consider 60 days in addition to the ranges of time already taken into consideration (from March 9 to July 4) in both 2019 and 2020. e Italian National Health Service works on the principle that the local county will reimburse the hospital for the services provided. erefore, each patient is dismissed with a coded diagnosis, coded diagnostic and therapeutic procedures. e codes are called International Classification of Disease-9th Revision (ICD-9). For each complication we looked up the corresponding ICD-9 code and we counted how many times this code was used during both analyzed periods. e data was given to us by the administrative office that stores the clinical notes and sends the ICD-9 codes to the region for the payment of the various services that the hospital provided. It should be considered that the ICD codes have not been recently reviewed; therefore, the distinction between moderate pre-eclampsia and severe HELLP disease still exists.
We considered the following maternal-fetal and gynecological complications: We analyzed frequency distributions of the variables separately for each woman and calculated mean and standard deviations for continuous variables. We then analyzed the association between factors and outcomes for categorical variables (expressed as a number and percentage of the total) using the chi-square test (χ2). e level of significance was established with p < 0.05. Statistical analysis was performed using SPSS Statistics V20.0 (IBM Corporation, Armonk, NY, USA). Given the retrospective observational nature of the study which is also based on an anonymous routine database, approval by the Local Ethics Committee was not necessary.

Results
e population involved in this study included a total of 1857 women admitted to the OGED in the same period of 2019 and 2020 (resp., 1246 and 611) ( Table 2). e relative decrease of patients presenting to OGED was −50.96% (Table 2). e percentage of nonpregnant women was significantly lower in 2020 compared to 2019 (Table 2). We saw a significant decrease of the permanence times in OGED in 2020 mainly regarding stays that exceeded 120 minutes ( Table 2). e prevalence of yellow codes was higher in 2020 ( Table 2). Among the diagnoses of discharge, we could highlight reductions in bleeding, pain, urinary diseases, and gastrointestinal diseases (Table 3).
Among obstetrical patients, we found an important increase of the percentage of yellow codes and third-trimester pregnancy admission. e rate of hospitalization had a significant increase in percentage (Table 4).
About the amount of time in the OGED, it was clear that every code required less time to be managed, with a significant reduction of patients that waited more than 120 minutes. We found faster management within 30 minutes of all codes, in particular for the yellow one. We observed a reduction for all the admission symptoms and signs and for all the diagnoses at discharge. e decrease was more evident for bleeding, pain, and gastrointestinal and respiratory issues ( Table 5).
In respect of the specific obstetrical symptoms, there was a percentage increase of patients that were admitted for uterine contractions and then hospitalized (Table 6).
We also noted an increased number of patients admitted for vaginal discharge, but a decreased number of patients with diagnosis of miscarriage, threatened abortion, or threatened preterm labor. We observed a reduction of gynecological patients discharged with diagnosis of pain and a rise in the percentage of distal genital tract issues ( Table 7).
None of the time-related obstetric and gynecological complications showed significant changes despite the significant reduction in the number of visits during lockdown, even if the number of cases was too small and statistical significance was not demonstrated (Table 8)

Discussion
Unprecedented public health interventions were put into action to reduce the risk of SARS-CoV-2 transmission [6]. A sudden and sharp reduction in ED attendance began immediately after the spread of the first SARS-CoV-2 infection in Italy. Starting on February 21st, overall, ED attendance abruptly decreased by 41.8% in comparison to the previous year [7]. Elective surgery and outpatient visits were reduced; both medical and surgical wards were shifted to COVID-19 units; new supplementary sub-intensive and ICUs were created for COVID-19 patients [8]. e number of visits during lockdown decreased; in particular, for nonpregnant patients, the reduction was highly significant in comparison to previous data [9]; that reduction could be due to fear of contagion, transportation limitations, fear of getting infected, and increased use of telemedicine. Moreover, visits to private doctors' offices decreased [10,11].
During the last twenty years, OGEDs have been utilized for non-emergent medical care more and more, especially during pregnancy [4]. Emergency care settings in the US treat approximately 750000 patients annually for chief complaints related to gynecology and obstetrics. Interactions with the healthcare system during pregnancy and the postpartum period represent opportunities for health promotion [12], preventative health screenings [13], and engagement in long-term behavioral changes to reduce the future risk [14]. Some women have multiple barriers to outpatient follow-up, including lack of childcare, difficulties in accessing care, and lack of understanding about the longterm health risks associated with pregnancy complications [15]. ese conditions may increase women's reliance on OGED settings [16]. is aspect could explain why in the lockdown period there was a reduction in the access of Italian patients, while this reduction was not equally present in patients with other origins, as also shown in another study [9]. e reduction of improper visits allowed the staff to attend to the real emergency codes, such as red and yellow, within the recommended time. Prompt ED service is expected when patients present to an emergency room with urgent health problems [17]. However, various factors can increase the waiting time for access to health services and also can prevent timely intervention for patients consulting emergency rooms [18]. As a result, the probability of serious complications, such as disability and death, has increased [19]. It is difficult to prevent emergency room congestion and overcrowding because patients, who do not require emergency care, also consult ED [20]. Extended ED waiting times result in a longer duration of pain experienced by patients, negative health effects, and reduced patient satisfaction [21]. Moreover, because of increased waiting times, growing numbers of patients leave the emergency room without receiving medical care [22]. As expected, the waiting time increases with increasing numbers of patients in the emergency room and reduces with decreasing numbers of patients consulting the emergency room [23]. Moreover, ED visits are often less efficient and less equipped for longitudinal care of chronic medical diagnoses [24]. In 2019, the percentage of LWBS was 8.2%; during 2020 lockdown period, it was 0.3%, with a significant decrease of −98%. LWBS is a major concern for physicians, healthcare providers, and hospitals [22]. e LWBS population, varying from less than <1% to more than 10% of all triaged patients (depending on the ED), has been considered in the literature as a shortfall in healthcare access, as these patients do not receive the care they originally sought [25]. Some studies suggest that LWBS rates may be an expression of the patient's level of safety as some patients who did not receive medical treatment had avoidable health outcomes. Conversely, it may also be hypothesized that the decision to LWBS reflects a lower-acuity complaint that has been resolved or will be resolved without medical intervention [26]. LWBS patients are high ED utilizers, accounting for nearly three times the number of visits in comparison to the general ED patients. ey could represent a patients' population with persistent lack of access to reliable outpatient healthcare options or to adequate management of chronic pain and psychiatric conditions. During our research, we saw an increase in the percentage of hospitalizations. is could be related to a reduction in less severe cases with a consequent proportional increase in situations of greater complexity. In support of this hypothesis, there was an increase in the percentage of yellow codes and a decrease in green and white codes. Moreover, a previous study on gynecological visits suggests that, during the COVID-19 lockdown, real emergencies have been screened from more deferrable cases.
is led to a proportional increase of the number of hospitalized women, especially for emergent surgeries, and a decrease of the number of women discharged from the ED [27].
Pain seems to be the reason for which the most substantial reduction in OGED visits has occurred. is leads us to hypothesize that this symptom is considered by patients to be less alarming than bleeding, causing less discomfort, compared to vaginal infections. e evidence of less visits to OGED for nonspecific symptoms suggests that the OGED was used more appropriately, due to the reduction of overcrowding, a decrease of inappropriate visits, and shorter waiting times.
For a more appropriate use of OGED, some changes could be useful: firstly, implementation of local health services to manage nonspecific symptoms in pregnancy; secondly, a more clear communication regarding how to manage pregnancy related symptoms during planned check-ups; thirdly, more expensive bills for accessing OGED for nonurgent reasons. is could allow less crowding and shorter waiting times, with staff being at ease during working hours.
One unexpected result, which amazed us, was the reduction in patients who tried to take medication, for example, a simple painkiller, before coming to the emergency room. In this regard, the hypotheses are the following: firstly, there was an increase in anxiety, so patients felt that it is safer to go to ED; secondly, it was more difficult to access a pharmacy and telephone advice of a doctor; thirdly, in the first months of the pandemic, fake news regarding the use of nonsteroidal anti-inflammatory drugs was published online [28].
As for second aim of our study, there were no differences between the time related to maternal-fetal or gynecological complications in the two periods under study, although there was a significant reduction in the number of visits for obstetric and gynecological reasons. In a previous published study, we demonstrated that there were no significant differences either for antepartum or intrapartum complications [29]. We found a high score for anxiety and depression, although it cannot be compared to the same score on the same population before the pandemic [30]. e high level of anxiety and depression we found is consistent with other studies [31,32]. e high prevalence of anxiety and depressive symptoms in pregnant women and new mothers should be a public health issue; in addition, screening for perinatal depression and anxiety should be considered during a pandemic. Data related to changes in obstetric or gynecological complications during lockdown are extremely conflicting [33].
An Italian study showed an increase of about three times in stillbirths, a decrease in the percentage of late preterm births, and an increase in full-term births.
ere was a nonsignificant increase in very preterm births and a nonsignificant reduction in caesarean sections [34].
Dell'Utri et al. showed a significant increase in the incidence of stillbirth, which did not occur in our setting. is could be explained as follows: either their catchment area is more extended than ours, or in Lombardy the percentage of sick professionals was higher than in Piedmont, and therefore our patients were able to contact their physicians for advice with less difficulty [9]. A Spanish study did not find any link between prematurity and lockdown, nor between stillbirths and lockdown [35]. A reduction in the rate of preterm birth was observed in France and Denmark [36]. Another study showed a decrease in iatrogenic preterm births during the initial COVID-19-related lockdown in the Netherlands in singletons [37]. An Iranian study showed, on the one hand, no significant differences in stillbirth rates or pregnancy complications (including preeclampsia, pregnancy-induced hypertension, and gestational diabetes) and, on the other hand, a decrease in preterm births and low birth weight in the pre-and intra-pandemic periods [38]. In Jordan, there were no significant differences in preterm birth and stillbirth rates, neonatal mortality, or perinatal mortality before and during the COVID-19 lockdown, but there was a significantly lower incidence of extreme low birth weight (ELBW) infants (<1 kg) during the COVID-19 lockdown period than that before the lockdown [39]. In India there was a reduction of 45.1% in institutional deliveries, a percentage point increase of 7.2 in high-risk pregnancy, and a 2.5-fold rise in admission to the intensive care unit of pregnant women during the pandemic. One-third of women had inadequate antenatal visits. e main reasons for delayed health-seeking were lockdown and fear of contracting infection, resulting in 44.7% of pregnancies with complications [40]. e major strength of our study is that we analyzed in detail and compared the symptoms for which patients, both obstetrical and gynecological, visited OGED by examining the variations of the assigned triage codes.

Study Limitations.
e retrospective collection of patient data could have been biased by possible errors in the recording of data. e definition of the final diagnoses was highly heterogeneous in the electronic records. Although the definition of the principal complaint at triage was consistently quite homogeneous, it is also possible that the triage midwifes interpreted the patients' complaints in a different way. We chose not to analyze in this study the data relating to the activity of the delivery room, because the complexity of the variations requires a separate study, which is now under review. We describe what happened during the outbreak in Piedmont. It seems reasonable to speculate that this small-scale scenario could repeat itself in other, similar epidemiological contexts.

Conclusions
An evident change emerged in relation to the dynamics between the local obstetrical and gynecological population, and OGED resources. e COVID-19 lockdown greatly reduced the rate of admission to OGED, and this reduction suggests a more appropriate use of OGED that may inspire future policies for the implementation of emergency services. e reduction in the admission rate did not lead to an increase of time-related obstetric and gynecological complications, but further studies are needed to evaluate this.
Data Availability e data can be requested by contacting Dr. Roberta Amadori (roberta.amadori@uniupo.it).

Conflicts of Interest
e authors declare that there are no conflicts of interest regarding the publication of this article.

Authors' Contributions
R. Amadori contributed to protocol/project development, manuscript editing, and data analysis. R. Buscemi, A. Desando, and F. Grillo participated in data collection/ management and manuscript writing/editing. D. Surico and V. Remorgida supervised the work.