Impact of COVID-19 on the Incidence and Severity of Obstetric and Gynecologic Emergency Department Visits in an Integrated Health Care System
Cassidy E Tierney, MD1; Mary Kathryn Abel, AB2; Mubarika M Alavi, MS3; Miranda Ritterman Weintraub, PhD, MPH4; Andrew Avins, MD, MPH5; Eve Zaritsky, MD6 Perm J 2022;26:21.136 • E-pub: 04/05/2022 • https://doi.org/10.7812/TPP/21.136 Volume 26, Issue 1Corresponding Author Author Affiliations 2University of California, San Francisco, School of Medicine, San Francisco, CA, USA 3Division of Research, Kaiser Permanente Northern California, Oakland, CA, USA 4Department of Graduate Medical Education, Kaiser Permanente Northern California, Oakland, CA, USA 5Division of Research, Kaiser Permanente Northern California, Oakland, CA, USA 6Department of Obstetrics and Gynecology, Kaiser Permanente Northern California, Oakland, CA, USA Author Contributions: Disclosures Copyright Information AbstractOBJECTIVE: COVID-19 has had an unprecedented impact on medical care use and delivery, including stark reductions in emergency department (ED) volume. The aim of this study was to assess changes in incidence of OB/GYN ED visits and disease severity at time of presentation during the COVID-19 pandemic. STUDY DESIGN: We conducted a multicenter retrospective study of OB/GYN-related ED visits before and during the COVID-19 pandemic. Incidence rates (IRs) and severity measures were compared across time periods and years. RESULTS: A total of 18,668 OB/GYN ED encounters occurred between January 1 and December 31, 2020, compared to 21,014 encounters between January 1 and December 31, 2019. During shelter-in-place, visits decreased by 41% compared to the pre-pandemic period in 2020 before returning to typical rates (incidence rate ratio (IRR) = 0.98 in fall/winter). We found a similar proportion of patients with hemoglobin < 7 g/dL for diagnoses associated with bleeding and patients with white blood cell count > 12,000 per μL in the setting of infection comparing corresponding time periods in 2019 and 2020. There were fewer formal OB/GYN consults, hospital admissions at time of presentation, and urgent surgical procedures performed across all periods in 2020; however, hospitalization within 7 days substantially increased in the first half of 2020. CONCLUSION: The incidence of OB/GYN ED visits declined substantially between March and August 2020 but then returned to pre-pandemic levels by fall/winter 2020. The decreased incidence was not accompanied by an increase in severity of presentation. IntroductionThe COVID-19 pandemic has had an unprecedented impact on medical care use and delivery. In addition to navigating the challenges of COVID-19, health care systems were forced torethink how to deliver non-COVID-19–related health care. In-person visits were replaced with virtual visits, surgical procedures were postponed, and preventive screenings, such as Papanicolaou tests and colonoscopies, were deferred.1 Previous studies have shown stark reductions in emergency department (ED) visits for serious conditions, such as myocardial infarction and stroke.2–5 Reports of variation in disease severity during the pandemic have been mixed, with some studies suggesting an increase in adverse outcomes in the setting of delayed presentation.6–8 Myriad OB/GYN concerns bring patients to the ED, ranging from miscarriage and ovarian torsion to fibroids. In some cases, such as ectopic pregnancy, delayed treatment can be life threatening. An investigation by this group and others found that ED visits for OB/GYN concerns declined substantially during the early pandemic period9,10; however, to our knowledge, no studies have assessed long-term changes in emergent OB/GYN care. There is also scant literature on the pandemic’s impact on severity of illness at time of presentation specifically for OB/GYN-related concerns.11 The objective of this study was to further elucidate trends in OB/GYN ED visits and changes in the severity of presentation during the COVID-19 pandemic, as a proxy for potentially delayed care. MethodsSTUDY POPULATIONDemographic and clinical characteristics among female patients 18 years or older presenting to a Kaiser Permanente Northern California ED for OB/GYN emergencies were examined. Kaiser Permanente Northern California is an integrated health care system that served 1.8 million women in 2020. It includes 21 hospitals and encompasses facilities across the northern and central parts of the state, including the San Francisco Bay Area, Silicon Valley, Sacramento, and much of the Central Valley. STUDY DESIGNThis study employed a multiple cross-sectional design comparing 2 consecutive years. For each encounter, diagnoses were classified as OB or GYN according to International Classification of Diseases, 10th Revision, Clinical Modification (ICD-10-CM) diagnostic codes. OB diagnoses included bleeding < 20 weeks and postpartum, endometritis/pyelonephritis infections, spontaneous/missed abortion, ectopic pregnancy, molar pregnancy, and nausea/vomiting of pregnancy. GYN diagnoses included fibroids/ abnormal uterine bleeding (AUB), pelvic inflammatory disease, ovarian torsion/cysts, and pelvic pain. Pelvic pain, molar pregnancy, and nausea/vomiting of pregnancy were excluded from additional analyses after the overall incidence rates (IRs) were calculated because they did not have reliable measures of severity. All medical and demographic data were extracted from the electronic medical record. COVID-19 case and hospitalization data were extracted from the California Open Data Portal through the Government Operations Agency.12 Four time periods in 2020 and corresponding periods in 2019 were compared: pre-pandemic (January 1–March 3), initial California shelter-in-place (March 4–May 31), summer (June 1–August 31), and fall/winter (September 1–December 31). March 4, 2020, marks the first documented COVID-19 death in California and the day California Governor Gavin Newsom declared a state of emergency. A statewide shelter-in-place order was later enacted on March 19, 2020. Our study was deemed exempt with waiver of consent from Institutional Review Board review by the Kaiser Permanente Northern California Research Determination Office. STATISTICAL ANALYSESIRs (per 100,000 person-weeks at risk) during the pandemic period were compared with the pre-pandemic period. Person-weeks were calculated according to the total adult (≥ 18 years old) membership of Kaiser Permanente Northern California at the middle of each monthly period. Patients who were not Kaiser Permanente Northern California members at the time of ED presentation were included in the overall Kaiser Permanente IR trend analysis (Figure 1) but excluded from all other analyses. Differences among time periods were compared with incidence rate ratios (IRRs), overall and stratified by key patient characteristics and diagnoses (Table 1). Attributable risk percentage changes between the time periods were calculated with 95% confidence intervals (CIs) for demographic characteristics and illness-severity measures, including rates of OB/GYN consult, hospitalization (both at time of encounter and within 7 days), surgery within 7 days, Intensive Care Unit admission within 2 days, and death within 7 days. Mean lengths of hospital stay pre-COVID and during the COVID-19 pandemic were also compared using the two-sample t-test (and validated with nonparametric Mann-Whitney tests). Only encounters with an OB/GYN primary diagnosis were further analyzed for differences in severity. Relevant vital sign and laboratory measurements by condition, including temperature ≥ 38°C (100.4°F), hemoglobin < 7 g/dL, white blood cell count (WBC) > 12,000 per μL, lactate > 2 mmol/L, and human chorionic gonadotropin > 5000 IU/mL, were compared across time periods using the Pearson Chi-squared test and validated with the Fisher exact test. A two-sided p value < 0.05 was considered significant. All analyses were conducted with SAS (version 9.4; SAS Institute, Inc., Cary, NC).
Figure 1: Incidence of ED visits for obstetric and gynecologic conditions before and during COVID-19. Encounters for members and non-members were included. Person-weeks were calculated according to the total adult (18 years of age or older) membership of Kaiser Permanente Northern California at the middle of each monthly period. Vertical bars indicate 95% CIs. The data in grey represent the number of patients diagnosed with COVID-19 in Northern California per 100,000 person-weeks, and the data in black represent the number of patients hospitalized with COVID-19 in Northern California per 100,000 person-weeks. These rates are calculated using data from the official California Department of Public Health COVID-19 statistics, limited to Northern California counties and the current population of Northern California (approximately 15.7 million). ED = emergency department.
Table 1: Characteristics of patients presenting to the ED for OB/GYN conditions before and during COVID-19 aICD-10-CM codes for conditions are as follows: fibroids/AUB (D25.x, N92.4x, N93.9x, N95.0x); infection, including pelvic inflammatory disease, tubo-ovarian abscess, cervicitis, endometritis, or vulvovaginal cellulitis (N76.4x, A18.1x, A54.2x, A56.1x, N70.0x, N70.9x, N71.0x, N71.9x, N72.x, N73.0x, N73.2x, N73.3x, N73.5x, N73.8x, N73.9x, N75.1x, N82.x); ovarian torsion/cysts (N83.5x, N83.2x); ectopic pregnancy (O00.x); spontaneous or missed abortion (O02.1x, O03.x, O20.0x, Z33.2x); and pregnancy-related bleeding prior to 20 weeks or postpartum (O04.x, O20.9x, O26.85x, O44.1x, O44.3x, O45.x). bBold numbers indicate statistically significant difference in percent change between 2019 and 2020. AUB = abnormal uterine bleeding; BMI = body mass index in kg/m2. ResultsINCIDENCEA total of 35,475 OB/GYN ED encounters occurred between January 1 and December 31, 2020, compared to 42,509 encounters between January 1 and December 31, 2019, a 16.6% reduction in overall visits. Excluding pelvic pain diagnoses and limiting to Kaiser Permanente members, the final cohort consisted of 18,668 encounters in 2020 and 21,014 in 2019. Overall incidence of OB/GYN ED encounters for 2020 decreased dramatically during the start of the COVID pandemic, rising steadily during the summer period corresponding with an increase in COVID-19 cases and hospitalizations before returning to pre-pandemic levels in fall/winter (Figure 1). Among women who had a diagnosis for which a reliable severity measure existed (see above), visits decreased by 41% (IRR = 0.59) during the early pandemic period (March 4–May 31) and by 8% (IRR = 0.92) during the summer (June 1–August 31) compared to the pre-pandemic period (Table 2). Findings were similar when analyzing OB and GYN ED encounters separately. There was a sharp decline in OB and GYN visits during the early pandemic period (IRRs = 0.59 and 0.60, respectively). By Period 3 (summer), GYN ED visits had returned to pre-pandemic levels (IRR = 0.96), while OB visits remained significantly lower (IRR = 0.82). The incidence of OB and GYN visits were similar to the pre-pandemic period (IRRs = 0.94 and 1.01, respectively) by Period 4 (fall/winter). There were no clinically substantial differences in age or race/ethnicity across the 2 years (Table 1).
Table 2: OB/GYN ED visits during early pandemic and late pandemic periods compared to the pre-pandemic period in 2020 bObstetric ED visit indications were quantified as bleeding (ICD-10-CM codes O20.0, O20.9, O26.8, O44.1, O44.3, O45.x), infection (O23.0), and other (O00.x, O02.x, O03.x, O04.x). Events by indication sum to greater than total events because some patients presented with multiple indications. cper 100,000 person-weeks at risk, with person-weeks calculated according to the total adult (≥ 18 years old) membership of Kaiser Permanente Northern California at the middle of each monthly period. dGynecologic ED visit indications were quantified as nonobstetric bleeding (N92.4, N93.9, N95.0, D25.x), infection (A18.1, A54.2, A56.1, N73.0, N73.2, N73.3, N73.5, N73.8, N73.9, N70.0, N70.9, N71.0, N71.9, N72, N75.1, N76.4, N82), and other (N83.2, N83.5). Events by indication sum to greater than total events because some patients presented with multiple indications. ED = emergency department; IR = incidence rate; IRR = incidence rate ratio. SEVERITYAn OB/GYN specialty consult was placed less frequently in the ED across all time periods in 2020 compared to 2019, and fewer patients received surgery within 7 days of ED presentation (Table 3 and Figure 2). There were also fewer hospitalizations associated with the ED encounter between January 1 and May 31, 2020, compared to the corresponding period in 2019; however, individuals were more likely to be hospitalized within 7 days after the initial ED encounter during the first surge of the pandemic between March 4 and May 31, 2020. Mean length of hospital stay during COVID (March 4, 2020–December 2020) was significantly shorter than during the pre-COVID period (January 2019–March 3, 2020) (76.8 hours [95% CI 70.0, 83.6] vs. 86.4 hours [95% CI 81.3, 91.4], p < 0.03). Intensive Care Unit admission and death were both extremely rare outcomes, with Intensive Care Unit admission occurring in 0.1% to 0.2% of all cases and death occurring in 0% to 0.1% of all cases across all time periods. There were no statistically significant differences in vital sign and laboratory measurements of severity for conditions with primary diagnoses, including temperature, hemoglobin level, WBC, lactate level, and human chorionic gonadotropin level, between the 2 years (Table 4).
Table 3: Changes in presentation and outcome characteristics for OB/GYN ED encounters at Kaiser Permanente Northern California in 2020 compared to 2019a aResults are reported as absolute percent change (% in 2020 − % in 2019) and 95% CIs. ICD-10-CM codes for conditions are as follows: fibroids/AUB (D25.x, N92.4x, N93.9x, N95.0x); infection, including pelvic inflammatory disease, tubo-ovarian abscess, cervicitis, endometritis, or vulvovaginal cellulitis (N76.4x, A18.1x, A54.2x, A56.1x, N70.0x, N70.9x, N71.0x, N71.9x, N72.x, N73.0x, N73.2x, N73.3x, N73.5x, N73.8x, N73.9x, N75.1x, N82.x); ovarian torsion/cysts (N83.5x, N83.2x); ectopic pregnancy (O00.x); spontaneous or missed abortion (O02.1x, O03.x, O20.0x, Z33.2x); and pregnancy-related bleeding prior to 20 weeks or postpartum (O04.x, O20.9x, O26.85x, O44.1x, O44.3x, O45.x). bBold numbers indicate statistically significant difference in percent change. AUB = abnormal uterine bleeding; ED = emergency department; ICU = Intensive Care Unit.
Figure 2: Disease severity measures for OB/GYN ED encounters at Kaiser Permanente Northern California in 2020 compared to 2019. Graph results are percentages of total (N) encounters for each time period. Statistically significant differences (p < 0.05) between years are noted in bold in Table 3. ED = emergency department.
Table 4: Laboratory measurements of illness severity by OB/GYN emergency subtype in 2019 and 2020a aTest of 2 proportions were performed to assess differences in laboratory values in 2019 versus 2020. No analyses were statistically significant below an alpha level of < 0.05. bInfection defined as pelvic inflammatory disease, tubo-ovarian abscess, cervicitis, endometritis, or vulvovaginal cellulitis. AUB = abnormal uterine bleeding; CI = confidence interval; hCG = human chorionic gonadotropin; Hgb = hemoglobin; WBC = white blood cell. DiscussionIn this integrated health care system, the incidence of OB/GYN ED visits declined precipitously between March and May 2020 but then returned to near-pre-pandemic levels despite the increasing rate of COVID-19 cases across California. The reduction in ED OB visits persisted longer than GYN visits, with IRs remaining substantially below the pre-pandemic period until fall/winter. The underlying reasons for this trend are unclear. One possible explanation is that shelter-in-place mandates as well as concerns regarding safety of medical facilities may have initially shifted thresholds for patients seeking emergent care; however, as government policies changed and “pandemic fatigue” set in, health care utilization patterns returned to baseline, even with COVID-19 cases on the rise. Several additional factors may have contributed to these findings, such as improved telehealth care decreasing the need for ED visits as well as messaging surrounding the safety of seeking emergency care. The delayed normalization of OB ED visits may be explained by an overall decrease in pregnancies in the early pandemic period and possibly increased virtual prenatal visits initiated with the pandemic. An alternative explanation is that pregnant patients avoided seeking care in an ED setting longer than their nonpregnant counterparts due to fear of exposure, a hypothesis supported by evidence of patients choosing to forgo antenatal visits or even considering home birth to avoid entering a medical setting.13–15 We found no evidence that this period of decrease in visits was accompanied by an increase in severity of presentation. There were fewer OB/GYN consults and surgical procedures performed within 7 days of presentation across all periods of 2020 compared to 2019. Notably, however, the lowest chance of hospitalization at time of ED encounter was between January 1 and May 31, 2020 (Periods 1 and 2), but the highest chance of hospitalization within 7 days of the initial ED encounter was between March 4 and May 31, 2020 (Period 2). There was also a substantial decrease in length of hospital stay between the pre-pandemic period and during the COVID-19 pandemic period, with a mean difference of about 10 hours, a difference of uncertain clinical significance. With COVID-19, Kaiser Permanente Northern California rapidly expanded its virtual care visit opportunities for patients, potentially decreasing the overall need to visit the ED. However, there may also have been some hesitation to admit or operate on patients during the pandemic, particularly when COVID-19 testing ability was limited and operating room staff had been reassigned, leading physicians to favor outpatient medical management over surgical intervention. This does not explain the initial decline in these parameters, which was noted prior to the first documented case of COVID-19 in California. The increase in hospitalization within 7 days of a patient’s initial ED visit suggests that, in some cases, threshold for admission may have been unusually high to minimize patient exposure to COVID-19 and keep inpatient censuses lower for expected COVID-acquired patients. The shorter mean hospital stay during the pandemic, albeit small and of uncertain clinical significance, is consistent with potential increased focus on shortening inpatient stays for those admitted with non-COVID-19 conditions to both reduce chance of transmission and prepare for potential COVID-19 admissions. A major strength of our study is the ability to present multicenter data from a wide network of northern California hospitals. To our knowledge, this is the largest study to date specifically assessing trends in OB/GYN emergency care during the pandemic. It is also one of the only studies of its kind, across all disciplines, to present data past the acute phase of COVID-19 (through December 2020 for this study). This gives us the unique ability to observe both the acute impact of the early pandemic as well as changes over time as medical and social norms and care patterns shifted. This study also has limitations. Given that these data are from an integrated health care system in a single state, conclusions may not be readily generalizable to other health care systems in other parts of the country. We are also limited in our ability to report specifically on OB emergencies. A subset of Kaiser Permanente Northern California facilities have labor and delivery units that address all OB emergencies in pregnant patients ≥ 20 weeks gestation. Because of this, we only included obstetrical conditions that would present to the ED, regardless of the presence of a labor and delivery unit. Therefore, we cannot draw any conclusions about incidence or severity of OB emergencies in the second half of pregnancy. There were several limitations in our statistical analyses. Calculation of IRs using an estimated time-at-risk inherently assumes the probability of disease during the study period is constant for each individual, which is not true in the case of many OB/GYN emergencies, which are more prevalent (or exclusively seen) in certain age categories. There is, however, consistency in this confounding factor across all calculations. Furthermore, we did not adjust for multiple statistical comparisons, but we were careful not to assert potential differences unless they were consistent across time periods or showed monotonic change over time. Finally, while overall statistical power was usually high, as evidenced by generally narrow CIs, some comparisons had lower power, and findings of no statistically significant differences should be interpreted cautiously. Finally, limitations also include those inherent to the retrospective nature of the study, including variations in electronic medical record data quality and availability. Mortality data are limited to those deaths that occurred while hospitalized, so we cannot comment on any differences in death occurring outside of the hospital setting. Because our person-time rate calculations only included Kaiser Permanente Northern California members, ED visits of patients without Kaiser Permanente Northern California insurance were excluded from all statistical analyses, limiting our ability to comment on trends for uninsured or otherwise-insured populations. Our findings suggest that the decrease in ED visits was not accompanied by an increase in acuity; however, it is unclear whether this was because patients were accessing alternative care venues during that time. Current literature suggests that decreases in ED volume were accompanied by increased telehealth usage in several health care systems.16,17 Future research is needed to assess whether the initial decline in OB/GYN ED encounters was similarly accompanied by an increase in other care encounter types, such as virtual visits or outpatient office visits. ConclusionThe incidence of OB/GYN ED visits declined substantially during the early pandemic period but then returned to near-2019 levels by summer 2020, despite a rise in COVID-19 cases and hospitalizations. Illness severity upon presentation to the ED did not appear to change substantially during the pandemic. References
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ETOC
Click here to join the eTOC list or text ETOC to 22828. You will receive an email notice with the Table of Contents of The Permanente Journal.
CIRCULATION
2 million page views of TPJ articles in PubMed from a broad international readership.
Indexing
Indexed in MEDLINE, PubMed Central, EMBASE, EBSCO Academic Search Complete, and CrossRef.
