COVID-19疫情對剖腹產及生產住院醫療耗用的影響-高風險與非高風險族群之差異

Abstract

背景: COVID-19帶給整體社會、經濟以及醫療體系莫大的衝擊。孕產婦的健康一直是國際長期重視的議題之一，然而面對醫院配合防疫政策及整體醫療量能減少之下，國外研究顯示非高風險的孕產婦有可能因為確診或害怕染疫，改變生產方式或生產過程中接受更多的產科介入，且文獻對於疫情是否造成剖腹產上升的證據並不一致。儘管近十幾年來台灣政府致力於改善周產期的照護，但剖腹產率依舊高居不下，且於疫情期間仍有上升趨勢。因各國疫情爆發時間、實施的政策、醫療資源不同，有關疾病防治的規定或啟動時機亦有所差異。據此，疫情可以做為一個很好的挑戰，透過分析可以了解疫情是否影響了不同風險孕產婦的生產行為和資源耗用。

目的: 藉由估算疫情前後不同風險的孕產婦之間的差距，探討疫情本身是否會加劇持續上升的剖腹產率，並分析疫情下在孕產有沒有增加醫療資源耗用。

方法: 台灣於2020年1月21日爆發首例確診，並自2021年5月15日進入三級緊戒。為了觀測疫情爆發前後的實際影響，本研究運用衛生福利資料科學中心之資料進行次級資料分析，以2020年為基準，選擇2019年與2021年兩年期間在出生通報檔有分娩事實之孕產婦做為主要研究對象。主要自變項為是否為高風險族群，根據孕齡（35歲以上/未滿20歲）或有任一孕期間危險因子（包括抽菸、酗酒、藥癮、妊娠糖尿病、妊娠高血壓等不良健康行為與妊娠併發症）來判定是否為高風險孕產婦。控制變項共有八項，包括年齡、投保金額、原始國籍別、產檢次數、居住地區都市化程度、共病指標、COVID-19相關症狀與婦產科醫師人力。依變項包含有無剖腹產、有無催生或引產、住院天數和生產住院總費用，均使用差異中的差異法並控制醫院特性，分別以多元線性機率模型和廣義估計方程式，分析疫情是否會影響不同風險族群間剖腹產利用與醫療資源耗用的差異。

結果: 研究共納入246,660位產婦，高風險族群的比例在疫情前為32.32%，疫情後為36.49%。透過雙變項分析，得知高風險產婦與非高風險產婦疫情前後各自剖腹產率和醫療資源耗用皆有顯著的改變。疫情前後，高風險產婦剖腹產從43.70%增加至44.56%，非高風險產婦剖腹產從30.67%增加至31.57%。至於催生或引產的比例，高健康風險產婦從10.55%增加至15.35%，非高風險產婦從11.22%增加至16.74%。不同風險族群疫情期間的平均住院天數與費用都相比疫情前有顯著的成長。多變項分析結果顯示，產婦的風險程度各自與剖腹產的利用、催生或引產、住院天數與住院醫療費用均有顯著的正向關係；疫情各自與催生或引產及住院醫療費用有著正向顯著關係，與住院天數則是正相關但無統計上無顯著。對於疫情是否會影響高風險族群與非高風險族群之間的差距，結果顯示在剖腹產利用和住院天數上無顯著差異，但在催生或引產和住院醫療費用上有統計上顯著差異(p<0.05)。

結論： 整體剖腹產率和生產住院醫療耗用在疫情前後存在顯著差異，均有上升的趨勢，並且催生或引產和費用的差異在控制各變項後依舊存在。產婦的風險程度彼此之間存在差異，且會影響剖腹產的利用和生產醫療資源耗用。此外，疫情會使不同風險族群間使用催生或引產的差距縮小、生產費用的差距擴大。總結來說，台灣的醫療衛生體系在面對疫情壓力時，雖然有能力去應變，但額外的產科介入和花費較多的資源可能是疫情的代價之一。疫情前後整體生產模式保持穩定，但仍不可忽視非高風險產婦上升的剖腹產率，以及在無醫學適應症下進行剖腹產或催生等產科介入對孕產婦和新生兒健康帶來的潛在風險。未來政策建議除了持續進行高風險孕產婦的周產期照護之外，同時也應針對非高風險孕產婦進行相對應的孕期管理，減少原可避免的額外產科介入。

Background: The COVID-19 pandemic has imposed significant challenges on society and healthcare systems globally. Lockdown policies and the implementation of COVID-19 pre-delivery screening may have changed labor and delivery practices during the pandemic. Some pregnant women may have chosen to undergo a cesarean section (C-section) or receive more obstetric interventions out of fear of COVID-19 and due to restrictions on their partners’ access to maternity wards. Previous evidence is inconsistent on whether the pandemic has led to increased C-section and induction rates. While some research in high-income countries showed no changes in delivery mode, recent data from Taiwan indicate an increase in cesarean rates during the COVID-19 pandemic. Given the variations in pandemic onset, policies, and healthcare resources across countries, the regulations and timing of disease prevention measures differ. Therefore, the pandemic presents a unique opportunity to analyze its impact on mode of delivery and healthcare resource utilization among different risk groups of pregnant women.

Objective: This study aims to compare C-section and healthcare resource utilization (includes induction rates, length of hospital stay, and medical utilization), assessing variations between different health risk groups in Taiwan before and during the COVID-19 pandemic. The goal is to determine whether the pandemic has contributed to the increasing trend in C-section rates and how it has impacted healthcare resource utilization among pregnant women with varying risk.

Study Design: Using the Taiwan National Health Insurance Research Database (NHIRD), this retrospective cohort analysis compares women who had live singleton-term deliveries before and during the COVID-19 pandemic. Since Taiwan reported its first COVID-19 case on January 21, 2020, and announced the nationwide Level 3 epidemic alert on May 15, 2021, we selected pregnant women in 2019 as the pre-COVID period and in 2021 as the COVID period to observe the actual impact of pandemic. Linked with the Birth Registry to inpatient admissions and outpatient service claims, the dataset includes information on infants' birth methods and mothers' health conditions before and during pregnancy. The primary independent variable is whether the woman is considered high-risk, defined by medical risk factors including maternal age (<20 or >35 years old), pregnancy complications (e.g., gestational diabetes, hypertension), and unhealthy behaviors during pregnancy (e.g., smoking, substance use). All models adjust for maternal characteristics, including age, ethnicity, number of prenatal visits, COVID-19 symptoms, comorbidities, doctors, and socioeconomic background. The study uses a difference-in-difference method, employing linear probability models with a hospital fixed-effect model for binary outcomes. For continuous outcomes, generalized estimating equations (GEE) treating hospitals as the unit of repeated measures to control for hospital characteristics. The differences between different risk groups before and during the pandemic are compared using marginal effects to estimate the probability of occurrence.

Principal Findings: The study included 246,660 women who gave birth in 2019 and 2021. The proportion of high-risk pregnancies increased from 32.32% to 36.49% during the pandemic. Bivariate analysis revealed significant changes in C-section rates and healthcare resource utilization among both risk groups during the pandemic. High-risk pregnancies was found an increase in C-section rates from 43.70% to 44.56% , while among low-risk women, the rates rose from 30.67% to 31.57%. The induction rates for high-risk women increased from 10.55% to 15.35%, and for low-risk women from 11.22% to 16.74%. The length of stay and medical costs for both risk groups significantly increased during the pandemic. Multivariable analysis showed significant positive relationships between different risk groups and the outcomes. The pandemic was also positively associated to induction and medical costs but not significantly related to hospital stay length. Concerning the impact of the pandemic on the disparities between high-risk and low-risk groups, no significant differences were observed in C-section utilization and hospital stay length. However, significant differences were found in induction rates and medical costs (p<0.05).

Conclusions: Overall, the C-section rates and healthcare resource utilization exhibited significant differences before and during the pandemic, showing an upward trend. The differences in induction rates and medical costs persisted even after controlling for various variables. Maternal health risk influenced C-section utilization and healthcare resource consumption. Additionally, the pandemic narrowed the disparities in induction use between different risk groups while widening the disparities in medical costs. Induction of labor may bear a stiff price when considering increases in costs related to maternal morbidity and cesarean delivery. These findings may not only reflect the adaptations made in maternity units but also highlight the price that we need to pay when there is an external pandemic stresses on health systems. In summary, despite the mode of delivery remained stable during pandemic, it is worth noting the rising C-section rate among low-risk pregnancies and increased obstetric interventions. Unnecessary obstetric interventions, such as C-sections or inductions without medical indications, pose potential risks to maternal and neonatal health. Future policies should emphasize tailored risk management for pregnant women, encompassing not only continued perinatal care for high-risk individuals but also appropriate management for low-risk pregnancies to mitigate unnecessary obstetric interventions.