Estimating national-level measles case-fatality ratios in low-income and middle-income countries: an updated systematic review and modelling study.
Estimating national-level measles case-fatality ratios in low-income and middle-income countries: an updated systematic review and modelling study.
复制标题
DOI：
10.1016/s2214-109x(23)00043-8
复制
发表时间：
2023-04
影响因子：
34.3
通讯作者：
Portnoy, Allison
中科院分区：
文献类型：
--
作者：
Sbarra, Alyssa N.;Mosser, Jonathan F.;Jit, Mark;Ferrari, Matthew;Ramshaw, Rebecca E.;O'Connor, Patrick;Krause, L. Kendall;Rogowski, Emma L. B.;Portnoy, Allison
为了了解当前的 麻疹死亡负担，并减轻未来的负担，对麻疹病例死亡人数进行可靠的估计至关重要。对特定于年龄、地点和时间的麻疹病死率(CFRs)的估计对于捕捉潜在人口因素的变化至关重要，如疫苗接种覆盖率和麻疹发病率，这些因素导致CFRs的增加或减少。在这项研究中，我们更新了对麻疹CFRs的估计，方法是在之前的系统综述的基础上进行扩展，并实施Meta回归模型。我们的目标是使用所有可用的信息来按国家、年龄和年份估计低收入和中等收入国家(LMIC)的麻疹病死率。对于这项系统回顾和元回归模型研究，我们在2020年12月31日检索PubMed上发表的1980年1月1日至2020年12月31日期间LMICs中截至2019年12月31日的麻疹病例和死亡病例的所有可用原始数据。我们纳入了一些研究，这些研究以前的系统评价已经包括或包含了来自基于医院、基于社区或基于监测的报告的麻疹病例和死亡的主要数据，包括疫情调查。我们排除了非人类研究，或只报告了仅关于非主要人群或受限人群(例如艾滋病毒携带者)或长期麻疹死亡率(例如亚急性硬化性全脑炎死亡)的数据，以及不包括国家一级数据或有关麻疹病例和死亡的相关信息的研究，或者是针对高收入国家的研究。我们从符合纳入和排除标准的研究中提取了有关麻疹病例和麻疹死亡的汇总数据。使用这些数据和一组与麻疹CFRs相关的协变量，我们实施了贝叶斯Meta回归模型，以产生按地点和年龄组从1990年到2019年的麻疹CFRs估计。这项研究没有在普罗斯佩罗或其他地方注册。我们确定了2705条记录，其中208条来源包含关于低收入国家麻疹病例和麻疹死亡的信息，并被纳入审查范围。1990至2019年，社区和医院的病死率大幅下降，各年龄段的模式一致。对于0-34岁的人群，我们估计2019年社区环境的平均病死率为1.32%(95%不确定区间[UI]1·28-1·36)，医院环境的平均病死率为5.35%(5.08-5.)。我们估计2019年社区环境中1岁以下儿童的CFR为3.03%(UI 2.89-3.16)，1-4岁儿童为1.63%(1.58-1.68)，5-9岁儿童为0.84%(0.80-0.87)，10-14岁儿童为0.67%(0.-0.70)。尽管CFR在1990年至2019年期间有所下降，但不同地点和年龄的CFR仍然存在很大的异质性。这种系统性审查的一个局限性是，我们无法评估特定人群中的麻疹病死率，例如难民和国内流离失所者。我们最新的方法框架和估计数可用于评估麻疹控制和疫苗接种方案在减少可预防的麻疹死亡负担方面的效果。比尔和梅琳达·盖茨基金会、疫苗联盟GAVI和美国国立卫生研究院。
To understand the current measles mortality burden, and to mitigate the future burden, it is crucial to have robust estimates of measles case fatalities. Estimates of measles case–fatality ratios (CFRs) that are specific to age, location, and time are essential to capture variations in underlying population-level factors, such as vaccination coverage and measles incidence, which contribute to increases or decreases in CFRs. In this study, we updated estimates of measles CFRs by expanding upon previous systematic reviews and implementing a meta-regression model. Our objective was to use all information available to estimate measles CFRs in low-income and middle-income countries (LMICs) by country, age, and year. For this systematic review and meta-regression modelling study, we searched PubMed on Dec 31, 2020 for all available primary data published from Jan 1, 1980 to Dec 31, 2020, on measles cases and fatalities occurring up to Dec 31, 2019 in LMICs. We included studies that previous systematic reviews had included or which contained primary data on measles cases and deaths from hospital-based, community-based, or surveillance-based reports, including outbreak investigations. We excluded studies that were not in humans, or reported only data that were only non-primary, or on restricted populations (eg, people living with HIV), or on long-term measles mortality (eg, death from subacute sclerosing panencephalitis), and studies that did not include country-level data or relevant information on measles cases and deaths, or were for a high-income country. We extracted summary data on measles cases and measles deaths from studies that fitted our inclusion and exclusion criteria. Using these data and a suite of covariates related to measles CFRs, we implemented a Bayesian meta-regression model to produce estimates of measles CFRs from 1990 to 2019 by location and age group. This study was not registered with PROSPERO or otherwise. We identified 2705 records, of which 208 sources contained information on both measles cases and measles deaths in LMICS and were included in the review. Between 1990 and 2019, CFRs substantially decreased in both community-based and hospital-based settings, with consistent patterns across age groups. For people aged 0–34 years, we estimated a mean CFR for 2019 of 1·32% (95% uncertainty interval [UI] 1·28–1·36) among community-based settings and 5·35% (5·08–5·64) among hospital-based settings. We estimated the 2019 CFR in community-based settings to be 3·03% (UI 2·89–3·16) for those younger than 1 year, 1·63% (1·58–1·68) for age 1–4 years, 0·84% (0·80–0·87) for age 5–9 years, and 0·67% (0·64–0·70) for age 10–14 years. Although CFRs have declined between 1990 and 2019, there are still large heterogeneities across locations and ages. One limitation of this systematic review is that we were unable to assess measles CFR among particular populations, such as refugees and internally displaced people. Our updated methodological framework and estimates could be used to evaluate the effect of measles control and vaccination programmes on reducing the preventable measles mortality burden. Bill & Melinda Gates Foundation; Gavi, the Vaccine Alliance; and the US National Institutes of Health.