文献速递流行病学与生物统计

利用口头尸检识别印度比哈尔邦死产相关因素:一项基于人群的研究

Identification of factors associated with stillbirth in the Indian state of Bihar using verbal autopsy: A population-based study.
PubMed 文献 · PLoS Med · 2017-08-01

该研究在印度比哈尔邦109,689户家庭中开展口头尸检访谈,使用SmartVA算法判定死因,分析死产相关因素。年化死产发生率为每1000例出生21.2例(95%CI 19.7至22.6),农村更高。研究识别出孕产妇贫血、产时发热、高血压及产科并发症等可能风险因素,并基于开放叙述提出卫生系统改进建议。

要点

  • 年化死产发生率为每1000例出生21.2例(95%CI 19.7至22.6),农村地区更高。
  • 仅6.1%的女性报告曾接受梅毒筛查试验,33%报告孕期未服用铁叶酸片。
  • 产科并发症和分娩时过度出血作为可能风险因素约占近30%的病例。

核心结论

  • 本研究是印度邦级层面利用口头尸检开展的大规模基于人群的死产评估之一,为调查死产可能风险因素及卫生系统所需的地面层面变革提供了详细见解。
  • 大多数死产可通过在孕前、孕期及产时和分娩时提供高质量、循证干预来预防,印度应以系统化、标准化方式记录死产,以弥合知识差距并采取适当行动减少死产。

研究设计与样本

  • 研究在印度比哈尔邦1,017个代表该邦的集群中、109,689户家庭样本(参与率87.1%)开展口头尸检访谈,覆盖2011年1月至2014年3月发生的包括死产在内的死亡。
  • 每次访谈使用人口健康指标研究联盟简版口头尸检问卷,死因通过SmartVA自动化算法判定。
  • 死产定义为妊娠≥28周、胎儿无任何生命迹象的胎儿死亡。

主要发现

  • 年化死产发生率为每1000例出生21.2例(95%CI 19.7至22.6),农村地区更高。
  • 共识别1,132例死产,其中686例(62.2%)为男孩,327例(29.7%)为第一胎,760例(68.9%)在卫生机构分娩。
  • 全部死产中54.5%估计为产前死产。
  • 仅6,161名女性(55.9%)报告至少1次产前保健就诊,33%的女性报告孕期未服用铁叶酸片。
  • 仅6.1%的女性报告曾接受排除梅毒的试验;34.2%的死产可能风险因素为不明原因。

关键数值与推荐

  • 在报告妊娠最后3个月至少1种并发症的其余65.8%女性中,贫血、产时发热和高血压等孕产妇状况占大多数并发症。
  • 产科并发症和分娩时过度出血作为可能风险因素约占近30%的病例。
  • 开放叙述中识别的5个主要主题中有3个与医疗保健提供者相关:缺乏及时关注、技能差(知识或实施)、不愿分娩死婴。

临床怎么用

  • 作者基于研究提出若干建议,可进一步加强印度新生儿行动计划(INAP)方法,以提高死产数据的质量和数量,避免这种不必要的生命损失。
  • 研究强调需要更好的熟练分娩照护,并利用病例研究记录的临床照护 breakdown 或错失机会来改善高质量熟练照护的提供。

局限性

  • 这些数据的主要局限性在于死产是基于妊娠期而非出生体重定义的,但这是在出生体重不可用或不准确的发展中国家环境中的多项研究采用的做法。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

BACKGROUND: India was estimated to have the largest numbers of stillbirths globally in 2015, and the Indian government has adopted a target of <10 stillbirths per 1,000 births by 2030 through the India Newborn Action Plan (INAP). The objective of this study was to use verbal autopsy interviews to examine factors associated with stillbirth in the Indian state of Bihar and make recommendations for the INAP to better inform the setting of priorities and actions to reduce stillbirths. METHODS AND FINDINGS: Verbal autopsy interviews were conducted for deaths including stillbirths that occurred from January 2011 to March 2014 in a sample of 109,689 households (87.1% participation) in 1,017 clusters representative of the state of Bihar. The Population Health Metrics Research Consortium shortened verbal autopsy questionnaire was used for each interview, and cause of death was assigned using the SmartVA automated algorithm. A stillbirth was defined as a foetal death with a gestation period of ≥28 weeks wherein the foetus did not show any sign of life. We report on the stillbirth epidemiology and present case studies from the qualitative data on the health provider interface that can be used to improve success of improved, skilled care at birth and delivery interventions. The annualised stillbirth incidence was 21.2 (95% CI 19.7 to 22.6) per 1,000 births, with it being higher in the rural areas. A total of 1,132 stillbirths were identified; 686 (62.2%) were boys, 327 (29.7%) were firstborn, and 760 (68.9%) were delivered at a health facility. Of all the stillbirths, 54.5% were estimated to be antepartum. Only 6,161 (55.9%) of the women reported at least 1 antenatal care visit, and 33% of the women reported not consuming the iron folic acid tablets during pregnancy. Significant differences were seen in delivery-related variables and associated maternal conditions based on the place of delivery and type of stillbirth. Only 6.1% of the women reported having undergone a test to rule out syphilis. For 34.2% of the stillbirths, the possible risk factor for stillbirth was unexplained. For the remaining 65.8% of the women who reported at least 1 complication during the last 3 months of pregnancy, maternal conditions including anaemia, fever during labour, and hypertension accounted for most of the complications. Of importance to note is that the maternal conditions overlapped quite significantly with the other possible underlying risk factors for stillbirth. Obstetrics complications and excessive bleeding during delivery contributed to nearly 30% of the cases as a possible risk factor for stillbirth, highlighting the need for better skilled care during delivery. Of the 5 major themes identified in open narratives, 3 were related to healthcare providers-lack of timely attention, poor skills (knowledge or implementation), and reluctance to deliver a dead baby. The case studies document the circumstances that highlight breakdowns in clinical care around the delivery or missed opportunities that can be used for improving the provision of quality skilled care. The main limitation of these data is that stillbirth is defined based on the gestation period and not based on birth weight; however, this is done in several studies from developing country settings in which birthweight is either not available or accurate. CONCLUSIONS: To our knowledge, this study is among the few large, population-based assessments of stillbirths using verbal autopsy at the state level in India. These findings provide detailed insight into investigating the possible risk factors for stillbirths, as well as insight into the ground-level changes that are needed within the health system to design and implement effective preventive and intervention policies to reduce the burden of stillbirths. As most of the stillbirths are preventable with high-quality, evidence-based interventions delivered before and during pregnancy and during labour and childbirth, it is imperative that with INAP in place, India aspires to document stillbirths in a systematic and standardised manner to bridge the knowledge gap for appropriate actions to reduce stillbirths. We have made several recommendations based on our study that could further strengthen the INAP approach to improve the quality and quantity of stillbirth data to avoid this needless loss of lives.

来源:PLoS Med,查看原文。版权归原作者所有。

引用

  1. Dandona R, Kumar GA, Kumar A, et al. Identification of factors associated with stillbirth in the Indian state of Bihar using verbal autopsy: A population-based study. PLoS Med. 2017. PMID: 28763449. doi:10.1371/journal.pmed.1002363

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。