Background: The burden of stroke is increasing in India; stroke is now the fourth leading cause of death and the fifth leading cause of disability. Previous research suggests that the incidence of stroke in India ranges between 105 and 152/100,000 people per year. However, there is a paucity of available data and a lack of uniform methods across published studies.

Aim: To identify high-quality prospective studies reporting the epidemiology of stroke in India.

Summary of review: A search strategy was modified from the Cochrane Stroke Strategy and adapted for a range of bibliographic databases from January 1997 to August 2020. From 7,717 identified records, nine studies were selected for inclusion; three population-based registries, a further three population-based registries also using community-based ascertainment and three community-based door-to-door surveys. Studies represented the four cities of Mumbai, Trivandrum, Ludhiana, Kolkata, the state of Punjab and 12 villages of Baruipur in the state of West Bengal. The total population denominator was 22,479,509 and 11,654 (mean 1,294 SD 1,710) people were identified with incident stroke. Crude incidence of stroke ranged from 108 to 172/100,000 people per year, crude prevalence from 26 to 757/100,000 people per year and one-month case fatality rates from 18% to 42%.

Conclusions: Further high-quality evidence is needed across India to guide stroke policy and inform the development and organisation of stroke services. Future researchers should consider the World Health Organisation STEPwise approach to Surveillance (STEPS) framework, including longitudinal data collection, the inclusion of census population data and a combination of hospital-registry and comprehensive community ascertainment strategies to ensure complete stroke identification.

背景：印度的中风负担正在增加；中风现在是第四大死亡原因和第五大残疾原因。先前的研究表明，印度每年的中风发病率在 105 到 152/100,000 人之间。然而，在已发表的研究中缺乏可用的数据并且缺乏统一的方法。

目的：确定报告印度中风流行病学的高质量前瞻性研究。

综述总结：从 1997 年 1 月至 2020 年 8 月，从 Cochrane 卒中策略修改了检索策略，并适用于一系列书目数据库。从 7,717 条已确定的记录中，选择了 9 项研究；三个基于人口的登记处，另外三个基于人口的登记处也使用基于社区的确定和三个基于社区的挨家挨户调查。研究代表了孟买、特里凡得琅、卢迪亚纳、加尔各答、旁遮普邦四个城市和西孟加拉邦巴鲁伊普尔的 12 个村庄。总人口分母为 22,479,509 人，11,654 人（平均 1,294 SD 1,710）被确定为卒中事件。中风的粗发病率范围为每年 108 至 172/100,000 人，粗发病率范围为 26 至 757/100，

结论：印度需要更多高质量的证据来指导卒中政策，并为卒中服务的发展和组织提供信息。未来的研究人员应考虑世界卫生组织逐步监测方法 (STEPS) 框架，包括纵向数据收集、人口普查数据的纳入以及医院登记和综合社区确定策略的结合，以确保完整的中风识别。