Assessing the ability of health information systems in hospitals to support evidence-informed decisions in Kenya
Assessing the ability of health information systems in hospitals to support evidence-informed decisions in Kenya
复制标题
DOI：
10.3402/gha.v7.24859
复制
发表时间：
2014-01-01
影响因子：
2.6
通讯作者：
English, Mike
中科院分区：
文献类型：
--
作者：
Kihuba, Elesban;Gathara, David;English, Mike
背景：医院管理信息系统（HMIS）是国家卫生信息系统（HIS）的关键组成部分，具体的政策文件阐明了医院管理部门支持肯尼亚信息生成所需采取的行动。我们对肯尼亚医院内数据生成和报告的核心功能进行了评估，以促进国家报告的解释，并就需要改进的关键领域提供指导，以支持决策中的数据使用。 设计：该调查是在肯尼亚 22 家医院进行的一项横断面、整群样本研究。统计分析是描述性的，并进行了聚类调整。结果：大多数 HMIS 部门都遵守正式指南来制定部门计划。然而，只有少数 (3/22) 在调查前 12 个月内进行了数据质量审核。医院总收入的平均 3%（范围 1-8%）分配给 HMIS 部门。大约一半的记录官员职位已被填补，大约一半 (13/22) 的医院实施了某种形式的电子健康记录，主要侧重于改善患者账单，与地区 HIS 无关。手动患者登记的完整性各不相同，妇幼保健诊所、妇产科和儿科病房的手动患者登记完整性分别为 90% (95% CI 80.1-99.3%)、75.8% (95% CI 68.7-82.8%) 和 58% (95% CI 50.4-65.1%)。生命事件通知率较低，记录的新生儿死亡、婴儿死亡和活产分别为 25.7%、42.6% 和 71.3%。医院常规报告显示，活产数略有多报，新鲜死产和新生儿死亡少报。结论：研究结果表明，HMIS 未提供高质量数据。数据质量保障、监管支持、信息通信技术应用的数据基础设施、人力资源、财力资源、整合等方面存在较大制约。
Background: Hospital management information systems (HMIS) is a key component of national health information systems (HIS), and actions required of hospital management to support information generation in Kenya are articulated in specific policy documents. We conducted an evaluation of core functions of data generation and reporting within hospitals in Kenya to facilitate interpretation of national reports and to provide guidance on key areas requiring improvement to support data use in decision making.Design: The survey was a cross-sectional, cluster sample study conducted in 22 hospitals in Kenya. The statistical analysis was descriptive with adjustment for clustering.Results: Most of the HMIS departments complied with formal guidance to develop departmental plans. However, only a few (3/22) had carried out a data quality audit in the 12 months prior to the survey. On average 3% (range 1-8%) of the total hospital income was allocated to the HMIS departments. About half of the records officer positions were filled and about half (13/22) of hospitals had implemented some form of electronic health record largely focused on improving patient billing and not linked to the district HIS. Completeness of manual patient registers varied, being 90% (95% CI 80.1-99.3%), 75.8% (95% CI 68.7-82.8%), and 58% (95% CI 50.4-65.1%) in maternal child health clinic, maternity, and pediatric wards, respectively. Vital events notification rates were low with 25.7, 42.6, and 71.3% of neonatal deaths, infant deaths, and live births recorded, respectively. Routine hospital reports suggested slight over-reporting of live births and under-reporting of fresh stillbirths and neonatal deaths.Conclusions: Study findings indicate that the HMIS does not deliver quality data. Significant constraints exist in data quality assurance, supervisory support, data infrastructure in respect to information and communications technology application, human resources, financial resources, and integration.