Kenya Malaria Indicator Survey 2015 - Kenya

Abstract
---------------------------

Malaria is a significant public health problem in Kenya. More than 70 percent of the population live in malaria risk areas, including those most vulnerable to the disease: children and pregnant women. In the last 5 years, tremendous efforts have been made to combat malaria with prevention and treatment interventions such as mass and routine mosquito net distribution programs to attain universal coverage, intermittent preventive treatment for malaria during pregnancy, and parasitological diagnosis and management of malaria cases.

The Kenya Malaria Indicator Survey is one of the key performance monitoring tools that are periodically used to provide an in-depth assessment of malaria control over time. This is the third survey undertaken; the previous two were in 2007 and in 2010.

This report provides information on the performance of the key malaria control interventions as experienced by communities across the country. The results contained in this report are crucial to the evaluation of interventions, planning for the future, and understanding the dynamics that affect malaria control programme efforts.

The report shows that with concerted efforts and effective partnerships we can reduce the impact of malaria in the country. A clear indication of this is the overall reduction in malaria prevalence in Kenya as compared with the 2010 survey results. Other key results include the increased uptake in ownership and use of nets as well as improved availability of recommended medicines for the treatment of malaria. The survey results are similar to those for malaria control indicators reported by the 2014 Kenya Demographic Health Survey.

The report has come at an opportune time, and the government urges stakeholders at all levels to embrace the report, assess its implications on malaria control, and chart the way forward. The report will form the platform for our malaria control strategy in the coming years. It is clear that with continued investments we can make substantial progress toward the objective of eliminating communicable diseases, and thus the Ministry of Health is committed to further reducing the malaria burden in the coming years. A malaria-free Kenya is possible.

Geographic coverage
---------------------------

It was a National survey

The survey used National Sample Survey and Evaluation Programme (NASSEP) V sampling frame, the frame was nationally representative and was developed by the KNBS after the 2009 Census.

Kind of data
---------------------------

Sample survey data [ssd]

Sampling procedure
---------------------------

The estimates from a sample survey are affected by two types of errors: nonsampling errors and sampling errors. Nonsampling errors are the results of mistakes made in implementing data collection and data processing, such as failure to locate and interview the correct household, misunderstanding of the questions by either the interviewer or the respondent, and data entry errors. Although numerous efforts were made during the implementation of the 2015 Kenya Malaria Indicator Survey (KMIS) to minimize this type of error, nonsampling errors are impossible to avoid and difficult to evaluate statistically.

Sampling errors, on the other hand, can be evaluated statistically. The sample of respondents selected in the 2015 KMIS is only one of many samples that could have been selected from the same population, using the same design and expected size. Each of these samples would yield results that differ somewhat from the results of the actual sample selected. Sampling errors are a measure of the variability among all possible samples. Although the degree of variability is not known exactly, it can be estimated from the survey results.

Sampling error is usually measured in terms of the standard error for a particular statistic (mean, percentage, etc.), which is the square root of the variance. The standard error can be used to calculate confidence intervals within which the true value for the population can reasonably be assumed to fall. For example, for any given statistic calculated from a sample survey, the value of that statistic will fall within a range of plus or minus two times the standard error of that statistic in 95 percent of all possible samples of identical size and design.

If the sample of respondents had been selected as a simple random sample, it would have been possible to use straightforward formulas for calculating sampling errors. However, the 2015 KMIS sample is the result of a multi-stage stratified design, and, consequently, it was necessary to use more complex formulas. Sampling errors are computed in either ISSA or SAS, using programs developed by ICF International. These programs use the Taylor linearization method of variance estimation for survey estimates that are means, proportions, or ratios.

The Taylor linearization method treats any percentage or average as a ratio estimate, r = y/x, where y represents the total sample value for variable y, and x represents the total number of cases in the group or subgroup under consideration. The variance of r is computed using a formula, with the standard error being the square root of the variance:

The 2015 KMIS sample was designed to produce estimates for key indicators for the country as a whole, for urban and rural areas separately, and for each of the malaria epidemiologic zones: highland epidemic; lake endemic; coast endemic; semi-arid, seasonal; and low risk.

The sampling frame used for the 2015 KMIS was the Fifth National Sample Survey and Evaluation Program (NASSEP V) master sampling frame, which is created and maintained by KNBS for household-based surveys in Kenya. Development of the frame started in 2012. It contains a list of all enumeration areas (EAs) created for the 2009 census and covers the entire country. The frame is split into four equal subsamples, from one of which the 2015 KMIS sample was drawn. Kenya is administratively divided into 47 counties, created in the 2010 Constitution; within the frame, each county is stratified into urban and rural areas and is contained within one or two of the five malaria endemic zones.

The survey used a two-stage stratified cluster sampling design. In the first stage, 246 clusters (131 rural, 115 urban) were selected with equal probability from the NASSEP V. The second stage involved selection of a uniform sample of 30 households using systematic sampling from each of the selected clusters. Prior to household selection, all the clusters were updated by KNBS. This entailed undertaking a household listing in each of the selected clusters in order to update the list of residential households within it. As part of the listing, KNBS also updated the necessary maps and recorded the geographic coordinates of each cluster. Only selected households were interviewed, and replacement of nonresponding households was not allowed.

Mode of data collection
---------------------------

Computer Assisted Personal Interview [capi]

Cleaning operations
---------------------------

The 2015 KMIS used ASUS Transformer T100 tablet computers with data entry programs developed in CSPro by The DHS Program at ICF International. Tablets were Bluetooth-enabled to facilitate the electronic transfer of household assignment among field team members and the transfer of completed questionnaires to team supervisors for transfer to the central office. Code division multiple access wireless technology via Internet File Streaming System (IFSS) developed by The DHS Program was used to transfer encrypted data from the field to the central office in Nairobi. Each tablet was fitted with a micro-SD card for encrypted data back-up.

To facilitate communication and monitoring, each field worker was assigned a unique identification number. In the central office, data received from the field team supervisors’ tablets were registered and checked against any inconsistencies and outliers. Data editing and cleaning included range checks and structural and internal consistency checks. Any anomalies were communicated to the respective team through their team supervisor. The corrected results were re-sent to the central processing office.

Response rate
---------------------------

A total of 7,313 households were selected for the study, of which 6,667 were occupied at the time of fieldwork. Of these, 6,481 households were successfully interviewed, yielding an overall household response rate of 97 percent.

疟疾在肯尼亚是一项重大的公共卫生问题。超过70%的人口居住在疟疾风险区域，其中包括那些最易受疾病侵害的人群：儿童和孕妇。在过去五年中，为应对疟疾，我们付出了巨大的努力，包括大规模和常规蚊帐分发计划以实现全面覆盖、孕期间歇性预防治疗疟疾以及疟疾病例的寄生虫学诊断和管理等预防和治疗干预措施。

肯尼亚疟疾指标调查是定期用于提供疟疾控制长期评估的关键绩效监测工具之一。这是第三次进行的调查；前两次分别是在2007年和2010年。

本报告提供了关于全国范围内社区体验的关键疟疾控制干预措施性能的信息。本报告中的结果对于评估干预措施、规划未来以及理解影响疟疾控制项目努力的动态至关重要。

报告显示，通过齐心协力和有效的合作伙伴关系，我们可以减少该国的疟疾影响。这一明确迹象是，与2010年调查结果相比，肯尼亚疟疾的总体患病率有所下降。其他关键结果包括蚊帐拥有率和使用率的提高以及治疗疟疾推荐药物的可获得性改善。调查结果与2014年肯尼亚人口健康调查报告的疟疾控制指标结果相似。

报告适时发布，政府敦促各层面的利益相关者接受报告，评估其对疟疾控制的影响，并规划前进的道路。该报告将成为我们未来几年疟疾控制策略的平台。显而易见，通过持续投资，我们可以朝着消除传染病的既定目标取得实质性进展，因此卫生部致力于在未来的几年内进一步减少疟疾负担。一个无疟疾的肯尼亚是可能的。

地理覆盖范围
---------------------------
这是一次全国性调查

调查使用了国家抽样调查和评估计划（NASSEP）V抽样框架，该框架在全国范围内具有代表性，由KNBS在2009年人口普查后制定。

数据类型
---------------------------
样本调查数据 [ssd]

抽样程序
---------------------------
样本调查的估计受到两种类型误差的影响：非抽样误差和抽样误差。非抽样误差是数据收集和数据处理中出现的错误的结果，例如未能找到并采访正确的家庭、采访者或受访者对问题的误解，以及数据输入错误。尽管在实施2015年肯尼亚疟疾指标调查（KMIS）期间做出了众多努力以最大限度地减少此类错误，但非抽样误差是不可避免的，并且难以从统计上进行评估。

另一方面，抽样误差可以从统计上进行评估。2015年KMIS中选择的受访者样本只是从同一人口中可以选出的许多样本之一，使用相同的设计和预期规模。这些样本中的每一个都会产生与实际选定样本的结果有所不同的一些结果。抽样误差是衡量所有可能样本之间变异性的一个指标。尽管变异的程度并不完全清楚，但它可以从调查结果中估计出来。

抽样误差通常以特定统计量（均值、百分比等）的标准误差来衡量，这是方差的平方根。标准误差可用于计算置信区间，其中可以合理地假设真实值落在该区间内。例如，从任何给定的样本调查计算出的任何统计量，该统计量的值将在95%的所有可能样本的相同大小和设计中，该统计量的标准误差的两倍范围内。

如果受访者样本被选为简单随机样本，则可以使用简单的公式来计算抽样误差。然而，2015年KMIS样本是多阶段分层设计的产物，因此有必要使用更复杂的公式。抽样误差是在ISSA或SAS中计算的，使用ICF国际开发的程序。这些程序使用泰勒线性化方法对调查估计进行方差估计，这些估计是均值、比例或比率。

泰勒线性化方法将任何百分比或平均值视为一个比率估计，r = y/x，其中y代表变量y的总样本值，x代表考虑的组或子组中的总案例数。r的方差使用公式计算，标准误差是方差的平方根：

2015年KMIS样本旨在产生针对整个国家的关键指标估计值，针对城市和农村地区分别，以及针对每个疟疾流行病学区域：高山流行区；湖地区内疫区；海岸地区内疫区；半干旱，季节性；和低风险区。

2015年KMIS使用的抽样框架是第五个国家抽样调查和评估计划（NASSEP V）主抽样框架，该框架由KNBS创建和维护，用于肯尼亚基于家庭的调查。该框架的开发始于2012年。它包含为2009年人口普查创建的所有调查区域的列表，并覆盖整个国家。该框架分为四个相等的子样本，其中之一用于抽取2015年KMIS样本。肯尼亚在2010年宪法中被行政划分为47个县；在框架中，每个县被分为城市和农村地区，并包含在五个疟疾疫区中的任何一个或两个中。

调查使用了两阶段分层聚类抽样设计。在第一阶段，从NASSEP V中随机选择了246个聚类（131个农村，115个城市）。在第二阶段，从每个选定的聚类中使用了系统抽样选择了一个均匀的30户家庭样本。在家庭选择之前，KNBS对所有聚类进行了更新。这包括在每个选定的聚类中进行家庭清单，以更新其内的住宅家庭清单。作为清单的一部分，KNBS还更新了必要的地图并记录了每个聚类的地理坐标。只有选定的家庭被采访，不允许替换未响应的家庭。

数据收集方式
---------------------------
计算机辅助个人采访 [capi]

数据清洗操作
---------------------------
2015年KMIS使用了ASUS Transformer T100平板电脑，数据输入程序由ICF国际在CSPro中开发。平板电脑具有蓝牙功能，以方便在实地团队成员之间电子传输家庭分配以及在将完成的问卷传输到团队主管处，再由其转至中央办公室。使用由DHS计划开发的互联网文件流系统（IFSS）的码分多址无线技术将加密数据从现场传输到内罗毕的中央办公室。每个平板电脑都配备了微型SD卡，用于加密数据备份。

为了便于沟通和监控，每位实地工作人员都被分配了一个唯一的识别号码。在中央办公室，从实地团队主管的平板电脑接收到的数据被注册并检查是否存在任何不一致和异常。数据编辑和清洗包括范围检查和结构性和内部一致性检查。任何异常情况都会通过其团队主管通知相应的团队。更正后的结果会重新发送到中央处理办公室。

响应率
---------------------------
总共选择了7,313个家庭进行研究，其中在实地工作时，6,667个家庭被占用。在这些家庭中，6,481个家庭成功接受了采访，从而产生了97%的总体家庭响应率。