Center for Global Health, Kenya Medical Research Institute, Kisumu, Kenya
Health Economics, Liverpool School of Tropical Medicine, Liverpool, UK.
BMJ Open. 2019 Sep 20;9(9):e033883. doi: 10.1136/bmjopen-2019-033883.
The objective of this analysis was to examine trends in malaria parasite prevalence and related socioeconomic inequalities in malaria indicators from 2006 to 2013 during a period of intensification of malaria control interventions in Siaya County, western Kenya.
Data were analysed from eight independent annual cross-sectional surveys from a combined sample of 19 315 individuals selected from 7253 households. Study setting was a health and demographic surveillance area of western Kenya. Data collected included demographic factors, household assets, fever and medication use, malaria parasitaemia by microscopy, insecticide-treated bed net (ITN) use and care-seeking behaviour. Households were classified into five socioeconomic status and dichotomised into poorest households (poorest 60%) and less poor households (richest 40%). Adjusted prevalence ratios (aPR) were calculated using a multivariate generalised linear model accounting for clustering and cox proportional hazard for pooled data assuming constant follow-up time.
Overall, malaria infection prevalence was 36.5% and was significantly higher among poorest individuals compared with the less poor (39.9% vs 33.5%, aPR=1.17; 95% CI 1.11 to 1.23) but no change in prevalence over time (trend p value <0.256). Care-seeking (61.1% vs 62.5%, aPR=0.99; 95% CI 0.95 to 1.03) and use of any medication were similar among the poorest and less poor. Poorest individuals were less likely to use Artemether-Lumefantrine or quinine for malaria treatment (18.8% vs 22.1%, aPR=0.81, 95% CI 0.72 to 0.91) while use of ITNs was lower among the poorest individuals compared with less poor (54.8% vs 57.9%; aPR=0.95; 95% CI 0.91 to 0.99), but the difference was negligible.
Despite attainment of equity in ITN use over time, socioeconomic inequalities still existed in the distribution of malaria. This might be due to a lower likelihood of treatment with an effective antimalarial and lower use of ITNs by poorest individuals. Additional strategies are necessary to reduce socioeconomic inequities in prevention and control of malaria in endemic areas in order to achieve universal health coverage and sustainable development goals.
本分析旨在探讨在肯尼亚西部 Siaya 县加强疟疾控制干预期间（2006 年至 2013 年）疟疾寄生虫流行率的变化趋势，以及疟疾指标的相关社会经济不平等。
从 7253 户家庭中抽取的 19315 名个体的 8 项独立年度横断面调查中收集数据。研究地点是肯尼亚西部的一个健康和人口监测区。收集的数据包括人口统计学因素、家庭资产、发热和用药情况、显微镜检查疟原虫寄生虫血症、经杀虫剂处理的蚊帐（ITN）使用和就医行为。家庭被分为五个社会经济地位组，并分为最贫困家庭（最贫困的 60%）和不太贫困家庭（最富有的 40%）。使用多元广义线性模型计算调整后的患病率比（aPR），该模型考虑了聚类，并对汇总数据使用 Cox 比例风险假设恒定随访时间。
总体而言，疟疾感染率为 36.5%，最贫困个体的感染率明显高于不太贫困个体（39.9%比 33.5%，aPR=1.17；95%CI 1.11 至 1.23），但流行率随时间没有变化（趋势 p 值<0.256）。最贫困和不太贫困个体的就医（61.1%比 62.5%，aPR=0.99；95%CI 0.95 至 1.03）和使用任何药物的情况相似。最贫困个体使用青蒿琥酯-甲氟喹或奎宁治疗疟疾的可能性较低（18.8%比 22.1%，aPR=0.81，95%CI 0.72 至 0.91），而最贫困个体使用 ITN 的比例低于不太贫困个体（54.8%比 57.9%；aPR=0.95；95%CI 0.91 至 0.99），但差异可以忽略不计。
尽管随着时间的推移，在 ITN 使用方面实现了公平，但疟疾的分布仍存在社会经济不平等。这可能是由于最贫困个体接受有效抗疟药物治疗的可能性较低，以及使用 ITN 的可能性较低。在疟疾流行地区，为了实现全民健康覆盖和可持续发展目标，需要采取额外的策略来减少预防和控制疟疾方面的社会经济不平等。
