Carabin H, Guyatt H, Engels D
The Wellcome Trust Centre for the Epidemiology of Infectious Disease, Department of Zoology, University of Oxford, Oxford, UK.
Trop Med Int Health. 2000 Mar;5(3):192-202. doi: 10.1046/j.1365-3156.2000.00530.x.
To assess the cost-effectiveness of three alternative screening strategies in delivering treatment to patients presenting with symptoms suggestive of S. mansoni, using data collected from 17 PHCCs in the Rusizi Plain, Burundi. The three strategies were: (1) screening all symptomatic patients using a Kato-Katz smear and treating only the ones found positive; (2) treating all symptomatic patients or (3) treating only those presenting with symptoms of severe diarrhoea (blood in stool).
The database consisted of 41 051 visits of symptomatic patients to the 17 PHCCs during 1990. Effectiveness was measured as the number of infected patients treated (gold standard: 25-mg Kato-Katz smear). Cost-effectiveness ratios (CERs) (cost in US$ per infected patient treated) were used to compare the control strategies. Sensitivity analysis was performed to assess the effect of drug price and prevalence of infection on the CERs.
The overall prevalence of S. mansoni infection was 9.5% (95% CI = 9.1%, 9.9%). Treating only those with severe symptoms of infection would have left 92% of infected patients untreated. Performing a Kato-Katz smear to confirm the diagnosis was more cost-effective than treating all symptomatic patients on a presumptive basis, with CERs of 4.2 US$ and 12.43 US$ per infected person treated, respectively. Sensitivity analysis showed that, for a cost of 0.99 US$ per dose of praziquantel, the Kato-Katz option remained the most cost-effective approach for prevalences under 76%. For a drug price of 0.21 US$ per dose, both strategies would have become equivalent. The latter value varied extensively per PHCC (range 0.17 US$ to 0.51 US$), due to the different prevalences of infection (range 0.5% to 34.3%).
We found that using severe diarrhoea as an indicator for infection was not appropriate, and that screening symptomatic patients with the Kato-Katz method remained the most cost-effective approach in the given setting. The CER of treating all symptomatic patients on a presumptive basis depended very much on the drug price and the endemic level.
利用从布隆迪鲁济济平原17个初级卫生保健中心收集的数据，评估三种替代筛查策略在为出现曼氏血吸虫病症状的患者提供治疗方面的成本效益。这三种策略分别是：（1）使用加藤厚涂片法对所有有症状的患者进行筛查，仅治疗检测呈阳性的患者；（2）治疗所有有症状的患者；（3）仅治疗那些出现严重腹泻（粪便带血）症状的患者。
数据库包含1990年期间17个初级卫生保健中心41051例有症状患者的就诊记录。有效性以接受治疗的感染患者数量来衡量（金标准：25毫克加藤厚涂片法）。成本效益比（每治疗一名感染患者的成本，单位为美元）用于比较对照策略。进行敏感性分析以评估药品价格和感染率对成本效益比的影响。
曼氏血吸虫病感染的总体患病率为9.5%（95%置信区间 = 9.1%，9.9%）。仅治疗那些有严重感染症状的患者会导致92%的感染患者未得到治疗。采用加藤厚涂片法确诊比基于推测治疗所有有症状的患者更具成本效益，每治疗一名感染患者的成本效益比分别为4.2美元和12.43美元。敏感性分析表明，对于每剂吡喹酮成本为0.99美元的情况，加藤厚涂片法选项在患病率低于76%时仍是最具成本效益的方法。对于每剂药品价格为0.21美元的情况，两种策略将变得等效。由于不同的感染率（范围为0.5%至34.3%），后一个值在每个初级卫生保健中心差异很大（范围为0.17美元至0.51美元）。
我们发现将严重腹泻作为感染指标并不合适，在给定环境下，使用加藤厚涂片法筛查有症状的患者仍是最具成本效益的方法。基于推测治疗所有有症状患者的成本效益比在很大程度上取决于药品价格和地方流行程度。
