Demarteau Nadia, Morhason-Bello Imran O, Akinwunmi Babatunde, Adewole Isaac F
Health Economics, Global Vaccines Development, GlaxoSmithKline Vaccines, Avenue Fleming 20 B-1300, Wavre, Belgium.
BMC Cancer. 2014 May 24;14:365. doi: 10.1186/1471-2407-14-365.
This study aims to assess the most efficient combinations of vaccination and screening coverage for the prevention of cervical cancer (CC) at different levels of expenditure in Nigeria.
An optimization procedure, using a linear programming approach and requiring the use of two models (an evaluation and an optimization model), was developed. The evaluation model, a Markov model, estimated the annual number of CC cases at steady state in a population of 100,000 women for four alternative strategies: screening only; vaccination only; screening and vaccination; and no prevention. The results of the Markov model for each scenario were used as inputs to the optimization model determining the optimal proportion of the population to receive screening and/or vaccination under different scenarios. The scenarios varied by available budget, maximum screening and vaccination coverage, and overall reachable population.
In the base-case optimization model analyses, with a coverage constraint of 20% for one lifetime screening, 95% for vaccination and a budget constraint of $1 per woman per year to minimize CC incidence, the optimal mix of prevention strategies would result in a reduction of CC incidence of 31% (3-dose vaccination available) or 46% (2-dose vaccination available) compared with CC incidence pre-vaccination. With a 3-dose vaccination schedule, the optimal combination of the different strategies across the population would be 20% screening alone, 39% vaccination alone and 41% with no prevention, while with a 2-dose vaccination schedule the optimal combination would be 71% vaccination alone, and 29% with no prevention. Sensitivity analyses indicated that the results are sensitive to the constraints included in the optimization model as well as the cervical intraepithelial neoplasia (CIN) and CC treatment cost.
The results of the optimization model indicate that, in Nigeria, the most efficient allocation of a limited budget would be to invest in both vaccination and screening with a 3-dose vaccination schedule, and in vaccination alone before implementing a screening program with a 2-dose vaccination schedule.
本研究旨在评估在尼日利亚不同支出水平下,预防宫颈癌(CC)的疫苗接种和筛查覆盖率的最有效组合。
开发了一种优化程序,采用线性规划方法,需要使用两个模型(评估模型和优化模型)。评估模型是一个马尔可夫模型,针对四种替代策略估计了100,000名女性群体在稳态下每年的CC病例数:仅筛查;仅接种疫苗;筛查和接种疫苗;以及不进行预防。马尔可夫模型针对每种情况的结果被用作优化模型的输入,以确定在不同情况下接受筛查和/或接种疫苗的人群的最佳比例。这些情况因可用预算、最大筛查和疫苗接种覆盖率以及总体可及人群而异。
在基础案例优化模型分析中,对于一生一次的筛查覆盖率限制为20%,疫苗接种覆盖率限制为95%,且每位女性每年预算限制为1美元以最小化CC发病率,与接种疫苗前的CC发病率相比,预防策略的最佳组合将使CC发病率降低31%(有3剂次疫苗可用)或46%(有2剂次疫苗可用)。采用3剂次疫苗接种方案时,不同策略在人群中的最佳组合为20%仅筛查、39%仅接种疫苗和41%不进行预防,而采用2剂次疫苗接种方案时,最佳组合为71%仅接种疫苗和29%不进行预防。敏感性分析表明,结果对优化模型中包含的限制条件以及宫颈上皮内瘤变(CIN)和CC治疗成本敏感。
优化模型的结果表明,在尼日利亚,有限预算的最有效分配方式是采用3剂次疫苗接种方案时同时投资于疫苗接种和筛查,以及在采用2剂次疫苗接种方案实施筛查计划之前仅投资于疫苗接种。