Institute of Health Economics and Health Care Management, Helmholtz Zentrum München, Neuherberg, Germany.
Institute for Medical Information Processing, Biometry, and Epidemiology (IBE), LMU Munich, Munich, Germany.
Adv Nutr. 2021 Oct 1;12(5):1957-1995. doi: 10.1093/advances/nmab028.
Simulation modeling can be useful to estimate the long-term health and economic impacts of population-based dietary policies. We conducted a systematic scoping review following the PRISMA-ScR (Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews) guideline to map and critically appraise economic evaluations of population-based dietary policies using simulation models. We searched Medline, Embase, and EconLit for studies published in English after 2005. Modeling studies were mapped based on model type, dietary policy, and nutritional target, and modeled risk factor-outcome pathways were analyzed. We included 56 studies comprising 136 model applications evaluating dietary policies in 21 countries. The policies most often assessed were reformulation (34/136), taxation (27/136), and labeling (20/136); the most common targets were salt/sodium (60/136), sugar-sweetened beverages (31/136), and fruit and vegetables (15/136). Model types included Markov-type (35/56), microsimulation (11/56), and comparative risk assessment (7/56) models. Overall, the key diet-related risk factors and health outcomes were modeled, but only 1 study included overall diet quality as a risk factor. Information about validation was only reported in 19 of 56 studies and few studies (14/56) analyzed the equity impacts of policies. Commonly included cost components were health sector (52/56) and public sector implementation costs (35/56), as opposed to private sector (18/56), lost productivity (11/56), and informal care costs (3/56). Most dietary policies (103/136) were evaluated as cost-saving independent of the applied costing perspective. An analysis of the main limitations reported by authors revealed that model validity, uncertainty of dietary effect estimates, and long-term intervention assumptions necessitate a careful interpretation of results. In conclusion, simulation modeling is widely applied in the economic evaluation of population-based dietary policies but rarely takes dietary complexity and the equity dimensions of policies into account. To increase relevance for policymakers and support diet-related disease prevention, economic effects beyond the health sector should be considered, and transparent conduct and reporting of model validation should be improved.
模拟建模可用于估计基于人群的饮食政策对长期健康和经济的影响。我们按照 PRISMA-ScR(系统评价和荟萃分析扩展的首选报告项目用于范围综述)指南进行了系统的范围综述,以绘制和批判性评估使用模拟模型进行的基于人群的饮食政策的经济评估。我们在 2005 年后在 Medline、Embase 和 EconLit 中搜索了英文发表的研究。根据模型类型、饮食政策和营养目标对建模研究进行了映射,并分析了建模的风险因素-结果途径。我们纳入了 56 项研究,其中包括 136 项评估 21 个国家饮食政策的模型应用。评估最多的政策是配方改革(34/136)、税收(27/136)和标签(20/136);最常见的目标是盐/钠(60/136)、含糖饮料(31/136)和水果和蔬菜(15/136)。模型类型包括马尔可夫型(35/56)、微观模拟(11/56)和比较风险评估(7/56)模型。总体而言,关键的饮食相关风险因素和健康结果都得到了建模,但只有 1 项研究将整体饮食质量作为一个风险因素。只有 19 项研究(56 项中的 19 项)报告了验证信息,很少有研究(56 项中的 14 项)分析了政策的公平影响。通常包括的成本组成部分是卫生部门(56/56)和公共部门实施成本(35/56),而不是私营部门(18/56)、生产力损失(11/56)和非正式护理成本(3/56)。大多数饮食政策(136/136)被评估为独立于应用成本视角的节省成本。对作者报告的主要限制的分析表明,模型有效性、饮食效果估计的不确定性以及长期干预假设需要仔细解释结果。总之,模拟建模广泛应用于基于人群的饮食政策的经济评估,但很少考虑饮食的复杂性和政策的公平性维度。为了提高决策者的相关性并支持与饮食相关的疾病预防,应考虑卫生部门以外的经济影响,并改进模型验证的透明性和报告。