Applying quantitative health impact assessment for evidence-based policy-making in public health

Abstract

The prevention of noncommunicable diseases (NCDs) remains a major challenge to public health. Since unhealthy diets, tobacco use, the harmful use of alcohol as well as low physical activity are the key drivers of NCDs, a range of public health interventions exist to target these lifestyle related risk factors. However, the effectiveness of these interventions depends on underlying population-specific characteristics such as the age structure, the existing risk-factor exposure, as well as the incidence-prevalence-mortality profile of the respective chronic diseases. In order to identify the best course of action among competing interventions before implementing them, prospective health impact assessments (HIAs) can be applied to quantify and compare the potential future effect of an intervention on population health. Therefore, the overall aim of this dissertation was to assess the impact public health interventions targeting lifestyle risk factors have on NCDs at the population-level in the long run, in order to contribute to evidence-based public health decision-making.

In the context of this dissertation, four HIAs of policies were conducted, which seem to be especially promising namely (i) a tax on processed meat in Germany, (ii) a tax on saturated fat in European countries, (iii) the removal of the value-added tax (VAT) and a subsidy on fruits and vegetables in Germany, and (iv) a rolling out of a digital intervention promoting physical activity among older adults at the national level in Germany while accounting for differential effectiveness across education groups. The simulations were conducted with the software tool DYNAMO-HIA, which dynamically projects deaths, ischemic heart disease (IHD), stroke, diabetes and/or colorectal cancer cases under the different policies over a projection period of 10 years.

Cumulated over the 10-year projection period, a tax on processed meat would lead to 9,300 less male deaths and 4,500 less female deaths under the lowest tax of 4%, up to 76,700 less male deaths and 37,100 less female deaths under the highest tax of 33.3%. A tax on saturated fat would reduce the prevalent IHD cases in projection year 10 by a minimum of 500 among males and 300 among females in Denmark, up to a maximum of 5,600 and 4,000 among males and females in the UK. The removal of the VAT on fruits and vegetables would prevent an estimated 13,960 deaths, cumulated over the 10-year projection period, while a 20 % subsidy on fruits and vegetables would avert 54,880 deaths. Under the physical activity intervention, approximately 3,589–5,829 incident IHD, stroke and diabetes cases could be avoided among males aged ≥55 years over a 10-year projection period, as well as 4,381–7,163 disease cases among females aged ≥55 years. Strikingly, the highest reduction for males would be achieved under the intervention that is most effective for those with a high education, whereas for females, the highest reduction would be achieved under the intervention that is most effective amongst those with a low level of education.

Findings of the four quantitative HIAs conducted in the context of this dissertation suggest that a tax on processed meat, a tax on saturated fat, a removal of the VAT and a subsidy on fruits and vegetables as well as a digital physical activity intervention can with a varying degree contribute to the reduction of NCDs in Germany and at the European level. Especially a tax on processed meat, perhaps with a simultaneous subsidy on fruits and vegetables, may be a promising approach, as both provide comparatively large health benefits and are likely to reduce inequalities as they do not rely on individual’s agency and commitment. At the same time, we identified a range of barriers in the process of conducting the HIAs and therefore formulate the following research recommendations: a) improve data availability, b) assess inequalities between population groups when conducting HIAs, and c) conduct HIAs on more complex interventions and risk factors.