medRxiv preprint doi: https://doi.org/10.1101/2022.05.18.22275284; this version posted May 21, 2022. The copyright holder for this preprint 
          (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. 
                              It is made available under a CC-BY-NC-ND 4.0 International license . 
         1  PRIMEtime: an epidemiological model for informing diet and obesity policy 
         2   
         3  Linda J Cobiac 
         4  Senior Research Fellow 
         5  School of Medicine and Dentistry 
         6  Griffith University 
         7   
         8  Cherry Law 
         9  Lecturer 
        10  Department of Agri-Food Economics and Marketing 
        11  University of Reading 
        12   
        13   
        14  Peter Scarborough 
        15  Professor of Population Health 
        16  Nuffield Department of Population Health 
        17  University of Oxford 
        18   
        19   
        20  Funding:  
             NOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice.
                                                                                       1 
            medRxiv preprint doi: https://doi.org/10.1101/2022.05.18.22275284; this version posted May 21, 2022. The copyright holder for this preprint 
             (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. 
                                       It is made available under a CC-BY-NC-ND 4.0 International license . 
          21    This project was supported by the NIHR Biomedical Research Centre at Oxford (IS-BRC-1215-20008) 
          22    and an NIHR project grant for an evaluation of the UK Soft Drinks Industry Levy (16/130/01).  
          23     
          24     
          25     
          26                                     
                                                                                                                   2 
            medRxiv preprint doi: https://doi.org/10.1101/2022.05.18.22275284; this version posted May 21, 2022. The copyright holder for this preprint 
             (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. 
                                       It is made available under a CC-BY-NC-ND 4.0 International license . 
          27    Abstract 
          28    Background: Mathematical modelling can play a vital role in guiding public health action. In this paper, 
          29    we provide an overview of the revised and updated PRIMEtime model, a tool for evaluating health 
          30    and economic impacts of policies impacting on diet and obesity. We provide guidance on populating 
          31    PRIMEtime with country-specific data; and illustrate its validation and implementation in evaluating a 
          32    combination of interventions in the UK: a sugar-sweetened beverage (SSB) tax; a ban on television 
          33    advertising of unhealthy foods; and a weight loss program. 
          34    Methods: PRIMEtime uses routinely available epidemiological data to simulate the effects of changes 
          35    in diet and obesity on 19 non-communicable diseases, in open- or closed-population cohorts, over 
          36    time horizons from 1 year to a lifetime. From these simulations, the model can estimate impact of a 
          37    policy on population health (obesity prevalence, cases of disease averted, quality-adjusted life years), 
          38    health and social care costs, and economic measures (net monetary benefit, cost-effectiveness ratios). 
          39    We populated  PRIMEtime  with  UK  data  and  validated  epidemiological  predictions  against  two 
          40    published data collections. We then evaluated three current obesity intervention policies based on 
          41    estimates of effectiveness from published evaluation studies. 
          42    Results: There was considerable variation in the modelled impact of interventions on prevalence of 
          43    obesity and subsequent changes in health and the need for health care: restrictions on TV advertising 
          44    of unhealthy foods to children led to the largest reductions in obesity prevalence; but the SSB tax, 
          45    which also targeted adults, had the biggest benefits in reducing obesity-related disease; and the 
          46    weight loss program, while having very small impact on obesity prevalence at the population scale, 
          47    had large and immediate benefits in improving health and reducing health sector spending. From a 
          48    health sector perspective, the combination of interventions produced a favourable net monetary 
          49    benefit of £31,400 (12,200 to 50,700) million. But the combined effect in reducing prevalence of 
          50    overweight  and  obesity,  was  not  estimated  to  reach  more  than  0.81  percentage  points  (95% 
                                                                                                                   3 
            medRxiv preprint doi: https://doi.org/10.1101/2022.05.18.22275284; this version posted May 21, 2022. The copyright holder for this preprint 
             (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. 
                                       It is made available under a CC-BY-NC-ND 4.0 International license . 
          51    uncertainty interval: 0.21 to 1.4) for males and 0.95 percentage points (0.24 to 1.7) for females by 
          52    2050.  
          53    Conclusions: Diet and obesity interventions have the potential to improve population health and 
          54    reduce health sector spending both immediately and in the long-term. Models such as PRIMEtime can 
          55    be used to evaluate the economic merits of intervention strategies and determine how best to 
          56    combine interventions to achieve maximum population benefit. But with almost a third of children 
          57    and two-thirds of adults currently overweight or obese, we need to broaden the application of public 
          58    health models to evaluating the structural and systemic changes that are needed in our society to 
          59    address the underlying drivers of the obesity epidemic. 
          60                                     
                                                                                                                   4