Table 1

Key input data and assumptions

InputStratificationValuesSourceNote
Preintervention TFA intake, E%Age, sexOnline supplemental table S12019 Global Burden of Disease (GBD) studyFor each model iteration, random draws from age-sex-specific lognormal TFA distributions were made.
Postintervention TFA intake, E%n/a0 (primary analysis) Mean ± SD: 0.1±0.01 (sensitivity analysis)2018 Global Dietary Database (estimates of dairy intake)Given the minimal intake of naturally occurring trans-fatty acids (eg, from dairy) compared to countries like Australia, UK and Denmark, the intervention was assumed to virtually eliminate trans-fatty intake in Nigeria. However, we explored the impact of a low but existing TFA intake post intervention in a sensitivity analysis.
Theoretical minimum risk distribution of TFA intake, E%n/a0Marklund et al, Plos Med 2020The theoretical minimum risk distribution of TFA intake was assumed to equal the intake of natural occurring TFA.
Population sizeAge, sexOnline supplemental table S22019 Global Burden of Disease (GBD) study
Mortality rateAge, sexOnline supplemental table S22019 Global Burden of Disease (GBD) study
IHD incidence, prevalence and case fatality ratesAge, sexOnline supplemental table S32019 Global Burden of Disease (GBD) study
Disability weightsAge, sexOnline supplemental tables S4 and S52019 Global Burden of Disease (GBD) study
Healthcare costsAge, sexOnline supplemental table S6Rosendaal et al, PLOS ONE 2016.We used cost estimates of acute care after an IHD event per patient (ie, incident IHD costs) and costs of follow-up care after an IHD event per patient per year (ie, prevalent IHD costs) from Rosendaal et al20 and estimates of total health expenditure from the Nigerian National Health Accounts for 2017 report.22 In the absence of a Nigerian source, we derived age and sex distributions of total health expenditure in Nigeria by assuming the relative expenditure by age and sex from the Kenya Household Health Expenditure and Utilisation Survey, 2013. The total health expenditure was then divided by population numbers in each sex and age-group to derive the per capita expenditure. Non-IHD costs were calculated for each age-sex group, subtracting IHD costs from total healthcare expenditure for that group, and was then divided by the group’s population size to estimate per capita non-IHD costs. All healthcare costs were inflated to 2019. For each model iteration, random draws from triangular distributions of IHD costs and normal distributions of non-IHD costs (assuming SD equal to 20% of central estimates) were made.
RR for IHD per 2%E from TFAAgeOnline supplemental table S7Afshin et al, Lancet 2019For each model iteration, random draws from age-specific lognormal RR distributions were made.
Government policy implementation costsn/aOnline supplemental table S8NAFDAC estimatesThe following costs were considered: national legislation costs; costs for NAFDAC (ie, work force expenses, enlightenment activities, enforcement activities, workshops activities, monitoring, and other activities including research); FMoH costs for legislation; SON costs for revising food standard regulations; and CPC costs for enlightenment activities. For each model iteration, random draws from normal distributions of implementation costs were made, assuming SD equal to 20% of central estimates.
Industry reformulationn/aOnline supplemental table S8Marklund et al, Plos Med, 2020Reformulation costs were calculated using equivalent USD costs from UK estimates (25 000 GBP per product)5 multiplied by the number of products in the Nigerian food supply potentially containing industrial TFA (primary model: n=331; sensitivity analysis: n=662). Annual cost to industry equaling 1% of the initial reformulation cost was assumed. For each model iteration, random draws from normal distributions of initial and annual reformulation costs were made, assuming SD equal to 20% of central estimates.
  • Table created by the authors.

  • CPC, Consumer Protection Commission; E%, energy percentage; FMoH, Federal Ministry of Health; GBP, British pound; IHD, ischaemic heart disease; n/a, not available/applicable; NAFDAC, National Agency for Food and Drug Administration and Control; RR, relative risk; SON, Standard Organisation of Nigeria; TFA, trans-fatty acid; USD, United States dollar.