Evidence and the policy cycle

A robust and broad ranging evidence base has been critical to determine and shape policies on diet and obesity and physical activity, allowing the development of policy that is well targeted and likely to achieve improved health outcomes – for example, incremental salt and sugar reduction^(8,9) , the rationale for the establishment of the soft drinks industry level (‘sugar tax’⁽¹⁰⁾), implementation of restrictions on advertising^(11,12) and the fortification of non-wholemeal wheat flour with folic acid⁽¹³⁾. It is also critical in order to robustly defend policy decisions; SACN's risk assessments were fundamental in Department of Health and Social Care defending a recent judicial review⁽¹⁴⁾ on the implementation of the nutrient-profiling model in relation to breakfast cereals.

There are many interfaces between evidence production, policy development and implementation. Evidence has a role at all stages of the policy cycle – from identifying a risk, developing dietary recommendations, making the case for intervention and considering options for intervention/management to evaluation and surveillance^{(15,Reference Cairney and Oliver16)} . Evidence developed or considered by policy makers may be directly requested by ministers, instigated by an independent advisory committee (such as SACN or the Committee on Toxicity) or other organisations (such as the National Institute for Health and Care Excellence or WHO), requested or recommended by devolved agencies, such as Food Standards Scotland, or other government departments, or be part of wider research initiatives. However, the type of evidence required will depend on the stage of the process, for example risk assessment or risk management, or the public health question being asked. Furthermore, policy development is not solely dependent on evidence, and even good evidence may not be enough to ensure that a policy is progressed – politics and delivery considerations are also critical. Controversial, politically unfavourable, or costly policies are likely to require much stronger evidence than those which are relatively cheap, easy to implement and generally considered a good idea (or at least uncontroversial) to the majority of people. A good example of a more complex situation is in relation to folic acid. Fig. 1 shows the multiple risk assessmentst, policy deliberations and public consultation required before a policy to mandate the fortification of non-wholemeal wheat flour with folic acid was announced in 2021^(17–22).

Fig. 1. Timeline to date on evidence to policy on mandatory fortification of non-wholemeal wheat flour with folic acid to help prevent neural tube defects (NTD)^(17–22). COMA, Committee on Medical Aspects of Food and Nutrition Policy; COT, Committee on Toxicity; FSA, Food Standards Agency; FSS, Food Standards Scotland; SACN, Scientific Advisory Committee on Nutrition; TUL, tolerable upper level.

Risk assessment by Scientific Advisory Committee on Nutrition

SACN has advised the UK government on nutrition-related matters since 2001, superseding the Committee on Medical Aspects of Food and Nutrition Policy⁽²³⁾. SACN was established as part of a wider government approach to separating risk assessment from risk management (policy making) following the BSE crisis and subsequent Philips enquiry⁽²⁴⁾. A clear conclusion from the enquiry was that expert scientific committees should be restricted to giving advice and should not be setting policy. Therefore, the development of government food and nutrition policy, while based on advice from SACN, is separate to the scientific assessment^(7,25) . However, that is not to say there is not an iterative process and SACN can be asked to advise on possible practical solutions, as is currently the case in relation to its consideration of international approaches to vitamin D fortification⁽²⁶⁾ and plans for monitoring the forthcoming mandatory fortification of non-wholemeal wheat flour with folic acid. SACN's remit is to assess the risks and benefits of nutrients, dietary patterns, food or food components to health by evaluating scientific evidence and to make dietary recommendations for the UK population (including vulnerable and/or diverse groups) based on its assessment. SACN has a public health focus, therefore the treatment of disease is outside SACN's remit unless specifically requested. Consideration of issues related to alcohol, other than as a source of energy, is also outside SACN's remit⁽⁷⁾.

All government scientific advisory committees must have clear governance and oversight in place alongside rigorous quality assurance/assessment process(es), which ensure the quality of the evidence produced and which is important for wider trust. The SACN Code of Practice⁽⁷⁾ sets out how the committee reflects the guidelines governing scientific advisory committees, as set by the Government Office for Science Code of Practice for Scientific Advisory Committees⁽²⁷⁾. Members are recruited in line with the public appointment process and their declarations of interest are published⁽²⁸⁾ with a historical record provided in SACN Annual Report⁽²⁹⁾. Meeting minutes are published⁽³⁰⁾ and some meetings held in open session allowing members of the public to observe. Draft reports undergo consultation, and all responses are published. Representatives from the four UK governments can observe SACN but they must treat the discussions and papers shared in strict confidence. SACN is supported by an Office for Health Improvement and Disparities secretariat who are able to undertake systematic reviews, technical report writing and translation into population dietary recommendations⁽⁷⁾. SACN sets its own work programme, predominantly through a horizon scan process⁽³¹⁾ but also in response to ad hoc requests – for example from ministers, Chief Medical Officers, devolved administrations or policy teams. SACN's framework for the evaluation of evidence⁽²⁵⁾ ensures that its risk assessment processes are robust, reproducible and transparent. SACN generally undertakes systematic reviews of reviews and issues accompanying calls for evidence. Identified evidence is quality assessed and evidence is graded⁽²⁵⁾. Draft reports are consulted on, and responses are published on the gov.uk webpages to all comments received⁽³²⁾.

National datasets, in particular the National Diet and Nutrition Survey, are also fundamental to SACN's risk assessment process, as well as policy development and monitoring (Box 1).

Food-based dietary recommendations

To date SACN's recommendations have predominantly focused on individual nutrients rather than foods or dietary patterns. Reflecting the available evidence, SACN has noted the complexities and weakness of current evidence base, which is largely observational and prone to residual confounding and reverse causality, and that published data may not be directly relevant to the UK population. For example, in relation to Mediterranean diets the committee have noted that there is no single Mediterranean diet (a multitude of scores and analytical approaches are applied), scores may not be validated for all age or population groups, findings tend to be heterogeneous and while the dietary components that are characteristic of Mediterranean dietary patterns broadly align with current UK healthy eating recommendations, as illustrated by the Eatwell Guide, there are few studies that compare these dietary patterns⁽⁴⁴⁾.

SACN (or SACN's predecessor, the Committee on Medical Aspects of Food and Nutrition Policy) has been able to make specific food-based recommendations in some instances:

1. (1) Recommending eating at least five portions of a variety of fruit and vegetables each day ⁽⁴⁵⁾

2. (2) Recommending two portions fish, one of which is oily, per week ⁽⁴⁶⁾

3. (3) Recommending that adults who regularly consume more than 90 g of red and processed meat daily reduce their consumption to no more than the population average of 70 g daily ⁽⁴⁷⁾

4. (4) Recommending that the consumption of sugars-sweetened beverages should be minimised in children and adults ⁽⁴⁸⁾

SACN's current and future work programme may also include dietary pattern and/or food-based recommendations – SACN's current work programme includes risk assessments on vitamin D fortification⁽²⁶⁾, plant-based drinks⁽⁴⁹⁾ and nutrition and maternal health⁽⁵⁰⁾. SACN has also agreed to undertake further consideration of iodine and health, iron bioavailability and a stepwise approach to considering wholegrain, firstly agreeing definitions⁽³¹⁾.

Processed foods

It is estimated that 50–60 % of UK dietary intakes may be classified as ‘ultra-processed’^{(Reference Rauber, da Costa Louzada and Martinez Steele51,Reference Adams and White52)} . The most commonly used definition – NOVA – includes foods that are clearly less healthy – such as sugar-sweetened drinks, confectionary and salty snacks – but also foods that would be encouraged as part of a healthier diet such as wholemeal sliced bread, canned beans or wholegrain breakfast cereals^{(Reference Sadler, Sokolović and Grassby53,Reference Monteiro, Cannon and Lawrence54)} .

Given the increasing interest in this issue, SACN has published a position statement on processed foods and health⁽⁵⁵⁾. SACN evaluated existing classifications of processed foods, including ultra-processed foods; evaluated the suitability of food processing definition(s) as a dietary exposure and considered the availability and quality of evidence considering food processing with health outcomes. SACN concluded that observed associations between higher consumption of (ultra-)processed foods and adverse health outcomes are concerning, however it is unclear to what extent these associations are explained by established relationships between nutritional factors and health outcomes. SACN have made a range of research recommendations to explore relationships between (ultra-)processed foods and health outcomes, based on a classification system that can reliably be applied in the UK. This is a complex issue and further scrutiny, with input from other specialisms such as food scientists and toxicologists, is likely to be required.

Translation of food and nutrient recommendations

In 1994, the Committee on Medical Aspects of Food and Nutrition Policy published dietary recommendations to prevent CVD that are reflective of the present recommendations on macronutrients⁽⁵⁶⁾. To support implementation, the report included illustrative food-based examples utilising information on household food consumption at the time. However, what was intended as a helpful, illustrative translation of complex science was considered by the popular press as fastidious and the evidence-based basis of the nutrient recommendations were overlooked. More successfully, the Committee on Medical Aspects of Food and Nutrition Policy recommendations were translated into a visual food model, ‘The Balance of Good Health’ with segment sizes based upon a model average diet. The visuals were then updated for the Eatwell Plate, published in 2007, which used simpler visuals, a wider range of real foods and more intuitive names for the food groups^(57,58) .

The UK's current national food model, the Eatwell Guide was published in its current form in 2016, based on updated recommendations from SACN on free sugars and fibre. Alongside translation of nutrient recommendations, it also incorporates specific recommendations made by SACN on fruit and vegetable, red and processed meat and oily fish. Developed with an external reference group, it is based on modelled food patterns in the UK using data from the National Diet and Nutrition Survey, thereby reflecting realistic food choices. The Eatwell Guide underwent consumer testing and consultation. It is of note that the Eatwell Guide clearly shows that foods commonly considered ‘ultra-processed’, high in saturated fat, salt or sugar, should be eaten less often or in small amounts, and diets largely plant-based with fruit and vegetables and starchy, wholegrain carbohydrates being the major contributors to our meals^(57,58) .

Government implementation of dietary recommendations based on advice from SACN includes encouraging the food industry to reformulate foods high in sugars, salt and energy^(8,9,59) , the introduction of regulations restricting the placement of products high in saturated fat, salt or sugar in key selling locations in store and online^(11,12) , and the soft drinks industry levy⁽¹⁰⁾. The UK government's activities to translate dietary recommendations make use of the 2004/5 nutrient-profiling model which aims to identify products which are ‘healthier options’⁽⁶⁰⁾. The model was developed through rigorous scientific scrutiny and extensive consultation, including academic workshops and advice from SACN. A research report on the nutrient-profiling model was published in 2005 and following further consultation and scientific scrutiny, technical guidance was first published in 2009⁽⁶¹⁾. The nutrient-profiling model considers the balance of the ingredients and nutritional content of a food, using a scoring system in which points are allocated for certain ingredients or nutrients within a food or drink per 100 g. The model attempts to balance the contribution made by ‘beneficial’ nutrients (i.e. protein, fibre, fruit, vegetables and nuts), alongside the ‘negative’ contributions from nutrients for which intakes are higher than recommended (i.e. energy, saturated fat, sugar and salt).

The Government Buying Standards for Food and Catering Services nutrition standards, first introduced in 2011, implement SACN's dietary recommendations within public sector settings⁽⁶²⁾. The standards are mandatory for all central government departments and their agencies, as well as prisons, the armed forces and National Health Services trusts and foundation trusts in England. School food-based standards to ensure children meet their nutritional requirements across the school day are also mandatory for maintained schools⁽⁶³⁾.

Way forward

Collaboration with other disciplines is an important way forward for risk assessment in nutrition science. SACN is currently undertaking a joint risk assessment with the Committee on Toxicity on plant-based drinks⁽⁴⁹⁾. Such products are not nutritionally equivalent to cow's milk and are known to contain a number of toxicants, but for some population groups such as vegans they may be the only acceptable alternative to cow's milk^(64,65) . The joint risk assessment requires consideration of both the nutritional benefits and toxicological risks and uses the BRAFO (Benefit Risk Analysis for Foods) framework⁽⁶⁶⁾, previously applied by SACN and the Committee on Toxicity on their assessment of the introduction of peanut and hens' egg to infant diets⁽⁶⁷⁾. This type of ‘risk-benefit’ analysis may also be useful to employ for consideration of food or nutrient recommendations in other areas, for example in conjunction with another issue – such as sustainability or aspects of food processing.

Implementation of SACN's risk assessments on nutrition and health – ‘risk management’ – takes an evidence-based approach, as demonstrated by the production methods for the Eatwell Guide. These have generally made use of ad hoc expert groups and consultation with scientists working in the relevant fields. However there is no standing advisory committee for dietary risk management (as SACN provides for risk assessment) and this may be worth exploring in future.