*Geek Box: Diet Quality Indices
Most approaches to dietary assessment in nutritional epidemiology focus on individual foods and nutrients as the exposure of interest. However, there are several different methods of analysing total dietary patterns or characteristics of dietary patterns, known as “diet quality indices”, or “index” in the singular. This involves creating a numeric scoring system to represent the quality of the overall dietary pattern based on specific levels of intake of different food groups, foods, and nutrients, in that dietary pattern.
There are many examples of diet quality indices, including the Diet Quality Index, Healthy Eating Index and Alternate Healthy Eating Index, and the Mediterranean Diet Score. They provide a means to quantitatively assess the healthfulness of a dietary pattern, given “quality” is an ambiguous term for research purposes. The overall scoring range may then be used as a numeric variable for a statistical analysis, i.e., by comparing levels of adherence quantified numerically as “low”, “moderate”, or “high” adherence to that dietary pattern.
There are several ways an index may assess diet quality. Basing the index on food groups and nutrient intakes is the most common, e.g., an index may have “low-fat dairy” as a food group, and then thresholds in grams per day upon which higher or lower scores are assigned. Alternatively, polyunsaturated fats [PUFA] may have a threshold in percentages, to which higher or lower scores are assigned based on PUFA intake as percentage of total energy intake.
In contrast, dietary components associated with negative health outcomes, including sugar-sweetened beverages and fruit juices, red/processed meats, trans fats, and sodium have points scores inverse to consumption. For example, the maximum score for sodium intake in the Alternate Healthy Eating Index is awarded for <1,612mg/d sodium intake, while >5,271mg/d sodium intake results in a score of 0.
Where evidence exists of the effects of a specific diet, such as in the case of a Portfolio Diet, the diet quality index for that diet may be based on evidence from interventions of the effects of the component parts of the diet. In the case of the Portfolio Diet, the diet quality index was created by categorising the diet into 6 components: plant protein, nuts, viscous fibre, plant sterols, MUFA, and saturated fat and cholesterol sources.
One of the major advantages to scoring indices like this is that they are inherently adaptable to different dietary patterns, and may need to evolve over time with further knowledge. For example, a max score is provided in the Alternate Healthy Eating Index if trans fats are <0.5%, but due to changes in the food supply this score is often a default 10 points, and not necessarily reflective of diet quality.
There is also debate about whether whole-milk dairy [i.e., “full fat”] should be scored as a positive, negative, or neutral score for a dietary pattern. The MIND dietary pattern score also scores positively for red wine, but some would debate the merits of any alcohol. The point to bear in mind here is that any dietary pattern score is not necessarily a sacrosanct representation of a particular dietary pattern, and they are modifiable as the wider evidence develops.