Protein consumption in Canadian habitual diets: usual intake, inadequacy, and the contribution of animal- and plant-based foods to nutrient intakes

Protein consumption has important implications for nutrition, human health, and environmental sustainability. The Dietary Reference Intakes (DRIs) express protein recommendations as the Estimated Average Requirement (EAR) and the Acceptable Macronutrient Distribution Range (AMDR), which do not vary by age or sex for individuals aged ≥19 y (Institute of Medicine 2006). The EAR is the median daily intake of protein needed to meet the needs of half the healthy individuals in a particular life stage and age-sex group, which is set at 0.66 g/kg of body weight (BW)/d. The AMDR is the range of calories, expressed as a percentage of total energy intake, that is associated with a lower risk of chronic disease while providing adequate intake of a macronutrient. For protein, the AMDR is set at 10 to 35% of total energy intake. Data from the United States National Health and Nutrition Examination Survey (NHANES) 2011–2014 revealed that usual protein intake among American adults averaged 80 g/d and 16% of total energy intake (Berryman et al. 2018). The percent of total energy intake from protein averaged 17% for Canadian adults in 2015 (Statistics Canada 2017b). However, information pertaining to inadequate protein intake, which is determined by the percentage of the population below the EAR (Institute of Medicine 2006), is currently lacking.

Animal- and plant-based foods differ in their quantity and quality of protein and content of other essential nutrients. Animal protein is deemed high quality because it provides all 9 essential amino acids and is more bioavailable compared with plants (Institute of Medicine 2006; Lonnie et al. 2018). Consuming plant-based protein sources with complementary amino acid profiles (Arentson-Lantz et al. 2015; Gardner et al. 2019) is encouraged to ensure adequate intake of all essential amino acids, particularly for vegetarians (Institute of Medicine 2006). Commonly consumed animal- and plant-based protein sources also supply a range of essential nutrients (Phillips et al. 2015), some of which are deemed of public health concern in Canada (calcium, vitamin D, iron, and potassium) as individuals fall below recommendations (Health Canada 2014). However, they also confer nutrients to limit (sodium, added sugars, and saturated fat), among which animal-based protein sources contribute significantly to intake of saturated fat, a nutrient that is often scrutinized for its putative harmful association with cardiovascular health (Health Canada 2019b). The choice of protein source is an important determinant of human health, as epidemiological studies have shown that high intakes of plant protein are inversely associated with cardiometabolic indicators and mortality (Lin et al. 2015; Song et al. 2016). Yet, there is also evidence that replacement of protein from red and processed meats with that from other animal sources such as fish, poultry, and low-fat dairy products may help lower cardiometabolic risk factors (Zhubi-Bakija et al. 2020). Moreover, animal sources have a substantially greater carbon footprint than plants per gram of protein (Carlsson-Kanyama and Gonzalez 2009; Pimentel and Pimentel 2003; Sabaté et al. 2014). However, modeling studies have demonstrated that partial replacement of animal-based foods with plant-based alternatives resulted in lower carbon footprint diets that included moderate amounts of nutrient dense animal-source foods (Macdiarmid et al. 2012; Seves et al. 2017). Despite the importance of dietary protein sources to human and planetary health, the respective contribution of animal- and plant-based foods to total protein intake in Canadian habitual diets is unknown.

In January 2019, Health Canada published the first revamp of the nation’s food guide in over a decade. The new Canada’s Food Guide (hereafter referred to as the 2019 CFG) encourages largely plant-based diets, including vegetables and fruit, whole grains, and protein foods. Although the guide does not provide a definition for protein foods, emphasis is placed on consuming protein from plants more often (Health Canada 2019b). At present, the implications of the 2019 CFG on intakes of shortfall nutrients is not known, yet available evidence raises concern as to its nutritional adequacy. Barr (2019) estimated the nutrient content of foods in the food guide snapshot based on a standard 2000 kcal diet and found that the percent Daily Value was not met for calcium, vitamin D, and potassium. Moreover, a recent report on a scientific expert meeting identified knowledge gaps brought about by the new protein foods group, including protein inadequacy among vulnerable populations and intake of other essential nutrients by Canadians (Fernandez et al. 2020). To assess the implications of the 2019 CFG’s protein recommendations, it is important to first characterize protein intake trends in Canadian habitual diets as a baseline with which to compare future dietary shifts. In particular, estimating the contribution of animal- and plant-based foods to intake of protein and other nutrients may help define targets to shape public policy interventions aimed at meeting current dietary guidance. Therefore, the primary objective of this study was to assess usual protein intake and inadequacy for adults using data from the 2015 Canadian Community Health Survey (CCHS) – Nutrition. The secondary objective was to determine the contribution of animal- and plant-based foods to intake of protein and nutrients, particularly nutrients of public health concern and to limit.

Given the recent changes to the 2019 CFG that place emphasis on plant protein, population diet studies are necessary to provide information as to which nutrient-rich protein sources Canadians should be incorporating into their diets (Fernandez et al. 2020). The present study addressed many of the uncertainties and knowledge gaps that arose from the introduction of the new protein foods group in the 2019 CFG, particularly protein intake and inadequacy in the Canadian population with a particular focus on animal- and plant-based foods. Based on our findings, most Canadians were in line with protein recommendations except for females aged ≥71 y, who had the highest prevalence of inadequacy. Moreover, two-thirds of Canadians’ total protein intake derived from animal-based foods for which dairy was a top source of nutrients of concern (calcium and vitamin D) and a nutrient to limit (saturated fat).

Despite assumptions of protein adequacy in the Canadian population (Fernandez et al. 2020), our findings confirm that older adults are most at risk of not meeting recommendations. The percent of the population below the EAR for protein was similar to the pattern observed in the United States (Berner et al. 2013; Berryman et al. 2018; Phillips et al. 2015). Phillips et al. (2016) proposed that higher protein intake within the range of 1.2 to 1.6 g/kg of BW/d may help promote healthy ageing, appetite control, and weight management. For the elderly, higher protein intake may prevent sarcopenia, a disease characterized by the progressive loss of muscle mass and strength that leads to impaired physical function, frailty, and mortality (Cruz-Jentoft and Sayer 2019; Phillips et al. 2016). Current protein requirements based on nitrogen balance studies may underestimate actual requirements, especially for the elderly population (Phillips et al. 2016). As food intake decreases with age (Berner et al. 2013), the body resorts to utilizing protein for energy; since energy to protein ratios are highest for individuals with the lowest energy requirements, sedentary elderly women with higher BMIs are likely to have higher protein requirements compared with other demographics (Nowson and O’Connell 2015). The 2019 CFG may encourage Canadians to consume more protein-rich foods, but the lack of suggested serving sizes and age- and sex-specific recommendations limits its usefulness, particularly for those most at risk of falling below the EAR.

Total protein intake from animal-source foods was more than double that contributed by plant-based foods, similar to observations in the United States and United Kingdom (Berner et al. 2013; Lonnie et al. 2018; Phillips et al. 2015). We found that chicken and beef alone contributed the most to total protein intake, whereas protein from plant sources derived from a wider range of food categories, albeit contributing relatively little protein. Similarly, data from NHANES 2007–2010 revealed that chicken and beef contributed one-quarter of protein intake from animal sources and 13% of total protein intake (Pasiakos et al. 2015). We also found that cheese and fluid milk were the top dairy sources of total protein, although cheeses contributing the most protein were also particularly high in fat (>25% and 10 to 25% milk fat). Among plant-based foods, cereals, grains, and breads contributed the most to total protein intake. However, top sources were refined products as opposed to whole grain food categories. One analysis using data from the 2015 CCHS – Nutrition classified respondents into clusters based on their consumption of grain-based foods and found that only 8% of Canadian adults followed a ‘Whole Wheat & Whole-Grain Bread’ dietary pattern (Hosseini et al. 2019), which did not align with recommendations to consume most grains as whole (Health Canada 2011, 2019b). Although nuts, seeds, and legumes have a higher overall protein content compared with other plant-based foods, they are consumed in smaller quantities and thus contributed negligible amounts to total protein intake. Although current dietary guidance promotes consumption of protein from plants more often, particularly from sources exemplified by this group (Health Canada 2019a), our findings revealed that Canadians obtained relatively little protein from nuts, seeds, and legumes compared with animal-based foods and even cereals, grains, and breads. Substantial shifts in Canadian dietary patterns are required to increase the prominence of plant-based protein from a variety of sources, particularly from nuts, seeds, and legumes, as recommended in the 2019 CFG.

One of the major knowledge gaps regarding population-wide protein intake in Canada is the contribution of various protein sources to intakes of nutrients of public health concern (Fernandez et al. 2020). Our findings show that dairy contributed the most to intake of calcium and vitamin D, cereals, grains, and breads to intake of iron, and vegetables and fruit to intake of potassium. In the United States, animal-based protein sources contributed greater amounts of iron, zinc, vitamin B12, and phosphorus compared with plant sources, which contributed more dietary fibre, vitamin E, and magnesium (Phillips et al. 2015). Our previous work showed that milk and alternatives contributed 53% of calcium and 39% of vitamin D in Canadian habitual diets, in addition to a range of other essential nutrients (Auclair et al. 2019). In Canada, mandatory fortification of milk with vitamin D under the Food and Drugs Act (Health Canada 1999) explains its contribution to intake of this nutrient. Diet modeling of NHANES 2003–2006 revealed that it would be difficult to replace nutrients from dairy with non-dairy foods (Fulgoni et al. 2011). For example, replacing the calcium from dairy with a non-dairy calcium composite (either fortified soy-based beverage or orange juice, bony fish, or leafy greens) would result in lower overall intake of protein, total fat, vitamin B12, riboflavin, phosphorus, zinc, saturated fat, and sodium, whereas intake of magnesium, potassium, and vitamin A would increase, with no change in vitamin D. Based on our results, there is concern as to whether dietary shifts aligning with those in the 2019 CFG, particularly with regards to protein foods, may further compromise intake of certain nutrients of public health concern provided no changes in the mandatory fortification of foods or additional dietary guidance.

Animal- and plant-based foods also contributed to intake of nutrients to limit. Our results show that dairy was the top source of saturated fat, which was 2 and a half times that contributed by red and processed meat. The 2019 CFG encourages the replacement of foods that are high in saturated fat with foods containing unsaturated fat as a means of promoting cardiovascular health (Health Canada 2019b). A similar rationale was proposed in the World Health Organization’s draft guidelines on saturated fat. However, Astrup et al. (2019) argue that the type of saturated fatty acid and the food matrix are both critical factors for informing dietary recommendations. Substantial evidence points to an inverse association between plant protein and cardiometabolic health (Mariotti 2019; Song et al. 2016), which have also been shown to have a lesser environmental impact than animal protein (Auestad and Fulgoni 2015). Therefore, the type of animal- and plant-based protein making up habitual diets is fundamental in addressing human and planetary health in tandem.

Strengths of this study include the use of data from a nationally representative survey and the estimation of usual protein intake using the NCI method. Moreover, to our knowledge, this is the first study to provide estimates of protein inadequacy in the Canadian population. However, limitations include the self-reported nature of 24-hour dietary recalls, which are prone to bias through misreporting. However, according to Garriguet (2018), energy misreporting was not a major source of bias in the 2015 CCHS – Nutrition. Furthermore, the 2019 CFG does not yet provide a definition for protein foods; therefore, we used examples in the guide to broadly classify all foods and beverages reported in the CCHS into animal- and plant-based food groupings. Although protein recommendations are based on consumption of high-quality protein (WHO/FAO/UNU Expert Consultation 2007), there is a lack of data pertaining to protein quality for foods based on the Digestible Indispensable Amino Acid Score (Hodgkinson et al. 2020). At present, there is insufficient information to assess the influence of protein quality on protein inadequacy. Moreover, the degree of detail with which we categorized foods was limited (e.g., grouping eggs with poultry) due to the computationally intensive nature of the NCI method (Davis et al. 2019). Finally, since the Canadian Nutrient File does not distinguish between total and free sugars, the present analysis accounts solely for total sugars, despite recommendations in the 2019 CFG that are geared towards free sugars.

In conclusion, most Canadian adults were in line with the DRIs for protein; however, special attention should be warranted to older adults and females who were more prone to fall short of requirements. Except for cereals, grains, and breads, the majority of protein intake was derived from animal-based foods. Yet, animal sources were not top contributors of nutrients of concern, with the exception of dairy, which also contributed significant amounts of saturated fat. Moreover, miscellaneous foods and beverages contributed negligible amounts of protein but were top sources of sodium and total sugars. Based on Canadian habitual diets in 2015, our results show that major adjustments are needed to meet the recommendations in the 2019 CFG, specifically regarding the shift towards plant-based protein foods. Despite the guide’s holistic approach to healthy eating, such transitions may pose implications on future prevalence of inadequacy for protein and nutrients of concern in Canada, particularly for older adults and females.

The analysis presented in this paper was conducted at the McGill−Concordia Laboratory of the QICSS, which is part of the Canadian Research Data Centre Network (CRDCN). The services and activities provided by the McGill−Concordia Laboratory of the QICSS are made possible by the financial or in-kind support of the Social Sciences and Humanities Research Council, the Canadian Institutes of Health Research, the Canadian Foundation for Innovation, Statistics Canada, and McGill University. The views expressed in this paper do not necessarily represent that of CRDCN or its partners. The authors would like to thank Didier Garriguet from Statistics Canada for his advice and guidance regarding the estimation of usual intakes using the NCI method. The authors would also like to thank Danielle Forest from Statistics Canada at the McGill−Concordia Laboratory of the Quebec Inter-University Centre for Social Statistics for processing vetting requests. O.A. was recipient of the Margaret A. Gilliam Fellowship in Food Security.