Changing Landscape from Nutrients to Dietary Patterns: Implications for Child Health

Introduction

Diet has undergone a lot of changes in the last several decades, from a purely nutritional function to a platform to mitigate and protect against chronic diseases. A healthy diet has now become the goal to achieve this. The World Health Organization defines a healthy diet as one that helps to protect against malnutrition in all its forms, whether undernutrition, overnutrition, or micronutrient deficiency, and noncommunicable diseases (NCDs) like diabetes, heart disease, stroke, and cancer ^[1].

For infants and young children, the aim is for optimal nutrition in the first 2 years of life to achieve enhanced growth and development as well as reduce the risk of becoming overweight or obese and developing NCDs later in life ^[1]. That diet is a significant risk factor for occurrence of NCDs was emphasized in the study that analyzed the impact of suboptimal diet and NCD mortality and morbidity ^[2]. Using a comparative risk assessment approach from data gathered from consumption of major foods and nutrients from 195 countries, 11 million deaths were attributable to dietary risk factors led by high sodium intake and low intake of whole grains and fruits.

Changes in Focus: From Nutrients to Dietary Patterns

The emphasis on the diet–disease relationship has acquired a new dimension recently. Previously, it focused on individual nutrients. This single nutrient and outcome diet quality approach is being challenged because of certain disadvantages. It is difficult to attribute health effects to a single dietary component. Individual dietary components interact with each other in the form of food synergy. If there is a need to manipulate a diet, there may be substitution effects: a high or low intake of one dietary component may produce changes in other aspects of the diet. From a narrow perspective involving micronutrients then macronutrients to foods and food groups, the view has become wider in scope. This resulted in a change in the nutrition landscape to involve dietary patterns (DPs) and how they affect dietary quality, which in the long term can relate to health and disease outcomes (Fig. 1).

Dietary Patterns

DPs are quantities, proportions, variety, or combination of different foods, drinks, and nutrients in diets and the frequency by which they are habitually consumed ^[3]. A healthy DP is composed of nutrient-dense foods and beverages across all food groups in recommended amounts based on age and physiologic status and within calorie limits. It should support present and future health emphasizing on protection against serious health outcomes or chronic diseases later in life. It should be established early, even prenatally with good maternal nutrition, and maintained throughout different life stages. Maintenance is the key to minimize diet-related chronic diseases risks. Conversely, consuming unhealthy DP throughout life can increase the risks of chronic diseases like cardiovascular diseases, obesity, diabetes, and some forms of cancer.

DPs as Assessment Tool for Diet Quality

DPs can be used to assess diet quality. Currently, two methods of derivation of DPs are being used, the index-based approach and the data-driven approach.

Index-Based Approach
This method is called “a priori” approach since it relies on scientific knowledge from previous investigations into effects of diet in health promotion or disease prevention. Then compliance is measured with a preexisting diet quality index, current guidelines, or recommendations by assigning diet scores which can then reflect the level of adherence. Examples include the Healthy Eating Index, the Mediterranean Diet Score (MDS), and the Dietary Approaches to Stop Hypertension (DASH) score.

Data-Driven Approach
This method is called “a posteriori” approach to distinguish it from the index based approach. Available data determine the patterns using statistical methods to generate them. It provides information about existing DPs within a specific population group. It does not identify an ideal DP but captures the primary source of variation within the study population. Examples include principal component analysis, cluster analysis, and factor analysis. The steps include first systematically reducing the number of foods reported to reach an optimal combination of food groups. Then the inputs or predictors are aggregated into linear combinations that best explain the maximum amount of variation in decreasing importance across all variables. These principal components, factors, or clusters are now referred to as the DPs. Examples include “snack pattern,” “sweets pattern,” “healthy pattern,” and “western dietary pattern.”

Comparisons of the two approaches are listed in Table 1. There are certain drawbacks with the use of each approach. The index-based approach focuses on selected aspects of diet and does not consider the correlation of food or nutrient items. It does not reflect the overall effect of the diet in general but rather is just a formal sum of single effects ^[4]. Also, in the calculation of diet score, the definition of cutoff points for high or low consumption is arbitrary and subjective. In the data-driven approach, the decision as to what principal components or factors are going to be included to describe a DP is often biased toward the beliefs of the authors and not on evidence-based criteria and independent from the relevance for any disease outcome ^[5]. In addition, DPs extracted from one study population are not reproducible and comparable since estimates are study specific.

Directions of DPs
DPs can have two different directions, either high or low quality. As expected, high-quality DP will result in good outcome, but the reverse is true for low quality DP. High-quality DPs are expected to reduce outcome risks as exemplified by the DASH diet and the Healthy Mediterranean diet. The higher the score utilizing these diets, the better is the outcome. On the other hand, some dietary recommendations utilize DPs that are of low quality to emphasize their deleterious effects. An example is the NOVA food classification, which started in Brazil. With this classification, it assigns a group to food products based on how much processing they have undergone, from unprocessed to ultra-processed food and drinks. The more processed the food, the higher is the score on NOVA classification, but the lower is its quality.

DPs and Health Outcomes

Studies on DPs and health outcomes mainly concentrated on adults from both index-based and data-driven approaches. An updated systematic review and meta-analysis ^[6] showed that those with higher scores from healthy DPs using Healthy Eating Index, Alternative Healthy Eating Index, and DASH score had reduced risk for all-cause mortality, cardiovascular disease incidence or mortality, cancer incidence or mortality, type 2 diabetes, and neurodegenerative diseases.

Using data-driven DPs, an umbrella review of 18 meta-analyses ^[7] showed that from studies with a moderate level of evidence healthy DPs, described as “healthy,” “prudent,” “fruits and vegetables” predominantly plant-based with olive oil, seeds, beans, fruits, vegetables, and nuts, reduced the risk for fractures, type 2 diabetes, breast cancer, and colorectal cancer, while unhealthy DPs described as “western,” “traditional,” “high fat,” or “sweets and fats” increased the risk for type 2 diabetes, metabolic syndrome, and fractures.

Studies in children pose certain inherent difficulties. Children’s diet is evolving and transitioning to adult type of diet. Diet prescription should account for rapid growth rate, higher energy, and nutrient requirements, but with limited capacity for food intake especially in the first year of life. More than the quantity and quality of food, early childhood is also critical in developing healthy eating habits that transcend into adulthood and may impact also in later health. The role of milk, as the predominant source of nutrition in early childhood, should be considered in the recommendation. In older children and adolescents, compliance may present as a problem like when prescribing Mediterranean or DASH diets to them.

There are few available studies on DPs and health outcomes in children. Most are descriptive in nature. There are also a variety of indices using different quality of unvalidated dietary assessment tools, making conclusions uncertain with limited applicability and generalizability. In developed countries, most studies are focused on the association of dietary intake with dietary quality and only a few on diet-related chronic diseases. In developing countries, most studies concentrated on assessment of nutrient adequacy and growth ^[8].

A summary of the studies on children’s DPs and health outcomes shows that unhealthy DP is related to: 

• Increased risk for overweight and higher fat mass and body mass index (BMI) in adolescence ^[9] 
• Higher cardiometabolic markers independent of BMI: systolic blood pressure, insulin resistance, and triglycerides ^[10]
• Increased odds of attention deficit hyperactivity disorder ^[11]
• Poor social skills and problematic behavior ^[12]
• Increased risk for asthma ^[13]

Features of Studies on DPs in Children

Tracking
Tracking can be viewed in two ways: first, early dietary exposure may be a prelude to later DP, and second, DPs developed early in life can persist through later childhood. In the first scenario, studies to prove this rely mainly on association at two or more time points. An example is from a study of children in Brazil ^[14], which showed that those who were not exclusively breastfed until 4 months of age had higher adherence to “unhealthy” and “snack” patterns in later childhood. In Australia, there was a positive association between breastfeeding during infancy and healthy, meat and vegetable patterns at 2 to 8 years of age ^[15].

In the second scenario, evidence gathered from long-term cohort studies will reveal whether DPs developed early in life can persist throughout different life stages. Several examples exist from large longitudinal studies performed in several countries: Avon Longitudinal Study of Parents and Children (ALSPAC) ^[16], the Childhood Obesity Project of the European Union (EU-CHOP) ^[17], the Western Australian Pregnancy Cohort Study ^[18], the Melbourne Infant Feeding Activity and Nutrition Trial (InFANT) ^[19], the EDEN Mother-Child Cohort ^[20], the Danish SKOT-1 study ^[21], Young Finns Study ^[22], and the Saskatchewan Pediatric Bone Mineral Accrual Study ^[23]. Majority of these utilized data-driven techniques with very few using the index-based approaches. Although the results were population specific, the conclusions were similar across all the studies that DPs, whether healthy or unhealthy, can develop early in life and may persist through childhood, adolescence, or adulthood. This may have an impact on later risks for diseases such that early intervention may be warranted to foster a healthy eating pattern geared toward better future health outcomes.

Tracking and Association with Disease Risks
Available studies have implicated that tracking of DPs from early childhood predisposes to disease risks at a later age. The two important disease risks centered on cardiometabolic markers and adiposity. Certain disease markers do not translate to actual disease but may be surrogate indicators of possible disease in the future. Using a posteriori approaches, adherence to less healthy DPs (protein DP, fats and sugars DP in the EU-CHOP Project ^[24]; “sweet tooth, drinker/social DP, western DP in the Northern Ireland Young Hearts Project [25]; transition DP in the Danish SKOT-1 study ^[21]; and the traditional DP in the Young Finns Study ^[26]) showed direct association with systolic blood pressure, triglycerides, low-density lipoprotein and total cholesterol, HOMA-IR, homocysteine, apolipoprotein B, and C-reactive protein in later childhood and adolescence. The reverse was true with the healthier DP generally giving inverse association.

The Young Finns Study ^[26] went further to evaluate some adult subclinical outcomes like adult arterial pulse wave velocity (PMV) and carotid intima media thickness suggestive of vascular aging and showed that prolonged preferential consumption of fruits and vegetables from childhood to adulthood was associated with lower PMV and carotid intima media thickness.

Few a priori studies also showed the benefits of adhering to certain diet with trend toward lower PMV in young adults in the Northern Ireland Young Hearts Project among those with high MDS ^[25]. From the ALSPAC ^[27], high MDS at 13 years was associated with significant reduction in cardiometabolic risk score at 24 years.

A similar trajectory is observed regarding DPs and association with adiposity. Children with lower adherence to transition DP in the Danish study ^[21] or high in “energy-dense, low-fiber, high-fat” DP in the ALSPAC ^{[16, 28]} were associated with increasing adiposity as measured by BMI or fat mass index. In a systematic review of 16 studies, diet with lower percentage of obesogenic factors has the potential to reduce the risk of obesity ^[9].

Whether these markers and indicators will translate to actual diseases later in life remains to be proven. But what is important is early detection and surveillance that will lead to early prevention.

Effects of Sociodemographic Factors
Sociodemographic factors appear to correlate with diet quality in children and include age, gender, income, parental education, maternal age, socioeconomic status, residence, marital status of parents, presence of other siblings, and parental smoking to name a few. What is consistent among the studies is that higher scores were observed among girls, younger children, and those from high-income bracket ^[8]. Maternal education constitutes the strongest predictor of eating and maintaining healthier DPs. Lower maternal education (with higher BMI scores in some studies) significantly predicted adherence to processed energy-dense DP ^[16], “western” diet ^[19], “low fruit and vegetables diets” ^[28], “processed and fast foods” ^[20], and “junk pattern” ^[29], “snacky diet” ^[30].

Conclusion

A shift in focus has emerged from nutrients to DPs and rightfully so since diet is consumed as a synergy of individual components. DPs correlate with diet quality. Two approaches are used to derive DPs: index-based and data-driven approaches, each with its own advantages and disadvantages. DPs are related to chronic health outcomes later in life, more evident in adults, but with fewer studies in children. Most current pediatric studies focused on tracking of certain DPs and their association with disease risks using proxy indicators notably cardio metabolic markers and adiposity. Tracking and adherence to specific DPs from early childhood may influence future disease causation, but stronger evidence is needed.

References

1. World Health Organization. Healthy diet, 2020. Available from: https://www.who.int/news-room/ fact-sheets/detail/healthy-diet. Accessed May 1, 2022. 
2. GBD 2017 Diet Collaborators. Health effects of dietary risks in 195 countries 1990–2017:a systematic analysis for the Global Burden of Disease Study 2017. Lancet. 2019;393:1958–72.
3. US Department of Agriculture, Center for Nutrition Policy and Promotion, Nutrition Evidence Library, Dietary Patterns Technical Expert Collaborative. A series of systematic reviews on the relationship between dietary patterns and health outcomes. 2014. Available from: https://nesr. usda.gov/sites/default/files/2019–06/DietaryPatternsReport-FullFinal2.pdf. Accessed May 1, 2022.
4. Hoffmann K, Schulze MB, Schienkiewitz A, et al. Application of a new statistical method to derive dietary patterns in nutritional epidemiology. Am J Epidemiol. 2004;159:935–44.
5. Trichopoulos D, Lagiou P. Dietary patterns and mortality. Br J Nutr. 2001;85:133–4.
6. Morze J, Danielwicz A, Hoffmann G, et al. Diet quality as assessed by the Healthy Eating Index, Alternate Healthy Eating Index, Dietary Approaches to Stop Hypertension score, and health outcomes: a second update of a systematic review and meta-analysis of cohort studies. J Acad Nutr Diet. 2020;120:1998–2021.
7. Jayedi A, Soltani S, Abdolshahi A, et al. Healthy and unhealthy dietary patterns and the risk of chronic disease: an umbrella review of metaanalyses of prospective cohort studies. Br J Nutr. 2020;124:1133–44.
8. Lazarou C, Newby PK. Use of dietary indexes among children in developed countries. Adv Nutr. 2011;2(4):295–303.
9. Liberali R, Kupek E, de Assis MAA. Dietary patterns and childhood obesity risk: a systematic review. Child Obes. 2020;16:70–85. 
10. Rocha NP, Milagres LC, Longo GZ, et al. Association between dietary pattern and cardiometabolic risk in children and adolescents: a systematic review. J Pediatr (Rio J). 2017;93:214–22. 
11. Howard AL, Robinson M, Smith GJ, et al. ADHD is associated with a “Western” dietary pattern in adolescents. J Atten Disord. 2011;15:403–11. 
12. Oh SY, Ahn H, Chang N, et al. Dietary patterns and weight status associated with behavioural problems in young children. Public Health Nutr. 2014;17:2563–9. 
13. D’Innocenzo S, Matos SM, Prado MS, et al. Dietary pattern, asthma, and atopic and non-atopic wheezing in children and adolescents: SCAALA study, Salvador, Bahia State, Brazil. Cad Saude Publica. 2014;30(9):1849–60.
14. Vieira SA, de Almeida Fonseca PC, Andreoli CS, et al. Exclusive breast-feeding and sociodemographic characteristics are associated with dietary patterns in children aged 4–7 years. Public Health Nutr. 2019;22:1398–1405 
15. Grieger JA, Scott J, Cobiac L. Dietary patterns and breast-feeding in Australian children. Public Health Nutr. 2011;14:1939–47. 
16.  Emmett PM, Jones LR, Northstone K. Dietary patterns in the Avon Longitudinal Study of Parents and Children. Nutr Rev. 2015;73(Suppl. 3); 207–30. 
17. Luque V, Escribano J, Closa-Monasterolo R, at al. Unhealthy dietary patterns established in infancy track to mid-childhood: the EU childhood obesity project. J Nutr. 2018;148:752–9. 
18.  Appannah G, Murray K, Trapp G, et al. Dietary pattern trajectories across adolescence and early adulthood and their associations with childhood and parental factors. Am J Clin Nutr. 2021;113:36–46. 
19. Lioret S, McNaughton SA, Spence AC, et al. Tracking of dietary intakes in early childhood: the Melbourne InFANT Program. Eur J Clin Nutr. 2013;67:275–81. 
20. Lioret S, Betoko A, Forhan A, et al. Dietary patterns track from infancy to preschool age: crosssectional and longitudinal perspectives. J Nutr. 2015;145:775–82. 
21. Andersen LB, Mølgaard C, Ejlerskov KT, et al. Development of dietary patterns spanning infancy and toddlerhood: relation to body size, composition and metabolic risk markers at three years. AIMS Public Health. 2015;27:332–57. 
22. Mikkilä V, Räsänen L, Raitakari OT, et al. Consistent dietary pattern identified from childhood to adulthood: the cardiovascular risk in Young Finns Study. Br J Nutr. 2005;93:923–31. 
23. Movassagh EZ, Baxter-Jones ADG, Kontulainen S, et al. Tracking dietary patterns over 20 years from childhood through adolescence into young adulthood: the Saskatchewan Pediatric Bone Mineral Accrual Study. Nutrients. 2017;9:990– 1003. 
24. Luque V, Closa-Monasterolo R, Grote V, et al. Dietary patterns acquired in early life are associated with cardiometabolic markers at school age. Clin Nutr. 2021;40:4606–14.
25. McCourt HJ, Draffin CR, Woodside JV, et al. Dietary patterns and cardiovascular risk factors in adolescents and young adults: the Northern Ireland Young Hearts Project. Br J Nutr. 2014;112:1685–98. 
26. Kaikkonen JE, Mikkilä V, Magnussen CG, et al. Does childhood nutrition influence adult cardiovascular disease risk? Insights from the Young Finns Study. Ann Med. 2013;45:120–28. 
27. Buckland G, Taylor CM, Emmett PM, et al. Prospective association between Mediterranean-style dietary score in childhood and cardiometabolic risk in young adults from the ALSPAC birth cohort. Eur J Nutr. 2022;61:737–52. 
28. Ambrosini GL, Emmett PM, Northstone K, et al. Tracking dietary pattern associated with increased adiposity in childhood and adolescence. Obesity. 2014;22:458–65. 
29. Northstone K, Emmett P. Multivariate analysis of diet in children at four and seven years of age and associations with socio-demographic characteristics. Eur J Clin Nutr. 2005;59:751–60. 
30. Leventakou V, Sarri K, Georgiou V, et al. Early life determinants of dietary patterns in preschool children: Rhea mother-child cohort, Crete, Greece. Eur J Clin Nutr. 2016;70:60.