Nutrition and healthy ageing: a matter of reciprocity?

Those over age 65 years constitute the fastest growing segment of populations worldwide. This change in the world's age distribution and consequent demographic shift can be attributed to a large extent to reduced birth rate, greater economic development, more effective medical interventions (hence reductions in mortality), and to improvements in hygiene, nutrition, lifestyle and in general preventive care⁽ Reference Salomon, Wang and Freeman ^{1 )}.

Ageing is often defined as a continuous process that includes loss of functional capability and the increase in risk of morbidity and mortality⁽ Reference Franco, Karnik and Osborne ^{2 )}. Although ageing is a risk factor for chronic diseases, ageing is not necessarily defined by the presence of morbidities but is a process that starts early in life and which is modulated by the accumulation of experiences and exposures throughout the life-course⁽ Reference Franco, Karnik and Osborne ^{2 )}. As suggested earlier, a ‘healthy ageing phenotype’ can be defined as ‘the condition of being alive, while having highly preserved functioning metabolic, hormonal and neuro-endocrine control systems at the organ, tissue and molecular levels‘⁽ Reference Franco, Karnik and Osborne ^{2 )}. This phenotype represents a state of adequate resilience that minimises the risk of major chronic conditions such as CVD and dementia, but also represents the maintenance of cognitive and physical functioning⁽ Reference Franco, Karnik and Osborne ^{2 )}. Although life expectancy has increased in the past decade, following improved care and diagnosis, the years lived with disabilities and diseases are increasing⁽ Reference Salomon, Wang and Freeman ^{1 )}, emphasising that the maintenance of a healthy phenotype in later life has become a major public health challenge. Dietary habits are one of the important modifiable factors that can affect the maintenance of a healthy ageing phenotype⁽ Reference Mathers ^{3 )}. For example, high intake of both n-6 and n-3 PUFA have been found to reduce the risk of CHD⁽ Reference Willett ^{4 )}, whereas trans-fatty acids intake is considered to increase CVD risk⁽ Reference Willett ^{4 ,} Reference Aronis, Khan and Mantzoros ^{5 )}.

One for all and all for one

Although nutrition has long been among the main areas of interest for scientists trying to identify the lifestyle factors that exacerbate or reduce the risk of age-related disorders, studies of the specific nutrients, single foods or food groups have multiple challenges⁽ Reference Hu ^{6 )}. A major difficulty in interpreting the outcomes of such studies in relation to health and disease is the high degree of correlation among dietary components. Moreover, the assumption that single foods or nutrients have isolated effects may not be valid when it comes to maintaining a healthy ageing phenotype since this may cover a combination of different pathways. In addition, foods and nutrients interact, or act in synergy, when influencing numerous metabolic processes. As a result, the focus of nutritional research in the last two decades has shifted to the analyses of whole diets or dietary patterns⁽ Reference Hu ^{6 )}. A dietary pattern approach could better capture the totality of diet, and may be a more appropriate approach for investigating diet–disease associations than focusing on single foods or nutrients⁽ Reference Hu ^{6 –} Reference Jones-McLean, Shatenstein and Whiting ^{8 )}.

A posteriori approaches to dietary pattern analysis are driven by the dietary data per se of the population under study and patterns are derived on the basis of correlations or clustering of various dietary factors (e.g. using principal component analysis or k-means clustering)⁽ Reference Ocke ^{9 )}. Conversely, in a priori approaches, researchers define scores or indices of the overall dietary quality on the basis of foods or diets associated with health or according to the specific dietary guidelines⁽ Reference Ocke ^{9 )}. While such dietary guidelines can play important roles in health promotion policies and in individual dietary counselling, they are often based on evidence of decreased risk of chronic diseases and/or prevention of nutrient deficiencies⁽ Reference Ocke ^{9 )} and their utility in promoting healthy ageing is uncertain. In several countries, adherence of these dietary guidelines is monitored regularly through national food surveys, but less is known about how well adherence to these dietary guidelines predicts health and development of age-related diseases. In recent years, several indices have been developed to assess adherence to these specific dietary guidelines in relation to health outcomes⁽ Reference Ocke ^{9 )}. In this review, we describe the existing evidence of several dietary patterns and dietary guidelines in relation to aspects of healthy ageing and we attempt to identify the key characteristics, which they have in common. Emphasis has been placed on life expectancy, cardiometabolic diseases and cognitive outcomes, but not on cancer because of its heterogeneity in the relationship between diet and different cancer sites.

A Posteriori defined regional dietary patterns

A posteriori defined dietary patterns have frequently been used in epidemiological studies. One advantage of this approach is that it is data-driven, and, thus, independent of previous hypotheses and not limited by the current knowledge in nutrition⁽ Reference Ocke ^{9 )}. However, due to the different populations that are studied, the specific dietary patterns that are associated with health outcomes can be cultural-specific⁽ Reference Tucker ^{10 )} and thus difficult to translate to other settings.

Earlier regional dietary patterns have already been described to explain some of the cross-cultural variation in the prevalence of age-related chronic diseases. For example, the original Eskimo diet and Japanese diet are both characterised by relatively high intakes of fish, which is thought to explain the low rate of CHD in these populations⁽ Reference Kromhout ^{11 ,} Reference Feskens and Kromhout ^{12 )}. Later cross-cultural studies, such as the Seven Countries Study initiated in the late 1950s, suggested that fish intake per se may not be that important (fish consumption was not consistently associated with mortality patterns across countries), whereas low intakes of saturated fats may explain reduced mortality risk⁽ Reference Keys, Menotti and Karvonen ^{13 )}. The Mediterranean diet (MD), which can be characterised by high intake of vegetables, fruit, grains, fish and olive oil⁽ Reference Haber ^{14 )}, has received most attention in relation to mortality because of intriguing differences in chronic disease prevalence and mortality statistics in Mediterranean populations relative to other parts of the world^{( 15 )}, and its potential health effects have been acknowledged since Renaissance⁽ Reference Haber ^{14 )}.

More recently, the importance of diet has been further recognised in other parts of the world where populations exhibit greater healthy ageing, i.e. living longer without disabilities and being more physically active even when aged over 90 years. Population groups in Sardinia (Italy), Nicoya Peninsula (Costa Rica), Loma Linda (California) and Okinawa (Japan), have reputedly longer life expectancy and lower rates of age-related diseases (including CVD and dementia) compared with other countries⁽ Reference Buettner ^{16 )}. These areas have been designated as the ‘Blue Zones’ and are part of a large research project to discover lifestyle factors that explain healthy longevity⁽ Reference Buettner ^{16 )}. Despite their different ethnic and socioeconomic backgrounds, those living in these ‘Blue Zones’ share some common lifestyle characteristics including diets rich in plant-based foods (notably vegetables, beans, soyabean and lentils) and low in meat⁽ Reference Buettner ^{16 )} . These findings are consistent with most of the studies using the a posteriori dietary patterns approach within populations where lower risks of mortality and also of cardiometabolic disease and of poor cognitive outcomes are associated with higher intakes of plant-based foods and relatively low meat consumption⁽ Reference Fung, Schulze and Manson ^{17 –} Reference Ozawa, Ninomiya and Ohara ^{22 )}.

A priori defined dietary patterns: dietary guidelines and dietary scores

A priori defined dietary patterns are generally based on evidence showing that the dietary patterns are important to health. In practical terms, diets are assessed for the presence or absence of a certain amount of foods or nutrients and the resulting score is then operationalised as a dietary index. Generally, three groups of a priori defined dietary patterns can be distinguished: (a) dietary scores based on the national or international dietary guidelines, (b) scores derived from assessment of concordance with nutrients or foods associated with the specific health outcomes and (c) scores based on characteristics of regional diets, e.g. an MD pattern (Table 1)⁽ Reference Ocke ^{9 )}.

Table 1.

Examples of a priori developed dietary patterns

HEI, healthy eating index; MD, Mediterranean diet; HDI, healthy diet indicator.

* With emphasis on mortality, cardiometabolic disease and cognitive outcomes.

From observation to intervention?

Although, several indices of MD patterns as well as adherence to the DASH diet and other healthy diet scores are associated with a lower risk of stroke, CHD and CVD in numerous prospective cohort studies in different populations⁽ Reference Salehi-Abargouei, Maghsoudi and Shirani ^{69 ,} Reference Sofi, Cesari and Abbate ^{80 –} Reference Martinez-Gonzalez, Garcia-Lopez and Bes-Rastrollo ^{84 )}, residual confounding by other lifestyle and socioeconomic factors remains a major challenge in determining causality. While a trial of the DASH diet has demonstrated that adherence to DASH can reduce numerous cardiovascular risk factors⁽ Reference Chen, Maruthur and Appel ^{85 )} the causal effects on mortality and on other age-related end points are unknown.

The benefit of adherence to an MD pattern on primary prevention of cardiovascular endpoints was confirmed recently in a randomised controlled trial (PREDIMED) conducted in Spain. In this study, an MD supplemented with either extra-virgin olive oil or nuts reduced the risk of major cardiovascular events by almost 30%⁽ Reference Estruch, Ros and Salas-Salvado ^{86 )}. Also, results from the PREDIMED trial showed that this intervention improved cognition (0·6 points increase in score on the Mini-Mental State Examination and Clock Drawing Tests) in participants aged 75+ years with high vascular risk⁽ Reference Martinez-Lapiscina, Clavero and Toledo ^{87 )}. Although whether the PREDIMED trial is a test of an MD per se or of the effects of supplemental extra-virgin olive oil or nuts can be debated, the study provides strong evidence that greater adherence to this dietary pattern can have substantial beneficial effects on health in the older people and reduces the risk of common age-related diseases⁽ Reference Estruch, Ros and Salas-Salvado ^{86 ,} Reference Martinez-Lapiscina, Clavero and Toledo ^{87 )}. However, before such findings can be translated to public health-level interventions regionally or nationally, more evidence is needed about the most effective intervention modalities for achieving the necessary changes in eating behaviour and to sustain those behavioural changes in the long term⁽ Reference Curtis, Adamson and Mathers ^{88 )}. In addition, there is a need for an objective panel of outcome measures, which can be used to ascertain whether the interventions have been successful in enhancing healthy ageing. Recently, a preliminary panel of such measures (based on the healthy ageing phenotype concept⁽ Reference Franco, Karnik and Osborne ^{2 )}) which is designed to assess the efficacy of lifestyle-based interventions in promoting healthy ageing in community settings has been proposed⁽ Reference Lara, Godfrey and Evans ^{89 )}.

Nutrition and healthy ageing: agreement on the key ingredients

Dietary patterns define relevant and modifiable risk factors for deviations from a healthy ageing phenotype. Using both a posteriori and a priori dietary patterns, some key ‘ingredients’ (food groups) can be identified that are associated consistently with longevity and with better cardiometabolic and cognitive health. These include: high intakes of fruits, vegetables, fish, (whole) grains and legumes/pulses and potatoes (Fig. 1).

Fruit, vegetables and (whole) grains are components included in the HEI-2005, HEI-1995, Recommended Food score and the WHO HDI, but are also found in indices of the MD (Table 1). All of these indices and also a posteriori defined dietary patterns that include these food groups have been associated with lower risk of mortality, cardiometabolic disease and adverse cognitive outcomes in middle-aged and older people.

Although fish has been included recently in the HEI-2010, we are not aware that this new index has yet been studied in relation to healthy ageing outcomes. However, fish is a component in most Mediterranean diets and has been included in other indices such the DASH diet, the PNNSG score and the Polymeal as well as in a posteriori defined dietary patterns that have been linked with healthy ageing outcomes.

Finally, legumes/pulses and potatoes are not included in all a priori defined dietary patterns but those, which included one or more of these components (e.g. HEI-2005, PNNSG, HDI and MD scores) have found that they are associated with lower mortality rates and decreased risk of cardiometabolic diseases and of poor cognitive outcomes.

In contrast, dietary patterns rich in red meat and sugar-rich foods have been associated with an increased risk of mortality, type 2 diabetes mellitus and CHD in studies that defined these patterns a posteriori ⁽ Reference Fung, Schulze and Manson ^{17 –} Reference Hu, Rimm and Stampfer ^{20 ,} Reference Anderson, Harris and Tylavsky ^{25 ,} Reference Heidemann, Schulze and Franco ^{28 ,} Reference Meyer, Doring and Herder ^{31 ,} Reference Heroux, Janssen and Lam ^{32 ,} Reference Hlebowicz, Persson and Gullberg ^{35 ,} Reference Malik, Fung and van Dam ^{37 –} Reference Odegaard, Koh and Butler ^{39 )}. There is a general agreement on the potential deleterious effect of higher intakes of these two food items and most of the dietary guidelines and scores designate these foods as unfavourable components (Table 1). However, not all a priori defined dietary patterns include these components (e.g. they are not included in the Recommended Food Score, the Danish Diet quality Index, the PNNSG score and WHO HDI). It can be argued that these guidelines might be improved (made more sensitive) by adding specific recommendations on the intake of red meat and of sugar-rich foods and that such modified guidelines may be better at identifying those with eating behaviours more likely to lead to a healthy ageing phenotype.

Fig. 1.

(colour online) Nutrition and healthy ageing: the key ingredients.

Other food groups/items are the subject of controversy. For example, the role of dairy products in healthy ageing remains unclear. Although, some a priori dietary patterns (e.g. HEI-1995, -2005, -2010, Recommended Food Score, PNNSG and DASH) have included dairy foods as a beneficial component, none of the MD scores includes dairy products as a beneficial component. In addition, a posteriori derived dietary patterns do not show consistent relationships between patterns rich in dairy and beneficial health outcomes⁽ Reference Anderson, Harris and Tylavsky ^{25 ,} Reference Maruyama, Iso and Date ^{27 ,} Reference Waijers, Ocke and van Rossum ^{30 ,} Reference Granic, Andel and Dahl ^{90 )}.

It could be argued that the relationship between dairy-rich dietary patterns and health outcomes are culturally specific and are influenced by the habitual dairy consumption of the population under study and the type of dairy products consumed. For example, the Netherlands and USA are countries with the highest milk consumption worldwide^{( 91 )}, but increased mortality risk was associated with dietary patterns characterised particularly by fat-rich dairy products⁽ Reference Anderson, Harris and Tylavsky ^{25 )}. Future studies in this field need to address the effects of dairy-rich dietary patterns on health outcomes in specific subpopulations.

The totality of available evidence suggests that dietary patterns rich in fruit, vegetables, fish, whole grains and starchy low-fat staple foods are likely to play a key role in promoting aspects of healthy ageing including life expectancy and lower risk of cardiometabolic diseases and adverse cognitive outcomes. However, the current evidence is derived mainly from observational studies and more evidence from intervention studies on the effect of whole diets and on how to achieve the necessary sustained dietary behaviour changes is needed to support the development of soundly based public health interventions to promote healthy ageing at the population level. Furthermore, although regional diets such as the MD may be beneficial in other geographical locations, interventions promoting their use may be influenced by local availability, population-specific effects and by culture-specific acceptability of particular foods and food groups. A more tailored approach is needed.

Personalised nutrition: the next frontier?

Most of the current dietary guidelines adopt a ‘one size fits all’ approach and this needs to change since not all individuals within populations will share the same dietary risks and there is considerable heterogeneity in food preferences. Indeed, a more personalised approach to dietary interventions may be both more acceptable and more efficacious.

Future nutritional research should recognise this challenge and aim to disentangle the complex interactions among genotype, diet, lifestyle and environmental factors to understand the basis for varying vulnerabilities and the different responses to dietary patterns⁽ Reference Joost, Gibney and Cashman ^{92 )}. Addressing these interactions may provide the basis for a more personalised approach to nutritional recommendations and to nutritional counselling to improve eating patterns and to promote healthy ageing.