Inadequate dietary habits are the most preventable risk factor with impact on worldwide non-communicable diseases morbidity and mortality(1,Reference Roth, Abate and Abate2) , reflecting the importance of studying and exploring dietary choices and their determinants. Meanwhile, home cooking practices are declining, while the consumption of food prepared away from home is gaining importance in human diets worldwide(Reference Fanzo and Davis3–Reference Díaz-Méndez and García-Espejo6). Eating out of home has been identified as a risk factor for higher intakes of energy, total and saturated fat, sugar and sodium and lower intakes of fibre and micronutrients, as well as higher intakes of soft drinks, sugar-sweetened beverages (SSB), fruit juices, sweets, beer and other alcoholic beverages and lower intakes of dairy products, fruit, fish and seafood and vegetables(Reference Lachat, Nago and Verstraeten7–Reference Orfanos, Naska and Rodrigues11).
Overall, eating out of home usually translates into poorer diet quality. The definition of what constitutes eating out of home, however, lacks methodological consensus. Two distinct definitions have been considered in the literature: (i) all food and beverages consumed away from home, regardless of whether they were prepared at or away from home; and (ii) all food and beverages purchased away from home, irrespective of the consumption place(Reference Orfanos, Naska and Trichopoulos12). This lack of a common definition makes direct comparisons of out-of-home intakes across population groups difficult – which, consequently, does not facilitate the design of public health policies to encourage consumers to eat healthier when they are eating out of home. Moreover, meals are often not consumed at the same place where they were prepared; meals consumed at home, for instance, may be acquired from a restaurant, canteen, cafeteria or similar establishments or from takeaway or home delivery services(Reference Jekanowski13,Reference Guthrie, Lin and Frazao14) .
Studies of patterns of food preparation at the population level are geographically limited and mainly focused on adults(Reference Möser4–Reference Díaz-Méndez and García-Espejo6,Reference Virudachalam, Long and Harhay15–Reference Adams, Goffe and Brown21) , while studies of their associations with dietary intakes are equally scare(Reference Wolfson and Bleich22–Reference O’Dwyer, McCarthy and Burke30). In Portugal, one study investigated the time adults spent on food preparation at home in 2015(Reference Perista, Cardoso and Brázia31), and another examined adult food intake per location between 1999 and 2003(Reference Naska, Katsoulis and Orfanos9,Reference Orfanos, Naska and Rodrigues11,Reference Orfanos, Naska and Trichopoulos12) . More recently, a study identified patterns of food preparation for both adults and children and uncovered their associations with sociodemographics, health status and lifestyle factors(Reference Rei, Correia and Torres32). Still, there is a lack of research in Portugal to understand dietary intake based on whether food and beverages were prepared at or away from home. In view of this, the aim of the present study was to investigate associations between dietary intakes and patterns of food preparation – defined as all food and beverages purchased away from home, irrespective of the place of consumption – by age group, drawing from data collected by the most recent National Food, Nutrition and Physical Activity Survey (IAN-AF 2015/2016).
Methods
Study design and participants
This cross-sectional study included participants from the IAN-AF 2015/2016, aged between 3 and 84 years old and with two completed dietary intake assessments (n 5005). The IAN-AF participants were randomly selected from the National Health Registry by multistage sampling and comprised a representative sample of the general (non-institutionalised) resident Portuguese population(Reference Lopes, Torres and Oliveira33).
Data collection
Data were collected through interviews (8–15 d apart) conducted by trained researchers (with a background in Nutrition or Dietetics), using an electronic platform (You eAT&Move), between October 2015 and September 2016. This period covered the four seasons and included all days of the week (randomly selected), to incorporate seasonal effects and day-to-day variation in food consumption(Reference Lopes, Torres and Oliveira33). For children (< 10 years old), food consumption data were collected by two non-consecutive days of 24-h food diaries followed by a face-to-face interview with the main caregivers to gather details related to food description and quantification(Reference Lopes, Torres and Oliveira33). For other age groups, food consumption data were collected by two non-consecutive 24-h recalls(Reference Lopes, Torres and Oliveira33).
The eAT24 module, incorporated in the You eAT&Move platform used to collect and describe food consumption data, utilises the Portuguese food composition table by default for nutrient analysis. Portion sizes were calculated using multiple methods available within this module, specifically the food picture book method (which includes a set of 186 food photo series with six portions each), as well as the household measure photo series. Additionally, other methods such as weight/volume, standard unit and household measure methods were also available. Further details about the IAN-AF protocol and methodology can be found elsewhere(Reference Lopes, Torres and Oliveira33).
Patterns of food preparation
Dietary intake of participants were previously cluster analysed according to the place of food preparation, allowing the identification of four distinct patterns of food preparation in the Portuguese population(Reference Rei, Correia and Torres32): (1) the pattern At home by themselves – characterised by having the highest proportion (38·0 %) of intakes from foods and beverages prepared at home by individuals; (2) the pattern At home by relatives or friends – characterised by the highest proportion (44·7 %) of intakes from foods and beverages prepared at home by relatives or friends of individuals; (3) the pattern By food retail – characterised by the highest proportion (65·3 %) of intakes of foods and beverages prepared or acquired from food retail operators (including those consumed in natura); and (4) the pattern By restaurants, canteens and other away-from-home establishments – characterised by the highest proportions of intakes of foods and beverages prepared by restaurants (13·9 %), canteens (10·8 %) and other foodservice establishments (9·6 %).
Sociodemographic characteristics
Participants’ sex, age (grouped as children (3–9 years), adolescents (10–17 years), adults (18–64 years) or elderly (65–84 years)), area of residence (classified as Predominantly Urban, Medium Urban or Predominantly Rural according to the Portuguese National Institute of Statistics’ Classification of Urban Areas)(34) and highest education completed (categorised as No Education/Primary Education, Secondary Education or Tertiary Education). In the case of children and adolescents, the highest education completed by their parents was considered.
Diet quality
The Healthy Eating Score (HES) was used to evaluate dietary adequacy. This score, previously used with Portuguese children and adolescents(Reference Vilela, Muresan and Correia35), is based on dietary recommendations proposed by the WHO. This score comprises intake information from the following nine groups: (1) FV – fresh fruits, vegetables and legumes; (2) cereals and potatoes – rice, pasta, potatoes, bread and other grains; (3) dairy products – milk, yogurt and cheese; (4) white meat, fish and eggs – eggs, chicken, turkey, rabbit, fresh, dry and canned fish; (5) red and processed meat – for example, pork, beef, ham, sausage, bacon; (6) salty snacks – salty snacks, pizzas and chips; (7) SSB – sugar-sweetened beverages, such as fruit nectars and soft drinks; (8) sugar and honey – added/free sugar, honey, molasses and syrups; (9) sweets – cakes, candies, sweet pastry, chocolate, biscuits and ice cream, breakfast cereals and cereals bars. Groups 1–4 are considered healthier, so lower consumption scored lower and higher consumption scored higher. Contrariwise, the remaining five groups with less/non-healthy foods were assigned inverse scores. The HES final score ranges from 9 to 36, with higher scores translating into healthier and higher-quality diets.
Statistical analysis
All statistics were performed using the R Software version 3.6.3(36), the sample was weighted for the distribution of the Portuguese population using the library ‘survey’(Reference Lumley37), and the significance level considered was α = 0·05. The distribution of the patterns of food preparation was cross-tabulated with the distribution of participants’ sex, age group, area of residence and completed education. As less than 0·5 % of children and adolescents were clustered in the pattern of food preparation At home by themselves, this pattern was concatenated with the pattern of food preparation At home by relatives or friends, to form an overall pattern of food preparation At home (specifically, for these age groups). Due to the different number of patterns of food preparation for children and adolescents and for adults and elderly, the subsequent analyses were performed independently per non-adult (< 18 years) and adult (≥ 18 years) participants.
Linear regression models were used to obtain crude and adjusted standardised beta coefficients (β) and the respective 95 % CI, to study the association between the patterns of food preparations and 2-d mean daily intake of energy, of some nutrients and of nine HES food and beverage groups and the mean HES. All final models were adjusted for sex, age group, area of residence and education. For fibre, sodium and the nine HES food and beverage groups, the final models were additionally adjusted for total energy intake. Moreover, in adults and elderly, logistic regression models were used to obtain crude and adjusted OR and the respective 95 % CI to study the association between the patterns of food preparation and consuming alcoholic beverages, in particular wine, beer and other alcoholic beverages. The final models were adjusted for total energy intake, sex, age group, area of residence and education.
Results
The most prevalent pattern of food preparation is the pattern By restaurants, canteens and other away-from-home establishments (45·9 %, 95 % CI = 43·8, 48·1 %), being mostly adopted by men, younger individuals, residents in urban areas or more educated ones (Table 1).
Distribution of the patterns of food preparation according to sociodemographic characteristics, weighted for the distribution of the Portuguese population (Percentages and 95 % confidence intervals)
n, absolute frequency; %, relative frequency.
* Education completed by parents for participants < 18 years.
The boldface highlights the statistically significant results for the associations under study.
Dietary intake according to patterns of food preparation in children and adolescents
Table 2 shows the associations between mean daily intake of energy and nutrients and the patterns of food preparation among children and adolescents in Portugal, taking the pattern of preparation At home as reference. Adjusted results show that non-adults in the pattern By food retail (including food consumed in natura) have significant positive associations for daily intakes of saturated and trans fatty acids (respectively, β = 0·91 %TEI, 95 % CI = 0·40, 1·42 %TEI and β = 0·14 %TEI, 95 % CI = 0·06, 0·21 %TEI), total carbohydrate (β = 1·43 %TEI, 95 % CI = 0·23, 2·63 %TEI) and free sugars (β = 3·34 %TEI, 95 % CI = 2·14, 4·53 %TEI) and significant negative associations for daily intake of protein (β = –1·82 %TEI, 95 % CI = –2·50, –1·14 %TEI), fibre (β = –1·7 g, 95 % CI = –2·5, –0·8 g) and sodium (β = –181 mg, 95 % CI = –308, –54 mg). Moreover, non-adults in the pattern By restaurants, canteens and other away-from-home establishments have significant positive associations for daily intakes of energy (β = 166 kcal, 95 % CI = 77, 256 kcal) and total carbohydrate (β = 1·52 %TEI, 95 % CI = 0·45, 2·59 %TEI) and a significant negative association for daily intake of protein (β = –1·45 %TEI, 95 % CI = –2·06, –0·85 %TEI). No significant associations between other nutrients and the patterns of food preparation were observed.
Associations between mean daily intake of energy and nutrients and the patterns of food preparation in children and adolescents, weighted for the distribution of the Portuguese population (Mean values with their standard errors; standardised beta coefficient and 95 % confidence intervals)
β, standardised beta coefficient; %TEI, percentage of total energy intake; g, grams; mg, milligrams.
* Model adjusted to sex, age group, area of residence and parents’ education.
** For fibre and sodium, model 1 was adjusted for total energy intake, and model 2 was adjusted additionally for sex, age group, area of residence and parents’ education.
The boldface highlights the statistically significant results for the associations under study.
Table 3 shows the associations between mean daily intake of some IAN-AF 2015–2016 food and beverage groups, the HES and the patterns of food preparation among children and adolescents in Portugal, taking the pattern of preparation At home as reference. Adjusted results show that non-adults in the pattern By food retail (including food consumed in natura) have significant positive associations for daily intakes of dairy products (β = 80·4 g, 95 % CI = 40·4, 120·3 g), salty snacks (β = 14·9 g, 95 % CI = 4·1, 25·7 g), SSB (β = 112·3 g, 95 % CI = 72·1, 152·6 g) and sweets (β = 14·1 g, 95 % CI = 1·5, 26·7 g) and significant negative associations for daily intake of FV (β = –49·5 g, 95 % CI = –79·4, –19·6 g), cereals and potatoes (β = –79·3 g, 95 % CI = –98·3, –60·3 g) as well as white meat, fish and eggs (β = –31·4 g/d, 95 % CI = –43·9, –19·0 g). Furthermore, non-adults in the pattern By restaurants, canteens and other away-from-home establishments have a significant positive association for daily intake of dairy products (β = 53·7 g, 95 % CI = 17·9, 89·4 g) and significant negative associations for daily intake of cereals and potatoes (β = –32·2 g, 95 % CI = –49·2, –15·2 g) and white meat, fish and eggs (β = –28·4 g, 95 % CI = –39·6, –17·3 g). No significant associations between other food groups and the patterns of food preparation were observed.
Associations between mean daily intake of some IAN-AF 2015–2016 food and beverage groups, the Healthy Eating Score (HES) and the patterns of food preparation in children and adolescents, weighted for the distribution of the Portuguese population (Mean values with their standard errors; standardised beta coefficient and 95 % confidence intervals)
β, standardised beta coefficient; g, grams; FV, fresh fruits, vegetables and legumes; SSB, nectars and soft drinks.
* Model adjusted to total energy intake, sex, age group, area of residence and parents’ education.
The boldface highlights the statistically significant results for the associations under study.
Finally, both children and adolescents falling the pattern By food retail (including food consumed in natura) or the pattern By restaurants, canteens and other away-from-home establishments have significant negative association for HES (respectively, β = –1·8, 95 % CI = –2·4, –1·2 and β = –0·7, 95 % CI = –1·3, –0·2).
Dietary intake according to patterns of food preparation in adults and elderly
Table 4 shows the associations between mean daily intake of energy and nutrients and the patterns of food preparation among adults and elderly in Portugal, taking the pattern of preparation At home by themselves as reference. Adjusted results show that all adults in the pattern At home by relatives or friends have significant positive associations for daily intakes of fat (β = 0·92 %TEI, 95 % CI = 0·13, 1·71 %TEI), saturated and trans fatty acids (respectively, β = 0·61 %TEI, 95 % CI = 0·26, 0·96 %TEI and β = 0·05 %TEI, 95 % CI = 0·02, 0·08 %TEI) and free sugars (β = 1·02 %TEI, 95 % CI = 0·35, 1·68 %TEI) and a significant negative association for daily intake of fibre (β = –1·4 g, 95 % CI = –2·0, –0·7 g). Those in the pattern By food retail (including food consumed in natura) have significant positive associations for saturated and trans fatty acids (respectively, β = 0·82 %TEI, 95 % CI = 0·54, 1·11 %TEI and β = 0·08 %TEI, 95 % CI = 0·05, 0·11 %TEI), total carbohydrate (β = 1·00 %TEI, 95 % CI = 0·19, 1·80 %TEI) and free sugars (β = 1·95 %TEI, 95 % CI = 1·41, 2·50 %TEI) and significant negative associations for daily intakes of protein (β = –0·43 %TEI, 95 % CI = –0·84, –0·01 %TEI), fibre (β = –1·3 g, 95 % CI = –1·9, –0·8 g) and sodium (β = –123 mg, 95 % CI = –200, –46 mg). Additionally, adults and elderly in the pattern By restaurants, canteens and other away-from-home establishments have significant positive associations for daily intakes of energy (β = 95 kcal, 95 % CI = 37, 153 kcal), fat (β = 1·53 %TEI, 95 % CI = 0·90, 2·15 %TEI), saturated and trans fatty acids (respectively, β = 0·98 %TEI, 95 % CI = 0·70, 1·26 %TEI and β = 0·05 %TEI, 95 % CI = 0·02, 0·08 %TEI) and free sugars (β = 1·74 %TEI, 95 % CI = 1·26, 2·31 %TEI) and significant negative associations for daily intakes of total carbohydrate (β = –1·65 %TEI, 95 % CI = –2·43, –0·87 %TEI) and fibre (β = –2·2 g, 95 % CI = –2·7, –1·6 g). No significant associations between other nutrients and the patterns of food preparation were observed.
Associations between mean daily intake of energy and nutrients and the patterns of food preparation in adults and elderly, weighted for the distribution of the Portuguese population (Mean values with their standard errors; standardised beta coefficient and 95 % confidence intervals)
β, standardised beta coefficient; %TEI, percentage of Total Energy Intake; g, grams; mg, milligrams.
* Model adjusted to sex, age group, area of residence and education.
** For fibre and sodium, models 1 were adjusted for total energy intake and models 2 were adjusted additionally for sex, age group, area of residence and education.
The boldface highlights the statistically significant results for the associations under study.
Tables 5 and 6 show the associations between mean daily intake of some IAN-AF 2015–2016 food and beverage groups, consuming alcoholic beverages, the HES and the patterns of food preparation among adults and elderly in Portugal, taking the pattern of preparation At home by themselves as reference. Adjusted results show that adults in the pattern At home by relatives or friends have a significant positive association for daily intake of SSB (β = 25·7 g, 95 % CI = 4·7, 46·7 g) and a significant negative association for daily intake of FV (β = –49·1 g, 95 % CI = –73·0, –25·2 g). Adults in the pattern By food retail (including food consumed in natura) have significant positive associations for daily intake of dairy products (β = 81·7 g, 95 % CI = 64·6, 98·8 g), SSB (β = 56·0 g, 95 % CI = 38·8, 73·1 g) and sweets (β = 12·4 g, 95 % CI = 6·2, 18·4 g) and significant negative association for daily intake of FV (β = –37·8 g, 95 % CI = –57·3, –18·2 g), cereals and potatoes (β = –54·1 g, 95 % CI = –65·0, –43·3 g) and white meat, fish and eggs (β = –13·7 g, 95 % CI = –21·7, –5·6 g). Moreover, those in the pattern By restaurants, canteens and other away-from-home establishments have significant positive associations for daily intake of red and processed meat (β = 9·1 g, 95 % CI = 2·5, 15·6 g), salty snacks (β = 11·6 g, 95 % CI = 6·7, 16·5 g), SSB (β = 56·2 g, 95 % CI = 39·6, 72·9 g) and sweets (β = 9·6 g, 95 % CI = 3·5, 15·6 g) and significant negative associations for daily intake of FV (β = –68·8 g, 95 % CI = –87·8, –49·9 g) and cereals and potatoes (β = –48·0 g, 95 % CI = –58·4, –37·4 g). These adults and elderly also have significant higher odds of consuming alcoholic beverages such as spirits and liquors (OR = 1·82, 95 % CI = 1·11, 2·99). No significant associations between other food groups and the patterns of food preparation were observed.
Associations between mean daily intake of some IAN-AF 2015–2016 food and beverage groups, the Healthy Eating Score (HES) and the patterns of food preparation in adults and elderly, weighted for the distribution of the Portuguese population (Mean values with their standard errors; standardised beta coefficient and 95 % confidence intervals)
β, standardised beta coefficient; g, grams; FV, fresh fruits, vegetables and legumes; SSB, nectars and soft drinks.
* Model adjusted to total energy intake, sex, age group, area of residence and education.
The boldface highlights the statistically significant results for the associations under study.
Associations between consuming alcoholic beverages and the patterns of food preparation in adults and elderly, weighted for the distribution of the Portuguese population (Numbers and percentages; odds ratios and 95 % confidence intervals)
n, absolute frequency; %, relative frequency.
* Model adjusted to total energy intake, sex, age group, area of residence and education.
The boldface highlights the statistically significant results for the associations under study.
Lastly, all adults falling the pattern At home by relatives or friends, the pattern By food retail (including food consumed in natura) or the pattern By restaurants, canteens and other away-from-home establishments have significant negative association for HES (respectively, β = –0·5, 95 % CI = –0·8, –0·1 and β = –0·5, 95 % CI = –0·8, –0·2 and β = –1·2, 95 % CI = –1·5, –0·9).
Discussion
Studies exploring dietary intakes based on distinct patterns of food preparation are still scarce, especially those using national representative samples covering children and adult diets(Reference Goffe, Rushton and White26,Reference Burke, McCarthy and O’Neill28–Reference O’Dwyer, McCarthy and Burke30) . Therefore, the present study represents an important attempt to close this gap by performing a cross-sectional analysis of the IAN-AF 2015/2016 data to investigate associations between dietary intakes and patterns of food preparation, previously uncovered in the Portuguese population(Reference Rei, Correia and Torres32). Its findings indicate that children and adolescents eating more foods prepared or acquired from food retail operators have higher intakes of saturated and trans fatty acids, total carbohydrate and free sugars and lower intakes of protein, fibre and sodium than those eating more foods prepared at home. Moreover, they have higher intakes of dairy products, salty snacks, SSB and sweets and lower intakes of FV, cereals and potatoes, white meat, fish and eggs. Similarly, children and adolescents eating more foods prepared by restaurants, canteens and other foodservice establishments have higher intakes of energy and total carbohydrate and lower intakes of protein. They also have higher intakes of dairy products and lower intakes of cereals and potatoes, white meat, fish and eggs. In children (aged 5–12 years) from the Republic of Ireland, the findings are comparable: the percentage of energy from protein was significantly higher at home, while the percentage of energy from fat was significantly higher outside the home(Reference Burke, McCarthy and O’Neill28). Additionally, they have significantly higher intakes of fibre at home than at other people’s homes or outside the home(Reference Burke, McCarthy and O’Neill28). In the United Kingdom, children (aged 1·5–18 years) eating takeaway meals at home at least weekly have significantly higher intakes of energy than those eating takeaway meals at home rarely; however, the same association was not found regarding the frequency of consumption of meals out(Reference Goffe, Rushton and White26).
The results of the present study also indicate that adults and elderly eating more foods prepared at home by relatives or friends have higher intakes of total fat, saturated and trans fatty acids and free sugars and lower intakes of fibre compared with adults and elderly eating more foods prepared at home by themselves. They also have higher intakes of SSB and lower intakes of FV. Furthermore, those eating more foods prepared or acquired from food retail operators have higher intakes of saturated and trans fatty acids, total carbohydrate and free sugars and lower intakes of protein, fibre and sodium, as well as higher intakes of dairy products, SSB and sweets and lower intakes of FV, cereals and potatoes, white meat, fish and eggs. Lastly, adults and elderly eating more foods prepared by restaurants, canteens and other foodservice establishments have higher intakes of energy, total fat, saturated and trans fatty acids and free sugars and lower intakes of total carbohydrate and fibre. They also have higher intakes of red and processed meat, salty snacks, SSB, sweets and alcoholic beverages such as spirits and liquors and lower intakes of FV and cereals and potatoes. Findings from other countries are quite comparable: in adults (aged ≥ 20 years) from the USA, cooking dinner at home rarely (0–1 times per week) is associated with higher intakes of energy, carbohydrates, fat and sugar but lower intakes of fibre than cooking dinner at home more frequently (6–7 times per week), and it is also associated with eating more fast-food meals as well as frozen/ready-to-eat meals(Reference Wolfson and Bleich22). In Korean adults (aged 20–64 years), eating two or more meals per day prepared outside home reflects into higher intakes of energy, protein and fat and lower intakes of fibre and micronutrients (such as phosphorus, potassium, niacin and vitamin C – highly presented in FV) than eating two or more meals per day prepared at home(Reference Lee, Song and Cho24). In the United Kingdom, adults (aged ≥ 18 years) who eat meals out or takeaway meals at home at least weekly have higher intakes of energy(Reference Goffe, Rushton and White26), while adults (aged 29–64 years) who eat home-cooked meals more frequently have higher intakes of fruit and vegetables(Reference Mills, Brown and Wrieden25). In adults (aged 18–64 years) from the Republic of Ireland, eating more food prepared at home reflects into higher intakes of fibre, while eating more food prepared outside the home reflects into higher intakes of alcohol and a contribution of fat to energy above the recommendations(Reference O’Dwyer, Gibney and Burke29,Reference O’Dwyer, McCarthy and Burke30) .
Overall, in Portuguese children and adolescents, the pattern of food preparation associated with poorer diet quality was the pattern By food retail (including food consumed in natura), while in Portuguese adults and elderly, it was the pattern By restaurants, canteens and other away-from-home establishments. This study thus highlights that consuming food and beverages prepared away from home reflects in unhealthier and lower-quality diets among Portuguese non-adults and adults. Evidence from other countries with adult populations support our findings: in adults (aged ≥ 20 years) from the USA, cooking more frequently at home is associated with higher Healthy Eating Index-2015 score(Reference Wolfson, Leung and Richardson23), and in adults (aged ≥ 19 years) from the United Kingdom, eating more home-prepared food is associated with higher odds of being in the quintile most accordant with the Dietary Approaches to Stopping Hypertension (DASH) diet(Reference Clifford Astbury, Penney and Adams27). Also, in a population-based cohort study of adults (aged 29–64 years) from the United Kingdom, eating more frequently home-cooked meals is associated with higher adherence to DASH and Mediterranean diets(Reference Mills, Brown and Wrieden25). Nevertheless, as far as authors know, there is a lack of evidence regarding associations between patterns of food preparation and diet-quality scores (such as Healthy Eating Index-2015 score, DASH and Mediterranean Diet Score) in nationwide non-adult populations.
Strengths and limitations
This study used data from a representative sample, enabling a wider application of the findings. It is important to note that the data used was gathered by the IAN-AF 2015/2016, which might not accurately portray the current dietary status of the Portuguese population. Still, these data remain the most recent and comprehensive dietary data that are available for the country. Furthermore, despite the potential for residual confounding, regression models were adjusted to account for several relevant potential confounders.
Conclusion
Eating more foods prepared at home is associated with healthier and higher-quality dietary intakes; however, within the Portuguese population, the consumption of food prepared away from home prevails. Overall, eating more foods prepared away from home is associated with higher intakes of energy, saturated and trans fatty acids and free sugars and lower intakes of protein, fibre and sodium. It is also associated with higher intakes of dairy products, salty snacks, SSB and sweets and with lower intakes of FV, cereals and potatoes and white meat, fish and eggs. Moreover, in adults and elderly, it is associated with higher intakes of red meat and processed meat as well as alcoholic beverages such as spirits and liquors. Hence, in upcoming public health nutrition initiatives and dietary guidelines, there should be a greater emphasis on advocating for home cooking as a way of fostering healthier dietary choices as well as on ensuring the nutritional quality of food service meals, which could be much improved, for situations when it is not possible to cook and eat meals at home.
Acknowledgements
The authors gratefully acknowledge the participants enrolled in IAN-AF 2015/2016, all members of the research team and the institutional support from the General Directorate of Health (DGS), the Regional Health Administration Departments, the Central Administration of the Health System (ACSS) and the European Food Safety Authority.
Financial support
The IAN-AF 2015/2016 received funding from the EEA Grants Program, Public Health Initiatives (grant number: PT06-000088SI3), and this study was supported by the Fundação Francisco Manuel dos Santos (https://www.ffms.pt/en) throughout the ‘How We Eat What We Eat – A Portrait of Meal Consumption in Portugal’ Project (grant number: 1042382FCEEINV328).
Competing interests
The authors declare that they have no conflicts to report.
Authorship
The authors’ contributions are as follows: Mariana Rei contributed to the conception and design of the study, data analysis, interpretation of the findings and wrote the manuscript; Daniela Correia contributed to the data analysis and writing-review and editing of the manuscript; Duarte Torres, Carla Lopes, Ana Isabel A Costa and Sara SP Rodrigues contributed to the conception and design of the study, interpretation of the findings and writing-review and editing of the manuscript. Ana Isabel A Costa and Sara SP Rodrigues were coordinators of the Project ‘How We Eat What We Eat – A Portrait of Meal Consumption in Portugal’. All authors read and approved the final manuscript.
Ethics of human subject participation
Ethical approval was obtained from the National Commission for Data Protection, the Ethical Committee of the Institute of Public Health of the University of Porto and the Ethical Commissions of each of the Regional Administration of Health. All participants were also asked to provide their written informed consent for participation according to the Ethical Principles for Medical Research involving human subjects expressed in the Declaration of Helsinki and the national legislation. Written agreements from the legal representative were required for children and adolescents below 18 years, and adolescents were also asked to sign the consent form together with their legal representative.
Open access
