Cheers, proost, saúde: Cultural, contextual and psychological factors of wine and beer consumption in Portugal and in the Netherlands

ABSTRACT Wine and beer consumption are an integral part of European culture: Southern Europe is associated with wine and Northern Europe is associated with beer. When consumed in moderation, these alcoholic beverages can be part of a balanced and healthy diet. In the 1990s, non-alcoholic beer (NAB), which has no cultural roots, became available in the market. This review identifies determinants for consumption of wine, beer, and NAB, using data on consumption patterns from Portugal and the Netherlands. Since the 1960s the image of Portugal as a wine country declined, whereas the image of the Netherlands as a beer country remained stable. In each country beer is now the most consumed alcoholic beverage and is mainly a men's beverage, whereas wine is the second most consumed and is consumed by both genders. Cultural differences define Portuguese as “outdoors, everyday drinkers”, within a meal context, and Dutch as “at home, weekend drinkers.” Wine is perceived as the healthiest beverage, followed by NAB, and regular beer. Motivation for consumption is related to context: wine for special occasions, beer for informal occasions, and NAB for occasions when alcohol is not convenient. Moderate wine and beer consumption seems to be surrounded by positive emotions. This review is relevant for public health, for industry market strategies, and identifies opportunities of future research on drinking behaviour.


Introduction
Wine and beer consumption are an integral part of European life and culture, dating back to before the Roman era (Baeza, 2000;Poelmans and Swinnen, 2011). Wine and beer are embedded within social and cultural traditions that, in part, define the diets of many Europeans. From a nutritional perspective these are interesting beverages to study since patterns of consumption determine whether their effects on health and well-being are positive or negative. Excessive alcohol consumption is well recognized to increase the risk of certain diseases, such as cancers, cirrhosis and neurological problems (Foster and Marriott, 2006). However, when consumed in moderation wine and beer are accepted beverages that form part of a healthy, balanced diet, and lifestyle (Lindberg and Ezra, 2008;Arranz et al., 2012). Presently the median contribution of alcohol to the total energy intake (expressed as per adult per day) is 20.5 g (7%) in Portugal (BAP, 2003(BAP, -2008 and 15.1 g (5%) in the Netherlands (DNFCS, 2007(DNFCS, -2010. Drinking guidelines of both countries for daily alcohol consumption are: up to 42 g for men and 28 g for women in Portugal and, up to 20 g for men and 10 g for women in the Netherlands (ICAP, 2010).
Since the nineties, non-alcoholic beer (NAB) has become more present in the market (Nederlandse Brouwers: NB), partly as a result of drinking and driving concerns (social and legislative) but also as a response to general health concerns, and for religious reasons. This beverage has no cultural background and it is certainly not as popular as wine or beer itself. Nevertheless its presence in the market is an interesting development because NAB provides some of the sensory rewards and health benefits of beer itself, without the delivery of alcohol (Martin et al., 1990;Sohrabvandi et al., 2012).
In order to better understand how wine, beer and NAB consumers make their choices, it is necessary to know the main drivers of consumption of these beverages.
Many studies describe different aspects of alcohol consumption across the world, and these studies reveal that cultural, contextual and psychological factors are determinants of beverage choice (SIRC, 1998;Pettigrew, 2003;McCluskey and Shreay, 2011). However, there is no comprehensive overview of studies giving a clear picture of the different determinants of wine, beer and NAB consumption.
The objective of this paper is to review the scientific and public domain literature concerning wine, beer and NAB consumption in two countries that can be considered representative examples of Southern and Northern European culture, namely Portugal and the Netherlands. Within the scope of this review, an attempt is made to characterize the consumption determinants for these beverages which are prevalent in the two countries.
Knowledge of the determinants of consumption is relevant for public health policy and to help companies design marketing strategies.
This review begins by describing the intake of wine, beer and NAB in Portugal and in the Netherlands, covering cultural factors such as consumption patterns and consumer profile (gender, age, income, level of education). Subsequently, contextual factors of consumption are described, concerning occasions and places. Finally, information regarding psychological factors such as motivation for consumption is explored in addition to the perception of consumers concerning health and emotional attributes related to consumption.
2. The evolution of cultural differences of consumption 2.1 Per capita wine and beer consumption over time In terms of consumption, Portugal is classically perceived to be a wine country and the Netherlands a beer country. In the paragraphs below we will analyse whether or not these stereotypes concur with present consumption figures.
Estimates of the per capita consumption of wine and beer come from different sources. The World Health Organization (WHO) provides the contribution of wine and beer to the total per capita alcohol consumption (Fig. 1). This indicator is published per year and per country since the 1960s (WHO, 2013).
The WHO figures show that over the period 1960-2010, marked changes occurred in the patterns of alcohol consumption, in both countries (WHO, 2013). In Portugal the contribution of wine to the total per capita alcohol consumption decreased 39%, whereas beer consumption increased 29%. In the Netherlands, wine consumption increased 26% at the cost of spirit consumption (WHO, 2013). However the contribution of beer consumption remained stable and is still around 50% (WHO, 2013).
Data on the consumption per capita can also be obtained from European or global wine and beer industry organisations-e.g. Brewers of Europe (BoE) and the International Organisation of Vine and Wine (OIV). The data presented in Table 1 are derived from these organisations and in this paper they will be called commercial figures.
The consumption per capita published from commercial figures (Table 1) is calculated dividing the total of each beverage sold in the country by the total number of inhabitants, including people of all ages, independently of being drinkers or non-drinkers. Per capita consumption inferred in this way is an overestimation, in one hand, as it does not account for storage and waste. On the other hand is an underestimation because the number of drinkers is far less than the total number of people in a country. These commercial figures, however, can be considered to generally compare between countries.
Based Beer consumption across Europe remained relatively stable over the same period, showing a slight decrease in both Portugal and the Netherlands. In Portugal, per capita beer consumption declined 12% over 2003-2011, placing in 2011, the country in fourteenth position in Europe in terms of beer consumption, corresponding to 145 mL/day per capita. Over the same period, in the Netherlands per capita beer consumption declined by 10%, placing it in 2011 in ninth position in Europe, corresponding to 196 mL/day per capita beer consumption (Table 1). Its per capita consumption of 196 mL/day in 2011 placing it in ninth position in Europe in terms of beer consumption (Table 1).  Another source of consumption figures comes from government statistics. For this review we used the Portuguese Health National Survey (PHNS, 2005(PHNS, -2006 and the Dutch National Food Consumption Survey (DNFCS, 2007(DNFCS, -2010. These are reports based on national surveys focusing on individual dietary intake data reported by consumers. In these studies as in those above inferred from sales data, wine or beer drinking are not mutually exclusive. As shown in Table 2, the two countries applied different methodologies for data collection. In Portugal, in the PHNS all family members were interviewed face-to-face about the average per day consumed over the previous week. In the Netherlands, the DNFCS performed individual telephone interviews using two 24-hour recalls to selected people by invitation. Data collection based on interviews may have the disadvantage of under-reporting, a well-known source of systematic error in individual surveys (Ferro-Luzzi, 2002). The underreported foods vary from study to study and the reason behind it might be because consumers have low involvement with some foods, like snack foods, and those are easily forgotten when reporting daily food consumption (Dop, 2002). Regarding alcohol consumption underreporting can also be explained because it is socially a sensitive topic (Dop, 2002), and people may not wish to admit to drink as much as they do. Table 3 shows the consumption per capita as calculated from national surveys. In Portugal, on average, daily consumption of wine drinkers is around 300 mL and 600 mL for beer drinkers. In the Netherlands, the equivalent figures are around 200 mL for wine drinkers and 750 mL for beer drinkers (PHNS, 2005(PHNS, -2006DNFCS, 2007DNFCS, -2010. From these data values per capita consumption were extrapolated for both countries bearing in mind the percentage of drinkers from the adult population. Comparing the values from the two sources, i.e. commercial figures (Table 1) and national surveys (Table 3), there is a notable difference in the figures given of daily consumption, as can be seen in Table 4.
As previously mentioned, the calculation of per capita consumption was performed considering different primary data, and this has an understandable effect on the final outcome. Values derived from commercial figures are around twice those from national surveys (Table 4). This discrepancy may be explained by two main reasons. On one hand, data provided from self-reporting questionnaires might lead to an underestimation of consumption, especially as related to alcoholic beverages. On the other hand, the commercial figures are based on the whole population of a country (including children and other non-drinkers) to calculate per capita consumption. Nevertheless, the ratio between both sources (surveys and commercial figures) is relatively constant between countries and beverages. Hence, it seems acceptable to consider both sources representing a valid basis for comparison purposes even if in absolute terms there is a discrepancy between the data.
The following conclusions are observed from both sources: in Portugal wine consumption is twice as much as in the Netherlands. Beer consumption is far higher in the Netherlands than in Portugal. It is also higher in volume terms, than wine consumption in both countries.
From a health perspective, and in addition to patterns of consumption, it is important to factor in the relative alcohol contents of wine and beer. Wine contains 8-15% of alcohol by volume (German and Walzen, 2000) whereas beer contains 3-6% alcohol by volume (Bamforth, 2002). For Portugal wine contributes more than twice as much to the total alcohol consumed than beer. In the Netherlands beer is the greatest contributor to total alcohol consumption.
Looking at these figures, Portugal may still be considered as a wine drinking country, and the Netherlands as one which prefers beer. However, the difference in consumption patterns between the two countries is smaller than 50 years ago.

Non-alcoholic beer
A beer can be called non-alcoholic depending on the legal definitions, which can vary between countries. In Portugal NAB must contain less than 0.5% alcohol by volume (Portaria n. 1/ 96), whereas in the Netherlands up to 0.1% alcohol by volume is permitted (Bierverordening, 2002). This difference in alcohol content (0.4%) might have some effect on consumerspreferences. Not only because alcohol is an important flavour Table 2. Methodology for data collection, total number of participants and number of wine and beer drinkers, of Portuguese and Dutch national surveys (PHNS, 2005(PHNS, -2006DNFCS, 2007DNFCS, -2010 contributor but also because during the removal of alcohol inevitably other flavour substances are removed (Mieth, 1996). NAB was introduced to the market in the 1990s (BoE). It is a new beverage compared to wine or beer. NAB was developed mainly to appeal to consumers who would appreciate its health and wellbeing attributes, being strategically positioned between beer and soft drinks, even being presented by one brand as a regenerative sports beverage (DM, 2012). Despite the technological effort in creating innovative dealcoholisation solutions which are less compromising to the flavour of the original beer, NAB has not been a successful product in European markets. In both countries the market share of beer sales for NAB is still very small, corresponding to 2% in Portugal and to 1% in the Netherlands (2011). An exception to this is Spain in which NAB has a market share of beer sales of 10% (2011) (Associação Portuguesa de Cervejeiros: APCV; NB).
It is not possible to present useful data of NAB consumption on the basis of publicly available data. In Portugal no data has been published whilst in the Netherlands, the DNFCS returned 2523 respondents of which only 11 (0.4%) admitted being NAB consumers.

Socio-demographic profile of wine and beer consumers
Understanding the relations between consumption and consumer drinking profile is helpful for commercial strategies and public policies to be effectively targeted. Consumers' drinking profiles are often described in terms of gender, age, level of education and monthly income (e.g. Pettigrew, 2003;Ramful and Zhao, 2006;Colen and Swinnen, 2011;McCluskey and Shreay, 2011). From Table 5 it can be seen that in Portugal, men are the major consumers of wine and they drink twice as much as women. In the Netherlands the gender distribution for wine consumption is fairly equal, both in number of consumers and in amount of wine consumed. In both Portugal and the Netherlands, beer is largely consumed by men and this is also the case in other western societies (Pettigrew, 2002;Ramful and Zhao, 2006;Colen and Swinnen, 2011). The majority of consumers (mode is presented in bold numbers in Table 5) drink two glasses of wine or beer per day, with the exception of wine consumption by Portuguese, who drink three. In a worldwide comparison, according to the WHO (2013), Portugal and the Netherlands are countries that present low risk (1 in a 1 to 5 scale) regarding alcohol consumption-related health problems. The likeliness of an adult being a wine consumer apparently increases with age, starting with 30 years of age or older, reaching the highest percentage in the age group between 50-54 years old (Ramful and Zhao, 2006;Melo et al., 2010). Portuguese and Dutch seem not to be exceptions, as it is shown in Table 6 that wine consumers tend to be adults (35 years). Beer is most likely to be consumed until the middle age (54 years), in which the younger age group (34 years) has a relevant contribution, especially in the Netherlands (37%). A plausible explanation for this is that beer is normally a comparatively inexpensive beverage with a good price / quality relation which might be more relevant to younger consumers (DM, 2011b). In addition, the language of much beer marketing is considered easy to understand by young people, making it more accessible (DM, 2011b), whereas wine marketing language tend not to be focused on their generation and needs (Thach and Olsen, 2006). Moreover, several studies suggest that flavour seems to be the main driver for wine consumption (van Zanten, 2005;Pettigrew and Charters, 2006;Charters and Pettigrew, 2008;Jaeger et al., 2009;Duarte et al., 2010;Melo et al., 2010;Bruwer et al., 2012). It is also suggested that age and level of experience play an important role in wine flavour preference (Blackman et al., 2010;Melo et al., 2010). Thus, these arguments may help to understand why younger consumers are more likely to drink beer than wine.
Typically, Portuguese consumers of wine and beer have primary education and low monthly income, whereas the majority of Dutch consumers have an intermediate level of education and higher monthly incomes (Table 6). These socioeconomic indices are consistent with the general characterization of the whole population of both countries. For comparison, the modal value of income for Portugal is 905€/month in 2011 (BDPC, 2011) and for the Netherlands 2750€/month in 2013 (Expatax, 2013). Therefore income and level of education are variables that can distinguish between Portuguese and Dutch consumers, but not between wine and beer consumers. Age and gender Table 5. Percentage of consumers, frequency distribution and mean number of glasses of wine (100 mL) and beer (200 mL) consumed per day by gender in Portugal and in the Netherlands (PHNS, 2005(PHNS, -2006DNFCS, 2007DNFCS, -2010 seem to be the most differentiating factors between wine and beer consumers.

Occasions of consumption
In characterizing drinking behaviour it is interesting to consider consumption over different parts of the week, and in particular weekend versus weekdays (Fig. 2). The weekends are a prime period of alcohol consumption and in the sample of each country only 6% of consumers drink less or did not drink during the weekends. For Dutch consumers, a difference is noted in that they drink more during weekends than weekdays or only drink during weekends (61%) (DNFCS, 2007(DNFCS, -2010. In contrast, Portuguese consumers tend to drink the same amount (wine and beer) on weekdays as during the weekend (62%) (PHNS, 2005(PHNS, -2006. In the Netherlands the preferred time of the day for alcohol consumption is the late afternoon and during the evening. Beer is commonly consumed during the late afternoon whereas wine is associated with evening meals. According to the Dutch eating habits, lunch is in general a light and quick meal (Yuksel, 2007).
Alcohol consumption during lunchtime in the Netherlands is not common (AICEP, 2008;DNFCS, 2007DNFCS, -2010. This is different from Portuguese eating habits where lunch and dinner are considered to be of equally importance, and for that reason, have equal relevance in relation to wine or beer consumption (PHNS, 2005(PHNS, -2006. Besides main meals, Portuguese also commonly consume alcoholic beverages at night (IPSOS, 2009).

Wine
In general, wine is perceived as complementary to food and is associated with sit-down meals, either at home or in restaurants (Pettigrew and Charters, 2006;DM, 2011b, Thach, 2012. Wine consumption is described as being related to more formal occasions, special occasions and celebrations, being considered a social facilitator to interact with other people (Pettigrew, 2003;Ritchie, 2007;Thach, 2012). Choosing a wine is considered a ritual and wine choice and willingness to pay are influenced by the nature of the occasion (Quester and Smart, 1998;Pettigrew, 2003;Aqueveque, 2006;Ritchie, 2007;Thach, 2012). Wine consumers are willing to pay more for a bottle of wine to take to a dinner party at a friend's house at Saturday night, compared to a bottle of wine to drink at home during the week, by oneself or with one's family over dinner (Quester and Smart, 1998). Worldwide wine is the preferred choice with a meal as a relaxing beverage, at home or in a bar (DM, 2011b).

Beer
Similar to wine, beer consumption is also associated with social events but with more informal occasions, such as barbeques, large music festivals, sports events, high energetic night out with friends, as well as pub culture and friendships (Pettigrew and Charters, 2006;SIRC, 2008;DM, 2011b, McCluskey andShreay, 2011).
Beer is predominantly perceived to be an everyday beverage for relaxation-oriented occasions as the drinkers commonly use the beverage as a symbol of demarcation between work and non-work hours (Pettigrew, 2003;Pettigrew and Charters, 2006). Beer is also consumed in solitary occasions, like when at home relaxing and watching TV, whereas drinking wine alone is reported less (Pettigrew and Charters, 2006; Kim and   (PHNS, 2005(PHNS, -2006DNFCS, 2007DNFCS, -2010 Chintagunta, 2012). Beer is classified as a "flexible beverage" as its consumption might or might not be related with eating contexts, alone or with others (Pettigrew and Charters, 2006).

Non-alcoholic beer
NAB is apparently unrelated to specific consumption places. Consumers do not choose it based on the type of occasion, as with wine or beer, but as an alternative for when they cannot drink alcohol. In general, the main reasons for drinking NAB are because consumers have to drive, for medical reasons or pregnancy (Vlek and Peters, 2012).

Places of consumption
Place of consumption is a reflection of habits of different cultures and contributes to consumption decisions (Pettigrew, 2002). Referring to wine, it seems that domestic settings, including own homes, and the homes of friends or family, are the most common places for consumption in Portugal and in the Netherlands (AESBUC 1 , 2003;DNFCS, 2007DNFCS, -2010DM, 2011b). Additionally, in Portugal, restaurants are also considered as an important consumption place (AESBUC, 2003), whereas in the Netherlands restaurants are less representative for wine consumption (DNFCS, 2007(DNFCS, -2010.
In Portugal, beer sales take place mainly in the on-trade (69%, 2011), which include pubs, clubs, bars, and restaurants. Only 31% of beer sales take place off-trade, through supermarkets, shops and other retail outlets. This is in contrast with the European trend also reflected in the Netherlands, where ontrade beer sales have been decreasing to 26% and off-trade beer sales increasing to 74% (2011). At home is, according to the latest data, the preferred place for beer consumption in the Netherlands (BoE;DNFCS, 2007DNFCS, -2010.
Overall it seems that beverages are used according to the nature of the occasion: wine for special and relaxing moments and with a meal, beer for informal and more energetic occasions and NAB when it is not convenient to drink alcohol. For Portuguese, wine consumption takes place mainly at home and in restaurants whereas beer consumption takes place mainly in bars, pubs and restaurants. Portuguese consume wine and beer either during lunch, dinner or in the evening, during weekdays or weekends. For Dutch, the preferred moment for wine and beer consumption is in the evening, at weekends and often home related.

Motivation for consumption
Factors of impact on beverage choice and motivation for consumption may be derived from the drink by itselfdrink relatedand/or are context related (Fig. 3).
Wine and beer have different sensory attributes and the choice for one or the other, as a complement of food, seems to be based on creating a balance between food and beverage, providing pleasure (Charters and Pettigrew, 2008;Donadini et al., 2008). Nevertheless, wine is by far the beverage of choice for meal occasions and wine drinkers are more knowledgeable about food and alcohol combinations than beer drinkers (Pettigrew and Charters, 2006;DM, 2011b). The variety of tastes and flavours provided by wine are probably the most significant reasons for wine purchase and consumption for the majority of consumers (van Zanten, 2005;Pettigrew and Charters, 2006;Charters and Pettigrew, 2008;Jaeger et al., 2009;Duarte et al., 2010;Melo et al., 2010;Bruwer et al., 2012). For young generations (24-35 years age), however, wine flavour does not seem to be that important and their primary focus is the conviviality and enjoyment that wine drinking provides (Barrena and Sanchez, 2009;Duarte et al., 2010).
Wine and beer have different functional connotations, fulfilling different consumer needs. These functional connotations play a role in choice behaviour. For instance, wine is seen as a social facilitator to interact with other people whereas thirst relief is often mentioned in relation to beer consumption (Yang et al., 2002;Donadini et al., 2008;Barrena and Sanchez, 2009).
The anticipated emotions and feelings associated with the consumption of a beverage is also a motivation for the choice. For instance, wine consumers feel enjoyment when planning to drink wine with food (Pettigrew and Charters, 2006). Comparably beer consumers may expect to relax after drinking a beer (Yang et al., 2002;Kim and Chintagunta, 2012).
There is some evidence showing that wine consumers perceive an emotional benefit from drinking wine that predominates over functional factors (Charters and Pettigrew, 2008;Barrena and Sanchez, 2009). This suggests that consumers' choice might include an emotional component in addition to the product knowledge (Barrena and Sanchez, 2009).
Wine and beer are often consumed in a social context, mainly with friends or family, to socialize, to celebrate and to have fun (SIRC, 1998;McCluskey and Shreay, 2011). In a social context consumption is, to some extent, influenced by peers, friends or family, either to seek approval of the behaviour (van Zanten, 2005), image management (Pettigrew, 2002;McCluskey and Shreay, 2011) or to impress others (Melo et al., 2010). Social pressure, mainly used by friends and family, are especially influential in young adults' alcohol consumption (Hall et al., 2004;Ahlstr€ om and Osterberg, 2005;Thach and Olsen, 2006;Olsen et al., 2007;Silva et al., 2014).
Rituals are symbolic attitudes involving wine or beer consumption. Those are used to distinguish certain moments or specific contexts, such as the end of the working day or the end of the week, before and after football games, to toast and celebrate special moments (SIRC, 1998;Pettigrew and Charters, 2006;Charters and Pettigrew, 2008;SIRC, 2008;McCluskey and Shreay, 2011).
Motivations for consumption are certainly associated with the development of preferences through experience, which seem to be different between wine and beer. This might be influenced by socialisation but might also be related to the effect of learned liking, as a result of repeated consumptionexposure effectas happens with other foods (Yeomans, 2006). There is some evidence that, for wine drinkers, the evolution of preference is primarily inversely related to sweetness and later directly related with perceived "body" in a wine. In this way consumers' preferences move first over the dimension sweet-not-sweet and later on, along the dimension light-heavy/full (Blackman et al., 2010;Melo et al., 2011). Concerning beer consumption it is commonly argued that beer has an acquired taste and most consumers do not like their first taste of it (McCluskey and Shreay, 2011). Still, to our knowledge, there is no scientific evidence published to demonstrate this effect. The differences in regional beer production might have an influence in beer consumption preferences, especially in countries like America and Germany. The locally developed beer styles are more available in their region of origin (McCluskey and Shreay, 2011). Also in Northern Europe many pubs offer only two or three local beers which can limit the experience of consumption and consequently limit the exposure effect (McCluskey and Shreay, 2011). This might be applicable, but to a lesser extent to the Netherlands, where beer production is dominated by eight breweries, which together account for over 95% of beer production. The remaining five percent is manufactured by over 120 microbreweries (NB). In Portugal there is a different situation. As 90% of the national market is dominated by two breweries, located in the north and in the centre (APCV), most cafes and bars tend to sell only one brand, so choice is very limited. In general, alcoholic drink preferences tend to vary with age as the taste of the consumers develops. Whereas younger consumers tend to look for a sweeter product the older consumers prefer a more complex taste and unique flavours (Datamonitor, 2012).
From a marketing point of view a distinction is often made between motivations for consumption and motivations for purchase. Referring to purchase behaviour, price and habit or preferred brand are reported to be factors that exert the greatest influence on choice of wine and beer in 19 countries worldwide, including Portugal and the Netherlands (Duarte et al., 2010;DM, 2011c). Overall, consumers perceive price as a quality indicator. Its relative importance, however, is influenced by the context of consumption and level of product involvement (Quester and Smart, 1998;Ritchie, 2007;Charters and Pettigrew, 2008).
The motivations for NAB consumption are less documented, but those reported are very different from those known for alcoholic beverages. Drinking NAB can be related to occasions where consumers cannot or do not want to drink alcohol (Thompson and Thompson, 1996;Vlek and Peters, 2012). Flavour and healthy benefits were found to be the most important motives behind purchase (Thompson and Thompson, 1996;Chrysochou, 2014). In addition, as for wine and beer, the choice of NAB is influenced by drinking companions, family and social groups (Thompson and Thompson, 1996). Amongst Dutch NAB nonconsumers this beverage has a drowsy image and flavour is the key factor not to choose this beverage (Vlek and Peters, 2012).
In summary, it appears that wine, beer and NAB serve different purposes and that motivation for consuming either beverage depends on the context of consumption, although sensory and social aspects also carry some weight in the decision.

Perception of health benefits associated with moderate alcohol consumption
Consumers in general are more conscious of the health risks of alcohol consumption than the suggested health benefits (DM, 2011a). Nonetheless, consumers appear to be well aware that drinking wine or beer may lead to positive health outcomes and that the key factor is moderation in consumption (AES-BUC, 2003;van Zanten, 2005;Ritchie, 2007;Barreiro-Hurle et al., 2008;Wright et al., 2008a;Saliba and Moran, 2010;Yoo et al., 2013). In addition, the presence of alcohol does not seem to be a decisive factor defining the perception of healthiness of beverages, since wine and beer have been considered healthier than soft drinks (Wright et al., 2008a). In general, wine (red or white) is perceived as healthier than beer (Wright et al., 2008b). This might be due to a relative lack of information on the possible health benefits of moderate beer consumption, since more scientific work has been published on wine than on beer, especially oriented towards the consumer. A popular perception is that beer consumption is responsible for causing a paunch or "beer belly", although scientific evidence does not support this perception when beer is consumed in moderation (Bendsen et al., 2011). Another reason might be that wine consumption is strongly related to meals, which results in a stronger association with healthiness (DM, 2011a). From the published literature different attitudes emerge on how consumers perceive the healthiness of wine which may be culturally related: American consumers do not significantly change their behaviour on the basis of health perception and the main attribute that guides wine choice is flavour (van Zanten, 2005;Wright et al., 2008b); Spanish would be willing to pay more for resveratrol enriched wine (Barreiro-Hurle et al., 2008) and Australian consumers, who perceived wine as healthy, have a higher frequency but not volume of consumption (Saliba and Moran, 2010).
In conclusion, consumers are aware of possible health benefits of alcoholic beverages consumption, but that information in itself does not seem to be the main driver of consumption (van Zanten, 2005;Wright et al., 2008b;DM, 2011a, Chrysochou, 2014. Ranking the three beverages, wine is perceived as the healthiest beverage, followed by non-alcoholic beer, and regular beer is the least healthy (Wright et al., 2008b;Chrysochou, 2014).

Wine and beer related emotional attributes
Emotions and mood play an important role in food choice (Lyman and McCloskey, 2001;Canetti et al., 2002;Gibson, 2006). The interaction between mood and food is based on a bidirectional relation, which is influenced by a variety of factors such as appetite, sensory effects and social context (Gibson, 2006). For instance a good mood can induce a consumer to choose and drink a high quality wine (Ritchie, 2007). Likewise different foods evoke different emotions both in nature and in intensity (Cardello et al., 2012). Relatively few references exist in the literature concerning emotions and wine or beer consumption (Desmet and Schifferstein, 2008;Ferranini et al., 2010;Kaneda et al., 2011). The few existing ones mainly report on positive emotions in relation to wine and beer ( Table 7).
Emotions that affect consumers' choices can be linked to intrinsic product attributes such as sensory characteristics (Kuenzel et al., 2010;Thomson et al., 2010). It has been shown that by modulating beer flavour properties, it is possible to accentuate positive emotions elicited while drinking beer (Kaneda et al., 2011). Extrinsic attributes, however, like product name (Cardello et al., 2012) or label (King et al., 2010), can also be linked to emotions. In addition, it has been shown that food products have emotional connotations that when linked to the product identity will assign a meaning to what we experience while eating or drinking (Thomson et al., 2010). This is known as product conceptualisations, which have not yet been explored with wine, beer, or NAB.
It seems that the emotional attributes that are related to beer consumption are more of the functional type compared to those related to wine. For instance, attributes elicited by beer consumption such as refreshed or relief were not cited in connection with wine. Moreover, elegant and interesting are perhaps the two words cited that are most exclusively linked to wine in published studies (Table 7). Emotions reported for both wine and beer were peace/peaceful and amusing/amusement.
In summary, the study of emotions elicited by wine and beer consumption is not well explored and seems a promising area of research to gain more insight into beverage choice.

Conclusions
This review summarizes the main determinants of wine, beer and NAB consumption, covering cultural, contextual and psychological aspects, in Portugal and in the Netherlands, as countries representing Southern and Northern European culture.
Wine is a symbol of Portugal, but since the 1960s wine consumption decreased significantly and beer consumption increased. Perhaps unexpectedly, nowadays beer production and consumption per capita is greater, in volume terms, than for wine. Still, and in the terms of health benefits, it is important to highlight that wine is the greater contributor to total alcohol consumption. For Portuguese, wine and beer consumption is part of their daily habits and is mainly associated with lunch and dinnertime. Beer is mainly consumed outside home whereas wine is mainly consumed at home and in restaurants.
Beer is a symbol of the Netherlands and patterns of consumption remained almost untouched over the years. Wine consumption however increased. Both beverages significantly contribute to the total alcohol consumption, although beer is the greatest contributor. Dutch are more likely to drink wine or beer at weekends and mainly at home and in the evening.
NAB it is a relatively new beverage, with no cultural roots and consumption is minimal in both countries compared to wine or beer.
Wine is for the older generation and for both genders. Indeed, it is perceived as the healthier beverage, but this is far from being the prime reason for consumption. Flavour, food pairing and sociability seem to be more fundamental drivers for wine consumption. Wine is consumed at formal and special occasions. Interesting, passionate and elegant are some of the positive emotions evoked by wine consumers. Beer is a men's beverage, and is more likely to be chosen by the younger generations. It has a high connection with sociability, amusement and enjoyment, and a close relation with informal and energetic occasions. NAB represents an alternative for occasions where alcohol drinking is not appropriate and/or convenient. It is perceived as healthier than beer and flavour plays an important positive role for consumers and negative role for non-consumers.
Beer (with different hop and ester aromas) Measurement of brainwaves related with psychologically evaluated values (positive/negative, arousal and degree of comfort) and sensory evaluation Relaxed, Refreshed, Peace and Relief None Kaneda et al. (2011) This review illustrates the disparity between the amount of research that exists in the public domain concerning wine and beer consumption, where studies on wine outweigh those on beer. Work on NAB is almost non-existent.
The determinants of consumption described in this review contribute to define the actual identity of moderate wine, beer and NAB consumption giving insights to understand patterns of consumption over time and changes in drinking behaviour, in two different cultures. Emotional attributes associated with consumption of these beverages are not yet well explored and, in combination with context of consumption and cultural differences, might be an interesting topic to gain more knowledge on beverage choice.