Regular adequate fruit and vegetable intake contributes to good nutrition and general health and protects against common chronic diseases including CHD(1–14), hypertension(Reference Bazzano2, Reference Miura, Greenland, Stamler, Liu, Daviglus and Nakagawa9, Reference Svetky, Simons-Morton and Vogt15), stroke(Reference Gillman, Cupples, Gagnon, Posner, Ellison, Castelli and Wolf6), diabetes(Reference Colditz, Manson, Stampfler and Rosner16–Reference Williams, Wareham and Cox21) and some cancers(22), as well as overweight and obesity and a number of other diseases(14). The WHO and FAO have called for nations to increase fruit and vegetable consumption through targeted campaigns(14). Multi-component approaches, addressing individual factors (e.g. knowledge, attitudes, skills, social influences and behaviours) as well as environmental factors (e.g. access, cost, quality and supply), appear to be more effective(Reference Glanz and Hoelscher23–Reference van der Horst, Oenema, Ferreira, Wendel-Vos, Giskes, van Lenthe and Brug27). There is a lack of published information about the effectiveness of population-based interventions promoting fruits and vegetables(24).
Health communication ‘has the capacity to create awareness, improve knowledge and induce long-term changes in individual and social behaviours’(28). There is a need for effective educational messages to encourage the consumption of fruit and vegetables(Reference Ledikwe, Ello-Martin and Rolls29, Reference Lea, Crawford and Worsley30). Knowledge of the recommended fruit and vegetable intake may be motivational, leading to self-evaluation of intake, influencing social norms and increasing the expectation and approval for that level of consumption (Reference Reynolds, Bishop, Chou, Xie, Nebeling and Perry31). There might be agreement on the need to increase fruit and vegetable consumption; however, there are differences between countries in their classification, what constitutes a serving and the recommended servings(Reference Miller, Pollard and Coli32), see Table 1.
Standard and recommended fruit and vegetables servings per day for good health in Australia, Canada and USA
na, not applicable.
*4–7 years, †8–11 years, ‡12–18 years, §19–60 years, ||60+ years.
The Australian recommendation of at least 675 g daily (including potatoes) is consistent with the minimum 400–600 g daily (excluding potatoes) recommended by health authorities to protect against disease(14, 22, Reference Lock, Pomerleau, Causer, Altmann and McKee33). Australian fruit and vegetable recommendations have been separate since 1994, based on the differing nutrient profiles of fruits and vegetables; the practical aspects of eating them; current household consumption levels; and food supply(Reference Cashel and Jefferson34). Australia’s food selection guide, The Australian Guide to Healthy Eating(Reference Smith, Kellett, Schmerlaib and Smith35), is incorporated into Australian dietary guidelines(36–38) and is used as the basis for most nutrition education initiatives.
Developing suitable messages to assist consumers to understand and accept the importance of healthy eating is complex. The way in which recommended food intake amounts are presented, as number of servings or specified size, is often misinterpreted by consumers(Reference Britten, Haven and Davis39). Focus group research to assist the development of educational messages for the American MyPyramid Food Guidance System found people considered fruits and vegetables should be part of a healthy diet; however, they were confused about serving sizes and the recommended daily intake(Reference Britten, Haven and Davis39).
The ‘5-a-day’ message has been used as part of campaigns to encourage increased fruit and vegetable consumption(Reference Ashfield-Watt40–42). Consumer research suggests that the ‘5+ a-day’ message is understood and appropriate to encourage increased frequency of consumption; however, consumers have a limited understanding of servings size(Reference Ashfield-Watt40, Reference Ashfield-Watt, Welch, Day and Bingham41). The Western Australian Health Department (WAHD) developed the Fruit ‘n’ Veg with Every Meal social marketing campaign to encourage fruit and vegetable consumption in 1990. Evaluation found that although consumers were aware of the campaign and interpreted the message as needing to eat more fruit and vegetables, they were not prompted to action(Reference Miller, Pollard and Paterson43), suggesting that a message specifying an optimal intake was preferable to one saying ‘just eat more’.
The Go for 2&5® social marketing campaign launched in March 2002 with mass media advertising (television, radio, press and point-of-sale), public relations, publications, website (www.gofor2and5.com), and school and community activities(Reference Pollard, Miller, Daly, Crouchley, O’Donoghue, Lang and Binns44). The campaign increased awareness of the need to eat more fruit and vegetables by increasing knowledge of the recommended number of servings, with a corresponding increase in consumption(Reference Pollard, Miller, Daly, Crouchley, O’Donoghue, Lang and Binns44). The Go for 2&5® logo mnemonic device reminded consumers of the target (Fig. 1).
Go for 2&5® logo
The objective of the present study was to determine consumer understanding of what constitutes a serving and the relationship to current fruit and vegetable intake. The timing of the research, six months into a high-profile social marketing campaign, allows analysis of the association between serving size understanding and campaign awareness.
Methods
In March 2002, the WAHD commenced the Health & Wellbeing Surveillance System (HWSS) continuous data collection system. Computer-assisted telephone interview was used to interview over 550 Western Australian people aged 16 years and over each month. The survey asked about a range of issues including health conditions, lifestyle risk factors, protective factors and sociodemographics. Monthly samples were extracted using the Electronic White Pages telephone numbers as the sample frame and stratifying by rural, remote and metropolitan areas. Within each stratum, random samples were selected. From mid-September until the end of November 2002, all respondents over 16 years (n 1108) were asked four additional questions relating to fruit and vegetable serving size and campaign awareness.
Respondents were asked, ‘What do you think a serve of vegetables/[fruit] equals?’ no alternatives were given and there was no prompting. Responses were then coded into pre-designated categories (one piece, one type, e.g. apple or carrot, ½ cup, other amount [Specify], ‘what I put on my plate’ [vegetables], other [Specify] or don’t know).
Next they were asked about their usual fruit and vegetable intake: ‘How many serves of vegetables/[fruit] do you usually eat each day? A serve of vegetable is equal to ½ cup of cooked vegetables or 1 cup of salad’ and ‘A serve of fruit is equal to one medium piece, two small pieces of fruit or one cup of diced fruit’. They were then asked about awareness of the campaign; ‘The Department of Health has recently conducted a campaign about fruit and vegetables. Do you recall hearing or seeing anything about this?’
Prevalence and mean estimates, with 95 % CI, were calculated using the Statistical Package for the Social Sciences version 15.0.1 (SPSS Inc., Chicago, IL, USA). Generalised linear models analysis was conducted using the survey data analysis module(Reference StataCorp45) in STATA 10 (StataCorp, College Station, TX, USA). Differences were reported as statistically significant when the confidence intervals did not overlap, where they are reported. Inferential statistical analyses provide P values as the basis for statistical significance. The WAHD-approved research met standard social marketing research criteria.
Results
The final sample of adults aged 16 years and over represents 81 % of contacts made. The data were weighted to correct for over-sampling in rural and remote areas and then adjusted to the age and sex distribution of the Western Australian Estimated Resident Population for 2002.
Knowledge of fruit serving size
Table 2 displays the self-reported serving size for fruit and vegetables. Most respondents, 83 %, nominated a serving size for fruit. More men (22 %) than women (12 %) said they were unsure about the fruit serving size and there were no significant differences based on age. ‘One piece’ was the most commonly identified serving size for fruit (42 %), with women (48 %) more likely than men (34 %) to suggest it. People aged 16–24 years were least likely to select ‘one piece’, 32 % compared to 48 % of 25- to 39-year-olds.
Perceived fruit and vegetable serving size, by gender and age, of persons aged 18 years and older, Western Australia, September to November 2002
na, not applicable.
*The column headings refer to the categories for the questions asking what constitutes a serving of fruit or vegetables.
‘One type of fruit’ was identified as the serving size for fruit by 11 % of respondents, with no significant difference based on gender or age. Other serving size amounts were specified by 16 % of respondents.
Knowledge of vegetable serving size
Table 2 displays the self-reported serving size for vegetables. Most respondents, 78 %, thought they knew the vegetable serving size. Significantly more men than women said they were unsure about the vegetable serving size, 28 % and 19 %, respectively. There were no significant differences based on age. Fourteen per cent of respondents, 20 % of women and 7 % of men identified ‘½ cup’ (the standard vegetable serving size). People over 40 years old were more likely to identify ‘½ cup’ than those who were younger. Twenty-eight per cent of respondents nominated a different amount for the vegetable serving size. Respondents less than 40 years old were more likely than those aged over 40 to specify a different amount. Men (10 %) were more likely than women (4 %) to identify ‘the amount I put on my plate’ as a serving of vegetables.
Fruit and vegetable consumption
Table 3 displays knowledge of standard serving size by the mean fruit and vegetable intake. The mean fruit intake was 1·8 (95 % CI 1·7, 1·8) servings/d, with women consuming more than men, 1·9 (95 % CI 1·8, 2·0) and 1·6 (95 % CI 1·5, 1·7) servings/d, respectively. Those who knew the standard fruit serving size had higher intakes, 1·9 compared to 1·7 servings/d; however, the difference was not significant. Respondents who knew the standard serving size for fruit were more likely to eat the recommended two servings per day than who did not, 57·2 % (95 % CI 52·0, 62·2) compared to 46·3 % (95 % CI 42·0, 50·7).
Daily fruit and vegetable consumption, by serving size and gender, of persons aged 18 years and older, Western Australia, September to November 2002
*The consumption levels are equal to or above/below the recommended Australian guidelines of at least two servings of fruit and five servings of vegetables per day.
†One piece of fruit or ½ cup of vegetables.
‡Another amount or ‘other’.
The mean vegetable intake was 2·8 (95 % CI 2·7, 3·0) servings/d. Women had higher intake than men, 3·1 (95 % CI 2·9, 3·0) and 2·6 (95 % CI 2·6, 2·8) servings/d, respectively. Men who knew the standard serving size had 1·1 servings more than those who did not. Respondents who knew the standard serving size for vegetables were more likely to have the recommended five servings per day than those who did not, 19 % compared to 11 %; however, the difference was not significant. Women who knew the standard serving size for vegetables were more likely to eat five or more servings daily than men who knew the standard serving size, 22 % compared to 9 %, respectively. Respondents who were aware of the campaign reported higher mean vegetable intake than those who were not, 3·0 (95 % CI 2·8, 3·1) compared to 2·5 (95 % CI 2·3, 2·7) servings/d, respectively. There was no significant difference for fruit consumption. Table 4 displays factors influencing fruit and vegetable consumption. Vegetable intake was significantly associated with being female (P = 0·006), increasing age (P < 0·0001), awareness of the campaign (P = 0·031), and knowledge of standard serving size (P = 0·006). Fruit consumption had a significant association with gender (P = 0·007). There were no significant associations based on educational attainment or household income.
Parameter estimates by number of servings of fruit (vegetables) consumed per day of persons aged 18 years and older, Western Australia, September to November 2002
Analysis done using STATA 10 survey data analysis module, generalised linear models.
Discussion
Developing suitable messages to assist consumers to understand and accept the importance of healthy eating is complex. Knowledge gaps associated with the recommendation to eat more fruit and vegetables were evident in this study, particularly for vegetables. Knowledge of the recommended amount may reflect a key skill needed to perform the behaviour, it may serve a motivational function leading to a self-evaluation of intake, and it may provide a normative influence, increasing the expectation and approval for that level of consumption(Reference Reynolds, Bishop, Chou, Xie, Nebeling and Perry31). To understand dietary recommendations, consumers needed to know the type and amount of recommended foods, and to assess the adequacy of their current intake they needed to know what constituted a serving(Reference Britten, Haven and Davis39). Britten et al. (2006) found consumers could apply this information by mentally adding up the amounts they consumed at each meal and comparing it to daily recommendations(Reference Britten, Haven and Davis39). Overly optimistic assessment of current intakes results in complacency about the need to eat more fruit and vegetables(Reference Pollard, Miller, Daly, Crouchley, O’Donoghue, Lang and Binns44, Reference Dibsdall, Lambert, Bobbin and Frewer46, Reference Cox, Anderson, Lean and Mela47). Respondents were more confident in assigning a serving size to fruit than to vegetables. The most common fruit serving size of ‘one piece’ supports the finding that consumers prefer recommendations expressed in pieces of fruit(Reference Britten, Haven and Davis39). Respondents reported vegetable serving sizes in ‘amounts’, consistent with previous findings that common household units or measures, for example a cup, were preferred for vegetables(Reference Britten, Haven and Davis39).
Many theoretical models can be used to guide food choice research(Reference Anderson, Winett and Wojcik48, Reference Glanz49). The Go for 2&5® campaign used a model of adapted phases between knowledge and behaviour(Reference Fishbein and Ajzen50, Reference Egger51) to examine behavioural beliefs, attitudes, influencers and intentions(Reference Pollard, Miller, Daly, Crouchley, O’Donoghue, Lang and Binns44). How or whether consumers use serving size information to make their food choices is still not clear; however, the information may be useful to assist with the assessment of current intake. Policy recommendations and assessment of intervention effectiveness rely on accurate dietary assessment. The measurement method influences the proportion of the population categorised as meeting fruit and vegetable guidelines(Reference Michels, Welch, Luben, Bingham and Day52–Reference Pomerleau, Lock, McKee and Altmann54).
The main strength of the present study is that it provides a quantified population-level snapshot of consumer understanding of fruit and vegetable serving sizes six months after the commencement of the Go for 2&5® campaign. The results are representative of the Western Australian population as a whole but may not represent subgroups within the population, such as Aboriginal people. There were also study limitations, the analysis did not distinguish amounts other than ‘½ cup’; other amounts, particularly for vegetables, are required. Additional questions relating to knowledge of recommended intake, or asking current intake before and after defining serving sizes are suggested; however, minimal additional questions can be added to an existing surveillance system. Knowledge of serving size or dietary behaviour may merely be a marker for a cluster of ‘healthy lifestyle’ behaviours (e.g. smoking, exercise). Recent New Zealand research found little or no clustering of healthy behaviours, and refuted that fruit and vegetable consumption is merely a marker of healthy lifestyle(Reference Tobias, Jackson, Yeh and Huang55). Further analysis to determine the influence of other lifestyle factors is suggested. More community education is required on serving size to maximise the impact of health promotion campaigns such as the Go for 2&5®. More research is required to establish the best method for this education.
Conclusions
Consumers differentiate between serving sizes for fruit and vegetables. Knowledge of the standard fruit and vegetable serving sizes is related to the consumption of the recommended amounts. Respondents’ understanding of fruit and vegetable serving sizes suggests it is important to separate fruit and vegetable recommendations, using common household measures to convey serving sizes, and using prescriptive messages, for example Go for 2&5® to encourage fruit and vegetable consumption.
Acknowledgements
Funding source:The development, implementation and evaluation of the Go for 2&5® campaign was funded by the Western Australian Health Department (WAHD). Horticulture Australia Limited funded Curtin University of Technology to assist with the preparation of publications. The WAHD developed and owns the Intellectual Property for the Go for 2&5® campaign.
Conflict of interest:Nil.
Author contributions:A.M.D. and C.M.P. worked for the WAHD who funded independent market research companies to conduct the HWSS survey and produce original data sets. None of the authors were involved in any interviews. C.M.P. was also responsible for managing the development, implementation and evaluation of the Go for 2&5® fruit and vegetable campaign.
Acknowledgements:The authors acknowledge the staff at the former Nutrition, Marketing and Communications Branches of the Health Promotion Directorate and the Health Information Centre at the WAHD; and 303 Advertising Agency (who developed the creative and campaign strategy). The WAHD contracted independent market research companies to administer each survey. Horticulture Australia Limited (HAL) supported Curtin University of Technology to analyse the data and prepare papers for publication.
