Dietary fats have long been investigated in relation to various health and disease outcomes, in particular CVD. Epidemiological evidence from recent systematic reviews and meta-analyses suggests that excessive intakes of saturated and trans-fats are associated with higher all-cause mortality and CVD risk(Reference Kim, Je and Giovannucci1,Reference Hooper, Martin and Jimoh2) . Much of this early evidence and knowledge on the effects of dietary lipids on outcomes related to CVD risk was derived from the Seven Countries Study(Reference Keys3) which showed that per capita intake of saturated fat, but not total fat, was strongly associated with incidence of CVD and CVD-related mortality. Despite decades of ongoing research and inconsistent findings between saturated fat intake and coronary outcomes(Reference de Souza, Mente and Maroleanu4–Reference Siri-Tarino, Sun and Hu6), limiting saturated fat intake remains almost universally agreed upon across most global food-based dietary guidelines and recommendations for CVD risk reduction(Reference Herforth, Arimond and Álvarez-Sánchez7,Reference Bowen, Sullivan and Kris-Etherton8) . In contrast, reductions in cardiovascular events have been strongly observed in studies where saturated fats are replaced with unsaturated fatty acids, which has been demonstrated in several systematic reviews that have included evidence from both observational studies and randomised controlled trials(Reference Hunter, Zhang and Kris-Etherton9–Reference Clifton and Keogh12). Given that certain cardiovascular risk factors are modifiable by improving diet quality, the use of dietary oils and fats is of clinical and public health importance.
Since dietary oils and fats possess different fatty acid profiles, there is renewed interest in their cardiometabolic health effects. For example, in persons at high CVD risk, investigators from the PREDIMED (Prevención con Dieta Mediterránea) study(Reference Estruch, Ros and Salas-Salvadó13) reported a 30 % reduction in the risk of a primary cardiovascular event in participants randomised to receive either a Mediterranean diet supplemented with either extra virgin olive oil (EVOO) or nuts over a 5-year period compared to those assigned to a control (low-fat) diet. There is also great interest and controversy in the consumption of dairy fats (a source of saturated fat) and CVD risk(Reference Bhupathi, Mazariegos and Cruz Rodriguez14–Reference Astrup, Geiker and Magkos17). An important aspect of such controversies is that individual variations in fatty acids may have different metabolic pathways and cardiometabolic health outcomes. For example, compared with a variety of plant-based oils, there is meta-analytic evidence to support that olive oil modulates blood lipid profiles, in particular, by decreasing total cholesterol (TC), LDL-cholesterol and triglycerides and increasing HDL-cholesterol(Reference Ghobadi, Hassanzadeh-Rostami and Mohammadian18). In other meta-analyses, olive oil consumption has been inversely associated with diabetes risk, improvements in endothelial function and attenuation of inflammatory biomarkers(Reference Schwingshackl, Christoph and Hoffmann19,Reference Schwingshackl, Lampousi and Portillo20) . As such, the potential cardiometabolic benefits of olive oil may in part be due to its predominant fatty acid composition of oleic acid and extends beyond the modulation of plasma lipids and lipoproteins alone(Reference Kris-Etherton21). Furthermore, the plant-based omega-3 (n-3) fatty acid α-linolenic acid, a precursor for the long-chain n-3 PUFA, has also been associated with improved cardiometabolic health outcomes in both observational studies and randomised controlled trials(Reference Bork, Lundbye-Christensen and Venø22–Reference Tindall, Petersen and Skulas-Ray26) as well as the attenuation of inflammatory mediators(Reference Caughey, Mantzioris and Gibson27). In a meta-analysis of randomised controlled trials, Amiri et al.(Reference Amiri, Raeisi-Dehkordi and Sarrafzadegan28) showed that compared with other edible plant-based oils, canola oil (characterised by its high MUFA and α-linolenic acid composition) significantly improved cardiometabolic risk markers including TC, LDL-cholesterol, TC/HDL, LDL/HDL and apolipoprotein B. Moreover, the investigators further reported that the lipid-modulating effect of canola oil was greater when ∼15 % of total energetic intake from other oils was replaced with canola oil. Despite evidence suggesting that coconut oil (a rich source of saturated fat) raises TC, LDL-cholesterol and HDL-cholesterol compared with unsaturated plant oils(Reference Eyres, Eyres and Chisholm29), when compared against butter, coconut oil has been shown to be beneficial for cardiovascular health by raising HDL-cholesterol and lowering HDL-cholesterol(Reference Teng, Zhao and Khoo30).
Despite their potential impact on cardiometabolic risk indices, use of dietary oils and fats in Western countries remains largely unknown and has likely changed in recent decades due to multicultural societies and population-based health promotion messaging specifically targeting reductions in saturated fat intake. It is also unclear whether the use of dietary oils and fats for different cooking methods has altered in response to supplier messaging; in recent years, some vegetable oils have been promoted as having a high smoke point for the purpose of certain cooking methods including frying(Reference Eyres31). Nevertheless, there is recent evidence from the United States of America (USA) showing that olive oil is now the most commonly used dietary fat source for cooking and food preparation, followed by vegetable oil and canola oil(Reference Spearpoint and Hopkin32). However, preferences for dietary fats, oil use and the cooking methods associated with their use in Australia remain scant. From a population health perspective, knowledge of dietary oil and fat use is an important step towards facilitating more targeted and effective implementation strategies to encourage consumers to adopt healthier choices that are consistent with dietary guidelines and recommendations.
Therefore, the aim of the present study is to explore the use of dietary oils and fats in cooking and food preparation amongst Australian adults.
Methods
Participants
A cross-sectional study using a mixed methodological approach was undertaken amongst Australian males and females aged ≥ 18 years. Australian adults who were permanent residents of Australia and could complete an anonymous online survey in English were invited to participate. Participants were recruited via social media platforms including Facebook, Twitter, Instagram and LinkedIn from July 2021 to December 2021 requesting voluntary participation. QualtricsXM survey software was used to construct and distribute the survey. A link to the survey was disseminated via social media platforms where the study protocol and potential risks were clearly outlined to all interested participants. The investigators disseminated the survey link on social media platforms (Facebook, Twitter, Instagram and LinkedIn) weekly from the dates previously outlined. This study was conducted in accordance with the guidelines laid down in the Declaration of Helsinki, and all procedures involving study participants were approved by the Human Research Ethics Committees at the University of South Australia (203972) on 16 June 2021 and the University of the Sunshine Coast (A211607) on 6 July 2021. As the research involved the completion of a self-administered online survey, participants were informed that all information collected would remain confidential. Participants acknowledged an informed consent statement to participate in the study.
Data collection
Dietary oil and fat use amongst Australian adults was assessed using a nineteen-item self-administered online questionnaire, designed to be completed in approximately 10 minutes. Given the lack of a previously validated and reliable survey instrument, the authors developed a prototype questionnaire that was initially piloted against a separate representative sample for face validity. For the assessment of face validity, a convenience sample of volunteers across a range of ages (18–75 years) was invited to complete the prototype questionnaire that was administered online. A total of n 12 participants completed the prototype questionnaire (75 % female; 25 % male). Nil changes to the readability of the questions were required following administration of the prototype questionnaire. The online questionnaire was divided into two parts (see online Supplementary Material 1). Part A consisted of open- and closed-ended questions related to participant demographic characteristics. Part B of the online questionnaire consisted of open- and closed-ended questions related to type and household quantity of daily oil used, main fat and oil use for different cookery methods (e.g. shallow frying, deep frying, stir frying, stewing, etc.) as well as perceived attitudes and beliefs related to oil use within the household (Table 1). There were no time restrictions applied to complete the questionnaire and participants were not required to answer all questions before proceeding to subsequent questions. The link to the survey was recognisable once only to the server it was sent thus preventing duplication when responding to the survey. The investigators also screened all of the participant responses (IP address and postcode viewed to ensure participants were from Australia) to ensure all responses were consistent with the eligibility criteria.
Summary of questions from part B of the online survey evaluating the types of oils used in cooking and food preparation in Australia
Data analysis
Qualitative data were analysed using conventional content analysis(Reference Hsieh and Shannon33). Data were read for familiarisation and to determine initial codes. A recursive process was undertaken independently by two authors (NW and EM) during content analysis to maintain the rich detail of the data(Reference Anzul, Downing and Ely34,Reference Braun and Clarke35) and descriptions and rationale for codes were documented to confirm the reliability of the data(Reference Fade and Swift36). This iterative process was continued by the same two authors (NW and EM) until the research team was in agreement, with the addition of a third researcher (AV) to resolve any discrepancies. Related codes were grouped into representative themes. Representative quotations that illustrate the themes are presented alongside each theme and referenced with the participant number (in brackets). Furthermore, quantitative content analysis of response data was undertaken(Reference Bryman37). Identified themes were expressed as frequencies and percentages of responses using Microsoft Excel software. Statistical Package for the Social Sciences (SPSS) for Windows 27.0 software(38) was also used to perform analysis of descriptive data and expressed as means ± standard deviations for continuous data and frequencies and percentages for categorical data. A one-way ANOVA was used to explore differences in oil use according to age categories. Additionally, χ 2 analysis using Fisher’s exact test was conducted to explore potential differences in the frequency of reported oil use type and the presence of self-reported medical conditions (e.g. medical condition v. no medical condition).
Results
Participant characteristics
A total of n 1248 participants completed the survey and were included in the final analyses. More than 90 % of the survey questions were completed by all participants. Participant characteristics are outlined in Table 2. Participants were predominantly female (females n 1143, males n 99) with more than half of participants (56·7 %) aged between 25 and 44 years. The majority of participants were born in Australia (n 1046; 83·8 %); however, participants reported a diverse range of cultural backgrounds including 72·8 % who identified as Australian, 8·0 % identified as English, 3·1 % identified as Greek, 2·4 % identified as Italian and 1·2 % identified as Chinese. More than half of participants (59·0 %) reported a total annual household income of between $75 000 and $199 999 AUD. Less than half of the participants (43·3 %) reported no medical conditions, 25·4 % reported one medical condition and 14·6 % reported two medical conditions. Anxiety, depression, hypertension, arthritis and high cholesterol were the most reported medical conditions amongst participants.
Demographic characteristics of participants (n 1248)
Quantity and type of oil use in cooking and food preparation
Quantity and reported use of all oil types is presented in Table 3. Approximately half of all participants (n 629; 50·4 %) reported using less than one tablespoon per day in cooking and food preparation. In contrast, more than one-third of participants (n 459; 36·8 %) reported using 1–2 tablespoons of oil per day. More than three-quarters of participants (n 1055; 84·5 %) reported using some form of olive oil as their main source of oil for cooking and food preparation, with almost two-thirds of the sample (n 816; 65·4 %) using EVOO (Table 3). Of those who reported using some form of olive oil, almost half (n 514; 48·7 %) reported using less than one tablespoon per day. The next commonly used oils were rice bran (n 57; 4·6 %) and canola (n 54; 4·3 %) (Table 3). Rice bran oil (n 39; 68·4 %) and canola oil (n 32; 59·3 %) were also reported to be consumed in amounts of less than one tablespoon per day. No significant differences in oil use were observed according to age category (F(16, 1231) = 1·25, P = 0·22). No significant differences were observed according to the presence of a medical condition in those that used EVOO, olive oil, light olive oil, rice bran or canola oil as the main household oil (medical condition: n 626; no medical condition: n 540; not reported n 82; all P values = >0·05). Walnut oil, safflower oil and animal fat/lard were not reported as the main household oils used by participants; however, they were used for specific food preparation and cooking methods.
Frequency and percentage of main household oil use differentiated by daily serve size (n 1248)
Abbreviations: EVOO, extra virgin olive oil.
*Quantity of EVOO used was not reported by n 6 participants.
Participants reported using a range of different oils for various cooking and food preparation methods (Table 4). Oil use was reported across all food preparation and cooking techniques. However, over three-quarters of participants (n 1036; 83·0 %) reported nil use of oil for boiling. EVOO was the most frequently reported oil used for raw food preparation methods (e.g. salad/vegetable dressings) (n 882; 71·5 %), followed by savoury baking or roasting (n 724; 58·0 %) and sauteing (n 679; 54·4 %). In contrast, butter was most frequently used by participants for sweet baking (n 471; 37·7 %). Canola oil and rice bran oil were both most frequently used for shallow frying (canola oil: n 140, 11·2 %; rice bran oil: n 117, 9·4 %), while sesame oil was most frequently used for stir frying (n 123; 9·9 %). However, EVOO was still the main oil used by most participants for shallow and/or stir frying (Table 4).
Frequency n (%) of oil use for different cooking methods (n 1248)
EVOO, extra virgin olive oil.
Perceived motivators for main household oil consumption
Survey responses on the perceived motivators for main oil use were analysed for themes and presented in combination with representative quotes, frequency and percentage of responses (Table 5). A total of seventeen themes were identified as perceived motivators. Almost half of all participants (n 619; 49·6 %) identified perceived health benefits as the primary motivating factor for the main choice of oil in cooking and food preparation methods. Sensory preference including flavour, taste and smell was also frequently reported as an influential factor determining choice of oil (n 583; 46·7 %). Versatility (n 127; 10·2 %) and convenience (n 110; 8·8 %) were also frequently reported.
Perceived motivators of the main oil used by participants in their household (n 1248)
Given that some form of olive oil (e.g. EVOO, olive and light) was identified as the predominant oil used in Australian households (n 1055; 84·5 %), the perceived motivators for its use were akin to those outlined in Table 5. As such, the most frequently reported motivator for olive oil use included the perceived health benefits of olive oil (n 561; 53·2 %), followed by flavour and taste, including the flavour of infused olive oils (n 503; 47·7 %), and its versatility for use across different cooking methods (n 112; 10·6 %). Convenience was also frequently reported as a perceived motivator for using olive oil (n 98; 9·3 %).
Of the n 193 participants who reported not using any form of olive oil as their main dietary oil source, sensory preference, in particular taste and flavour, was the most frequently reported reason for not using this type of oil (n 58; 30·1 %). Limiting cooking problems (e.g. EVOO having a lower smoke point and potential damage to cooking utensils) was also frequently reported (n 49; 25·4 %). Moreover, expense (n 32; 16·6 %) was also identified as an important barrier for not using a form of olive oil.
Discussion
Results from this large cross-sectional analysis demonstrated that olive oil, in particular EVOO, was the main choice of oil for use in cooking and food preparation in this sample of Australian households. The perceived motivators for the use of the main cooking oil in the household included its potential for health benefits, sensory preference and its versatility.
Our findings are indeed novel as this, to the best of our knowledge, is the first Australian study to investigate oils used in cooking and food preparation in Australian households. Nevertheless, our data are representative of previous apparent consumption from international data on the use of olive oil for cooking and food preparation in Western countries(Reference Uylaşer and Yildiz39,40) . Specifically, current available data on trends related to olive oil consumption indicate that over the past 30 years, there has been a 75 % increase in the consumption of olive oil in Australia(40). In comparison, olive oil consumption in the USA has also seen a 78 % increase from 1991 to 2021. In contrast, olive oil consumption in some Mediterranean countries, including Italy and Greece, has seen between a 13 and 45 % decline in olive oil consumption over this timeframe(40).
The rise in EVOO use in Australian households has the potential to contribute to healthier dietary patterns and positively impact population health due to the myriad of health benefits associated with its high nutritional quality and its unique composition. In particular, oleic acid in addition with bioactive polyphenolic molecules (such as oleuropein, hydroxytyrosol and tyrosol) are important constituents of EVOO which may help to explain its cardioprotective role including reductions in TC and LDL-cholesterol, blood pressure, pro-inflammatory markers, improved insulin sensitivity and endothelial function and reductions in oxidative stress(Reference George, Marshall and Mayr41–Reference De Santis, Cariello and Piccinin45). Regular consumption of EVOO has also been associated with reductions in diabetes mellitus, inflammatory bowel disease, obesity and some cancers(Reference Jiménez-Sánchez, Martínez-Ortega and Remón-Ruiz46). As such, it is not surprising that perceived health benefits were identified as the primary motivator for olive oil use in Australian households. Our findings are in agreement with a smaller study involving n 35 men and women in the USA who also identified the perceived health benefits of olive oil as an important factor influencing consumer choice(Reference Santosa, Clow and Sturzenberger47). Importantly, however, results from the present study as well as the aforementioned study conducted by Santosa et al.(Reference Santosa, Clow and Sturzenberger47) did not further articulate the specific details related to perceived health benefits such as cardiovascular benefits and/or reductions in disease risk. Nevertheless, despite the perceived health benefits reported by participants in the present study, the majority reported using less than one tablespoon of olive oil per day, which is markedly less than reported in previously published clinical trials investigating cardiovascular benefits from the use of EVOO(Reference Estruch, Ros and Salas-Salvadó13,Reference George, Marshall and Mayr41,Reference Davis, Hodgson and Woodman48) . Whether lower quantities of EVOO incorporated into an otherwise healthy dietary pattern will exert the same cardiovascular benefits remains largely unknown and is subject to ongoing research. Nevertheless, given the potential cardioprotective health benefits associated with EVOO consumption, it would be prudent for health care clinicians to advocate the use of EVOO, particularly in high-risk patients, at quantities that are acceptable to individual, social and health-related needs.
Sensory preference was also a determining factor in the main choice of oil used. In particular, participants favoured a ‘neutral’ oil including EVOO or olive oil. Nevertheless, some participants reported olive oil as having ‘too strong’ a flavour and favoured alternative oils depending on cuisine type. Previous evidence suggests that the fatty acid composition of oils may be responsible for taste perception and can play a role in determining taste intensity and perceived retention of taste(Reference Koriyama, Wongso and Watanabe49). Furthermore, some studies have indicated that the phenolic compounds present in EVOO, namely oleuropein, may influence the bitterness and thus perceived taste of the oil(Reference Campestre, Angelini and Gasbarri50). In a sample of n 60 Italian adults, Barbieri et al.(Reference Barbieri, Bendini and Valli51) reported that participants did indeed appreciate the fruity flavour when sampling EVOO; however, this was superseded by the perceived bitterness of the oil. Furthermore, sensory perception has been reported to be influenced by more than taste and smell alone, with brand, price, country of origin, and label information and presentation playing a key role in sensory perception(Reference Fernandes, Ellis and Gámbaro52). In the present study, data on the sensory perception of olive oil beyond comments related to the perceived neutrality of the oil were not collected, and as such, we have limited information regarding respondents’ motivations on choosing olive oils based on perceptions of specific flavours, pungency or other sensory characteristics.
Limiting cooking problems (e.g. burning food, damage to cookware) was frequently reported as a perceived motivator for choosing the main household cooking oil. Specifically, respondents reported selecting an oil with a high smoke point for the style of cooking (e.g. frying) that would not damage cookware or utensils. As such, this was more frequently reported amongst participants who did not use a form of olive oil as the main oil in the household. Indeed there is a perceived perception that olive oils do not have a high smoke point and are not suitable for certain cooking methods, such as frying; however, evidence continues to mount suggesting that EVOO in particular can be heated to temperatures of 180 degrees Celsius for deep frying and up to 240 degrees Celsius for shallow frying while maintaining stability and producing less polar by-products than its polyunsaturated counterparts and in turn protecting cookware(Reference De Alzaa, Guillaume and Ravetti53–Reference Ramírez-Anaya, Castañeda-Saucedo and Olalla-Herrera56). Furthermore, there is also a common perception that heating olive oil to high temperatures may damage the properties of the oil. However, previous evidence suggests that the phenolic compounds within EVOO play a protective role in minimising oxidation during frying(Reference Servili, Sordini and Esposto57,Reference Santos, Cruz and Cunha58) , and EVOO with a high oleic acid composition demonstrates a lower formation of toxic compounds during frying(Reference Procida, Cichelli and Compagnone59). Nevertheless, in the present study, EVOO continued to be the most frequently used oil for sautéing, shallow frying and stir frying which is a finding consistent with a previous cross-sectional survey of n 2234 participants in the USA which reported that 86 % of participants sautéed with olive oil despite suggestions of the proposed low smoke point associated with the oil(Reference Wang, Moscatello and Flynn60). Despite the high use of EVOO across these cooking techniques, we did observe differences between sauteing, stir frying and shallow frying. For example, we observed that a greater proportion of participants (74·3 %) used a form of olive oil for sauteing rather than for stir frying (46·3 %), which is considered a similar cooking technique albeit at a lower temperature. However, this may also reflect the interpretive differences in oil use for different cooking methods across different ethnicities. Additionally, the theme of minimising damage to cookware was more frequently reported as a perceived motivator for oil choice than cost. This is indeed a novel finding given that some form of olive oil was the most commonly reported oil used in Australian households, despite its high costs relative to other common cooking oils(Reference D’Adamo, Falcone and Gastaldi61,Reference Martínez, Aragonés and Poole62) . Our findings are in contrast to those previously reported by Erraach et al.(Reference Erraach, Sayadi and Gomez63) and Santosa et al.(Reference Santosa, Clow and Sturzenberger47) who identified that cost was one of the main determining factors in participants’ decisions to purchase olive oil in n 439 adult consumers of olive oil (52·5 % female; 47·5 % male) in Spain and n 35 adults (82·9 % female; 17·1 % male) in the USA, respectively. However, more recent findings suggest that when choosing a household oil, factors, including sustainability, are more considered than the cost of olive oil (Reference Erraach, Jaafer and Radić64). In contrast to these findings, supporting local production of oils and environmental sustainability were not as frequently reported in the present study.
Cultural background and family preferences were also identified by some participants as motivators for choice of oil. The link between cultural background and the main household oil was perhaps somewhat expected given the well-established associations between culture and food choice(Reference Pollard, Kirk and Cade65). This theme was also similar to previous research conducted in Europe which reported that culture and family tradition were important determining factors for consumption of olive oil among Austrian, German, British, Russian, Italian and Croatian citizens(Reference llak Peršurić and Težak Damijanić66). Nevertheless, in the present study, cultural background and family preferences were not reported as frequently in comparison with the aforementioned study.
This study is not without limitations. Firstly, we recruited a convenience sample of Australian adults who were generally younger, well-educated and from a high-income bracket, which was not generalisable to the wider population of Australian adults. Given that we used social media platforms for recruitment, this approach resulted in selection bias and limited our ability to capture oil use and preferences amongst more disadvantaged populations. Therefore, future research into a more socio-economically diverse sample of Australian adults would further identify oil use in cooking and food preparation in Australia and provide a more definitive understanding of the oils used and the perceived motivators for their use. This would also allow further analysis of the links between oil use and health-related outcomes. In addition, females responded to the online questionnaire at much higher rates compared with males. This may not be unexpected given that females may be more likely to participate in cooking-related activities such as watching cooking programmes and participating in online cooking surveys(Reference Villani, Egan and Keogh67) as well as cooking and preparing meals in the household(Reference Storz, Beckschulte and Brommer68,Reference Lupton69) . However, these findings are unlikely to be generalisable to the wider population of Australian adults. Lastly, reporter bias was also likely given that the quantity of oil use, and the method of food preparation in the household was self-reported.
Conclusions
Findings from this cross-sectional analysis suggest that the majority of households in this sample of Australian adults use some form of olive oil as the main oil in cooking and food preparation, primarily due to its perceived health benefits, sensory preference and versatility. However, although olive oil use was frequently reported in this population, the amount of olive oil used was markedly lower than those amounts needed to elicit cardiovascular benefits. As this is the first study to investigate oil use and consumption in Australian households, further investigation into the quantities consumed and current consumer understanding of the potential health benefits, sensory properties and versatility of olive oil use for different cooking methods will assist in developing more targeted strategies for health care professionals to disseminate advice related to the use of olive and other cooking oils for optimal health and disease prevention.
Acknowledgements
This research received no specific grant from any funding agency, commercial or not-for-profit sectors.
All authors were involved in the conception and design of the study, data analysis and development of the manuscript. All authors read and approved the final version of the manuscript.
There are no conflicts of interest.
Supplementary material
For supplementary material/s referred to in this article, please visit https://doi.org/10.1017/S0007114522003798
