Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

The Risks of Everyday Life: Fat content of chips, quality of frying fat and deep‐frying practices in New Zealand fast food outlets

The Risks of Everyday Life: Fat content of chips, quality of frying fat and deep‐frying practices... Judith Morley-John, Boyd A. Swinburn, Patricia A. Metcalf and Fezeela Raza Department of Community Health, University of Auckland, New Zealand Abstract Objectives: To collect baseline data on the fat content of hot chips, quality (degradation) of cooking fat, deep-frying practices and related attitudes in fast food outlets in New Zealand. To identify the key determinants of the fat content of chips and quality of cooking fat. Methods: A nationally representative sample of fast food outlets (n=150, response rate 80%) was surveyed between September 1998 and March 1999. Data collected included a questionnaire, observation of cooking practices and analysis of cooked chips and frying fat. Heather Wright Public Health Protection, Auckland Healthcare Services Ltd, New Zealand he prevalence of obesity is increasing in New Zealand,1,2 as it is in man y countries,3 and is now widely regarded as a pandemic.3,4 A reduction in the prevalence of obesity and related diseases will require a broader public health approach with an emphasis on making the environment less ‘obesogenic’ 5 by ensuring that healthy food choices and physical activity choices are made the easy choices.6,7 The fat content of deep-fried food is likely to contribute to obesity by increasing the intake of total fat and energy. The 1997 National Nutrition Survey reported that deepfried hot chips (French fries) are eaten by 47% of New Zealanders at least once a week.1 Many aspects of deep fr ying have been reported to influence the fat content of chips – chip size and shape,8 post-fr ying practices (such as shaking and hanging the basket),9,10 fat temperature and fat quality (i.e. the degree of fat degradation).11-14 Fat quality is influenced by frying temperature,15 vat cleaning, fat turnover, filtering and the use of f iltering agents.16 There is signif icant potential for public health inter ventions to promote best practice deep-frying techniques and this survey collected the baseline data needed to target and monitor those interventions. The specif ic aims of the hot chips study were to: • Collect baseline data on the fat content of chips, quality of cooking fat, deepfr ying practices and related attitudes and beliefs in fast food outlets throughout New Zealand. • Compare chain and independent outlets. • Identify the key determinants of fat content of chips and quality of cooking fat. • Identify opportunities for action to reduce the fat content of hot chips and improve the fat quality of the frying fat. Results: Only 8% of independent operators had formal training in deep frying practices compared with 93% of chain operators. There was a wide range of fat content of chips (5%-20%, mean 11.5%). The use of thinner chips, crinkle cut chips and lower fryer fat temperature were associated with higher chip fat content. Eighty-nine per cent of chain outlets used 6-10 mm chips compared with 83% of independent outlets that used chips ≥12 mm. A wide range of frying temperatures was recorded (136233°C) with 58% of outlets frying outside the reference range (175-190°C). As indices of fat degradation, fat acid and polar compound values above the recommended levels occurred in 54% and 5% of outlets respectively. Operators seemed willing to learn more about best practice techniques, with lack of knowledge being the main barrier to change. Conclusions and implications: Deep frying practices could be improved through operator training and certification options. Even a small decrease in the mean fat content of chips would reduce the obesogenic impact of this popular food. ( Aust N Z J Public Health 2002; 26: 101-7) Methods Sample Fast food outlets were randomly selected from a list of all takeaways and unlicensed restaurants listed in the Ne w Zealand Telecom Yellow Pages (n=3,301). They were stratif ied by geographical re gion and target numbers were defined to obtain a representative sample. Only the outlets that sold takeaway hot chips were eligible. On telephoning the selected outlets, 63% were found to be eligible and 80% agreed to take part. It was assumed that the vast majority of fast food outlets were listed in the Yellow Pages (although this was not independently verif ied). Survey process The data were collected during the period September 1998 to March 1999. An interviewer-administered questionnaire (29 closed questions) was designed to investigate chip-cooking practices and attitudes related to cooking practices. This instrument was developed on the basis of: • A literature review identifying key Correspondence to: Boyd Swinburn, School of Health Sciences, Deakin University, 221 Burwood Highway, Burwood, Victoria 3125. Fax: (03) 9244 6017; e-mail: swinburn@deakin.edu.au Submitted: April 2001 Revision requested: November 2001 Accepted: March 2002 2002 VOL. 26 NO . 2 AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH Morley-John et al. Article cooking procedures that can be measured and that are amenable to change. • Data from the pilot sur vey of 24 independent tak eaway bars within the Auckland region. • Infor mation obtained through consultation with fast food outlet operators and food technology experts in the area. The attitudes and perceptions of operators were measured using Liker t scales. Fat and chip turnovers were assessed by asking the operators about their monthly order volumes for each. Many of the operators were Chinese immig rants and translations of the participant infor mation sheet and interpreters were used as required. The study w as approved by the University of Auckland Human Subjects Ethics Committee. A 200 mL sample of fat from the main chip fryer was collected for fat analysis. The thermostat setting and actual f at temperature once the fryer had reached temperature (thermostat light off) were recorded. The temperature range of 180-185°C is considered optimal based on recommendations in other countries14,18 and we considered that 5°C either side of this (i.e. 175-190°C) would be an appropriate reference range for this study. The operator was asked to cook a serving of chips. Observations of the post-cooking procedure (banging, shaking and hanging of the chip basket) were recorded. Post-cooking drainage of fat from the chips was considered adequate if the basket was shaken and/or banged and hung for at least 20 seconds before being packaged. The chips were immediately transfer red from the packaging to a plastic bag. The questionnaire was then administered. The f at and chip samples were chilled and then frozen (-18°C) as soon as possible until analysis. Table 1: Characteristics of fast food outlets and operators. Characteristic Number, n(%) English first language, n(%) T raining, n(%) Informal, on-the-job Formal, on-the-job Formal institution Fat turnover, kg/wk Mean±SD Range Chip turnover, kg/wk Mean±SD Range 126±105 10-425 925±623 75-2400 60±43b 4-250 316±321c 5-2000 72±64 Chains 27 (18) 27 (100) 2 (7) 24 (89) 1 (4) Independents 123 (82) 66 (54) 113 (92)a 3 (2)a 7 (6) Total 150 94 (63) 115 (77) 27 (18) 8 (5) 426±456 Notes: (a) Significantly different from chains (two-tailed Fisher’s Exact Test): p<0.0001. (b) & (c) Significantly different from chains (two-tailed Student’s t-test): b p<0.01, c p<0.0001. Table 2: Fat content of chips and quality of cooking fat. Characteristic Number, n(%) Fat content, % wet weight Mean±SD Range Moisture content, % wet weight Mean±SD Range Acid value, mg per g fat Mean±SD Range Value >2.5% Polar compounds , % Mean±SD Range Value >25% Chains (n=27) 27 (18) Independents (n=123) 123 (82) Total (n=150) 12.9±3.1 7-19 11.2±2.7a 5-20 11.5±2.9 Chemical analyses The chips were analysed for total fat and moisture.19 The fr yer fat was analysed for the acid value20 as a reflection of the presence of free f atty acid levels and polar compounds as a reflection of polymerisation.21 Acid values and polar compound values above the commonly recommended levels of 2.5% and 25% respectively, indicate fat degradation.14 43.2±6.8 32-58 49.9±5.6b 33-67 48.7±6.3 4.8±5.0 0.2-17.9 56 11.5±6.6 1-26 4 3.6±3.2 0.1-15.4 54 12.4±7.2 1-37 6 3.8±3.6 54 12.2±7.1 5 Statistical analyses Standard univariate analyses were used to obtain the descriptive baseline data on cooking practices and related attitudinal variables. T-tests w ere conducted on mean responses to identify any significant differences between chain and independent outlets. The chi-square test or Fisher’ Exact Test (where cell numbers s were low) was used to compare categorical variables. Multiple linear regression models were used to explain the variance in fat content of chips, fat acid values and fat polar compounds. Notes: (a) p<0.01. (b) p<0.0001 significantly different from chain outlets (two-tailed Student’s t-test). Results Characteristics of outlets and operators Of the participating outlets (n=150, response rate 80%), 27 were chain outlets and 123 were owned by independent operators (see Table 1). All of the chain operators had English as their f irst language compared with only about half the independent operators. Most of the remainder spoke Chinese (Cantonese/Mandarin) as their f irst language. A Chinese interpreter was used for 23% of the interviews. Previous training of the operators was mainly ‘formal, on-the-job’ for the chain operators and ‘informal, on-thejob’ for the independent operators. The mean fat turnover in the chain outlets was twice as high as the independent outlets and chip turnover was about three times as high. Thus, while chain outlets represented only 18% of the total outlets, they contributed 39% of the chips sold. VOL. AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH 26 NO. 2 The Risks of Everyday Life Fat content of chips in fast-food outlets Table 3: Frying temperature in relation to the reference range (175-190°C). Chains (n=27) Perceived optimal temperature, % outside range Thermostat setting, % outside range Fat temperature just before cooking chips, % outside range Deviation between thermostat and fat temperature, % > 20°C deviation Table 4: Deep frying practices and chip size. Chains (n=27) Filter at least once/day, % Use of filter aids, % Adequate post-cook procedure,b % Chip width, % d ≥12 mm 10 mm 6-8 mm Cr inkle cut, % Ser ve size, g/serve Mean±SD Range Cost of chips, g/$f Mean±SD Range 82±30 47-171 238±87e 59-559 209±100 78 78 30 11 48 41 7 190±73 92-329 Independents (n=123) Total (n=150) Independents (n=123) 24a 2a 62c 83 13 3 15 326±109e 119-749 Total (n=150) 34 16 56 71 19 10 14 302±115 41a 46b,c 47 51c 26b,c,d 22c,d Notes: (a) & (b) Significantly different from chain operators (chi square): a p<0.01, b p<0.05. (c) In addition to 22 unreadable dials (22/123=18% of independents, 22/150=15% of total). (d) In addition to six outlets where steady state was not reached (6/123=5% of independents, 6/150=4% of total). Fat content of chips and quality of cooking fat There was a large range of chip fat content (5-20%) in both types of outlets with the mean fat content being higher from the chain outlets (12.9%) than the independent outlets (11.2%, p<0.01) (see Table 2). More than half of the outlets had high acid values, indicating early fat degradation, and about 5% had high polar compounds, indicating more advanced fat degradation. Notes: (a) p<0.0001, significantly different to chains (chi square test for ‘filter at least once a day’, two-tailed Fisher’s Exact Test for ‘use of filter aids’). (b) Bang and/or shake plus hang f or at least 20 seconds. (c) Significantly different from chain outlets (chi square test). p<0.01. (d) Significant difference between independents and chains across the r ange of chip width, p<0.001. (e) p<0.0001 significantly different from chain outlets (two-tailed Student’s t-test). (f) NZ dollar. Attitudes and perceptions of operators More than two-thirds of the outlet operators thought that a reduction in the fat content of chips would have at least some influence on the health of all New Zealanders (see Table 5). Lack of deep fr ying knowledge was the most impor tant barrier to change. In terms of changing their techniques, the majority did not consider the f inancial cost, customer preferences, or perceived reduced chip flavour to be bar riers to making changes. There were no differences between the responses from the operators of chains, apar t from more chain outlets disagreeing with the statement ‘There are no guidelines or standards for deep frying techniques that I know of ’. Frying temperature The perceived optimal cooking temperature fell outside the reference range of 175-190°C for 70% of chain operators and 41% of independent operators (p<0.05) (see Table 3). There was a wide range of measured fat temperature (136-233°C) with a mean temperature (SD) of 181.9°C (15.3). Temperature deviations of greater than 20°C between the dial setting and the actual fat temperature were found in 26% of the independent outlets. Therefore, 44% of the independent outlets had thermostats that were unreadable or inaccurate. Deep frying practices and chip size Signif icantly fewer independent operators filtered the fat at least once a day compared with chain operators (24% vs. 77%, p<0.0001) (see Table 4). Filter aids are products such as commercial fry powders that are used to prolong the useful life of the frying f at16 and they were used in more than three-quarters of the chain outlets compared with only 2% of the independent outlets. Adequate draining of the chip basket after cooking was observed in 30% of the chains and 62% of the independents (p<0.01). Most of the chain outlets used chips of 10 mm or 6 mm width (89%) while most of the independent outlets used chips of at least 12 mm width (83%, p<0.0001). There were large differences between the types of outlets in the serve size and the cost of chips. The independent outlets sold larger ser ves and sold about three times the amount of chips per dollar compared with the chain outlets. 2002 VOL. 26 NO . 2 Multiple regression models Chip fat content A multiple regression model was developed using the fat content as the dependent variable (see Table 6). Four factors e xplained 24% of the variance in chip fat content. Every 1 mm increase in chip size was associated with 0.18 percentage point decrease in fat content. The thin 6 mm chips f avoured by the chain outlets were therefore over one percentage point fattier than the >12mm chips favoured by independent outlets. Crinkle cut added about the same extra fat content (1.3 percentage points). Ever y 1°C decrease in fat temperature was associated with 0.04 percentage point increase in f at content. Serving size was also significantly negatively associated with chip f at content. The insertion of a variable AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH Morley-John et al. Article Table 5: Attitudes and perceptions of operators. Question “If most fast food outlets in NZ reduced the fat content in their chips, how much would that influence the health of all New Zealanders?” “I don’t really know how to change my deep-frying techniques to reduce the fat content of foods.” “It would cost too much to change my deep-frying techniques.” “My customers do not want foods that are lower in fat.” “If I changed my deep-frying techniques to reduce the fat content, it would make the foods less tasty.” “There are no guidelines or standards for deep-frying techniques that I know of.” Response Some/lot of influence Agree/strongly agree Agree/strongly agree Agree/strongly agree Agree/strongly agree Disagree/strongly disagree Chains Independents 69 48a Notes: (a) Significant difference between chains and independents (two-tailed Fisher’s Exact Test): p<0.0001. for independent/chain outlet into the model resulted in the chip size becoming non-signif icant but the association between a large serving size and lower fat content remained. The reason for serving size being associated with fat content was not clear, but it could be a marker for other practices. Fat acid value Table 6: Multivariate analysis. Variable Chip fat content (%) Chip size (mm) Cr inkle cutb Fat temp (°C) Ser ving size (g) Fat acid value, (mg per g fat) Par tial discard Vinegar useb ß coefficient -0.18 1.3 -0.04 -0.006 p value Model R2 0.24a Partial discard of the fryer fat (i.e. not discarding all of the fr yer fat) had a markedly positive association with a high fat acid value. The use of vinegar in the cleaning process was also associated with high f at acid value. Reasons for this relationship between vinegar use and fat acid value may be due to the vinegar dissolving traces of metals from the vats which accelerate fat degradation, or the vinegar maybe acting as an indicator of some other cleaning or deep frying practice. A high chip tur nover had a marginal association with a high acid value. Polar compounds Partial discarding of the fryer fat had the strongest association with higher polar compounds. The use of vinegar was also associated with high polar compounds and again this could be a 0.16a 4.06 1.84 0.001 8.11 4.47 0.53 2.43 <0.0001 0.0037 0.0739 0.31a <0.0001 <0.0001 <0.0001 0.0238 Chip usage (kg/wk) Polar compounds (%) Par tial discard Vinegar useb Fat age (days) Chinese operatorb Notes: (a)p<0.0001. (b) 1=yes, 0=no . direct effect or a marker for other practices. For e ver y one day increase in fat age there was a 0.5 percentage point increase in polar compounds. Chinese operators had a higher polar compound content, suggesting a marker for other deep frying practices. Discussion This study surveyed a representative sample of New Zealand fast food outlets, with about four -fifths of them being independent outlets. There was a wide range of chip fat content (5-20%) with the average being 11.5%. There were many signif icant differences in operating practices between independent and chain outlets and there was substantial scope for decreasing fat content in both outlet types. The key factors related to this were the low frequency of formal training and best-practice techniques in the independent outlets and the use of thin chips in the chain outlets. Operators in both types of outlets seemed willing to learn more about best-practice techniques and the fat content of their chips. Although 18% of the outlets were chains, their turnover of chips was 39% of the total turnover, indicating their disproportionate contribution to the hot chip market. While most chain operators obviously had systems for providing formal in-house or institution-based training, the lack of formal training was striking among the independent operators. About half the independent operators did not have English as their first language, and this will need to be catered for in any future training programs. The higher fat content of chips from chains was mainly due to their use of thin-cut chips. The greater surface area to volume ratio from thin cut or crinkle cut chips results in a proportionall y larger absor ption of fat.8 The four-fold range in fat content suggests there is plenty of opportunity for improving operating practices. About half the outlets had high acid values. Although traditionally used as an indication of fat degradation, the usefulness VOL. AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH 26 NO. 2 The Risks of Everyday Life Fat content of chips in fast-food outlets of acid value has been questioned.22 Only a small number of outlets had high polar compound values, a mark er of high level of fat deterioration. Frying temperature affects the fat content of the chips. Lower temperatures require longer frying times and the lack of crust formation on the chip surface results in more fat being absorbed.10 A higher cooking temperature leads to greater fat degradation. Excess amounts of degradation products (such as soaps and pol ymer compounds) act as surfactants and increase the fat content of the chips.15 Degraded fat also will not drain as easily off the chips after removal from the vat because of the increased viscosity of the fat.11,16,17 Recommended optimal frying temperatures are around 180-185°C.14,18 In the present study, a ver y wide range of temperatures was recorded (136-233°C). The extremely high temperatures indicate that some operators were unnecessarily degrading their fat and the ver y low temperatures may have contributed to a high chip fat content. Most of the chain outlets cooked chips at temperatures above or below our defined reference range of 175-190°C. Further published research on frying temperatures specific to chip size and fat type may help to tailor the optimal temperature range for each outlet. However, the main problem, especially for independent operators, was that they often didn’t know what temperature their fat was or what it should be. The chain outlets obviously had better quality control systems because the temperature range was much smaller than the independents. With 44% of the independent outlets using unreliable thermostats (either unreadable or inaccurate), an y future intervention should include quality control of thermostats or the use of thermometers. Much of the f at is absorbed by the chip after being removed from the fryer and is related to the adhesion of the fat to the chip surface.10 As only 56% of the outlets fulf illed our criteria of adequate drainage, post-cooking technique may be a fruitful area for training. With short frying times of 3-4.5 minutes, an extra bang of the basket can reduce fat content of chips by two percentage points.9 Fat tur nover and f iltration are related to fat quality.16 Where f at turnover is higher, the frequent replenishment with fresh fat dilutes the old fat and extends its frying life.13 Sixty-f ive per cent of the operators changed the fat at least once a week but many of the operators relied on late signs of degradation (fat and chip colour, fat smoking) in their decision to change the fat. Unfor tunately, there is no widely availab le simple measure for testing the degree of fat degradation. Filtration enhances the fat life by removing crumbs and other particulate matter.16 In the current study, chain outlets usually used automatic filtration units and filtered at least once per day, but independent operators f iltered less frequently using manual methods. Most of the chain operators used adsorbent materials (filter aids) to remove soluble compounds that are formed during the fr ying process and adversely affect f at quality. Filter aids have been reported to extend the life of the fat although more extensive study may be required to assess their value.16 It is recommended that after cleaning the vat with detergent or caustic compounds that the vat is rinsed with a vinegar solution 2002 VOL. 26 NO . 2 (weak acid) to avoid the formation of alkaline soaps that increase the de gradation of the fat.16 A f inal flushing of residual chemicals with water and drying of the vat is then required. The association between vinegar use and a higher degree of degradation found in this study needs further investigation to determine if this is cause and effect or confounding. There was a large range of chip serving size, with chains selling signif icantly more expensive and smaller chip servings. This financial benef it of using thin-cut chips will present a major barrier for chain outlets moving to thick-cut chips. Many of the operators thought the health of New Zealand people could be influenced by a reduction in the fat of their chips. There appeared to be a willingness to learn new techniques, with the main bar rier to changing their cooking practices being their lack of knowledge. Implications Small changes in the fat content of chips can have an important impact at a population level because it is such a high-volume food. Estimates of the total amount of potatoes used for hot chips in New Zealand (150,000 tonnes/y) minus waste (35%) and exports (40,000 tonnes/y) leaves the NZ consumption at about 60,000 tonnes/y (R. Gall, Vegfed Nov 2001). A reduction in mean fat content of chips by 20% (from 11.5% to 9.2%) translates into a reduction of about 0.4 kg/capita/y. Some outlets can achieve a 5% fat content, and while this is ideal, it is unlikely to be a realistic target for a mean fat content. Some verif ication of these estimates comes from the National Nutrition Survey where the daily repor ted total fat intake for adults was 94 g/d, of which 6.6 g came from potatoes (presumed to be almost all hot chips and potato crisps). A 20% reduction in fat content of hot chips (11.5% to 9.2%) and of potato crisps (about 32% to 26%) are both achievable through changed deep frying and post-fr ying techniques. This would result in a reduction of fat intak e per adult per year of about 0.5 kg. The annual weight gain of the adult New Zealand population is about 0.4 kg/adult/y. 1 The main point about the public health significance of this hot chips study is that relatively small achievable changes in the fat content of high volume fatty foods can have important population effects. The magnitude of a 20% decrease in the fat content of the chips is similar to the weight gain of the population. Further controlled experiments on factors related to the f at content of chips need to be done to identify optimal practices. In addition, f ield-based cost-benefit analyses of changing techniques or frying fat will also be impor tant because of the financial implications for the operators. There are opportunities to reduce the fat content of chips through operator training and perhaps cer tif ication of outlets. Certif ication of operators for deep fr ying, similar to certif ication for food safety, is one option and endorsement of outlets by organisations such as the National Heart Foundation is another option for promoting best practice deep frying techniques. Promoting the use AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH Morley-John et al. Article of unsaturated fats for frying is also of major public health importance (although beyond the scope of this report). This study has provided baseline data of current takeaway practices that can be compared with future monitoring surveys in response to interventions and training programs. Commentary: Chips with everything P. A. Baghurst Public Health Research Unit, Women’s and Children’s Hospital, South Australia Acknowledgements This study was funded through grants from the National Hear t Foundation and the Ministry of Health. Auckland Healthcare supported the work of Heather Wright on this project and the National Heart Foundation supported Boyd Swinburn’s time on the project. The authors gratefully acknowledge the chemical analysis and advice given by Jill Schumacher and Jim Mitchell, ESR . http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Australian and New Zealand Journal of Public Health Wiley

The Risks of Everyday Life: Fat content of chips, quality of frying fat and deep‐frying practices in New Zealand fast food outlets

Loading next page...
 
/lp/wiley/the-risks-of-everyday-life-fat-content-of-chips-quality-of-frying-fat-rFIZ5JhxlI

References (30)

Publisher
Wiley
Copyright
Copyright © 2002 Wiley Subscription Services
ISSN
1326-0200
eISSN
1753-6405
DOI
10.1111/j.1467-842X.2002.tb00900.x
Publisher site
See Article on Publisher Site

Abstract

Judith Morley-John, Boyd A. Swinburn, Patricia A. Metcalf and Fezeela Raza Department of Community Health, University of Auckland, New Zealand Abstract Objectives: To collect baseline data on the fat content of hot chips, quality (degradation) of cooking fat, deep-frying practices and related attitudes in fast food outlets in New Zealand. To identify the key determinants of the fat content of chips and quality of cooking fat. Methods: A nationally representative sample of fast food outlets (n=150, response rate 80%) was surveyed between September 1998 and March 1999. Data collected included a questionnaire, observation of cooking practices and analysis of cooked chips and frying fat. Heather Wright Public Health Protection, Auckland Healthcare Services Ltd, New Zealand he prevalence of obesity is increasing in New Zealand,1,2 as it is in man y countries,3 and is now widely regarded as a pandemic.3,4 A reduction in the prevalence of obesity and related diseases will require a broader public health approach with an emphasis on making the environment less ‘obesogenic’ 5 by ensuring that healthy food choices and physical activity choices are made the easy choices.6,7 The fat content of deep-fried food is likely to contribute to obesity by increasing the intake of total fat and energy. The 1997 National Nutrition Survey reported that deepfried hot chips (French fries) are eaten by 47% of New Zealanders at least once a week.1 Many aspects of deep fr ying have been reported to influence the fat content of chips – chip size and shape,8 post-fr ying practices (such as shaking and hanging the basket),9,10 fat temperature and fat quality (i.e. the degree of fat degradation).11-14 Fat quality is influenced by frying temperature,15 vat cleaning, fat turnover, filtering and the use of f iltering agents.16 There is signif icant potential for public health inter ventions to promote best practice deep-frying techniques and this survey collected the baseline data needed to target and monitor those interventions. The specif ic aims of the hot chips study were to: • Collect baseline data on the fat content of chips, quality of cooking fat, deepfr ying practices and related attitudes and beliefs in fast food outlets throughout New Zealand. • Compare chain and independent outlets. • Identify the key determinants of fat content of chips and quality of cooking fat. • Identify opportunities for action to reduce the fat content of hot chips and improve the fat quality of the frying fat. Results: Only 8% of independent operators had formal training in deep frying practices compared with 93% of chain operators. There was a wide range of fat content of chips (5%-20%, mean 11.5%). The use of thinner chips, crinkle cut chips and lower fryer fat temperature were associated with higher chip fat content. Eighty-nine per cent of chain outlets used 6-10 mm chips compared with 83% of independent outlets that used chips ≥12 mm. A wide range of frying temperatures was recorded (136233°C) with 58% of outlets frying outside the reference range (175-190°C). As indices of fat degradation, fat acid and polar compound values above the recommended levels occurred in 54% and 5% of outlets respectively. Operators seemed willing to learn more about best practice techniques, with lack of knowledge being the main barrier to change. Conclusions and implications: Deep frying practices could be improved through operator training and certification options. Even a small decrease in the mean fat content of chips would reduce the obesogenic impact of this popular food. ( Aust N Z J Public Health 2002; 26: 101-7) Methods Sample Fast food outlets were randomly selected from a list of all takeaways and unlicensed restaurants listed in the Ne w Zealand Telecom Yellow Pages (n=3,301). They were stratif ied by geographical re gion and target numbers were defined to obtain a representative sample. Only the outlets that sold takeaway hot chips were eligible. On telephoning the selected outlets, 63% were found to be eligible and 80% agreed to take part. It was assumed that the vast majority of fast food outlets were listed in the Yellow Pages (although this was not independently verif ied). Survey process The data were collected during the period September 1998 to March 1999. An interviewer-administered questionnaire (29 closed questions) was designed to investigate chip-cooking practices and attitudes related to cooking practices. This instrument was developed on the basis of: • A literature review identifying key Correspondence to: Boyd Swinburn, School of Health Sciences, Deakin University, 221 Burwood Highway, Burwood, Victoria 3125. Fax: (03) 9244 6017; e-mail: swinburn@deakin.edu.au Submitted: April 2001 Revision requested: November 2001 Accepted: March 2002 2002 VOL. 26 NO . 2 AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH Morley-John et al. Article cooking procedures that can be measured and that are amenable to change. • Data from the pilot sur vey of 24 independent tak eaway bars within the Auckland region. • Infor mation obtained through consultation with fast food outlet operators and food technology experts in the area. The attitudes and perceptions of operators were measured using Liker t scales. Fat and chip turnovers were assessed by asking the operators about their monthly order volumes for each. Many of the operators were Chinese immig rants and translations of the participant infor mation sheet and interpreters were used as required. The study w as approved by the University of Auckland Human Subjects Ethics Committee. A 200 mL sample of fat from the main chip fryer was collected for fat analysis. The thermostat setting and actual f at temperature once the fryer had reached temperature (thermostat light off) were recorded. The temperature range of 180-185°C is considered optimal based on recommendations in other countries14,18 and we considered that 5°C either side of this (i.e. 175-190°C) would be an appropriate reference range for this study. The operator was asked to cook a serving of chips. Observations of the post-cooking procedure (banging, shaking and hanging of the chip basket) were recorded. Post-cooking drainage of fat from the chips was considered adequate if the basket was shaken and/or banged and hung for at least 20 seconds before being packaged. The chips were immediately transfer red from the packaging to a plastic bag. The questionnaire was then administered. The f at and chip samples were chilled and then frozen (-18°C) as soon as possible until analysis. Table 1: Characteristics of fast food outlets and operators. Characteristic Number, n(%) English first language, n(%) T raining, n(%) Informal, on-the-job Formal, on-the-job Formal institution Fat turnover, kg/wk Mean±SD Range Chip turnover, kg/wk Mean±SD Range 126±105 10-425 925±623 75-2400 60±43b 4-250 316±321c 5-2000 72±64 Chains 27 (18) 27 (100) 2 (7) 24 (89) 1 (4) Independents 123 (82) 66 (54) 113 (92)a 3 (2)a 7 (6) Total 150 94 (63) 115 (77) 27 (18) 8 (5) 426±456 Notes: (a) Significantly different from chains (two-tailed Fisher’s Exact Test): p<0.0001. (b) & (c) Significantly different from chains (two-tailed Student’s t-test): b p<0.01, c p<0.0001. Table 2: Fat content of chips and quality of cooking fat. Characteristic Number, n(%) Fat content, % wet weight Mean±SD Range Moisture content, % wet weight Mean±SD Range Acid value, mg per g fat Mean±SD Range Value >2.5% Polar compounds , % Mean±SD Range Value >25% Chains (n=27) 27 (18) Independents (n=123) 123 (82) Total (n=150) 12.9±3.1 7-19 11.2±2.7a 5-20 11.5±2.9 Chemical analyses The chips were analysed for total fat and moisture.19 The fr yer fat was analysed for the acid value20 as a reflection of the presence of free f atty acid levels and polar compounds as a reflection of polymerisation.21 Acid values and polar compound values above the commonly recommended levels of 2.5% and 25% respectively, indicate fat degradation.14 43.2±6.8 32-58 49.9±5.6b 33-67 48.7±6.3 4.8±5.0 0.2-17.9 56 11.5±6.6 1-26 4 3.6±3.2 0.1-15.4 54 12.4±7.2 1-37 6 3.8±3.6 54 12.2±7.1 5 Statistical analyses Standard univariate analyses were used to obtain the descriptive baseline data on cooking practices and related attitudinal variables. T-tests w ere conducted on mean responses to identify any significant differences between chain and independent outlets. The chi-square test or Fisher’ Exact Test (where cell numbers s were low) was used to compare categorical variables. Multiple linear regression models were used to explain the variance in fat content of chips, fat acid values and fat polar compounds. Notes: (a) p<0.01. (b) p<0.0001 significantly different from chain outlets (two-tailed Student’s t-test). Results Characteristics of outlets and operators Of the participating outlets (n=150, response rate 80%), 27 were chain outlets and 123 were owned by independent operators (see Table 1). All of the chain operators had English as their f irst language compared with only about half the independent operators. Most of the remainder spoke Chinese (Cantonese/Mandarin) as their f irst language. A Chinese interpreter was used for 23% of the interviews. Previous training of the operators was mainly ‘formal, on-the-job’ for the chain operators and ‘informal, on-thejob’ for the independent operators. The mean fat turnover in the chain outlets was twice as high as the independent outlets and chip turnover was about three times as high. Thus, while chain outlets represented only 18% of the total outlets, they contributed 39% of the chips sold. VOL. AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH 26 NO. 2 The Risks of Everyday Life Fat content of chips in fast-food outlets Table 3: Frying temperature in relation to the reference range (175-190°C). Chains (n=27) Perceived optimal temperature, % outside range Thermostat setting, % outside range Fat temperature just before cooking chips, % outside range Deviation between thermostat and fat temperature, % > 20°C deviation Table 4: Deep frying practices and chip size. Chains (n=27) Filter at least once/day, % Use of filter aids, % Adequate post-cook procedure,b % Chip width, % d ≥12 mm 10 mm 6-8 mm Cr inkle cut, % Ser ve size, g/serve Mean±SD Range Cost of chips, g/$f Mean±SD Range 82±30 47-171 238±87e 59-559 209±100 78 78 30 11 48 41 7 190±73 92-329 Independents (n=123) Total (n=150) Independents (n=123) 24a 2a 62c 83 13 3 15 326±109e 119-749 Total (n=150) 34 16 56 71 19 10 14 302±115 41a 46b,c 47 51c 26b,c,d 22c,d Notes: (a) & (b) Significantly different from chain operators (chi square): a p<0.01, b p<0.05. (c) In addition to 22 unreadable dials (22/123=18% of independents, 22/150=15% of total). (d) In addition to six outlets where steady state was not reached (6/123=5% of independents, 6/150=4% of total). Fat content of chips and quality of cooking fat There was a large range of chip fat content (5-20%) in both types of outlets with the mean fat content being higher from the chain outlets (12.9%) than the independent outlets (11.2%, p<0.01) (see Table 2). More than half of the outlets had high acid values, indicating early fat degradation, and about 5% had high polar compounds, indicating more advanced fat degradation. Notes: (a) p<0.0001, significantly different to chains (chi square test for ‘filter at least once a day’, two-tailed Fisher’s Exact Test for ‘use of filter aids’). (b) Bang and/or shake plus hang f or at least 20 seconds. (c) Significantly different from chain outlets (chi square test). p<0.01. (d) Significant difference between independents and chains across the r ange of chip width, p<0.001. (e) p<0.0001 significantly different from chain outlets (two-tailed Student’s t-test). (f) NZ dollar. Attitudes and perceptions of operators More than two-thirds of the outlet operators thought that a reduction in the fat content of chips would have at least some influence on the health of all New Zealanders (see Table 5). Lack of deep fr ying knowledge was the most impor tant barrier to change. In terms of changing their techniques, the majority did not consider the f inancial cost, customer preferences, or perceived reduced chip flavour to be bar riers to making changes. There were no differences between the responses from the operators of chains, apar t from more chain outlets disagreeing with the statement ‘There are no guidelines or standards for deep frying techniques that I know of ’. Frying temperature The perceived optimal cooking temperature fell outside the reference range of 175-190°C for 70% of chain operators and 41% of independent operators (p<0.05) (see Table 3). There was a wide range of measured fat temperature (136-233°C) with a mean temperature (SD) of 181.9°C (15.3). Temperature deviations of greater than 20°C between the dial setting and the actual fat temperature were found in 26% of the independent outlets. Therefore, 44% of the independent outlets had thermostats that were unreadable or inaccurate. Deep frying practices and chip size Signif icantly fewer independent operators filtered the fat at least once a day compared with chain operators (24% vs. 77%, p<0.0001) (see Table 4). Filter aids are products such as commercial fry powders that are used to prolong the useful life of the frying f at16 and they were used in more than three-quarters of the chain outlets compared with only 2% of the independent outlets. Adequate draining of the chip basket after cooking was observed in 30% of the chains and 62% of the independents (p<0.01). Most of the chain outlets used chips of 10 mm or 6 mm width (89%) while most of the independent outlets used chips of at least 12 mm width (83%, p<0.0001). There were large differences between the types of outlets in the serve size and the cost of chips. The independent outlets sold larger ser ves and sold about three times the amount of chips per dollar compared with the chain outlets. 2002 VOL. 26 NO . 2 Multiple regression models Chip fat content A multiple regression model was developed using the fat content as the dependent variable (see Table 6). Four factors e xplained 24% of the variance in chip fat content. Every 1 mm increase in chip size was associated with 0.18 percentage point decrease in fat content. The thin 6 mm chips f avoured by the chain outlets were therefore over one percentage point fattier than the >12mm chips favoured by independent outlets. Crinkle cut added about the same extra fat content (1.3 percentage points). Ever y 1°C decrease in fat temperature was associated with 0.04 percentage point increase in f at content. Serving size was also significantly negatively associated with chip f at content. The insertion of a variable AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH Morley-John et al. Article Table 5: Attitudes and perceptions of operators. Question “If most fast food outlets in NZ reduced the fat content in their chips, how much would that influence the health of all New Zealanders?” “I don’t really know how to change my deep-frying techniques to reduce the fat content of foods.” “It would cost too much to change my deep-frying techniques.” “My customers do not want foods that are lower in fat.” “If I changed my deep-frying techniques to reduce the fat content, it would make the foods less tasty.” “There are no guidelines or standards for deep-frying techniques that I know of.” Response Some/lot of influence Agree/strongly agree Agree/strongly agree Agree/strongly agree Agree/strongly agree Disagree/strongly disagree Chains Independents 69 48a Notes: (a) Significant difference between chains and independents (two-tailed Fisher’s Exact Test): p<0.0001. for independent/chain outlet into the model resulted in the chip size becoming non-signif icant but the association between a large serving size and lower fat content remained. The reason for serving size being associated with fat content was not clear, but it could be a marker for other practices. Fat acid value Table 6: Multivariate analysis. Variable Chip fat content (%) Chip size (mm) Cr inkle cutb Fat temp (°C) Ser ving size (g) Fat acid value, (mg per g fat) Par tial discard Vinegar useb ß coefficient -0.18 1.3 -0.04 -0.006 p value Model R2 0.24a Partial discard of the fryer fat (i.e. not discarding all of the fr yer fat) had a markedly positive association with a high fat acid value. The use of vinegar in the cleaning process was also associated with high f at acid value. Reasons for this relationship between vinegar use and fat acid value may be due to the vinegar dissolving traces of metals from the vats which accelerate fat degradation, or the vinegar maybe acting as an indicator of some other cleaning or deep frying practice. A high chip tur nover had a marginal association with a high acid value. Polar compounds Partial discarding of the fryer fat had the strongest association with higher polar compounds. The use of vinegar was also associated with high polar compounds and again this could be a 0.16a 4.06 1.84 0.001 8.11 4.47 0.53 2.43 <0.0001 0.0037 0.0739 0.31a <0.0001 <0.0001 <0.0001 0.0238 Chip usage (kg/wk) Polar compounds (%) Par tial discard Vinegar useb Fat age (days) Chinese operatorb Notes: (a)p<0.0001. (b) 1=yes, 0=no . direct effect or a marker for other practices. For e ver y one day increase in fat age there was a 0.5 percentage point increase in polar compounds. Chinese operators had a higher polar compound content, suggesting a marker for other deep frying practices. Discussion This study surveyed a representative sample of New Zealand fast food outlets, with about four -fifths of them being independent outlets. There was a wide range of chip fat content (5-20%) with the average being 11.5%. There were many signif icant differences in operating practices between independent and chain outlets and there was substantial scope for decreasing fat content in both outlet types. The key factors related to this were the low frequency of formal training and best-practice techniques in the independent outlets and the use of thin chips in the chain outlets. Operators in both types of outlets seemed willing to learn more about best-practice techniques and the fat content of their chips. Although 18% of the outlets were chains, their turnover of chips was 39% of the total turnover, indicating their disproportionate contribution to the hot chip market. While most chain operators obviously had systems for providing formal in-house or institution-based training, the lack of formal training was striking among the independent operators. About half the independent operators did not have English as their first language, and this will need to be catered for in any future training programs. The higher fat content of chips from chains was mainly due to their use of thin-cut chips. The greater surface area to volume ratio from thin cut or crinkle cut chips results in a proportionall y larger absor ption of fat.8 The four-fold range in fat content suggests there is plenty of opportunity for improving operating practices. About half the outlets had high acid values. Although traditionally used as an indication of fat degradation, the usefulness VOL. AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH 26 NO. 2 The Risks of Everyday Life Fat content of chips in fast-food outlets of acid value has been questioned.22 Only a small number of outlets had high polar compound values, a mark er of high level of fat deterioration. Frying temperature affects the fat content of the chips. Lower temperatures require longer frying times and the lack of crust formation on the chip surface results in more fat being absorbed.10 A higher cooking temperature leads to greater fat degradation. Excess amounts of degradation products (such as soaps and pol ymer compounds) act as surfactants and increase the fat content of the chips.15 Degraded fat also will not drain as easily off the chips after removal from the vat because of the increased viscosity of the fat.11,16,17 Recommended optimal frying temperatures are around 180-185°C.14,18 In the present study, a ver y wide range of temperatures was recorded (136-233°C). The extremely high temperatures indicate that some operators were unnecessarily degrading their fat and the ver y low temperatures may have contributed to a high chip fat content. Most of the chain outlets cooked chips at temperatures above or below our defined reference range of 175-190°C. Further published research on frying temperatures specific to chip size and fat type may help to tailor the optimal temperature range for each outlet. However, the main problem, especially for independent operators, was that they often didn’t know what temperature their fat was or what it should be. The chain outlets obviously had better quality control systems because the temperature range was much smaller than the independents. With 44% of the independent outlets using unreliable thermostats (either unreadable or inaccurate), an y future intervention should include quality control of thermostats or the use of thermometers. Much of the f at is absorbed by the chip after being removed from the fryer and is related to the adhesion of the fat to the chip surface.10 As only 56% of the outlets fulf illed our criteria of adequate drainage, post-cooking technique may be a fruitful area for training. With short frying times of 3-4.5 minutes, an extra bang of the basket can reduce fat content of chips by two percentage points.9 Fat tur nover and f iltration are related to fat quality.16 Where f at turnover is higher, the frequent replenishment with fresh fat dilutes the old fat and extends its frying life.13 Sixty-f ive per cent of the operators changed the fat at least once a week but many of the operators relied on late signs of degradation (fat and chip colour, fat smoking) in their decision to change the fat. Unfor tunately, there is no widely availab le simple measure for testing the degree of fat degradation. Filtration enhances the fat life by removing crumbs and other particulate matter.16 In the current study, chain outlets usually used automatic filtration units and filtered at least once per day, but independent operators f iltered less frequently using manual methods. Most of the chain operators used adsorbent materials (filter aids) to remove soluble compounds that are formed during the fr ying process and adversely affect f at quality. Filter aids have been reported to extend the life of the fat although more extensive study may be required to assess their value.16 It is recommended that after cleaning the vat with detergent or caustic compounds that the vat is rinsed with a vinegar solution 2002 VOL. 26 NO . 2 (weak acid) to avoid the formation of alkaline soaps that increase the de gradation of the fat.16 A f inal flushing of residual chemicals with water and drying of the vat is then required. The association between vinegar use and a higher degree of degradation found in this study needs further investigation to determine if this is cause and effect or confounding. There was a large range of chip serving size, with chains selling signif icantly more expensive and smaller chip servings. This financial benef it of using thin-cut chips will present a major barrier for chain outlets moving to thick-cut chips. Many of the operators thought the health of New Zealand people could be influenced by a reduction in the fat of their chips. There appeared to be a willingness to learn new techniques, with the main bar rier to changing their cooking practices being their lack of knowledge. Implications Small changes in the fat content of chips can have an important impact at a population level because it is such a high-volume food. Estimates of the total amount of potatoes used for hot chips in New Zealand (150,000 tonnes/y) minus waste (35%) and exports (40,000 tonnes/y) leaves the NZ consumption at about 60,000 tonnes/y (R. Gall, Vegfed Nov 2001). A reduction in mean fat content of chips by 20% (from 11.5% to 9.2%) translates into a reduction of about 0.4 kg/capita/y. Some outlets can achieve a 5% fat content, and while this is ideal, it is unlikely to be a realistic target for a mean fat content. Some verif ication of these estimates comes from the National Nutrition Survey where the daily repor ted total fat intake for adults was 94 g/d, of which 6.6 g came from potatoes (presumed to be almost all hot chips and potato crisps). A 20% reduction in fat content of hot chips (11.5% to 9.2%) and of potato crisps (about 32% to 26%) are both achievable through changed deep frying and post-fr ying techniques. This would result in a reduction of fat intak e per adult per year of about 0.5 kg. The annual weight gain of the adult New Zealand population is about 0.4 kg/adult/y. 1 The main point about the public health significance of this hot chips study is that relatively small achievable changes in the fat content of high volume fatty foods can have important population effects. The magnitude of a 20% decrease in the fat content of the chips is similar to the weight gain of the population. Further controlled experiments on factors related to the f at content of chips need to be done to identify optimal practices. In addition, f ield-based cost-benefit analyses of changing techniques or frying fat will also be impor tant because of the financial implications for the operators. There are opportunities to reduce the fat content of chips through operator training and perhaps cer tif ication of outlets. Certif ication of operators for deep fr ying, similar to certif ication for food safety, is one option and endorsement of outlets by organisations such as the National Heart Foundation is another option for promoting best practice deep frying techniques. Promoting the use AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH Morley-John et al. Article of unsaturated fats for frying is also of major public health importance (although beyond the scope of this report). This study has provided baseline data of current takeaway practices that can be compared with future monitoring surveys in response to interventions and training programs. Commentary: Chips with everything P. A. Baghurst Public Health Research Unit, Women’s and Children’s Hospital, South Australia Acknowledgements This study was funded through grants from the National Hear t Foundation and the Ministry of Health. Auckland Healthcare supported the work of Heather Wright on this project and the National Heart Foundation supported Boyd Swinburn’s time on the project. The authors gratefully acknowledge the chemical analysis and advice given by Jill Schumacher and Jim Mitchell, ESR .

Journal

Australian and New Zealand Journal of Public HealthWiley

Published: Jan 1, 2002

There are no references for this article.