Cost and convenience influence meal choice( Reference Glanz, Basil and Maibach 1 , Reference McDermott and Stephens 2 ). Consumption of fast foods and use of pre-prepared items and ready meals are common, and increasing worldwide( Reference Lachat, Nago and Verstraeten 3 – Reference Steyn, Mchiza and Hill 6 ). Takeaway meals are meals obtained quickly, without wait service, purchased in self-serve or carry-out venues. These meals tend to be energy dense, nutrient poor and high in saturated fat and Na( Reference Dunford, Webster and Barzi 7 , Reference Nielsen and Popkin 8 ).
It is perceived that healthy food costs more than unhealthy food, that convenience foods are better value for money, and there is a lack of time for food preparation( Reference Lee, Ni Mhurchu and Sacks 9 – Reference Lopéz-Azpiazu, Martìnez-González and Kearney 13 ). Time is an often overlooked social determinant of health( Reference Venn and Strazdins 14 ) not usually factored into the cost of food preparation, but is a barrier to preparing meals( Reference Muth, Karns and Zmuda 15 – Reference Davis and You 17 ). The evidence on incorporating time in meal costing studies is limited( Reference Davis and You 18 ).
The International Network for Food and Obesity/non-communicable diseases Research, Monitoring and Action Support (INFORMAS) aims to monitor key aspects of food environments, including food prices and affordability( Reference Lee, Ni Mhurchu and Sacks 9 ). The present study focuses on the meals component of the INFORMAS food price and affordability module by comparing the cost of takeaway v. home-cooked and home-assembled meals. The study aimed to assess differences in cost between popular takeaway meals compared with similar, but healthier home-made and home-assembled meals, with and without the inclusion of time.
Methods
Frequently consumed takeaways were identified and matched to similar, but healthier, home-made and home-assembled meals. The home-made meals used common ingredients and required basic cooking skills and standard kitchen equipment.
The methods are summarised below. The detailed protocol for the study can be found in the online supplementary material, Supplemental Material 1.
Takeaway meals
Popular takeaway meals (Tables 1 and 2) were identified from the 2013 New Zealand (NZ) Household Expenditure Survey( 19 ), the 2008/09 Adult NZ Nutrition Survey( 20 ) and a survey of a convenience sample of 144 takeaway outlets. Popular fast-food outlets in NZ were identified using Euromonitor data( Reference Euromonitor 21 ). The nutrient compositions of the Domino’s, McDonald’s and KFC meals were sourced from the respective websites( 22 – 24 ) and for the remaining meals from the NZ Food Composition Database( 25 ).
The meals from chain restaurants (Domino’s, McDonald’s and KFC) were value meals for four with set prices and sizes across stores; weights and prices were collected from one outlet of each chain. The remaining takeaway meals were from independent outlets, so varied in size, price and components; these outlets were randomly sampled and selected. Seven census area units with a reasonable number of takeaway outlets in Auckland city were selected; two areas with lower, three with medium and two with higher deprivation scores as per the NZ Deprivation Index 2013( Reference Atkinson, Salmond and Crampton 26 ). All takeaway outlets in these census area units serving the identified popular takeaway meals were enumerated and two outlets of each type were selected for each popular takeaway meal (fourteen prices for each meal).
Meals were purchased from takeaway outlets between November 2015 and March 2016. The meal and the main components (rice/noodles/potatoes, meat and vegetables) were weighed.
Home-prepared meals
For the purpose of the study, a meal included vegetables, a protein source and a carbohydrate component. The nutrition targets to select the home-made and home-assembled meals were guided by criteria for healthy recipes and ready meals identified in the literature( 27 – 32 ) and the NZ Eating and Activity Guidelines( 33 ).
Meal nutrition targets for two adults and two children were:
-
∙ minimum 600 g non-starchy vegetables;
-
∙ maximum raw weight 500 g red meat, 600 g skinless poultry, 600 g seafood;
-
∙ ≥20 g protein;
-
∙ ≤24 g saturated fat; and
-
∙ ≤3600 mg Na.
The rationale for decisions for the home-prepared meals is outlined in Table 1. Recipes from popular NZ recipe books, magazines and websites were used to identify key ingredients and flavours of each home-made meal so the characteristics were similar to their takeaway counterpart. The ingredients of each home recipe were entered into the nutrient analysis software FoodWorks 7 Professional (Xyris Software (Australia) Pty Ltd, 2012) with the NZ Food Composition Database and the mean nutrient content for each meal option was calculated.
For the home-assembled meals, the main ingredients of the home-made meals were replaced by pre-prepared items wherever possible. For example, fresh vegetables were replaced with frozen vegetables. Some preparation and assembling was required. Combinations of the meal components were checked against the nutrition targets.
The process is shown in online supplementary material, Supplemental Material 2.
Cost of meals
The cheapest available item, including generic brands, for each ingredient was selected and priced at six supermarkets from major chains in similar areas to the takeaway outlets.
The time to prepare the meal was tested in a home kitchen, or estimated from a tested meal with a similar method. The time to order and to wait for each takeaway meal was recorded. The cost of time was selected using the standard market substitute approach, which values food preparation time at the amount the labour could be purchased on the market( Reference Davis and You 18 ). The minimum wage of $NZ 15·25( 34 ) was selected to cost time, as this is similar to the hourly wage of a food preparer.
The price, mean cost, se, 95 % CI (±1·96 se) and range were calculated for each meal and costs compared between the different types of meal (takeaway, home-made, home-assembled). The analysis was conducted with and without including the cost of time. As the weights of meals varied, the price per kilogram was calculated.
Results
Seven to ten distinct recipes were selected for each home-made meal. The combinations for the home-assembled meals ranged from three combinations for pizza to thirty combinations for beef chow mein.
Healthiness of meals
The home-prepared meals were designed to be healthy so provided more vegetables (at least 600 g non-starchy vegetables except pizza) and less saturated fat and Na than their takeaway counterparts (Table 2).
The home-prepared meals did not exceed the saturated fat target, except for the burgers and home-assembled chicken meal (Table 2). All of the takeaway meals exceeded the saturated fat target, except pizza.
None of the home-made meal recipes exceeded the Na target (Table 2). Three of the home-assembled meals (chicken meal, beef chow mein, burger) exceeded the Na target due to the high Na content of specific components. All but one takeaway meal (fish and chips) exceeded the Na target.
Cost of meals
Time exclusive
The cost of the home-made meals was significantly cheaper than the takeaway counterparts for all but one meal option (fish and chips; Table 3). The cost of the home-assembled meals was significantly cheaper than the takeaway counterparts. Three home-made meals (chicken meal, beef chow mein, pizza) were the cheapest options. Both home-prepared butter chicken meals were the cheapest options.
* Number of recipes, meal combinations or takeaway outlets with a price.
† Time in minutes; preparation time for home-made and home-assembled meals and waiting time for takeaway meals.
‡ When the meal was a fixed price in different outlets, there are no confidence intervals.
§ Cost per kilogram does not include the cost of time.
Time inclusive
When the cost of time was added (Table 3) all home-assembled meals were significantly cheaper than other options, except pizza. The takeaway meal was the most expensive option for the chicken meal, burger and pizza (32, 27 and 19 % more than the home-made meal, respectively). The home-made meal was the most expensive option for the fish and chip meal (14 % more than the takeaway meal). Including the cost of time reduced the relative difference between the cost of takeaway and home-made meals; the takeaway meals cost from 14 % less to 32 % more than the home-made meals (11–100 % more without time). There was little change in the relative difference in cost between takeaways and home-assembled meals.
Discussion
In general, healthier home-cooked and home-assembled meals were cheaper than their takeaway counterparts, when either the cost of the complete meal, or the cost standardised for weight, was calculated. As the home-made meals required at least 45 % more preparation time than the other meals, adding the cost of relevant preparation (home meals) and waiting time (takeaway meals) made the home-assembled meals the cheapest option and either the home-made or takeaway meals the most expensive.
Home-assembled meals are potentially a better option than takeaway meals, as they are 15–48 % cheaper, have similar preparation/waiting time, and can provide a healthy meal if pre-prepared ingredients lower in saturated fat and Na are chosen.
Although the difference between the cost of home-made, home-assembled and takeaway meals was checked for statistical significance, what is important is the meaningful difference in cost between the meals that would influence the consumer’s decision to choose one meal type over another. This is challenging to quantify, as meal preparation is a trade-off between the cost of purchasing food and time available( Reference Becker 35 ) with influences such as taste and culture( Reference Yang, Davis and Muth 36 ). Households differ on the value placed on nutrition, providing a home-made meal, the priority of food in the budget, available time, and whether meal preparation is experienced as a pleasure or a chore( Reference Caraher, Dixon and Lang 37 – Reference Guthrie, Lin and Frazao 42 ). Some are motivated to purchase takeaway meals, paying for the cost of service to save time, but may trade off healthiness, quality and taste( Reference Costa, Dekker and Beumer 43 – Reference Ryan 46 ). A study using two US survey data sets estimated the price elasticity of demand for different types of food purchased away from home and concluded that an increase in the price of fast food may shift consumption to meals prepared at home( Reference Richards and Mancino 47 ).
Time
Similar to the methodology in another study( Reference Yang, Davis and Muth 36 ), hands-on preparation time was used rather than the full cooking time as the meal preparer could conduct other activities during cooking time. The time to shop for ingredients and transport time to food stores or takeaway outlets were not calculated; this time will vary between households, and it was assumed purchase of ingredients would be part of regular household shopping. Cooking fuel was not included; the cost of electricity to cook meals was estimated at $NZ 0·26 per meal( 48 ).
A US study( Reference Yang, Davis and Muth 36 ) reported pre-prepared items (e.g. apple sauce) and meals (e.g. lasagne) cost less than the home recipe when the cost of time was included. The cost to prepare the US Thrifty Food Plan was met by 62 % of low-income households, but when time costs were included only 13 % could afford the required foods( Reference Davis and You 17 ).
Strengths and limitations
To account for possible variations in price, weight and composition of takeaway meals from independent outlets, the takeaway outlets were selected from a range of census area units.
It was challenging to determine the appropriate meal size. There is no consensus on what is considered a meal( Reference Murakami and Livingstone 49 , Reference Leech, Worsley and Timperio 50 ), so the home-prepared meals were matched to their takeaway counterparts for the amount of rice, noodles or potatoes. The meal cost was also standardised for weight (1 kg) and the pattern of results was similar to results for the total meal cost. When the cost per kilogram was calculated without vegetables, the pattern of results was similar to the price per kilogram with vegetables, but the relative differences were smaller.
The preparation time may not be an accurate indication for an average household or those with low cooking skills. It was assumed waiting times would be similar across different outlets of a fast-food chain. Ideally, a sensitivity analysis should have been conducted to account for other time factors such as grocery shopping and transportation time.
Implications
Monitoring whether the cost to the consumer of home-cooked meals increases at a faster rate than the cost of takeaway meals is important, as changing relative costs is one factor that could affect the consumption of takeaways. This research provides a method to compare the cost of meals across the spectrum of preparation from home-made to home-assembled to takeaway meals. The cost differential between each meal option can be compared at one point, monitored over time and compared with cost differentials between types of meals in other countries. These results can be utilised for education purposes to encourage households to prepare home-made meals and to address the barrier of time.
Conclusion
Healthier options of home-prepared meals were generally cheaper than their takeaway counterparts, for the cost of the complete meal and the cost standardised for weight. Adding the cost of preparation and waiting time made the home-assembled meals the cheapest and either the home-made or takeaway meal the most expensive option. Home-made meals can be healthy and cheap but do require time. Home-assembled meals are quicker to prepare and can be cheaper and healthier than takeaways, so is a recommended option if time is limited. This research adds to the sparse research reported in the literature comparing the cost of meals with varying degrees of convenience, accounting for time. Further research on the price elasticity of healthy meals and takeaways is required.
Acknowledgements
Financial support: This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors. Conflict of interest: None. Authorship: S.M. led the study conception and design, data collection, analysis and writing of the manuscript. S.V. contributed to the study conception and design, data analysis and critically revised the manuscript. P.X. contributed to the data acquisition and data analysis and critically revised the manuscript. A.L. contributed to the study conception and critically revised the manuscript. B.S. contributed to the study conception and design and critically revised the manuscript. Ethics of human subject participation: Not applicable.
Supplementary material
To view supplementary material for this article, please visit https://doi.org/10.1017/S1368980017000805