Grains on the brain: A survey of dog owner purchasing habits related to grain-free dry dog foods

Sydney Banton; Andrew Baynham; Júlia G Pezzali; Michael von Massow; Anna K Shoveller

doi:10.1371/journal.pone.0250806

. 2021 May 19;16(5):e0250806. doi: 10.1371/journal.pone.0250806

Grains on the brain: A survey of dog owner purchasing habits related to grain-free dry dog foods

Sydney Banton ¹, Andrew Baynham ², Júlia G Pezzali ¹, Michael von Massow ², Anna K Shoveller ^1,^*

Editor: Juan J Loor³

PMCID: PMC8133411 PMID: 34010328

Abstract

Grain-free pet food options abound in the pet food market today, representing more than 40% of available dry dog foods in the United States. There is currently a dearth of information about the factors that contribute to a dog owner’s choice of a grain-free dry dog food and if those factors are similar among countries. Therefore, the primary objective of the current survey was to identify the variables that are predictive of a dog owner’s choice of a grain-free dry food across North America (Canada and the United States) and Europe (France, the United Kingdom and Germany). The survey consisted of 69 questions, took less than 15 minutes to complete and was distributed virtually via Qualtrics (Qualtrics XM, Utah, USA). A total of 3,298 responses were collected, equally distributed between countries. Multinomial logistic regression was performed in SPSS Statistics (Version 26, IBM Corp, North Castle, New York, USA). Male respondents, people from France, people who ranked the importance of ingredients in a pet food in the lower quartiles and people who do not rotate their dog’s diet to provide variety were less likely to select ‘no grain’ when choosing a pet food. In contrast, people who believe that their dog has a food allergy, follow more than 5 specific dietary routines in their own diet, do not try to include grains in their own diet, get their information about pet food from online resources or pet store staff and look for specific claims on pet food (such as ‘no fillers’), were all more likely to select ‘no grain’ when choosing a pet food. This survey provides insight into the similarities and differences in decision making among dog owners in North America and Europe and should be considered when exploring the effects of grain-free dog foods on canine health and well-being.

Introduction

Many dog owners today treat their dogs like family members and consequently, they consider a wide range of important variables, other than price, when selecting a diet for their dog. A survey conducted by Boya et al. [1] reported that those who anthropomorphize their dogs to a greater extent, rate price as less important when buying dog food than when buying food for their own consumption. The same survey also found that the anthropomorphization of dogs was associated with placing equal value on food quality and ‘holistic’, ‘natural’, and ‘organic’ claims of dog food and food for their own consumption [1].

In the human food industry, segments of the population have placed higher value on ‘clean-label’ foods, as they are often perceived as healthier by consumers across the globe [2]. ‘Clean-label’ foods often carry claims, such as, ‘natural’, ‘organic’, ‘free from artificial ingredients’, ‘no GMOs’, etc. [2]. As a result, many of these human food trends have been translated to the pet food industry, giving consumers the option to choose pet foods that carry similar claims, such as, ‘no fillers’, ‘no artificial flavours’, ‘gluten-free’, ‘organic’, and perhaps the most popular, ‘grain-free’. Grain-free diets have been on the market for over a decade and make up more than 40% of dry dog foods available in the United States (USA) [3]. Dog foods that carry the claim, ‘grain-free’, have seen significant growth in the last decade, with sales increasing by 221% from 2012 to 2016 in the USA [4]. In a recent survey of USA pet owners, approximately 50% of respondents perceived that grain-free diets were healthier for their pet [4]. In 2015, when grain-free was nearing its peak in the USA, almost 30% of market share was composed of grain-free pet food, compared to only 15% in the United Kingdom (UK) and 1% in France in the same year [5]. This suggests that the desire for grain-free pet foods differs between countries and is valued more by North American consumers.

Although many factors, such as ingredients, price, health/nutrition and freshness are important to dog owners when choosing a dog food [1, 6, 7], all of these reports surveyed residents of the USA only. Currently, little is known about consumer habits related to the purchase of grain-free dog diets in North America and other countries around the world. In order to understand differences in consumer attitudes towards grain-free dog food, it is necessary to first understand differences in consumer demographics, diet and purchasing habits that may contribute to that. Thus, the primary objective of the current survey was to identify the factors that are predictive of a dog owner’s choice of a grain-free dry dog food across Europe (France, Germany and the UK) and North America (USA and Canada). The second objective was to compare dog owner and dietary practices among countries. We hypothesized that dog owners in North America would be more likely to select grain-free compared to dog owners in Europe and that respondents who followed a more strict routine in their own diet would be more likely to select grain-free for their dog.

Materials and methods

Survey

The survey, titled “Pet Food Consumer Habit Survey,” consisted of 69 questions, took under 15 minutes to complete and was available in English, French and German languages. The first section consisted of questions related to the respondents’ demographics and diet. The diet section consisted of questions that assessed the respondent’s dietary routine, if they try to eat grains as part of a healthy diet and if they try to eat more whole grains than refined grains. The second section consisted of questions related to the respondent’s dog, including demographics and diet. The diet section included information on how much and how often they feed their dog via a series of true/false questions and how often they feed their dog treats, table scraps, fruits/vegetables and other items. There were also minor sections related to the dog’s exercise routine and allergies, and minor sections related to the dog owner’s pet food purchasing habits and factors important to dog ownership. In the allergy section, respondents were asked if their dog has ever experienced itchy skin, hair loss, smelly skin, smelly stool or soft stool and if they have ever addressed these symptoms via diet change, if they believe their dog has a food allergy, and if their dog has been diagnosed by a veterinarian with a food allergy. The purchasing habits section consisted of questions related to where they get their information about dog food from, where they purchase their dog food from and what factors are most important to them when choosing their dog’s food via a ‘constant sum’ question, where they were given 100 points to allocate to price, ingredient list, claims made on the bag, if your pet likes it, brand and sustainability. There were three separate questions that asked what features of pet food respondents look for when selecting a pet food. One of these questions contained ‘no’ statements, including ‘no grain.’ The term ‘no grain’ was intentionally used instead of the term ‘grain-free’ in order to avoid introducing any negative bias towards the term ‘grain-free’ due to the 2018 FDA report [8] suggesting a link between grain-free diets and the development of canine dilated cardiomyopathy (DCM). The last section consisted of four questions where the respondent was asked to rate how important nutrition, exercise, veterinary care and socialization are to their dog’s overall health.

Data collection

This study and survey was approved by the University of Guelph’s Research Ethics Board (REB 19-12-026). Written consent was obtained from each respondent and all data was analyzed anonymously. Before completing the survey, respondents were instructed to read the “Consent to Participate in Research” form that outlined the purpose of the study, the researchers, procedures, risks and benefits, participation and withdrawal, their rights, confidentiality and inclusion/exclusion criteria. The survey was anonymous and no identifying information was collected.

The survey was distributed across Canada, the US, Germany, France and the UK via an online survey platform, Qualtrics (Qualtrics XM, Utah, USA). Qualtrics recruited participants via email and this was independent of the researchers. To be eligible for the survey, respondents had to be over the age of 18 years, have at least one dog that was consuming a non-prescription dry dog food (kibble) and be the primary person responsible for selecting the dog food. In addition, respondents could not be a veterinarian, dog breeder or work in the pet food industry. If the respondent had more than one dog that was consuming a non-prescription dry food, they were instructed to answer the questions according to the older dog. In order to make sure respondents were paying attention and not answering the survey questions randomly, two attention check questions were asked where the respondent was told to select a specific response. If that response was not selected, the survey terminated, and the data were not used. Those who were eligible and completed the entire survey were compensated by Qualtrics with cash, vouchers, gift-cards, or points towards rewards, etc.

In order to confirm the survey was functioning as designed, a soft launch was conducted on June 2, 2020, where 59 responses were collected from Canada, the USA and the UK, 49 responses from France and 50 responses from Germany (n = 158). The full launch occurred on June 3, 2020 and all data were collected within 24 hours of launching. The male: female quota was set to 50: 50 and the study aimed at recruiting approximately 645 responses from each country.

Statistical analysis

All survey data were analyzed using SPSS Statistics (Version 26, IBM Corp, North Castle, New York, USA). Categorical data were analyzed from all countries together, as frequencies, and a two-sided test of equality for column proportions was done to compare proportions between countries. All pairwise comparisons were adjusted using the Bonferroni correction. Multinomial logistic regression was used to determine which variables were significant in a consumer’s choice in selecting ‘no grain’ as a criterion for dog food. The demographic variables that were of interest to the authors were the respondent’s age, sex, country and type of dog, therefore, they were included in each model. A preliminary review of the data revealed that more than 50% of respondents in the USA were over the age of 65 years, therefore, the interaction term, USA*age 65 plus was also included in all models in order to see if the skewed age data had any effect on the outcome of the model. Age and country were input as dummy variables, where 18 to 24 years and the UK were always treated as the base case. The dependent variable was the binary outcome: selected ‘no grain’ or did not select ‘no grain’. Statistical significance was declared at P≤0.05.

Results

Demographics of respondents

A total of 657 responses were collected from Canada, 657 from the USA, 656 from Germany, 661 from France and 667 from the UK (n = 3,298). Demographics of the respondents are outlined in Table 1. Results pertinent to the current discussion are outlined in this report and the complete survey can be found in Supporting Information.

Table 1. Demographic data, including sex, age, country of residence, income, education and children living in the household from all respondents (n = 3, 298).

Demographics	n	%
Sex
Male	1648	50.0
Female	1650	50.0
Age
18–24 years	137	4.2
25–34 years	524	15.9
35–44 years	680	20.6
45–54 years	670	20.3
55–64 years	666	20.2
65 years or older	621	18.8
Country
France	661	20.0
Germany	656	19.9
United Kingdom	667	20.2
Canada	657	19.9
United States	657	19.9
Income
0–24,999	580	17.6
25,000–49,999	1030	31.2
50,000–74,999	671	20.3
75,000–99,999	457	13.9
100,000–124,999	216	6.5
125,000–149,999	152	4.6
150,000 or more	192	5.8
Education
Less than high school diploma	215	6.5
High school diploma	902	27.3
College degree	732	22.2
Bachelor’s degree	781	23.7
Master’s degree	486	14.7
Professional degree	99	3.0
PhD	83	2.5
Live with children between 0–17 years
Yes	1224	37.1
No	2074	62.9

Open in a new tab

The age distribution within countries was similar between France, Germany, the UK and Canada. However, in the USA more than 50% of the respondents were 65 years or older. To account for this, an interaction term of being from the USA and being 65 years or older was added to each multinomial regression model (see statistical analysis section). The sex distribution was 50% male and 50% female within each country.

Diet of respondents

Overall, 52.4% of respondents reported that they do not follow any dietary routines themselves, while 47.6% of respondents reported that they follow at least one of the following diet regimens; low sugar or sugar-free diet (12.9%), low fat/ low calorie diet (11.5%), high protein diet (9.6%), high fiber diet (8.9%), no processed foods (i.e. no additives, minimal ingredient diet; 8.6%), organic (8.6%), low sodium diet (7.6%), vegetarian (7.1%), low or carbohydrate-free diet (5.7%), dairy-free (2.9%), diabetic diet (2.8%), gluten-free or Celiac diet (2.6%), other (2.3%), vegan (2.2%), raw (1.8%), grain-free (1.7%), Halal (1.2%), ketogenic (1.1%), Kosher (0.9%), low fiber diet (0.8%) and renal or kidney diet (0.5%). People from France (60.2%) selected none of the above more than people from Germany (40.4%) or the USA (51.9%; P<0.0001) but were not different than the UK (55.6%) or Canada (53.7%; Fig 1). In contrast, more people from Germany (7.5%) selected that they follow 5 or more dietary routines compared to France (2.1%) or the UK (1.9%; P<0.0001), but were not different from Canada (5.3%) or the USA (5.9%; Fig 1).

Fig 1 — Proportions in the same category that do not share a common letter are significantly different at P<0.05.

When asked to assign a value of 1 to 10 (0 being strongly disagree, 5 being neutral and 10 being strongly agree) to the statement, “I try to eat grains as part of a healthy diet,” 68.4% of respondents assigned a value between 6–10 (considered to agree), 20.8% of respondents assigned a value of 5 (neutral), and 10.9% of respondents assigned a value between 0–4 (considered to disagree). More people from France (20.7%) disagreed with the statement than any other country (Germany: 7.2%, UK: 9.1%, Canada: 8.4%, USA: 8.8%; P<0.0001; Fig 2). Similarly, 62.6% of respondents assigned a value between 6–10 (agree) to the statement, “I try to eat more whole grains than refined grains,” 19.9% of respondents assigned a 5 (neutral) and 17.5% of respondents assigned a value between 0–4 (disagree).

Fig 2 — Proportions in the same category that do not share a common letter are significantly different at P<0.05.

Demographics of dogs

The sex, age, breed and size of the dogs owned by the respondents are reported in Table 2. A majority of respondents owned only 1 dog (78.1%), however, if the respondent owned more than one dog, they were asked to respond to the questions according to the oldest dog in their household.

Table 2. Demographic data, including sex, age, breed and size for all dogs (n = 3, 298).

Variable	n	%
Sex
Male	1969	59.7
Female	1329	40.3
Age
0–2 years	324	9.8
2–5 years	1097	33.3
5–8 years	892	27.1
8–11 years	563	17.1
Older than 11 years	422	12.8
Breed
Purebred	1780	54.0
Mixed breed	1518	46.0
Size
X-small (up to 8 lbs)	150	4.5
Small (8–22 lbs)	1036	31.4
Medium (22–55 lbs)	1299	39.4
Large (55–100 lbs)	741	22.5
Giant (over 100 lbs)	72	2.2

Open in a new tab

Diet of dogs

When asked about what type and how frequently other foods are offered to the dog, only 2.3% of respondents reported that no other foods are offered to their dog (daily or occasionally), while 97.7% of respondents selected at least one of the following; daily dog treats (48.1%), occasional dog treats (48.3%), daily table scraps (9.3%), occasional table scraps (38.1%), daily fruits and vegetables (11.7%), occasional fruits and vegetables (26.2%), daily other foods (12.3%), and occasional other foods (13.7%). People from France were seemingly more strict with offering other foods, as 55.8% of respondents reported giving no other foods to their dog on a daily basis, which was significantly higher than the other countries (Germany: 37.3%, UK: 37.0%, Canada: 38.2%, USA: 33.3%; P<0.0001; Fig 3).

Fig 3 — Proportions in the same category that do not share a common letter are significantly different at P<0.05.

A majority of respondents answered ‘true’ to the following true/false questions, respectively, “I provide a fixed amount of food to my dog every day,” (86.9%), “I restrict my dog’s food intake to control weight,” (58.4%), “I follow the feeding guidelines on my dog’s pet food bag,” (60.7%), “I feed my dog an amount of food based on my veterinarian’s recommendation,” (56.5%), and “I purposely rotate my dog’s dry food to provide variety” (55.7%). More people in Europe (France: 64.4%, Germany: 64.9%, UK: 61.6%) responded that they rotate their dogs diet to provide variety compared to people in North America (Canada: 45.6%, USA: 41.6%; P<0.0001, Fig 4). Other differences among countries in their response to the above questions are outlined in Fig 4.

Fig 4 — Proportions in the same category that do not share a common letter are significantly different at P<0.05.

Incidence of allergy symptoms in dogs

When asked about any issues related to allergies, 40.4% of respondents reported that their dog had experienced at least one of the following; itchy skin (19.8%), soft stool (19.1%), hair loss (8.6%), smelly stool (6.7%), and smelly skin (4.9%). The number of allergy symptoms reported did not differ among countries when no symptoms, one symptom and 2 or more symptoms were compared. Of the respondents who selected one or more of the symptoms listed above, 66.4% of them reported that they have tried changing the dog’s diet to address the issue. Across all respondents, only 10.9% reported feeding their dog a specific diet because they believe that their dog has a food allergy, compared to 7.8% of people who reported feeding their dog a specific diet because their dog has been diagnosed by a veterinarian with a food allergy. People from Canada (13.2%) reported feeding their dog a specific diet because they believe their dog has an allergy more often than people from the UK (8.2%; P = 0.044), but were not different from France (11.8%), Germany (9.9%) or the USA (11.0%).

Pet food purchasing habits

A majority of respondents selected that they get their information about pet food from their veterinarian (57.6%), followed by online resources (40.9%), pet store staff (29.3%), friends and family (27.9%), online/television advertisements (11.1%), breeders (11.0%), other resources (10.3%), animal shelters/rescues (5.5%) and from pet food companies (5.0%). More people from France (67.3%) indicated that they get their information about pet food from a veterinarian than any other country (Germany: 53.7%, UK: 48.3%, Canada: 59.7%, USA: 59.2%; P<0.0001; Fig 5). In contrast, German respondents received their information about pet food from various sources and reported getting information from online resources (55.9%) and from pet store staff (49.8%) more often than any other country (France: 29.8%, UK: 41.5%, Canada: 41.2%, USA: 35.8%; P<0.0001; France: 15.6%, UK: 30.3%, Canada: 32.4%, USA: 18.4%; P<0.0001, respectively, Fig 5). Pet food was most commonly purchased from a pet specialty store (39.9%), followed by a grocery store (31.4%), online (16.3%), other (8.1%), and veterinary clinics (4.3%).

Fig 5 — Proportions in the same category that do not share a common letter are significantly different at P<0.05.

There were three separate questions that assessed what attributes respondents consider when choosing a pet food. Claims related to targeted nutrition resonated with a majority of respondents (84.5%) who selected at least one of the following options: ‘age specific nutrition’ (47.6%), ‘size specific nutrition’ (47.2%), ‘digestive care’ (26.9%), ‘dental care’ (22.3%), ‘joint care’ (20.7%), ‘breed specific nutrition’ (19.1%), ‘weight control’ (17.5%) and ‘sensitive skin/stomach’ (10.7%). More respondents in France (91.7%) selected at least one of the above options compared to respondents in the UK (83.4%), Canada (83.9%) or the USA (75.5%; P<0.0001; Fig 6) but did not differ from Germany (88.0%).

Fig 6 — Proportions in the same category that do not share a common letter are significantly different at P<0.05.

With respect to protein related claims, 89.8% of respondents selected that they look for at least one of the following options: ‘poultry’ (68.5%), ‘beef’ (55.8%), ‘fish’ (32.6%), ‘protein #1 ingredient’ (24.1%), ‘organic/natural’ (21.8%), ‘pork’ (21.2%), ‘limited ingredient diet’ (13.6%), ‘exotic proteins’ (9.7%) and ‘vegetarian/vegan diet’ (3.2%). Similar to the above question, more people in France (92.7%) selected at least one of the above options compared to people in Germany (88.1%) or the USA (86.3%; P<0.0001; Fig 6) but did not differ among respondents in the UK (92.1%) or Canada (89.5%).

Compared to the attributes listed above, the claims containing ‘no’ statements held the lowest value among respondents, but still a majority (61.8%) of respondents selected at least one of the following options: ‘no artificial colours or flavours’ (38.6%), ‘no fillers’ (30.1%), ‘no by-products’ (28.3%), ‘no grain’ (21.3%), ‘no corn’ (16%), ‘no wheat’ (15.7%), ‘no soy’ (14.5%), ‘no pulse ingredients’ (9.1%) and ‘gluten-free’ (7.7%). In contrast to the above questions, fewer respondents in France (44.9%) selected at least one of ‘no’ statements compared to the other four countries (P<0.0001; Fig 6). However, there was no difference among Germany (68.8%), Canada (69.5%) or the USA (65.8%; Fig 6). When just the claim, ‘no grain’, was compared between countries, again, respondents from France (8.0%) demonstrated the lowest tendency to select ‘no grain’ as an attribute that influences the purchase of pet food compared to the other countries (Germany, 30.0%, UK, 19.6%, Canada, 21.8%, USA, 27.4%; P<0.0001, Fig 7).

Fig 7 — Proportions that do not share a common letter are significantly different at P<0.05.

When asked to allocate 100 points between six attributes of pet food, in order of importance, owners indicated that the perception that their dog enjoys consuming the food was most important (29.4 ± 21.4 points), followed by the ingredient list (26.1 ± 21.4 points), the price (19.5 ± 18.9 points), the brand of dog food (11.6 ± 14.8 points), claims made on the bag (7.3 ± 9.5 points) and sustainability (6.1 ± 8.7 points). When country-specific rankings were evaluated, respondents in France (22.7 ± 0.8 points) allocated more points to the price of the dog food compared to Germany (17.4 ± 0.7 points), the USA (17.5 ± 0.8 points) or the UK (17.4 ± 0.7 points; P<0.05), while respondents in Germany (28.9 ± 0.8 points), Canada (28.6 ± 0.9 points) and the USA (29.0 ± 1.0 points) allocated more points to the ingredient list compared to France (21.5 ± 0.7 points) or the UK (22.7 ± 0.7 points; P<0.05; Fig 8). Other differences among countries in importance of specific attributes are outlined in Fig 8.

Fig 8 — Proportions in the same category that do not share a common letter are significantly different at P<0.05.

Factors important to dog ownership

When asked to rank the following items on a scale of 0 to 10, 97% of respondents indicated that nutrition was an important component of dog ownership (assigned a score between 6 to 10), with 40.7% of them indicating that nutrition was extremely important (score of 10). Similar scores were assigned to exercise and regular veterinary care/vaccines, where 94.7% and 90.2% of respondents assigned a score between 6 to 10, respectively, and 39.8% and 44.3% of respondents assigned a score of 10, respectively. Finally, although scores were still high, socialization was shown to be the least important factor for dog ownership. Across all respondents, 85.8% assigned a score between 6 to 10, with 28.9% of them selecting 10, indicating that socialization is extremely important. While all five countries assigned similar scores, when compared, there were some differences, outlined in Fig 9.

Fig 9 — Proportions in the same category that do not share a common letter are significantly different at P<0.05.

Multinomial logistic regression

The models presented below were developed with the goal of understanding what variables are predictive of a consumer’s choice of a grain-free dog diet. Seven models were developed using questions from the survey related to allergies in dogs, human and dog diet, pet food purchasing habits, veterinary care/advice, dog ownership, demographics and exercise/feeding practices. The models with a McFadden Pseudo R-Square value above 0.10 are discussed in detail and results are reported in tables. The other models are discussed briefly and tables can be found in Supplementary Data.

The Allergy Model (χ²(23) = 383.24, P<0.0001, Table 3) had a McFadden Pseudo R-Square value of 0.11. Male respondents were less likely to select ‘no grain’ than female respondents (P = 0.015). People from France were less likely to select ‘no grain’ than people from the UK (P<0.0001). People who selected ‘yes’ to feeding their dog a specific diet because it has been diagnosed by a veterinarian with a food allergy were less likely to select ‘no grain’ than those who selected ‘no’ (P = 0.041). In contrast, people who selected that their dog has experienced two or more allergy symptoms were 1.3 times more likely to select ‘no grain’ than those who selected no allergy symptoms (P = 0.047). People who selected ‘yes’ to feeding their dog a specific diet because they believe it has a food allergy were 4 times more likely to select ‘no grain’ than those who selected ‘no’ (P<0.0001). When each country was looked at separately for this question, people from Germany and France were not significantly different from UK. People from the USA were still more likely to select ‘no grain’ if they said ‘yes’ to feeding a specific diet because they believe their dog has an allergy, but significantly less likely than those from the UK (calculated Odds Ratio: 1.79, P = 0.047). People who selected that they look for claims that are often linked to allergies in dogs: sensitive skin/stomach formulas, exotic proteins or limited ingredient diets when choosing a pet food were 1.7, 1.9 and 3 times more likely, respectively, to select ‘no grain’ than those who do not look for these options (P<0.0001). Age, type of dog and USA by 65 years plus interaction was not significant (P>0.05).

Table 3. Multinomial logistic regression parameter estimates for the Allergy Model.

Variable	β¹	Std. Error	P-Value	OR²	95% CI³
Variable	β¹	Std. Error	P-Value	OR²	Lower Bound	Upper Bound
Age
25 to 34 years	-0.040	0.261	0.880	0.961	0.576	1.605
35 to 44 years	-0.100	0.256	0.694	0.904	0.548	1.493
45 to 54 years	0.128	0.255	0.616	1.137	0.689	1.875
55 to 64 years	0.181	0.256	0.479	1.199	0.726	1.979
65 years or older	0.178	0.282	0.528	1.195	0.687	2.078
18 to 24 years	.	.	.	.	.	.
Sex
Male	-0.224	0.092	0.015	0.799	0.667	0.958
Female	.	.	.	.	.	.
Country
Germany	0.281	0.150	0.062	1.324	0.986	1.778
France	-1.042	0.199	<0.0001	0.353	0.239	0.521
USA	0.113	0.189	0.550	1.119	0.773	1.620
Canada	0.016	0.157	0.916	1.017	0.748	1.382
UK	.	.	.	.	.	.
Type of dog
Purebred	-0.026	0.092	0.780	0.975	0.813	1.168
Mixed breed	.	.	.	.	.	.
Allergy Symptoms
One	0.090	0.112	0.422	1.094	0.879	1.362
Two or more	0.278	0.140	0.047	1.321	1.003	1.740
None	.	.	.	.	.	.
Do you feed your dog a specific diet because you believe your dog has a food allergy?
Yes	1.389	0.327	<0.0001	4.010	2.110	7.618
No	.	.	.	.	.	.
Do you feed your dog a specific diet because your dog has been diagnosed by a veterinarian with a food allergy?
Yes	-0.458	0.224	0.041	0.633	0.408	0.982
No	.	.	.	.	.	.
When choosing a pet food, I look for …	0.526	0.139	<0.0001	1.692	1.288	2.222
Sensitive skin/stomach	0.526	0.139	<0.0001	1.692	1.288	2.222
When choosing a pet food, I look for …	1.105	0.117	<0.0001	3.019	2.401	3.795
Limited ingredient diet	1.105	0.117	<0.0001	3.019	2.401	3.795
When choosing a pet food, I look for …	0.644	0.138	<0.0001	1.904	1.452	2.496
Exotic protein	0.644	0.138	<0.0001	1.904	1.452	2.496
*Age 65 plusUSA**	0.425	0.253	0.093	-	-	-
*GermanyDo you feed your dog a specific diet because you believe your dog has a food allergy?**	-0.101	0.425	0.813	-	-	-
*FranceDo you feed your dog a specific diet because you believe your dog has a food allergy?**	-0.320	0.466	0.493	-	-	-
*USADo you feed your dog a specific diet because you believe your dog has a food allergy?**	-0.805	0.406	0.047	-	-	-
*CanadaDo you feed your dog a specific diet because you believe your dog has a food allergy?**	-0.730	0.401	0.069	-	-	-

Open in a new tab

¹ Estimated multinomial logistic regression coefficient

² Odds Ratio or exponentiation of the coefficient (β)

³ 95% Confidence Interval of the Odds Ratio

McFadden Pseudo R-Square = 0.11

Dependent variable categories, 1 = selected ‘no grain’, 0 = did not select ‘no grain’

The Diet Model (χ²(29) = 587.81, P<0.0001, Table 4) had a McFadden Pseudo R-Square value of 0.17. Male respondents were less likely to select ‘no grain’ than female respondents (P = 0.046). People from France (P<0.0001) and people from Canada (P = 0.024) were less likely to select ‘no grain’ than people from the UK. When allotting points to importance of the ingredient list on a pet food label (more points being more important), people that were in the first, second and third quartiles were less likely to select ‘no grain’ than those in the fourth quartile (P<0.0001). People from Germany were 1.4 times more likely to select ‘no grain’ than those from the UK (P = 0.019). People who selected that they follow 5 or more dietary routines for themselves were 2.3 times more likely to select ‘no grain’ than those that do not follow any dietary routines (P<0.0001). People who selected between 0–4 (disagree) or 5 (neutral) in response to, “I try to eat grains as part of a healthy diet” were 1.6 and 1.4 times more likely to select ‘no grain’ than those that selected between 6–10 (P = 0.003). People who selected between 1–3 or 4 or more options in the protein ingredient pet food question were 1.6 and 1.8 times more likely to select ‘no grain’ than those who selected no options (P = 0.019 and P = 0.010, respectively). People who selected that they look for ‘no fillers’ or ‘no by-products’ when choosing a pet food were 2.6 and 1.6 times more likely to select ‘no grain,’ respectively, than those who do not look for ‘no fillers’ or ‘no by-products’ (P<0.0001). People who give their dog 2 or more different foods (treats, table scraps, fruits/veggies, other foods) on a daily basis were 1.4 times more likely to select ‘no grain’ than those that do not give their dog any other foods on a daily basis (P = 0.006). Age, type of dog, number of choices selected in the targeted nutrition pet food question, importance of nutrition and the USA by 65 plus interaction were not significant (P>0.05). In addition, when the 20 human diet options were the only variables in the model, only those that selected that they follow a grain-free, vegetarian, vegan, ketogenic or no processed foods diet were more likely to select ‘no grain’ when choosing a pet food (P<0.05).

Table 4. Multinomial logistic regression parameter estimates for the Diet Model.

Variable	β¹	Std. Error	P-Value	OR²	95% CI³
Variable	β¹	Std. Error	P-Value	OR²	Lower Bound	Upper Bound
Age
25 to 34 years	0.032	0.267	0.904	1.033	0.612	1.743
35 to 44 years	-0.181	0.263	0.490	0.834	0.499	1.396
45 to 54 years	-0.049	0.262	0.851	0.952	0.570	1.590
55 to 64 years	-0.050	0.264	0.850	0.951	0.567	1.595
65 years or older	-0.077	0.290	0.790	0.926	0.524	1.634
18 to 24 years	.	.	.	.	.	.
Sex
Male	-0.191	0.096	0.046	0.826	0.685	0.997
Female	.	.	.	.	.	.
Country
Germany	0.337	0.144	0.019	1.401	1.057	1.856
France	-0.976	0.187	<0.0001	0.377	0.261	0.543
USA	-0.148	0.189	0.433	0.862	0.595	1.249
Canada	-0.344	0.152	0.024	0.709	0.526	0.955
UK	.	.	.	.	.	.
Type of dog
Purebred	0.005	0.096	0.958	1.005	0.833	1.213
Mixed breed	.	.	.	.	.	.
Which one of the following best reflects your own dietary routine (select all that apply)
Selected 1 option	-0.021	0.119	0.859	0.979	0.775	1.237
Selected 2–4 options	0.131	0.127	0.304	1.140	0.888	1.462
Selected 5 or more options	0.825	0.202	<0.0001	2.282	1.535	3.394
Selected no options	.	.	.	.	.	.
I try to eat grains as part of a healthy diet
0–4 (Disagree)	0.498	0.169	0.003	1.646	1.183	2.290
5 (Neutral)	0.364	0.124	0.003	1.439	1.129	1.835
6–10 (Agree)	.	.	.	.	.	.
When choosing a pet food, I look for (size, breed, age specific nutrition, digestive, weight, dental, joint, sensitive skin/stomach care)
1–3 options selected	0.054	0.149	0.718	1.056	0.788	1.415
4 or more options selected	0.212	0.176	0.227	1.236	0.876	1.744
No options selected	.	.	.	.	.	.
When choosing a pet food, I look for (poultry, beef, fish, pork, exotic protein, organic/natural, protein #1 ingredient, vegetarian/vegan, limited ingredient diet)
1–3 options selected	0.485	0.207	0.019	1.625	1.082	2.440
4 or more options selected	0.565	0.220	0.010	1.759	1.143	2.707
No options selected	.	.	.	.	.	.
When choosing a pet food, I look for …
No fillers	0.964	0.111	<0.0001	2.621	2.108	3.260
No by-products	0.440	0.111	<0.0001	1.553	1.248	1.932
Other food items (dog treats, table scraps, fruits/veggies, other) given on a daily basis
One option selected	0.054	0.109	0.619	1.056	0.853	1.307
Two or more options selected	0.364	0.131	0.006	1.439	1.113	1.862
No options selected	.	.	.	.	.	.
How important are the following items when purchasing pet food …
Ingredients
First Quartile	-1.326	0.150	<0.0001	0.266	0.198	0.357
Second Quartile	-0.827	0.136	<0.0001	0.437	0.335	0.571
Third Quartile	-0.433	0.124	<0.0001	0.649	0.509	0.827
Fourth Quartile	.	.	.	.	.	.
Importance of nutrition in terms of your dog’s overall health
2–5 (Not important/neutral)	0.273	0.348	0.433	1.314	0.664	2.598
6–10 (Important)	.	.	.	.	.	.
*Age 65 plusUSA**	0.300	0.266	0.259	-	-	-

Open in a new tab

¹ Estimated multinomial logistic regression coefficient

² Odds Ratio or exponentiation of the coefficient (β)

³ 95% Confidence Interval of the Odds Ratio

McFadden Pseudo R-Square = 0.17

Dependent variable categories, 1 = selected ‘no grain’, 0 = did not select ‘no grain’

The Purchasing Habits Model (χ²(26) = 438.75, P<0.0001, Table 5) had a McFadden Pseudo R-Square value of 0.13. Again, people from France were less likely to select ‘no grain’ than people from the UK (P<0.0001) and male respondents were less likely to select ‘no grain’ than female respondents (P = 0.002). People who purposely rotate their dog’s dry food to provide variety were less likely to select ‘no grain’ (P = 0.021). Similar to the Diet Model, people who ranked the importance of ingredients on a pet food label in the first, second and third quartiles were less likely to select ‘no grain’ than those who ranked ingredients in the fourth quartile (P<0.0001). People who selected that they get their information about pet food from online resources and/or pet store staff were 1.6 and 1.3 times more likely to select ‘no grain’ than those that do not get their information about pet food online or from pet store staff (P<0.0001 and P = 0.006, respectively). People who purchase their dog food from a veterinary clinic, a pet specialty store and online were 2.2, 1.6 and 2.4 times more likely to select ‘no grain’ than people who purchase their pet food from a grocery store (P = 0.002, P<0.0001 and P<0.0001, respectively). Age, type of dog, importance of price when selecting pet food and the interaction term, USA by 65 plus were not significant (P>0.05).

Table 5. Multinomial logistic regression parameter estimates for the Purchasing Habits Model.

Variable	β¹	Std. Error	P-Value	OR²	95% CI³
Variable	β¹	Std. Error	P-Value	OR²	Lower Bound	Upper Bound
Age
25 to 34 years	0.039	0.263	0.883	1.039	0.621	1.739
35 to 44 years	-0.148	0.258	0.565	0.862	0.520	1.429
45 to 54 years	0.084	0.257	0.745	1.087	0.657	1.799
55 to 64 years	0.160	0.259	0.537	1.174	0.706	1.952
65 years or older	0.151	0.287	0.598	1.163	0.663	2.041
18 to 24 years	.	.	.	.	.	.
Sex
Male	-0.283	0.093	0.002	0.753	0.628	0.903
Female	.	.	.	.	.	.
Country
Germany	0.214	0.141	0.130	1.239	0.939	1.633
France	-1.009	0.185	<0.0001	0.365	0.254	0.524
USA	0.051	0.182	0.779	1.052	0.736	1.505
Canada	-0.044	0.149	0.768	0.957	0.714	1.282
UK	.	.	.	.	.	.
Type of Dog
Purebred	-0.061	0.093	0.514	0.941	0.784	1.129
Mixed breed	.	.	.	.	.	.
I purposely rotate my dog’s dry food to provide variety
True	-0.224	0.097	0.021	0.799	0.661	0.967
False	.	.	.	.	.	.
Where do you get your information about dog food from?
Veterinarian	-0.116	0.097	0.232	0.891	0.737	1.077
Online	0.452	0.096	<0.0001	1.571	1.301	1.897
Pet store staff	0.290	0.105	0.006	1.337	1.087	1.643
Where do you purchase your pet food from?
Vet clinic	0.787	0.250	0.002	2.197	1.347	3.583
Pet specialty store	0.452	0.123	<0.0001	1.572	1.235	2.001
Online	0.863	0.144	<0.0001	2.371	1.789	3.143
Other	0.304	0.215	0.156	1.355	0.890	2.064
Grocery store	.	.	.	.	.	.
How important are the following items when purchasing pet food … Price
First Quartile	0.024	0.145	0.866	1.025	0.772	1.361
Second Quartile	0.167	0.139	0.228	1.182	0.901	1.551
Third Quartile	0.090	0.144	0.533	1.094	0.825	1.450
Fourth Quartile	.	.	.	.	.	.
How important are the following items when purchasing pet food … Ingredients
First Quartile	-1.552	0.147	<0.0001	0.212	0.159	0.283
Second Quartile	-0.876	0.138	<0.0001	0.416	0.318	0.545
Third Quartile	-0.439	0.123	<0.0001	0.645	0.507	0.820
Fourth Quartile	.	.	.	.	.	.
*Age 65 plusUSA**	0.370	0.258	0.151	-	-	-

Open in a new tab

¹ Estimated multinomial logistic regression coefficient

² Odds Ratio or exponentiation of the coefficient (β)

³ 95% Confidence Interval of the Odds Ratio

McFadden Pseudo R-Square = 0.13

Dependent variable categories, 1 = selected ‘no grain’, 0 = did not select ‘no grain’

The other models all had a McFadden Pseudo R-Square value less than 0.10 (The Vet Model: χ²(21) = 169.02, P<0.0001, McFadden Pseudo R-Square value of 0.049, The Dog Ownership Model: χ²(29) = 226.90, P<0.0001, McFadden Pseudo R-Square value of 0.066, The Demographics Model: χ²(33) = 208.24, P<0.0001, McFadden Pseudo R-Square value of 0.061, the Exercise/Feeding Model: χ²(29) = 204.07, P<0.0001, McFadden Pseudo R-Square value of 0.060, S1–S4 Tables). In general, across these models, male respondents were less likely to select ‘no grain’ than female respondents, people from France were less likely to select ‘no grain’ than people from the UK and people from Germany were more likely to select ‘no grain’ than people from the UK (P<0.05). Interestingly, from the Vet Model, only getting your information about pet food from the vet was not significant and from the Demographics Model, education level and income were also not significant (P>0.05).

Discussion

To the authors’ knowledge, this is the first large scale, published survey investigating consumer habits related to purchasing grain-free dry dog food across the UK, Germany, France, the USA and Canada. In addition, this survey was entirely recruited via Qualtrics, therefore there was no bias in social media recruitment or location bias. As expected, many factors contribute to a dog owner’s choice of grain-free dry dog food. Perhaps the most noteworthy from this study, the incidence of allergies, owner purchasing habits and the dietary routine of owners appear to have the highest explanatory power for choice of grain-free dry dog food, according to results from multinomial logistic regression.

From the Allergy Model, this study found that people who reported that their dog experienced two or more allergy symptoms were more likely to select ‘no grain’ than those that reported no symptoms. The most common food allergens in dogs with diagnosed food allergies are beef and dairy, accounting for more than 60% of food allergy cases reviewed by Verlinden et al. [9], followed by wheat, egg and chicken, while the least common food allergens in dogs are soy, corn, fish, pork, and rice (each accounting for less than 6% of all food allergies diagnosed) [10]. While animal protein allergies are far more common than grain allergies in dogs [9, 10], the results from the current study suggest that consumers may believe that a grain-free diet can address allergy symptoms in their dog. In addition, selecting a claim that is often advertised as being beneficial to the treatment of allergies in dogs, such as, ‘limited ingredient diet’, ‘sensitive skin/stomach’ and ‘exotic protein’ was also positively associated with selecting a grain-free diet.

Another finding in the present study was that people who believe their dog has a food allergy were four times more likely to select ‘no grain’ when choosing a pet food. In contrast, owners whose dog has been diagnosed by a veterinarian with a food allergy were less likely to select ‘no grain’. Veterinarians are trained to diagnose a food allergy via an elimination trial where a hydrolyzed protein diet is fed exclusively for 6–8 weeks. However, this can be difficult for a dog owner to comply with since only one specific diet and no other foods can be fed for the duration of the trial. Therefore, it is possible that dog owners are turning to grain-free as their own remedy to a suspected food allergy. One flaw in interpreting this allergy data is the fact that the survey excluded people feeding a prescription diet to their dog and likely, if a dog is being fed a specific diet because their vet diagnosed an allergy, it would be a prescription diet. Another possibility is that, although 7.8% of people reported that they feed their dog a specific diet because their vet diagnosed a food allergy, these may be prescription diets, as we did not define what a prescription diet was to the respondent or ask what diet they are currently feeding, in order to confirm.

We hypothesized that those who follow a stricter dietary routine themselves would be more likely to impose dietary restrictions on their dogs and select a grain-free diet, since grain-free diets are often perceived as healthier by pet food consumers [4]. Modelling data from Kumcu & Woolverton [11] demonstrated that people who purchase ‘premium’ food for themselves (defined as USDA certified organic food) are more likely to purchase ‘premium’ pet food (defined according to brand and keywords such as natural, organic, etc). In a survey by Dodd et al. [12], people who reported feeding their dog a plant-based diet were all either vegan or vegetarian themselves. These results align with our findings where people who follow 5 or more dietary routines themselves and people who do not try to eat grains as part of a healthy diet were more likely to select ‘no grain’ when choosing a pet food. Furthermore, those who follow a grain-free, vegetarian, vegan, ketogenic or no processed foods dietary routine themselves were all more likely to select ‘no grain’ when choosing a pet food. It is unclear why people who are vegetarian or vegan would be more likely to select a grain free diet for their dog, however, this highlights the idea that many variables, not only owner dietary habits, contribute to selection of a dog food. It is necessary to point out that the current survey only included those feeding a dry food (kibble) diet and these data may be quite different if owners who feed a different dietary format are included.

People who allocate more points to the importance of the ingredients in a pet food and people who look for the terms, ‘no fillers’ and ‘no by-products’ were more likely to select ‘no grain’ when choosing a pet food, supporting the statement by LaFlamme et al. [13] that grains are often considered ‘fillers’ by pet owners. In addition, Conway & Saker [14] reported that respondents in a survey often selected that grain-free diets were “diets free of fillers and by-products”. This is not surprising, given that ingredients in pet food are consistently ranked within the top three most important factors by consumers [7, 14] and were ranked as the second most important factor in the current survey. In addition, respondents were willing to pay more for natural and organic ingredients compared to conventional ingredients in a discrete choice experiment conducted by Simonsen et al. [6]. This suggests that those who value organic and natural ingredients share a similar mind set with those who select grain-free diets. Furthermore, this suggests that those who believe ingredients in a dog food are more important than price may be more susceptible to advertising claims on dog foods, such as ‘no fillers’ and ‘no by-products’. Although it seems that ingredients are one of the most important factors driving pet food purchasing choices, animals require nutrients, not ingredients [15]. Taking the above findings together, it is possible that dog owners are using ingredients as a proxy for nutrition and if they perceive the ingredients as healthy, they are more likely to feed it to their dog. This may also help explain the finding that those who selected that they give two or more other foods on a daily basis were more likely to select ‘no grain’ when choosing a pet food. It is possible that these people are supplementing their dog’s diet with other foods they perceive as healthy, similar to how those that feed a raw diet will often supplement with oils, fruits, vegetables, etc. [16]. However, since we did not ask more specific questions about adding additional foods to a dry dog food, we cannot draw any concrete conclusions related to the potential effect of these additional foods.

The above conclusions highlight an important disconnect between human and pet nutrition. While the majority of European and North American food-based dietary guidelines recommend consuming whole grains as part of a healthy diet [17], grains appear to be perceived as unhealthy in a dog’s diet by pet owners. There is ample evidence that consuming whole grains and fibre can lower human risk for cardiovascular disease [18] and no scientific evidence that grains are detrimental to a dog’s health. This suggests that the beliefs dog owners have about grains may be largely based on marketing strategies, which is consistent with the findings that selecting claims, ‘no fillers’ and ‘no by-products’ are associated with selecting ‘no grain.’ However, allowing consumers to drive ingredient composition of pet food could potentially lead to nutrient imbalances if amino acid balance is not considered [15, 19]. Recently, regular consumption of a grain-free diet has been associated with the development of canine dilated cardiomyopathy (DCM) [8]. Grain-free diets replace grain ingredients with pulse ingredients. Although high in protein, the amino acid profile of a pulse is different than that of a grain. Thus, careful consideration should be taken with regards to the amino acid composition and balance of the diet when grains are completely replaced by legumes and/or pulses. Furthermore, the Purchasing Habits Model found that those who rotate their dogs dry food to provide variety are less likely to select ‘no grain’, suggesting that grain-free diets may be selected as a static and regimented dietary routine. A recent clinical study by Kaplan et al. [20] revealed that in 24 Golden Retrievers diagnosed with DCM, the median length of time consuming the same grain-fee diet was 814 days, and the maximum, 3,558 days. In human nutrition, a diverse diet is encouraged to ensure breadth of nutrient intake to maintain health. It has been well established that those in developing countries, particularly women, who lack a diverse diet are more likely to develop micronutrient deficiencies [21]. Furthermore, using data from more than 1,000 French adults, Bianchi et al. [22] showed that adults with a higher diet diversity score were more likely to have a higher nutritional adequacy score. Although the importance of a diverse diet is established in human nutrition, there is a dearth of data with regards to the metabolic and health impact of a static feeding regimen in the domestic dog. Thus, future studies should investigate the effects of static feeding of a grain-free diet on micronutrient imbalances and its impact on the development of nutritionally-mediated DCM.

Evaluation of consumer purchasing behaviours across countries and between sexes is a strength of this study. Across all models, men were less likely to select ‘no grain’ when choosing a pet food. Several human food purchasing studies around the world have found that women are more likely to purchase organic foods than men [23–25], thus it was expected that this trend would translate to purchasing pet food. Similarly, respondents from France were always less likely to select ‘no grain’. More than half of the respondents from France did not follow any dietary routine themselves, they were less likely to get information about pet food from online resources and pet store staff and they rated the ingredient list on pet foods as less important, all of which were factors that were inversely associated with selecting a grain-free dog food. In contrast, people from France rated price as more important and selected targeted nutrition options (i.e.–size, breed, age specific nutrition, etc.) more often, suggesting that other factors may be more important to them than whether or not the diet contains grains.

Divergence in opinions about grain-free dog foods in Europe are mirrored in sales data. Although four of the top ten European pet food companies have plants that are located in Germany or France and produce grain-free diet options [26], the number of pet foods with the claims ‘no additives/preservatives’ and ‘all natural product’ were 4 and 7 times higher, respectively, in Germany compared to France from 2015 to 2018 [27]. In this survey, respondents from Germany were more likely to select ‘no grain’ in four of seven regression models. In terms of human food, France and Germany are the top two largest markets for organic food consumption, following the USA [28], so perhaps the desire to impart your own dietary choices on your dog is not the same among countries. Further to that idea, more respondents in France disagreed with the statement, “I try to eat grains as part of a healthy diet” compared to any other country, despite the food-based dietary guidelines in France recommending consumption of starchy foods, especially whole grains, at each meal [29]. This implies that people from France do not impart their own dietary choices on their pets, compared to the other countries. In addition, it is possible that countries in Europe view their pet dogs differently, however, the human-animal bond in different countries around the world needs further investigation.

Conclusion

Overall, while it is clear that many factors contribute to a consumer selecting a grain-free dry dog food, in the current study, some factors such as: sex, perception of allergies, a consumer’s own dietary routine, different pet food information resources, the importance of the ingredient list and looking for specific claims, may be more predictive than others. Furthermore, the choice of grain-free is not the same between continents or even between countries, as people from France were always less likely to select ‘no grain’, while people from Germany were often more likely to select ‘no grain’. These factors should be considered when exploring the population level effects of grain-free dog foods on canine health and well-being. In addition, the data from the present survey lends insights into effective avenues for the scientific community to facilitate knowledge transfer between themselves and consumers.

Supporting information

S1 Table. Multinomial logistic regression parameter estimates for the Vet Model.

¹ Estimated multinomial logistic regression coefficient. ² Odds Ratio or exponentiation of the coefficient (β). ³ 95% Confidence Interval of the Odds Ratio. McFadden Pseudo R-Square = 0.050. Dependent variable categories, 1 = selected ‘no grain’, 0 = did not select ‘no grain’.

(DOCX)

Click here for additional data file.^{(18.1KB, docx)}

S2 Table. Multinomial logistic regression parameter estimates for the Dog Ownership Model.

¹ Estimated multinomial logistic regression coefficient. ² Odds Ratio or exponentiation of the coefficient (β). ³ 95% Confidence Interval of the Odds Ratio. McFadden Pseudo R-Square = 0.066. Dependent variable categories, 1 = selected ‘no grain’, 0 = did not select ‘no grain’.

(DOCX)

Click here for additional data file.^{(19.9KB, docx)}

S3 Table. Multinomial logistic regression parameter estimates for the Demographics Model.

¹ Estimated multinomial logistic regression coefficient. ² Odds Ratio or exponentiation of the coefficient (β). ³ 95% Confidence Interval of the Odds Ratio. McFadden Pseudo R-Square = 0.061. Dependent variable categories, 1 = selected ‘no grain’, 0 = did not select ‘no grain’.

(DOCX)

Click here for additional data file.^{(22.3KB, docx)}

S4 Table. Multinomial logistic regression parameter estimates for the Exercise/Feeding Model.

¹ Estimated multinomial logistic regression coefficient. ² Odds Ratio or exponentiation of the coefficient (β). ³ 95% Confidence Interval of the Odds Ratio. McFadden Pseudo R-Square = 0.060. Dependent variable categories, 1 = selected ‘no grain’, 0 = did not select ‘no grain’.

(DOCX)

Click here for additional data file.^{(20.8KB, docx)}

S1 File

(PDF)

Click here for additional data file.^{(292.6KB, pdf)}

Acknowledgments

We thank Dr. Chris Marinangeli (Pulse Canada) and Dr. Adronie Verbrugghe (Ontario Veterinary College) for their comments on the manuscript.

Data Availability

All raw survey data files are available from the Agri-environmental Research Data Repository (DOI: https://doi.org/10.5683/SP2/JZUOMW).

Funding Statement

This research was funded by Rolf C. Hagen, Inc. (Grant 053974, website: http://caen.hagen.com) and awarded to AKS. The funders had no role in study design, data collection and analysis, or decision to publish. The funders did review the manuscript before it was submitted for publication.

References

1.Boya UO, Dotson MJ, Hyatt EM. A comparison of dog food choice criteria across dog owner segments: An exploratory study. Int J Consum Stud. 2015;39(1):74–82. [Google Scholar]
2.Asioli D, Aschemann-Witzel J, Caputo V, Vecchio R, Annunziata A, Næs T, et al. Making sense of the “clean label” trends: A review of consumer food choice behavior and discussion of industry implications. Food Res Int. 2017;99(1);58–71. 10.1016/j.foodres.2017.07.022 [DOI] [PubMed] [Google Scholar]
3.Plantz, B. Grain-free dry dog, cat food no longer a niche market. [Internet]. 2017. Available from: https://www.petfoodindustry.com/articles/6576-grain-free-dry-dog-cat-food-no-longer-a-niche-market.
4.Facts Packaged. U.S. Pet Market Outlook, 2018–2019 [Internet]. 2017. 277 p. Report No.: LA15754465. Available from: https://www.packagedfacts.com/Pet-Outlook-11819832/. [Google Scholar]
5.Growth from Knowledge. Natural and grain-free pet food: Serious contenders. Nuremburg: Growth from Knowledge; 2016. June. 2 p. [Google Scholar]
6.Simonsen JE, Faskeno GM, Lillywhite JM. The value-added dog food market: Do dog owners prefer natural or organic dog foods? J Agric Sci. 2014;6(6):86–97. [Google Scholar]
7.Schleicher M, Cash SB, Freeman LM. Determinants of pet food purchasing decisions. Can Vet J. 2019;60(6):644–50. [PMC free article] [PubMed] [Google Scholar]
8.FDA. FDA Investigating potential connection between diet and cases of canine heart disease [Internet]. 2018. Available from: https://www.fda.gov/animal-veterinary/cvm-updates/fda-investigating-potential-connection-between-diet-and-cases-canine-heart-disease
9.Verlinden A, Hesta M, Millet S, Janssens GPJ. Food allergy in dogs and cats: A review. Crit Rev Food Sci Nutr. 2006;46(3):259–73. 10.1080/10408390591001117 [DOI] [PubMed] [Google Scholar]
10.Mueller RS, Olivry T, Prélaud P. Critically appraised topic on adverse food reactions of companion animals (2): Common food allergen sources in dogs and cats. BMC Vet Res. 2016;12(9):1–4. 10.1186/s12917-016-0633-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Kumcu A, Woolverton AE. Feeding Fido: Changing consumer food preferences bring pets to the table. J Food Prod Mark. 2015;21(2):213–30. [Google Scholar]
12.Dodd SAS, Cave NJ, Adolphe JL, Shoveller AK, Verbrugghe A. Plant-based (vegan) diets for pets: A survey of pet owner attitudes and feeding practices. PLoS One. 2019;14(1):e0210806. 10.1371/journal.pone.0210806 [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Laflamme D, Izquierdo O, Eirmann L, Binder S. Myths and misperceptions about ingredients used in commercial pet foods. Vet Clin North Am Small Anim Pract. 2014;44(4):689–98. 10.1016/j.cvsm.2014.03.002 [DOI] [PubMed] [Google Scholar]
14.Conway DMP, Saker KE. Consumer attitude toward the environmental sustainability of grain-free pet foods. Front Vet Sci. 2018;5(170):1–8. 10.3389/fvets.2018.00170 [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Mansilla WD, Marinangeli CPF, Ekenstedt KJ, Larsen JA, Aldrich G, Columbus DA, et al. Special topic: The association between pulse ingredients and canine dilated cardiomyopathy: Addressing the knowledge gaps before establishing causation. J Anim Sci. 2019;97(3):983–97. 10.1093/jas/sky488 [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Morelli G, Bastianello S, Catellani P, Ricci R. Raw meat-based diets for dogs: Survey of owners’ motivations, attitudes and practices. BMC Vet Res. 2019;15(74):1–10. 10.1186/s12917-019-1824-x [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Montagnese C, Santarpia L, Buonifacio M, Nardelli A, Caldara AR, Silvestri E, et al. European food-based dietary guidelines: A comparison and update. Nutrition. 2015;31(7–8):908–915. 10.1016/j.nut.2015.01.002 [DOI] [PubMed] [Google Scholar]
18.Ruxton C, Derbyshire E. The health benefits of whole grains and fibre. Nutr Food Sci. 2014;44(6):492–519. [Google Scholar]
19.Freeman LM, Stern JA, Fries R, Adin DB, Rush JE. Diet-associated dilated cardiomyopathy in dogs: What do we know? J Am Vet Med. 2018;253(11):1390–4. 10.2460/javma.253.11.1390 [DOI] [PubMed] [Google Scholar]
20.Kaplan JL, Stern JA, Fascetti AJ, Larsen JA, Skolnik H, Peddle GD, et al. Taurine deficiency and dilated cardiomyopathy in golden retrievers fed commercial diets. PLoS One. 2018;13(12):e0210233. 10.1371/journal.pone.0210233 [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Arimond M, Wiesmann D, Becquey E, Carriquiry A, Daniels MC, Deitchler M, et al. Simple food group diversity indicators predict micronutrient adequacy of women’s diets in 5 diverse, resource-poor settings. J Nutr. 2010;140(11):2059S–2069S. 10.3945/jn.110.123414 [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Bianchi CM, Egnell M, Huneau JF, Mariotti F. Plant protein intake and dietary diversity are independently associated with nutrient adequacy in French adults. J Nutr. 2016;146(11):2351–60. 10.3945/jn.116.236869 [DOI] [PubMed] [Google Scholar]
23.Nasir VA, Karakaya F. Consumer segments in organic foods market. J Consum Mark. 2014;31(4):263–77. [Google Scholar]
24.Onyango BM, Hallman WK, Bellows AC. Purchasing organic food in US food systems: A study of attitudes and practice. Br Food J. 2007;109(5):399–411. [Google Scholar]
25.Hamzaoui-Essoussi L, Zahaf M. Canadian organic food consumers’ profile and their willingness to pay premium prices. J Int Food Agribus Mark. 2012;24(1):1–21. [Google Scholar]
26.Wall, T. 10 top European pet food companies 2019. [Internet]. 2020. Available from: https://www.petfoodindustry.com/articles/9555-top-european-pet-food-companies-2019.
27.Thériault F. Sector trend analysis–Pet food trends in France [Internet]. Ottawa; 2018. Available from: https://www.agr.gc.ca/eng/international-trade/market-intelligence/reports/sector-trend-analysis-pet-food-trends-in-france/?id=1538496137531&fbclid=IwAR1ALuBZIPXeAZV9A2Vq26Ko3U-T-0yT-eZcq_M1ZnVEKBxtO_W43zTGCRg#e
28.FiBL. The ten countries with the largest markets for organic food 2016. [Internet]. 2018. Available from: https://statistics.fibl.org/europe/key-indicators-europe.html.
29.High Council for Public Health. Statement related to the update of the French Nutrition and Health Programme’s dietary guidelines for adults for the period 2017–2021. Paris: High Council for Public Health; 2017. Feb. 7 p. [Google Scholar]

PLoS One. doi: 10.1371/journal.pone.0250806.r001

Decision Letter 0

Juan J Loor

29 Jan 2021

PONE-D-20-39049

Grains on the brain: A survey of dog owner purchasing habits related to grain-free diets

PLOS ONE

Dear Dr. Shoveller,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Mar 15 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Juan J Loor

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2.Thank you for stating the following financial disclosure:

"This research was funded by Rolf C. Hagen, Inc. (Grant 053974, website: http://ca-en.hagen.com) and awarded to AKS. Warren House and Tiana Owens, employees of Rolf C. Hagen, Inc. both reviewed the manuscript."

Please state what role the funders took in the study. If the funders had no role, please state: "The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript."

If this statement is not correct you must amend it as needed.

Please include this amended Role of Funder statement in your cover letter; we will change the online submission form on your behalf.

3.Thank you for stating the following in the Competing Interests section:

"I have read the journal's policy and the authors of this manuscript have the following competing interests: Sydney Banton was funded by Rolf C. Hagen, Inc. to complete the research. "

Please confirm that this does not alter your adherence to all PLOS ONE policies on sharing data and materials, by including the following statement: "This does not alter our adherence to PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests). If there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include your updated Competing Interests statement in your cover letter; we will change the online submission form on your behalf.

Please know it is PLOS ONE policy for corresponding authors to declare, on behalf of all authors, all potential competing interests for the purposes of transparency. PLOS defines a competing interest as anything that interferes with, or could reasonably be perceived as interfering with, the full and objective presentation, peer review, editorial decision-making, or publication of research or non-research articles submitted to one of the journals. Competing interests can be financial or non-financial, professional, or personal. Competing interests can arise in relationship to an organization or another person. Please follow this link to our website for more details on competing interests: http://journals.plos.org/plosone/s/competing-interests

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

Reviewer #3: No

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: PONE-D-20-39049 Grains on the brain: A survey of dog owner purchasing habits related to grain-free diets

I found the manuscript interesting and well organized. I think the work has value and will be useful to many in the pet food area. My only concern is that the paper is quite long; however, I have no concrete suggestions for shortening. My comments were quite limited.

Line

348 ‘about pet food from a various sources and’ awkward? Perhaps delete ‘a’

382 Suggest ‘or the USA’

466 Suggest ‘from the UK.’

678 ‘pet foods with the claims ‘no additives/preservatives’ and ‘all natural product’ were 4 and 7 times higher’ Suggest ‘the number of pet foods…’

Reviewer #2: The authors have described the results of dog owner survey focused on feeding habits of people living in a few countries in Europe and North America. The topic is of interest and a lot of data were generated, but feel major revisions are needed. Comments are provided below.

General comments:

1. The title and introduction are focused on grain-free diets (most likely to drive reader interest), but much of the data have nothing to do with this topic. I think a much more appropriate and useful approach is the focus on differing responses across countries, owner sex, etc. and to point out how much pet owners rely on fad diet, trends, and marketing strategies – many that have no scientific basis – to make their purchases. This is interesting, but a huge problem that the industry has. This dataset shows a lot of these problems in the human and pet food industries. While the authors report the data obtained, they have not tried too hard to point out that the lack of education, science, and facts that contribute to these choices is big problem. Only 1-2 sentences in the discussion highlight this problem (lines 625-628; lines 638-643). I would encourage the authors to take on that challenge.

2. If the authors feel they need to focus on grain-free, the first 7-8 figures should be moved to supplementary files. If that is the topic, it just distracts from the main point of the paper. It provides a foundation and background of the survey participants, but most of that data has nothing to do with that topic. The text can refer to the figures, but provide that information.

3. The manuscript is much too long. If data are provided in tables and figures, every data point is not necessary in the text of the Results section. The main differences should be provided in the text with reference to figures and tables.

4. The manuscript contains too much slang. Phrases like we found, pet owners get, etc. should be replaced with technical terms.

5. It is stated that because an independent firm provided the survey that it was unbiased. However, there is no evidence to back up this comment. How were people recruited? What groups were targeted? Unless this can be described, this statement is incorrect. Also, a huge bias is that only kibble feeders completed the survey. I think much of the data provided would be quite different for those feeding other dietary formats. This needs to be highlighted more prominently in the abstract, limitations section and conclusion.

6. In many countries, the consumption of whole grains is promoted as part of a healthy diet in humans. The benefits of consuming whole grains, which is science-based, is important information that should be highlighted and referenced. The disconnect on facts like this between human and pet nutrition is confusing and frustrating. If the authors do not provide science-based information like this, the data may actually feed the negative, ingredient-based trends that we are seeing in the industry today.

7. Figures must be reordered so that they are in order as presented. Figure 3 is the main problem.

Specific comments:

1. Title is inaccurate as written. It is suggested to remove the emphasis on grain-free, as it is only part of the dataset. If the focus remains on grain-free, than much of the results should be minimized and moved to supplemental files because it distracts from this point.

2. Abstract: the fact that the survey was only completed by kibble feeders is an important point that should be included. The responses may be drastically different for those feeding other diet formats.

3. Line 54-67 and 701-704: is the “natural” and “organic” trend really linked with the “grain-free” trend? The natural and organic trends are based on how plants are grown, how animals are fed and raised, and how ingredients are processed regardless of type. Grain-free is completely based on ingredient source and type.

4. Line 125 and other places: data are plural so replace “was” with “were”

5. Results section should be shortened dramatically. Exact data numbers are not needed in the figures, tables, and text. The text should provide the main differences and referring to tables and figures – not repeating them.

6. Line 196, 199 and other places: use “or” instead of “and”

7. Line 206-218: shorten significantly. Many of the specific points (lines 211-213) are inconsequential to the big picture and can be removed.

8. Line 272-292: shorten significantly.

9. Line 368: how are protein ingredients defined here? Many ingredients contain protein and AA but are not considered protein-rich ingredients and are not included for that purpose.

10. Discussion: can shorten by one third.

11. Line 553-554: see comments about bias above. Nearly everything has bias. It is not a problem for this study, but is important to highlight what they are and take them into account when interpreting the data.

12. 566-567: need to revise. Allergies to grain are based on their protein content. What are you meaning to say here?

13. Line 588-590: see comments above about this. Why does science demonstrate that whole grain consumption in humans is positive and part of the healthy dietary guidelines provided by governmental agencies, yet grains are perceived to be a problem with pet owners? This is a very important disconnect that needs to be addressed.

14. Line 595-597: this suggests that this group is the “fad diet” group, a group that probably tries a lot of strategies based on marketing rather than the boring, science-based strategy of eating a balanced diet, regular exercise, and low quantities of negative behaviors (drinking, smoking). This problem should be called out.

15. Line 603: the term “since” refers to time. It should be replaced with a more appropriate term.

16. Line 611-614: This again highlights one of the major problems with pet food consumers today. Completely stuck on ingredients rather than scientific facts. I would encourage the authors to not only present these data, but discuss why that type of thinking is problematic.

17. Line 625-628 and line 638-643: important points that deserve more attention.

18. Line 643-654: delete. This supposed problem has been ingredient based from the beginning and lacking basic science principles. This section continues this problematic narrative.

19. Line 684-686: what do French human dietary recommendations say about grains? This information is necessary in order to properly interpret these responses.

20. Line 703: natural and grain-free trends are quite different. Why are the authors trying to link them together?

21. Line 709-716: a stronger conclusion is needed, stressing the problems with ingredient-based marketing and purchasing trends. If researchers don’t highlight these problems, who will? Also, this dataset is completely based on opinions of kibble diet feeders. Data from those feeding other formats may be quite different.

22. Figure 7: spell out veterinarian

Reviewer #3: The manuscript reports a survey based on the classification of grain free food for companion animals. The criticism is that the term "grain free" is not defined for the owners and has not scinetific basis, is only a marketing matter. In other words, the report is useful for a feed company and for marketing purpose but does not add anything for the scientific community.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2021 May 19;16(5):e0250806. doi: 10.1371/journal.pone.0250806.r002

Author response to Decision Letter 0

22 Feb 2021

Response to reviewer #1

Thank you for the comprehensive review. We believe that addressing your queries significantly improved our manuscript. We specifically worked to shorten the results and discussion section. We also made the suggested word or phrase changes outlined in your feedback.

Response to reviewer #2

General comments:

Response to 1-3:

Thank you for your thoughtful comments. We understand your point of view regarding fad diets, trends and differing approaches across countries; however, our a priori hypothesis and intent was directly related to grain-free diets due to the concern in North America that feeding these diets is associated with dilated cardiomyopathy. Furthermore, we used the consumer data in the multinomial models where many factors are associated with the choice of a grain-free diet, including the owner’s own dietary regimen, perception of allergies in dogs and even demographic variables such as sex and country. Consequently, it is necessary to report the data that focus on the basic understanding of the population as they are the foundations behind the statistical models. Our hypothesis states that we suspect those with more strict dietary regimens would be more inclined to select ‘no grain’ and that is what we found. Consumer research dictates that it is necessary to present the initial country differences in the variables that were used in the 3 models so that the reader can understand the differences between countries before the variables are incorporated into the models. In addition, we have added a line to the introduction to make this more clear: “In order to understand differences in consumer attitudes towards grain-free dog food, it is necessary to first understand differences in consumer demographics, diet and purchasing habits that may contribute to that.” (Line 71-74 in new manuscript).

However, we understand that presenting all data may divert the focus of the manuscript, and thus, we have significantly decreased the data in the results section that is unrelated to the three models presented (mainly feeding frequency, dog’s body weight and dog’s exercise) and removed two figures (feeding frequency and dog’s exercise) that were unrelated to the variables in the models. We have also removed some of the text in the Results and referred to the figures instead as well as removed some of the numbers in the text that are already highlighted in the tables (specifically the ORs and CI in the regressions).

4. The manuscript contains too much slang. Phrases like we found, pet owners get, etc. should be replaced with technical terms.

We have made phrases like this more formal. For example, line 40 changed from “this survey gives us insight” to “this survey provides insight,” line 497 removed “we found that.”

We did not state that the survey was completely unbiased, we highlighted the fact that a third party did all of the recruitment via email which does decrease the location/distribution bias. Most of the published manuscripts that use surveys to understand any aspect regarding the pet food industry and/or pet food owners use social media platforms to recruit participants (Dodd et al., 2019, Schleicher et al., 2019, Morelli et al., 2019), which does not represent the general population. Qualtrics is a recognized survey software that is widely used in other areas of research to understand consumer habits.

We do appreciate your comment about this being biased towards pet owners who choose dry dog food. Seeing as we excluded respondents who feed other dietary formats, we have added “grain-free dry dog food” to the title and made an effort to point out more often in the text that these were respondents who feed dry dog food only.

Thank you for addressing this, it is a great discussion point. We agree that this is an important disconnect between human and pet food. We have made an effort to address this in the discussion section, Line 529-536 in new manuscript:

7. Figures must be reordered so that they are in order as presented. Figure 3 is the main problem.

Thank you for pointing this error out. We have gone through the figures and renumbered them to ensure they are ordered correctly.

Specific comments:

See our response to 1-3 above. Additionally we have added “dry dog food” to the title: “Grains on the brain: A survey of dog owner purchasing habits related to grain-free dry dog food.”

Thank you for highlighting this, it is an important aspect of the current study. We have now added “dry dog food” to the title and pointed out more explicitly that we are investigating how respondents select a grain-free dry dog food in the abstract and throughout the text. Instead of saying “grain-free diets” we have changed it to “grain-free dry dog food” throughout the text.

Thank you for this comment, we agree. We have removed the sentence in the introduction where we classify grain-free as ‘natural’ pet food. Additionally, we have changed the first sentence of the abstract from, “Natural pet food options abound in the pet food market…” to “Grain-free pet food options abound in the pet food market today, representing more than 40% of available dry dog foods in the United States.” The paragraph that contains line 701-704 has been removed in order to decrease the length of the discussion as it was focused on a result from the supplementary models. We have changed the sentence comparing natural pet food market share between the US and UK to a comparison of grain-free market share: “In 2015, when grain-free was nearing its peak in the USA, almost 30% of market share was composed of grain-free pet food, compared to only 15% in the United Kingdom and 1% in France in the same year [5].” (Line 64-66 in new manuscript). Consequently reference number 5 was changed to a report from Growth from Knowledge on the pet food sector in 2016.

It is also important to note, we never actually used or defined the term “grain-free” in the survey and instead used “no grain”. We did so intentionally to avoid any bias consumers may have when they hear the term “grain-free”. The term also carries some negative connotations in the USA after the FDA report so we intentionally did not define. The only term that appears as one of the options in the pet food questions is “no grain.” We have added this to the methods section:

“There were three separate questions that asked what features of pet food respondents look for when selecting a pet food. One of these questions contained ‘no’ statements, including ‘no grain.’ The term ‘no grain’ was intentionally used instead of the term ‘grain-free’ in order to avoid introducing any negative bias towards the term ‘grain-free’ due to the 2018 FDA report [8] suggesting a link between grain-free diets and the development of canine dilated cardiomyopathy.” (Line 103-108 in new manuscript)

4. Line 125 and other places: data are plural so replace “was” with “were”

Thank you, this has been corrected.

We have removed significant portions of the results section, mainly results that are not pertinent to the models like dog’s body weight, feeding frequency and exercise as well as the two figures that were associated with these results. Additionally, we have removed all ORs and CIs from the text and referred to the table where the data in presented for all regression models.

6. Line 196, 199 and other places: use “or” instead of “and”

Thank you, this has been corrected.

7. Line 206-218: shorten significantly. Many of the specific points (lines 211-213) are inconsequential to the big picture and can be removed.

This has been shortened. The specific country comparisons have been removed and only the main distinction between France and the other countries has been called out in the text.

8. Line 272-292: shorten significantly.

This has been shortened. All of the data related to feeding frequency has been removed.

9. Line 368: how are protein ingredients defined here? Many ingredients contain protein and AA but are not considered protein-rich ingredients and are not included for that purpose.

We did not define the term protein in this context, the question in the survey simply stated, “When choosing a pet food, I look for… (select all that apply).” There were 3 of these questions in the survey, so we assigned a broad term to each one (targeted nutrition related claims, protein related claims and ‘no’ statements) for the purposes of the results section only. This was done to inform the reader of the types of attributes that were listed in each question. We have also attached the full survey as part of supplemental material if the reader wanted to refer to the question in the context of the survey. We have also changed our wording to “protein related claims” in Line 291 of the new manuscript.

10. Discussion: can shorten by one third.

We have made an effort to shorten the discussion by removing the paragraph that was discussing results related to demographics because those results were presented in the supplemental material. The discussion is now only 6 pages.

We agree that everything has some form of bias. In this statement, we were trying to point out a strength of this survey being recruited by a third party rather than by social media, therefore eliminating much of the recruitment bias. We have also made more of an effort to address the dry food related bias throughout the text and how these results may be different if people who feed other dietary formats are included.

12. 566-567: need to revise. Allergies to grain are based on their protein content. What are you meaning to say here?

Thank you for pointing this out, we were referring to animal protein sources being more common allergens than grain sources and have added the words ‘animal protein’ to make this more clear. The new sentence reads: “While animal protein allergies are far more common than grain allergies in dogs [9, 10], the results from the current study suggest that consumers may believe that a grain-free diet can address allergy symptoms in their dog.” Line 469-471 in new manuscript.

Fantastic comment, thank you. This is a nice contrast between human and pet food that adds to the discussion. We have brought this point in and highlighted the disconnect between human and pet nutrition. We have also reworked this section of the discussion to address comments (16, 17 and 18) below as well. The following is the paragraph we added/re-worked:

“The above conclusions highlight an important disconnect between human and pet nutrition. While the majority of European and North American food-based dietary guidelines recommend consuming whole grains as part of a healthy diet [17], grains appear to be perceived as unhealthy in a dog’s diet by pet owners. There is ample evidence that consuming whole grains and fibre can lower human risk for cardiovascular disease [18] and no scientific evidence that grains are detrimental to a dog’s health. This suggests that the beliefs dog owners have about grains may be largely based on marketing strategies, which is consistent with the findings that selecting claims, ‘no fillers’ and ‘no by-products’ are associated with selecting ‘no grain.’ However, allowing consumers to drive ingredient composition of pet food could potentially lead to nutrient imbalances if amino acid balance is not considered [15, 19]. Recently, regular consumption of a grain-free diet has been associated with the development of canine dilated cardiomyopathy (DCM) [8]. Grain-free diets replace grain ingredients with pulse ingredients. Although high in protein, the amino acid profile of a pulse is different than that of a grain. Thus, careful consideration should be taken with regards to the amino acid composition and balance of the diet when grains are completely replaced by legumes and/or pulses. Furthermore, the Purchasing Habits Model found that those who rotate their dogs dry food to provide variety are less likely to select ‘no grain’, suggesting that grain-free diets may be selected as a static and regimented dietary routine. A recent clinical study by Kaplan et al. [20] revealed that in 24 Golden Retrievers diagnosed with DCM, the median length of time consuming the same grain-fee diet was 814 days, and the maximum, 3,558 days. In human nutrition, a diverse diet is encouraged to ensure breadth of nutrient intake to maintain health. It has been well established that those in developing countries, particularly women, who lack a diverse diet are more likely to develop micronutrient deficiencies [21]. Furthermore, using data from more than 1,000 French adults, Bianchi et al. [22] showed that adults with a higher diet diversity score were more likely to have a higher nutritional adequacy score. Although the importance of a diverse diet is established in human nutrition, there is a dearth of data with regards to the metabolic and health impact of a static feeding regimen in the domestic dog. Thus, future studies should investigate the effects of static feeding of a grain-free diet on micronutrient imbalances and its impact on the development of nutritionally-mediated DCM.” Line 529-557 in new manuscript.

This is a great observation and could be the case. However, in the question where we asked about diet of the respondent, there were several options that would not be considered ‘fad diets,’ such as, Kosher, Halal, low sodium, celiac, diabetic. Therefore, this group of respondents who selected they are following 5 or more dietary regimens could be following a number of diets for medical or religious reasons. Because we did not ask about fad diets specifically, we believe it may be difficult to infer this.

15. Line 603: the term “since” refers to time. It should be replaced with a more appropriate term.

This has been corrected.

See #13 above and additional changes in this area of the discussion. Line 529-557 in new manuscript.

17. Line 625-628 and line 638-643: important points that deserve more attention.

See #13 above and additional changes in this area of the discussion. Line 529-557 in new manuscript.

18. Line 643-654: delete. This supposed problem has been ingredient based from the beginning and lacking basic science principles. This section continues this problematic narrative.

See #13 above and additional changes in this area of the discussion. Line 529-557 in new manuscript.

19. Line 684-686: what do French human dietary recommendations say about grains? This information is necessary in order to properly interpret these responses.

This is a great observation and surprisingly to us, the French dietary guidelines also suggest consumption of whole grains with every meal so it is unclear why so many more French people would disagree with the statement, “I try to eat grains as part of a healthy diet.” We have added a line to address this (Line 578-581: Further to that idea, more respondents in France disagreed with the statement, “I try to eat grains as part of a healthy diet” compared to any other country, despite the food-based dietary guidelines in France recommending consumption of starchy foods, especially whole grains, at each meal [29].).

20. Line 703: natural and grain-free trends are quite different. Why are the authors trying to link them together?

The paragraph that contains line 703 has been removed in order to decrease the length of the discussion as it was focused on a result from the supplementary models.

While we do understand your point about these problems, we believe we have addressed these in the discussion now. We have identified in the conclusion that consumers that value ingredients are more likely to select a grain-free diet and that this should be considered when the health and well-being of dogs consuming a grain-free diet are examined. In addition, we have added a line to address the value of consumer surveys in relation to knowledge transfer: “In addition, the data from the present survey lends insights into effective avenues for the scientific community to facilitate knowledge transfer between themselves and consumers.” Line 595-597 in the new manuscript.

22. Figure 7: spell out veterinarian

This change has been made. In addition, we uploaded new figures with the letters used to assign significance changed to descending order.

Response to Reviewer #3

The manuscript reports a survey based on the classification of grain free food for companion animals. The criticism is that the term "grain free" is not defined for the owners and has not scientific basis, is only a marketing matter. In other words, the report is useful for a feed company and for marketing purpose but does not add anything for the scientific community.

Defining grain-free and providing various levels of information on a scientific basis would have been an interesting addition to this study. The goal of the study was to understand who is purchasing grain-free, where they purchasing it, why they are purchasing it and where they are getting their information about pet food from. If we were to provide a definition of grain-free, it may bias some of the responses, especially if the information was in contrast to what respondents currently understood as being “grain-free”. While the definition itself is not defined within the survey, the term “grain-free” was never actually used in the survey either. We intentionally used the term ‘no grain’ instead so that we did not introduce any bias. This is because, the term “grain-free” has gained some negative attention lately due to its possible connection to dilated cardiomyopathy (DCM) in dogs. We have also added a line to the methods to point this out:

The factors described within this paper, such as the demographic information and avenues which pet food purchasers get their information, dictates how types of pet foods are purchased. This has an advantage in lending insights and identifying effective avenues for the scientific community to facilitate knowledge outreach between themselves and the consumer. Having this understanding of consumer data is critical to the scientific community in understanding real world consumer applications.

Attachment

Submitted filename: Response to reviewers.docx

Click here for additional data file.^{(30.8KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0250806.r003

Decision Letter 1

Juan J Loor

14 Apr 2021

Grains on the brain: A survey of dog owner purchasing habits related to grain-free dry dog foods

PONE-D-20-39049R1

Dear Dr. Shoveller,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Juan J Loor

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #3: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #1: Yes

Reviewer #3: No

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #1: Yes

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #1: Yes

Reviewer #3: Yes

**********

6. Review Comments to the Author

Reviewer #1: (No Response)

Reviewer #3: Dear Authors,

concerning my previous comments, the title and the manuscript still contain "grain free" term. I read the comments of the other two reviewers and your answers. The manuscript was improved, but the main critcism I raised was not resolved. The aim of the paper is market oriented and useless from a scientific point of view.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #3: No

PLoS One. doi: 10.1371/journal.pone.0250806.r004

Acceptance letter

Juan J Loor

22 Apr 2021

PONE-D-20-39049R1

Grains on the brain: A survey of dog owner purchasing habits related to grain-free dry dog foods

Dear Dr. Shoveller:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Juan J Loor

Academic Editor

PLOS ONE

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

S1 Table. Multinomial logistic regression parameter estimates for the Vet Model.

(DOCX)

Click here for additional data file.^{(18.1KB, docx)}

S2 Table. Multinomial logistic regression parameter estimates for the Dog Ownership Model.

¹ Estimated multinomial logistic regression coefficient. ² Odds Ratio or exponentiation of the coefficient (β). ³ 95% Confidence Interval of the Odds Ratio. McFadden Pseudo R-Square = 0.066. Dependent variable categories, 1 = selected ‘no grain’, 0 = did not select ‘no grain’.

(DOCX)

Click here for additional data file.^{(19.9KB, docx)}

S3 Table. Multinomial logistic regression parameter estimates for the Demographics Model.

¹ Estimated multinomial logistic regression coefficient. ² Odds Ratio or exponentiation of the coefficient (β). ³ 95% Confidence Interval of the Odds Ratio. McFadden Pseudo R-Square = 0.061. Dependent variable categories, 1 = selected ‘no grain’, 0 = did not select ‘no grain’.

(DOCX)

Click here for additional data file.^{(22.3KB, docx)}

S4 Table. Multinomial logistic regression parameter estimates for the Exercise/Feeding Model.

¹ Estimated multinomial logistic regression coefficient. ² Odds Ratio or exponentiation of the coefficient (β). ³ 95% Confidence Interval of the Odds Ratio. McFadden Pseudo R-Square = 0.060. Dependent variable categories, 1 = selected ‘no grain’, 0 = did not select ‘no grain’.

(DOCX)

Click here for additional data file.^{(20.8KB, docx)}

S1 File

(PDF)

Click here for additional data file.^{(292.6KB, pdf)}

Attachment

Submitted filename: Response to reviewers.docx

Click here for additional data file.^{(30.8KB, docx)}

Data Availability Statement

All raw survey data files are available from the Agri-environmental Research Data Repository (DOI: https://doi.org/10.5683/SP2/JZUOMW).

[pone.0250806.ref001] 1.Boya UO, Dotson MJ, Hyatt EM. A comparison of dog food choice criteria across dog owner segments: An exploratory study. Int J Consum Stud. 2015;39(1):74–82. [Google Scholar]

[pone.0250806.ref002] 2.Asioli D, Aschemann-Witzel J, Caputo V, Vecchio R, Annunziata A, Næs T, et al. Making sense of the “clean label” trends: A review of consumer food choice behavior and discussion of industry implications. Food Res Int. 2017;99(1);58–71. 10.1016/j.foodres.2017.07.022 [DOI] [PubMed] [Google Scholar]

[pone.0250806.ref003] 3.Plantz, B. Grain-free dry dog, cat food no longer a niche market. [Internet]. 2017. Available from: https://www.petfoodindustry.com/articles/6576-grain-free-dry-dog-cat-food-no-longer-a-niche-market.

[pone.0250806.ref004] 4.Facts Packaged. U.S. Pet Market Outlook, 2018–2019 [Internet]. 2017. 277 p. Report No.: LA15754465. Available from: https://www.packagedfacts.com/Pet-Outlook-11819832/. [Google Scholar]

[pone.0250806.ref005] 5.Growth from Knowledge. Natural and grain-free pet food: Serious contenders. Nuremburg: Growth from Knowledge; 2016. June. 2 p. [Google Scholar]

[pone.0250806.ref006] 6.Simonsen JE, Faskeno GM, Lillywhite JM. The value-added dog food market: Do dog owners prefer natural or organic dog foods? J Agric Sci. 2014;6(6):86–97. [Google Scholar]

[pone.0250806.ref007] 7.Schleicher M, Cash SB, Freeman LM. Determinants of pet food purchasing decisions. Can Vet J. 2019;60(6):644–50. [PMC free article] [PubMed] [Google Scholar]

[pone.0250806.ref008] 8.FDA. FDA Investigating potential connection between diet and cases of canine heart disease [Internet]. 2018. Available from: https://www.fda.gov/animal-veterinary/cvm-updates/fda-investigating-potential-connection-between-diet-and-cases-canine-heart-disease

[pone.0250806.ref009] 9.Verlinden A, Hesta M, Millet S, Janssens GPJ. Food allergy in dogs and cats: A review. Crit Rev Food Sci Nutr. 2006;46(3):259–73. 10.1080/10408390591001117 [DOI] [PubMed] [Google Scholar]

[pone.0250806.ref010] 10.Mueller RS, Olivry T, Prélaud P. Critically appraised topic on adverse food reactions of companion animals (2): Common food allergen sources in dogs and cats. BMC Vet Res. 2016;12(9):1–4. 10.1186/s12917-016-0633-8 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0250806.ref011] 11.Kumcu A, Woolverton AE. Feeding Fido: Changing consumer food preferences bring pets to the table. J Food Prod Mark. 2015;21(2):213–30. [Google Scholar]

[pone.0250806.ref012] 12.Dodd SAS, Cave NJ, Adolphe JL, Shoveller AK, Verbrugghe A. Plant-based (vegan) diets for pets: A survey of pet owner attitudes and feeding practices. PLoS One. 2019;14(1):e0210806. 10.1371/journal.pone.0210806 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0250806.ref013] 13.Laflamme D, Izquierdo O, Eirmann L, Binder S. Myths and misperceptions about ingredients used in commercial pet foods. Vet Clin North Am Small Anim Pract. 2014;44(4):689–98. 10.1016/j.cvsm.2014.03.002 [DOI] [PubMed] [Google Scholar]

[pone.0250806.ref014] 14.Conway DMP, Saker KE. Consumer attitude toward the environmental sustainability of grain-free pet foods. Front Vet Sci. 2018;5(170):1–8. 10.3389/fvets.2018.00170 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0250806.ref015] 15.Mansilla WD, Marinangeli CPF, Ekenstedt KJ, Larsen JA, Aldrich G, Columbus DA, et al. Special topic: The association between pulse ingredients and canine dilated cardiomyopathy: Addressing the knowledge gaps before establishing causation. J Anim Sci. 2019;97(3):983–97. 10.1093/jas/sky488 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0250806.ref016] 16.Morelli G, Bastianello S, Catellani P, Ricci R. Raw meat-based diets for dogs: Survey of owners’ motivations, attitudes and practices. BMC Vet Res. 2019;15(74):1–10. 10.1186/s12917-019-1824-x [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0250806.ref017] 17.Montagnese C, Santarpia L, Buonifacio M, Nardelli A, Caldara AR, Silvestri E, et al. European food-based dietary guidelines: A comparison and update. Nutrition. 2015;31(7–8):908–915. 10.1016/j.nut.2015.01.002 [DOI] [PubMed] [Google Scholar]

[pone.0250806.ref018] 18.Ruxton C, Derbyshire E. The health benefits of whole grains and fibre. Nutr Food Sci. 2014;44(6):492–519. [Google Scholar]

[pone.0250806.ref019] 19.Freeman LM, Stern JA, Fries R, Adin DB, Rush JE. Diet-associated dilated cardiomyopathy in dogs: What do we know? J Am Vet Med. 2018;253(11):1390–4. 10.2460/javma.253.11.1390 [DOI] [PubMed] [Google Scholar]

[pone.0250806.ref020] 20.Kaplan JL, Stern JA, Fascetti AJ, Larsen JA, Skolnik H, Peddle GD, et al. Taurine deficiency and dilated cardiomyopathy in golden retrievers fed commercial diets. PLoS One. 2018;13(12):e0210233. 10.1371/journal.pone.0210233 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0250806.ref021] 21.Arimond M, Wiesmann D, Becquey E, Carriquiry A, Daniels MC, Deitchler M, et al. Simple food group diversity indicators predict micronutrient adequacy of women’s diets in 5 diverse, resource-poor settings. J Nutr. 2010;140(11):2059S–2069S. 10.3945/jn.110.123414 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0250806.ref022] 22.Bianchi CM, Egnell M, Huneau JF, Mariotti F. Plant protein intake and dietary diversity are independently associated with nutrient adequacy in French adults. J Nutr. 2016;146(11):2351–60. 10.3945/jn.116.236869 [DOI] [PubMed] [Google Scholar]

[pone.0250806.ref023] 23.Nasir VA, Karakaya F. Consumer segments in organic foods market. J Consum Mark. 2014;31(4):263–77. [Google Scholar]

[pone.0250806.ref024] 24.Onyango BM, Hallman WK, Bellows AC. Purchasing organic food in US food systems: A study of attitudes and practice. Br Food J. 2007;109(5):399–411. [Google Scholar]

[pone.0250806.ref025] 25.Hamzaoui-Essoussi L, Zahaf M. Canadian organic food consumers’ profile and their willingness to pay premium prices. J Int Food Agribus Mark. 2012;24(1):1–21. [Google Scholar]

[pone.0250806.ref026] 26.Wall, T. 10 top European pet food companies 2019. [Internet]. 2020. Available from: https://www.petfoodindustry.com/articles/9555-top-european-pet-food-companies-2019.

[pone.0250806.ref027] 27.Thériault F. Sector trend analysis–Pet food trends in France [Internet]. Ottawa; 2018. Available from: https://www.agr.gc.ca/eng/international-trade/market-intelligence/reports/sector-trend-analysis-pet-food-trends-in-france/?id=1538496137531&fbclid=IwAR1ALuBZIPXeAZV9A2Vq26Ko3U-T-0yT-eZcq_M1ZnVEKBxtO_W43zTGCRg#e

[pone.0250806.ref028] 28.FiBL. The ten countries with the largest markets for organic food 2016. [Internet]. 2018. Available from: https://statistics.fibl.org/europe/key-indicators-europe.html.

[pone.0250806.ref029] 29.High Council for Public Health. Statement related to the update of the French Nutrition and Health Programme’s dietary guidelines for adults for the period 2017–2021. Paris: High Council for Public Health; 2017. Feb. 7 p. [Google Scholar]

PERMALINK

Grains on the brain: A survey of dog owner purchasing habits related to grain-free dry dog foods

Sydney Banton

Andrew Baynham

Júlia G Pezzali

Michael von Massow

Anna K Shoveller

Roles

Abstract

Introduction

Materials and methods

Survey

Data collection

Statistical analysis

Results

Demographics of respondents

Table 1. Demographic data, including sex, age, country of residence, income, education and children living in the household from all respondents (n = 3, 298).

Diet of respondents

Fig 1. Number of dietary routines followed by respondents in each country (n = 3,298).

Fig 2. Score assigned to the statement, “I try to eat grains as part of a healthy diet” between respondents in each country (n = 3,298).

Demographics of dogs

Table 2. Demographic data, including sex, age, breed and size for all dogs (n = 3, 298).

Diet of dogs

Fig 3. Amount of other foods (respondents had the options of dog treats, table scraps, fruits/vegetables and other foods) given on a daily basis among respondents in each country (n = 3,298).

Fig 4. Percent of respondents that answered ‘true’ to the five feeding statements, among countries (n = 3,298).

Incidence of allergy symptoms in dogs

Pet food purchasing habits

Fig 5. Percent of respondents that get their information about pet food from a veterinarian, online and/or from pet store staff among countries (n = 3,298).

Fig 6. Percent of respondents that look for at least one of the above options when selecting a pet food (n = 3,298).

Fig 7. Percent of respondents in each country that look for “no grain” when selecting a pet food (n = 3,298).

Fig 8. Mean (±SD) score assigned to each factor involved in selecting a pet food among countries (n = 3,298).

Factors important to dog ownership

Fig 9. Percent of respondents who assigned a value of 6–10 to the factors important to dog ownership among countries (n = 3,298).

Multinomial logistic regression

Table 3. Multinomial logistic regression parameter estimates for the Allergy Model.

Table 4. Multinomial logistic regression parameter estimates for the Diet Model.

Table 5. Multinomial logistic regression parameter estimates for the Purchasing Habits Model.

Discussion

Conclusion

Supporting information

Acknowledgments

Data Availability

Funding Statement

References

Decision Letter 0

Juan J Loor

Roles

Author response to Decision Letter 0

Decision Letter 1

Juan J Loor

Roles

Acceptance letter

Juan J Loor

Roles

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases