Sex-Based Differences in Food Consumption: Foodborne Diseases Active Surveillance Network (FoodNet) Population Survey, 2006–2007

Abstract

Background. This analysis used data from the most recent Foodborne Diseases Active Surveillance Network (FoodNet) Population Survey (May 2006 through April 2007) to examine differences in the consumption of various types of foods between men and women.

Methods. Participants were surveyed by telephone and asked whether or not they had consumed certain foods in the past 7 days, including the following “high-risk” foods commonly associated with foodborne illness: pink hamburger, raw oysters, unpasteurized milk, cheese made from unpasteurized milk, runny eggs, and alfalfa sprouts. Data were weighted to adjust for survey design and to reflect the age and sex distribution of the population under FoodNet surveillance.

Results. A total of 14 878 persons ≥18 years were interviewed, of whom 5688 (38%) were men. A higher proportion of men reported eating meat and certain types of poultry than women, whereas a higher proportion of women ate fruits and vegetables. A higher proportion of men than women reported consuming runny eggs (12% versus 8%), pink hamburger (7% versus 4%), and raw oysters (2% versus 0.4%). A higher proportion of women than men ate alfalfa sprouts (3% versus 2%). No differences by sex were observed for consumption of unpasteurized milk or cheese.

Conclusions. Data from the FoodNet Population Surveys can be useful in efforts to design targeted interventions regarding consumption of high-risk foods. Moreover, understanding the background rates of food consumption, stratified by sex, may help investigators identify the kinds of foods likely to be associated with outbreaks in which a preponderance of cases occur among members of one sex.

Several factors can influence the type and quantity of food a person consumes, including income, sex, age, and health status [1]. Studies that report on consumption patterns in population subgroups often focus on specific subgroups of the population (eg, college students, children, and pregnant women), and most have been carried out in countries other than the United States [2, 3]. Knowledge of food consumption patterns by sex may help in targeting health education campaigns that promote healthy eating habits. It can also help in hypothesis generation during outbreak investigations by providing background rates of consumption of certain foods, which can be compared to consumption patterns among cases linked to the outbreak and thereby suggest what the food vehicle may be.

The Foodborne Diseases Active Surveillance Network (FoodNet) conducts periodic surveys of residents of the FoodNet surveillance area, FoodNet Population Surveys, as well as ongoing population-based surveillance for infections transmitted commonly through food. The most recent FoodNet Population Survey was administered from May 2006 through April 2007, gathering data on food-consumption patterns and other topics related to foodborne illness. Approximately 46 million people, or 15% of the US population, reside within the FoodNet surveillance area. Thus, the FoodNet Population Survey can be used to better understand how patterns of food consumption in the general population are similar or differ between men and women. This analysis examined differences in consumption of various types of foods between men and women.

METHODS

FoodNet is the principal foodborne disease component of the Centers for Disease Control and Prevention’s (CDC) Emerging Infections Program. The program is a collaborative effort between the CDC, public health departments in 10 states (Connecticut, Georgia, Maryland, Minnesota, New Mexico, Oregon, Tennessee, and selected counties in California, Colorado, and New York), the US Department of Agriculture’s Food Safety and Inspection Service, and the Food and Drug Administration [4]. Participants in the 12-month 2006–2007 FoodNet Population Survey were randomly selected using a 2-stage, disproportionate, stratified sampling scheme. In the first stage, households within the surveillance area were contacted by telephone, using random-digit dialing. In the second stage, by use of a computer algorithm that took into account the number of males and females in the household, an individual household member ≥1 year old was randomly selected to participate in the survey.

The survey included questions on sex, age, state of residence, income, and education level. Two sets of questions were used to determine whether, during the 7 days before the interview, the respondent had consumed selected food items from the following categories: dairy, meat and poultry, seafood, fresh vegetables, fresh fruits, processed and dried foods, and frozen foods. Respondents were randomly assigned to receive 1 set of questions. Identical questions about a subgroup of foods commonly associated with outbreaks of foodborne illness, or “high-risk” foods, appeared in both sets and therefore were asked of all respondents. These foods included pink (or undercooked) hamburger, raw oysters, unpasteurized milk, cheese made from unpasteurized milk, runny eggs, and alfalfa sprouts.

The analysis was limited to persons ≥18 years of age. Proportions were estimated for each food item and compared using a χ² test. All estimates were weighted to the FoodNet population by age and sex and to account for the sampling design. A P value of <.05 was considered statistically significant. SAS 9.2 survey procedures were used for the analysis (SAS Institute; Cary, NC).

The FoodNet Population Survey was approved by the CDC’s institutional review board, and by the local institutional review board at each site.

RESULTS

A total of 14 878 persons ≥18 years of age were interviewed from May 2006 through April 2007. The overall response rate was 18%. Of these respondents, 5688 (38%) were men and 9190 (62%) were women; most survey participants (86%) were white, and most (65.6%) lived in an urban or suburban area (Table 1). As shown in Tables 2 and 3, general consumption patterns for selected foods and consumption patterns for high-risk foods varied by sex.

A higher proportion of men than women reported eating meat, such as steak and roast, duck and game hen, and ham, whereas a higher proportion of women reported eating fruits and vegetables (Table 2). Brussels sprouts and corn were the only vegetables for which men reported significantly higher consumption rates than women. Most men and women reported consuming chicken; game (eg, duck and game hen) was consumed by more men, although still at a relatively low rate. A higher proportion of men than women reported eating shrimp, oysters (including raw oysters), and other shellfish. A higher proportion of women than men reported eating frozen berries and frozen vegetables. For dried foods, more women reported eating almonds and walnuts, whereas more men reported consuming peanuts.

A greater proportion of men than women reported consuming several high-risk foods. These included runny eggs (12% versus 8%; P < .0001), pink hamburger (7% versus 4%; P < .0001), and raw oysters (1.9% versus 0.4%; P < .0001) (Table 3). A significantly higher proportion of women than men reported eating alfalfa sprouts (3% versus 2%; P = .01). No significant differences by sex were seen for consumption of unpasteurized milk or cheese.

Sex-based differences for most of the food items persisted after analyses were controlled for age.

DISCUSSION

Our analysis demonstrated important differences in food consumption patterns between men and women interviewed in the 2006–2007 FoodNet Population survey, a large survey of US adults. A greater proportion of men than women reported eating meat, such as steak and roast, duck and game hen, and ham, whereas more women than men reported eating fruits and vegetables in the week prior to interview. Also, of note, more women than men reported consumption of organic produce. Men were also generally more likely to report consumption of high-risk foods, with the exception of alfalfa sprouts.

These findings are generally consistent with findings from other studies [5–8] and extend those findings both to the general population and to specific high-risk foods of public health concern. The 1987 National Health Interview Survey found that, on average, women consume more fruits and vegetables, less meat, and fewer high-fat foods than men [7]. Similarly, a study of African Americans demonstrated that older age and female sex were associated with higher intake of fruits and vegetables [8]. A report by the Department of Agriculture Economic Research Service showed that women are more knowledgeable than men about diet and nutrition [9]. In a study examining sex-based differences in dieting trends, eating habits, and nutrition beliefs among college students, significantly more women than men reported trying a low-fat diet or a low-carbohydrate diet. Additionally, more women than men believed that it was important to limit consumption of carbohydrates and fat to lose weight and that they needed to lose weight [6]. This could be one reason for the differences in eating habits between men and women. Among elementary school students in Ohio, a correlation was observed between nutrition knowledge and food choices for girls in the 7th and 8th grades [10]. Nutrition knowledge scores were higher for girls than boys. They concluded that nutrition knowledge might have a positive impact on eating behavior.

Similar findings have also been reported in studies conducted outside the United States. In Finland and the Baltic countries, men more frequently consume meat and meat products, while women more frequently consume fruits and vegetables [5]. An Australian study showed that women consumed more fruits and vegetables than men and that women have more knowledge about nutrition, which may result in more-healthful eating habits. [11]. Additionally, a study involving participants from 23 countries demonstrated that women were more likely than men to report eating high-fiber foods and fruits and avoiding high-fat foods; approximately 50% of the sex-based differences in food choices were attributable to a desire for greater weight control and a stronger commitment to making healthy food choices among women, compared with men [1]. A similar conclusion was reached in a Norwegian study, in which women were found to be more health-conscious than men and more likely to adhere to recommended dietary practices [12]. A study among Swedish university students showed that male students had higher rates of overweight and obesity than female students and that male students were less interested in nutrition advice and health-enhancing activities [2].

Our analysis of high-risk foods revealed that men were more likely to consume undercooked hamburger, raw oysters, and runny eggs. Only 1 high-risk food, alfalfa sprouts, was consumed more frequently by women. This could be because alfalfa sprouts are more likely consumed with salads, since more women reported eating green leafy vegetables than men. These findings are consistent with those from studies that used data from the 1998–1999 FoodNet Population Survey, which demonstrated that 42% of men compared with 34% of women ate ≥1 risky food in the week before survey participation [13]. A similar pattern was also reported in 2 studies using a meta-analysis by Patil and colleagues, in which a higher proportion of men reported consumption of raw or undercooked ground beef and eggs than women (27% versus 21% for undercooked ground beef and 54% versus 47% for undercooked eggs) [14, 15].

Some of the limitations of this study are that the surveys are based on self-report, and recall bias could be an issue. It may be a food preference rather than what they actually ate in the 7 days before the interview. Samples for this survey were drawn from published telephone directories, and so this excluded people with cellular phones only, which is a growing population. Another limitation is the low response rate. But, the response rate in this study is not different from that in other telephone-based surveys.

The results of this analysis may be useful in the design of nutrition campaigns and targeted interventions related to the consumption of high-risk foods. The results may also be pertinent to the investigation of foodborne illness outbreaks. For example, in a multistate outbreak of Salmonella serotype Enteritidis infection associated with consumption of raw almonds, comparison of what the cases ate with the background rate from the FoodNet Population Survey gave an insight to the vehicle [16]. In outbreak situations, comparison of the food consumption patterns among outbreak-associated cases to the background rate of consumption in the population may help provide initial insight as to whether a particular food item merits further study as a possible cause of the outbreak.

Notes

Disclaimer.

The contents of this work are solely the responsibility of the authors and do not necessarily represent the official views of the Centers for Disease Control and Prevention.

Financial support.

This work was supported in part by the Centers for Disease Control and Prevention (CDC; Cooperative Agreement U60/CD303019). FoodNet is funded by the Food Safety Office and the Emerging Infections Program of the CDC, the Department of Agriculture Food Safety and Inspection Service, and the Food and Drug Administration.

Supplement sponsorship.

This article was published as part of a supplement entitled “Studies From the Foodborne Diseases Active Surveillance Network,” sponsored by the Division of Foodborne, Waterborne, and Environmental Diseases of the National Center for Emerging and Zoonotic Infectious Diseases from the Centers for Disease Control and Prevention, and the Association of Public Health Laboratories.

Potential conflicts of interest.

All authors: No reported conflicts.

All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.

References