Patients’ Understanding and Use of Snack Food Package Nutrition Labels

Methods

A convenience sample of adult patients between the ages of 18 and 65 presenting for medical care at the University of Tennessee/St. Francis Family Practice Center in June 2001 were invited to participate in the study. The University of Tennessee Health Sciences Center Institutional Review Board granted approval of this study. Informed consent was obtained from all patients who volunteered to participate in the study. A total of 100 patients were enrolled, and 90 patients completed the entire survey with the investigator. Participants were first given a questionnaire collecting demographic data and patient health information (self-reported). Participants were asked to state their estimated consumption per week of snack foods and beverages, television viewing habits, and weekly amount of exercise.

Participants were interviewed in person by one of the authors (WWC). Each participant was shown photocopies of nutrition information from an assortment of snack food packages. We selected one beverage, bottled for convenience sale, and a selection of 3 prepackaged, commonly available snack food items. Each product package contained multiple servings (ranging from 2 to 3.5 servings per package). Information taken directly from package labels of the products we presented is reproduced in Table 1. Participants were asked a series of questions to determine their ability to correctly interpret nutrition information from the product labels on display. The interview was conducted with minimal prompting and assistance from the interviewer. Initial incorrect interpretations of the nutrition label were not corrected but simply noted. The participants who initially failed to recognize the packages contained multiple servings were prompted to re-evaluate, and their response was again noted. The structure of the complete interview process is outlined in Table 2.

The primary outcome measure was the participants’ final assessment of the total calories in each snack food package and beverage container, based on their interpretation of the nutrition information label. Patients who read the “calories per serving line” as equal to the total package calorie content were scored as “no” (for incorrect interpretation). Patients who ultimately noted that there are multiple servings per package were scored as “yes” for a correct interpretation, even if they had given an initially incorrect interpretation or had difficulty calculating the exact number of calories in the package.

The interview process also asked the participant to interpret the fat content on the nutrition label. However, it quickly became apparent that confusion about fat calories per package paralleled confusion about calorie content. Therefore, only data on participants’ understanding on calories per package was collected and reported as a proxy to understand how well these patients understood the nutrition information label.

Secondary outcome measures were based on correlations between correct or incorrect understanding (of total calories per package) with data collected from the patient survey. Data were analyzed using SAS, version 8, with χ² analysis and Fischer exact test for correlation of label understanding with educational background, exercise, television viewing habits, obesity, and comorbid medical conditions.

Results

Patients reported consuming, on average, 2 fast-food type snacks per day (beverage or packaged snack foods). The majority of our sample patients are sedentary, exercising fewer than 3 times per week. The mean self-reported TV viewing of our sample was 25.4 hours per week. Our patient sample included a significant number that were overweight. The BMI was calculated for 83 patients, and the mean BMI was 28.5. Many of our patients had one or more chronic medical disorders in which lifestyle, diet, and nutrition are known to play a major role, such as diabetes, hypertension, and cardiovascular heart disease (CHD). Demographic and socioeconomic data, self-reported habits, and self-reported health information is summarized in Table 3.

Eighty-one of 90 patients interviewed (91%) knew where to find nutrition information on packaged foods and could correctly identify the line on the nutrition label indicating calories per serving. However, 77 (86%) initially incorrectly identified the total package calorie content equal to “calories per serving.” With prompting, some patients corrected this initial error, but a large majority (57 of 90, or 63%) still in the end confused calories per serving with total calories per package. Patients who had not completed high school were less likely to correctly identify total calories in a package based on the label information ( P = .011). Inability to correctly identify total calories in the package was also associated with the presence of CHD in our patient sample ( P < .05) (Table 4). No significant association was found between incorrect interpretation of package information and race/ethnicity, gender, marital status, television viewing habits, exercise, obesity, and self-reported presence of chronic diseases other than CHD.

Discussion

A recent study has suggested a strong link between snack food consumption and adolescent obesity.10 Like other studies, however, this was based on a recall of food consumption. No studies that we are aware of have examined how a typical general practice patient population understands and uses (or does not) snack food nutrition information. An overwhelming majority of patients in this sample does not understand the distinction between “serving size” and “package/container size” of prepackaged snack foods and beverages. Thus, they believe they are consuming a single serving of snack food per package when in fact they are not. Patients with less education seem to be especially at risk of making this mistake.

Based on self-reported snack food intake of 2 units daily, our patients consume an average of 4 to 6 servings of snack food per day without realizing that they are doing so. The average patient in our study thus consumes (conservatively) 600 to 1000 calories daily from snack food sources. The consequences of this excess caloric intake over time can be highly significant. Because few physicians routinely ask patients about snack food intake, the impact of this on weight and weight-related disorders is very likely to be hidden or overlooked.

Many physicians can relate tales of obese patients who consume large amounts of sugar and caffeine-laden beverages or multiple packages of prepackaged snacks, yet these same patients genuinely seem to have no concept of the effect that this has on their health. Our study may provide an explanation for this phenomenon. It is surprising how little systematic study has been done on snack food intake and its impact on health in actual medical practice. This study suggests the need for better education of patients on this matter, and the need for physicians to specifically ask about snack food intake among patients in their practice. Finally, this study calls into question the utility of snack food package labeling itself.

Our study relies in part on patients’ self-report of food intake and related habits, which may be misleading. However, recent studies have shown that overweight persons tend to under-report caloric energy intake.5 In addition, the personal interviews allowed us to correct for any recall bias by directly testing participant’s knowledge about serving versus package size.

The assumption has been that consumers can correctly use nutrition label information to distinguish between serving size and unit/package size, and use this information to make informed choices about their eating habits. Our survey suggests that this assumption is incorrect. We find that the overwhelming majority of our patients do not understand snack food nutrition labels well enough to make informed dietary choices. This is especially striking in our patients with limited education, who tend to be most at risk for nutrition-related disorders. Further research into this area is needed. The role of advertising and the effects of the widespread availability of snack food are other factors to consider. The results of such clinical research could lead to better understanding and management of obesity in this country.