In 2009–10, 35.7% of US adults were obese.¹ Although there are multiple causes of obesity, one potential contributor is regularly consuming foods prepared away from home, such as those eaten at fast-food or chain restaurants. Foods from these venues are often high in calories,² thus, not surprisingly, there is an association between fast-food consumption and excessive energy intake³ and obesity.^{4 – 6} Further, many customers underestimate the number of calories in restaurant items and meals.^{7 – 9}

Before 2007, nutrition information was seldom available in restaurants at the point of purchase.¹⁰ When it was available, it was often difficult to find.^11,12 Displaying calorie information on menus and menu boards has been hypothesized as a strategy that may influence energy intake by increasing consumer awareness of the caloric content of menu items to inform their decision-making. In recent years, local municipalities in the USA have passed policies requiring restaurants to post-calorie and other nutritional information.¹³ In 2010 Congress passed the Patient Protection and Affordable Care Act which, when implemented, will require chain restaurants with 20 or more locations to make certain nutritional information publicly available.¹⁴

Research conducted in jurisdictions that have implemented calorie labelling polices have found mixed results for adults’ use of calorie information at fast-food and chain restaurants. Dumanovsky et al.¹⁵ found that the percentage of adults in New York City (NYC) who reported seeing calorie information increased from 25% before the regulation requiring calories be posted on menu boards to 64% after implementation. Among customers who saw calorie information post-enforcement, 27% said they used the information.¹⁵ In a separate study, Dumanovsky et al.¹⁰ conducted cross-sectional surveys on adults before and 9 months after enforcement of NYC’s calorie labelling regulation to determine the mean caloric content purchased among customers who said that they used the calorie information when deciding what to order. The 15% of customers who reported using calorie information purchased 106 fewer calories than customers who did not see or use the calorie information. However, in King County, WA, researchers found no difference in the number of calories purchased following menu labelling legislation.¹⁶

The purpose of our research was to (i) determine the proportion of US adults that read calorie information when it is available at fast-food or chain restaurants and describe the sociodemographic and behavioural characteristics associated with reading the calorie information and (ii) determine the prevalence of using this information to help select their food purchases among those who read it.

Our cross-sectional study was based on the 2009 HealthStyles Survey. HealthStyles is a national mail survey administered annually as a follow-up survey to ConsumerStyles, a consumer panel survey administered by Synovate, Inc. ConsumerStyles is sent to a stratified random sample of 21 420 US adults in Synovate’s panel of participants. Low income and minority groups are oversampled in ConsumerStyles to have sufficient representation.¹⁷ Respondents receive a small monetary incentive. The response rate for the 2009 ConsumerStyles Survey was 49.4%.

HealthStyles is based on a random sample of panel households that return ConsumerStyles. HealthStyles surveys US adults (≥18 years) and is designed to assess health-related attitudes, knowledge and behaviours through a mail survey. The response rate for 2009 Health Styles was 65.0% (4556/7004). The data were weighted on gender, age, income, race and household size to match the 2008 US Current Population Survey to make the sample representative of the US population.

To determine the proportion of adults who read calorie information when it is available at fast-food and chain restaurants, participants who go to fast-food or chain restaurants were asked: ‘Do you typically read calorie information for foods and drinks when it is available at fast-food and chain restaurants?’ Response options were ‘Yes’, ‘No’, ‘Never noticed or looked for calorie information’, ‘Usually cannot find calorie information’ and ‘Don’t Know’.

To determine the proportion of adults who use calorie information to decide their order, those who reported ‘Yes’ to reading calorie information to the question described above were asked ‘How often does this calorie information help you decide what to order?’ Response options were ‘Always’, ‘Most of the time’, ‘About half of the time’, ‘Sometimes’, ‘Never’ and ‘Don’t Know’. We dichotomized responses as ‘Yes’ (‘Always’, ‘Most of the time’, ‘Half of the time’, ‘Sometimes’) and ‘No’ (‘Never’).

Covariates included gender, age group (18–34, 35–44, 45–54, 55–64 and 65+ years), race/ethnicity (non-Hispanic white, non-Hispanic black, Hispanic and non-Hispanic other), marital status (married/domestic partnership and not married, which included widowed, divorced and single persons), household income (<$30 000, $30 000 to <$60 000, $60 000 to <$85 000 and ≥$85 000), education (high school or less, some college, college graduate or more), region of the country (New England, Middle Atlantic, East North Central, West North Central, South Atlantic, East South Central, West South Central, Mountain and Pacific) and frequency of eating at a fast-food or chain restaurant per week (never, less than four times per month, one to two times per week and three or more times per week).

Selection of several of the explanatory variables was based on previous findings for the use of calorie information when provided on menu labelling or on the Nutrition Facts Panel. For example, women are more likely to read calorie information compared with men,¹⁰ adults in wealthier neighbourhoods are more likely to report using calorie information than those in poorer neighbourhoods¹⁰ and those who go to fast-food restaurants are less likely to use calorie information compared with those who do not go to fast-food restaurants.¹⁸

The data set included 4556 respondents. We excluded 193 respondents because of missing data or a non-classifiable response (i.e. ‘don’t know’) leaving an analytic sample of 4363. Specifically, we excluded respondents with missing data for marital status (n =5), education (n =39), fast-food or chain restaurant frequency (n =30), read calorie information when available (n =72) and use calorie information (n =13) and respondents for selecting ‘Don’t know’ to the read calorie information (n =22) and use calorie information (n =12) questions.

We assessed the prevalence of reading calorie information when available among adults who go to fast-food or chain restaurants. The analytic sample used (n =3512) included those who go to fast-food or chain restaurants and responded ‘Yes’ or ‘No’ to the read calorie information when available question. From the sample of 4363 adults, we excluded those who do not go to fast-food or chain restaurants (n =441), those who responded ‘Never noticed or looked for calorie information’ (n =202) and those who reported ‘Usually cannot find calorie information’ (n =208).

We conducted multivariable logistic regression to estimate odds ratios (ORs) and 95% confidence intervals (CIs) for the explanatory variables in the model. The multivariable logistic regression model adjusted for gender, age, race/ethnicity, marital status, annual household income, education, region and frequency of eating at a fast-food or chain restaurant. We also assessed the prevalence of using calorie information while ordering among those who read calorie information when available. Weighted percentages of using calorie information were compared by each sociodemographic and behavioural characteristic using X² tests (unadjusted) and a P-value of 0.05; we also calculated the standard error (SE) for each sociodemographic and behavioural characteristic for using calorie information. All statistical analyses were performed using the Statistical Analysis Software (SAS), version 9.3, which accounted for the sample design.

The sociodemographic and behavioural characteristics are described for the total sample in Table 1 (first numeric column). Just over half were women (51.7%), and 30.8% were in the 18–34 year age group, 69.4% were non-Hispanic White and 59.9% were married or in a domestic partnership. One-tenth of the sample (10.1%) reported never going to fast-food or chain restaurants, over half (55.6%) reported going to fast-food or chain restaurants less than four times per month, and 10.4% reported going three or more times per week.

Among those who went to fast-food or chain restaurants and noticed calorie information when it was available, the prevalence of reading calorie information was 36.4% (Table 1, second numeric column). Those who ate at fast-food or chain restaurants three or more times per week had a lower prevalence of reading calorie information when available compared with those who went less often (25.9% for ≥3 times per week versus 34.4% for one to two times per week versus 39.3% for <4 times per month). Among those who went to fast-food and chain restaurants, multivariable logistic regression found that women were more likely than men to read calorie information when available [adjusted OR (aOR) =1.8, 95% CI =1.5–2.1; Table 1, last numeric column]. In addition, those who ate at a fast-food or chain restaurant three or more times per week were less likely to read calorie information (aOR =0.6, 95% CI =0.4–0.8) than patrons who went less than four times per month.

Among those who reported going to fast-food or chain restaurants and reading calorie information, the proportion of adults who reported using calorie information when available for each response option was: 13.8% ‘always’, 40.3 ‘most of the time’, 15.9% ‘half of the time’, 25.4% ‘sometimes’ and 4.6% ‘never’ (data not shown). Thus, 95.4% (1251/1309) used calorie information at least sometimes (Table 2). Chi-square tests found that the only statistically significant difference among subcategories for the sociodemographic and behavioural characteristics was for region of the country, although the difference between males and females approached significance (P =0.06).

The findings of this study raise additional research questions. First, there is a need to understand why two-thirds of fast-food or chain restaurant attendees do not read calorie information when available. Secondly, there is need to understand how readers are using calorie information and how use can be improved or expanded as needed. The findings and conclusions in this report are those of the authors and do not necessarily represent the views of the Centers for Disease Control and Prevention.