Americans’ Perceptions about Fast Food and How They Associate with Its Consumption and Obesity Risk

Abstract

We aimed to systematically examine Americans’ perceptions of fast food (FF) and how these perceptions might affect fast food consumption (FFC) and obesity risk. We searched PubMed and Google for studies published in English until February 17, 2017 that reported on Americans’ perceptions (defined as their beliefs, attitudes, and knowledge) regarding FF as well as those on their associations with FFC and obesity risk. Thirteen articles met inclusion criteria. Limited research has been conducted on these topics, and most studies were based on convenience samples. A 2013 nationally representative phone survey of about 2000 subjects showed that one-fifth of Americans thought FF was good for health, whereas two-thirds considered FF not good. Even over two-thirds of weekly FF consumers (47% of the total population) thought FF not good. Americans seem to have limited knowledge of calories in FF. Negative and positive FF perceptions were associated with FFC. Those who consumed less FF seemed more likely to view FF negatively. When Americans valued the convenience and taste of FF and preferred FF restaurants with kid's menus and play areas, they were likely to purchase more FF. Available research indicates neither perceived availability of FF nor Geographical Information System (GIS)-based FF presence in the neighborhood has significant associations with weekly FFC. No studies examined potential links between FF perceptions and obesity risk. Americans’ perceptions of FF and how they might associate with FFC and obesity risk are understudied. Considerable variation was observed in Americans’ perceptions and FFC.

Introduction

Consumption of fast food (FF, food being mass-produced and served quickly) is common in the United States and many other industrialized countries, and it has been increasing steadily in some developing countries as well owing to factors such as its convenience, low cost, consistent taste, easy access through a variety of restaurant chains, and the FF industry's marketing effort (1, 2). A 2013 survey showed that about half of Americans reported eating at FF restaurants at least weekly (3), and ∼80% of adults went to FF restaurants at least once per month (3, 4). FF is one of the major components of Americans’ diet, as about 11% of their calories are derived from fast food consumption (FFC) among adults (5).

FF is often made with ingredients such as high-fat meat, refined grains, and added sugar and fats (6) and is high in sodium, saturated fat, and cholesterol, which are harmful to health (7, 8). FF often contributes to higher energy intake compared with traditional food items (9). Multiple studies have found that frequent FFC is associated with elevated BMI and cholesterol, and increased risk for obesity, hypertension, and diabetes (2, 10–12). As obesity is becoming a global public health crisis and has resulted in high health and economic costs (13), researchers have recommended avoiding frequent FFC (14, 15).

Although some FF industries have altered their menus to include healthier options (e.g., whole-wheat buns, low-fat milk, grilled chicken nuggets), their aggressive FF marketing via television, mobile devices, and social media promotes liking of FF, but not specifically healthier FF choices (16). Research suggests that heavy television viewing might lead to less negative and more positive beliefs about FF's health consequences (17). Compared with adults, children are more susceptible to frequent FFC owing to peer influence and impulsive actions (18). Also, subgroups of American adults who have the highest percentage of total energy intake from FF include young adults (16%), non-Hispanic blacks (15%), and obese individuals (13%) (5). Young non-Hispanic blacks have the greatest intake from FF, contributing 21% of their total energy intake.

An inverse socioeconomic gradient in FFC among adults and children has also been reported in the United States (19, 20). Low-income and food-insecure people consume more FF owing in part to lack of access to healthy and affordable foods (21) and perceived barriers of higher prices for healthy foods and lower prices for FF (22), which also contributes to sustaining FF as a major contributor to Americans’ diets. Additionally, a systematic review found that FF restaurants are more prevalent in low-income areas compared with middle- to higher-income areas and in areas with higher concentrations of ethnic minority groups compared with Caucasians (23). Disproportionately high access to FF in the United States is prevalent in predominantly black neighborhoods (24).

Some research suggests that individuals’ beliefs, attitudes, and knowledge, such as perceived barriers, self-confidence, or risks/benefits in dietary behaviors, statistically predict eating behavior (25); those having strong confidence in diet control are less likely to consume high-calorie diets (26). Impulsive individuals prefer convenience meals more than health-conscious individuals (27), whereas those who pay more attention to food labels are more likely to consume a healthy diet. Yet, no study has systemically examined how Americans think about FFC, and whether they want to limit or continue their FFC, as reducing FFC has been recommended for preventing obesity. Differences in nutrition- and health-related attitudes, beliefs, and knowledge, including making informed food choices and awareness of nutrition-related health risks (28), could be implicated in high FFC in the United States.

This study aimed to systematically examine Americans’ perceptions of FF, defined as their beliefs, attitudes, and knowledge regarding FF, and to assess how FF perceptions are related to FFC and obesity risk.

Methods

Literature search

We searched PubMed and Google (http://google.com/) for relevant studies published in English through February 17, 2017. We used the following related keywords with Medical Subject Heading (MeSH) terms and conducted title and abstract searches: fast food, intake, consumption, knowledge, attitude, belief, perception, obesity, overweight, weight, BMI, the United States, and America. We also consulted several experts in the field. The search yielded 169 titles. We reviewed the titles and abstracts to exclude duplicates and publications not related to the topic, which resulted in 69 publications. We then reviewed the full papers. Thirteen publications met our inclusion criteria, with 3 of them brought to our attention by experts we consulted (Figure 1).

FIGURE 1.

Flowchart of literature search and study selection procedure. FF, fast food; FFC, fast food consumption.

Study inclusion and exclusion criteria

Inclusion criteria were: 1) an original study (i.e., based upon quantitative data analysis), 2) conducted in the United States, 3) focused on FF, 4) that reported results on people's perceptions, attitudes, knowledge, or beliefs about FF or FFC, 5) included healthy people without disease, and 6) had a sample size ≥100.

Our study's exclusion criteria included: 1) review papers or qualitative research without structured data collection, 2) studies conducted outside the United States, 3) studies focused only on selected food groups (i.e., sugar-sweetened beverages, unhealthy snacks, or convenience foods), 4) absence of results on people's perceptions, attitudes, knowledge, or beliefs about FF or FFC, 5) participants having diet-related diseases, and 6) sample size <100.

Data extraction

We extracted information including study setting, study design and objective, sample characteristics (e.g., age, sex, sample size, race/ethnicity, sampling area, prevalence of overweight/obesity), FF definition in the study, level of FFC, assessment and level of attitudes, knowledge, beliefs, and perceptions of FF and its associations with FFC and obesity, while following PRISMA guidelines, into a standardized data extraction form. Two co-authors carried out the literature search and data extraction.

Study quality assessment

We modified the NIH quality assessment tool for observational cohort and cross-sectional studies (29) to assess the quality of each included study by 7 criteria according to the study characteristics. For each criterion, a score of 1 was assigned if “yes” was the response, whereas “0” was assigned otherwise (i.e., an answer of “no”, “not applicable”, “not reported”, or “cannot determine”). A study-specific global score, ranging from 0 to 7 (a higher score indicates a better rank than a lower score), was calculated by summing scores across all criteria. This quality assessment helped to describe the strength of the scientific evidence, but was not used to determine the inclusion of studies (Supplemental Table 2).

Results

Main characteristics of the included studies

Study design and quality

From a pool of 13 studies, there was 1 longitudinal study and 12 cross-sectional studies. The cohort study collected information from paper questionnaires administered in person. Seven cross-sectional surveys were web-based or random-digit-dial telephone-based, whereas the remaining studies were conducted through intercept interviews at FF restaurants, colleges, and public health centers from the following states: California, Maryland, Minnesota, New Jersey, New York, North Carolina, Philadelphia, and South Carolina (Table 1). The study quality assessment score ranged from 3 to 6 out of 7 (Supplemental Tables 1 and 2). All studies included study samples of >100 participants [the smallest sample was 107 (30) and the largest 10,450 (31)], and specified measures of FFC and FF-related knowledge, attitudes, beliefs, and perceptions.

TABLE 1.

Main characteristics of the 13 studies regarding Americans’ perceptions of fast food (FF) and FF consumption (FFC)¹

Reference	Research focus/aims	Study design and setting2	Study subjects3	FF studied	Reported FFC	Measures of FF or FF-related perception	Reported FF perceptions	FFC associated demographics	Study quality4
1. Lee-Kwan et al. (39)	Reasons associated with purchasing kids’ meals	Web-based national survey	n = 1147, M, F, parents (>18 y), OW: 30.2%, OB: 32.3%	Kids’ meals at a FF or chain restaurant	51% of parents purchased kids’ meals in the past month	Reasons for purchasing kids’ meals with 11-option response	Reasons parents buy kids’ meals at FF restaurants: 1) child asked for it (48%); 2) usual habit (42%); 3) healthier sides (e.g., fruits; 25%)	Young parents (OR: 3.4; 95% CI: 2.1, 5.6 in 18–34 y vs. 50 y) and SSB daily drinkers (OR: 2.7; 95% CI: 1.7, 4.3 vs. none) were more likely to purchase kids’ meal.	4
2. Lorts and Ohri-Vachaspati (38)	Weight loss attempts and eating behaviors	Random-digit-dial household phone survey; NJ	n = 548, M, F, obese adults, low income	FF, energy-dense snack foods, SSB	# of FFC = 0.99 ± 1.80/wk	The link between weight loss attempts and FFC5	FFC was perceived as bad for weight loss	NR	4
3. Oexle et al. (40)	Neighborhood FF availability and FFC	Telephone survey; SC	n = 838, M, F, adults, OW: 36%, OB: 33%	McDonald's, KFC, Taco Bell, takeout pizza	Never: 59.4%, 1 time/wk: 18.3%, >1 time/wk: 22.3%	5-point Likert scale on perceived FF availability in neighborhood; GIS-based presence of FF outlets	Perceived FF availability: low (score: 0, 1)—56.0%, high (3, 4)—44.0%	Younger age (<45 y: 41.5% vs. ≥65 y: 10%), white (24.7% vs. others: 17.5%), being employed (34.2% vs. unemployed: 19.3%) had higher risk of weekly FFC	4
4. Russell and Buhrau (17)	Television viewing and adolescents’ beliefs about the health risks of FFC	2 online surveys	Survey 1/2: n = 445/1048, M, F, 14–17 y/12–17 y	FF	# of days/month eating FF in survey 1/2 = 3.32 ± 1.17/3.42 ± 1.24	5-point Likert scales [very unlikely (1) to very likely (5)] on beliefs about health risks (e.g., harm their health)/benefits (e.g., feel good) of eating FF after eating in a FF restaurant every day	Positive health consequences score of FFC: 2.8 ± 1.1, negative health consequences score of FFC: 3.5 ± 1.2	Heavy TV viewing related to more positive beliefs (β ± SE = −0.01 ± 0.02), but less negative beliefs on FF health consequences (β ± SE = −0.08 ± 0.03; all P values <0.01)	5
5. Gallup daily tracking survey (3)	Americans’ perception of FF	Telephone interviews; nationally representative sample in 50 states and DC	n = 2027, M, F, adults	FF	≥ 1 time/wk: 47%	4-point Likert scales on the health value of food served in FF restaurants	Good: 22%, not good: 76%	NR	4
6. Lucan and Mitra (31)	Perceptions of the food environment and FFC	Random-digit-dialing telephone survey; PA	n = 10,450, M, F, adults	FF restaurants such as McDonald's, Pizza Hut, Crown Fried Chicken	Mean: 0.77 times/wk; 1–2 times/wk: 33.5%; no FF in past wk: 58.7%	Questionnaire on food environment perception5	Difficulty finding produce in neighborhood (4.7%), poor supermarket accessibility (31.6%), and poor grocery quality (13.9%)	Greater FFC was associated with 1) difficulty vs. easy finding produce availability (IRR: 1.31; 95% CI: 1.19, 1.45), 2) with vs. without supermarket accessibility (IRR: 1.06; 95% CI: 1.00, 1.11), 3) poor vs. good grocery quality (IRR: 1.20; 95% CI: 1.12, 1.28)	4
7. Elbel (35)	Calorie labeling and calorie knowledge at FF restaurants	Survey outside FF restaurants in low-income communities before and after calorie labeling implementation; NY and NJ	N = 1122 from 19 FF restaurants, M, F, age = 38.2 ± 14.1 y, Black, Hispanic, Latino, low SES	McDonald's, Burger King, Wendy's, and KFC	Usual FF consumers	Knowledge of FF calories using open-ended questions	After implementing calorie labeling on menus, the correct estimation of FF calorie counts increased 60% (15%→24%) in NY. 23% correctly estimated daily caloric requirements	NR	4
8. Piron et al. (34)	Knowledge, and attitudes to FF calorie labeling	Calorie and Nutrition Information Survey; CA	N = 639, M, F, age = 34.9 ± 11.6 y, mostly African American and Latino, public health clinic patients, underserved population, OW: 34%, OB: 21%	McDonald's	Ate at McDonald's: ≥1 time/wk, 22%; ≥1 time/mo, 28%; few times a year, 25%; almost never/never, 26%	Questionnaire on calorie knowledge and attitudes toward FF calorie labeling	23% correctly estimated daily caloric requirements. 93% thought FF calorie labeling important. 85% thought FF calorie info should be posted on the menu	NR	4
9. Dave et al. (37)	Relationship of attitudes toward FF and FFC in adults	Random-digit-dial telephone survey; MN	N = 530, M, F, adults, white (94%)	Burger King, Hardee's, Kentucky Fried Chicken, Pizza Hut, etc.	All ate at FF restaurant ≥1 time/wk	Attitude toward FFC measured by 13 items with 5-point Likert scale response	Determining factors of FFC: perceived convenience, entertainment, and unhealthfulness of FF	Male (OR: 1.9; 95% CI: 1.5, 2.6 vs. female) and married (OR: 0.5; 95% CI: 0.4, 0.7 vs. single) were more likely to have FFC >1 time/wk	4
10. Larson et al. (32)	Factors associated with FFC	Longitudinal study in high school classrooms; MN	N = 1686, M, F, age = 15.9 ± 0.8 y at baseline; 20.5 ± 0.9 y at follow-up	McDonald's, Burger King, Hardee's, etc.	Never/wk: 21.8%, 1–2 times/wk: 50.9%, >3 times/wk: 27.3% at follow-up	Questionnaire on taste/time barriers or benefits of healthy eating, self-efficacy for healthy eating, and concern about health5 at baseline	NR	Frequent FFC was most common among males of low-middle SES (43.8%), least common among females attending college (14.5%)	6
11. Bryant and Dundes (30)	FF perceptions	Survey; MD	N = 107, M, F, college students	FF	NR	Rank the 5 inducing factors to buy FF	Taste: 84%; convenience: 70% (M), 68% (F); cheap price: 61% (M), 34% (F)	NR	3
12. Ayala et al. (33)	Restaurant and food shopping practices among Latino women	Recruited via random-digit dial; home-based interview; CA	N = 357, F, adults, predominately Mexican immigrants, OW: 39%, OB: 41.6%	FF restaurants	# of FFC: 1.0 ± 1.3 times/wk	5-point Likert scale to assess 7 reasons for choosing FF over other restaurant	Preferred FF restaurant to other restaurant (43.1%)	FFC was higher among those younger, employed, lower income, and lived in the US longer (all P values <0.001)	4
13. Satia et al. (36)	African Americans’ use of FF restaurants	Survey; NC	N = 568, M, F, adults, black, OW: 35%, OB: 40%	McDonald's, Pizza Hut, or fast Chinese restaurants	FFC in last 3 mo: usually, 4%; often, 22%; sometimes, 50%; rarely/never, 24%	Questionnaire on diet-related psychosocial factors5	Self-efficacy to eat less fat: not confident 12%; affordable to purchase healthy foods: sometimes/no 27%	FFC was higher among those younger, never married, obese, and physically inactive (all P values <0.001); not significantly different in FFC by sex, education, smoking, or urban/rural residence	4

¹Values are means ± SDs, unless otherwise indicated. #, frequency; FF, fast food; FFC, fast food consumption; IRR, incidence rate ratio; NR, not reported; OB, obese; OW, overweight; SSB, Sugar-Sweetened Beverages; SES, socioeconomic status.

²One study was a cohort study (32); all others were cross-sectional studies.

³Majority race/ethnic group(s) were specified, otherwise multiple race/ethnic groups were involved in the study.

⁴This study quality assessment tool was adopted from the NIH's quality assessment tool for observational cohort and cross-sectional studies (29). A study-specific global score, ranging from 0 to 7 (a higher score indicates a better rank than a lower score), was calculated by summing up scores across all criteria.

⁵No direct FF perception was accessed.

Participant characteristics

Demographics of study subjects were heterogeneous in the 13 studies. Two studies targeted children (<18 y) (17, 32). The range of mean age in the 11 other studies was from 34.9 y to 57.6 y. Except for 1 study (33), the studies included both males and females. Five (42%) studies were race/ethnicity-based: 2 studies included mostly African-Americans and Latinos (34, 35), 3 studies were specifically composed of African Americans (36), whites (94%) (37), or first-generation Mexican immigrant Latinos (33), whereas 8 studies (55%) had diverse race/ethnic groups (3, 17, 30, 31, 32, 38, 39, 40).

Among the 6 studies reporting participants’ weight status, 1 study had all obese participants (38), and in the majority of the studies at least two-thirds of participants were either overweight or obese (33, 34, 36, 38–40). Three studies reported data from low-SES populations (34, 35, 38). A single study collected data from public health clinic clients (34).

Definition of FF in the studies reviewed

FF refers to foods that are mass-produced and can be served quickly. They often provide lower nutritional value and higher sugar, fat, or sodium content compared with other dishes. In these studies, the most common definition of FF was foods derived from FF restaurants such as McDonald's, Pizza Hut, Burger King, Kentucky Fried Chicken, Taco Bell, Wendy's, and other similar establishments (Table 1), unlike Piron and colleagues who focused solely on McDonald's (34). The consumption of FF was measured through various FFQs or 5-point Likert scales. Studies reported FFC with a frequency of 1 wk, 1 mo, 3 mo, or 1 y. Percentages or mean number of FF meals per week were reported [e.g., 50% of subjects ate FF 1–2 times/wk (32), having FF 1.0 time/wk on average (33), see Table 1].

Main research focus of the studies

The majority of the 13 studies investigated Americans’ psychosocial factors/personal preference in FF purchases (30, 32, 33, 36, 37, 39) or perceived food environments (31, 40), such as FF availability in the neighborhood. The link between weight loss attempts (38) and individuals’ health concerns about FF (3, 17) with FFC were addressed in 3 studies. Two studies examined the association of nutrition knowledge [e.g., required daily calorie level (34, 35) and FF calorie estimation (35)] with FFC. No studies directly compared obesity risk by different FF perceptions (Figure 2).

FIGURE 2.

Number of fast food perception studies by research focus and study design. (A) Research focus. (B) Sampling pool. (C) Age group of study subjects.

Perceptions of FF

Interviews, questionnaires, and focus group discussions were used to assess perceptions of FF regarding beliefs, attitudes, and knowledge. A nationally representative survey (which included >2000 adults matched to national sociodemographic characteristics) reported that in 2003, 23% of Americans considered FF to be “good” (i.e., “very good” or “fairly good”), which did not change over 10 y. About two-thirds of Americans thought FF was “not good” (i.e., “not too good”, “not good at all”) in both 2003 and 2013, while the percentage who considered FF “not good at all” increased from 23% to 28% (Figure 3) (3). One study directly measured adolescents’ beliefs about risks and benefits of eating FF in online panels using a 5-point Likert scale (range: 1–5). The mean score of adolescents’ reported experience of positive health consequences (e.g., feeling good) after eating at FF restaurants every day was 2.8 (SD: 1.1) and that for negative health consequences (e.g., harming their health) was 3.5 (SD: 1.2) (17).

FIGURE 3.

Americans’ perception and consumption of fast food based on Gallup daily tracking survey in 2003, 2006, and 2013 (3). (A) Americans’ perception of fast food by year (%). (B) Americans’ frequency of eating fast food by year (%). (C) Americans’ perception of fast food by fast food consumption level in 2013 (%). Data source: Gallup daily tracking survey based on telephone interviews conducted in 2003, 2006, and 2013 with a random sample of >2000 adults (aged ≥18 y old) per wave, living in all 50 US states and the District of Columbia. The samples were weighted to match the national demographics of gender, age, race, Hispanic ethnicity, education, region, population density, and phone status (3). This is the best related study we identified that reported on Americans’ perceptions of FF, although it was not published in a peer-reviewed journal.

Seven studies examined reasons or personal factors for purchasing FF with multiple-choice or open-ended questions, regarding negative compared with positive attitudes/beliefs toward FF and related health concerns, such as 1) preferred convenience [e.g., I eat at FF restaurants because they are easy to get to, quick, no need to cook (37), fewer time barriers to FFC (32), I buy FF since it is convenient (30), and being not far away from home/work (33)], 2) affordability [e.g., I eat at FF restaurants because they are inexpensive (37) and affordable with low price (33, 36), I buy FF due to its amount of food for the money (30)], 3) good taste [e.g., I buy FF since it is tasty (30), I have less taste barrier to FF vs. healthy foods (32), I'm not fully confident to eat less fat (56%) or keep healthy eating (40%) (36)], 4) more fun and socializing opportunities than other eating places [e.g., I eat at FF restaurants because they are fun and entertaining, other people in my life chose to eat there; it is a way of socializing with friends or family (37), my child loves the kids’ menu and play area (33, 39)], and, on the other hand, 5) perceptions of FF as unhealthful [e.g., I think FF restaurants serve not good food for health (3), I think FF restaurants serve mostly high-fat foods or unsafe foods (37), I have concerns about health issues of FF (32), FF is bad for weight loss (38)], etc. However, these studies were unable to fully document the diverse perspectives of Americans’ FF perceptions in the studies owing to the limitations of survey questionnaires.

People's perceptions of FF availability [high: 44% (40)] and food environments (e.g., poor supermarket accessibility: 31.6%, poor grocery quality: 13.9%) (31), knowledge of daily caloric requirements [23% correctly answered (34, 35)], and estimation of FF calories [15% (35)] were estimated with a Likert scale or open-ended questions (Table 1).

Associations between FF perceptions, FFC, and obesity risk

No published studies examined the association between FF perception and obesity risk. Some reported on diverse associations between Americans’ perceptions of FF and FFC (see Table 2). Several salient points emerged regarding FF and obesity risk factors. First, an individual's positive attitudes about FF may associate with a high level of FFC. People had more frequent FF intake when they thought of FF as convenient (likelihood of FFC >1 time/wk, OR: 1.2; 95% CI: 1.1, 1.2 compared with not perceived as convenient) (37), and preferred having a kid's menu and play area [data were not shown, P < 0.05 (33)]. Americans who consumed less FF seemed more likely to view FF negatively. Those who ate FF least often (< a few times per year) were about 2 times more likely to report FF as “not good at all” than those who ate FF at least weekly (43% compared with 19%) (3). In addition, girls who reported barriers in time (per 1 SD change, OR: 1.18; 95% CI: 1.04, 1.34) (32) and taste (OR: 1.26; 95% CI: 1.11, 1.44) (32) to healthy eating compared with FF had more frequent consumption of FF than others. Having low self-efficacy to avoid fat or the unhealthfulness of FF were also predictors of frequent FF restaurant use [risk of more frequent FFC per 1 SD change of self-efficacy for healthy eating, OR: 0.80; 95% CI: 0.75, 0.98 (32); data were not shown, all P values <0.05 (36)]. Those who attempted weight loss ate less FF [likelihood of weekly FFC, OR: 0.77, 95% CI: 0.62, 0.97 compared with not trying to lose weight (38)], which indicated weight concerns surrounding FFC. However, health care provider's advice to lose weight among obese subjects was not associated with lower FFC (likelihood of weekly FFC, OR: 0.82; 95% CI: 0.66, 1.02) (38). In another study, neither ease of access nor fastness of FF had a significant association with FFC among Latina women (33).

TABLE 2.

The association between Americans’ perceptions of FF and FFC¹

Studies/references	Association between FF perception and FFC	Direction of FFC changes
1. Lorts and Ohri-Vachaspati (38)	FFC was perceived as bad for weight loss (those who attempted weight loss were less likely to have weekly FFC, OR: 0.77; 95% CI: 0.62, 0.97).	↓: perceived FF as bad for weight loss
2. Oexle et al. (40)	High perceived FF availability (OR: 1.04; 95% CI: 0.73, 1.50 vs. low FF availability) and having ≥1 FF outlet present ≤1 mile from home in GIS analysis (OR: 0.64; 95% CI: 0.38, 1.09 vs. none) were not significantly associated with higher risk of weekly FFC.	−: perceived FF availability, GIS-based FF outlet presence around home
3. Gallup daily tracking survey, 2013 (3)	73% of those who eat FF at least weekly believed FF is not good. 81% of those who eat FF less than once a month believed FF is not good.	↓: perceived FF as not good for health
4. Russell and Buhrau (17)	FFC was higher along with perceived health benefits of FF (number of FFC days in the past month per perception score, β ± SE = 0.23 ± 0.03, P < 0.01), but did not differ by perceived health risks of FF (all P values >0.05).	↑: perceived health benefits of FF; −: perceived health risks of FF
5. Dave et al. (37)	Those who perceived FF as convenient (OR: 1.2; 95% CI: 1.1, 1.2 vs. not convenient) were more likely to have FFC >1 time/wk. FFC was not different by perceived unhealthfulness of FF (P = 0.20).	↑: perceived convenience of FF; −: perceived unhealthfulness of FF
6. Larson et al. (32)	1) Girls: The risk of more frequent consumption of FF increased with the degree of perceived time barriers (per 1 SD change, OR = 1.18; 95% CI: 1.04, 1.34) and taste barriers (OR: 1.26; 95% CI: 1.11, 1.44) to healthy eating at baseline; but the risk of more frequent consumption of FF lowered with the degree of concern about health (OR: 0.83; 95% CI: 0.73, 0.95) and self-efficacy for healthy eating (OR: 0.85; 95% CI: 0.75, 0.98) at baseline. 2) Boys: None were significant.	In girls, ↑: having less barriers in time and taste to FF vs. healthy eating; ↓: concerning health issues of FF; ↓: ability to avoid unhealthfulness of FF
7. Ayala et al. (33)	FF restaurant users thought 1) distance, 2) price, and 3) kid-friendly menu/play area were more important, whereas 1) quality of food, 2) familiarity with food options, and 3) service were less important for their restaurant selection compared with other restaurant users (all P values <0.05). FF and other restaurant users did not feel differently about easy access in their restaurant selection.	↑: preferring low price and kid's menu/play area, and FF not far away from home/work; −: preferring easy access of FF
8. Satia et al. (36)	Those having low self-efficacy to eat less fat (vs. being confident) were more likely to have higher FFC (all P values <0.05). FFC was not different by perceived barriers to healthy eating owing to higher price of healthy foods vs. FF.	↓: ability to avoid fat of FF; −: affordable, low price of FF

¹Only 3 studies directly accessed Americans’ perceptions of FF. None of the studies examined the association between FF perceptions and the risk of obesity. ↑, higher; ↓, lower; −, no significant difference; FF, fast food; FFC, fast food consumption; GIS, Geographical Information System.

Reported associations between beliefs about health consequences of FF and FFC were inconsistent by gender. One study reported that FFC was higher in individuals reporting positive perceptions about FF's impact on health [number of FFC days in the past month per perception score about the health benefit of FF, β ± SE = 0.23 ± 0.03, P < 0.01 (17)]. However, FFC did not differ by perceived health risks of FF (P > 0.05) in 2 cross-sectional studies among adolescents and adults (17, 37), whereas the risk of more frequent FFC lowered with the degree of concern about health (per 1 SD change, OR: 0.83; 95% CI: 0.73, 0.95) among girls in a longitudinal study (32).

Nutrition knowledge could influence Americans’ FF perceptions and consumption. About four-fifths of FF restaurant users have an incorrect estimation of daily caloric requirements (34, 35). On the other hand, the mandatory menu labeling in FF restaurants in New York City increased the correct estimation for FF calories among consumers in a low-income neighborhood by 60%, although the correct rate was still low (before menu labeling: 15%; after menu labeling: 24%) (35). The majority of individuals thought FF calorie labeling was important (93%), and that FF calorie information should be posted on the menu next to the food items (85%) (34).

Overall, the link between FFC, FF establishments, and other food restaurants/market options being reported seems inconsistent. One study reported higher FFC among those with a negative perception of the food environment, such as poor produce availability [among individuals nested in census tracts, incidence rate ratios (IRRs) of FFC, IRR: 1.31; 95% CI: 1.19, 1.45], supermarket accessibility (IRR: 1.06; 95% CI: 1.00, 1.11), and grocery quality (IRR: 1.20; 95% CI: 1.12, 1.28) around the neighborhood (31). However, another study found that neither personal perception of FF restaurants in neighborhoods (OR: 1.04; 95% CI: 0.73, 1.50) nor the Geographic Information System (GIS)-based presence of FF outlets within 1 mile of home (OR: 0.64; 95% CI: 0.38, 1.09) was associated with a higher weekly FFC (40).

Factors that might affect Americans’ FF perception and FFC

FF perceptions among obese individuals may be critical for diet behavior change and weight loss efforts. One study showed that obese adults attempting to lose weight consumed less FF than those not trying to lose weight (OR: 0.77; 95% CI: 0.62, 0.97) after adjusting for health care provider's advice to lose weight. However, health care provider's advice to lose weight had insignificant differences on obese individuals’ weekly FFC after adjusting for the respondent's weight loss attempts (38). These findings indicate that FFC was perceived as bad for weight loss among obese adults. It may be that receiving health care providers’ advice to lose weight is not associated with a further reduction in FFC among those who are trying to lose weight, as they already consume energy-dense foods less often. On the other hand, health care providers may not educate patients well about FF as a typical example of energy-dense food, since those having a health care provider's advice but no weight loss attempts did not show a significant difference in FFC in that study.

In addition, different levels of FFC across age, gender, and race/ethnicity may indicate demographic differences in FF perceptions. Most of the studies consistently found that the young (33, 36, 40) [young parents (39)], males (32, 37, 41), African Americans and Hispanics (41), the employed (33, 40), the unmarried (36, 40), obese or physically inactive people (36), or people who were not attempting to lose weight (38) were more likely to have higher FFC (or ate more frequently in FF restaurants) than old, female, white, unemployed, married, non-obese, or physically active people or people who were attempting to lose weight. In contrast, 1 study found a higher likelihood of weekly FFC in white compared with African American, Hispanic, and other race/ethnic groups (40); but no significant sex difference in FFC among African Americans (36) (Table 1).

Discussion

Our systemic review found that only a relatively small number of studies (n = 13) have examined Americans’ perceptions of FF, and only 1 was a longitudinal study. Some of them assessed how Americans’ perceptions about FF might associate with FFC, but none examined whether or not Americans’ perceptions of FF affect obesity risk. Americans had both positive and negative beliefs and attitudes about FF. However, thus far, only 1 study has reported such specific results; and it was based on a phone survey. Future studies are needed in this field.

A 2013 nationally representative phone survey of about 2000 American adults found that about 22% of Americans considered FF good, whereas 28% thought FF was “not good at all” (3). The others (about 50%) thought FF was “not too good”. Some of these perceptions had significant associations with the level of FFC. Those who perceived FF as convenient and beneficial to health, preferred the taste of FF compared with healthy foods, or had low self-efficacy for eating less fat were more likely to frequently eat FF than their counterparts. Americans who consumed less FF seemed more likely to view FF negatively. However, inconsistent findings were shown in the association between perceived FF health risks and FFC. Perceived FF availability in neighborhoods was not associated with FFC.

Our systematic review found that Americans do not have good knowledge of FF calories, or daily caloric requirements (34, 35). However, two-thirds of Americans thought the food served at FF restaurants is not good for them, and even 73% of weekly FF consumers did so (3). Nonetheless, Americans continue to consume FF and think of FF as one of their casual dining options (42). FF images on television and frequent exposures to FF may induce people to perceive FF more positively by appealing to the perceptions of low cost, good taste, and convenience and desensitizing viewers to the possible health risks of FF. One study found that heavy television viewers perceived more positive health consequences but fewer health risks of eating in a FF restaurant every day than others (17).

Furthermore, perceived health benefits of FF (17) were associated with higher FFC, although perceived health risks of FF did not result in a significant difference in FFC in 2 studies (17, 37). One study concluded that educating the public about the health risks of FF may not bring either people's attention to or desired changes in dietary behavior (37). Not just the convenience, entertainment, and socializing opportunities of FF (32, 33, 37), but also Americans’ low self-efficacy to eat less fat and avoid the unhealthfulness of FF (32) may influence FFC. However, African Americans and Latina women did not take into account low price (36), easy access, and fastness of FF (33) when choosing a FF restaurant; instead the taste and familiarity of FF guided their choices.

Previous studies have noted that the presence of FF in the neighborhood was not associated with individuals’ FFC (43, 44), although higher BMI was found among subjects living with FF restaurants in the neighborhood (40). Yet, 1 study found that FF outlet density in low-income neighborhoods was positively associated with FFC (45). The negative perception of other food environments (e.g., poor produce availability and supermarket accessibility, and poor grocery quality in the surrounding area) was also associated with greater FFC (31). We speculate that the conflicting research findings could be due to several reasons. First, methodological issues in Oexle and colleagues’ study (40), e.g., the conventional study design, examining a selective overweight/obese population, and not a using rigid boundary for FFC near the home, could have biased the findings. Second, individual perceptions and awareness of food environments are not fully captured by GIS-based data (31, 46). Third, results may indicate that perceiving a short distance to FF availability in the neighborhood is not the major factor Americans use to decide their amount of FFC. Compared with others in developing countries who were exposed to FF later, Americans are less sensitive to nearby FF availability (47). Also Americans’ typical food shopping behaviors by car (48) may lead them to feel no barriers to visiting relatively faraway FF restaurants. A study found that Americans travel an average of 2.6 (SD: 3.7) miles from home to food establishments (49).

A longitudinal study in the United States reported that instead of FF availability, FF price had a significant effect on adolescents’ BMI (11). Another study reported that a 10% higher FF outlet density in the neighborhood indicated by zip code was associated with a 0.3% higher frequency of children's weekly FFC; but a 10% higher FF price resulted in 5.3% lower FFC (all P values <0.01) (41). Overweight/obese people can be more vulnerable to the FF availability in their neighborhood than others regarding their FFC (41). Only overweight/obese children had a sensitivity to FF availability (indicated by FF restaurant density) influencing their FFC compared with non-overweight children (β ± SE = 0.08 ± 0.03, P < 0.01 compared with −0.40 ± 0.33, P > 0.05) (41). Obese people may purchase more calories when they go to FF restaurants than non-overweight individuals do (41).

Future research is needed to confirm the association between perceived FF availability and FFC in the United States. Regulating accessibility of FF chains in the United States might not be as effective as raising FF price for obesity interventions (11, 15, 50), although the sustainable effects of health taxes on reducing consumption of targeted products and related harms are not known (51). Demonstrating quick and convenient preparation of nutritious alternatives (e.g., healthier prepacked meals) to FF (37) or providing restricted or guided choices in food outlets (52) could be more promising interventions for the obesity epidemic in the United States as shown in intervention studies.

Our reviewed research findings have several limitations. The quality of the studies reviewed was relatively low, e.g., predominately based on convenience sampling, local, cross-sectional, and race/ethnicity-based study designs, which may bias the results due to under/overrepresentation of certain groups. Also, the definition of FF used varied across the studies, and some of the studies did not directly measure FF perceptions. Cross-sectional data cannot assess causal relationships. None of the studies examined the association between FF perception and obesity risk.

Nevertheless, to our knowledge, our study is the first to systematically examine people's perception of FF and its association with FFC in the United States, and provides useful insight into obesity intervention strategies considering FF. To improve research into Americans’ FF perception and its impact on FFC, future studies should use longitudinal study designs with representative samples to examine the causal relationship between FF perceptions and FFC, and directly measure Americans’ perceptions regarding FF. For example, questions should assess whether Americans think FFC is good or bad for health and should be limited, or whether Americans think frequent FFC increases the risk of obesity and reducing FFC helps with weight loss. As shown here, perceptions may be more proximal drivers of FFC behavior than external factors, such as FF availability. For example, changing the perceived importance of calorie counts on menus for weight loss strategies may be needed to complement the environmental changes of posting the information. This would allow for the development of behavioral intervention strategies regarding food choice dictated by positive/negative perceptions, knowledge and attitude, and food environment/prices.

In conclusion, only a small number of studies have examined Americans’ perceptions of FF and their associations with FFC. None have examined how they might link to obesity risk. Americans have both negative and positive perceptions of FF, and are likely to purchase more FF as they valued the convenience and taste of FF, and preferred kid's menus and play areas in FF restaurants over other eating places. Those who consumed less FF seemed more likely to view FF negatively. Further studies with longitudinal study designs and standardized direct measures of FF perceptions are needed.

Notes

Supported in part by research grants from the US NIH (U54HD070725), the Eunice Kennedy Shriver National Institute of Child Health & Human Development (NICHD, 1R01HD064685-01A1), the USDA Agricultural Research Service (3062-51000-051-00D), and the National Aeronautics and Space Administration (NASA) Mission X program. The content of the paper is solely the responsibility of the authors and does not necessarily represent the official views of the funders.

Author disclosures: JM, LJ, HX, JK, and YW, no conflicts of interest.

Supplemental Tables 1 and 2 are available from the “Supplementary data” link in the online posting of the article and from the same link in the online table of contents at https://academic.oup.com/advances/.

Abbreviations used:

FF

fast food

FFC

fast food consumption

GIS

Geographical Information System