Abstract
OBJECTIVES. The objectives of this study were (1) to determine fast food restaurant prevalence in hospitals with pediatric residencies and (2) to evaluate how hospital environment affects purchase and perception of fast food.
METHODS. We first surveyed pediatric residency programs regarding fast food restaurants in their hospitals to determine the prevalence of fast food restaurants in these hospitals. We then surveyed adults with children after pediatric outpatient visits at 3 hospitals: hospital M with an on-site McDonald’s restaurant, hospital R without McDonald’s on site but with McDonald’s branding, and hospital X with neither on-site McDonald’s nor branding. We sought to determine attitudes toward, consumption of, and influences on purchase of fast food and McDonald’s food.
RESULTS. Fifty-nine of 200 hospitals with pediatric residencies had fast food restaurants. A total of 386 outpatient surveys were analyzed. Fast food consumption on the survey day was most common among hospital M respondents (56%; hospital R: 29%; hospital X: 33%), as was the purchase of McDonald’s food (hospital M: 53%; hospital R: 14%; hospital X: 22%). McDonald’s accounted for 95% of fast food consumed by hospital M respondents, and 83% of them bought their food at the on-site McDonald’s. Using logistic regression analysis, hospital M respondents were 4 times more likely than respondents at the other hospitals to have purchased McDonald’s food on the survey day. Visitors to hospitals M and R were more likely than those at hospital X to believe that McDonald’s supported the hospital financially. Respondents at hospital M rated McDonald’s food healthier than did respondents at the other hospitals.
CONCLUSIONS. Fast food restaurants are fairly common in hospitals that sponsor pediatric residency programs. A McDonald’s restaurant in a children’s hospital was associated with significantly increased purchase of McDonald’s food by outpatients, belief that the McDonald’s Corporation supported the hospital financially, and higher rating of the healthiness of McDonald’s food.
There were 195133 “limited-service restaurants” in the United States in 2004.1 Commonly called fast food restaurants (or “quickservice restaurants” by the industry), they had sales in excess of $135 billion in 2005,2 having increased steadily since 1990.3 At fast food restaurants, the food is paid for before eating, whereas full-service restaurants have waitstaff take orders from seated patrons who pay after eating.1,2 Fast food is found in virtually every American community; along highways; and in airports, shopping malls, office buildings, museums, and hospitals. Fast food companies attract children by including toys with meals, operating playgrounds at their restaurants, and choosing locations that are frequented by children. In Chicago, fast food restaurants cluster near schools.4
The prevalence of childhood obesity has been increasing in recent years.5–7 Overweight now is the most common medical condition of childhood,8 with approximately one third of US children aged 2 to 19 years overweight or at risk for overweight in 2003–2004.7,9 Obesity increasingly is prevalent among black and Hispanic children.5,7,10 Consumption of fast food is considered one of several contributors to the overweight epidemic.6,11,12 Fast food is eaten by 30% to 50% of children on any given day, and on days when they eat fast food, children consume 420 to 525 kJ (100–125 kcal) more than on days without fast food.11,13
Many hospitals embrace fast food companies in their marketing strategies. Hospitals establish on-site fast food outlets to promote their physical environments, attract patients, and boost revenue.14 Among “Honor Roll” hospitals listed by US News & World Report as America’s best hospitals in 2001, 38% had fast food restaurants on their premises.15 Fast food outlets are perceived as sites for quick, low-cost food with familiar, reliable menus for outpatient visitors and employees.16,17
Marketing aims to improve the perception of a corporation or its products through association with a desirable outcome, activity, image, or location. Branding, a form of marketing, uses corporate imagery, logos, trademarks, names, and/or associated characters (eg, Tony the Tiger) on a place, product, or event (eg, US Cellular Field, Ford Expedition Eddie Bauer, Macy’s Thanksgiving Day Parade, respectively) to enhance the reputation or recognition of the corporation and/or its products. Linking a product with socially acceptable or desirable settings is intended to increase product acceptance and sales. The presence of fast food restaurants in hospitals places fast food in a milieu that is perceived as credible and beneficial to good health14 and serves to market the fast food brand.
We designed and conducted 2 related studies to explore relationships between hospitals and fast food restaurants. Study 1 examined the prevalence of franchised or brand-name fast food outlets at US hospitals that sponsor pediatric residency programs. Study 2 compared purchase and perceptions of fast food at children’s hospitals by surveying adults who were accompanied by children as they were exiting 3 Chicago-area children’s hospitals: one with an in-hospital McDonald’s franchise, one with McDonald’s branding, and one with neither a McDonald’s franchise nor branding.
METHODS
Study 1
The department of pediatrics office or the residency program office at all 200 US pediatric residency programs listed in the 2001–2002 Accreditation Council for Graduate Medical Education’s Graduate Medical Education Directory18 was contacted in 2002–2003. A departmental or program staff member or a chief resident provided responses by telephone, fax, or e-mail to questions about hospital and program demographics, the presence of any fast food restaurants in the hospital, and affiliation with a Ronald McDonald House, a temporary local residence for families of children who are receiving treatment at children’s hospitals. There are 245 Ronald McDonald Houses in 26 countries, each owned and operated by a local chapter of Ronald McDonald House Charities. Ronald McDonald House Charities receives support from McDonald’s Corporation.19 This survey was approved by the Institutional Review Board of Children’s Memorial Hospital.
Study 2
Adults who were accompanied by children after pediatric outpatient visits were surveyed at 3 Chicago-area children’s hospitals. Each hospital has a pediatric residency program and provides all levels of care from primary through tertiary care, including various transplant services. Each has an affiliated Ronald McDonald House. The hospitals represented 3 distinct conditions: (1) hospital M had a McDonald’s restaurant on site; (2) hospital R did not have a McDonald’s restaurant on site but has prominent branding, as it is named for McDonald’s trademark character, Ronald McDonald, and has McDonald’s corporate imagery throughout the facility; and (3) hospital X did not have a McDonald’s restaurant or any reference to McDonald’s on site. None of the hospitals had franchise or brand-name fast food facilities other than McDonald’s, and each had its own cafeteria. Surveys were distributed on 2 afternoons at each hospital between December 2002 and April 2003 and were completed by a convenience sample of adults who were accompanied by children as they were leaving the hospital. All respondents were at the hospital for a pediatric outpatient service and were accompanied by at least 1 child who was aged ≥2 years. Only surveys that were completed by adults who were the parent, guardian, or other relative of a child with them that day were analyzed. Surveys were available in Spanish and English. The institutional review board of each participating hospital approved this study.
Surveys were anonymous, had 21 questions, and took ∼5 minutes to complete. Questions included demographic information concerning the adult and accompanying children. Race/ethnicity, self-assigned by each respondent, was included because it has been associated with the prevalence of overweight and obesity5,7,9,10 and frequency of fast food consumption.20 Two questions asked directly about purchase of fast food on the day of survey. The first was, “Did you buy food from a fast food restaurant today?” Those who responded affirmatively checked any of 10 named restaurant options or “other.” The second question asked, “Did you eat or purchase McDonald’s food today?” and, if yes, “Was the food you purchased today from a McDonald’s located in the hospital?” Other questions determined the last time “you gave your children food from a McDonald’s restaurant” (choices were today, yesterday, within the past 2–7 days, within the past 30 days, >30 days ago, and never), how often the children eat fast food, use of rewards for good behavior that day, familiarity with the hospital, and reasons for eating McDonald’s food that day.
“Buying or eating fast food” on the survey day was defined by a positive response to either of the 2 questions that asked whether the respondent bought fast food today or ate or purchased McDonald’s food today or by choosing “today” as the last time they gave their child food from McDonald’s. “Buying or eating McDonald’s food” on the survey day was defined by a “yes” response to bought fast food today with McDonald’s as the designated brand or location or a positive response to either of the 2 McDonald’s-specific items noted above.
The survey included an unmarked line ∼12.5 cm in length. The words “Not healthy” were written at the left end of the line, and “Healthy” was written at the right end. Respondents were asked to indicate their perception of the healthiness of McDonald’s food by placing a mark on the line. Inadvertently, a translation error on the Spanish version of this question substituted the words “En una escala del 1 al 10…” (translation: “On a scale of 1–10…”), for the English version, which stated, “On the scale below… .” Therefore, this item was scored by dividing the line into 10 equal parts (1 = least healthy; 10 = most healthy) and noting the area within which the mark fell or by using the number written on the line by some of the Spanish-speaking respondents. We divided respondents into 2 groups at the median health rating, yielding a lower health rating (scores of 1–3) and a higher health rating (scores of 4–10).
Data Analysis
Data were analyzed using SPSS 12.0.0 (SPSS, Inc, Chicago, IL). Significance was set at P < .05. In study 1, analyses included χ2 and Mann Whitney U tests to assess factors that were associated with the presence of on-site fast food restaurants. In study 2, Mann-Whitney U, χ2, or Fisher’s exact tests, as indicated, were used to evaluate associations. Logistic regression analyses were conducted to examine associations with the dependent variables “McDonald’s health rating” and “buying or eating McDonald’s today.” We report adjusted odds ratios (AORs) with 95% confidence intervals (CIs). Factors that were considered in analyses included hospital site, parental education, race/ethnicity, insurance, frequency of buying fast food, age group of oldest child (2–4 years, 5–10 years, or 11 years or older), and number of children present (1 or >1). McDonald’s health rating also was considered in the model that predicted buying or eating McDonald’s today. Hospital site and all independent variables that were significant at P < .1 were entered into multiple logistic regression models. Final models showed good fit using the method of Hosmer and Lemeshow.21
RESULTS
Study 1
A representative at each of the 200 US pediatric residency programs was contacted; item response rates varied from 183 to 200, because some respondents were not familiar with all of the information requested at their hospitals. The programs had 12 to 126 pediatric residents (median: 36; 25th percentile: 21 residents; 75th percentile: 48 residents; n = 199 responses to this item). Their hospitals (n = 183 responses) had a median of 106 pediatric beds (range: 12–715 beds; 25th percentile: 60 beds; 75th percentile: 187 beds). Fast food restaurants were reported to be present in 29.5% (59 of 200) of the hospitals, with McDonald’s alone in 17 (8.5%), McDonald’s and another brand in 5 (2.5%), and other brands alone in 37 (18.5%) hospitals. On-site fast food was more prevalent at hospitals with more pediatric beds (n = 183; median: 150 beds in hospitals with fast food versus 101 beds in those without fast food; Mann-Whitney U, P = .013). Similarly, hospitals with fast food on site had larger residency programs (n = 199; median: 42 residents in hospitals with fast food versus 36 residents in hospitals without fast food; Mann-Whitney U, P = .023). A Ronald McDonald House was affiliated with 120 (60%) of 200 hospitals, but the presence of a Ronald McDonald House was not significantly associated with a McDonald’s restaurant in the hospital (χ2, P = .196).
Study 2
Study Population
A total of 404 surveys were completed by adults with children who were ≥2 years old leaving the 3 children’s hospitals. Eighteen were excluded because the adult who completed the survey was not a parent, guardian, or relative. A total of 386 surveys were analyzed: 179 completed at hospital M, 107 at hospital R, and 100 at hospital X. Most (88%) surveys were completed in English. Respondents at hospital X were more likely to be visiting that hospital for the first time (hospital X: 53%; hospital M: 14%; hospital R: 6%; χ2, P < .001). Response rates for individual items varied minimally and was >97% for most items, except where noted.
There were significant differences in the race/ethnicity and education of the respondents at the 3 hospitals (Table 1), but they did not differ with respect to insurance status or frequency of fast food consumption. In addition, the respondents did not differ by hospital in age (mean: 36 years ± 9 years), percentage who were the mother (83%), number of children present (67% 1 child, 23% 2 children, 10% 3–5 children), or the age group of the oldest child present (32.2% 2–4 years, 36.4% 5–10 years, 31.4% ≥11 years).
Demographic Characteristics of Respondents by Hospital (n = 386)
Marketing
Most hospital M respondents (88%) correctly reported that there was a McDonald’s restaurant at that site, whereas few respondents at hospital R (5%) and hospital X (8%) thought (incorrectly) that there was a McDonald’s restaurant at those sites. Significantly more respondents at hospitals M (78%) and R (73%) than at hospital X (44%) thought that McDonald’s Corporation was a financial supporter of the hospital (n = 359 respondents; χ2, P < .001).
Use of Fast Food Restaurants
Overall, 42% of respondents bought fast food on the day of the survey, and McDonald’s was chosen by 82% of fast food purchasers. There were significant hospital-associated differences in the purchase of any type of fast food and purchase of McDonald’s food among the respondents on the day of the survey (Fig 1). At hospital M, 56% had fast food and 53% had McDonald’s food on the day of the survey, whereas purchase of any fast food and McDonald’s food was considerably lower at the other hospitals. Among the 164 respondents with fast food restaurant use on the day of survey, McDonald’s was the restaurant used by 95% of hospital M respondents (at hospital R, 48% used McDonald’s, and at hospital X, 67% used McDonald’s; χ2, P < .001). Moreover, 83% of McDonald’s purchasers who visited hospital M bought their food at the McDonald’s restaurant in the hospital. Among the 111 (84%) McDonald’s visitors who responded to the question asking why they chose McDonald’s on the survey day, 46% (39 of 87) at hospital M indicated that location was important; location was less important among McDonald’s buyers at the other hospitals (hospital R: 9%; hospital X: 27%; Fisher’s exact test, P = .032). Child’s preference was more likely to have influenced purchase of McDonald’s among those who exited hospitals R (64%) and M (46%) than for those who exited hospital X (13%; Fisher’s exact test, P = .017). Other influences that were listed less commonly were price (14%), to reward the child (12%), toy or prize (9%), and menu selections (6%).
Reported purchase of fast food by hospital outpatient visitors on survey day. ▪, hospital M; 
, hospital R; □, hospital X. aP values using χ2 statistics.
We used logistic regression to examine influences on consumption of McDonald’s on the survey day (Table 2). Respondents who visited hospital M were 4 times more likely than those at hospitals R and X to have bought McDonald’s food on the day of survey (AOR: 4.1; 95% CI: 2.0-8.4).
ORs of Variables Associated With Buying or Eating McDonald’s food on Survey Day (n = 340)
Perception of Fast Food
Respondents rated their perception of the healthiness of McDonald’s food; 341 surveys could be scored from 1 (not healthy) to 10 (healthy). This question was answered by 72% who took a Spanish-language survey and 91% who selected an English-language survey (χ2, P < .001). Most (86%) respondents who rated the healthiness of McDonald’s food gave it a rating between 1 and 5, which was closer to “not healthy” than to “healthy” (Fig 2). We divided respondents into 2 groups using the median rating: those with scores of 1 to 3 (49% of respondents) and those with scores of 4 to 10 (51%). Respondents at hospital M were twice as likely as those at hospital X to rate the healthiness of McDonald’s food above the median (logistic regression, AOR: 2.1; 95% CI: 1.1–4.1); respondents’ scores at hospitals R and X did not differ (Table 3). Frequency of fast food consumption, race/ethnicity, and education had AORs similar to the AOR for hospital in the perception of healthiness of McDonald’s food.
Health rating of McDonald’s food by hospital outpatient visitors at the 3 hospitals combined (n = 341). Respondents were asked to place a mark on a line labeled “Unhealthy” on the left and “Healthy” on the right. See Methods for full description of rating scale.
ORs of Variables Associated With Perception of McDonald’s as More Healthful (Score 4–10 on Healthy Scale; n = 334)
There were no significant interactions between hospital site and other variables in the logistic regression model that evaluated fast food use or the model that evaluated perception of fast food. Restricting the analyses to those who completed the survey in English did not alter the results (data not shown).
DISCUSSION
We found that fast food restaurants were reported in 29.5% of hospitals that sponsor pediatric residency programs, a figure that is similar to that reported previously in general hospitals.14 We also present the first documentation that the presence of a McDonald’s restaurant in a children’s hospital was associated with (1) increased fast food purchasing by parents, (2) the belief that McDonald’s Corporation was a hospital benefactor, and (3) more positive perceptions of the healthiness of McDonald’s food.
The proportion of parents who reported going to a fast food restaurant on the survey day at hospitals R and X (29% and 33%, respectively) was similar to fast food consumption for all US children: 30% of children aged 4 to 19 years in 1 study.11 Others have reported higher consumption rates, from 42% to 50% of children, depending on age, on any given day.13 Fast food, specifically McDonald’s food, consumption rates were nearly twice as high among respondents at the hospital with the on-site McDonald’s restaurant compared with visitors who were surveyed at the other 2 hospitals. Most of the increase was accounted for by increased purchase of McDonald’s food; respondents at hospital M were 4 times more likely to buy McDonald’s on the survey day than those at hospital X. Convenient location and child preferences were reasons to go to McDonald’s, making restaurant placement in a pediatric hospital particularly attractive to parents. McDonald’s branding (hospital R) in the absence of a restaurant was not associated with increased consumption of McDonald’s or other fast food.
All 3 Chicago hospitals, along with 60% of US hospitals that sponsor pediatric residency programs, had a Ronald McDonald House nearby, a benefit to patients, patients’ families, and communities. The presence of a Ronald McDonald House was not associated with a hospital-based McDonald’s restaurant in our survey of pediatric residency programs. However, in the Chicago outpatient study, having either an on-site McDonald’s restaurant or McDonald’s branding was associated with increased perception that the McDonald’s Corporation supports the hospital financially (in fact, McDonald’s Corporation or Ronald McDonald House Charities provides support to all 3 hospitals). This link suggests that a beneficent association between McDonald’s and the hospital is marketed when a McDonald’s restaurant or branding is present.
Hospitals and hospital-based nutritionists have a role in promoting healthy lifestyles and food choices.15,16 Fast food restaurants in children’s hospitals may have an impact on patient and family adherence to nutritional advice. Fast food is associated with consumption of more calories, fat, and carbohydrates and reduced intake of fiber, milk, fruits, and vegetables.11,13,22 Among adolescent girls, fast food consumption is an important determinant of diet quality20 and can have an impact on the rate of weight gain and risk for obesity.23 Nevertheless, visitors to the hospital with the on-site McDonald’s restaurant had a more positive view of the healthiness of McDonald’s food than did visitors to the 2 hospitals without McDonald’s on site, providing additional evidence of successful marketing.
The family has a powerful influence in shaping and maintaining the eating habits and food preferences of its members.24,25 The American Academy of Pediatrics recommends that health supervision include encouragement to parents to promote healthy eating patterns and model healthy food choices.26 Food choices also are affected by ethnicity and acculturation.27,28 The actual food choices that are made by parents and children in hospitals probably depend on the availability and the promotion of both healthy and less healthy foods in the context of family experience, expectations, and culture. Unfortunately, fast food restaurants traditionally have not focused their menus on the types of foods that are found most commonly in healthy eating plans. Attempts to introduce more healthy options sometimes have had lackluster success, whereas marketing of new “super-sized,” high-calorie options are prominent.29
Convenient location and child preference influenced many parents to buy McDonald’s food. This choice may be modulated by reluctance to seem less caring or to enforce rules after a stressful visit or a difficult medical procedure. It also may be related to parental stress or feelings of guilt and the need for parents to comfort themselves and their children with familiar food. In any case, parents may have difficulty helping a child make healthy choices when faced with the ready availability of fast food in the hospital. Moreover, because the presence of McDonald’s on site was associated with perception of greater nutritional value of the food, parents in that setting may be more willing to purchase McDonald’s food.
The American Academy of Pediatrics recommends development of social marketing (the application of commercial marketing techniques to promote voluntary behaviors that are intended to improve personal and/or community well-being) to promote healthy food choices.26 However, the presence of fast food restaurants in pediatric hospitals markets dietary choices that are contrary to desired health messages.29 In the case of hospital-based fast food, promotion of health may be secondary to other reasons to locate fast food restaurants at the hospital, including promotion of the hospital, improving the financial performance of the hospital and its outpatient food service, or other concerns.14
We have shown that the presence of a fast food franchise in a hospital positively influences parental perceptions of fast food. We do not know what food was actually purchased by any of the respondents on the survey day; it is conceivable that some fast food buyers chose more healthy items. Our survey was created for this investigation, and it has unknown validity and reliability. Finally, we gathered responses from individuals at 3 hospitals, but because hospital was the independent factor of interest, the analysis did not account for the clustering of individuals by hospital.
CONCLUSIONS
Fast food restaurants are found in 29.5% of US hospitals with pediatric residency programs. The presence of a McDonald’s restaurant in a children’s hospital significantly increased the purchase of fast food on the day of a hospital outpatient visit, mostly as a result of purchase of food at the in-hospital McDonald’s. Moreover, the presence of an on-site McDonald’s restaurant was significantly associated with a more positive perception of McDonald’s food and with the belief that McDonald’s Corporation is a financial supporter of the hospital.
Fast food restaurants are ubiquitous. Therefore, we think it unlikely that removing fast food restaurants from children’s hospitals (which in this study amounted to only 59 of the >195000 fast food restaurants nationwide1) would directly alter overall fast food availability or intake in the United States, but the presence of fast food franchises in hospitals contributes to parental perceptions of fast food quality at a time when childhood obesity is increasing. In the midst of increasing rates of childhood obesity, efforts are needed to balance the dietary choices and social messages consequent to the presence of fast food restaurants in children’s hospitals. The presence of fast food restaurants in children’s hospitals should be reconsidered.
Acknowledgments
This work was supported by a Senior Resident Research Project Grant to Dr Sahud from the Children’s Memorial Institute for Education and Research (now the Children’s Memorial Research Center).
We thank Roopa Seshadri, PhD, for statistical consultation and Joel Frader, MD, for critical review of the manuscript. Elise Grenier, MD, assisted with survey distribution at the 3 Chicago-area children’s hospitals. Jonathon Guryan, PhD (University of Chicago Graduate School of Business), and Clarissa Yeap, PhD (University of Minnesota Department of Applied Economics), helped with acquisition of restaurant industry data.
Footnotes
- Accepted August 10, 2006.
- Address correspondence to Robert R. Tanz, MD, Children’s Memorial Hospital, 2300 Children’s Plaza, Box 16, Chicago, IL 60614. E-mail: rtanz{at}northwestern.edu
The authors have indicated they have no financial relationships relevant to this article to disclose..
Presented in part at the annual meeting ot the Pediatric Academic Societies; May 3, 2004; San Francisco, CA.
Dr Sahud’s current address is Department of Pediatrics, Allegheny General Hospital, 320 N East St, Pittsburgh, PA 15212.
REFERENCES
- Copyright © 2006 by the American Academy of Pediatrics
In this issue
Jump to section
Related Articles
Cited By...
- Downsizing: policy options to reduce portion sizes to help tackle obesity
- It's time to ban junk food on hospital premises
- Health Care Providers' Perceived Role in Changing Environments to Promote Healthy Eating and Physical Activity: Baseline Findings From Health Care Providers Participating in the Healthy Eating, Active Communities Program
- The Food and Beverage Vending Environment in Health Care Facilities Participating in the Healthy Eating, Active Communities Program