Research Study Background

A. Purpose

The "TrEAT Yourself Well" program was a social marketing campaign and research study designed to encourage the selection of healthier meals by consumers in Southern California restaurants. The project was carried out between January 1999 and December 2001 by Juárez & Associates, a Los Angeles-based research company, and Accents On Health, a San Diego-based nutrition company that had been promoting restaurant nutrition for twelve years through their "Healthy Dining" program. Additional team members included Dr. Paul Patterson, agricultural economist at the University of Arizona at Mesa and his colleagues, Social Marketing Consultant Regino Chávez, and the Tom Feltenstein Neighborhood Marketing Institute of Palm Beach, Florida. The social marketing interventions and promotions were implemented in San Diego during a fifteen-month period between March 2000 and May 2001.

Sponsored by the State of California, Department of Health Services’ Cancer Prevention and Nutrition Section, the program was intended to contribute to the effectiveness and impact of ongoing California DHS campaigns (including the California 5 a Day Campaign, California Project LEAN and the California Nutrition Network for Healthy, Active Families) to prevent cancer and other life-threatening diseases by reducing nutrition-related risk factors. The originating funding source was the National Cancer Institute.

Improving the nutritional content of restaurant meals is becoming increasingly important because, according to research, the frequency of dining out is on the upswing, especially in California -- now averaging four to five meals per week. Busy schedules, travel, two-career families, business entertaining, socializing, celebration, relaxation and convenience have made dining out a way of life for time-starved consumers. Unfortunately, most restaurant meals contain insufficient quantities of fruits and vegetables and are high in fat. In this campaign, a healthy entrée was defined as including two or more servings of fruits and/or vegetables and containing under 30 percent of calories from fat (or under 20 grams of fat).

Restaurateurs do not consider it their responsibility or role to generate demand for healthy menu items. Rather they offer what they believe consumers want and leave it up to their patrons to make their selections. Restaurateurs believe that better-educated and/or health-conscious patrons will find the items they desire on the menu and will order them if they wish. But consumers have little chance of making informed decisions because even nutrition-savvy diners face an almost impossible task of deciphering which choices are in fact healthy, due to hidden fats and poor availability of appealing fruit/vegetable dishes and options. Health educators see the need for providing more accurate and complete information, not merely the subtle clues that the industry favors (using such words as "fresh," "seasonal" and "delightful").

The goals of this program, then, were

• To provide target consumers with information to create awareness of healthy menu choices in restaurant settings
• To position healthy menu entrées as positive choices among consumers
• To increase customers’ positive attitudes regarding the purchase of healthful menu items
• To increase the percentage of sales of healthy menu selections in selected chains

The target consumer was defined as a frequent restaurant diner who eats out at least three times per week, between the ages of thirty and seventy, likely middle-to-high income and likely female. Specific sociodemographic variables were determined by the type of chain implementing the project since, at the outset, a variety of concepts (restaurant classifications or themes) was considered desirable.

B. Approach/Implementation

The research design called for several restaurant chains, in which sales of identical menu items could be compared between experimental (intervention) sites in San Diego and control sites located outside of the San Diego campaign area. Within each chain, both experimental and control locations offered the same healthy menu items which were promoted in the same way on the menus. A multifaceted social marketing campaign was implemented in San Diego, including media advertising, in-restaurant and neighborhood promotions, and networking with health professionals as well as various health organizations.

Juárez & Associates, the primary grantee, was responsible for grant administration, developing the social marketing plan, conducting and evaluating the focus groups, collecting diner surveys and restaurant sales data, and submitting all reports. Accents On Health, the major subgrantee, was responsible for coordinating the campaign with the restaurant marketing departments and managers, and implementing the campaign in the San Diego locations. Statistical analysis and evaluation of survey and sales data were carried out by Dr. Paul Patterson and colleagues. The Tom Feltenstein Neighborhood Marketing Institute acted as consultant on restaurant promotions.

Shortly after the grant’s award, appropriate forms were submitted to the Committee for the Protection of Human Subjects of the State of California to gain approval for the consumer research. The project was granted exemption status, due to the fact that research subjects could not be identified, either directly or through links, and any disclosure of their responses outside the research could not reasonably place them at risk of criminal or civil liability or damage their financial standing, employability or reputation.

Six California-based restaurant chains, which were current clients of Accents On Health, originally agreed in writing to work with the California state-sponsored campaign to offer and promote healthier menu choices during the fifteen-month period between March 1, 2000, and May 31, 2001. They were offered the benefits of advertising and promotions for their San Diego units in exchange for their cooperation with recipe development/modification, instituting in-restaurant promotions, granting permission to conduct research (customer surveys) in the stores, and access to sales data during the campaign. Confidentiality or non-disclosure agreements were required by several corporate entities to obtain the latter.

The participating chains included

Restaurant A, a family-style Mexican concept

(Two experimental locations in San Diego, two control locations elsewhere in California)

Restaurant B, a quick-service California-style chicken and fresh vegetable concept
(Two experimental locations in San Diego, two control locations elsewhere in California)
Restaurant C, a 1940s style diner
(Two experimental locations in San Diego, one control elsewhere in CA)
Restaurant D, an upscale casual concept
(Two experimental locations in San Diego, two control locations elsewhere in California)
Restaurant E, a pair of fine-dining restaurants with individual names under the same corporate ownership
(Two experimental locations in San Diego, two control locations elsewhere).
Restaurant F, a quick-service Mexican concept
(Two experimental locations in San Diego and two control locations elsewhere in California)

The experimental locations were identified by corporate management as being the most amenable to and appropriate for implementing the social marketing program. During the course of the campaign, Restaurant Chain F dropped out due to a change in corporate ownership and direction, as did one of the control locations for restaurant E, due to a lack of commitment by management.

Upon selection of the restaurants, Accents On Health staff consulted with each of the chefs to identify appropriate menu items for the study, a minimum of six dishes per chain. In some cases, the selected items were ones which already met the study criteria, but in most cases the healthy items chosen by the chefs still required modification, such as the addition of fruits or vegetables and/or reduction in the amount of fat used in preparation. Restaurant B, which had on its menu up to fourteen side dishes that qualified as containing one or more servings of fruits or vegetables and less than 30 percent fat, agreed to designate and promote several TYW combination meals, each of which featured lean meats and two specific healthy sides dishes.

Within each chain or concept, measurements (sales data and customer surveys) were collected in the two experimental San Diego County locations that were subject to the social marketing promotions, and in the one or two control locations situated some distance from the San Diego region. Control locations were selected far away from the San Diego area of dominant influence (ADI) to ensure that diners would not be exposed to area-wide media promotions, i.e., television and magazine advertising, or to the Healthy Dining books that were also produced for the Los Angeles and Orange County markets. (In retrospect, this strategy proved to be difficult to implement and to monitor, as well as unnecessary, and therefore it is not recommended for future campaigns.) All locations offered the healthy items, albeit with varying degrees of visibility, so all were monitored for compliance with the program.

Focus groups were conducted with restaurant management, chefs, waitstaff and frequent diners before the campaign was instituted to refine the methodology and to define the slogan (TrEAT Yourself Well), the logo and the target audience. After the completion of the campaign, additional focus groups were held with frequent diners from the targeted restaurants to explore consumer reaction to the various promotions and the TYW meals.

A social marketing plan was developed in the fall of 1999 by the team’s consultant, Regino Chávez, in preparation for a spring 2000 launch. Before the campaign got under way, a San Diego-based advertising agency (Third Eye Advertising) and a public relations firm (Nuffer, Smith, Tucker, Inc.) were selected through a bidding process to develop a mass media advertising campaign, including assistance in designing, preparing and disseminating the promotions such as television spots and magazine ads. Shortly after the inception of the campaign, color brochures with coupons (e.g., buy one entrée and get one free, or $1.00 off a TYW item) were developed and distribution began through partners in the health community, hospitals, community events, the Accents On Health website and mailings.

The social marketing program implemented in the greater San Diego area included

• In-restaurant and neighborhood promotions, such as special menus, table tents, posters, etc. Additional promotions included waitstaff incentives and contests, a custom newspaper insert and direct mail campaign, samples of the TrEAT Yourself Well dishes distributed at neighborhood apartments and businesses, and frequency/discount cards. Different chains implemented various combinations of the in-store promotion strategies, as discussed below. Menus identifying the healthy dishes were available in both control and experimental locations; all other promotions occurred at experimental locations only.

  Promotion	Restaurant	Dates
  Menus	All restaurants	March 00 – May 01
  Posters	Restaurant B	Aug 00, March - May 01
  	Restaurant C	May – Dec 00
  	Restaurant E	July – Nov 00, April 01
  	Restaurant E	Oct. 00
  Table tents/cards	Restaurant F	July – Oct 00
  	Restaurant E	June – July 00, April 01
  Bingo cards	Restaurant B	April 01
  Takeout flyers	Restaurant D	April 01
  Healthy employee Meals	Restaurant E	March 01
  Server buttons	Restaurant C	April - May 01
  Free food samples	Restaurant D	Aug – Sep 00
  Business blitzes	Restaurant E	March 01
  	Restaurant B	March 01
  Apartment blitzes	Restaurant D	Jan. 24 - April 14, 01
  Community events	Several	March – July 00
  Waitstaff incentives	Restaurant D	Aug 00 (loc.#1), Sep/Oct 00, Mar 01
  	Restaurant C	Oct – Nov 00, April 01
  	Restaurant E	March 01
  	Restaurant E	March 01

• Paid advertising with KFMB-TV (Channel 8) and San Diego Magazine. KFMB produced 15- and 30-second television spots that aired during appropriate time segments, as agreed by the project team. Over1,000 spots were aired as planned, along with additional ones when time became available. KFMB’s programming and news directors also offered opportunities for interviews and programming emphasizing the campaign message.
• Over a twelve-month period, San Diego Magazine featured twelve full-color advertisements in the publication, which has a monthly circulation of approximately 47,000. In addition, the magazine produced table tents for in-restaurant promotions in all participating San Diego restaurants as well as extra column space periodically to promote the TrEAT Yourself Well message.
• Community events (educational seminars and food tastings). Called "Extravaganzas," these events included a seminar entitled "The Fat and Skinny on Restaurant Dining," a discussion about the TYW Campaign, a chef demonstration and a food tasting, as well as instructional materials. Community events were held between January 2000 and July 2000 at San Diego hospitals and medical centers, churches, YMCAs, fitness centers and businesses. They were discontinued in order to focus efforts in the target restaurants.
• Liaisons with health professionals in the community (dietitians, physicians, nurses, health educators, fitness trainers, etc.). Brochures were made available to the public through professional offices, clinics and hospitals. In addition, special gift certificates for a complimentary TYW meal at Restaurants A and D were mailed to dietitians, physicians and other health professionals to introduce the TYW menu offerings, in the hopes that the dishes would be recommended in turn to clients and patients. The gift certificates included a questionnaire on the reverse side requesting feedback on the TYW dishes.
• More than 200,000 copies of the custom newspaper, a four-page full color newspaper-sized promotion piece, were distributed in zip codes surrounding the experimental restaurant locations through inserts in the major San Diego daily newspaper, advertising packets and direct mailings to health-conscious households. These inserts were developed, produced and distributed by Horizon’s Publishing of Jupiter, Florida, under the supervision and assistance of the project team.
• Also part of the campaign was a website (www.treat-yourself-well.com), which was established relatively late, in April 2001. During the entire period between 1999 and 2001, however, Accents On Health’s website (www.healthy-dining.com) devoted several pages to TrEAT Yourself Well and the California 5 a Day program. All advertising and promotional materials encouraged the public to visit the website for restaurant and nutrition information, menu advice, and instructions for obtaining coupons and additional information. The KFMB-TV and San Diego Union Tribune websites also had links to the healthy-dining.com website throughout most of the campaign. At most of these websites, links were made to partnering organizations, including www.ca5aday.com.

C. Progress Relative to Specific Aims

Table 1 on the following page lists the estimated number of people reached by various media and print promotions during each month of the campaign. This table shows that by far the largest numbers of people were reached via television spots. However, since these spots were primarily 30 seconds or less in length, the total impact on viewers would be less for each exposure than from a brochure or seminar through which more information may be communicated. Furthermore, many of the television viewers as well as the magazine readers (in particular, the subscribers) were most probably the same over time, i.e., the same viewers and/or readers were receiving additional exposure to the same ads.

Sales data from the first few months suggested minimal effectiveness of the media and community efforts. To boost sales, the in-restaurant server-incentive promotion was instituted in two chains beginning late in 2000, and this strategy greatly enhanced sales. Additional promotions implemented in early 2001 further increased sales during the final months of the campaign. These included the custom newspaper with coupons, gift certificates to health professionals, the database mailing to restaurant customers, and further waitstaff incentives. In the one chain in which the team was able to obtain post-implementation sales data it was shown that sales generally regressed to pre-implementation levels after the campaign had ended. In some measure, this may have been due to the fact that patrons do not always eat in the same locations month after month, and that new customers were not motivated to try the healthy items even though they were still on the menu.

The level of cooperation from restaurant management ranged from indifference to helpful support. In each case, the project team needed to deal with various staff, from the heads of the marketing departments to the local managers and assistant managers. As the restaurant business operates in a busy and chaotic environment, most of these people were very busy with their day-to-day activities. Yet, in the end, five of the original six concepts generally adhered to their agreements, even when their chain was acquired by another or when management structure was reorganized. For example, during the initial 12 months of the campaign, over half of the corporate contact people left the company or moved to other positions within the company.

TABLE 1. MEDIA AND MAIL PROMOTIONS

D. Evaluation Findings

Customer knowledge and attitudes were measured by means of in-store surveys before, during and after the implementation phase; monthly restaurant sales data provided information on the number and percentage of designated healthy dishes that were sold. Customer data were analyzed to determine 1) changes in customer purchases, knowledge and attitudes, and 2) which marketing or promotion strategies were most closely correlated with increased sales.

Sales data: Table 2, below, displays the number of TYW entrée sales as a percentage of all comparable entrée sales. The sales data from the first few months of the campaign showed that the healthy choices constituted only a small percentage of comparable sales, but varied considerably among chains. Most sales ranged from less than 1 percent to 6 percent during the baseline periods and from less than 2 percent to 8 percent during promotions. The more upscale restaurants and communities appeared to have higher sales, regardless of promotional activity, but this observation was not tested statistically.

Sales data from Restaurant B present a special case. As can be seen on the first page of Table 2, sales of B’s TYW combination items never reached over 0.3 percent, due to the fact that the designated healthy combo meals were not promoted adequately (nor were they even available in some locations at some times) and in fact were dropped from the menus in early 2001. Nevertheless the healthy side dishes were still available at all stores and constituted, in fact, a significant portion of sales. Analysis of these side dishes only (second page of Table 2) proved inconclusive. Because neither the combination meals nor the side dishes were effectively promoted during the campaign, and because these data sets are not equivalent or comparable to data collected from the other restaurants, this chain’s sales could not be used in the final statistical analysis.

In the four remaining chains, the campaign raised the share of TYW items sold in the experimental restaurants from an initial 1.89 percent to 2.61 percent of all items. Thus the campaign produced a 38 percent increase in the sales of TYI items.

Regression analysis of the data was used to determine the effectiveness of the various promotions in increasing percentage share of the TYW menu item sales. Clearly the effectiveness of the promotion strategies varied. General market interventions such as media advertising and community events appear to have been ineffective in increasing sales of the TYW items in the target restaurants. Brochures and the custom newspaper were somewhat more effective, as were gift certificates sent to health professionals and database mailings to the restaurants’ mailing lists. By far the most effective strategy was the waitstaff incentive program, in which servers were paid a monetary incentive for each TYW meal they sold.

TABLE 2. SALES DATA

A blank square signifies that data were not available.

* Sales data include entrées only, not sides (except Restaurant B where sides are included as part of the meal), banquets, a la carte, catering, children's meals, beverages, desserts, comps, takeout/to-go meals, breakfasts, etc.

A "0" signifies that less than 0.1 percent of meals served were TrEAT Yourself Well dishes.

RESTAURANT B: TYW SIDES AS A PERCENTAGE OF ALL SIDES

These results suggest that certain promotional activities may encourage diners to eat more healthfully when dining out. However, the methods employed in this campaign were primarily designed to induce a short-term change in dining habits. For instance, the waitstaff incentive program produced a dramatic increase at the time; however these promotions had little or no carryover effect in the following months at the same locations. In an effort to improve the dietary habits of U.S. consumers, it is essential to find promotional and educational methods that have a lasting impact. It is dubious whether public health administrators could advocate promotion programs that rely on paying waitstaff to promote more healthful items, particularly if there is doubt that they produce sustained changes in diners’ habits.

Customer Surveys: More than 3,200 diners were surveyed by means of a self-administered questionnaire in three waves – before, during and after the implementation phase. Data were collected on Tuesdays, Wednesdays and Thursdays so as to be more likely to reach persons who were regular away-from-home diners as opposed to special occasion weekend diners. Approximately half the surveys were completed during the lunch hours (11:30 – 2:30) and half during the dinner hours (5:30 – 9:00). Trained data collectors approached customers entering the restaurant to ask if they would be willing to complete the survey, and to determine eligibility. Customers approached were determined by choosing every second to fifth diner, depending on the number of customers usually patronizing the individual store at that mealtime. To be eligible, respondents had to be 18 years of age or older and live or work in the general area, so as to exclude business travelers, tourists and others who would not fit the criteria of likely exposure to the media campaign in the experimental area. The data collectors were instructed not to interfere with the customers’ dining experience, the waitstaff, or the efficient functioning of the restaurant. A $5.00 gift certificate or cash incentive was offered to each person who turned in a valid survey.

Demographic information on the survey respondents showed that 57 percent were female, 76 percent were non-Hispanic whites, 11 percent were Hispanic, 6 percent were Asian and 4 percent African American. More than half (57%) had a college education, 46 percent were in the 30-50-year age group, and almost 30 percent had incomes over $100,000. Approximately 33 percent reported dining out six to ten times per month and an additional 26 percent reported dining out three to five times per month. The most common reason for dining out was convenience (29%), followed by business (19%), too busy to shop/cook (18%) and social occasions (17%). In terms of awareness of healthy items, 77 percent of respondents answered that they notice special menu items and 79 percent of those said they have ordered them in the past. Over half (55%) said they tend to order the same menu item(s) repeatedly, suggesting a possible reluctance to try new entrées.

Taste was listed as a significant factor in menu item selection by over 50 percent of respondents. Approximately 30 percent reported health/nutrition as an influencing factor, 18 percent reported placing a usual order, 12 percent listed speed, while only 9 percent indicated cost as an influencing factor. Surprisingly, other factors such as promotions or waiter recommendations were reported by fewer than 10 percent each.

Most of the demographics and reasons for dining out, as well as for choosing particular items, were consistent between the experimental and control locations. However, interesting differences were found during the course of the campaign with respect to diners’ awareness of healthy promotions. By the end of the campaign, significantly more respondents in the experimental locations indicated that they had noticed television, pamphlets, medical facilities and restaurant promotions. These were, in fact, major parts of the social marketing campaign in San Diego. Radio, newspaper and billboard advertising, which were not employed, showed no significant differences in responses between the beginning and end of the campaign, or between experimental vs. control locations.

The responses in the San Diego area were significantly different (P<0.05) with respect to awareness of 5 A Day and TYW, compared to the control locations, and comparing the beginning and the end of the campaign. In the experimental locations, those reporting familiarity with 5 A Day increased from 17 to 25 percent. The control locations also showed a significant increase, from 15 to 20 percent. Respondents indicating a familiarity with the TYW campaign also showed a significant increase (P<0.05) from 5 to 14 percent in the experimental sites and from 6 to 9 percent in the control sites. Presumably, the familiarity indicated in the control locations at the beginning of the TYW campaign was simply noticing the information that explained the criteria of the TYW designations on the menus. Not surprisingly, the familiarity with TYW was less than that of the 5 A Day Campaign, which has been ongoing in California for over twelve years. In terms of ordering healthy menu items, a significantly larger number of diners reported ordering TYW menu items in the experimental (13%) vs. control (7%) locations by the end of the campaign.

Awareness of the campaign had a significant influence on the beliefs consumers held regarding the importance of healthy dining. Consumers who tended to notice specially marked menu items or had purchased these items in the past were more likely to hold favorable beliefs about such dishes. These results suggest that introducing a trial of a healthy entrées can do much to change consumer beliefs about their appeal. Furthermore, these positive beliefs are seen to have a positive effect on consumer attitudes. However, after controlling for the influence of beliefs, it was found that time-pressured, convenience-seeking diners who place strong importance on taste continue to view healthy menu items as less appealing options. Taken together, the results on consumer beliefs and attitudes present some interesting contradictions. While higher income, time-pressured individuals recognize the importance of dining healthfully, they are less likely to view healthy menu items as appealing. This would suggest that although the dietary health message is understood, there is still some reluctance to accept such items in restaurants.

Those diners who do hold more positive attitudes about healthy menu items are more likely to purchase them. Similarly, those who hold nutrition and health as primary factors influencing their decisions are more likely to purchase these items. Consumers in the experimental stores were also found to be significantly more likely to purchase healthy entrées, providing some evidence supporting the direct effect of the campaign. Coupons were also found to be an effective way to induce purchase. In some instances, this campaign may have generated more positive beliefs and attitudes as well as future purchases in additional restaurants.

Indeed, introducing a trial of healthy entrées may prove beneficial in improving consumer beliefs and attitudes about such selections. Upon trying these items, the consumer must find them appealing. Consumers already exhibit generally positive beliefs about healthy menu items, and they certainly are aware of the fact that they should be consuming more fruits and vegetables, but most remain skeptical about ordering healthy items in restaurants, as reflected in their less than favorable predisposition at the moment of decision. Future research might focus on how healthful menu items can be made more universally appealing to consumers and, therefore, more commercially attractive to restaurant owners.

In summary, the results of the surveys showed significant improvements in consumers’ awareness, beliefs and positive attitudes towards ordering the healthier menu selections during the course of the TrEAT Yourself Well campaign. The sales data also demonstrated that consumers’ ordering behavior can be modified by implementing different promotions; for example, the waitstaff incentive programs increased monthly sales of the targeted items as much as eightfold in one chain. Overall, the most cost-effective interventions were the waitstaff promotions, the restaurants’ customer database mailings, gift certificates to health professionals, and the brochures.

E. Recommendations for Future Research

In terms of implementation details, the following are considerations for larger-scale restaurant nutrition projects that promote healthy menu items:

Establish realistic goals and objectives at the beginning of the project, e.g., a 10 percent increase in sales of healthy menu items during promotions, sustainable at a 5 percent increase once the promotions are no longer being implemented.

If possible, ensure that the healthy menu items will be available, unchanged, for the entire time the research is being conducted, both in experimental sites and in control sites, if the latter are utilized. It should be noted that keeping menu items constant enhances data collection and analysis, but precludes the ability of restaurants to use attractive seasonal dishes or add more desirable dishes in response to consumer reactions. It also tends to limit the number of healthy menu choices.

Enlist the commitment of the top corporate management from the beginning of the project. Work with a manageable number of chains and individual locations at any given time, and do not over commit resources and research staff. For logistical purposes, keeping all experimental and control sites within the same general geographical area is preferred. However, if promotions use mass media or materials distributed in locations other than the stores’ immediate neighborhoods, the control locations need to be far enough away that they will not be contaminated by effects of the promotions. While studying chain restaurants enables the research goal of having control locations and therefore an experimental research design, the complexity of working with chain restaurants presents numerous challenges, particularly with regard to the number of decision-makers and the need to adapt strategies to conform with established corporate marketing, sales, employee and accounting policies.

Establish personal relationships with corporate marketing managers as well as with individual restaurant managers and assistant managers. Follow up regularly in support of the objective of encouraging diners to order healthy entrées. Be prepared to work with frequent changes in staff and expect corporate staff turnover. Ensure that both the corporate and the restaurant managers are working with each other vis-à-vis the TYW program, but be sensitive to the fact that restaurant personnel are frequently busy. For example, do not call restaurant managers during the lunch or dinner hours. Be flexible.

In terms of accurately tracking sales, assure that the healthy items are unique, i.e., distinct from “non-healthy” items. Healthy dishes should be included in or attached to the standard menu, not displayed on table tents, posters or boards only, although these may be additional to the menu during promotional periods. Locate the healthy dishes prominently in the menu, not hidden on the last page. Be prepared to deal with changing menus and entrée selections.

Be sure that all promoted healthy items present as good-tasting a product as possible, and be realistic about what nutrition criteria are required. Meals that are low-fat or contain restricted calories are likely to be rejected as too small a portion size or not satisfying enough in terms of taste. Describe healthy items using words that sizzle, that make them sound fresh, tasty and exciting, with bold spices and flavors. Do not overemphasize their healthy aspects, and especially do not use the names of diseases such as cancer and diabetes.

Be prepared to sign confidentiality or non-disclosure agreements to obtain sales data. Ensure that the sales data are available per the desired time period (e.g., monthly) and are usable so as to enable aggregation of healthy items as a percentage of comparable items, if that is the standard to be used. Obtain sales data for several months during a pre-implementation period, when the healthy items are on the menus but are not being promoted, and for at least several months after the implementation period, when all the promotions have been completed.

Work through the restaurants’ current marketing efforts or utilize complementary strategies; consider carefully before introducing unfamiliar strategies. If a store already uses couponing, then healthy items could also be marketed with coupons. If a store rotates waitstaff recommendations, then one period could be devoted to recommending healthy items. If a store provides free samples, then taste tests of healthy items can be offered. Customize the promotions by individual store, if necessary.

To achieve short-term sales goals, focus on strategies that reach individual restaurants’ patrons and local neighborhoods, rather than on mass marketing efforts such as television, radio and area-wide print advertising. Print advertising and coupons can be distributed within the store and/or the neighborhood. For long-term results, more education may be necessary rather than, or in addition to, the one-time use of coupons or server promotions.

Schedule promotions to coincide with the time frame of the sales data; for instance, begin at the first of a given month and end at the end of that or another month if sales figures are available by the month. Spread out or stagger interventions and promotions both by chain and by individual restaurants, so that results can be isolated and analyzed independently, not in conjunction with other promotions, and can be compared with baseline periods as well as with the results from other restaurants. Be aware of spillover effects with other locations in a chain (e.g., coupons targeted for one location might be redeemed at another).

In general, do not schedule promotions for December, since restaurants are extremely busy during that time with parties, catering and special events.

Carefully track individual interventions and promotions by exact method, date, site and personnel involved. Keep a current log of all restaurant staff and contacts. Track visits to websites, if applicable, and set up web hosting to be able to do that. Keep track of all costs incurred per each intervention, including staff time, travel and communications.

Work with partners in the health community if possible; doctors, nurses, hospitals, dietitians, personal trainers, fitness clubs, weight loss clinics, pharmacies, chiropractors, etc., all have contact with many patients/clients each week. These health partners could be effective conduits for informing and encouraging the public about the TYW meals. In addition, state and local restaurant associations should be included as partners.

Monitor regularly using standardized forms and knowledgeable project staff. Ensure that 1) the healthy items are available on the menu and are prepared as specified, 2) the specific promotions are being implemented as agreed upon, 3) the waitstaff are aware of and knowledgeable about the items, and 4) they are coded correctly on the order forms or sales slips. Be prepared to train waitstaff and new managers if needed.

Conduct exit research with patrons who have ordered the healthy entrées to evaluate the degree of satisfaction with the items, including positive and negative reactions.

In summary, the results of this study suggest that diners’ behavior can be modified by various promotion techniques, but much remains to be done to improve out-of-the-home dining habits. Future studies should continue to investigate ways to overcome the negative perceptions of healthy restaurant meals. A campaign strategy should include a combination of promotions to encourage the trial of designated items as well as promotions to encourage longer-term changes in awareness, beliefs, attitudes and ordering behavior.