PERPECTIVES H PRACTICE An outcome evaluaon of afood bank program NANCY COTUGNA, DrPH,RD; CONNIE E. VICKERY, PhD, RD; MARY GLICK, MS ABSTRACT Nonprofit organizations such as the Food Bank of Delaware rely heavily on contributions as they strive to create a hungerfree society. To demonstrate to donors and the public at large the positive difference made by the Food Bank of Delaware, we developed an outcome evaluation method that would measure success in terms other than the amount of food distributed. A detailed list of foods available for 1 month was evaluated item by item in terms of the US Department of Agriculture's Food Guide Pyramid. From this list, we calculated the number of people who could potentially be provided with the minimum recommended number of servings per day for each food group. The highest amount of food distributed was from the Bread, Cereal, Rice, and Pasta Group; food items from this group could potentially meet recommended servings for more than 6,000 persons a day. The food bank distributed the least amount of food from the Milk, Yogurt, and Cheese Group. Food bank programs can use outcome evaluations to demonstrate to contributors and product providers the potential impact of the program and the qualitative and quantitative use of their donations. Results also identify areas for nutrition education and improved menu planning for the provider organizations who distribute these foods. JAm DietAssoc. 1994; 94:888890. he Food Bank of Delaware, one of 185 food banks in the United States (1), is a nonprofit, charitable organization dedicated to creating a hunger-free community. In pursuit of this goal, the food bank distributes food and other grocery items to qualified organizations that feed people in need. This food comes from three sources. The largest source of food is donations from food retailers, manufacturers, wholesalers, churches, and civic organizations. A second source of food is what the food bank purchases - generally staple items that are rarely donated. The food bank also receives food from participating in the Self-Help and Resource Exchange (SHARE) program, an international food cooperative specializing in fresh produce and frozen meat. In fiscalyear 1993, the Food Bank of Delaware distributed 3.5 million lb food to provider organizations who then made the food available to their constituents. It is important to demonstrate to donors and the general public that food banks make a positive difference in the community because, as a nonprofit organization, it relies on contributions to survive. In the past, success has been measured by pounds of food distributed or the dollar value of the pounds of food distributed. Although this kind of measure gives a sense of progress, it measures most accurately the output of product not the impact on the community. Outcome evaluation answers questions about the benefits and beneficiaries of a program. Well-designed outcome evaluations also permit researchers to draw inferences about the meaning of the outcome (2). For nonprofit organizations, measuring impact or outcome is difficult for a variety of reasons. The Food Bank of Delaware, for example, is an open system that has many intervening factors over which there is little or no control (eg, external economic forces such as downsizing of major corporations and food industry emphasis on zero defect, which affects quantity and quality of available foods). Because there are no easy solutions to the social problems of hunger and poverty, outcome evaluation is often overlooked. This becomes apparent during a review of the N. Cotugna and C. E. Vickery are associateprofessors in the Department of Nutrition and Dietetics at the University of Delaware, Newark. M. Glick is presidentand chief executive officer of the Food Bank ofDelaware, Newark. Address correspondence to: N. Cotugna, DrPH, RD, Department of Nutrition and Dietetics, 323 Alison Hall, University of Delaware, Newark, DE 19716-3301. 888 / AUGUST 1994 VOLUME 94 NUMBER 8 literature on food banks which yields little in the way of reported outcome evaluation (3). The outcome evaluation we describe in this article is an attempt to measure the potential nutritional impact of a food bank program on the community it serves. METHODS We obtained computer-generated output detailing the type and amount of product distributed for a single month in 1993 for each of the three Food Bank of Delaware sources- SHARE, purchased products, and donated products. Each food item distributed was broken down into a measurable unit based on label information. When label information was insufficient, we measured the items. For instance, a case of apple juice contains 12 46oz bottles and weighs 48 lb. One unit was considered to be a single 46-oz bottle of apple juice. We then determined the number of food servings that one unit item provided according to the UJS Department of Agriclture (USDA) Food Guide Pyramid (4). For example, the 46-oz bottle of apple juice would provide 7.7 :4-c servings front the Fruit Group. Mixed dishes were analyzed by the respective food groups the dishes comprised. For example, each 12.6-oz can of chicken and gravy provided three 2-oz servings of meat and three 2-oz portions of gravy. This was done for 169 different food items distributed for a single month. The servings for such foods as meats were based on edible portions as obtained from Bowes and Church's Food Values of PortionsCommonly Used (5), and IJSDA's Buying IFood (6). Servings of sweets and fats were also calculated. Portions were determined from servingsize information on product labels. Institutional-size containers were evaluated using guidelines from FoodforFifty (7). The net amount of food servings in relationship to the gross weight of items was also detern lled to compare number of servings to gross pounds of food distributed. The total servings of food provided from each food group was determined for a single month and on a per-day serving basis. A period of 1 month was selected for manageability of the calculations. Purchased products remain relatively consistent frommonth to month as do categories of SHARE foods (eg, 5 lb meat is distributed each month, though the type of meat will vary each month; beef might be distributed one month and pork another). Donated products vary the most from month to month, but past records show that a single month provides an adequate representation of products available on a yearly basis. The final calculation we made was to determine an outcome measure defined as the number of people who could potentially be provided with the minimum recommended servings per day for each food group in Ihe Food Guide Pyramid. This measure was labeled People Provided the Minimum Recommended (PPMR) The figure was derived by dividing the total number of servings distributed per day for each food group by the minimum recommended number of servings for that category. Table 1 Distribution of donated, purchased, and Self-Help and Resource Exchange (SHARE) program foods reported by servings for a month and a day, and People Provided the Minimum Recommended (PPMR) a Food group No. of servings Month Day PPMR Donated foods Milk, yogurt, and cheese Vegetables Fruit Bread, cereal, rice, and pasta Meat, poultry, fish, dry beans, eggs, and nuts 16,124 71.943 190,470 1,029,512 49,930 537 2.398 6.349 34,31 1.664 269 799 3.174 5.720 832 Purchased foods Milk, yogurt, and cheese Vegetables Fruit Bread, cereal, rice, and pasta Meat, poultry, fish, dry beans, eggs, and nuts 12.552 36,654 20,569 132.916 40 783 418 1.222 686 4,431 1 359 209 407 343 738 680 SHARE foods Milk, yogurt, and cheese Vegetables Fruit Bread, cereal, rice, and pasta Meat, poultry, fish, dry beans, eggs, and nuts 0 39,465 12.700 9.208 32,322 0 1,316 423 307 1,077 0 439 212 51 539 aNumber of people who could be provided with the minimum number of recommended servings per day for each food group of the Food Guide Pyramid (4) Table 2 Total servings of food distributed and People Provided the Minimum Recommended (PPMR)a inall three programs for the month monitored Food group Milk, yogurt, and cheese No. of servings Month Day PPMR 28,676 956 478 Vegetables 148,062 4,935 1,645 Fruit 223,739 7458 3,729 1 171 636 39,055 6,509 Meat, poultry, fish, dry beans, eggs, and nuts 123,034 4 101 2.051 Sweets 587 703 19,590 Fats 361 215 12,041 Bread, cereal, rice, and pasta a aNumber of people who could be provided with the minimum number of recommended servings per day for each food group of the Food Guide Pyramid (4). bSweets and fats technica ly are not food groups of the Food Gu de Pyramid so there isno PPMR. RESULTS Table 1 shows the total amount of food servings for each food group distributed for the month, the estimated daily distribution, and the PPMR for SHARE program food packages, donated products, and purchased products. During the month of analysis, 635 SHARE program food packages were distributed. Table 2 summarizes distributions for the three sources combined. The food group from which greatest amount of food was distributed (excluding sweets and fats) was Bread, Cereal, Rice, and Pasta, followed by Fruit; Vegetables; and Meat, Poultry, Fish, Dry Beans, Eggs, and Nuts. The least amount of food was distributed from the Milk, Yogurt, and Cheese Group. When divided into recommended mninhium Food Guide Pyramid servings, we found that it would be possible to provide the largest number of people with Bread Group servings (6,509 persons per day), followed in ,JOUIRNAL OF THE AMERICAN DIETETIC ASSO()(IATION / 889 PRISPECTIVES N PRACTICE descending order by Fruit, Meat, Vegetable, and Milk group servings (478 persons per day). For the month monitored, the food bank distributed 19,590 servings of sweets per day (587,703 for the month) and 12,041 servings of fats per day (361,215 for the month). When included in the total amount of foods distributed, sweets and fats ranked second and third, respectively, behind Bread Group distributions. DISCUSSION/IMPLICATIONS We used measured servings from the various food groups within the Food Guide Pyramid (including sweets and fats) to provide an outcome evaluation measure for the potential nutritional impact of a food bank program on the community. For instance, potentially more than 6,000 individuals per day could receive the minimum recommended number of bread servings with the amount of food that is being distributed to providers each day. We talk in terms of potential because, of course, in reality it is not feasible to measure the actual amounts of foods the provider organizations serve to their clients once the food supply has left the food bank. Indeed, one client may actually be provided with or select three slices of bread at a meal, whereas another chooses or is served only one slice. However, we believe that measuring the potential impact is important because the information can be conveyed to the providers for their consideration in areas such as menu planning. From a nutrition education perspective, a nutrition educator might suggest that more milk products be purchased and incorporated into menus. Portion sizes might also be a topic to address. Sweet foods and fats were the second and third largest quantities of foods distributed by the food bank. Although sweets and fats are considered "empty calories" from a nutrition standpoint, nevertheless they contribute energy, which is important from a hunger perspective. The greatest amount of fats and sweets comes from donated foods, which include large contributions of __ candy, desserts, and cream cheese. The SHARE program package provided only one dessert item. Purchased fats and sweets were solely condiment-type items such as jelly, pancake syrup, and gravy. Therefore, when donors are provided with these data, they may be persuaded to contribute less of these itents anld more front nutritious food groups. The issue then becomes presenting data such as ours to a wider donor audience. Dietitians could replicate our study in their local areas or simply share oulr data with their own local food banks as a way to appeal for appropriate donated foods. The actual calculations for each of the 169 foods offered by the Food Bank of Delaware were laborious to obtain and far too cumbersome to report; however, we are willing to share them with those who might be interested. We hope our effort will encourage others to consider different and creative types of outcome evaluations and report them so that all in the field may benefit from and build on those findings. References 1. Crane R, ed. A Second Harvest milestone. Second Hoar;est Update. 1994; Winter: 4. otiool 2. Coyle S, Boruch RF, Turner CF, eds. EoaluatingAI)S ree( Programs.Washington, DC: National Academy Press; 1991. ng. 1985; 57:92-9i. 3. The growing food bank network. FoodEngineeri 4. Food Guide Pyramid.Washington, DC: Human Nutrition Information Service, US Dept of Agriculture; 1992. Home a(nd Garden Bulletin No. 249. 5. Pennington JA, Church HN. Bowes and Church's Food Values of PortionsCommonly Used. 16th ed. Philadelphia, Pa: .. B. Lippincon.t: 1993. 6. Science and Education Administration. Buying Food. Waslington, DC: UIS Dept of Agriculture; 1978. Home Economics Research Repor No. 42. 7. Shugart G,Molt M.Food/firFift. 8th ed. New York, NY: Macmillan Publishing Company; 1989. _ OF ItEST TO YOU Dietetics profesmios lcawd to pintWcirto i Wld Feed DayIM orld Food Day 1994 will be observed on October 14. This international event provides an ideal focus for dietetics professionals interested in raising awareness in their communities about domestic and international hunger. To plan an event that garners broad-based community involvement and sponsorship, dietetics professionals should organize a committee and enlist the support of community leaders and resources, including city council members, legislators, college presidents, hospital chief executive officers, church leaders, media specialists, sports and fitness personalities, and community groups. The committee can plan a 1-day event for October 14, a series of activities spanning the month of October, or an ongoing program. Following are ideas for local programs: * Collect foodstuffs for a food pantry. * Sponsor monthly "brown-bag" hunger discussions or seminars. * Invite community persons to a "hunger fest"; plan a meal that reflects the value of one meal in food stamps (ie, 71 to 73 cents). · Hold a 1-day "hunger fast" at a university; donate money that would have been used on meals to a local food bank. * Organize a series of hunger talks at work sites, churches and synagogues, or university campuses in your community. (Obtain information on hunger in your community from local food banks, state public health and welfare offices, and Special Supplemental Food Program for Women, Infants, and Children clinics.) 890 / AUGUST 1994 VOLUME 94 NUMBER 8 * Write articles for your local newspaper. The US National Committee for World Food Day is organizing a teleconference that will be broadcast from 12 noon to 3 PM (EST) on October 14. The teleconference topic, Sharing Water: Farms, Cities, and Ecosystems, will highlight public policy debate concerning the increasing demand for water and the deteriorating quality of this resource in all regions of the world. Discussion, which will be led by an international panel of experts, will focus on the countries of the Western Hemisphere. Dietetics professionals across the country are encouraged to organize a group to participate in the teleconference discussion through a local satellite hook-up. For additional information on the World Food Day teleconference or otherWorld Food Dayactivities, contact: Patricia Young, national coordinator; US National Conunittee for World Food Day, 1001 22nd St NW, Washington, DC 20437; telephone, 202/653-2404; fax, 202/653-5760. The American Dietetic Association's Hunger and Malnutrition dietetic practice group (DPG) was organized by dietitians who strive to reduce domestic and international hunger through educational and social service events. Dietitians not currently involved with this group who want to organize a World Food Day event can find out about activities being organized by dietitians in their communities by calling Minkie Medora, MS, RD, chairman of the Hunger and Malnutrition DPG, at 406/728-4100.