oversight

Food Assistance Programs: Recipient and Expert Views on Food Assistance at Four Indian Reservations

Published by the Government Accountability Office on 1990-06-18.

Below is a raw (and likely hideous) rendition of the original report. (PDF)

       c
                      United   States     Genera1 Accounting   Office
                      Report to Congressional Requesters
GA!!0                                                                   s



June 1990
                      F’OODASSISTANCE                                       1
                      PROGMMS
                     Recipient and Expert                                   ‘I
                     Views on Food
                     Assistance at Four
                     Indian Reservations




                  RESl’RltXED--Not to      be released outside the
                  GeneraIAccounting    Omce unIesa apeciflcaUy
                  approved by the OffIce of ~ngresaionaI
                  Rd8tiOIU39


                                  *-rL*
GAO/RCED-90-152
       Resources, Community, and
       Economic Development Division

       B-2360 11

       June 18, 1990

       Congressional Requesters

       This is the second of two reports responding to your request for information on the
       availability and adequacy of food assistance on four Indian reservations: Fort Berthold in
       North Dakota; Pine Ridge in South Dakota; White Earth in Minnesota; and Navajo in Arizona,
       New Mexico, and Utah. Our first report issued on September 29, 1989, discussed the
       availability of food assistance on the four reservations and the nutritional basis of the Food
       Stamp Program and Food Distribution Program on Indian Reservations. This report provides
       recipient and expert views on (1) the reasons why households participate in the Food Stamp
       Program or the Food Distribution Program on Indian Reservations, (2) the effect of these
       programs on hunger and the health of Indians on the four reservations, and (3) the
       availability of nutrition education. It also discusses the characteristics of Indian households
       who most frequently participate in these programs.

       As arranged with your office, unless you publicly announce its contents earlier, we plan no
       further distribution of this report until 10 days from the date of this letter. At that time, we
       will send copies to the Secretary of Agriculture; the Director, Office of Management and
       Budget; tribal officials; and other interested parties.

       This work was done under the direction of John W. Harman, Director, Food and Agriculture
     n Issues, (202) 275-5138. Other major contributors are listed in appendix V.




                                        I
       J. Dexter Peach
i/     Assistant Comptroller General
B-23601 1

List of Requesters       The Honorable Quentin Burdick, Chairman
                         Committee on Environment and Public Works
                         United States Senate

                         The Honorable Daniel Inouye, Chairman
                         Select Committee on Indian Affairs
                         United States Senate

                         The Honorable Patrick Leahy, Chairman
                         Committee on Agriculture, Nutrition, and Forestry
                         United States Senate

                         The Honorable Jeff Bingaman
                         The Honorable Kent Conrad
                         The Honorable Tom Daschle
                         The Honorable Dennis DeConcini
                         The Honorable Tom Harkin
                         The Honorable John McCain
                         United States Senate




                Page 2
Page 3
Executive Summary


             In an August 3, 1988, letter, three Senate committees and seven Senators
Purpose      expressed concerns regarding the effectiveness of public and private
             programs in alleviating hunger and promoting the nutritional welfare of
             residents on Indian reservations. They requested that GAO evaluate the
             availability and adequacy of food assistance on four reservations they
             selected, representing about 25 percent of the total Indian reservation
             population: Fort Berthold, North Dakota; Pine Ridge, South Dakota;
             White Earth, Minnesota; and Navajo-Arizona,      New Mexico, and Utah.

             In response, GAO conducted two studies. The first (GAo/RcED89-177,Sept.
             29, 1989) identified available food assistance and examined the nutri-
             tional basis of the Food Stamp Program and the Food Distribution Pro-
             gram on Indian Reservations (FDPIR). In this second study, GAO obtained
             recipients’ and community officials’ views on the (1) ability of Indians to
             participate in these programs, (2) impact of the Food Stamp Program
             and FDPIR on hunger, (3) diet-related health problems of Indians on the
             four reservations, and (4) adequacy of nutrition education provided by
             the Food Stamp Program and FDPIR.


             The Food Stamp Program and WPIR are the two primary federal food
Background   assistance programs available to Indians on or near the four reserva-
             tions. The Food Stamp Program provides monthly coupons, redeemable
             for groceries to eligible households. As an alternative to food stamps,
             FDPIR provides commodities to eligible Indian and non-Indian households
             located on, and Indian households located near, reservations. The maxi-
             mum food stamp benefit is designed to provide households with no
             countable income an adequate quantity of food. However, most house-
             holds have some countable income and do not receive maximum bene-
             fits; for such households, food stamps are a supplemental food source.
             Similarly, FDPIR benefits are intended to be supplemental.

             To obtain the views of recipients, GAO used a focus group methodology
             to elicit spontaneous and candid discussion of specific topics. However,
             because the results discussed in this report are based on the views of
             selected recipients, they may not be generalized either to other food
             stamp and FDPIR recipients at the four reservations or to participants
             nationwide. Further, because environmental, cultural, economic, and
             other conditions vary on reservations nationwide, the information
             reported for these four reservations may not represent all Indian
             reservations.

             To obtain the perspective of community officials who are familiar with
             the problems of Indian diet and health conditions on the four reserva-


             Page 4           GAO/lKEMO-162   Recipient   and Expert   Viewe on hlian   Food Assistance
                          Jkecutive   Summary




                          tions, GAO convened panels of social service providers, program officials,
                          and health care professionals to obtain their expert views on these mat-
                          ters. As with recipients’ views, our results based on the views of these
                          community officials cannot be extrapolated.


                          According to the collective views of the health care providers and offi-
Results in Brief          cials interviewed, some hunger exists on all four reservations. For those
                          participating in federal food assistance programs, according to panelists,
                          hunger is more common among Food Stamp Program households than
                          IJDPIR households for two reasons. First, the Food Stamp Program’s
                          administrative requirements can contribute to breaks, losses, or vari-
                          ances in benefits. Second, allotment levels are too low to buy an ade-
                          quate low-cost diet. They also indicated that hunger exists in households
                          who cannot qualify, or are discouraged from applying, for assistance by
                          the administrative and eligibility requirements of the Food Stamp Pro-
                          gram. These requirements included household composition, income, and
                          asset limits and the application process itself.

                          Recipients and panelists were also concerned that the limited variety
                          and poor quality of some FDPIR foods and limited nutrition education
                          contribute to diet-related health problems, such as diabetes, prevalent
                          on the four reservations. Nutrition education that addresses the nutri-
                          tional needs of the general population, as well as diet-related health
                          problems, is offered on a limited basis through FDPIFL Little, if any, is
                          provided through the Food Stamp Program.

                          Improving the nutritional status of Indian households depends on many
                          economic and social factors. While it is difficult to fully address them
                          through federal food assistance programs, improving the accessibility
                          and quality of diet and the availability of nutrition information can help
                          to alleviate hunger and more effectively serve individuals with diet-
                          related health problems.


Principal Findings

Food Stamp Program        FDPIR  recipients at three reservations told GAO that they could not qual-
                          ify under stringent food stamp eligibility criteria because their house-
Participation Obstacles   hold composition, income, or resources made them ineligible for benefits
                          (this issue was not discussed at the fourth reservation). A frequently
                          cited obstacle was that households exceed the Food Stamp Program’s




                           Pae    5             GAO/acnrscrl52   Recipient   md Expert   Viem   on Indian   Pd   Ashmnce
                         Executive   Summary




                         resource limit because of the vehicle asset limit of $4,500, which has not
                         been changed since it was established by the Food Stamp Act of 1977.
                         Introduced on February 27, 1990, H.R. 4110 would raise the limit to
                         $5,500 for the period January 3 1 to September 30, 199 1. On October 1,
                         1991, and each year thereafter, the limit would be adjusted to reflect
                         changes in the Consumer Price Index.

                         The lengthy and complex application process and excessive verification
                         requirements of the Food Stamp Program discourage households from
                         participating. Furthermore, according to recipients and panelists, the
                         distant location of and lack of reliable transportation to food stamp
                         offices on the four reservations present other barriers to participation.
                         They noted that these obstacles contribute to hunger by preventing or
                         discouraging households from getting the needed food assistance.


Hunger Identified as a   Food stamp and FDPIR recipients at all four reservations told GAO that
Problem for SomeFood     they had few means beyond federal programs of obtaining food. Accord-
                         ing to panelists, because of high unemployment, most reservation house-
Assistance Recipients    holds rely heavily on federal food assistance programs for their dietary
                         needs. However, panelists considered food assistance benefit levels inad-
                         equate to buy a low-cost diet.

                         Panelists also believe that hunger is more common among food stamp
                         recipients than FDPIR beneficiaries because food stamp recipients have
                         difficulty in complying with the program’s monthly reporting require-
                         ments. As a result, they experience delays or losses in their benefits
                         because of failure to complete or errors in their monthly reports or are
                         terminated from the program for untimely reporting. Also, because food
                         stamp benefit levels for those with changing incomes are based on a
                         prior month’s income, they may not accurately reflect a household’s cur-
                         rent food assistance needs.

                         According to recipients, Food Stamp Program benefits are inequitably
                         distributed among reservation households. They told GAO that benefits
                         often understate the needs of households with older children. The food
                         stamp benefit further understates households’ needs, according to some
                         panelists and recipients, because it does not consider high reservation
                         food prices, and transportation costs to grocery stores. An official of the
                         U.S. Department of Agriculture agreed that the distribution of food
                         stamp benefits is affected by family composition, food prices, and trans-
                         portation costs.




                          Page 6               GAO/RCEDfBlSZ   Redpient   and Expert   Views on Indian   Food Assistance
                         Jhxutive   Summary




FDPIR Foods Can          Diets low in fat and salt and high in a variety of fruits and vegetables
Aggravate Diet-Related   can help minimize diet-related health problems at the four reservations.
                         While panelists agreed that the Food Stamp Program can better accom-
Health Problems          modate these special needs, they stated that FDPIR is less likely to
                         because the food package is high in fat and salt. Further, the consistent
                         absence of many commodities, particularly fruits and vegetables, and
                         the inedibility of other commodities limit the nutritional variety of the
                         package.


Nutrition Education Is   Although nutrition education is a component of both programs at the
Limited                  four reservations, according to recipients, FDPIR participants received
                         some nutrition information while Food Stamp Program participants
                         received little or no nutrition information. In addition, the type and
                         amount of nutrition education provided to FDPIR recipients varied by res-
                         ervation. Nutrition education activities, provided through both pro-
                         grams, ranged from written materials and monthly lectures to cooking
                         demonstrations and one-on-one counseling. These activities were most
                         evident at the Navajo Reservation. According to health officials,
                         expanded education services, tailored to reservation Indians, are needed
                         on the four reservations.


                         To enhance the overall effectiveness of the Food Stamp Program and
Recommendations          FDPIR in meeting the nutritional needs and diet-related health concerns
                         of Indian households, GAO recommends the Secretary of Agriculture (1)
                         monitor the availability of FDPIR commodities, (2) review the quality of
                         FDPIR commodities and determine improvements needed, and (3) empha-
                         size the importance of nutrition education and ensure that adequate
                         education services are provided to Food Stamp Program and FDPIR recip-
                         ients. In addition, GAO has made prior recommendations discussed in
                         chapter 5, which, if adopted, should help address some of the problems,
                         such as monthly reporting, faced by reservation food stamp recipients.


Agency Comments          during its review. These officials generally agreed with the report’s
                         findings and suggested several, technical changes that have been
                         included where appropriate. In accordance with the requesters’ wishes,
                         GAO did not obtain official written agency comments on this report.




                         Page 7               GAO/RCED9&162   Redpient   and Jkpert   Views on Indian   Food Assistance
Contents


Executive Summary                                                                                                 4

Chapter 1                                                                                                        10
Introduction             Food Assistance Programs
                         Previous GAO Report
                                                                                                                 10
                                                                                                                 12
                         Objectives, Scope, and Methodology                                                      12

Chapter 2                                                                                                        17
Obstaclesto              Application and Qualifying Criteria Viewed as Complex
                              and Stringent
                                                                                                                 17
Participation in the
Food Stamp Program
Chapter 3                                                                                                        24
Hunger and Diet-         Continuing Dependence on Federal Food Programs
                         Hunger Identified as More Common Among Food Stamp
                                                                                                                 25
                                                                                                                 26
Related Illnesses
              --              Recipients Than Beneficiaries of FDPIR
Continue for Those in    Additional Problems Affect the Elderly and Certain                                      32
                              Children
Food Assistance          Commodity Program Not Designed to Address Nutrition-                                    33
Programs                      Related Problems

Chapter 4                                                                                                        39
Nutrition Education Is   Nutrition Education Can Help Prevent and Treat Health
                              Problems
                                                                                                                 39
Limited                  Nutrition Education Is Limited and Varies by Program                                    41
                         Suggestions for Expanding Nutrition Education                                           43

Chapter 5                                                                                                        45
GAO Observations,        Hunger
                         Diet-Related Concerns
                                                                                                                 45
                                                                                                                 48
Conclusions, and         Recommendations to the Secretary of Agriculture                                         49
Recommendations
Appendixes               Appendix I: Participants of Panel Meetings at the Four                                  50
                             Reservations
                         Appendix II: Demographic Characteristics of Indian                                      52
                             Households Participating in FDPIR




                         Page 8          GAO/RCED9O-152   Redpient   and Ekpert   Views on Indian   Food Assistance
         Contents




         Appendix III: Demographic Characteristics of Indian                                   55
             Households Participating in the Food Stamp Program
         Appendix IV: Methodology and Sampling Plan for Food                                   59
             Stamp and FDPIR Demographics
         Appendix V: Major Contributors to This Report                                         63

Tables   Table 2.1: Distances to Food Stamp Offices Serving the                                20
             Four Reservations
         Table II. 1: Estimated Percentages of 1988 FDPIR                                      53
             Packages Provided to Households With or Without
             Children
         Table 11.2:Estimated Percentages of 1988 FDPIR                                        53
              Packages That Went to Different Household Types
         Table III. 1: Estimated Percentages of Fiscal Year 1988                               56
              Food Stamp Issuances That Went to Indian
              Households, by Different Household Types
         Table 111.2:Extent to Which the Maximum Food Stamp                                     58
              Benefit Provided the Household Food Standard, by
              Household Type
         Table IV. 1: FDPIR Packages Distributed in 1988 and                                    60
              Sampled
         Table IV.2: Estimate of Total Packages Issued in 1988 at                               60
              the Navajo Reservation
         Table IV.3: Sample of Indian Households Located on or                                  62
              Near Pine Ridge Reservation, South Dakota, and
              Navajo Reservation in Arizona and New Mexico




         Abbreviations
                    Aid to Families With Dependent Children
         FDPIR      Food Distribution Program on Indian Reservations
         FNS        Food and Nutrition Service
         GAO        U.S. General Accounting Office
         IHS        Indian Health Service
         TFP        Thrifty Food Plan
         USDA       U.S. Department of Agriculture
         WIC        Special Supplemental Food Program for Women, Infants, and
                        Children
                    Food Research Action Center


          Page 9          GAO/BCED9@152   Recipient   and Expert   Views on Indian Food Assistance
Chapter 1

Introduction


                  A range of issues are associated with the availability and adequacy of
                  federal and nonfederal food assistance on Indian reservations.’ Two pri-
                  mary programs provide federal food assistance to the Indian reserva-
                  tions-the Food Stamp Program and the Food Distribution Program on
                  Indian Reservations (Fur%)-both administered by the U.S. Department
                  of Agriculture (USDA) and authorized by the Food Stamp Act of 1977
                  (P.L. 95-l 13, Sept. 29, 1977), as amended. In September 1989, in
                  response to congressional requesters, we issued a report? which pro-
                  vided an overview of the nutritional adequacy of these two major fed-
                  eral programs and the availability of nonfederal food assistance at four
                  reservations: Fort Berthold in North Dakota; Pine Ridge in South
                  Dakota; White Earth in Minnesota; and Navajo in Arizona, New Mexico,
                  and Utah.

                  This report supplements our September 1989 report by presenting the
                  views of participants in these two programs and those of health care
                  and social service providers on the four reservations. We also profile the
                  characteristics of households that participate in the programs to gain
                  insight into the adequacy of their food assistance benefits. (See apps. II
                  and III.)


                  Participating Indian households on the four reservations rely primarily
Food Assistance   on either the Food Stamp Program or FDP~R for their overall dietary
Programs          needs. Both programs provide eligible households with the opportunity
                  to obtain a more nutritious diet. The Food Stamp Program provides par-
                  ticipants with redeemable food coupons that are intended for the
                  purchase of supplemental foods that they purchase out of family income
                  or other welfare program payments. In general, to be certified as eligible
                  for food stamps, a household must meet income and resource require-
                  ments unless all members receive Aid to Families with Dependent Chil-
                  dren (AFDC)or Supplemental Security Income. Unless exempted for
                  reasons such as age, disability, or current employment, household mem-
                  bers must register for work and comply with the requirements of a
                  training and employment program. In addition, the household must meet




                  ‘An Indian reservation is an area of land “reserved” through treaties, congressional acts, Execuwe
                  Orders, and agreements for Indian use. Reservation land may be owned and occupied by non-Indians,
                  and some reservations have a high percentage of non-Indian land owners.
                                                               acy of Primary Food Programs on Four Indian Reser-




                  Page 10               GAO/RCED-fW152      Recipient   and Erpert   Viewe on Indian   Foad Assistance
Chapter1
Wroduction




several other nonfinancial standards, which include citizenship or eligi-
ble alien status; provide social security numbers; and, if a student, meet
certain criteria.

FDPIR,  which serves eligible Indian and non-Indian households located on
and Indian households located near reservations, is designed to present
an acceptable nutritional alternative to food stamps. It provides benefits
in the form of a monthly food package. Eligibility for and participation
in FDPIR are based on application and certification of reservation or tri-
bal status, income and resource qualifications, and other nonfinancial
factors similar to those of the Food Stamp Program. Also, like the Food
Stamp Program, households composed entirely of AFDCand/or Supple-
mental Security Income recipients, automatically meet income and
resource eligibility requirements.

Depending on individual household characteristics and program criteria,
households may be eligible to participate in one program or the other.
Households that are both food stamp and FDPIR eligible may choose to
participate in either program, but not simultaneously in both. The pro-
grams differ in household definition and income and resource criteria.

Individuals residing, purchasing, and preparing meals together are gen-
erally required to apply for food stamp or FDP~R participation as one
household, and the income and resources of each member are combined
to determine eligibility. However, the two programs differ for related
household members. The Food Stamp Program requires specific related
members residing together to generally apply as one household without
regard to their purchase and meal preparation practices, whereas, FDPIR
allows individuals residing together to apply separately, without regard
to relationship, if they do not purchase and prepare meals with other
household members.

Although both programs provide benefits to low-income households,
income eligibility standards for both programs differ. Participation in
the Food Stamp Program is limited to households that meet both gross
and net income standards, whereas FDPIR household participation is
based solely on net income standards, which are more generous than
food stamp standards. For example, in fiscal year 1990 a four-person
household will be eligible for food stamps if its gross income is less than
$1,3 11 and net income is less than $1,009. On the other hand, to qualify
for FDPIR a four-person household must have a net income that is less
than $1,121.



Page 11          GAO/RcED9015I2   Recipient   and Expert   Viewe on Indian   Food Assistance
                          Chspter 1
                          Introduction




                          Both the Food Stamp Program and FDPIRplace limits on the value of a
                          participating household’s resources. However, the Food Stamp Program
                          limits the combined value of both liquid and nonliquid household
                          resources while F’DPIRlimits only the value of a household’s liquid
                          resources.


                          In our September 1989 report, we determined the following:
Previous GAO Report
                        . The two largest federal food assistance programs serving the four
                          Indian reservations are USDA’S Food Stamp Program     and FDPIR. Three of
                          the four reservations also receive some type of nonfederal food
                          assistance.
                        . The Food Stamp Program and FDPIRare designed to provide recipients
                          with benefits consistent with national dietary guidelines. However,
                          because many factors affect the nutritional value of the food that indi-
                          viduals consume, we were unable to determine the nutritional adequacy
                          of program benefits for specific individuals.
                        . Four major diet-related health conditions exist on the four reservations:
                          obesity, diabetes, heart disease, and hypertension. Although proper
                          nutrition may not cure these conditions, it can reduce their complica-
                          tions or help prevent their occurrence. The Food Stamp Program and
                          FDPIRare not designed to specifically address the special dietary needs
                          of Indian recipients; however, ensuring that program recipients receive
                          and apply adequate nutrition education can help accommodate these
                          needs. Other federal programs are available to Indians on reservations
                          that address the dietary needs of special groups.


Obj-ectives,Scope,and     the Chairmen of the Senate Committees on Agriculture, Nutrition, and
Methodology               Forestry; Environment and Public Works; and the Select Committee on
                          Indian Affairs; six Senators-Jeff Bigaman, Kent Conrad, Tom Das-
                          chle, Dennis DeConcini, Tom Harkin, John McCain-and former Senator
                          Daniel Evans, regarding food assistance on four Indian reservations: (1)
                          Fort Herthold in North Dakota; (2) Pine Ridge in South Dakota; (3)
                          White Earth in Minnesota; and (4) Navajo in Arizona, New Mexico, and
                          Utah.

                           Our specific objectives for this report were to solicit the views of recipi-
                           ents of federal food assistance programs-Food      Stamp Program and
                           FDPIR-and others in the community, knowledgeable about these pro-
                           grams, regarding


                           Page 12          GAO/RCED-B&l52   Recipient   and J3xpert Views on Indian Food Assistance
                    Chapter 1
                    Introduction




                l   the ability of reservation households to participate in the programs,
                l   the impact of the programs on hunger and diet-related health problems
                    that are prevalent on the four reservations, and
                l   the adequacy of nutrition education provided by the programs.

                    Additionally, we agreed to describe the characteristics of Indian house-
                    holds that participate in the Food Stamp Program and FDPIR on the four
                    reservations. (Apps. II and III provide this information.)

                    Because of the widely varying demographic conditions found on about
                    304 federal Indian reservations throughout the United States,3 the infor-
                    mation reported for these 4 selected reservations should not be consid-
                    ered as representative of all Indian reservations.

                    To solicit the views of recipients and others in the community, we con-
                    vened panels of individuals knowledgeable about Indian diet and health
                    and interviewed program participants in focus groups.4 We also inter-
                    viewed responsible federal, state, and local program administrators. In
                    addition, we identified and obtained household demographic character-
                    istic information from probability samples of (1) FDPIR food package
                    issuances and (2) FNSdata of issued food stamp benefits.


Use of Panels       To identify the nutritional concerns at the four reservations and the
                    impact of food programs in addressing these concerns, we convened
                    seven panels of social service providers, program officials, and health
                    care professionals from the reservation communities-one       panel each at
                    the Fort Berthold and White Earth reservations, two at the Pine Ridge
                    Reservation, and three at the Navajo Reservation. More panels were
                    held at the Pine Ridge and Navajo reservations because of their larger
                    geographic size and population. We recruited panel members who, based
                    on their profession or role in the community, were particularly knowl-
                    edgeable about Indian diet and health. A listing of reservation panelists
                    is provided in appendix I.




                    3American Indians Today: Answers to Your Questions, U.S. Department of the Interior, Bureau of
                    Indian Affairs, 1988.

                    4Focus groups are small homogeneous groups assembled to candidly discuss a topic under the con-
                    trolled guidance of a moderator. They are generally viewed as an effective way to capture mqor
                    themes related to a discussion topic.



                     P8ge 13              GAO/RCXD-90152      Recipient   and Expert   Views on Indian Food Assistance
                         Ch8pt.m 1
                         Introduction




Use of Focus Group and   To obtain recipient views on their ability to participate in these pro-
Individual Interviews    grams, how the food programs address nutritional needs, and the ade-
                         quacy of nutrition education, we conducted 10 “focus group” interviews
                         with food stamp and FDPIR recipients-2 food stamp and 2 FDPIR groups
                         at the Navajo Reservation and 1 each for the 2 programs on the other 3
                         reservations.

                         Focus group participants were recruited by the local food stamp and
                         FDPIR offices, and each group had between 8 and 12 participants, for a
                         total of 37 food stamp and 48 FDPIR participants. Because the spontane-
                         ity of the discussion and its focus are facilitated when participants have
                         common demographic or relevant characteristics, only adult females
                         who, for the most part, spoke English were recruited for our groups.
                         Participants ranged in age from 21 to 71 and represented various house-
                         hold compositions, ranging from 1 to 12 members.

                         We interviewed recipients new to food assistance as well as long-term
                         program participants. Food stamp recipients had an average of 6 years
                         of experience with the program, ranging from less than 1 year to 20
                         years. FDPIR recipients had an average of 8 years experience with the
                         present program or its predecessor, the Needy Family Program,s ranging
                         from less than 1 year to 40 years. Additionally, approximately 58 per-
                         cent of the focus group members had at one time or another participated
                         in both programs.

                         Because we used a style of moderation that would spontaneously elicit
                         opinion, specific topics of discussion varied within and between groups.
                         The absence of discussion about an issue implies nothing about its rela-
                         tive importance to the group at a particular reservation.

                         Because the focus group results discussed in this report are based on the
                         perceptions and experiences of selected recipients who may not be rep-
                         resentative of the general recipient population, they cannot be genera-
                         lized either to other food stamp and FDPIR recipients at the four
                         reservations or to participants nationwide.




                          5FDPIR is an outgrowth of the Needy Family program, established in 1936 as a state-admuustered
                          commodity distribution program.



                          Page 14               GAO/RCED9@152      Recipient   and Expert   Views on Indian Food Assistance
                             Ch8pter 1
                             lntroducdon




                             We also interviewed appropriate federal, state, and local administrators
                             of the federal food programs to clarify program rules regarding the pro-
                             cess of applying for and maintaining benefits and to ascertain the differ-
                             ences in local administration of the programs. The results of relevant
                             studies and reports were discussed with appropriate officials.


Use of Probability Samples   To provide information on households receiving FDPIR packages, we took
                             a probability sample of packages issued in calendar year 1988 at each of
                             the four reservations-Fort   Berthold, Pine Ridge, White Earth, and
                             Navajo-and studied the characteristics of the households receiving
                             these packages. (See app. IV for a detailed description of our sample
                             selection process.) Appendix II summarizes the characteristics of house-
                             holds that received FDPIR packages on the four reservations.

                             To provide information on Indian households receiving food stamps, we
                             also analyzed data from Food and Nutrition Service’s (FNS)automated
                             quality control data base on households with at least one Indian mem-
                             ber.6 Our analysis focused on households receiving food stamp benefits
                             in fiscal year 1988 and residing in areas that somewhat approximate the
                             boundaries of the Navajo Reservation in Arizona and New Mexico and
                             the Pine Ridge Reservation in South Dakota, the only areas of the 4 res-
                             ervations with at least 30 records in FNS’data base.

                             We use the term “somewhat approximate” for the following reasons.
                             The Navajo Reservation data in Arizona include three counties, each
                             containing land both on and off the reservation. Because the Navajo Res-
                             ervation in Arizona completely surrounds the Hopi Reservation, the
                             data from Arizona includes both Hopi and Navajo households. The Nav-
                             ajo Reservation data in New Mexico include two counties with land both
                             on and off the reservation and four additional counties that contain
                             Navajo tribal lands that are not part of the major Navajo Reservation.
                             The data for the Pine Ridge Reservation in South Dakota do not include
                             Bennett County-roughly     the southeastern quarter of the reservation.
                             The FNSdata base cannot distinguish between Indian tribes. Also
                             because the local agency/geographic codes did not exactly coincide with
                             reservation boundaries, the FNSdata base may also contain Indian
                             households living off the reservation. (See app. IV for a more detailed
                             discussion of the contents of this data file.) The data provided by FNS

                              6We did not review the internal controls established by FNS for its automated quality control data
                              base. We did, however, review documents describing record layout and defining data elements and
                              compare values of the data we received with documented values. Within this context, we observed no
                              data outside expected ranges.



                              Page 15               GAO/WED-g&l52       Redpient   and Expert   Viewe on Indian   Food Assistance
Chapter 1
Introduction




are from a stratified sample of food stamp issuances (see table IV.3).
Appendix III summarizes the characteristics of households that received
food stamps on these two reservations.

For each estimate based on either the FDPIR or food stamp samples dis-
cussed in this report, the associated sampling error at the g&percent
confidence level is given in parentheses following the estimate. For
example, “23 (+ 7) percent” means that the chances are 19 out of 20
that the true value could be as low as 16 percent-23 minus 7-or as
high as 30 percent-23 plus 7.

We performed our review between March and November 1989, in accor-
dance with generally accepted government auditing standards.




 Page 16        GAO/ECJZDg@152   Redpient   and Expert   Views on Indian   Food Assistance
Chapter 2

Obstaclesto Participation in the Food
stamp       Program




                      Although many Indian households prefer the Food Stamp Program over
                      FDPIR because it allows them to make their own food choices, panelists
                      and recipients told us that many households are discouraged or pre-
                      vented from participating in the Food Stamp Program because of admin-
                      istrative hindrances in applying and qualifying for benefits. In
                      particular, the food stamp application and documentation requirements
                      present obstacles to reservation Indians who may be illiterate or do not
                      keep the type of records needed to verify income and expenses. Further,
                      the distant location, from 30 to 165 miles away, and the absence of relia-
                      ble transportation for some recipients to food stamp offices present bar-
                      riers, even to program application, that could affect a household’s
                      ability to comply in a timely manner with the procedural requirements
                      of the program.

                      Some households cannot qualify for food stamp benefits because of pro-
                      gram regulations concerning either their household composition, income,
                      or resources. While it may be easier, depending on individual household
                      circumstances, to qualify for FDPIR, we were told that many households
                      that are ineligible for food stamps will not apply for commodities. They
                      perceive the eligibility criteria to be the same for both programs and,
                      therefore, believe a rejection by the Food Stamp Program disqualifies
                      them from FDPIR.

                      We were told that others wilI not participate in the Food Stamp Program
                      because they believe the benefit is so low that transportation costs to
                      the food stamp office would eliminate any benefit they might receive.
                      According to most panelists and recipients, these obstacles contribute to
                      hunger by preventing or discouraging households from getting the
                      needed food assistance.


                      Panel members and recipients at all four reservations told us that the
Application and       Food Stamp Program was less accessible than FDPIR because of the some-
Qualifying Criteria   times lengthy and complex food stamp application process and/or the
Viewed as Complex     program’s stringent eligibility criteria. They also mentioned that they
                      encountered transportation problems when going to the food stamp
and Stringent         office. According to FDPIR recipients at the Navajo, White Earth, and
                      Pine Ridge reservations, FDPIR was their only option because they could
                      not meet food stamp eligibility criteria. At the Fort Rerthold Reserva-
                      tion, FDPIR recipients did not indicate that they were ineligible for food
                      stamps but told us that the burdensome application process discouraged
                      their participation in the Food Stamp Program.



                      Page 17         GAO/WED-g@152   Recipient   and m   View   on Indian   Food Assistance
                      chapter 2
                      Obetacles to Participation   in the Food
                      stamp Program




Application Process   At all four reservations, food stamp and FJIPIR recipients described
                      applying for food stamps as a complex and sometimes lengthy process.
                      Panel members told us that the process is cumbersome, in part, because
                      of the complex process involved for determining eligibility. Traveling
                      long distances to the food stamp office also posed significant problems
                      for many applicants.

The Application       To participate in the Food Stamp Program, households, or their author-
                      ized representatives, must file an application form, be interviewed, and
                      provide documents to support their eligibility. Although the application
                      process often cannot be completed the day applicants first visit the
                      office, they can file a partial application to establish a filing date for
                      determining when benefits begin.

                      In the states administering the program for the four reservations, house-
                      holds applying for food stamp eligibility must file a multipurpose appli-
                      cation, which is used to determine eligibility and benefit levels for
                      several assistance programs at once. Although more efficient than sepa-
                      rate applications, the combined application is lengthier and more diffi-
                      cult to complete than a typical food stamp application. The application
                      forms at three reservations ranged from 21 pages at Pine Ridge to 38
                      pages at Fort Berthold, while at the Navajo Reservation, applicants use
                      a shorter form, between 8 to 11 pages, depending upon which of the
                      three states (that serve the reservation) with which the application is
                      filed. In some of these applications, households must provide at least 60
                      pieces of information about household composition, income, living
                      expenses, and assets.

                      The lengthy application form and the time required to complete it were
                      cited as stumbling blocks to participation by recipients and panels at all
                      four reservations. One FDPIR recipient told us that the food stamp appli-
                      cation scared her off after she had sat for hours trying to complete the
                      application. Others shared this sentiment.

                      At the Navajo, Pine Ridge, and Fort Berthold reservations, recipients
                      believed that the verification requirements-the     need for receipts, pay
                      stubs, and other documents to support or verify their statements on the
                      application form- were excessive and could unnecessarily delay deliv-
                      ery of benefits (this was not discussed by the White Earth recipients).
                      Some recipients claimed that they were required to bring in their chii-
                      dren’s report cards (to establish student status), residency testimonials
                      from neighbors, and other forms of verification before their application



                      Page 18                  GAO/ICEDWlSZ      Recipient   and Expert   Views on Indian   Food Assistance
                           Chapter 2
                           Obstacles to Participation   in the Food
                           S-P    Pwwm




                           for food stamps could be approved. Another recipient stated she had to
                           produce documents to verify that her husband was in prison.

                           Further, according to a Pine Ridge panelist, many people who seek emer-
                           gency food assistance from food banks are those who experienced
                           problems completing the food stamp application and as a result had ben-
                           efits delayed up to 4 weeks. The Food Stamp Program requires states to
                           provide benefits to destitute* households within 5 days of their applica-
                           tion. However, the Pine Ridge panelist did not indicate whether those
                           households seeking emergency assistance were eligible for or offered
                           expedited benefits by the local food stamp offices. In an earlier report,2
                           we found indications that eligible households elsewhere in the nation
                           were not always offered expedited benefits.

                           Panel members at the Navajo Reservation explained that the application
                           process is especially difficult for Indian households who may be illiter-
                           ate or who do not keep the types of records required to verify income
                           and expense information on their application. For example, medical
                           expenses for services rendered by a tribal “medicine man” may be diffi-
                           cult to document, according to panelists.

‘lkansportation Problems   Recipients at the Navajo and Pine Ridge reservations noted that the
                           location of the food stamp office, and the absence of reliable transporta-
                           tion to get there, presented barriers that could affect someone’s ability
                           to comply in a timely manner with the procedural requirements of the
                           program (this was not discussed by the White Earth and Fort Berthold
                           recipients). As shown in table 2.1, potential food stamp applicants could
                           have to travel anywhere from 30 to 165 miles one way to get to the
                           office to apply in person.




                           *A household is considered destitute if its expected monthly income is less than $150 and its liquid
                           assets are $100 or less.
                           2Food Stamp Program: Administrative Hindrances to Participation (GAO/RCED-S94, Oct. 21,19&B).



                           P8ge 19                GAO/BcEDB@152 Redpient and Expert Viewa on Indian Food AsdstMce
                                        Chapter 2
                                        Obstacles to Participation   in the Food
                                        stamp Program




Table 2.1: Distances to Food Stamp      - ,,-                                       - - - ,. .
Offices Serving the Four Reservations                                                 Offices sewing the             Range of miles to nearest
                                        Reservation                                           reservation                -- food stamp office
                                        Fort Berthold                                                         5                            30 .fio
                                        Pine Ridge                                                            2         ___--        -     30   50
                                        White Earth                                                           3                            40 .50
                                        Naval0
                                          Anzona                                                              8                            50 - 75
                                           New Mexico                                                         3                           75    120
                                           Utah                                                               3                           30 - 165

                                        Source Local food stamp offmals on the four reservations.


                                        According to panel members at the Pine Ridge and Navajo reservations,
                                        because the Food Stamp Program limits the value of resources] that can
                                        be owned by recipients, many participants do not have cars or have
                                        older, unreliable cars and, as a result, may have to pay neighbors or
                                        others for rides to the food stamp office. Further, according to recipients
                                        at the Navajo and Fort Berthold reservations, the transportation costs-
                                        because of distance and/or payment to others-that      would be incurred
                                        in visiting the food stamp office to apply and gather the required docu-
                                        mentation would eliminate any benefit they might get.

                                        According to local administrators of the Food Stamp Program, food
                                        stamp offices serving the Fort Berthold, Pine Ridge, and Navajo reserva-
                                        tions offer alternative services so that applicants do not have to travel
                                        to the main office. However, satellite offices operate for a few hours on
                                        1 to 4 days a month. For example, they told us that, the McKenzie
                                        County food stamp office, which serves the Fort Berthold Reservation,
                                        has a satellite location on the reservation that is open for 2 hours once a
                                        month. In Jackson County on the Pine Ridge Reservation where 90 per-
                                        cent of all recipients in the county rely on alternative services, the satel-
                                        lite food stamp office is open 2 days a month. At the Navajo
                                        Reservation, local program officials told us that the frequency of opera-
                                        tion of satellite offices depends on the number of clients in an area. They
                                        said that, at a minimum, satellite service is provided 2 days a month at
                                        chapter house locations.

                                         The Hunger Prevention Act of 1988 (P.L. 100-435) authorizes several
                                         administrative improvements to the Food Stamp Program that simplify
                                         the application process and reduce barriers to participation in rural

                                         ‘3Resourcesmclude liquid and nonliquid assets, such as cash on hand, money in checlang and savmgs
                                         accounts, stocks and bonds, licensed and unlicensed vehicles, and recreational property



                                         Page 20                 GAO/RCED9&152       Recipient   and Expert       Views on Indian Food .+.&stance
                       Chapter   2
                       Obstacles to Participation   in the Food
                       stamp program




                       areas. For example, the act requires state agencies to assist applicants in
                       obtaining the appropriate verification and completing the application
                       process. It also grants states permission to waive in-office interviews”
                       and to mail forms to households who live in a location not served by a
                       certification office or have transportation difficulties. Panelists and
                       recipients did not indicate whether in-office interviews were being
                       exempted. However, according to local administrators of the program,
                       households were mailed forms and provided assistance in obtaining ver-
                       ification documents when requested.


Eligibility Criteria   At all four reservations, panelists mentioned that the Food Stamp Pro-
                       gram has more stringent eligibility requirements than FDPIR, which pre-
                       clude or discourage many households from participating. FDPIR
                       recipients at the Navajo, White Earth, and Pine Ridge reservations told
                       us that Food Stamp Program regulations concerning either their house-
                       hold composition, income, or resources made them ineligible for benefits.

HouseholdMembers       Panel members at the Navajo and Fort Berthold reservations mentioned
                       that often extended families who live together cannot qualify for the
                       Food Stamp Program because they are required to apply as one house-
                       hold and their combined income exceeds the eligibility guidelines.

                       As a general rule, for both the Food Stamp Program and FDPIR, all indi-
                       viduals living together and purchasing food and preparing meals in com-
                       mon constitute a “household” and must apply together. The income and
                       assets of all household members are aggregated in determining eligibility
                       and benefits. However, for the Food Stamp Program, specified relatives
                       (i.e., parents, children, and siblings) generally must apply together with-
                       out regard to this “purchase and prepare” rule. Therefore, two adult
                       sisters, for example, who live in one household, but who purchase food
                       and prepare meals separately, must apply for food stamps together.
                       (However, elderly or disabled individuals and parents with minor chil-
                       dren, as well as some others, can apply as separate households.) In con-
                       trast, FDPIR regulations allow related individuals living together, without
                       regard to relationship, who do not purchase and prepare meals in com-
                       mon, to apply separately for benefits.




                       *Waiver of inaffice interviews are to be granted, on request, if a household is unable to appomt an
                       authorized representative and all adult members are elderly or disabled, live in locations not served
                       by a certification office, or have transportation difficulties.



                       Page 21                GAO/RCRIMWlS2        Recipient   and Expert   Views on Indian Foal Assistance
          Chapter 2
          Obstacles to Participation   in the Food
          S-P     Prom




          According to panelists at the Navajo Reservation, it is common practice
          for relatives to live together. However, because these extended families
          may not share income for purposes of buying food, the aggregate income
          of the household does not accurately reflect the household’s ability to
          purchase food or its need for food assistance.

Income    Recipients at all four reservations also mentioned that the Food Stamp
          Program considers certain types of payments, such as tuition assistance,
          as income that are not available for food expenditures. As a result, some
          households exceed the income guidelines. For example, recipients at the
          Pine Ridge and Fort Rerthold reservations told us that portions of edu-
          cational scholarships and grants are included as income, although they
          are not available for food purchases (this was not discussed by recipi-
          ents at White Earth and Navajo).

          According to Food Stamp Program regulations regarding federal educa-
          tion assistance, amounts in excess of tuition and mandatory fee require-
          ments are generally counted as income; with respect to nonfederal
          education assistance, only amounts earmarked for living expenses are
          treated as income. Recipients believe that the excess amount available
          from federal grants should be income-exempt as it is used to pay for
          miscellaneous education expenses such as books, supplies, and
          transportation.

          Recipients at the Navajo, White Earth, and Pine Ridge reservations told
          us that they had chosen FDPIR over the Food Stamp Program because
          Food Stamp Program regulations regarding income calculation made
          their benefit amount so low that it was not worth going through the
          process.

Resourw   Another obstacle to participation for some Indian food stamp appli-
          cants, according to panelists and recipients at the Navajo and Pine Ridge
          reservations and recipients at the White Earth Reservation, is having
          resources, usually a vehicle, with a value that exceeds Food Stamp Pro-
          gram limits. The program sets maximum allowable resources that house-
          holds must meet to be eligible for benefits. The combined value of a
          household’s liquid and nonliquid resources-such as cash on hand,
          money in checking and savings accounts, stocks and bonds, unlicensed
          vehicles, and recreational property-cannot     exceed $2,000 unless the
          household has an elderly member, in which case the limit is $3,000.




           Page 22                 GAO/RCED9@162     Recipient   and b-pert   Views on Indian   Food Assistance
                               Chapter 2
                               Obstacles to Participation   in the Food
                               stamp Ro@am




                               Additionally, non-exempt licensed vehicles” are evaluated for fair mar-
                               ket value, and the portion of the value that exceeds $4,500-an amount
                               established by the Food Stamp Act of 1977-is attributed to the house-
                               hold’s resource level, regardless of any encumbrances on the vehicle.
                               According to a local food stamp official, if a family has a vehicle that is
                               less than 3 years old, the vehicle will in ail likelihood have a value too
                               high for the family to qualify for food stamps.

                               Recipients at the Navajo, White Earth, and Pine Ridge reservations
                               claimed that they all had been denied food stamp benefits at one time
                               due to the value of their vehicles. Some said they resorted to selling
                               their cars to qualify. This created a hardship for their families because
                               poor road conditions, inclement weather, and remote living locations on
                               the reservations make having reliable transportation necessary.


Belief That FDPIR Criter *ia   According to Navajo and Pine Ridge panelists, many households that are
                               ineligible for food stamps will not apply for commodities. Because they
Are Same as Food Stamp         perceive the eligibility criteria to be the same for both programs, they
Program’s                      believe that a rejection by the Food Stamp Program disqualifies them
                               from FDPIR as well. Panelists indicated that outreach activities are
                               needed to improve applicant understanding of the differences between
                               the two programs. Although federal cost sharing for outreach was rein-
                               stated by the Hunger Prevention Act of 1988 to promote informational
                               activities regarding program eligibility, benefits, and the application
                               process, panelists and program officials we spoke to were not aware of
                               whether an increase in outreach activities had occurred.




                               5Licen3ed vehicles exempt from this provision are those wed to produce earned income. or necessary
                               for the transportation of a physically disabled household member, or used as a home, or necessary for
                               long distance travel to employment.



                               Page 23                 GAO/RCTEIM@-162    Recipient   and &pert   Views on Indian Food Assistance
Chapter 3

Hunger and Diet-RelatedIllnessesContinue for
Thosein Food AssistancePrograms

               The Food Stamp Program and FDPIR have helped to alleviate hunger and
               improve the diets of Indian recipients on the four reservations. For
               many Indians, these programs continue to constitute their primary and
               long-term food supply because of persistent high unemployment on the
               reservations. However, because the Food Stamp Program and FDPIR are
               intended to supplement, rather than satisfy, total dietary needs of most
               households, some panel members and recipients believe they may not be
               nutritionally adequate for those Indians who rely completely on them.

               Hunger,’ which continues to be a concern on the reservations we visited,
               is more common among food stamp recipients than beneficiaries of
               FDPIR, according to recipients and panel members. We were told that the
               monthly reporting requirement of the Food Stamp Program contributes
               to hunger because (1) some households experience breaks or losses in
               benefits for procedural noncompliance with this requirement and (2) it
               creates variances in monthly allotments that make it difficult for house-
               holds to plan food expenditures. Further, benefit levels are perceived to
               be insufficient to purchase an adequate low-cost diet because of inequi-
               ties in the benefit calculation and the high cost of living on the reserva-
               tions. In addition, the elderly and children of alcoholic parents may
               experience hunger when the food assistance intended for their consump-
               tion is used by other household members or is exchanged to purchase
               nonfood items.

               The food programs have provided Indian households an opportunity to
               obtain a more adequate diet. However, many diet-related health
               problems, such as obesity, diabetes, heart disease, and hypertension,
               still prevail on the reservations we visited. While neither program is
               designed to address special dietary needs of recipients, the Food Stamp
               Program can better accommodate those with diet-related illnesses.
               According to recipients and panel members, FDPIR, on the other hand,
               contributes to these problems, because of the inconsistent availability
               and poor quality of some foods in the FDPIR package.




               ‘We used a definition of hunger developed by the Food Research Action Center (FRAC), a nonprofit
               organizUion in Washington, DC., for its Community Childhood Hunger Identification Protect. FRAC
               identifies hunger as the lack of resources to obtain food, food shortages in the household. changes in
               focd intake or eating habits of the household (such as skipping  meals), and reliance on a linuted
               number of foods or staples to stretch available rewurces.



               Page 24                GAO/RCEB9@152        Recipient   and Rxpert   Views on Indian Food Assistance
               Chapter 3
               Hunger and Met-Related    Illnesaea Continue
               for Thoee in Food Assistance Progrma




               Although the federal food assistance programs are, for the most part,
Continuing     intended to be supplemental, they are often the primary or only food
Dependenceon   source available to participating Indian households, because of continu-
Federal Food   ing high unemployment on the reservations. This heavy reliance, in the
               absence of other substantial resources, can contribute to hunger among
Programs       food stamp and FDPIR recipients.

               FDPIR is not intended to provide a complete 30day supply of food to
               eligible households and assumes that participants will purchase or rely
               on other resources for a portion of their monthly food supply. According
               to USDAofficials, food stamps are also supplemental benefits for most
               households. While the food stamp benefit is designed to provide house-
               holds with no countable income an adequate quantity of food and nutri-
               ents for an entire month, most food stamp households have some
               countable income, 30 percent of which the program expects to be con-
               tributed toward food purchases. However, food bank officials on the
               panels told us that many of their visitors are food stamp recipients who
               have used up their monthly allotment and have no other resources to
               buy food.

               Food stamp recipients at the Navajo Reservation and FDPIR recipients at
               the Navajo, Pine Ridge, and White Earth reservations indicated that fed-
               eral food assistance is their primary source of food. At the Fort Berthold
               Reservation, FDPIR recipients told us that the FDPIR package is their only
               source of food. Further, recipients at all reservations indicated that they
               had very little income with which to supplement their benefits.

               Our analysis confirms that food stamp and FDPIR benefits were issued to
               households that, on average, have limited gross income2 with which to
               pay for living expenses as well as food purchases needed to supplement
               their food assistance benefits. For food stamp issuances to households
               with any gross income over $0 in the general area of the Pine Ridge and
               Navajo reservations in fiscal year 1988, we estimate that the average
               gross monthly income per person was $102 ( f 7). (We did not have data
               for White Earth and Fort Berthold.) We also estimate that for house-
               holds with some gross income, the average gross monthly income per
               FDPIR household member on the four reservations in calendar year 1988
               was $177 ( f 15) at Fort Berthold, $188 ( + 12) at Pine Ridge, $283 ( + 12)
               at White Earth, and $198 ( f 15) at the Navajo Reservation.

                ‘We computed gross income for a FDPIR household by taking the sum of the household’s earned and
                unearned income before deductions. For food stamp households, we used the gross countable income
                recorded in the FNS automated data base.




                Page 25               GAO/RCEW@l52         Eedpient   and Expert   Views on Indinn Food Amistance
                        Chapter 3
                        Hunger and Diet-Related   Illnesses Continue
                        for Those in Food Assistance Programs




                        According to recipients at the Navajo, Pine Ridge, and White Earth res-
                        ervations, many households supplement their food assistance benefits
                        by depending on food banks for assistance, purchasing food with the
                        small disposable incomes they have, and relying on other federal pro-
                        grams like the Special Supplemental Food Program for Women, Infants,
                        and Children (WIG)and Aid to Families With Dependent Children (MIX).
                        These federal programs, however, are available only to special popula-
                        tions. For example, AFDC is available to eligible needy families with chil-
                        dren; and WIGis available to eligible low-income pregnant, breast-
                        feeding, and postpartum women and children up to 5 years old.

                        FDPIR  recipients at Fort Berthold told us that they try to stretch their
                        benefits by borrowing food, hunting, or relying on a limited number of
                        staple foods, such as macaroni and flour (for fry bread), to feed their
                        families for a portion of the month.

                        Recipients at all four reservations indicated that federal food programs
                        have been a permanent long-term source of food for their families. For
                        example, approximately 60 percent of the food stamp recipients that we
                        met with have been receiving food stamps between 4 and 20 years, and
                        approximately 60 percent of FDPIR recipients we talked to have been
                        using commodities between 4 and 40 years. According to panel members
                        at the Navajo, Pine Ridge, and Fort Berthold reservations, this heavy
                        reliance on food assistance will continue unless unemployment is
                        reduced on the reservations. As indicated by our September 1989 report,
                        unemployment rates on the four reservations ranged from 50 percent on
                        the Navajo Reservation to 79 percent on the Fort Berthold Reservation.


                        According to panelists, hunger is a major concern at the Pine Ridge, Nav-
Hunger Identified as    ajo, and Fort Berthold reservations and a lesser concern at the White
More   Common   Among   Earth Reservation. This information confirms the preliminary findings
Food Stamp Recipients   we noted in our September 1989 report with respect to the Fort Berthold
                        and Pine Ridge reservations. While our earlier study did not indicate
Than Beneficiaries of   that hunger was a problem at the Navajo Reservation, members of all
FDPIR                   three Navajo panels told us that, based on personal observations and the
                        increasing use of food banks, hunger does exist on their reservation.
                        Panelists, particularly food bank administrators, told us that although
                        hunger affects participants and nonparticipants of the federal programs
                        alike, nonparticipating households and food stamp recipients are groups
                        that are severely affected.




                         Page 20               GAO/RCED-9@152      Recipient   and Expert   Views on Indian   Food Assistance
                          chapter 3
                          Hanger  and Diet-Related Rln-    Continae
                          for Thoee in Food Assistance Progmms




                          According to panelists at the Navajo and White Earth reservations, hun-
                          ger is more common among food stamp households than FDPIR benefi-
                          ciaries. This can be attributed to (1) breaks or losses in food stamp
                          benefits due to noncompliance with the monthly reporting requirement,
                          (2) variances in the amount of stamps issued each month, and (3) bene-
                          fit levels that are believed to be insufficient to purchase an adequate
                          low-cost diet.


 Noncompliance With the   Panel members at the Navajo and Fort Ekrthold reservations and food
                          stamp recipients at all four reservations told us that the monthly report-
 Monthly Reporting        ing requirement? of the Food Stamp Program is for many households a
-Requirement
     _       Can Caldse   difficult and unnecessary administrative burden. For example, they said
Breaks or Losses          many food stamp-dependent Indian households experienced breaks or
                          losses in benefits for procedural noncompliance with the reporting
                          requirement.

                          Failure to accurately complete the monthly report or to submit it on time
                          can result in an interruption of benefits or can cause the participant to
                          be terminated from the program and lose benefits during the time it
                          takes to reapply. Although, states may adopt a monthly reinstatement
                          option that can prevent households from having to reapply and lose
                          benefits during a lengthy reapplication process: Minnesota (White
                          Earth), North Dakota (Fort Berthold), New Mexico and Utah (Navajo)
                          are among 13 states that have not done so. To reapply, a participant
                          must complete and file a new application, provide required verification,
                          and appear for an interview.

                          In an earlier report,5 we recommended that FNSencourage the 13 states
                          to adopt the reinstatement option, if practicable. In March 1990, USI%
                          directed the food stamp regional offices to contact these 13 states, as we
                          had recommended, and report the results to FNS by April 30,199O. FNS is
                          currently analyzing the responses.

                          %ates administering the Food Stamp Program CZUIrequire that certain households report and verify
                          income and household circumstances on a monthly basis. This information is used to retrospectively
                          calculate benefits. Every month households subject to this rq uirement must report and provide doc-
                          uments to support income, medical, dependent care, and shelter expenses; household composition;
                          and other circumstances relevant to the amount of the food stamp allotment.

                          4The reinstatement option allows the state to accept a monthly report in the month after it is due and
                          provides the recipient, if still eligible, a full month’s benefits and the opportunity to continue in the
                          program without reapplying.
                          5Food Stamp Program: Participants Temporarily Termiwted for Procedural Noncompliance (GAO/
                          m-81,       June 22, 1989).



                          Page 27                GAO/RCEDBtIl62        Recipient and Jhpert Viem on lndhn Food AaaLetance
Chapter 3
Hunger and Diet-Related   JIinesses Continue
for Those in Food Assistance Programs




Food stamp recipients at the Fort Berthold, Navajo, and Pine Ridge res-
ervations described the difficulty they had had in completing the
monthly report and providing the state-required documentation.
According to Food Stamp Program regulations, if a household has
presented insufficient documentation or documentary evidence is diffi-
cult to obtain, state officials are required to offer the household assis-
tance in obtaining the documents or use a collateral contact or home
visit to obtain the information. According to local administrators of the
program, this service is available if recipients are unable to provide the
documentation after having made every effort to do so.

Also, according to panel members at the Navajo and Pine Ridge reserva-
tions, many food stamp recipients have difficulty completing the
monthly report because they are illiterate or do not understand English
and therefore need assistance.

According to local food stamp officials at all four reservations, to ensure
that the monthly reports are received on time and thereby avoid breaks
in benefits, many recipients deliver their monthly reports in person even
though they are allowed to mail them in. For example, we were told that
at the Navajo Reservation, approximately 50 percent of the Indian food
stamp recipients in McKinley County that report monthly deliver their
reports in person, even though they may have to travel an average of
 100 miles one way to do so.

Some recipients at the Fort Berthold and Navajo reservations have had
their benefits terminated for untimely reporting and have had to reap-
ply to the program. Others at the Fort Berthold, Navajo, and Pine Ridge
reservations have experienced delays of anywhere from 10 days to 1
month in receiving benefits because they did not complete or made
errors in their monthly reports, which must be corrected before food
benefits can be provided. Because recipients are so dependent on food
stamps to supply the large majority of their food needs, delays in receiv-
ing benefits may affect their ability to feed their families.

 Further, delays or suspensions in benefits brought about by the monthly
 reporting requirement create hardships for households who, because
 they are certified for the Food Stamp Program, cannot approach FDPIR
 for assistance. Program regulations prohibit households from simultane-
 ously participating in FDPIR and food stamps, although households may
 from month to month choose to participate in one program and then the
 other.



 Page 28               GAO/RCED-!I@152         Recipient   and Expert   Views on Indian   Food Assistance
                                 Chapter 3
                                 Hunger and Diet-Related   Illnesses Continue
                                 for Those in Food Assistance Programm




Variances in Monthly Food        For those subject to the monthly reporting requirement, benefits are
Stamp Allotment                  adjusted to more closely reflect the current financial situation of the
                                 household and ensure more accurate payments. However, it can result in
                                 hardships for recipients in that it contributes to variances in monthly
                                 food stamp allotments.

                                 Although the program is designed to provide a means of obtaining mini-
                                 mum food requirements through a combination of stamps and income, in
                                 practice this does not happen when the household has fluctuating
                                 income and expenses. For those subject to the monthly reporting
                                 requirement, benefits are based on retrospective income and expenses.
                                 However, we were told that they often do not reflect recipients’ immedi-
                                 ate food needs. For example, a household’s prior month’s income could
                                 be high which would result in a small benefit amount when in actuality
                                 the household may need a larger benefit to make up for a loss of income
                                 or high shelter expense in the current month. According to recipients at
                                 White Earth, Pine Ridge, and Fort Berthold reservations, this mismatch
                                 of income and expenses with benefit amount creates hardships for
                                 many.

                                 Because households have difficulty complying with the monthly report-
                                 ing requirement and this noncompliance affects food stamp benefits,
                                 panelists at the Navajo and Fort Rerthold reservations recommended
                                 that this requirement be eliminated from the program. Recipients at Fort
                                 Berthold also suggested eliminating monthly reporting due to the admin-
                                 istrative burden it places on them and the variances in monthly benefits
                                 created by retrospective budgeting. It should be noted that the Hunger
                                 Prevention Act of 1988 allows states more flexibility in deciding
                                 whether households must report monthly.


Food Stamp Benefit               Both a perceived inequity in the calculation of food stamp benefits and
                                 the reservations’ high cost of living-cost of food on the reservations
Perceived to Be Too Low          and transportation to get to lower cost sources off the reservations-
to Purchase an Adequate          create problems in stretching benefits to obtain an adequate diet.
Low-Cost Diet
PerceivedInequities in Benefit   According to recipients at three reservations, inequities exist in the food
Calculation                      stamp benefit calculation because of unrealistic assumptions regarding
                                 standard household composition and household definition.

                                 Food stamp benefits are designed around the Food Stamp Program’s
                                 standard family of four members- a man and woman 20 to 50 years old


                                 Page 29               GAO/RCED9&152        Recipient   and Expert   Views on Indian Food Assistance
Chapter 3
Hunger and Diet-Related   Illuesaea Continue
for Those in Food Assistance Programa




and children ages 6 to 8 and 9 to 11. In our analysis of Food Stamp
households at the Pine Ridge and Navajo reservations, we found so few
occurrences of families that fit the standard family definition that we
were unable to make a meaningful estimate of the extent to which
households on these reservations matched the definition. Furthermore,
according to recipients at the Navajo and Fort Rerthold reservations,
their food stamps lasted only 1 to 2 weeks because they have older chil-
dren, usually teenagers,” in their family who have larger appetites than
do younger children.

We estimated that 67 ( + 5) percent of the food stamp issuances to
Indian households in the general area of the Navajo and Pine Ridge res-
ervations were at least 5 percent higher or lower than the cost of
purchasing the Thrifty Food Plan diet using nationwide average food
prices. We found that those households with younger children under 6
received benefits equaling or exceeding requirements, whereas those
households with only adults or teenagers received less than required to
purchase a low-cost diet. (See app. III for the details of this analysis.)

Panelists and recipients at the Pine Ridge Reservation also told us that
the food stamp allotment is inadequate because it does not consider the
needs of part-time household members, such as children in weekday
boarding schools who eat at home on weekends. According to Food
Stamp Program regulations, residents of institutions that receive over
half of their meals from the institution are not included as household
members.

Food stamp recipients at the Navajo and Pine Ridge reservations told us
that feeding older children or family members on weekends who are not
included when benefits are determined is a reason for their running out
of food. Recipients at the Navajo and Fort Berthold reservations sug-
gested that a more equitable determination of food stamp benefits could
be made by tailoring benefit levels to the food requirements of house-
holds based on the age of its members. Panelists at Pine Ridge suggested
that more equity could be built into the food stamp benefit by acknowl-
edging part-time meal eaters in the definition of the food stamp
household.




6Teenagers are defined as children ages 12-19.



 Page 30               GAO/RCED-So-162     Recipient   and Jhpert   Views on Indian   Food Assistance
                                 Chapter 3
                                 Hunger and Diet-Related   Ulnesaea Continue
                                 for Those in Food Aasiet~ce    Pmgrama




ProblemsFacedby Indians in       According to panelists at the Navajo and Pine Ridge reservations, Indi-
Stretching Food Stamp Benefits   arts face unique conditions related to the high cost of living on reserva-
                                 tions that erode the real value of the food stamp benefits and require
                                 them to stretch their food dollars more than other recipients do.

                                 Navajo and Pine Ridge panelists and recipients at all four reservations
                                 told us that Indian households cannot purchase as much food with their
                                 food stamp benefit7 as do other recipients because of the high cost of
                                 food on the reservation. According to panelists, food prices are high
                                 because of the cost of transporting food over vast distances and the lim-
                                 ited number of grocery stores on the reservations. To minimize erosion
                                 of food stamp benefits, panelists at the Navajo and Pine Ridge reserva-
                                 tions and recipients at Fort Berthold told us that benefits should be
                                 adjusted for high-cost areas such as the reservations.

                                 Additionally, Pine Ridge and Fort Berthold reservation recipients told us
                                 that since food stamps are issued to everyone on the same day of the
                                 month, grocers can and do increase food prices the week of issuance.
                                 However, food stamp regulations allow states to stagger issuance of
                                 stamps through the fifteenth of the month, provided that each house-
                                 hold receives its coupons on the same day every month. We noted that
                                 only the states serving the Navajo Reservation stagger issuance of
                                 stamps, while at the Pine Ridge, White Earth, and Fort Rerthold reser-
                                 vations, issuance dates are the same for all recipients.

                                 Recipients at Fort Berthold and Pine Ridge told us that often to get bet-
                                 ter value from their food stamp allotments they must travel 50 to 75
                                 miles off the reservation to purchase food at lower prices. Those recipi-
                                 ents who do not have cars or whose cars are unreliable may have to
                                 incur the expense of finding alternative ways to travel to a grocery store
                                 off the reservation. For this reason, the panelists at the Navajo Reserva-
                                 tion suggested allowing a deduction from income for transportation
                                 expenses.

                                 Further according to panel members from the Navajo and Pine Ridge
                                 reservations, the Food Stamp Program does not adequately consider the
                                 high shelter cost on the reservations, such as rent and utilities, and its
                                 impact on disposable household income. The Food Stamp Program
                                 allows an excess shelter cost deduction of up to $177* in fiscal year 1990

                                 7The value of the food stampbenefit is based on the cost of purchming the Thrifty     Food Plan at
                                 current average national food prices.

                                 *This amount applies to households in the contiguous 48 states or the District of Columbia.



                                 Page 31                GAO/RCED9@152        Redpient   and Expert   Viewa on Indian   Food Assistance
                         Chapter 3
                         Hunger and Diet-Related   Illnesees Continue
                         for Thoee in Food Assistance Programa




                         when shelter costs exceed 50 percent of the households adjusted
                         income.g However, panelists believe that the nondeductible portion of
                         shelter costs decreases the amount of disposable income households
                         have available for food purchases. For example, a household with
                         adjusted income of $500 and shelter expenses of $400 can deduct $150
                         of its shelter costs (i.e., $400 less half of adjusted income ($2501). The
                         remaining $250 of shelter expenses theoretically is paid out of the $350
                         of net income (i.e., $500 less $150 of excess shelter costs). As a result,
                         households may not be able to devote the 30 percent of disposable
                         income that the program assumes it can if they have to pay for nonde-
                         ductible shelter costs first.


                         Additional problems exist for food assistance recipients who are elderly
Additional Problems      or for children of alcoholic parents. Panel members at all four reserva-
Affect the Elderly and   tions indicated that many elderly people and children not only experi-
Certain Children         ence a scarcity of food, as do other recipients who lack resources beyond
                         their food assistance benefits, but also go without any food for days
                         because the food assistance intended for their consumption may be used
                         by other household members. For example, Elderly Nutrition Program
                         representatives on the panels at the Navajo, Pine Ridge, and White
                         Earth reservations told us that many of the elderly follow the ways of
                         the traditional Indian family by allowing children and grandchildren to
                         eat first. As a result, they will give a portion of their food assistance to
                         their immediate family and go hungry for a part of the month.

                         Similarly, Child Welfare representatives on the panels at all four reser-
                         vations told us that alcoholic parents will use any cash that the house-
                         hold may have, as well as sell or trade food stamps, to buy alcohol. It is
                         common, they said, to see children left on their own for many days with-
                         out any food. In fact, White Earth panelists indicated that those most
                         affected by hunger on their reservation were children of alcoholic par-
                         ents. Officials from the School Lunch Programs at the Navajo and Pine
                         Ridge reservations also told us that they see many hungry children for
                         whom the only meal of the day is the one they receive from either the
                         School Lunch or Headstart programs.




                          gA~usted  household income is gross income reduced by a standard deduction and a dependent care
                          cost deduction.



                          Page 32               GAO/RCED9@152       Recipient   and Expert   Viewa on hihn   Food Assistance
                         chapter 3
                         Hunger and Met-Related   Ilhsses   Ckmtinue
                         for Those in Food Assistance Programa




                         Although many in the Indian population have the unique opportunity of
Commodity Program        participating in FDPIR as an alternative to the Food Stamp Program, FDPIR
Not Designed to          may not meet the needs of some households, particularly those with
Address Nutrition-       diet-related health problems. Obesity, diabetes, heart disease, and
                         hypertension are prevalent diet-related health problems on the four res-
Related Problems         ervations we visited. Recipients and panelists at the Navajo, Pine Ridge,
                         and Fort Berthold reservations told us that some of the food items
                         offered by FDPIR often contribute to these problems. In contrast, the
                         Food Stamp Program provides recipients with greater freedom to obtain
                         a wider variety of foods, including fresh fruit and vegetables, and to
                         accommodate special dietary needs, such as a need for foods low in fat
                         and/or salt.


Health Problems on the   Obesity, which is primarily caused by an excessive intake of calories
Four Reservations        and a lack of exercise, is perceived to be a major health problem on all
                         four reservations by panelists. For example, an Indian Health Service
                         (II-IS) official at Pine Ridge told us that 65 percent of all clinic patients
                         were obese according to a recent survey.

                         Diabetes is also a major concern on all four reservations. IHSofficials,
                         from the Aberdeen area (which encompasses the Pine Ridge, Fort Ber-
                         thold, and 13 other reservations) were especially concerned with the
                         near epidemic proportions of the illness. We were told that the Aberdeen
                         area leads all other areas nationwide in the number of diagnosed diabe-
                         tes cases, with about 30 percent of all adults over the age of 45 having
                         been diagnosed as diabetic. Moreover, the incidence of diabetes at the
                         Pine Ridge Reservation is six times the national average for people
                         between the ages of 45 and 64, and the Fort Berthold Reservation leads
                         all other Aberdeen area reservations in the number of diabetic cases.

                         In addition to obesity and diabetes, which are common to all four reser-
                         vations, panelists indicated the prevalence of reservation-specific com-
                         plications as well. Panelists from Navajo and Fort Berthold believe that
                         the high incidence of heart disease on both reservations is a complica-
                         tion resulting from obesity, and panelists from Pine Ridge and Fort Ber-
                         thold told us that the high incidence of hypertension on both
                         reservations also results from obesity.


FDPIR Package Lacks      Our review showed that recipients may not be getting the full variety of
                         foods authorized by FDPIR, because (1) many of the food items are not
Variety                  available for recipient selection, (2) the package lacks adequate servings


                         Page 33              GAO/RCED%%152 Recipient   and Expert   Views on Ind&qFood   A4stnnce
                                 chapter 3
                                 Hunger and Diet4Wated     Jllnesses Continue
                                 for Those in Food Assistance Prognuna




                                 of fruits and vegetables, and (8) some of the food items are of poor qual-
                                 ity and are inedible. Although panelists told us that this is a problem for
                                 all FDPIR recipients, it is of special concern for recipients with nutrition-
                                 related problems.

Unavailability   of Food Items   Although FDPIRiO is designed to offer recipients several choices within
                                 the four food groups, not all of the items are consistently available for
                                 recipient selection. Panelists and recipients at all the reservations,
                                 except White Earth, were concerned that, of the authorized items, only
                                 half are available each month. Moreover, recipients told us that many of
                                 the items do not vary from month to month. For example, often the only
                                 vegetable available is canned green beans, the only fruit available is
                                 canned pineapple, and the only meat available is canned luncheon meat.
                                 As a result, households eat the same foods throughout the month and
                                 often for many months at a time. We observed a lack of many author-
                                 ized items when we visited a distribution warehouse on the Navajo Res-
                                 ervation in June 1989. We found that only about 30 of the 60 authorized
                                 commodities from the 4 food groups were listed as available for selec-
                                 tion by recipients. We also noted that few offerings were available from
                                 the vegetable and fruit food groups and that chicken was not available.

                                 In contrast, the Food Stamp Program enables households to overcome
                                 the lack of variety experienced by FDPIR recipients. Food stamp recipi-
                                 ents at all four reservations told us that the Food Stamp Program pro-
                                 vides them the flexibility to choose what to buy and when to shop.
                                 Notably, they cited their ability to purchase fresh fruits, vegetables,
                                 meats, and dairy products, as well as other items not offered by FDPIR, a.s
                                 advantages of the Food Stamp Program. Panelists at the Pine Ridge and
                                 Fort Berthold reservations also agreed that the Food Stamp Program
                                 offers recipients a much greater variety of food than that authorized by
                                 FDPIR, and therefore, it is better suited for recipients with special dietary
                                 needs.

                                 Panelists and recipients from the Navajo, Pine Ridge, and Fort Berthold
                                 reservations recommended that usr% ensure that the majority of all
                                 authorized food items be available on a regular basis. Recipients in par-
                                 ticular stated that they should be issued “rainchecks” so that they may
                                 receive out-of-stock items whenever they become available.



                                  loIn 1977, USDA expanded the food package to include over 60 food items and represent thr 1 basic
                                  food groups. In 1986, USDA increased the quantity and nutrient value of the foods authored for
                                  distribution.



                                  Page 34               GAO/ltCEMO-152      Recipient   and Jkpert   Views on Indian   Food .Amsiatance
                               Chapter 3
                               Hunger and Diet-Related   Illnesses Continue
                               for Those in Food Assistance Programs




                               According to USDA, its annual commodity purchase plan determines what
                               commodities will be available for distribution. FNS develops the annual
                               plan by balancing legislative requirements with agricultural market
                               information, available funds, and recipient commodity preferences. The
                               plan may be modified throughout the year to take into account changing
                               market conditions. If a food item is not available because of market con-
                               ditions, there is, in most cases, an alternative commodity of equal nutri-
                               tional makeup from the same food group available for distribution.
                               However, according to USDA, local FDPIR staff do not always order the
                               complete variety of food items even when choices are available. Our
                               September 1989 report also identified other factors-such as the order-
                               ing pattern of local FDPIR staff, price fluctuations, and storage limita-
                               tions at the state and reservation level-that limit the variety of food
                               available at specific reservations for specific months.

Inadequate!3ervingsof Fruits   We also found that the FDPIR food package lacks adequate servings of
and Vegetables                 fruits and vegetables. According to criteria from the American Red
                               Cross’ Food Wheel, healthy individuals need approximately 92 servings
                               of fruits and 122 servings of vegetables per month.ll However. according
                               to FNS, the FDPIR package is designed to provide only 67 servings of fruits
                               (or 73 percent of that recommended) and 34 servings of vegetables (or
                               28 percent of that recommended) per month. Moreover, a comparison of
                               the number of servings of fruits and vegetables actually issued to recipi-
                               ents to those recommended, shows that recipients were taking home
                               only 59 percent of the recommended servings of fruits and 25 percent of
                               the recommended servings of vegetables.12 Although, USDA concurs that
                               the food package is lacking in adequate servings of fruits and vegeta-
                               bles, it believes that because the food package is supplemental. it is not
                               required to provide a complete diet.

                               However, this deviation from the recommended serving criteria is of
                               great concern to panelists and recipients alike at the Fort Berthold.
                               White Earth, and Navajo reservations, because many recipients rely on
                               the food package for their total diet. They recommended that LXDA
                               increase the quantities of fruits and vegetables in the FDPIR package.




                               ’ ‘The American Red Cross’ Food Wheel, developed in cooperation with USDA, gives a range f( lr ! he
                               number of servings needed to meet the nutritional needs of individuals since nutrient needs ba? with
                               age, sex, body build, and physical activity. Ninety-two and 122 represent the midpomt of the range of
                               number of servings from the food wheel for fruits and vegetables, respectively.

                               ‘*Actual   issuance is based on USDA data for June 1987 to May 1988.



                               Page 35                GAO/RCED-W152       Recipient   and Expert   Views on Indian   Food Cuumnce
                             Chapter 3
                             Hunger and Diet-Related   Illnesses Continue
                             for Those in Food Assistance F’rugrams




                             According to IHSofficials on the panels, a diet lacking in adequate serv-
                             ings of fruits and vegetables can significantly impact diet-related health
                             problems, such as diabetes.

Poor Quality of Food Items   Poor quality of available food items also contributes to inadequate food
                             package variety. Recipients at all four reservations told us that many of
                             the food items they receive, such as milk, butter, cheese, oatmeal, and
                             flour, are inedible because they are either spoiled, moldy, or infested
                             with bugs. Recipients also told us that they would like to see expiration
                             dates printed on commodities to help them determine if an item is out-
                             dated. Additionally, they said that many edible items are of very poor
                             quality. For example, foreign objects, like pebbles in the vegetables and
                             veins in the meats, are found in the items. Although, FDPIR regulations
                             provide for the replacement of damaged or inedible commodities, recipi-
                             ents at the Navajo Reservation told us that they could not obtain
                             replacements for inedible items but did not say why. Fort Rerthold
                             recipients told us that they did not return commodities because they
                             could not afford to go back to the warehouse, and White Earth recipi-
                             ents feared that returned items are redistributed to other households.

                             The lack of variety, inadequacy of servings, and inedibility of certain
                             foods ultimately reduce the overall nutritional value of the FDPIR pack-
                             age. According to an FNSnutritionist, the nutrient value of the food
                             package is based on the assumption of maximum variety (about 60
                             items). However, if maximum variety is not available, then recipients
                             may not receive either the intended full nutritional content of the pack-
                             age or the recommended daily allowances of essential nutrients, unless
                             they are able to supplement the package with other foods.


High Fat and Salt Content    Despite improvements made to the package,13 panelists from the Fort
of FDPIR Foods               Berthold, Pine Ridge, and Navajo reservations and recipients from all
                             four reservations concurred that some commodities still contain too
                             much fat and salt. Because a diet high in fat can contribute to obesity,
                             diabetes, and heart disease and a diet high in sodium can contribute to
                             hypertension, some foods in the FDPIR package may aggravate these con-
                             ditions. Panelists and recipients both commented on the high quantities
                             of fat and salt in the canned luncheon meat, the excessive fat in the
                             canned pork, and the excessive salt content of the canned vegetables.

                             13The FDPIR food package is the result of a number of arijustments made by FM to the commodity
                             program since 1977, to reflect tribal preferences and improve the nutritional profile of the package.
                             According to FM, the food package compares favorably with the recommended daily allowance goals
                             of the Thrifty Food Plan and is therefore an acceptable alternative to the Food Stamp Program.



                             Page 30                GAO/WED-90.152      Bedpient   and Expert   Views on Indian   Food Assistance
Chapter 3
Hunger and Diet-Related Illn~      Continue
for Those in Food Assistance ~rograma




Both recipients and panelists recommended that                       USDA   reduce the fat and
salt in the commodity package.

The level of sodium in the package may be of special concern to recipi-
ents that have hypertension .IJ According to USDA’S nutritional profile of
a hypothetical FDPIR package for a family of four, the package provides
2,241 milligrams of sodium per day. Although this sodium level meets
the recommended safe and adequate level for healthy individuals, it can
negatively affect some recipients with hypertension. Recent studiesis by
the National Institutes of Health and the National Research Council rec-
ommend that sodium levels for hypertensive individuals should be as
low as 1,500 to 1,800 milligrams per day, respectively.

To minimize the detrimental effects of the high fat and salt content of
FDPIR foods, recipients told us that they have to drain off the fat and
rinse other commodities prior to using them. Panelists from IHS told us
that recipients with special dietary needs are counseled on food prepa-
ration practices so that they may use the foods in the FDPIR package
without aggravating their health problems. Nutrition education availa-
ble to recipients on the reservations is discussed in chapter 4.

In addition, according to panelists at the Navajo and Pine Ridge reserva-
tions and recipients at the Pine Ridge and Fort Berthold reservations,
the FDPIR food package contains an overly large proportion of starchy
food, including macaroni, rice, and cornmeal. Recipients told us that
often during the last week of the month, these are the only commodities
left in their homes to feed their families. Consequently, many FDPIR
households subsist solely on a high starch diet for several days of the
month. Panelists believe that the high starch content of the commodity
package is a major contributor to the prevalence of obesity on the reser-
vations. I3Wh panelists and recipients recommended that USDA reevalu-
ate the starch content of the FDPIR package. However, FNSconsiders the
starch content of the FDPIR package to be within the prescribed ranges
for a nutritious diet.




14A hypertensive individual may or may not be sodium sensitive, dependlng on whether his/ her
blood pressure rises or decreases with sodium intake. Restriction of sodium in diets for some individ-
uals with mild hypertension has been found to be bene!icial.

“The 1988 Report of the Joint National Committee on Direction, Evaluation and Treatment of High
Blood pressure, National Institutes of Health, and Diet and Health: Implications for Reducing Chronic
Disease Risk, National Research Council, 1989.



Page 37                GAO/RCEJM@152        Recipient   and Expert   Views on lndian   Food Assistance
chapter 3
Hunger and Diet-Related   Illnesses Continue
for Those in Food Assistance programs




In 1985, USM reviewed the nutrient profile of the FDPIR package and con-
cluded that the package provided a nutritionally adequate, supplemen-
tal diet for healthy individuals but made further improvements in the
program to reduce the fat and salt in the package. USDA’Sconclusion
assumed that the maximum variety of foods is consistently available to
recipients. However, recipients’ experiences indicate that this may be
unrealistic; and in fact, the absence of maximum variety may change the
salt, fat, and starch content of the package. For example, as noted in our
September 1989 report, FDPIR recipients at the Navajo Reservation, una-
ble to obtain canned chicken at the time of our review, may have instead
consumed more canned beef or pork, which are about 63 and 136 per-
cent higher in total fat content, respectively, than canned chicken.

In December 1989, USDAproposed further changes to the commodity
package to make it more consistent with dietary guidelines and to be
more responsive to the special needs of its participants. These changes
when implemented will increase the quantities of fruits and vegetables
and reduce the fat content and the quantity of flour available in the
package.




 Page 58               GAO/RCEDWl52        Recipient   and Jhpert   Viewa on Indian   Food AAstMce
Chapter 4

Nutrition Education Is Limited


                        According to recipients at all four reservations, nutrition education
                        offered by FDPIR is limited, and little, if any, assistance is provided to
                        food stamp recipients to help them make knowledgeable food purchases.
                        Since many of the health problems of greatest concern for Indians
                        appear to be diet-related, panelists at all four reservations believe that
                        nutrition education is necessary to convince recipients-especially       those
                        who are obese or diabetic-to adopt food preparation and eating habits
                        that will prevent or minimize these problems. The panelists also sug-
                        gested ways for addressing such education to all household members.


                        Behavior change, especially in dietary practices, is a key element in
Nutrition Education     reducing the risk for chronic disease, according to the Surgeon General’s
Can Help Prevent and    1988 report on nutrition and health.’ The report recognizes that efforts
Treat Health Problems   to induce beneficial changes in dietary habits are based on an assump-
                        tion that people who understand the risks associated with their present
                        practices will alter them to prevent illness. However, it is also recog-
                        nized in the report that a number of societal and behavioral forces-
                        such as peer pressure, cultural and familial standards of appropriate
                        food intake, advertising of high-calorie foods and alcoholic beverages or
                        other determinants-inhibit    dietary change.

                        The report states that, despite these difficulties, considerable evidence
                        supports the effectiveness of nutrition education in changing dietary
                        intake to reduce risk factors for symptoms of conditions such as heart
                        disease, diabetes, and hypertension, among others.

                        Panelists at all four reservations told us that nutrition education can
                        play a significant role in the prevention and treatment of many of the
                        chronic, diet-related diseases that affect Indians living on reservations.
                        According to panel members, because obesity, diabetes, heart disease,
                        and hypertension may get their start at a very early age, these diseases
                        can be prevented through proper diet, especially one low in fat, sugar,
                        and salt.

                        Panelists at all four reservations view nutrition education as being
                        essential for food stamp recipients to help them make more knowledgea-
                        ble, economical, and nutritious decisions about their food purchases.
                        Panel members stated that nutrition education might not immediately


                        ‘The Surgeon General’s Report on Nutrition and Health, U.S. Department of Health and Human Ser-
                        vices, 1986, pp. 511-514.



                        Page 39              GAO/RCED-9@152      Recipient   and Expert   Views on Indian Food Assistance
Chapter 4
Nutrition Education   Is Limited




change food stamp households’ purchasing and eating habits but it will
make individuals more aware of the health impact of certain foods.

Nutrition education may even be more critical for FDPIR recipients, par-
ticularly those who are obese or diabetic. According to IHSnutritionists
on the Navajo, Pine Ridge, and White Earth panels, FDPIR recipients who
suffer from these problems are especially in need of nutrition counseling
because of the inadequate servings of fruits and vegetables in the FDPIR
package, the fat and salt content, and the limited availability of certain
commodities. For example, Navajo, Fort Berthold, and White Earth pan-
elists told us that recipients with diabetes need more fruits and vegeta-
bles than do healthy individuals. Since the package does not provide
adequate servings of these foods for healthy individuals, diabetic recipi-
ents need to be aware that they must supplement their commodity diet
with fruits and vegetables to stay healthy.

According to one Navajo nutrition educator for FDPIR, counseling house-
holds to stay away from fat and salt while providing them the commodi-
ties seemed contradictory. Other nutrition educators on the panels
agreed that commodities could aggravate health problems but thought
that these effects could be minimized through careful food preparation.
Furthermore, Navajo panelists told us that the limited variety of other
commodities from various food groups may compound these problems
unless participants understand nutrition concepts. For example, they
told us that many families do not know that corn and potatoes, which
are both starches, should not be served at the same meal

Panelists at the Fort Berthold and Pine Ridge reservations acknowl-
edged that while Indian households with health problems could better
accommodate their special diets through the Food Stamp Program they
feared that not all could switch to food stamps because of differences in
program eligibility criteria as previously discussed in chapter 2.

While some panelists stated that it was essential that food stamp and
FDPIR recipients comprehend the importance of and be able to obtain
nutritious diets, they all stated that, in general, many Indian families do
not understand the effects that certain foods have on their health. One
Navajo panelist told us that when his office set up nutrition displays at
different places on the reservation, people were surprised to learn how
much fat went into fried chicken and how much sugar was in sodas.
Others who confirmed this need for nutrition education reported seeing
food stamp recipients continuaIIy buying soda and potato chips, which
are high in sugar and fat, respectively.


Page 40                GAO/RCED90162   Redpient   and Expert   Viewa on Indian   Food Assistance
                             Chapter 4
                             Nutrition Education   Is Limited




                             Navajo and Pine Ridge panelists also believe poor nutrition results from
                             individual preferences for traditional foods and cooking methods, which
                             involves frying with either lard, shortening, or butter. For example, the
                             most popular traditional foods eaten in the Indian homes are fry bread,
                             Indian tacos, and Indian soup, which are usually cooked with grease,
                             fat, and salt. Other cultural habits are related to individual tribes. For
                             example, a mainstay of the Navajo diet is mutton, according to panelists;
                             the Navajo diet is very fat oriented; and a common practice at White
                             Earth is to fry macaroni. As a result, Navajo panelists told us that those
                             most unaware of poor eating habits and their effects are the older Indi-
                             ans and those living in remote areas who continue to use the traditional
                             methods of food selection and preparation. We were told that some of
                             the Navajo elderly still believe that being obese is a sign of good health.


                             Nutrition education is a component of both the Food Stamp Program and
Nutrition Education Is       FDPIR. Section 1 l(f) of the Food Stamp Act of 1977, as amended, autho-
Limited and Varies by        rizes USDAto extend food and nutrition education to food stamp program
Program                      participants. USDA has developed single-concept printed material, espe-
                             cially designed for persons with low reading and comprehension levels,
                             on how to buy and prepare more nutritious and economical meals and on
                             the relationship between food and good health. The act allows the Secre-
                             tary of Agriculture discretion in setting the form and content of nutri-
                             tion education programs and in determining where and how such
                             programs may best be used.

                             Nutrition education is also an integral part of FDPIR. FNSregulations stip-
                             ulate that state agencies administering FDPIR shall provide nutrition edu-
                             cation to participating households relative to the use and storage of USDA
                             commodities. Nutrition education activities must be identified in state
                             agency operating plans and are federally funded out of each agency’s
                             FDPIR administrative   budget. However, according to an FNSofficial, the
                             form, content, and amount of funds allocated to nutrition education
                             activities is determined by each state agency; and these activities are
                             limited and vary between the reservations.


Nutrition Education at the   Although both programs provide for nutrition education, we noted in
                             our September 1989 report that the amount and types of nutrition edu-
Four Reservations Varied     cation activities provided as part of the Food Stamp Program and FDPIR
                             at the four Indian reservations varied. According to Food Stamp Pro-
                             gram officials, nutrition education activities of local food stamp offices
                             serving the four reservations consist primarily of making nutrition


                              Page 41               GAO/RCED-g&152   Bedpient   and Expert   Views on Indian Food Assistance
                            Chapter 4
                            Nutrition Education   Is Limited




                            brochures and other literature available to food stamp recipients. Some
                            recipients receive food stamps at their residences and, therefore, may
                            not be exposed to this literature except when applying or reapplying for
                            benefits at the food stamp office.

                            In FDPIR, the Navajo program provided the most nutrition education
                            activities, which ranged from monthly lectures to cooking demonstra-
                            tions. At the White Earth, Fort Berthold, and Pine Ridge reservations,
                            nutrition education is provided primarily by IHS, tribal community
                            health representatives, or the home extension service’s home economist.
                            Nutrition education at these three reservations consisted of cooking
                            demonstrations and lectures at White Earth, counseling and visual
                            presentations at Fort Berthold, and the dissemination of nutrition litera-
                            ture at Pine Ridge. Although nutrition activities were present on these
                            reservations, IHS and tribal nutritionists expressed the need for
                            expanded services that are tailored to the specific needs of reservation
                            Indians.


Recipient Experiences       At the four reservations we visited, recipient nutrition education exper-
With Nutrition Educati.on   iences related to the recipients’ food assistance benefits were limited
                            and varied depending on which program they participated in. Food
Dependedon Program          stamp participants we interviewed at all four reservations were very
                            interested in receiving nutrition education from the program but neither
                            were aware of nor had attended any local food stamp office activities
                            related to nutrition education or food budgeting. The general sentiment
                            among recipients was that, beyond providing the basic benefit, the Food
                            Stamp Program does not provide adequate nutrition education.

                            In contrast, FDPIR recipients at all four reservations had received or were
                            aware of some nutrition education activities, particularly with respect to
                            the FDPIR cookbook. WhiIe some were not interested in receiving these
                            items from the program, others were very interested. Cooking demon-
                            strations were perceived as desirable by many recipients, although only
                            some recipients at the Fort Berthold and the Navajo reservations had
                            attended cooking demonstrations. At the Navajo Reservation, some of
                            the recipients stated that the cooking demonstrations already available
                            were not advertised as well as they could be.

                            In addition to nutrition education offered through FDPIR, recipients at
                            the Navajo, White Earth, and Pine Ridge reservations told us that coun-
                            seling is available through the IHSnutritionist for individuals with spe-
                            cial dietary needs, such as diabetics or the elderly. However, unless


                            Page 42                GAO/RCED-90-152   Redpient   and Expert   Views on Indian   Food .bsistance
                      chapter 4
                      Nutrition Education   b   Limited




                      recipients are being treated by IHS, they need to be referred by the Food
                      Stamp Program or FDPIR before they can receive counseling. We also
                      noted that at Pine Ridge, this counseling may not be available because
                      the IHS nutritionist position had been vacant for many months.


                      Panelists at all four reservations told us that because many of the Indian
Suggestionsfor        health problems require dietary modification for prevention and treat-
Expanding Nutrition   ment, they believe that nutrition education should be provided with fed-
Edbcatioh-            eraI food -assistance. In fact, panel members at the Navajo and White
                      Earth reservations believe nutrition education should be a prerequisite
                      to participation in the Food Stamp Program, by making benefits condi-
                      tional on recipients attending nutrition classes similar to those in the WIG
                      program. Unless the needed nutrition education accompanies food assis-
                      tance and an adequate food supply is available, these panelists thought
                      that the prevalence of obesity, diabetes, heart disease, and hypertension
                      is likely to continue.

                      The panelists offered a variety of suggestions for expanding and
                      improving nutrition education, The Navajo and Fort Berthold panelists
                      suggested that the federal health, welfare, and food assistance programs
                      pool their resources to provide nutrition education necessary to improve
                      the health of recipients of these programs, The Navajo panelists recom-
                      mended that the Food Stamp Program and FDPIR earmark funds for
                      nutrition education instead of leaving it to the discretion of local admin-
                      istrators of the programs. They also told us that federal agencies or pro-
                      grams within an agency should be allowed to consolidate their funds to
                      sponsor mass media campaigns and education programs that would pro-
                      mote understanding of good health and nutrition and to allow them to
                      use outside experts to effectively design these programs.

                      Panelists at all four reservations thought that one-on-one counseling
                      with a nutritionist would be the most effective way of providing nutri-
                      tion education through the Food Stamp Program and FDPIR, while
                      printed materials would be least effective. According to the 1988 Sur-
                      geon General’s report on nutrition and health, to do this, the programs
                      would have to hire more nutrition educators, which may be difficult in
                      remote areas, such as the Pine Ridge Reservation. Other effective nutri-
                      tion activities suggested by the panelists included (1) using public
                      broadcasting services, (2) showing videotapes at strategic locations like
                      FDPIR warehouses and food stamp offices, (3) expanding or reinstating
                      home extension services, (4) nutrition labeling of all commodities, and
                      (5) developing and distributing more recipes for FDPIR foods.


                       Page 43                  GAO/RCED4&152   Redpient   and Expert   Views on Indian Food Assistance
chapter 4
Nutrition Education   I.9 Limited




Because dietary behavior is also affected by psychological, cultural,
environmental, and economic factors, the panelists told us that nutrition
education provided through the federal programs needs to be tailored to
the behavior and knowledge of recipients regarding food and nutrients.
However, since it is difficult to change individual dietary habits in a
short period of time, in the interim, other changes are needed in FDPIR,
such as reducing the fat content of FDPIR meats. The White Earth, Fort
Berthold, and Navajo panel members also thought that nutrition educa-
tion should address cultural foods and their effects on health as well as
different ways to prepare commodities in times of limited nutritional
variety.

In addition to education provided through the food assistance programs
for heads of households or parents, panelists at all four reservations
thought that more nutrition education should be provided through the
public school curricula to address dietary habits at an early age. Addi-
tionally, they said, teachers, particularly in elementary schools, and par-
ents need extra training so that learning can be reinforced outside the
classroom as well as inside.




 Page 44                GAO/RCED9&152   Recipient   and Expert   VIewa on Indian   Food .bhstance
Chapter 5

GAO Observations,Conclusions,
and Recommendations

              The nutritional status of Indians is the product of complex interactions
              among environmental, cultural, economic, and other factors. The exact
              proportion of the effect that each of these factors has on an individual
              diet is uncertain. Some specific factors that we identified include the
              availability of an adequate food supply, nutritional content of foods
              consumed, individual food selection and preparation methods, accessible
              transportation, and individual diet-related health problems. While these
              and other factors cannot be fully addressed by existing federal food pro-
              grams, efforts to improve program services-especiaily      in providing
              adequate, consistent, and accessible food assistance and dietary educa-
              tion-can enhance the overall nutrition and health of reservation
              households.

              Federal food assistance programs, primarily the Food Stamp Program
              and FDPIR, have provided supplementary sources of food assistance for
              Indian households. However, there are indications, although difficult to
              quantify, that some hunger exists at all four reservations we visited.
              Also of concern to Indian households on all four reservations was the
              prevalence of diet-related diseases and the impact of federal food assis-
              tance programs on those diseases.


              Our September 1989 report found indications, including studies by res-
Hunger        ervation officials, of some hunger on the Fort Berthold and Pine Ridge
              reservations, the report also noted the growing dependence of Indian
              households on nonfederal food assistance. In this follow-on report, com-
              munity officials and federal food assistance recipients told us that some
              hunger exists on all four reservations, especially during the last week of
              the month.

              It is difficult to quantify the severity, both in the percent of the popula-
              tion affected and the duration to which hunger lasts each month. Fur-
              thermore, because of the diversity of factors that may contribute to
              hunger on the reservations, it is difficult to devise a comprehensive
              solution that would allow all reservation households to obtain an ade-
              quate diet.

              According to community representatives at all four reservations and
              administrators of nonfederal food assistance, hunger is common among
              households not receiving federal food assistance and more common
              among food stamp recipients than FDPIR participants. These obsematlons
              were confirmed by many food stamp recipients we talked to who



              Page 45          GAO/RCED-9@152   Recipient   and Expert   Vfewa on Indian   Food .bmsunce
chapter 5
GAO Observations,   Conclusions,
and Recommendations




described their food shortage experiences and limited resources to                          sup-
plement their benefits.

Our past’ and current work has shown that the administrative require-
ments of the application process and asset limitations can be participa-
tion obstacles in the Food Stamp Program. Some of the hindrances to
participation that we have identified are the unintentional results of
state or local office attempts to provide food stamps more efficiently.
For example, according to recipients and community representatives,
the lengthy application form, while helping to streamline the adminis-
trative process, can delay the delivery of benefits and discourage Indian
households from applying for food stamps. These administrative hin-
drances-coupled      with illiteracy, language barriers, a lack of informa-
tion about program eligibility and benefits, and poor physical access to
reservation food stamp offices-can prevent many needy Indian house-
holds from getting food assistance. As part of its annual review of state
food stamp operations, FNScould focus on the administrative hindrances
discussed in this report to identify these obstacles and ways to assist
states in overcoming them.

Poor physical access, according to recipients and community representa-
tives, is due, in part, to the Food Stamp Program’s $4,500 automobile
exclusion, as applied to the household resource limitation. They believe
that the $4,500 exclusion, which was established by the Food Stamp Act
of 1977, can deny rural participants the reliable transportation needed
to comply with food stamp procedural requirements as well as, limit
participation in the program. Although we did not determine how the
$4,500 vehicle exclusion affects participation nationwide, we believe
that if the $4,500 was considered to be a reasonable allowance for vehi-
cle asset value at the time of the 1977 act, because of inflationary
impacts it may no longer represent a reasonable allowance today. Intro-
duced on February 27, 1990, H.R. 4110 would among other things
increase the $4,500 limit to $5,500 for the period of January 31 to Sep-
tember 30, 1991. On October 1, 1991, and in each year thereafter, the
 limit would be adjusted to reflect changes in the Consumer Price Index.

 Further, with community representatives and program recipients citing
 hunger among nonparticipants and misunderstanding about program eli-
 gibility, improvements in outreach effectiveness may attract eligible
 people to participate in one of the federal food assistance programs. The
 Hunger Prevention Act of 1983 authorized federal funding to help pay

 ‘Food Stamp Program: Administrative Hindrances to Participation (GAO/RCED-S9-4, Oct. 21, 1988).



 Page 43                GAO/‘RCRl%BO-152   Recipient   and Expert   Views on Indian   Food Assistance
Chapter 5
GAO Observations,   Conciueions,
and Recommendations




for outreach services. That funding is available to states to help promote
informational activities regarding program eligibility, benefits, and the
application process.

Providing assistance, as currently required by food stamp regulations,
to applicants who have difficulty in obtaining the required documentary
evidence can make the application process less burdensome for those
households who are experiencing problems. Further, to help reduce bar-
riers in rural areas, the Hunger Prevention Act of 1988 allows states to
waive in-office interviews and mail application forms to applicants not
served by a certification office or who have transportation difficulties.
We believe that these types of services could make the Food Stamp Pro-
gram more accessible to some Indian households.

For Food Stamp Program participants, local food stamp office practices
that affect the delivery of benefits are also important factors in address-
ing the causes of hunger on the four reservations. In particular, monthly
reporting and recertification procedures that cause interruptions in ben-
efits, which we identified to be problems for other food stamp house-
holds in a June 1989 report,* also apply to Indians on the four
reservations. As we previously reported, broader implementation of the
monthly reporting reinstatement option can eliminate losses experienced
by many of the participants who file late monthly reports and receive
 benefits prorated from the late filing date. To minimize these losses, we
recommended that FNScontact the 13 states that had not adopted the
monthly reporting reinstatement option, including Utah, New Mexico,
 Minnesota, and North Dakota, and encourage its adoption where practi-
 cable. FNSis in the process of analyzing the responses from these states
 that had been contacted by its regional offices.

Similar to the reinstatement option, eliminating the proration provision
(section 8(c) of the Food Stamp Act of 1977, as amended) for recertifica-
tion in the Food Stamp Program, as recommended in our June 1989
report, would benefit those participants who are eligible for food stamps
but are temporarily terminated from the program for failing to meet its
recertification requirements in a timely manner. Elimination of the pro-
ration provision would allow eligible reservation households to receive a
full month’s food stamp benefits if they complete their recertification
requirements by the end of the following month.


 2F& bp       propram: Pti+tnts    Temporarily TemUated for Procedural Noncompliance (GAO/
 m-81,        June 22,1989).



 Page 47               GAO/RCED-B&l52    Recipient   and &pert   Views on Indian Food Assistance
                        Chapter 5
                        GAO Observations,   Conclusions,
                        and Recommendations




                        Another area in which administrative practices can improve delivery of
                        benefits involves the scheduling of food stamp issuances, States have
                        discretion in staggering the issuance of food stamps throughout the
                        month, instead of issuing them to all participants concurrently. Stag-
                        gered issuance, as is provided by the states serving the Navajo Reserva-
                        tion, may in fact help discourage grocers from allegedly raising their
                        prices when food stamps are issued. Such actions might be particularly
                        appropriate for North and South Dakota serving the Fort Berthold and
                        Pine Ridge reservations, where the number of food stores are limited
                        and increased food prices have been alleged.


                        Since many Indian households on the four reservations suffer from
Diet-Related Concerns   health problems associated with diet (obesity. diabetes, heart disease,
                        and hypertension) and rely heavily on federal food assistance, providing
                        appropriate commodity foods and dietary education is important in the
                        prevention and treatment of these diet-related health problems. Ensur-
                        ing (1) that commodity packages consistently contain adequate nutri-
                        tional variety, reduced levels of fat and salt, and expiration dates on
                        commodities and (‘2) the replacement of inedible commodities can
                        improve the nutritional quality of the Indian diet and is especially
                        important for persons with special diet-related needs. We are aware that
                        FNSis currently evaluating the effectiveness of FDPIR in providing par-
                        ticipants with an acceptable alternative to the Food Stamp Program.
                        However, because its study does not specifically address the quality and
                        nutritional variety of available commodity foods, we believe that FNS
                        should consider the issues identified in this report in its evaluations of
                        the program.

                        Educating reservation households about the dietary choices most condu-
                        cive to prevention and control of certain diet-related health problems is
                        essential, While including the importance of adequate physical activity,
                        educational efforts should focus on the potential benefits of eating a bal-
                        anced diet that is low in fat and salt.

                         Effective nutrition education should encourage the Indian population,
                         especially those with diet-related health problems, to adopt behavioral
                         changes in their food preparation and dietary consumption practices
                         that will benefit them and allow them to obtain the maximum value
                         from their benefits. Nutrition education directed at reservation house-
                         holds that participate in federal food assistance programs would be
                         most effective, we believe, if it were tailored to their food preferences



                         Page 48                GAO/l!CEDBO-162   Recipient   and Expefi   Views on Indian   Food Amsistnnce
                        Chapter 6
                        GAO Observations,   Conclwdons,
                        and Recommendationa




                        and lifestyles, emphasized the role of diet in health promotion and dis-
                        ease prevention, provided practical nutrition assessment methods, and
                        included dietary counseling.

                        Although some nutrition education is being provided at the four reserva-
                        tions, the amount and form of educational activities varied by reserva-
                        tion and by program. We were told that this is so because state and local
                        administrators determine the level of nutrition education effort for their
                        programs.


                        To enhance the overall effectiveness of the Food Stamp Program and
Recommendationsto       FDPIR in meeting the nutritional needs and diet-related health concerns
the Secretary of        of households located on the four Indian reservations, we recommend
Agriculture             that the Secretary of Agriculture direct the Administrator, Food and
                        Nutrition Service, to

                    . emphasize to the respective states the importance of promoting effective
                        informational activities, particularly outreach services, to reduce misun-
                        derstandings about program eligibility,
                    .   explore with state officials in North and South Dakota whether state
                        food stamp issuance practices result in increased area food prices and
                        take corrective action, as appropriate,
                    .   monitor on a continuing basis the availability of all FDPIR commodities
                        and, where variety is consistently limited, both (1) determine the nutri-
                        tional implications for recipients and the costs of improving commodity
                        availability and (2) take corrective action as appropriate,
                    .   include as part of its program evaluations a review of the quality of
                        FDPJR commodities, including reductions in the fat and salt content,
                        determine whether improvements are needed to ensure the edibility of
                        the commodities and their appropriateness for consumption, especially
                        by those persons with special diet-related health problems, and
                    .   emphasize the importance of nutrition education on Indian reservations
                         and work with tribal and program officials to ensure that adequate edu-
                         cation services are provided to recipients who participate in the Food
                         Stamp Program and FDPIR.




                        Page 49                GAO/RCEl%~l62   Recipient   and Expert   Views on Indian   Food Assistance
Appendix I

Participants of Panel Meetingsat the
Four Reservations

                       Director, Headstart Program
Fort Berthold          Director, Department of Social Services, Three Affiliated Tribes
Reservation            Director, Human Services Department, Three Affiliated Tribes
                       Director, FDPIR
                       Service Unit Director, Indian Health Service
                       Program Assistant, Home Extension Service
                       Nutritionist, wrc Program
                       Member, Tribal Council
                       Human Services Instructor, Fort Berthold Community College
                       Certifier, FDPIR
                       Caseworker, Mountrail County Department of Social Services
                       Representatives, Community Health Representative Program


                       Director, FDPIR
Pine Ridge             Director, Orte Food Bank
Reservation, Panel 1   Tribal Chief and Member, Foster Grandparents Senior Groups
                       Administrative Officer, Indian Health Service Hospital
                       Health Coordinator, Ogala Sioux Tribe
                       Social Services Caseworker, Bureau of Indian Affairs
                       Certifier, FDPm
                       Representative, Community Health Representative Program
                       Pastor, St. Agnes Parish


                       District Supervisor, Food Stamp Program
Pine Ridge             Acting Director of Social Services, Bureau of Indian Affairs
Reservation, Panel 2   Chief, Nutrition and Dietetics Branch, Aberdeen Area, Indian Health
                       service
                       Coordinator, Community Health Representative Program
                       Supervisor, Shannon and Bennett County Social Services
                       Diabetes Control Officer, Aberdeen Area Indian Health Service
                       Manager, Lakota Produce Growers
                       Certifier, FDpIR
                       Nutritionist, WIGProgram


                       Director, Health and Human Services, Chippewa Tribe
White Earth            Director of Education, Tribal Council
Reservation            Coordinator, FDPIR
                       Coordinator, Indian Child Welfare Program
                       Coordinator, Elderly Nutrition Program
                       Mahnomen County Coordinator, Mahube Community Council


                       Page 60          GAO/RCED4@162   Redpient   and Expert   Viewr on Indian Food Assistance
                      Appendix I
                      Participants  of Panel Meetings   at the
                      Four Resecvation~




                      Financial Assistance Supervisor, Becker County Human Services
                      Nutritionist, Indian Health Service
                      Medical Doctor, Indian Health Service


                      Director, Income Support Division, McKinley County
Navajo Reservation,   Director, Navajo wrc Program
Panel 1               Director, Navajo New Dawn Program
                      Co-Directors, Northern Arizona Food Bank in lagstaff
                      Associate Director, Navajo Direct Services Section, Navajo Department
                      of Health
                      Coordinator, McKinley County Student Nutrition Program
                      Eligibility Coordinator, Navajo FDPIR


                      Director of Human Services, Southwest Indian Foundation
Navajo Reservation,   Nutrition Coordinator, Navajo WIGProgram
Panel 2               Chief, Nutrition and Dietetics Branch, Navajo Indian Health Service
                      Manager, Navajo and Hopi Family Assistance Administration
                      Administrative Service Officer, Direct Services Section, Navajo Depart-
                      ment of Health
                      Administrative Officer, Community Health Service Program
                      Nutritionist, Navajo FDPIR
                      Manager, Fort Defiance Unit, Community Health Service Program


                      Director, Meals on Wheels
Navajo Reservation,   Executive Director, Echo Inc.
Panel 3               Manager, Shiprock Senior Center, Senior Citizen Centers Program
                      Coordinator, Senior Citizen Centers Program
                      Administrative Assistant, Navajo FDPIR
                      Supervisor, Teecnospos Warehouse, Navajo FDPIR
                      Eligibility Supervisor, New Mexico Income Support Division, San Juan
                      County
                      Supervisors, wIc Program
                      Eligibility Worker, New Mexico Income Support Division, San Juan
                      county
                      Nutrition Educator, Navajo FDP~R Program
                      Regional Nutritionist, wrc Program




                       Page 51               GAO/RCED9@152       Recipient   and Expert   Viewa on Indian   Food Assistance
Appendix II

DemographicCharacteristicsof Indian
HouseholdsParticipating in F’DPIR

                     To provide demographic information regarding characteristics of Indian
                     households receiving food packages in calendar year 1988 on the four
                     reservations, we took a probability sample of packages issued at each of
                     the four reservations.’ (See app. IV for a detailed description of our sam-
                     ple selection process.)

                     The results of our analysis of Indian households receiving FDPIR pack-
                     ages are presented by reservation. The numbers in parentheses follow-
                     ing each estimate are the sampling errors of the estimates at the 95-
                     percent level of confidence.

                     Because we sampled FDPIR packages, not households, a particular house-
                     hold can appear more than once over the time period of our review. For
                     example a household that received 12 monthly F+DPIR packages during
                     1988 at 1 of the 4 reservations might have been the recipient of 2 sam-
                     pled packages- 1 in April and 1 in September.


                     This section provides information regarding the size and composition of
Household Size and   households receiving FDPIR packages on the four reservations.
Composition
                     We estimate that the average number of people served by a FDPIR pack-
                     age was 3.2 (+- 0.2) for Fort Berthold, 3.0 (kO.2) for Pine Ridge, 2.5 (+
                     0.2) for White Earth, and 3.3 (+ 0.2) for the Navajo.

                     To provide information on household composition, we considered the
                     percent of households with and without children. This information is
                     provided in table II. 1.




                     ‘Our analyses of FDPIR and Food Stamp Program data are not comparable because they mclude
                     differences in (1) eligibility requirements of the programs (as detailed in ch. l), (2) time period differ-
                     ences (calendar year 1988 for FDPIR and fiscal year 1988 for food stamps), and (3) definitions and
                     geography (as detailed in app. IV).



                     Page 52                 GAO/RcEDoo-152        Recipient   and Expert   Viewn on Indbn    Food Aseistrn~e
                                      Appendix II
                                      Demographic Characteristica     of Indian
                                      Households Participating    in FDPJR




Table 11.1:Estimated Percentages of
1988 FDPIR Packages Provided to                                                        Fort
Households With or Without Children   Household type*                             Berthold     Pine Ridge       White Earth                 Navajo
                                      Households with children                      50(+6)          51 (25)           41 (k5)               61 (27)
                                      Households without children                   50(?6)          49(+-5)           59(?5)                39ik7)
                                      Notes:
                                      The numbers tn parentheses   are the sampling errors of the estimates at the 95percent     level of
                                      confidence.

                                      The number of packages represented rn our analysrs was 4,733 (+ 127) for Fort Berthold. 15,613 (&
                                      497) for Prne Ridge, 7,394 ( + 0) for Whrte Earth. and 71,170 (? 9,106) for Navajo.
                                      aAn adult IS defined as someone age 20 or older, and a chrld IS defined as someone age 19 or younger


                                      To provide more detailed information concerning the composition of
                                      households with children, we combined these households into four
                                      groups: (1) households in which all the children were less than 6, (2)
                                      households in which all the children were between 12 and 19, (3) all
                                      other households with children,2 and (4) all households without chil-
                                      dren. This information is provided in table 11.2.These groupings were
                                      chosen to highlight households with very different dietary require-
                                      ments. For example, the type and quantity of food required by teenagers
                                      is very different from that required by younger children.

Table 11.2:Estimated Percentages of
1988 FDPIR Packages That Went to                                                       Fort
Different Household Types             Household type.                             Berthold     Pine Ridge        White Earth                Navajo
                                      All children less than 6                       8(k3)            8(+3)             7(+2)                7(?5)
                                      All children between 12-19                    13 (It41         13(?3)             8(?2)               15(c6)
                                      All other households with
                                         children                                   29(?5)          30(?4)             27 (54)              39(r8)
                                      Households without children                   50(?6)          49(25)             59(-t5)              39(27)
                                      Notes:
                                      The numbers in parentheses    are the sampling errors of the estimates at the 95percent    level of
                                      confidence
                                      The number of packages represented by our analysis was 4,733 (+ 127) for Fort Berthold, 15,613 (2
                                      497) for Pine Ridge, 7,394 (2 0) for White Earth, and 71,170 (+ 9,106) for the Navafo.
                                      aAn adult is defined as someone age 20 or older, and a child is defined as someone age 19 or younger




                                       ‘%is groupincludes households in which all the children are between 6 and 11, as well as households
                                       with a mixture of children, for example, one teenager and one infant.




                                       Page 53                GAO/BCED4O-152 Xeciplent         and Expert     Viewe on Indian    Food Assistance
                  Appendix II
                  Denlographic Chaln~ri8tics     of Indian
                  Households Participating   in F-DPLB




                  According to FM, because the nutrient profile of the FDPIR package is
                  comparable to that of the Thrifty Food Plan (TIV>~and is an acceptable
                  alternative to the food stamp benefit, we were interested in estimating
                  the extent to which FDPIR households matched the standard family
                  structure as defined by the TFP.However, because there were so few
                  occurrences of the standard family in our sample, we were unable to
                  make a meaningful estimate of the extent to which they did match the
                  definition. The standard family accounted for 0 percent of our sample at
                  Fort Berthold, 0.5 percent at Pine Ridge, 1 percent at White Earth, and
                  0.2 percent at Navajo.


                  Because the F’DPIR package is intended to be a supplemental food source,
Gross Income of   we were interested in estimating what percentage of the FDPIR packages
Households        went to households with some gross income (any amount over SO). We
                  computed gross income by totaling each household’s earned and
                  unearned income.

                  We estimate that the percentage of FDP~R packages that went to house-
                  holds having some gross income (any amount over $0) was 87 ( + 4) per-
                  cent at Fort Berthold, 75 ( f 4) percent at Pine Ridge, 92 ( * 3) percent at
                  White Earth, and 96 ( + 3) percent for the Navajo.

                  To obtain the following averages, we included only those packages going
                  to households having some gross income. For these, we estimate that the
                  average gross monthly income per person served by a FDP~R package was
                  $177 (+ 15) at Fort Berthold, $188 (+ 12) at Pine Ridge, $283 ( 2 12) at
                  White Earth, and $198 ( f: 15) at the Navajo.




                  3The TFP for a family of four (a man and woman 20 to 60 years of age, and children ri to A and 9 to
                  11 years of age) by law constitutes the basis for allotments to households particrpatmg In the Fcod
                  Stamp program. Such a family c~r~~titutesthe definition of a “standard family.”



                  Page 54                GAO/RCEMO-152       Recipient   and Expert   Viewa on Indim   Food .An.sist~~~cc
Appendix III

DemographicCharacteristicsof Indian
HouseholdsParticipating in the Food
stamp Program
                     To provide information on the demographic characteristics of Indian
                     households receiving food stamps in fiscal year 1988 in areas approxi-
                     mating the Pine Ridge Reservation in South Dakota and the Navajo Res-
                     ervation in Arizona and New Mexico,’ we analyzed data from FNS’
                     quality control data base on households with at least one Indian mem-
                     ber.2 Our analysis focused on households receiving food stamp benefits
                     in fiscal year 1988 and residing in areas, administered by local agencies,
                     that somewhat approximate the boundaries of the Navajo Reservation
                     in Arizona and New Mexico and the Pine Ridge Reservation in South
                     Dakota. Because it is important to understand the geographic area cov-
                     ered by our data, a detailed description is provided in appendix IV.
                     However, the FNSdata base cannot distinguish among Indian tribes or
                     whether a household lives on or off the reservation. (See app. IV for a
                     detailed discussion of the contents of this data base and our
                     methodology.)

                     The results of our analysis of Indian households receiving food stamp
                     issuances at Pine Ridge in South Dakota and Navajo in Arizona and New
                     Mexico is presented below. The numbers in parentheses following the
                     estimates are the sampling errors of the estimates at the 95-percent
                     level of confidence.

                     Because we sampled food stamp issuances, not households, a particular
                     household can appear more than once over the time period of our
                     review. For example, a household that received 12 food stamp issuances
                     during fiscal year 1988 in the general area of the two reservations,
                     might have been the recipient of 2 sampled issuances-l in April and 1
                     in September.


Household Size and   people in the general area of the two reservations. An estimated 85 ( z 3)
Composition          percent of the food stamp issuances were to households with children.
                     For our analysis we define a child as anyone age 19 years or younger
                     and an adult as anyone 20 years or older.


                     ‘We received data from FNS only when ita data base contained at least 30 Indian (Native Amencan I
                     households in the local area code(s) within a state. We received no data on the Fort Berthold or Wute
                     Earth reservations. See app. IV for a more detailed description.
                     20ur analyses of FDPIR and Food Stamp Program data are not comparable, because they mclude
                     differences in (1) eligibility requirements of the programs (as detailed in ch. l), (2) time penod 1akn-
                     dar year 1988 for FDPIR and fiscal year 19St3for food stamps), and (3) definitions and geography a4
                     detailed in app. IV).



                     Page 55                GAO/ICED-~152        Redpient   and Expert   Views on Lndian Food Assistmce
                                        Appendix Ill
                                        Demographic Characteristks      of Indian
                                        Households Participating   in the Food
                                        S-P    PWDm




                                        To provide more detailed information concerning the composition of
                                        households with children, we combined these households into four
                                        groups: (1) households in which all the children were less than 6. (2)
                                        households in which all the children were between 12 and 19, (3) all
                                        other households with children,3 and (4) all households without chil-
                                        dren. This information is provided in table III. 1. These groupings were
                                        chosen to highlight households with very different dietary require-
                                        ments. For example, the type and quantity of food required by teenagers
                                        is very different from that required by younger children.

Table 111.1:Estimated Percentages of
Fiscal Year 1988 Food Stamp Issuances                                                                                                Estimated
That Went to Indian Households, by      Household type’                                                                               percentb
Different Household Types               All children   less than 6                                                                    20     (24
                                        All children   between      12-19                                                              9     (13
                                        All other households        with children                                                     56     (r5i
                                        All households    wlthout     children                                                        15     (23.
                                        Total                                                                                        101
                                        Source: GAO Analysts of FNS Data

                                        Notes:

                                        The estimate IS the percent of fiscal year 1988 food stamp issuances to Indian households in the geo-
                                        graph&    areas of the Navajo Reservation in Arizona and New Mexico and the Pine Ridge Reservation in
                                        South Dakota.

                                        The number of food stamp issuances represented by our analysis was 144429 (t 13,764)
                                        aAn adult is defined as someone age 20 or older, and a child is defined as someone age 19 or younger

                                        bathe numbers in parentheses are the sampling errors of the estimates at the 95.percent level of confl-
                                        dence. The total of the estimate column does not add to 100 percent because of rounding.


                                        We were also interested in estimating the extent to which the house-
                                        holds receiving food stamps matched the standard family structure as
                                        defined by the TFP.~However, we found so few occurrences of the stan-
                                        dard family in our sample that we were unable to make meaningful esti-
                                        mates. Within our sample, the standard family accounted for only 0.2
                                        percent of the food stamp issuances in the general area of the two
                                        reservations.




                                        3This group includes households in which all the children are between the ages of 6 and 11, as well as
                                        households with a mixture of children, for example, one teenager and one infant.

                                        4The TFF for a family of four (a man and woman 20 to 50 years of age and children 6 to 8 and 9 to 1 I
                                        years of age), by law, constitutes the basis for allotments to households participating In the Food
                                        Stamp Program. Such a family constitutes the definition of a “standard family.”



                                        Page 56                     GAO/RCEIMO-152   Redpient    and Expert   Viewe on Indian Food Assistancr
                     Appendix Ill
                     Demographic Characteristic      of Indian
                     Householda Participating   in the Food
                     stamp PwP--




                     Although the Food Stamp Program is intended to provide an adequate
Gross Income of      low-cost monthly diet for households with no countable income, it also
Households           provides a supplemental food source for households with some counta-
                     ble income. We estimated the percentage of the food stamp issuances
                     that went to households with some gross income (any amount over $0).
                     For the FNSdata, the term “gross income” refers to the data field in the
                     FNSdata base called “gross countable income,” which is defined by FNS
                     as the total monthly income of the food stamp household before apply-
                     ing any deductions. This monthly income is used to determine the food
                     stamp allotment.

                     We estimate that 94 ( f 2) percent of the food stamp issuances went to
                     households having some gross income (any amount over $0). Of these
                     households, we estimate that the average gross monthly income per per-
                     son receiving food stamps was $102 ( f 7).


                     Finally, we compared the household’s maximum food stamp benefit pro-
Comparison of        vided by the Food Stamp Program with the total expected food costs for
Maximum Food Stamp   the recipient households on the basis of the age and sex categories of
Benefit With Food    household members. We did this by computing a “household food stan-
                     dard” for each food stamp household. This standard is the sum of the
Cost Ejasedon        recommended TFYamounts for each member of the food stamp house-
Household            holde5 We then determined the extent to which the maximum food stamp
                     benefit provided the household food standard. We also estimated the
Composition          extent to which the maximum food stamp benefit provided the house-
                     hold food standard according to various household types (detailed in
                     table III. 1). The results of these analyses are provided in table 111.2.




                      5To provide consistency with the FNS data for fwai year 1988, we used TFP costs for various age
                      and sex categories published by USDA for June 1987. These amounts were used as the basis for the
                      food stamp allotments for fmai year 1988.



                      Page 57                GAO/aCED9@l52       Recipient   and Jkpert   Views on Indian Food Assistance
                                         Appendix III
                                         Demographic Characteristics      of Indian
                                         Households Participating    in the Food
                                         S-P        pl-w=m




Table 111.2:Extent to Which the Maximum Food Stamp Benefit Provided the Household Food Standard, by Household Type
                                                                         Household type’
Percent of hous?hold food standard                             All other                      Households
reM$d      by maxlmum food stamp       All children   households with           All children       without
                                       less than 6             children          age 12-19        children     Total
95 or less                                          c              2(-tl)O             5(+2)         6(&2)   13(?3)
96-99                                                c                  9(23)                3(+2)                     9(?3)         21(+4)
100-105                                    2(fl)d                      15(-t4)                       e                         e     17(t4)
106-115                                    6(+2)                      22(-t4)                         e                        e     26(r5)
116-140                                   12(+3)                       9(_+3)                         e                        e     21(i4)
Total                                     2O(-t4)                     57(?5)                 6(?3)                   15('4)         100

                                          Source: GAO analysis of FNS data.

                                          Notes:

                                          The numbers In parentheses are the sampling errors of the estimates at the 95percent       level of
                                          confidence.

                                          The estrmate IS the percent of fiscal year 1988 food stamp Issuances to lndlan households In the geo-
                                          graphlcal areas of the Navajo Reservation In Anzona and New Mexico and the Pine Ridge Reservatton In
                                          South Dakota

                                          The number of food stamp Issuances represented      In the analysis was 141,242 (? 13,642)
                                          Table III.2 may differ slightly from table III.1 because the number of cases excluded from analysis dlf-
                                          fered between the two tables (see noted, table IV.3) For example, the estimated percent of food stamp
                                          issuances to Indian households in which all the children were between the ages of 12 and 19 was 9
                                          (~3) percent, representlng 144,429 (+ 13,764) food stamp issuances In table 111.1,    and 8 ( ?3) percent,
                                          representing 141,242 (? 13,642) food stamp issuances In table Ill 2.
                                          aAn adult IS defined as someone age 20 or older, and a child IS defined as someone age 19 or younger

                                          bThe household food standard is based on the TFP amounts.
                                          % IS possible that a household In which all the children are less than 6 years old could have these
                                          values. Because we did not observe any such households, we cannot make meaningful estimates In
                                          these cells.
                                          dThe sampling error of this estimate may be somewhat understated because no occurrences             were
                                          observed in one of the four strata (see table IV.3 for the four strata).

                                          eNo households wtth all children between 12-19 or households wlthout children can have these values
                                          This occurs because the cost of the TFP for all persons age 12 and over exceeds the amount used as
                                          the basis of the food stamp allotment-the  total for the standard family divided by 4




                                          Page 56                 GAO/BcEDW152          Recipient    and Expert   Views on Indian    Food Assistance
Appendix IV

Methodology and SamplingPlan for Food
Stamp and FDPIR Demographics

                          This appendix details the methodology used to obtain samples of FDPIR
                          packages on the four reservations. For food stamp issuances, we were
                          able to obtain data for Indian (Native American) households on only the
                          Navajo Reservation in Arizona and New Mexico, the Pine Ridge Reserva-
                          tion in South Dakota, and their general surrounding areas. Because of its
                          importance, a detailed description of the geographic area covered by the
                          FNSdata is provided later in this appendix.

                          Because we reviewed probability samples of FDPIR packages and food
                          stamp issuances, each estimate developed from the samples has a mea-
                          surable precision, or sampling error. The sampling error is the maximum
                          amount by which the estimate obtained from a statistical sample can be
                          expected to differ from the true universe characteristics (value) we are
                          estimating. Sampling errors are stated at a certain confidence level-in
                          this case 95 percent. This means, for example, that the chances are 19
                          out of 20 that, if we applied our review procedures to all calender year
                          1988 FDPIR packages issued on a reviewed reservation, the results of
                          such a review would differ from the estimates obtained from our sample
                          by less than the sampling errors of such estimates.

                          Because we sampled (1) FDPIR packages and (2) food stamp issuances,
                          not households, a particular household can appear more than once over
                          the time period of our review. For example, a household that received
                          12 monthly FDPIR packages during 1988 at 1 of the 4 reservations might
                          have been the recipient of 2 sampled packages-l   in April and 1 in
                          September.


                          To provide information on the households receiving FDPIR packages, we
Selecting a Probability   took a probability sample of packages issued in 1988 at each of the four
Sample of F’DPIR          reservations and analyzed the characteristics of the households receiv-
Packages                  ing these packages, At the White Earth, pine Ridge, and F’t. Rerthold
                          reservations, we obtained lists of packages distributed in 1988 that
                          allowed us to take a separate simple random sample of packages at each
                          of these locations, The number of packages issued and sampled at the
                          three reservations where package distribution lists were available is
                          shown in table IV. 1.




                          Pye   59         GAO/XElMO-152   Recipient   and Ekpert   Viewo on Indian   Food Assistance
                                            Appendix N
                                            Methodology and Sampling Plan for Food
                                            Stamp and FDPIR Demographics




Table IV.1: FDPIR Packages Distributed
in 1988 and Sampled                                                                                                       Estimated 1988
                                                                           Packages                                             packages
                                                                       distributed in                                     represented by
                                            Reservation                          1988 Sampled             Reviewed                reviewa
                                            Ft. Berthold                         5,017          - 300              300                5,017
                                            Whtte Earth                          7.394            415              415                7.394
                                            Pine Ridge                          17,555            461              417b              15,679        (2465
                                            Total                               29,968          1,176           1,132                28,290 (+ 465)
                                            aThe numbers In parentheses are the sampling errors of the esttmates at the 95percent      level of
                                            confidence
                                            bWe were unable to locate any rnformatton other than the number of people in the household on 39
                                            cases servtced by the matn warehouse. In addrtron. the lrst of recrprents of 193 January packages from
                                            the Wamblee warehouse was unavarlable. whtch forced us to delete another 5 cases


                                            Because no package distribution list was available for the Navajo Reser-
                                            vation, we sampled at that reservation in a two-step procedure. First, we
                                            selected household casefiles from the active files and from the inactive
                                            files and recorded the number of packages issued to those households in
                                            1988. Second, we randomly selected one package for review from each
                                            household that received at least one package. We took a systematic sam-
                                            ple by selecting every 18th active casefile and every 114th inactive
                                            casefile. See table IV.2 for the number of casefiles and packages sampled
                                            for the Navajo Reservation.


Table IV.2: Estimate of Total Packages Issued in 1988 at the Navajo Reservation
                                                                   Sample casefiles 1988 packages                     Estimated 1988
                                     Number of     Casefiles         receiving 1988     in sampled                         Pa;Fuydt
Type                                  casefiles     sampled              packagesa        casef iles
Active                                     7,241             399                         336                 2,932               53,210           (~3,169
Inactive                                  13,547             113                          47                   260               31,170        (28,762
Total                                    20,768              612                         383                 3,192               64,380       (29,318)
                                             aThis IS the number of packages selected for revrew.
                                             bThe numbers in parentheses   are the sampltng errors of the estimate at the 95.percent level of confr.
                                             dence.


                                             We were able to review casefiles representing about 28,290 packages
                                             from Fort Berthold, White Earth, and Pine Ridge and 84,380 Navajo
                                             packages- a total of 112,670 packages. Because we were not always
                                             able to collect all the desired information from each casefile, the statisti-
                                             cal results in this report represent only those cases that contained suffi-
                                             cient information to perform the analysis. If more than 10 percent of the




                                             Page 60                GAO/RCED-9@152        Recipient   and Expert   Views on Indian    Food Assistnncr
                      Appendtx   N
                      Methodology and Sampling Plan for Food
                      Stamp and FDPIR Demographics




                      112,670 packages were not represented in an estimate because of miss-
                      ing information, the number of packages represented by the analysis is
                      provided in the tables in the appendixes II, III, and IV.


                      In order to provide information on the characteristics of Indian house-
Analysis of Data on   holds receiving food stamp benefits, we analyzed data, from ms’s Inte-
Indian Households     grated Quality Control System, on benefits issued in fiscal year 1988. We
Receiving Food        requested data for Indian households* located on the four reservations
                      in our review. FNS determined the local agency/geographic codes-
stamps                county or county equivalents-for     each of the four reservations. It then
                      provided us information for a given reservation located within a given
                      state if the data base contained at least 30 Indian (Native American)
                      households in the local agency code(s) in that state. FNSwas able to pro-
                      vide information only on areas that somewhat approximate the Pine
                      Ridge Reservation in South Dakota and the Navajo Reservation in New
                      Mexico and Arizona. A detailed explanation of the geographic areas cov-
                      ered by the FNSdata is given below. FNS provided us with no information
                      on the White Earth and Fort Berthold reservations or on the Navajo Res-
                      ervation located in Utah. The Pine Ridge data do not include Bennett
                      County-roughly      the southeastern quarter of the reservation.

                      F’NSprovided information related to Indian households receiving food
                      stamps, but the data were unable to distinguish households by Indian
                      tribe (for example, the Navajo Reservation in Arizona completely sur-
                      rounds the Hopi Reservation; therefore, data from Arizona include Hopi
                      as well as Navajo) or whether the household was located on or off the
                      reservation. Because the local agency code definitions did not exactly
                      coincide with reservation boundaries, the data they provided could also
                      contain Indian households located off the reservation. In South Dakota,
                      the two local agency codes comprise Shannon County (located totally on
                      the Pine Ridge Reservation) and Jackson County (of which only the
                      lower half of the county is located on the reservation). In Arizona, the
                      six local agency codes are located in three counties-Coconino, Apache,
                      and Navajo-each containing land both on and off the Navajo Reserva-
                      tion. The Hopi Reservation is located in Coconino and Navajo counties.
                      The local agency codes in New Mexico are located in six counties. Two of
                      the counties-McKinley    and San Juan-each contain land both on and
                      off the Navajo Reservation. Each of the other four counties-Bernallilo,


                      ‘FNS defined a household as Indian if at least one member of the household was coded as Native
                      American in its data base.



                      Page 61               GAO/RCED~152        Recipient   and Expert   Viewo on lndian   Food Assistance
                                             Appendix N
                                             Methodology and Sampling Plan for Food
                                             Stamp and FDPJR Demographics




                                             Sandoval, Socorro, and Cibola-contains                      Navajo tribal lands that are not
                                             part of the major Navajo reservation.

                                             We analyzed the data provided by FNS as a probability sample of food
                                             stamp issuances in a manner consistent with the sample selection infor-
                                             mation provided to us by FM. (See table IV.3 for information on sample
                                             and universe sizes.)


Table IV.3 Sample of Indian Households Located on or Near Pine Ridge Reservation, South Dakota, and Navajo Reservation in
Arizona and New Mexico
                                                              FY 88 food
                                                                   stamp Total quality        Indian Estimated issuances to
                                                            issuances in        control households      Indian households in
Reservation                   Stratum.                           stratum sample size     in sampieb      local agency areasC
Navajo                        Arizona/36                                   328,248            1,230                  94     25,086             (2 4,866
                              Arizona/37                                   493,086            1,616                 120      36,615            (r 6.295
                              New Mexico                                   568,896             1,188                145      69,436           (+ 10,583
Pine Ridge                    South Dakota                                 212,520               685                 50      15,512            (2 4.136.
Total                                                                                                               4094   146.64gd          (?13,672)
                                             %r data came from two strata In Anzona-36      and 37 New Mexico and South Dakota each had only
                                             one stratum

                                             ‘Because we took a sample of issuances not households, these may not be unique households             I e an
                                             tndlvldual household may have received more than one sampled issuance

                                             CThe numbers In parentheses   are the sampling errors of the estimates at the 95percent    level of
                                             confidence.
                                             *We excluded four of these cases, representing an estimated 1,915 (? 1,873) Issuances from analyses
                                             because the value In the fteld denoting the number of persons In the food stamp household did not
                                             match the number of individuals described In detail who were flagged as being part of the food stamp
                                             household. The remaimng 405 cases represent an estimated 144,734 (+ 13,775) Issuances in ftscal year
                                             1988 to lndlan households located In local agency areas approximating the Navajo Reservation In An-
                                             zona and New Mexico and the Pine Rdge Reservation In South Dakota. Depending on the analysis. ‘we
                                             also excluded at most 10 cases for which the state flndtng on the Quality Control Sample review was
                                             “Totally Ineligible” and 1 case in which the sex of 1 member of the food stamp household was unknown
                                             These exclusions represent an estimated 3,187 ( f 1,998) and at most 902 Issuances respectively




                                              Page 62               GAO/BcEDw)-lS2        Recipient    and Expert   Views on Indian    Food Assistancr
Appeudix V

Major Contributors to This Report


                      Gerald E. Killian, Assistant Director
Resources,            Peter M. Bramble, Jr., Assignment Manager
Community, and        Debra S. Ritt, Evaluator-in-Charge
                      Anu K. Mittal, Evaluator
Economic ” ’          Karen E. Bracey, Assistant Director
Development   Di~sW   CZU-O~YIIM. Boyce, Senior Social Science Analyst
Washington, D.C.      Judy K. Pagano, Senior Operations Research Analyst


                      David 0. Bourne, Regional Assignment Manager
Denver Regional
Office
                      Paul I. Wilson, Site Senior
Chicago Regional
Office




(028220)              Pyre 68
                                        :, .&+-

.




    Requests for copies of GAO reports            should be sent to:

    U.S. General Accounting    Office
    Post Office Box 6015
    Gaithersburg, Maryland     20877

    Telephone   202-275-6241

    The first five copies of each report          are free. Additional   copies are
    $2.00 each.

    There is a 25% discount    on orders for 100 or more copies mailed to a
    single address.

    Orders must be prepaid by cash or by check or money order made
    out to the Superintendent of Documents.
United States
General Accounting  Office
Washington, D.C. 20548

officialBllsiness
Penalty for Private   Use $300