NC219: Using Stage Based Interventions to Increase Fruit and Vegetable Intake in Young Adults
(Multistate Research Project)
Status: Inactive/Terminating
NC219: Using Stage Based Interventions to Increase Fruit and Vegetable Intake in Young Adults
Duration: 10/01/2001 to 09/30/2006
Administrative Advisor(s):
NIFA Reps:
Non-Technical Summary
Statement of Issues and Justification
Americans do not eat enough vegetables and fruits to promote optimal health and reduce the risk of chronic disease. The transition from parental control to independent adulthood that occurs for most young people between the ages of 18 and 24 is a critical stage for developing food habits affecting health throughout life. In addition, many young adults are starting their own families and are modeling behaviors that will influence the next generation. Intervention materials tailored to stage of change have been effective in promoting behavior change for a variety of behaviors, but have not been applied to increasing the consumption of vegetables and fruits in young adults. In addition, most stage-based interventions have focused on homogeneous populations and have not been replicated with diverse populations.
Justification
Consumption of vegetables and fruits strongly relates to a decreased risk of developing chronic diseases (1,2). The 2000 edition of the Dietary Guidelines for Americans places great importance on increasing consumption of foods from plant origins (3). However, it is clear from market research and national surveys that the typical American diet includes too few vegetables and fruits to promote optimal health (4-8). It is imperative that effective methods for promoting increases in both fruit and vegetable intakes are developed. Many Americans are familiar with the Food Guide Pyramid but they dont know important concepts such as what constitutes a serving (9). Few Americans are aware of the underlying message that consuming more than the minimal number of servings and consuming a greater variety of fruits and vegetables are among the most powerful dietary strategies for preventing chronic disease. Although it is possible that more targeted knowledge-based messages could be effective, researchers have found that knowledge alone is insufficient to promote behavior change for the majority of the population (10). While the Stages of Change Model shows promise for promoting change in dietary behavior, it has only recently been employed for promoting increases in fruit and vegetable intakes (11-13). Young adults between the ages of 18 to 24 are similar to older adults in their insufficient consumption of fruits and vegetables (4, 14-16). Young adults are a particularly vulnerable group because 1) the food habits they form at this transitional stage of life set the stage for their later years, 2) intake of fruit and vegetables starting at this age will have the maximum preventive potential, and 3) many of these young adults will have families and will model dietary behavior for their children.
Extent of the problem
Young adults are at a critical stage of development and are making the transition from parental control of their food intake to being responsible for their own, and often their children's, food intake (17-19). Results from national and regional surveys show the typical 18 to 24 year old diet is low in vegetables and fruits (4, 6, 14-16). Vegetables and fruits consumed tend to lack variety with habitual intake of a limited number of food items from each group (16). Food habits adopted during the young adult years have long-term effects on health throughout life and are the habits taught to the next generation of children. The results of the proposed research are intended to diminish the risk of chronic disease by promoting increased consumption of a wide variety of vegetables and fruits for young adults.
Need for cooperative work
In previous studies (NC 200 and 219 Regional Research) we have established a strong record of collaborative research as demonstrated by our publication record with multiple authors from different institutions (see Critical Review). These projects were developed and implemented collaboratively. For example, the survey completed last year was developed and implemented in ten states using the same instrument and survey design. We recruited 2,930 subjects with a range of 244 to 367 per state. Recruiting a sample of this size would have been prohibitive for any single institution. In addition, there were significant differences in demographic variables between states. An average of 28.5% were non-students with a high of 33% in ME and a low of 20% in NY. ME also had the highest proportion rural (52%) and NY the lowest (9%) with an average of 23%. Although overall 90% of the sample was Caucasian, AL had 45% African Americans and WI only 1%. The proposed project will focus on underserved populations such as African American, Hispanic, economically disadvantaged, and non-college students. Participating institutions will form cooperative arrangements with programs such as EFNEP, WIC, and Food Stamps to help in intervention development and recruitment. Using the same assessment methodology and intervention design, we can compare effectiveness of the intervention between populations as well as assess overall intervention effectiveness. This research could not be conducted without a collaborative effort across multiple states. The cooperation of a multi-state regional project benefits the proposed research by offering access to a wider variety of young adults than would be possible at a single institution. The larger and more diverse sample afforded through cooperative work increases the power and generalizibility of the results.
Benefits
In the proposed research we will demonstrate a benefit to consumers in the target audience by their increased consumption of vegetables and fruits. The qualitative study planned for Year 01 will provide data about preferred methods of delivery of nutrition information in diverse populations that can benefit programs working with these populations. Methods developed and found effective for promoting behavior change in food intake will benefit health educators who can adapt the methods for use with their audiences. This theory-based research will test the efficacy of stage tailored methods for behavior change in diverse populations. The methods developed may demonstrate a benefit to those in food marketing by providing tested ways to increase targeted food purchases. The methods used in this project should be of interest to agencies within USDA and NIH. Successful completion of this project would provide much needed information about the Stages of Change model and about the efficacy of the model for interventions to promote healthful food choices as part of a lifestyle pattern to prevent chronic diseases in diverse populations. The proposed project with its emphasis on diversity and systematic data collection and analysis will increase the possibility of obtaining additional grant funding in this area.
Impact on science
The proposed research is unique in at least two ways. Young adults as an age group have been neglected, perhaps due to their apparent health and vigor. The few investigators who have examined food habits and factors influencing food intake have focused primarily on college students. In our research, NC 200 and 219 gained access to young adults who were not college students. Our findings revealed strong influences on food intake and that food intake and influences on food intake differed depending on student status. The proposed research would further the advancement of our understanding of what motivates young adults to change food habits. Of greater importance, newer models for the study of behavior change, most notably the Stages of Change model, have shown promise for promoting some kinds of behavior change (20-23). Very few researchers have used the model to increase consumption of fruits and vegetables and fewer studies have been conducted with young adults as the target group. The proposed research would be one of the first to develop and test model-based nutrition interventions for young adults in diverse populations.
Other impacts
Early intervention with young adults for improvement in personal health status should lower total costs of health care for the nation over the long-term (24). Young adults who increase their consumption of vegetables and fruits would, by virtue of being a role model, transfer this food habit to their children and peers (17). There is evidence that health habits tend to cluster, suggesting the possibility that adopting healthier eating habits may lead to other positive health habits such as increased exercise or decreased smoking (24). Higher consumption of vegetables and fruits would provide an economic benefit to producers (25). While it is possible that certain individuals who increase their vegetable and fruit intake might concurrently decrease their intake of dairy and meat products, our research focuses on increasing consumption of plant products rather than decreasing consumption of other food products. An increased consumption of vegetables and fruits would result in an increased level of dietary fiber, folic acid and other nutrients in which the typical U.S. diet tends to be inadequate as well as increasing consumption of phytochemicals with disease preventing potential (4).
Related, Current and Previous Work
When it comes to food, young adults often say that time/convenience, money and health concerns top the list of reasons why they eat the way they do (19). However, when these concerns were examined in more detail, as was done in NC 200, force of habit seems to have a stronger influence on what young adults choose to eat. Gifft et al. (26), in their classic work, defined food habits not only as characteristic and repetitive ways to provide oneself with nourishment but also as ways to simultaneously attain social and emotional goals. Once formed, food habits tend to control food-related behaviors and are slow to change.
Data from NC 200 indicated that young adults who were not students ate higher fat meats and dairy products than did college students or college graduates (27). Also, cultural heritage may make some differences as Anglos were found to consume more high fat, quick preparation foods and less milk than Hispanics. Hispanics consumed more regular soft drinks and added cream and sugar to coffee and tea more often. For the young adults who responded to the NC 200 survey, only 28% of the males and 18% of the females reported eating three or more servings of vegetables daily and less than half of both genders (48% of males and 37% of females) reported 2 servings daily (28). By collapsing the factors influencing food intake for the NC 200 sample, the following four key factors emerged as strong perceptions held by young adults about foods (29): the social and physical aspects of food; the healthfulness of food with "fattening" attributes perceived as unhealthy; importance of nutrition tied with adequacy of money and food preparation/purchasing ability; and access to stores and cooking facilities. Various sub-samples showed slightly different perceptions but the main core of four are intact across subgroups. By targeting these perceptions, effective nutrition interventions can be developed which will promote positive dietary behavior changes. Results of the NC 219 survey will be presented briefly below and presented in greater detail in the Critical Review section.
Objectives
-
Qualitatively assess the preferred delivery method, as well as the acceptability of stage tailored newsletters in young adults in diverse populations.
-
Develop a sustained, 6-month stage-tailored intervention designed to increase consumption of vegetables and fruits tailored for diverse populations of young adults.
-
Test the efficacy of the intervention compared to a non-treatment control and compare the efficacy across diverse populations.
-
Methods
Objective 1In order to develop effective interventions for young adults, individually tailored to their stage of readiness to change, we will need to determine the optimal delivery method. Anticipated differences in demographic factors have prompted us to select six target groups; each group will be assessed by at least two different institutions following the same protocol. Non-students will only be targeted by one institution (ME) because it is anticipated that the other target groups will contain a large proportion of non-students. We will actively recruit an additional institution to ensure that African Americans are assessed by two institutions. Following is the list of participating institution and their target groups:
African American | AL |
Economically disadvantaged | MI, IA, SD |
Hispanic | RI, KS |
Non-student1 | ME |
Rural | NE, WI |
Student | OE, NY |
1Non-students are defined as 1) not currently in college, and 2) not a college graduate |
Current NC 219 Project
The assessment for the renewal will follow a similar protocol to the one used by the current projects 5th Year Committee. The Committee (KS, OR, MI, NY, IA, AL, SD) is currently testing stage-tailored newsletters for Precontemplation (PC), Preparation (P), and Maintenance (M) in focus groups of young adults categorized into stage of change for consuming 3 3 servings of vegetables a day. Newsletters, adapted by RI from previous materials, are brief, 24 page newsletters with stage matched messages designed to increase motivational readiness to increase intake (PC and Contemplation), to increase intake to meet the criterion (P), and to increase amount and variety of intake (Action and M). Material is presented in column style with color graphics and uses 14 point type. Although a set of 6 newsletters per stage have been adapted by RI, only the first two issues for PC, P, and M have been modified by the Committee and are being evaluated. These six newsletters were modified as follows: 1) tailored using data from the current project 2) tailored for states (logo on masthead and state specific resources included), 3) a clear theme per issue was inserted, 4) a story was included in each newsletter, and 5) language was modified as appropriate. A specific set of questions has been developed including questions assessing preferred method of delivery of nutrition information. Focus group leaders have been trained and are assisted by a recorder to tape record the session. Focus groups are using convenience samples and include 6 10 participants. Each state is conducting six focus groups (two per stage) except for KS which is conducting three focus groups (one per stage). KS is specifically targeting recruitment of non-students for their groups. Focus groups are being transcribed and the leader and recorder will informally summarize data about the preferred delivery method. Transcripts will be analyzed. If participants clearly identify another method of delivery than newsletters, additional institutions will be invited to join NC 219 based on expertise necessary for qualitative assessment of different delivery methods. Revisions to this protocol will be made based on results and a final protocol established at the Annual Meeting in Year 01.
NC-219 Renewal
Recruitment for focus groups in Year 01 will be targeted for the specific population being assessed and incentives for participation provided. Targeting specific populations will require formation of collaborative arrangements with EFNEP, the Food Stamp Program, the WIC program or other groups that will provide access to the target population of young adults. Each institution will develop specific plans for this collaboration. Newsletters or other sample intervention material will be modified to be appropriate for the specific target population and will be reviewed by collaborating groups. For example, the material will be translated into Spanish for Hispanic groups and will include appropriate examples of fruits and vegetables. Sample intervention materials will be compared (if more than one method is being evaluated) or the question about preferred method of delivery repeated. Participating institutions will conduct 6 focus groups (2 per stage PC, P, and M) in Year 01 and, as indicated above, at least two institutions will assess each target group. A minimum of 72 focus groups will be conducted in Year 01 following the same protocol. Tapes will be transcribed and reviewed by the group leader and recorder to correct or clarify as well as to identify nonverbal emphases, the data will then be analyzed. (NE will take the lead in all data analyses).There will be a conference call discussing results of the investigation into the preferred delivery method. If participants clearly identify another method of delivery than newsletters, additional institutions will be invited to join NC-219 based on expertise necessary for development of this different delivery method.
We anticipate writing a manuscript based on the qualitative data and will use the data for grant proposal development. We will begin seeking additional funding for the intervention during Year 02.
Objective 1 Procedures
Each participating institution will accomplish the following in Year 01:
- Tailor prototype intervention materials for one of the target groups listed above
- Conduct 6 focus groups with the target group and transcribe the data
In addition, we will recruit an additional institution to help conduct focus groups with African Americans.
Objective 2
Once a delivery method has been selected, intervention development can begin. Additional members of the NC 219 group may be invited with expertise in the selected delivery method. The focus of Year 02 will be development, prototype testing, and revision of the intervention materials. If the delivery method remains newsletters, the initial decision at the Annual Meeting will be to decide the optimal level of individual tailoring beyond stage of change and which variables will be included in tailoring. If the intervention is multimedia, the tailoring will include feedback on stage, behavior, decisional balance, and self-efficacy. However, if the intervention is multimedia, additional funding will be required to move beyond prototype testing.
Short forms of instruments assessing behavior, stage of change, and self-efficacy have been developed and validated in NC-219. These instruments, assessing fruit and vegetables separately, present minimal subject burden and can be used for tailoring interventions. The instrument assessing decisional balance has been developed and validated. The current decisional balance instrument will be reduced from 36 items to a short form (1214 items) to reduce subject burden. RI is developing this instrument as its 5th year objective, using the survey data base. If the group determines that tailoring should include feedback on use of processes of change, an additional instrument will need to be developed and validated. MI is currently working on assessing process of change use for fruits and vegetables.
If the results of Objective 1 indicate that newsletters are the optimal choice, then the remaining set of 6 newsletters per stage will be modified and tailored for specific populations as was done in the qualitative assessment described above. If the method changes based on the qualitative research, similar procedures will be employed to develop and tailor the intervention. In any case, active involvement in intervention development and tailoring is expected from collaborating groups. The goal of Year 02 will be to develop a sustained, 6-month intervention with monthly mailings of intervention materials. This will include assessment at 3 months followed by feedback on progress. If there has been stage movement, the materials will reflect the new stage.
Year 03 will focus on pilot testing and refining intervention materials. Each institution will initiate a small (N=25) pilot test of the intervention. This will include recruitment, provision of the intervention, and both qualitative and quantitative evaluation. It is anticipated that recruitment for certain target populations may be focused and utilize the collaborating programs to assist in recruitment. For example, Hispanic young adults may be recruited from WIC program clients at clinic visits. After the 6-month intervention, study variables will be reassessed and institutions will conduct brief process interviews about what intervention components were liked, what were not liked, and what could be done to improve the intervention. Based on the results of the pilot test, it is likely that there will be additional refinements made in the intervention. NE will enter the data and provide data sets to all participating institutions for data analysis.
Year 03 will include active proposal writing and manuscript development based on results of the pilot.
Objective 2 Procedures
Each participating institution will accomplish the following in Years 02 and 03:
- Complete intervention development for the target group
- Pilot test the intervention with 25 subjects from the target group and conduct post-intervention qualitative and quantitative assessment
Objective 3
Once the intervention has been fully developed, we will use an experimental design to test the efficacy of the intervention. Year 04 will focus on delivering the intervention with recruitment and intervention scheduled between October and November. The intervention will be 6 months with post treatment assessment in April and May. The final assessment will be at 12 months (during the 5th year). There will be two different methods of recruiting the sample. Institutions focusing on students and non-students will use the method utilized in the current project (random lists of young adults in different regions of the state) with proactive recruitment of eligible subjects. Proactive recruitment is important because Precontemplators tend not to respond to reactive recruitment procedures. The institutions will use a stratified sampling procedure to ensure an adequate representation of non-students. This will yield a representative sample, but the sample is likely to be primarily Caucasian. Therefore, it will be necessary to use targeted recruitment for specific additional populations. For example, Hispanic WIC clients may be recruited. Although the targeted recruiting may not be as generalizable, results should be valid for the specific population and can be rapidly disseminated to similar populations. Each state will recruit 250 subjects for a total sample size of 3000. This is similar to the sample recruited in the current project. Regardless of recruiting method, the design will be experimental with random assignment within state to experimental or control groups. The experimental group will receive a sustained 6-month, stage tailored intervention with assessment of both experimental and control groups at baseline, 6 and 12 months. The experimental group will be reassed at 3 months to provide feedback on progress. Attrition of 20% is anticipated leaving 2400 subjects at follow-up (400 subjects per target group). Primary outcome measures will be servings of fruits and vegetables measured on the food frequency questionnaire developed in NC-219. Our hypothesis is that the experimental group will increase consumption of fruits and vegetables combined by = serving more than the control group. In addition, we will calculate a variety score (number of different fruits and vegetables consumed) and anticipate that variety will be greater in the experimental group. As in the current project, NE will enter the data and will provide data sets to all participating institutions.
One of the primary aims of data analysis will be the comparison of the effectiveness of the intervention across diverse populations. With 400 subjects from each target population at follow-up, meaningful comparisons can be made between population groups. In addition, we will be able to assess differences in variables such as motivational readiness to change, decisional balance, and perceived self-efficacy to see if process-to-outcome patterns differ between groups.
We will conduct interviews with a randomly selected subsample of participants (10%) post intervention, as with the pilot, to collect process data.
Objective 3 Procedures
Each participating institution will accomplish the following in Years 04-05:
- Recruit a sample of 250 subjects, deliver the intervention, and evaluate the results at 6 and 12 months
- Conduct process interviews with a randomly selected sample of 25 subjects at 12 months
Proposed Timeline:
Year 01 |
October, 2001 - September, 2002 Conduct qualitative research exploring the best modalities of delivering interventions for young adults. |
Year 02 | October, 2002 - September 2003 Develop intervention materials Manuscripts on method of delivery research Grant proposal development |
Year 03 October, 2003 - September 2004 | Pilot and revise the intervention materials Manuscripts on the pilot Grant proposal development Process interviews |
Year 04 October, 2005 - September 2006 | October November: recruit study sample, assess and deliver the 6-month intervention. Assess immediate post-intervention impact Begin data processing and analysis Manuscript on baseline data Continue proposal development |
Year 05 October, 2006 - September 2007 | Complete 12 month follow-up Complete data analysis Conduct process interviews with young adults in the intervention group Outcome manuscript development Continue proposal development |